diff options
589 files changed, 23539 insertions, 4210 deletions
diff --git a/Documentation/ABI/testing/configfs-stp-policy b/Documentation/ABI/testing/configfs-stp-policy new file mode 100644 index 000000000000..421ce6825c66 --- /dev/null +++ b/Documentation/ABI/testing/configfs-stp-policy @@ -0,0 +1,48 @@ +What: /config/stp-policy +Date: June 2015 +KernelVersion: 4.3 +Description: + This group contains policies mandating Master/Channel allocation + for software sources wishing to send trace data over an STM + device. + +What: /config/stp-policy/<device>.<policy> +Date: June 2015 +KernelVersion: 4.3 +Description: + This group is the root of a policy; its name is a concatenation + of an stm device name to which this policy applies and an + arbitrary string. If <device> part doesn't match an existing + stm device, mkdir will fail with ENODEV; if that device already + has a policy assigned to it, mkdir will fail with EBUSY. + +What: /config/stp-policy/<device>.<policy>/device +Date: June 2015 +KernelVersion: 4.3 +Description: + STM device to which this policy applies, read only. Same as the + <device> component of its parent directory. + +What: /config/stp-policy/<device>.<policy>/<node> +Date: June 2015 +KernelVersion: 4.3 +Description: + Policy node is a string identifier that software clients will + use to request a master/channel to be allocated and assigned to + them. + +What: /config/stp-policy/<device>.<policy>/<node>/masters +Date: June 2015 +KernelVersion: 4.3 +Description: + Range of masters from which to allocate for users of this node. + Write two numbers: the first master and the last master number. + +What: /config/stp-policy/<device>.<policy>/<node>/channels +Date: June 2015 +KernelVersion: 4.3 +Description: + Range of channels from which to allocate for users of this node. + Write two numbers: the first channel and the last channel + number. + diff --git a/Documentation/ABI/testing/sysfs-bus-coresight-devices-etm3x b/Documentation/ABI/testing/sysfs-bus-coresight-devices-etm3x index d72ca1736ba4..67e3557bc4bf 100644 --- a/Documentation/ABI/testing/sysfs-bus-coresight-devices-etm3x +++ b/Documentation/ABI/testing/sysfs-bus-coresight-devices-etm3x @@ -8,13 +8,6 @@ Description: (RW) Enable/disable tracing on this specific trace entiry. of coresight components linking the source to the sink is configured and managed automatically by the coresight framework. -What: /sys/bus/coresight/devices/<memory_map>.[etm|ptm]/status -Date: November 2014 -KernelVersion: 3.19 -Contact: Mathieu Poirier <mathieu.poirier@linaro.org> -Description: (R) List various control and status registers. The specific - layout and content is driver specific. - What: /sys/bus/coresight/devices/<memory_map>.[etm|ptm]/addr_idx Date: November 2014 KernelVersion: 3.19 @@ -251,3 +244,73 @@ Date: November 2014 KernelVersion: 3.19 Contact: Mathieu Poirier <mathieu.poirier@linaro.org> Description: (RW) Define the event that controls the trigger. + +What: /sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmccr +Date: September 2015 +KernelVersion: 4.4 +Contact: Mathieu Poirier <mathieu.poirier@linaro.org> +Description: (RO) Print the content of the ETM Configuration Code register + (0x004). The value is read directly from the HW. + +What: /sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmccer +Date: September 2015 +KernelVersion: 4.4 +Contact: Mathieu Poirier <mathieu.poirier@linaro.org> +Description: (RO) Print the content of the ETM Configuration Code Extension + register (0x1e8). The value is read directly from the HW. + +What: /sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmscr +Date: September 2015 +KernelVersion: 4.4 +Contact: Mathieu Poirier <mathieu.poirier@linaro.org> +Description: (RO) Print the content of the ETM System Configuration + register (0x014). The value is read directly from the HW. + +What: /sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmidr +Date: September 2015 +KernelVersion: 4.4 +Contact: Mathieu Poirier <mathieu.poirier@linaro.org> +Description: (RO) Print the content of the ETM ID register (0x1e4). The + value is read directly from the HW. + +What: /sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmcr +Date: September 2015 +KernelVersion: 4.4 +Contact: Mathieu Poirier <mathieu.poirier@linaro.org> +Description: (RO) Print the content of the ETM Main Control register (0x000). + The value is read directly from the HW. + +What: /sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmtraceidr +Date: September 2015 +KernelVersion: 4.4 +Contact: Mathieu Poirier <mathieu.poirier@linaro.org> +Description: (RO) Print the content of the ETM Trace ID register (0x200). + The value is read directly from the HW. + +What: /sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmteevr +Date: September 2015 +KernelVersion: 4.4 +Contact: Mathieu Poirier <mathieu.poirier@linaro.org> +Description: (RO) Print the content of the ETM Trace Enable Event register + (0x020). The value is read directly from the HW. + +What: /sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmtsscr +Date: September 2015 +KernelVersion: 4.4 +Contact: Mathieu Poirier <mathieu.poirier@linaro.org> +Description: (RO) Print the content of the ETM Trace Start/Stop Conrol + register (0x018). The value is read directly from the HW. + +What: /sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmtecr1 +Date: September 2015 +KernelVersion: 4.4 +Contact: Mathieu Poirier <mathieu.poirier@linaro.org> +Description: (RO) Print the content of the ETM Enable Conrol #1 + register (0x024). The value is read directly from the HW. + +What: /sys/bus/coresight/devices/<memory_map>.[etm|ptm]/mgmt/etmtecr2 +Date: September 2015 +KernelVersion: 4.4 +Contact: Mathieu Poirier <mathieu.poirier@linaro.org> +Description: (RO) Print the content of the ETM Enable Conrol #2 + register (0x01c). The value is read directly from the HW. diff --git a/Documentation/ABI/testing/sysfs-bus-intel_th-devices-gth b/Documentation/ABI/testing/sysfs-bus-intel_th-devices-gth new file mode 100644 index 000000000000..22d0843849a8 --- /dev/null +++ b/Documentation/ABI/testing/sysfs-bus-intel_th-devices-gth @@ -0,0 +1,49 @@ +What: /sys/bus/intel_th/devices/<intel_th_id>-gth/masters/* +Date: June 2015 +KernelVersion: 4.3 +Contact: Alexander Shishkin <alexander.shishkin@linux.intel.com> +Description: (RW) Configure output ports for STP masters. Writing -1 + disables a master; any + +What: /sys/bus/intel_th/devices/<intel_th_id>-gth/outputs/[0-7]_port +Date: June 2015 +KernelVersion: 4.3 +Contact: Alexander Shishkin <alexander.shishkin@linux.intel.com> +Description: (RO) Output port type: + 0: not present, + 1: MSU (Memory Storage Unit) + 2: CTP (Common Trace Port) + 4: PTI (MIPI PTI). + +What: /sys/bus/intel_th/devices/<intel_th_id>-gth/outputs/[0-7]_drop +Date: June 2015 +KernelVersion: 4.3 +Contact: Alexander Shishkin <alexander.shishkin@linux.intel.com> +Description: (RW) Data retention policy setting: keep (0) or drop (1) + incoming data while output port is in reset. + +What: /sys/bus/intel_th/devices/<intel_th_id>-gth/outputs/[0-7]_null +Date: June 2015 +KernelVersion: 4.3 +Contact: Alexander Shishkin <alexander.shishkin@linux.intel.com> +Description: (RW) STP NULL packet generation: enabled (1) or disabled (0). + +What: /sys/bus/intel_th/devices/<intel_th_id>-gth/outputs/[0-7]_flush +Date: June 2015 +KernelVersion: 4.3 +Contact: Alexander Shishkin <alexander.shishkin@linux.intel.com> +Description: (RW) Force flush data from byte packing buffer for the output + port. + +What: /sys/bus/intel_th/devices/<intel_th_id>-gth/outputs/[0-7]_reset +Date: June 2015 +KernelVersion: 4.3 +Contact: Alexander Shishkin <alexander.shishkin@linux.intel.com> +Description: (RO) Output port is in reset (1). + +What: /sys/bus/intel_th/devices/<intel_th_id>-gth/outputs/[0-7]_smcfreq +Date: June 2015 +KernelVersion: 4.3 +Contact: Alexander Shishkin <alexander.shishkin@linux.intel.com> +Description: (RW) STP sync packet frequency for the port. Specifies the + number of clocks between mainenance packets. diff --git a/Documentation/ABI/testing/sysfs-bus-intel_th-devices-msc b/Documentation/ABI/testing/sysfs-bus-intel_th-devices-msc new file mode 100644 index 000000000000..b940c5d91cf7 --- /dev/null +++ b/Documentation/ABI/testing/sysfs-bus-intel_th-devices-msc @@ -0,0 +1,33 @@ +What: /sys/bus/intel_th/devices/<intel_th_id>-msc<msc-id>/wrap +Date: June 2015 +KernelVersion: 4.3 +Contact: Alexander Shishkin <alexander.shishkin@linux.intel.com> +Description: (RW) Configure MSC buffer wrapping. 1 == wrapping enabled. + +What: /sys/bus/intel_th/devices/<intel_th_id>-msc<msc-id>/mode +Date: June 2015 +KernelVersion: 4.3 +Contact: Alexander Shishkin <alexander.shishkin@linux.intel.com> +Description: (RW) Configure MSC operating mode: + - "single", for contiguous buffer mode (high-order alloc); + - "multi", for multiblock mode; + - "ExI", for DCI handler mode; + - "debug", for debug mode. + If operating mode changes, existing buffer is deallocated, + provided there are no active users and tracing is not enabled, + otherwise the write will fail. + +What: /sys/bus/intel_th/devices/<intel_th_id>-msc<msc-id>/nr_pages +Date: June 2015 +KernelVersion: 4.3 +Contact: Alexander Shishkin <alexander.shishkin@linux.intel.com> +Description: (RW) Configure MSC buffer size for "single" or "multi" modes. + In single mode, this is a single number of pages, has to be + power of 2. In multiblock mode, this is a comma-separated list + of numbers of pages for each window to be allocated. Number of + windows is not limited. + Writing to this file deallocates existing buffer (provided + there are no active users and tracing is not enabled) and then + allocates a new one. + + diff --git a/Documentation/ABI/testing/sysfs-bus-intel_th-devices-pti b/Documentation/ABI/testing/sysfs-bus-intel_th-devices-pti new file mode 100644 index 000000000000..df0b24fd0218 --- /dev/null +++ b/Documentation/ABI/testing/sysfs-bus-intel_th-devices-pti @@ -0,0 +1,24 @@ +What: /sys/bus/intel_th/devices/<intel_th_id>-pti/mode +Date: June 2015 +KernelVersion: 4.3 +Contact: Alexander Shishkin <alexander.shishkin@linux.intel.com> +Description: (RW) Configure PTI output width. Currently supported values + are 4, 8, 12, 16. + +What: /sys/bus/intel_th/devices/<intel_th_id>-pti/freerunning_clock +Date: June 2015 +KernelVersion: 4.3 +Contact: Alexander Shishkin <alexander.shishkin@linux.intel.com> +Description: (RW) 0: PTI trace clock acts as a strobe which only toggles + when there is trace data to send. 1: PTI trace clock is a + free-running clock. + +What: /sys/bus/intel_th/devices/<intel_th_id>-pti/clock_divider +Date: June 2015 +KernelVersion: 4.3 +Contact: Alexander Shishkin <alexander.shishkin@linux.intel.com> +Description: (RW) Configure PTI port clock divider: + - 0: Intel TH clock rate, + - 1: 1/2 Intel TH clock rate, + - 2: 1/4 Intel TH clock rate, + - 3: 1/8 Intel TH clock rate. diff --git a/Documentation/ABI/testing/sysfs-bus-intel_th-output-devices b/Documentation/ABI/testing/sysfs-bus-intel_th-output-devices new file mode 100644 index 000000000000..4d48a9451866 --- /dev/null +++ b/Documentation/ABI/testing/sysfs-bus-intel_th-output-devices @@ -0,0 +1,13 @@ +What: /sys/bus/intel_th/devices/<intel_th_id>-<device><id>/active +Date: June 2015 +KernelVersion: 4.3 +Contact: Alexander Shishkin <alexander.shishkin@linux.intel.com> +Description: (RW) Writes of 1 or 0 enable or disable trace output to this + output device. Reads return current status. + +What: /sys/bus/intel_th/devices/<intel_th_id>-msc<msc-id>/port +Date: June 2015 +KernelVersion: 4.3 +Contact: Alexander Shishkin <alexander.shishkin@linux.intel.com> +Description: (RO) Port number, corresponding to this output device on the + switch (GTH). diff --git a/Documentation/ABI/testing/sysfs-bus-mei b/Documentation/ABI/testing/sysfs-bus-mei index 20e4d1638bac..6bd45346ac7e 100644 --- a/Documentation/ABI/testing/sysfs-bus-mei +++ b/Documentation/ABI/testing/sysfs-bus-mei @@ -19,3 +19,10 @@ KernelVersion: 4.2 Contact: Tomas Winkler <tomas.winkler@intel.com> Description: Stores mei client device uuid Format: xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx + +What: /sys/bus/mei/devices/.../version +Date: Aug 2015 +KernelVersion: 4.3 +Contact: Tomas Winkler <tomas.winkler@intel.com> +Description: Stores mei client protocol version + Format: %d diff --git a/Documentation/ABI/testing/sysfs-class-fpga-manager b/Documentation/ABI/testing/sysfs-class-fpga-manager new file mode 100644 index 000000000000..23056c532fdd --- /dev/null +++ b/Documentation/ABI/testing/sysfs-class-fpga-manager @@ -0,0 +1,37 @@ +What: /sys/class/fpga_manager/<fpga>/name +Date: August 2015 +KernelVersion: 4.3 +Contact: Alan Tull <atull@opensource.altera.com> +Description: Name of low level fpga manager driver. + +What: /sys/class/fpga_manager/<fpga>/state +Date: August 2015 +KernelVersion: 4.3 +Contact: Alan Tull <atull@opensource.altera.com> +Description: Read fpga manager state as a string. + The intent is to provide enough detail that if something goes + wrong during FPGA programming (something that the driver can't + fix) then userspace can know, i.e. if the firmware request + fails, that could be due to not being able to find the firmware + file. + + This is a superset of FPGA states and fpga manager driver + states. The fpga manager driver is walking through these steps + to get the FPGA into a known operating state. It's a sequence, + though some steps may get skipped. Valid FPGA states will vary + by manufacturer; this is a superset. + + * unknown = can't determine state + * power off = FPGA power is off + * power up = FPGA reports power is up + * reset = FPGA held in reset state + * firmware request = firmware class request in progress + * firmware request error = firmware request failed + * write init = preparing FPGA for programming + * write init error = Error while preparing FPGA for + programming + * write = FPGA ready to receive image data + * write error = Error while programming + * write complete = Doing post programming steps + * write complete error = Error while doing post programming + * operating = FPGA is programmed and operating diff --git a/Documentation/ABI/testing/sysfs-class-mic.txt b/Documentation/ABI/testing/sysfs-class-mic.txt index 13f48afc534f..d45eed2bf128 100644 --- a/Documentation/ABI/testing/sysfs-class-mic.txt +++ b/Documentation/ABI/testing/sysfs-class-mic.txt @@ -41,18 +41,15 @@ Description: When read, this entry provides the current state of an Intel MIC device in the context of the card OS. Possible values that will be read are: - "offline" - The MIC device is ready to boot the card OS. On + "ready" - The MIC device is ready to boot the card OS. On reading this entry after an OSPM resume, a "boot" has to be written to this entry if the card was previously shutdown during OSPM suspend. - "online" - The MIC device has initiated booting a card OS. + "booting" - The MIC device has initiated booting a card OS. + "online" - The MIC device has completed boot and is online "shutting_down" - The card OS is shutting down. + "resetting" - A reset has been initiated for the MIC device "reset_failed" - The MIC device has failed to reset. - "suspending" - The MIC device is currently being prepared for - suspend. On reading this entry, a "suspend" has to be written - to the state sysfs entry to ensure the card is shutdown during - OSPM suspend. - "suspended" - The MIC device has been suspended. When written, this sysfs entry triggers different state change operations depending upon the current state of the card OS. @@ -62,8 +59,6 @@ Description: sysfs entries. "reset" - Initiates device reset. "shutdown" - Initiates card OS shutdown. - "suspend" - Initiates card OS shutdown and also marks the card - as suspended. What: /sys/class/mic/mic(x)/shutdown_status Date: October 2013 @@ -126,7 +121,7 @@ Description: the card. This sysfs entry can be written with the following valid strings: a) linux - Boot a Linux image. - b) elf - Boot an elf image for flash updates. + b) flash - Boot an image for flash updates. What: /sys/class/mic/mic(x)/log_buf_addr Date: October 2013 @@ -155,3 +150,17 @@ Description: daemon to set the log buffer length address. The correct log buffer length address to be written can be found in the System.map file of the card OS. + +What: /sys/class/mic/mic(x)/heartbeat_enable +Date: March 2015 +KernelVersion: 3.20 +Contact: Ashutosh Dixit <ashutosh.dixit@intel.com> +Description: + The MIC drivers detect and inform user space about card crashes + via a heartbeat mechanism (see the description of + shutdown_status above). User space can turn off this + notification by setting heartbeat_enable to 0 and enable it by + setting this entry to 1. If this notification is disabled it is + the responsibility of user space to detect card crashes via + alternative means such as a network ping. This setting is + enabled by default. diff --git a/Documentation/ABI/testing/sysfs-class-stm b/Documentation/ABI/testing/sysfs-class-stm new file mode 100644 index 000000000000..c9aa4f3fc9a7 --- /dev/null +++ b/Documentation/ABI/testing/sysfs-class-stm @@ -0,0 +1,14 @@ +What: /sys/class/stm/<stm>/masters +Date: June 2015 +KernelVersion: 4.3 +Contact: Alexander Shishkin <alexander.shishkin@linux.intel.com> +Description: + Shows first and last available to software master numbers on + this STM device. + +What: /sys/class/stm/<stm>/channels +Date: June 2015 +KernelVersion: 4.3 +Contact: Alexander Shishkin <alexander.shishkin@linux.intel.com> +Description: + Shows the number of channels per master on this STM device. diff --git a/Documentation/ABI/testing/sysfs-class-stm_source b/Documentation/ABI/testing/sysfs-class-stm_source new file mode 100644 index 000000000000..57b8dd39bbf7 --- /dev/null +++ b/Documentation/ABI/testing/sysfs-class-stm_source @@ -0,0 +1,11 @@ +What: /sys/class/stm_source/<stm_source>/stm_source_link +Date: June 2015 +KernelVersion: 4.3 +Contact: Alexander Shishkin <alexander.shishkin@linux.intel.com> +Description: + stm_source device linkage to stm device, where its tracing data + is directed. Reads return an existing connection or "<none>" if + this stm_source is not connected to any stm device yet. + Write an existing (registered) stm device's name here to + connect that device. If a device is already connected to this + stm_source, it will first be disconnected. diff --git a/Documentation/Changes b/Documentation/Changes index 6d8863004858..f447f0516f07 100644 --- a/Documentation/Changes +++ b/Documentation/Changes @@ -43,7 +43,7 @@ o udev 081 # udevd --version o grub 0.93 # grub --version || grub-install --version o mcelog 0.6 # mcelog --version o iptables 1.4.2 # iptables -V -o openssl & libcrypto 1.0.1k # openssl version +o openssl & libcrypto 1.0.0 # openssl version Kernel compilation diff --git a/Documentation/device-mapper/snapshot.txt b/Documentation/device-mapper/snapshot.txt index 0d5bc46dc167..ad6949bff2e3 100644 --- a/Documentation/device-mapper/snapshot.txt +++ b/Documentation/device-mapper/snapshot.txt @@ -41,9 +41,13 @@ useless and be disabled, returning errors. So it is important to monitor the amount of free space and expand the <COW device> before it fills up. <persistent?> is P (Persistent) or N (Not persistent - will not survive -after reboot). -The difference is that for transient snapshots less metadata must be -saved on disk - they can be kept in memory by the kernel. +after reboot). O (Overflow) can be added as a persistent store option +to allow userspace to advertise its support for seeing "Overflow" in the +snapshot status. So supported store types are "P", "PO" and "N". + +The difference between persistent and transient is with transient +snapshots less metadata must be saved on disk - they can be kept in +memory by the kernel. * snapshot-merge <origin> <COW device> <persistent> <chunksize> diff --git a/Documentation/devicetree/bindings/input/cypress,cyapa.txt b/Documentation/devicetree/bindings/input/cypress,cyapa.txt index 635a3b036630..8d91ba9ff2fd 100644 --- a/Documentation/devicetree/bindings/input/cypress,cyapa.txt +++ b/Documentation/devicetree/bindings/input/cypress,cyapa.txt @@ -25,7 +25,7 @@ Example: /* Cypress Gen3 touchpad */ touchpad@67 { compatible = "cypress,cyapa"; - reg = <0x24>; + reg = <0x67>; interrupt-parent = <&gpio>; interrupts = <2 IRQ_TYPE_EDGE_FALLING>; /* GPIO 2 */ wakeup-source; diff --git a/Documentation/devicetree/bindings/interrupt-controller/qca,ath79-misc-intc.txt b/Documentation/devicetree/bindings/interrupt-controller/qca,ath79-misc-intc.txt index 391717a68f3b..ec96b1f01478 100644 --- a/Documentation/devicetree/bindings/interrupt-controller/qca,ath79-misc-intc.txt +++ b/Documentation/devicetree/bindings/interrupt-controller/qca,ath79-misc-intc.txt @@ -4,8 +4,8 @@ The MISC interrupt controller is a secondary controller for lower priority interrupt. Required Properties: -- compatible: has to be "qca,<soctype>-cpu-intc", "qca,ar7100-misc-intc" - as fallback +- compatible: has to be "qca,<soctype>-cpu-intc", "qca,ar7100-misc-intc" or + "qca,<soctype>-cpu-intc", "qca,ar7240-misc-intc" - reg: Base address and size of the controllers memory area - interrupt-parent: phandle of the parent interrupt controller. - interrupts: Interrupt specifier for the controllers interrupt. @@ -13,6 +13,9 @@ Required Properties: - #interrupt-cells : Specifies the number of cells needed to encode interrupt source, should be 1 +Compatible fallback depends on the SoC. Use ar7100 for ar71xx and ar913x, +use ar7240 for all other SoCs. + Please refer to interrupts.txt in this directory for details of the common Interrupt Controllers bindings used by client devices. @@ -28,3 +31,16 @@ Example: interrupt-controller; #interrupt-cells = <1>; }; + +Another example: + + interrupt-controller@18060010 { + compatible = "qca,ar9331-misc-intc", qca,ar7240-misc-intc"; + reg = <0x18060010 0x4>; + + interrupt-parent = <&cpuintc>; + interrupts = <6>; + + interrupt-controller; + #interrupt-cells = <1>; + }; diff --git a/Documentation/devicetree/bindings/nvmem/imx-ocotp.txt b/Documentation/devicetree/bindings/nvmem/imx-ocotp.txt new file mode 100644 index 000000000000..383d5889e95a --- /dev/null +++ b/Documentation/devicetree/bindings/nvmem/imx-ocotp.txt @@ -0,0 +1,20 @@ +Freescale i.MX6 On-Chip OTP Controller (OCOTP) device tree bindings + +This binding represents the on-chip eFuse OTP controller found on +i.MX6Q/D, i.MX6DL/S, i.MX6SL, and i.MX6SX SoCs. + +Required properties: +- compatible: should be one of + "fsl,imx6q-ocotp" (i.MX6Q/D/DL/S), + "fsl,imx6sl-ocotp" (i.MX6SL), or + "fsl,imx6sx-ocotp" (i.MX6SX), followed by "syscon". +- reg: Should contain the register base and length. +- clocks: Should contain a phandle pointing to the gated peripheral clock. + +Example: + + ocotp: ocotp@021bc000 { + compatible = "fsl,imx6q-ocotp", "syscon"; + reg = <0x021bc000 0x4000>; + clocks = <&clks IMX6QDL_CLK_IIM>; + }; diff --git a/Documentation/devicetree/bindings/nvmem/mxs-ocotp.txt b/Documentation/devicetree/bindings/nvmem/mxs-ocotp.txt new file mode 100644 index 000000000000..daebce9e6b07 --- /dev/null +++ b/Documentation/devicetree/bindings/nvmem/mxs-ocotp.txt @@ -0,0 +1,25 @@ +On-Chip OTP Memory for Freescale i.MX23/i.MX28 + +Required properties : +- compatible : + - "fsl,imx23-ocotp" for i.MX23 + - "fsl,imx28-ocotp" for i.MX28 +- #address-cells : Should be 1 +- #size-cells : Should be 1 +- reg : Address and length of OTP controller registers +- clocks : Should contain a reference to the hbus clock + += Data cells = +Are child nodes of mxs-ocotp, bindings of which as described in +bindings/nvmem/nvmem.txt + +Example for i.MX28: + + ocotp: ocotp@8002c000 { + compatible = "fsl,imx28-ocotp", "fsl,ocotp"; + #address-cells = <1>; + #size-cells = <1>; + reg = <0x8002c000 0x2000>; + clocks = <&clks 25>; + status = "okay"; + }; diff --git a/Documentation/devicetree/bindings/nvmem/rockchip-efuse.txt b/Documentation/devicetree/bindings/nvmem/rockchip-efuse.txt new file mode 100644 index 000000000000..8f86ab3b1046 --- /dev/null +++ b/Documentation/devicetree/bindings/nvmem/rockchip-efuse.txt @@ -0,0 +1,38 @@ += Rockchip eFuse device tree bindings = + +Required properties: +- compatible: Should be "rockchip,rockchip-efuse" +- reg: Should contain the registers location and exact eFuse size +- clocks: Should be the clock id of eFuse +- clock-names: Should be "pclk_efuse" + += Data cells = +Are child nodes of eFuse, bindings of which as described in +bindings/nvmem/nvmem.txt + +Example: + + efuse: efuse@ffb40000 { + compatible = "rockchip,rockchip-efuse"; + reg = <0xffb40000 0x20>; + #address-cells = <1>; + #size-cells = <1>; + clocks = <&cru PCLK_EFUSE256>; + clock-names = "pclk_efuse"; + + /* Data cells */ + cpu_leakage: cpu_leakage { + reg = <0x17 0x1>; + }; + }; + += Data consumers = +Are device nodes which consume nvmem data cells. + +Example: + + cpu_leakage { + ... + nvmem-cells = <&cpu_leakage>; + nvmem-cell-names = "cpu_leakage"; + }; diff --git a/Documentation/devicetree/bindings/nvmem/vf610-ocotp.txt b/Documentation/devicetree/bindings/nvmem/vf610-ocotp.txt new file mode 100644 index 000000000000..56ed481c3e26 --- /dev/null +++ b/Documentation/devicetree/bindings/nvmem/vf610-ocotp.txt @@ -0,0 +1,19 @@ +On-Chip OTP Memory for Freescale Vybrid + +Required Properties: + compatible: + - "fsl,vf610-ocotp" for VF5xx/VF6xx + #address-cells : Should be 1 + #size-cells : Should be 1 + reg : Address and length of OTP controller and fuse map registers + clocks : ipg clock we associate with the OCOTP peripheral + +Example for Vybrid VF5xx/VF6xx: + + ocotp: ocotp@400a5000 { + compatible = "fsl,vf610-ocotp"; + #address-cells = <1>; + #size-cells = <1>; + reg = <0x400a5000 0xCF0>; + clocks = <&clks VF610_CLK_OCOTP>; + }; diff --git a/Documentation/devicetree/bindings/spi/sh-msiof.txt b/Documentation/devicetree/bindings/spi/sh-msiof.txt index 8f771441be60..705075da2f10 100644 --- a/Documentation/devicetree/bindings/spi/sh-msiof.txt +++ b/Documentation/devicetree/bindings/spi/sh-msiof.txt @@ -51,7 +51,7 @@ Optional properties, deprecated for soctype-specific bindings: - renesas,tx-fifo-size : Overrides the default tx fifo size given in words (default is 64) - renesas,rx-fifo-size : Overrides the default rx fifo size given in words - (default is 64, or 256 on R-Car Gen2) + (default is 64) Pinctrl properties might be needed, too. See Documentation/devicetree/bindings/pinctrl/renesas,*. diff --git a/Documentation/devicetree/bindings/usb/renesas_usbhs.txt b/Documentation/devicetree/bindings/usb/renesas_usbhs.txt index 64a4ca6cf96f..7d48f63db44e 100644 --- a/Documentation/devicetree/bindings/usb/renesas_usbhs.txt +++ b/Documentation/devicetree/bindings/usb/renesas_usbhs.txt @@ -5,6 +5,7 @@ Required properties: - "renesas,usbhs-r8a7790" - "renesas,usbhs-r8a7791" - "renesas,usbhs-r8a7794" + - "renesas,usbhs-r8a7795" - reg: Base address and length of the register for the USBHS - interrupts: Interrupt specifier for the USBHS - clocks: A list of phandle + clock specifier pairs diff --git a/Documentation/devicetree/bindings/w1/omap-hdq.txt b/Documentation/devicetree/bindings/w1/omap-hdq.txt index fef794741bd1..913c5f91a0f9 100644 --- a/Documentation/devicetree/bindings/w1/omap-hdq.txt +++ b/Documentation/devicetree/bindings/w1/omap-hdq.txt @@ -1,11 +1,15 @@ * OMAP HDQ One wire bus master controller Required properties: -- compatible : should be "ti,omap3-1w" +- compatible : should be "ti,omap3-1w" or "ti,am4372-hdq" - reg : Address and length of the register set for the device - interrupts : interrupt line. - ti,hwmods : "hdq1w" +Optional properties: +- ti,mode: should be "hdq": HDQ mode "1w": one-wire mode. + If not specified HDQ mode is implied. + Example: - From omap3.dtsi @@ -14,4 +18,5 @@ Example: reg = <0x480b2000 0x1000>; interrupts = <58>; ti,hwmods = "hdq1w"; + ti,mode = "hdq"; }; diff --git a/Documentation/fpga/fpga-mgr.txt b/Documentation/fpga/fpga-mgr.txt new file mode 100644 index 000000000000..ce3e84fa9023 --- /dev/null +++ b/Documentation/fpga/fpga-mgr.txt @@ -0,0 +1,171 @@ +FPGA Manager Core + +Alan Tull 2015 + +Overview +======== + +The FPGA manager core exports a set of functions for programming an FPGA with +an image. The API is manufacturer agnostic. All manufacturer specifics are +hidden away in a low level driver which registers a set of ops with the core. +The FPGA image data itself is very manufacturer specific, but for our purposes +it's just binary data. The FPGA manager core won't parse it. + + +API Functions: +============== + +To program the FPGA from a file or from a buffer: +------------------------------------------------- + + int fpga_mgr_buf_load(struct fpga_manager *mgr, u32 flags, + const char *buf, size_t count); + +Load the FPGA from an image which exists as a buffer in memory. + + int fpga_mgr_firmware_load(struct fpga_manager *mgr, u32 flags, + const char *image_name); + +Load the FPGA from an image which exists as a file. The image file must be on +the firmware search path (see the firmware class documentation). + +For both these functions, flags == 0 for normal full reconfiguration or +FPGA_MGR_PARTIAL_RECONFIG for partial reconfiguration. If successful, the FPGA +ends up in operating mode. Return 0 on success or a negative error code. + + +To get/put a reference to a FPGA manager: +----------------------------------------- + + struct fpga_manager *of_fpga_mgr_get(struct device_node *node); + + void fpga_mgr_put(struct fpga_manager *mgr); + +Given a DT node, get an exclusive reference to a FPGA manager or release +the reference. + + +To register or unregister the low level FPGA-specific driver: +------------------------------------------------------------- + + int fpga_mgr_register(struct device *dev, const char *name, + const struct fpga_manager_ops *mops, + void *priv); + + void fpga_mgr_unregister(struct device *dev); + +Use of these two functions is described below in "How To Support a new FPGA +device." + + +How to write an image buffer to a supported FPGA +================================================ +/* Include to get the API */ +#include <linux/fpga/fpga-mgr.h> + +/* device node that specifies the FPGA manager to use */ +struct device_node *mgr_node = ... + +/* FPGA image is in this buffer. count is size of the buffer. */ +char *buf = ... +int count = ... + +/* flags indicates whether to do full or partial reconfiguration */ +int flags = 0; + +int ret; + +/* Get exclusive control of FPGA manager */ +struct fpga_manager *mgr = of_fpga_mgr_get(mgr_node); + +/* Load the buffer to the FPGA */ +ret = fpga_mgr_buf_load(mgr, flags, buf, count); + +/* Release the FPGA manager */ +fpga_mgr_put(mgr); + + +How to write an image file to a supported FPGA +============================================== +/* Include to get the API */ +#include <linux/fpga/fpga-mgr.h> + +/* device node that specifies the FPGA manager to use */ +struct device_node *mgr_node = ... + +/* FPGA image is in this file which is in the firmware search path */ +const char *path = "fpga-image-9.rbf" + +/* flags indicates whether to do full or partial reconfiguration */ +int flags = 0; + +int ret; + +/* Get exclusive control of FPGA manager */ +struct fpga_manager *mgr = of_fpga_mgr_get(mgr_node); + +/* Get the firmware image (path) and load it to the FPGA */ +ret = fpga_mgr_firmware_load(mgr, flags, path); + +/* Release the FPGA manager */ +fpga_mgr_put(mgr); + + +How to support a new FPGA device +================================ +To add another FPGA manager, write a driver that implements a set of ops. The +probe function calls fpga_mgr_register(), such as: + +static const struct fpga_manager_ops socfpga_fpga_ops = { + .write_init = socfpga_fpga_ops_configure_init, + .write = socfpga_fpga_ops_configure_write, + .write_complete = socfpga_fpga_ops_configure_complete, + .state = socfpga_fpga_ops_state, +}; + +static int socfpga_fpga_probe(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct socfpga_fpga_priv *priv; + int ret; + + priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL); + if (!priv) + return -ENOMEM; + + /* ... do ioremaps, get interrupts, etc. and save + them in priv... */ + + return fpga_mgr_register(dev, "Altera SOCFPGA FPGA Manager", + &socfpga_fpga_ops, priv); +} + +static int socfpga_fpga_remove(struct platform_device *pdev) +{ + fpga_mgr_unregister(&pdev->dev); + + return 0; +} + + +The ops will implement whatever device specific register writes are needed to +do the programming sequence for this particular FPGA. These ops return 0 for +success or negative error codes otherwise. + +The programming sequence is: + 1. .write_init + 2. .write (may be called once or multiple times) + 3. .write_complete + +The .write_init function will prepare the FPGA to receive the image data. + +The .write function writes a buffer to the FPGA. The buffer may be contain the +whole FPGA image or may be a smaller chunk of an FPGA image. In the latter +case, this function is called multiple times for successive chunks. + +The .write_complete function is called after all the image has been written +to put the FPGA into operating mode. + +The ops include a .state function which will read the hardware FPGA manager and +return a code of type enum fpga_mgr_states. It doesn't result in a change in +hardware state. diff --git a/Documentation/input/multi-touch-protocol.txt b/Documentation/input/multi-touch-protocol.txt index b85d000faeb4..c51f1146f3bd 100644 --- a/Documentation/input/multi-touch-protocol.txt +++ b/Documentation/input/multi-touch-protocol.txt @@ -361,7 +361,7 @@ For win8 devices with both T and C coordinates, the position mapping is ABS_MT_POSITION_X := T_X ABS_MT_POSITION_Y := T_Y ABS_MT_TOOL_X := C_X - ABS_MT_TOOL_X := C_Y + ABS_MT_TOOL_Y := C_Y Unfortunately, there is not enough information to specify both the touching ellipse and the tool ellipse, so one has to resort to approximations. One diff --git a/Documentation/ioctl/ioctl-number.txt b/Documentation/ioctl/ioctl-number.txt index df1b25eb8382..3785b7e131f8 100644 --- a/Documentation/ioctl/ioctl-number.txt +++ b/Documentation/ioctl/ioctl-number.txt @@ -81,6 +81,9 @@ Code Seq#(hex) Include File Comments 0x22 all scsi/sg.h '#' 00-3F IEEE 1394 Subsystem Block for the entire subsystem '$' 00-0F linux/perf_counter.h, linux/perf_event.h +'%' 00-0F include/uapi/linux/stm.h + System Trace Module subsystem + <mailto:alexander.shishkin@linux.intel.com> '&' 00-07 drivers/firewire/nosy-user.h '1' 00-1F <linux/timepps.h> PPS kit from Ulrich Windl <ftp://ftp.de.kernel.org/pub/linux/daemons/ntp/PPS/> diff --git a/Documentation/mic/mic_overview.txt b/Documentation/mic/mic_overview.txt index 1a2f2c8ec59e..73f44fc3e715 100644 --- a/Documentation/mic/mic_overview.txt +++ b/Documentation/mic/mic_overview.txt @@ -28,6 +28,10 @@ The Symmetric Communication Interface (SCIF (pronounced as skiff)) is a low level communications API across PCIe currently implemented for MIC. More details are available at scif_overview.txt. +The Coprocessor State Management (COSM) driver on the host allows for +boot, shutdown and reset of Intel MIC devices. It communicates with a COSM +"client" driver on the MIC cards over SCIF to perform these functions. + Here is a block diagram of the various components described above. The virtio backends are situated on the host rather than the card given better single threaded performance for the host compared to MIC, the ability of @@ -51,19 +55,20 @@ the fact that the virtio block storage backend can only be on the host. | | | Virtio over PCIe IOCTLs | | | +--------------------------+ +-----------+ | | | +-----------+ -| MIC DMA | | +----------+ | +-----------+ | | MIC DMA | -| Driver | | | SCIF | | | SCIF | | | Driver | -+-----------+ | +----------+ | +-----------+ | +-----------+ - | | | | | | | -+---------------+ | +-----+-----+ | +-----+-----+ | +---------------+ -|MIC virtual Bus| | |SCIF HW Bus| | |SCIF HW BUS| | |MIC virtual Bus| -+---------------+ | +-----------+ | +-----+-----+ | +---------------+ - | | | | | | | - | +--------------+ | | | +---------------+ | - | |Intel MIC | | | | |Intel MIC | | - +---|Card Driver +----+ | | |Host Driver | | - +--------------+ | +----+---------------+-----+ - | | | +| MIC DMA | | +------+ | +------+ +------+ | | MIC DMA | +| Driver | | | SCIF | | | SCIF | | COSM | | | Driver | ++-----------+ | +------+ | +------+ +--+---+ | +-----------+ + | | | | | | | | ++---------------+ | +------+ | +--+---+ +--+---+ | +----------------+ +|MIC virtual Bus| | |SCIF | | |SCIF | | COSM | | |MIC virtual Bus | ++---------------+ | |HW Bus| | |HW Bus| | Bus | | +----------------+ + | | +------+ | +--+---+ +------+ | | + | | | | | | | | + | +-----------+---+ | | | +---------------+ | + | |Intel MIC | | | | |Intel MIC | | + +---|Card Driver | | | | |Host Driver | | + +------------+--------+ | +----+---------------+-----+ + | | | +-------------------------------------------------------------+ | | | PCIe Bus | diff --git a/Documentation/mic/mpssd/mpss b/Documentation/mic/mpssd/mpss index 582aad4811ae..09ea90931649 100755 --- a/Documentation/mic/mpssd/mpss +++ b/Documentation/mic/mpssd/mpss @@ -119,10 +119,10 @@ stop() # Wait for the cards to go offline for f in $sysfs/* do - while [ "`cat $f/state`" != "offline" ] + while [ "`cat $f/state`" != "ready" ] do sleep 1 - echo -e "Waiting for "`basename $f`" to go offline" + echo -e "Waiting for "`basename $f`" to become ready" done done diff --git a/Documentation/mic/mpssd/mpssd.c b/Documentation/mic/mpssd/mpssd.c index 3c5c379fc29d..aaeafa18d99b 100644 --- a/Documentation/mic/mpssd/mpssd.c +++ b/Documentation/mic/mpssd/mpssd.c @@ -43,7 +43,7 @@ #include <linux/mic_common.h> #include <tools/endian.h> -static void init_mic(struct mic_info *mic); +static void *init_mic(void *arg); static FILE *logfp; static struct mic_info mic_list; @@ -116,19 +116,18 @@ static struct { .num = htole16(MIC_VRING_ENTRIES), }, #if GSO_ENABLED - .host_features = htole32( + .host_features = htole32( 1 << VIRTIO_NET_F_CSUM | 1 << VIRTIO_NET_F_GSO | 1 << VIRTIO_NET_F_GUEST_TSO4 | 1 << VIRTIO_NET_F_GUEST_TSO6 | - 1 << VIRTIO_NET_F_GUEST_ECN | - 1 << VIRTIO_NET_F_GUEST_UFO), + 1 << VIRTIO_NET_F_GUEST_ECN), #else .host_features = 0, #endif }; -static const char *mic_config_dir = "/etc/sysconfig/mic"; +static const char *mic_config_dir = "/etc/mpss"; static const char *virtblk_backend = "VIRTBLK_BACKEND"; static struct { struct mic_device_desc dd; @@ -192,7 +191,7 @@ tap_configure(struct mic_info *mic, char *dev) return ret; } - snprintf(ipaddr, IFNAMSIZ, "172.31.%d.254/24", mic->id); + snprintf(ipaddr, IFNAMSIZ, "172.31.%d.254/24", mic->id + 1); pid = fork(); if (pid == 0) { @@ -255,8 +254,7 @@ static int tun_alloc(struct mic_info *mic, char *dev) return err; } #if GSO_ENABLED - offload = TUN_F_CSUM | TUN_F_TSO4 | TUN_F_TSO6 | - TUN_F_TSO_ECN | TUN_F_UFO; + offload = TUN_F_CSUM | TUN_F_TSO4 | TUN_F_TSO6 | TUN_F_TSO_ECN; err = ioctl(fd, TUNSETOFFLOAD, offload); if (err < 0) { @@ -332,7 +330,6 @@ static struct mic_device_desc *get_device_desc(struct mic_info *mic, int type) return d; } mpsslog("%s %s %d not found\n", mic->name, __func__, type); - assert(0); return NULL; } @@ -415,6 +412,13 @@ mic_virtio_copy(struct mic_info *mic, int fd, return ret; } +static inline unsigned _vring_size(unsigned int num, unsigned long align) +{ + return ((sizeof(struct vring_desc) * num + sizeof(__u16) * (3 + num) + + align - 1) & ~(align - 1)) + + sizeof(__u16) * 3 + sizeof(struct vring_used_elem) * num; +} + /* * This initialization routine requires at least one * vring i.e. vr0. vr1 is optional. @@ -426,8 +430,9 @@ init_vr(struct mic_info *mic, int fd, int type, int vr_size; char *va; - vr_size = PAGE_ALIGN(vring_size(MIC_VRING_ENTRIES, - MIC_VIRTIO_RING_ALIGN) + sizeof(struct _mic_vring_info)); + vr_size = PAGE_ALIGN(_vring_size(MIC_VRING_ENTRIES, + MIC_VIRTIO_RING_ALIGN) + + sizeof(struct _mic_vring_info)); va = mmap(NULL, MIC_DEVICE_PAGE_END + vr_size * num_vq, PROT_READ, MAP_SHARED, fd, 0); if (MAP_FAILED == va) { @@ -439,25 +444,25 @@ init_vr(struct mic_info *mic, int fd, int type, set_dp(mic, type, va); vr0->va = (struct mic_vring *)&va[MIC_DEVICE_PAGE_END]; vr0->info = vr0->va + - vring_size(MIC_VRING_ENTRIES, MIC_VIRTIO_RING_ALIGN); + _vring_size(MIC_VRING_ENTRIES, MIC_VIRTIO_RING_ALIGN); vring_init(&vr0->vr, MIC_VRING_ENTRIES, vr0->va, MIC_VIRTIO_RING_ALIGN); mpsslog("%s %s vr0 %p vr0->info %p vr_size 0x%x vring 0x%x ", __func__, mic->name, vr0->va, vr0->info, vr_size, - vring_size(MIC_VRING_ENTRIES, MIC_VIRTIO_RING_ALIGN)); + _vring_size(MIC_VRING_ENTRIES, MIC_VIRTIO_RING_ALIGN)); mpsslog("magic 0x%x expected 0x%x\n", le32toh(vr0->info->magic), MIC_MAGIC + type); assert(le32toh(vr0->info->magic) == MIC_MAGIC + type); if (vr1) { vr1->va = (struct mic_vring *) &va[MIC_DEVICE_PAGE_END + vr_size]; - vr1->info = vr1->va + vring_size(MIC_VRING_ENTRIES, + vr1->info = vr1->va + _vring_size(MIC_VRING_ENTRIES, MIC_VIRTIO_RING_ALIGN); vring_init(&vr1->vr, MIC_VRING_ENTRIES, vr1->va, MIC_VIRTIO_RING_ALIGN); mpsslog("%s %s vr1 %p vr1->info %p vr_size 0x%x vring 0x%x ", __func__, mic->name, vr1->va, vr1->info, vr_size, - vring_size(MIC_VRING_ENTRIES, MIC_VIRTIO_RING_ALIGN)); + _vring_size(MIC_VRING_ENTRIES, MIC_VIRTIO_RING_ALIGN)); mpsslog("magic 0x%x expected 0x%x\n", le32toh(vr1->info->magic), MIC_MAGIC + type + 1); assert(le32toh(vr1->info->magic) == MIC_MAGIC + type + 1); @@ -466,16 +471,21 @@ done: return va; } -static void +static int wait_for_card_driver(struct mic_info *mic, int fd, int type) { struct pollfd pollfd; int err; struct mic_device_desc *desc = get_device_desc(mic, type); + __u8 prev_status; + if (!desc) + return -ENODEV; + prev_status = desc->status; pollfd.fd = fd; mpsslog("%s %s Waiting .... desc-> type %d status 0x%x\n", mic->name, __func__, type, desc->status); + while (1) { pollfd.events = POLLIN; pollfd.revents = 0; @@ -487,8 +497,13 @@ wait_for_card_driver(struct mic_info *mic, int fd, int type) } if (pollfd.revents) { - mpsslog("%s %s Waiting... desc-> type %d status 0x%x\n", - mic->name, __func__, type, desc->status); + if (desc->status != prev_status) { + mpsslog("%s %s Waiting... desc-> type %d " + "status 0x%x\n", + mic->name, __func__, type, + desc->status); + prev_status = desc->status; + } if (desc->status & VIRTIO_CONFIG_S_DRIVER_OK) { mpsslog("%s %s poll.revents %d\n", mic->name, __func__, pollfd.revents); @@ -499,6 +514,7 @@ wait_for_card_driver(struct mic_info *mic, int fd, int type) } } } + return 0; } /* Spin till we have some descriptors */ @@ -575,9 +591,16 @@ virtio_net(void *arg) __func__, strerror(errno)); continue; } - if (!(desc->status & VIRTIO_CONFIG_S_DRIVER_OK)) - wait_for_card_driver(mic, mic->mic_net.virtio_net_fd, - VIRTIO_ID_NET); + if (!(desc->status & VIRTIO_CONFIG_S_DRIVER_OK)) { + err = wait_for_card_driver(mic, + mic->mic_net.virtio_net_fd, + VIRTIO_ID_NET); + if (err) { + mpsslog("%s %s %d Exiting...\n", + mic->name, __func__, __LINE__); + break; + } + } /* * Check if there is data to be read from TUN and write to * virtio net fd if there is. @@ -786,10 +809,16 @@ virtio_console(void *arg) strerror(errno)); continue; } - if (!(desc->status & VIRTIO_CONFIG_S_DRIVER_OK)) - wait_for_card_driver(mic, - mic->mic_console.virtio_console_fd, - VIRTIO_ID_CONSOLE); + if (!(desc->status & VIRTIO_CONFIG_S_DRIVER_OK)) { + err = wait_for_card_driver(mic, + mic->mic_console.virtio_console_fd, + VIRTIO_ID_CONSOLE); + if (err) { + mpsslog("%s %s %d Exiting...\n", + mic->name, __func__, __LINE__); + break; + } + } if (console_poll[MONITOR_FD].revents & POLLIN) { copy.iov = iov0; @@ -1048,8 +1077,9 @@ stop_virtblk(struct mic_info *mic) { int vr_size, ret; - vr_size = PAGE_ALIGN(vring_size(MIC_VRING_ENTRIES, - MIC_VIRTIO_RING_ALIGN) + sizeof(struct _mic_vring_info)); + vr_size = PAGE_ALIGN(_vring_size(MIC_VRING_ENTRIES, + MIC_VIRTIO_RING_ALIGN) + + sizeof(struct _mic_vring_info)); ret = munmap(mic->mic_virtblk.block_dp, MIC_DEVICE_PAGE_END + vr_size * virtblk_dev_page.dd.num_vq); if (ret < 0) @@ -1131,6 +1161,10 @@ write_status(int fd, __u8 *status) return ioctl(fd, MIC_VIRTIO_COPY_DESC, ©); } +#ifndef VIRTIO_BLK_T_GET_ID +#define VIRTIO_BLK_T_GET_ID 8 +#endif + static void * virtio_block(void *arg) { @@ -1297,12 +1331,7 @@ reset(struct mic_info *mic) mpsslog("%s: %s %d state %s\n", mic->name, __func__, __LINE__, state); - /* - * If the shutdown was initiated by OSPM, the state stays - * in "suspended" which is also a valid condition for reset. - */ - if ((!strcmp(state, "offline")) || - (!strcmp(state, "suspended"))) { + if (!strcmp(state, "ready")) { free(state); break; } @@ -1331,34 +1360,50 @@ get_mic_shutdown_status(struct mic_info *mic, char *shutdown_status) assert(0); }; -static int get_mic_state(struct mic_info *mic, char *state) +static int get_mic_state(struct mic_info *mic) { - if (!strcmp(state, "offline")) - return MIC_OFFLINE; - if (!strcmp(state, "online")) - return MIC_ONLINE; - if (!strcmp(state, "shutting_down")) - return MIC_SHUTTING_DOWN; - if (!strcmp(state, "reset_failed")) - return MIC_RESET_FAILED; - if (!strcmp(state, "suspending")) - return MIC_SUSPENDING; - if (!strcmp(state, "suspended")) - return MIC_SUSPENDED; - mpsslog("%s: BUG invalid state %s\n", mic->name, state); - /* Invalid state */ - assert(0); + char *state = NULL; + enum mic_states mic_state; + + while (!state) { + state = readsysfs(mic->name, "state"); + sleep(1); + } + mpsslog("%s: %s %d state %s\n", + mic->name, __func__, __LINE__, state); + + if (!strcmp(state, "ready")) { + mic_state = MIC_READY; + } else if (!strcmp(state, "booting")) { + mic_state = MIC_BOOTING; + } else if (!strcmp(state, "online")) { + mic_state = MIC_ONLINE; + } else if (!strcmp(state, "shutting_down")) { + mic_state = MIC_SHUTTING_DOWN; + } else if (!strcmp(state, "reset_failed")) { + mic_state = MIC_RESET_FAILED; + } else if (!strcmp(state, "resetting")) { + mic_state = MIC_RESETTING; + } else { + mpsslog("%s: BUG invalid state %s\n", mic->name, state); + assert(0); + } + + free(state); + return mic_state; }; static void mic_handle_shutdown(struct mic_info *mic) { #define SHUTDOWN_TIMEOUT 60 - int i = SHUTDOWN_TIMEOUT, ret, stat = 0; + int i = SHUTDOWN_TIMEOUT; char *shutdown_status; while (i) { shutdown_status = readsysfs(mic->name, "shutdown_status"); - if (!shutdown_status) + if (!shutdown_status) { + sleep(1); continue; + } mpsslog("%s: %s %d shutdown_status %s\n", mic->name, __func__, __LINE__, shutdown_status); switch (get_mic_shutdown_status(mic, shutdown_status)) { @@ -1377,94 +1422,110 @@ static void mic_handle_shutdown(struct mic_info *mic) i--; } reset: - ret = kill(mic->pid, SIGTERM); - mpsslog("%s: %s %d kill pid %d ret %d\n", - mic->name, __func__, __LINE__, - mic->pid, ret); - if (!ret) { - ret = waitpid(mic->pid, &stat, - WIFSIGNALED(stat)); - mpsslog("%s: %s %d waitpid ret %d pid %d\n", - mic->name, __func__, __LINE__, - ret, mic->pid); - } - if (ret == mic->pid) - reset(mic); + if (!i) + mpsslog("%s: %s %d timing out waiting for shutdown_status %s\n", + mic->name, __func__, __LINE__, shutdown_status); + reset(mic); } -static void * -mic_config(void *arg) +static int open_state_fd(struct mic_info *mic) { - struct mic_info *mic = (struct mic_info *)arg; - char *state = NULL; char pathname[PATH_MAX]; - int fd, ret; - struct pollfd ufds[1]; - char value[4096]; + int fd; snprintf(pathname, PATH_MAX - 1, "%s/%s/%s", MICSYSFSDIR, mic->name, "state"); fd = open(pathname, O_RDONLY); - if (fd < 0) { + if (fd < 0) mpsslog("%s: opening file %s failed %s\n", mic->name, pathname, strerror(errno)); - goto error; + return fd; +} + +static int block_till_state_change(int fd, struct mic_info *mic) +{ + struct pollfd ufds[1]; + char value[PAGE_SIZE]; + int ret; + + ufds[0].fd = fd; + ufds[0].events = POLLERR | POLLPRI; + ret = poll(ufds, 1, -1); + if (ret < 0) { + mpsslog("%s: %s %d poll failed %s\n", + mic->name, __func__, __LINE__, strerror(errno)); + return ret; + } + + ret = lseek(fd, 0, SEEK_SET); + if (ret < 0) { + mpsslog("%s: %s %d Failed to seek to 0: %s\n", + mic->name, __func__, __LINE__, strerror(errno)); + return ret; + } + + ret = read(fd, value, sizeof(value)); + if (ret < 0) { + mpsslog("%s: %s %d Failed to read sysfs entry: %s\n", + mic->name, __func__, __LINE__, strerror(errno)); + return ret; + } + + return 0; +} + +static void * +mic_config(void *arg) +{ + struct mic_info *mic = (struct mic_info *)arg; + int fd, ret, stat = 0; + + fd = open_state_fd(mic); + if (fd < 0) { + mpsslog("%s: %s %d open state fd failed %s\n", + mic->name, __func__, __LINE__, strerror(errno)); + goto exit; } do { - ret = lseek(fd, 0, SEEK_SET); + ret = block_till_state_change(fd, mic); if (ret < 0) { - mpsslog("%s: Failed to seek to file start '%s': %s\n", - mic->name, pathname, strerror(errno)); - goto close_error1; + mpsslog("%s: %s %d block_till_state_change error %s\n", + mic->name, __func__, __LINE__, strerror(errno)); + goto close_exit; } - ret = read(fd, value, sizeof(value)); - if (ret < 0) { - mpsslog("%s: Failed to read sysfs entry '%s': %s\n", - mic->name, pathname, strerror(errno)); - goto close_error1; - } -retry: - state = readsysfs(mic->name, "state"); - if (!state) - goto retry; - mpsslog("%s: %s %d state %s\n", - mic->name, __func__, __LINE__, state); - switch (get_mic_state(mic, state)) { + + switch (get_mic_state(mic)) { case MIC_SHUTTING_DOWN: mic_handle_shutdown(mic); - goto close_error; - case MIC_SUSPENDING: - mic->boot_on_resume = 1; - setsysfs(mic->name, "state", "suspend"); - mic_handle_shutdown(mic); - goto close_error; - case MIC_OFFLINE: + break; + case MIC_READY: + case MIC_RESET_FAILED: + ret = kill(mic->pid, SIGTERM); + mpsslog("%s: %s %d kill pid %d ret %d\n", + mic->name, __func__, __LINE__, + mic->pid, ret); + if (!ret) { + ret = waitpid(mic->pid, &stat, + WIFSIGNALED(stat)); + mpsslog("%s: %s %d waitpid ret %d pid %d\n", + mic->name, __func__, __LINE__, + ret, mic->pid); + } if (mic->boot_on_resume) { setsysfs(mic->name, "state", "boot"); mic->boot_on_resume = 0; } - break; + goto close_exit; default: break; } - free(state); - - ufds[0].fd = fd; - ufds[0].events = POLLERR | POLLPRI; - ret = poll(ufds, 1, -1); - if (ret < 0) { - mpsslog("%s: poll failed %s\n", - mic->name, strerror(errno)); - goto close_error1; - } } while (1); -close_error: - free(state); -close_error1: + +close_exit: close(fd); -error: +exit: init_mic(mic); pthread_exit(NULL); } @@ -1477,15 +1538,15 @@ set_cmdline(struct mic_info *mic) len = snprintf(buffer, PATH_MAX, "clocksource=tsc highres=off nohz=off "); - len += snprintf(buffer + len, PATH_MAX - len, + len += snprintf(buffer + len, PATH_MAX, "cpufreq_on;corec6_off;pc3_off;pc6_off "); - len += snprintf(buffer + len, PATH_MAX - len, + len += snprintf(buffer + len, PATH_MAX, "ifcfg=static;address,172.31.%d.1;netmask,255.255.255.0", - mic->id); + mic->id + 1); setsysfs(mic->name, "cmdline", buffer); mpsslog("%s: Command line: \"%s\"\n", mic->name, buffer); - snprintf(buffer, PATH_MAX, "172.31.%d.1", mic->id); + snprintf(buffer, PATH_MAX, "172.31.%d.1", mic->id + 1); mpsslog("%s: IPADDR: \"%s\"\n", mic->name, buffer); } @@ -1541,8 +1602,6 @@ set_log_buf_info(struct mic_info *mic) close(fd); } -static void init_mic(struct mic_info *mic); - static void change_virtblk_backend(int x, siginfo_t *siginfo, void *p) { @@ -1553,8 +1612,16 @@ change_virtblk_backend(int x, siginfo_t *siginfo, void *p) } static void -init_mic(struct mic_info *mic) +set_mic_boot_params(struct mic_info *mic) +{ + set_log_buf_info(mic); + set_cmdline(mic); +} + +static void * +init_mic(void *arg) { + struct mic_info *mic = (struct mic_info *)arg; struct sigaction ignore = { .sa_flags = 0, .sa_handler = SIG_IGN @@ -1564,7 +1631,7 @@ init_mic(struct mic_info *mic) .sa_sigaction = change_virtblk_backend, }; char buffer[PATH_MAX]; - int err; + int err, fd; /* * Currently, one virtio block device is supported for each MIC card @@ -1577,12 +1644,38 @@ init_mic(struct mic_info *mic) * the MIC daemon. */ sigaction(SIGUSR1, &ignore, NULL); +retry: + fd = open_state_fd(mic); + if (fd < 0) { + mpsslog("%s: %s %d open state fd failed %s\n", + mic->name, __func__, __LINE__, strerror(errno)); + sleep(2); + goto retry; + } + + if (mic->restart) { + snprintf(buffer, PATH_MAX, "boot"); + setsysfs(mic->name, "state", buffer); + mpsslog("%s restarting mic %d\n", + mic->name, mic->restart); + mic->restart = 0; + } + + while (1) { + while (block_till_state_change(fd, mic)) { + mpsslog("%s: %s %d block_till_state_change error %s\n", + mic->name, __func__, __LINE__, strerror(errno)); + sleep(2); + continue; + } + + if (get_mic_state(mic) == MIC_BOOTING) + break; + } mic->pid = fork(); switch (mic->pid) { case 0: - set_log_buf_info(mic); - set_cmdline(mic); add_virtio_device(mic, &virtcons_dev_page.dd); add_virtio_device(mic, &virtnet_dev_page.dd); err = pthread_create(&mic->mic_console.console_thread, NULL, @@ -1612,24 +1705,29 @@ init_mic(struct mic_info *mic) mic->name, mic->id, errno); break; default: - if (mic->restart) { - snprintf(buffer, PATH_MAX, "boot"); - setsysfs(mic->name, "state", buffer); - mpsslog("%s restarting mic %d\n", - mic->name, mic->restart); - mic->restart = 0; - } - pthread_create(&mic->config_thread, NULL, mic_config, mic); + err = pthread_create(&mic->config_thread, NULL, + mic_config, mic); + if (err) + mpsslog("%s mic_config pthread_create failed %s\n", + mic->name, strerror(err)); } + + return NULL; } static void start_daemon(void) { struct mic_info *mic; + int err; - for (mic = mic_list.next; mic != NULL; mic = mic->next) - init_mic(mic); + for (mic = mic_list.next; mic; mic = mic->next) { + set_mic_boot_params(mic); + err = pthread_create(&mic->init_thread, NULL, init_mic, mic); + if (err) + mpsslog("%s init_mic pthread_create failed %s\n", + mic->name, strerror(err)); + } while (1) sleep(60); diff --git a/Documentation/mic/mpssd/mpssd.h b/Documentation/mic/mpssd/mpssd.h index f5f18b15d9a0..8bd64944aacc 100644 --- a/Documentation/mic/mpssd/mpssd.h +++ b/Documentation/mic/mpssd/mpssd.h @@ -86,6 +86,7 @@ struct mic_info { int id; char *name; pthread_t config_thread; + pthread_t init_thread; pid_t pid; struct mic_console_info mic_console; struct mic_net_info mic_net; diff --git a/Documentation/power/pci.txt b/Documentation/power/pci.txt index 62328d76b55b..b0e911e0e8f5 100644 --- a/Documentation/power/pci.txt +++ b/Documentation/power/pci.txt @@ -979,20 +979,45 @@ every time right after the runtime_resume() callback has returned (alternatively, the runtime_suspend() callback will have to check if the device should really be suspended and return -EAGAIN if that is not the case). -The runtime PM of PCI devices is disabled by default. It is also blocked by -pci_pm_init() that runs the pm_runtime_forbid() helper function. If a PCI -driver implements the runtime PM callbacks and intends to use the runtime PM -framework provided by the PM core and the PCI subsystem, it should enable this -feature by executing the pm_runtime_enable() helper function. However, the -driver should not call the pm_runtime_allow() helper function unblocking -the runtime PM of the device. Instead, it should allow user space or some -platform-specific code to do that (user space can do it via sysfs), although -once it has called pm_runtime_enable(), it must be prepared to handle the +The runtime PM of PCI devices is enabled by default by the PCI core. PCI +device drivers do not need to enable it and should not attempt to do so. +However, it is blocked by pci_pm_init() that runs the pm_runtime_forbid() +helper function. In addition to that, the runtime PM usage counter of +each PCI device is incremented by local_pci_probe() before executing the +probe callback provided by the device's driver. + +If a PCI driver implements the runtime PM callbacks and intends to use the +runtime PM framework provided by the PM core and the PCI subsystem, it needs +to decrement the device's runtime PM usage counter in its probe callback +function. If it doesn't do that, the counter will always be different from +zero for the device and it will never be runtime-suspended. The simplest +way to do that is by calling pm_runtime_put_noidle(), but if the driver +wants to schedule an autosuspend right away, for example, it may call +pm_runtime_put_autosuspend() instead for this purpose. Generally, it +just needs to call a function that decrements the devices usage counter +from its probe routine to make runtime PM work for the device. + +It is important to remember that the driver's runtime_suspend() callback +may be executed right after the usage counter has been decremented, because +user space may already have cuased the pm_runtime_allow() helper function +unblocking the runtime PM of the device to run via sysfs, so the driver must +be prepared to cope with that. + +The driver itself should not call pm_runtime_allow(), though. Instead, it +should let user space or some platform-specific code do that (user space can +do it via sysfs as stated above), but it must be prepared to handle the runtime PM of the device correctly as soon as pm_runtime_allow() is called -(which may happen at any time). [It also is possible that user space causes -pm_runtime_allow() to be called via sysfs before the driver is loaded, so in -fact the driver has to be prepared to handle the runtime PM of the device as -soon as it calls pm_runtime_enable().] +(which may happen at any time, even before the driver is loaded). + +When the driver's remove callback runs, it has to balance the decrementation +of the device's runtime PM usage counter at the probe time. For this reason, +if it has decremented the counter in its probe callback, it must run +pm_runtime_get_noresume() in its remove callback. [Since the core carries +out a runtime resume of the device and bumps up the device's usage counter +before running the driver's remove callback, the runtime PM of the device +is effectively disabled for the duration of the remove execution and all +runtime PM helper functions incrementing the device's usage counter are +then effectively equivalent to pm_runtime_get_noresume().] The runtime PM framework works by processing requests to suspend or resume devices, or to check if they are idle (in which cases it is reasonable to diff --git a/Documentation/ptp/testptp.c b/Documentation/ptp/testptp.c index 2bc8abc57fa0..6c6247aaa7b9 100644 --- a/Documentation/ptp/testptp.c +++ b/Documentation/ptp/testptp.c @@ -18,6 +18,7 @@ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ #define _GNU_SOURCE +#define __SANE_USERSPACE_TYPES__ /* For PPC64, to get LL64 types */ #include <errno.h> #include <fcntl.h> #include <inttypes.h> diff --git a/Documentation/trace/intel_th.txt b/Documentation/trace/intel_th.txt new file mode 100644 index 000000000000..f7fc5ba5df8d --- /dev/null +++ b/Documentation/trace/intel_th.txt @@ -0,0 +1,99 @@ +Intel(R) Trace Hub (TH) +======================= + +Overview +-------- + +Intel(R) Trace Hub (TH) is a set of hardware blocks that produce, +switch and output trace data from multiple hardware and software +sources over several types of trace output ports encoded in System +Trace Protocol (MIPI STPv2) and is intended to perform full system +debugging. For more information on the hardware, see Intel(R) Trace +Hub developer's manual [1]. + +It consists of trace sources, trace destinations (outputs) and a +switch (Global Trace Hub, GTH). These devices are placed on a bus of +their own ("intel_th"), where they can be discovered and configured +via sysfs attributes. + +Currently, the following Intel TH subdevices (blocks) are supported: + - Software Trace Hub (STH), trace source, which is a System Trace + Module (STM) device, + - Memory Storage Unit (MSU), trace output, which allows storing + trace hub output in system memory, + - Parallel Trace Interface output (PTI), trace output to an external + debug host via a PTI port, + - Global Trace Hub (GTH), which is a switch and a central component + of Intel(R) Trace Hub architecture. + +Common attributes for output devices are described in +Documentation/ABI/testing/sysfs-bus-intel_th-output-devices, the most +notable of them is "active", which enables or disables trace output +into that particular output device. + +GTH allows directing different STP masters into different output ports +via its "masters" attribute group. More detailed GTH interface +description is at Documentation/ABI/testing/sysfs-bus-intel_th-devices-gth. + +STH registers an stm class device, through which it provides interface +to userspace and kernelspace software trace sources. See +Documentation/tracing/stm.txt for more information on that. + +MSU can be configured to collect trace data into a system memory +buffer, which can later on be read from its device nodes via read() or +mmap() interface. + +On the whole, Intel(R) Trace Hub does not require any special +userspace software to function; everything can be configured, started +and collected via sysfs attributes, and device nodes. + +[1] https://software.intel.com/sites/default/files/managed/d3/3c/intel-th-developer-manual.pdf + +Bus and Subdevices +------------------ + +For each Intel TH device in the system a bus of its own is +created and assigned an id number that reflects the order in which TH +devices were emumerated. All TH subdevices (devices on intel_th bus) +begin with this id: 0-gth, 0-msc0, 0-msc1, 0-pti, 0-sth, which is +followed by device's name and an optional index. + +Output devices also get a device node in /dev/intel_thN, where N is +the Intel TH device id. For example, MSU's memory buffers, when +allocated, are accessible via /dev/intel_th0/msc{0,1}. + +Quick example +------------- + +# figure out which GTH port is the first memory controller: + +$ cat /sys/bus/intel_th/devices/0-msc0/port +0 + +# looks like it's port 0, configure master 33 to send data to port 0: + +$ echo 0 > /sys/bus/intel_th/devices/0-gth/masters/33 + +# allocate a 2-windowed multiblock buffer on the first memory +# controller, each with 64 pages: + +$ echo multi > /sys/bus/intel_th/devices/0-msc0/mode +$ echo 64,64 > /sys/bus/intel_th/devices/0-msc0/nr_pages + +# enable wrapping for this controller, too: + +$ echo 1 > /sys/bus/intel_th/devices/0-msc0/wrap + +# and enable tracing into this port: + +$ echo 1 > /sys/bus/intel_th/devices/0-msc0/active + +# .. send data to master 33, see stm.txt for more details .. +# .. wait for traces to pile up .. +# .. and stop the trace: + +$ echo 0 > /sys/bus/intel_th/devices/0-msc0/active + +# and now you can collect the trace from the device node: + +$ cat /dev/intel_th0/msc0 > my_stp_trace diff --git a/Documentation/trace/stm.txt b/Documentation/trace/stm.txt new file mode 100644 index 000000000000..ea035f9dbfd7 --- /dev/null +++ b/Documentation/trace/stm.txt @@ -0,0 +1,80 @@ +System Trace Module +=================== + +System Trace Module (STM) is a device described in MIPI STP specs as +STP trace stream generator. STP (System Trace Protocol) is a trace +protocol multiplexing data from multiple trace sources, each one of +which is assigned a unique pair of master and channel. While some of +these masters and channels are statically allocated to certain +hardware trace sources, others are available to software. Software +trace sources are usually free to pick for themselves any +master/channel combination from this pool. + +On the receiving end of this STP stream (the decoder side), trace +sources can only be identified by master/channel combination, so in +order for the decoder to be able to make sense of the trace that +involves multiple trace sources, it needs to be able to map those +master/channel pairs to the trace sources that it understands. + +For instance, it is helpful to know that syslog messages come on +master 7 channel 15, while arbitrary user applications can use masters +48 to 63 and channels 0 to 127. + +To solve this mapping problem, stm class provides a policy management +mechanism via configfs, that allows defining rules that map string +identifiers to ranges of masters and channels. If these rules (policy) +are consistent with what decoder expects, it will be able to properly +process the trace data. + +This policy is a tree structure containing rules (policy_node) that +have a name (string identifier) and a range of masters and channels +associated with it, located in "stp-policy" subsystem directory in +configfs. The topmost directory's name (the policy) is formatted as +the STM device name to which this policy applies and and arbitrary +string identifier separated by a stop. From the examle above, a rule +may look like this: + +$ ls /config/stp-policy/dummy_stm.my-policy/user +channels masters +$ cat /config/stp-policy/dummy_stm.my-policy/user/masters +48 63 +$ cat /config/stp-policy/dummy_stm.my-policy/user/channels +0 127 + +which means that the master allocation pool for this rule consists of +masters 48 through 63 and channel allocation pool has channels 0 +through 127 in it. Now, any producer (trace source) identifying itself +with "user" identification string will be allocated a master and +channel from within these ranges. + +These rules can be nested, for example, one can define a rule "dummy" +under "user" directory from the example above and this new rule will +be used for trace sources with the id string of "user/dummy". + +Trace sources have to open the stm class device's node and write their +trace data into its file descriptor. In order to identify themselves +to the policy, they need to do a STP_POLICY_ID_SET ioctl on this file +descriptor providing their id string. Otherwise, they will be +automatically allocated a master/channel pair upon first write to this +file descriptor according to the "default" rule of the policy, if such +exists. + +Some STM devices may allow direct mapping of the channel mmio regions +to userspace for zero-copy writing. One mappable page (in terms of +mmu) will usually contain multiple channels' mmios, so the user will +need to allocate that many channels to themselves (via the +aforementioned ioctl() call) to be able to do this. That is, if your +stm device's channel mmio region is 64 bytes and hardware page size is +4096 bytes, after a successful STP_POLICY_ID_SET ioctl() call with +width==64, you should be able to mmap() one page on this file +descriptor and obtain direct access to an mmio region for 64 channels. + +For kernel-based trace sources, there is "stm_source" device +class. Devices of this class can be connected and disconnected to/from +stm devices at runtime via a sysfs attribute. + +Examples of STM devices are Intel(R) Trace Hub [1] and Coresight STM +[2]. + +[1] https://software.intel.com/sites/default/files/managed/d3/3c/intel-th-developer-manual.pdf +[2] http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ddi0444b/index.html diff --git a/Documentation/w1/masters/omap-hdq b/Documentation/w1/masters/omap-hdq index 884dc284b215..234522709a5f 100644 --- a/Documentation/w1/masters/omap-hdq +++ b/Documentation/w1/masters/omap-hdq @@ -44,3 +44,9 @@ e.g: insmod omap_hdq.ko W1_ID=2 inamod w1_bq27000.ko F_ID=2 +The driver also supports 1-wire mode. In this mode, there is no need to +pass slave ID as parameter. The driver will auto-detect slaves connected +to the bus using SEARCH_ROM procedure. 1-wire mode can be selected by +setting "ti,mode" property to "1w" in DT (see +Documentation/devicetree/bindings/w1/omap-hdq.txt for more details). +By default driver is in HDQ mode. diff --git a/MAINTAINERS b/MAINTAINERS index 9f6685f6c5a9..fe5bbfebc74b 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4003,7 +4003,7 @@ S: Maintained F: sound/usb/misc/ua101.c EXTENSIBLE FIRMWARE INTERFACE (EFI) -M: Matt Fleming <matt.fleming@intel.com> +M: Matt Fleming <matt@codeblueprint.co.uk> L: linux-efi@vger.kernel.org T: git git://git.kernel.org/pub/scm/linux/kernel/git/mfleming/efi.git S: Maintained @@ -4018,7 +4018,7 @@ F: include/linux/efi*.h EFI VARIABLE FILESYSTEM M: Matthew Garrett <matthew.garrett@nebula.com> M: Jeremy Kerr <jk@ozlabs.org> -M: Matt Fleming <matt.fleming@intel.com> +M: Matt Fleming <matt@codeblueprint.co.uk> T: git git://git.kernel.org/pub/scm/linux/kernel/git/mfleming/efi.git L: linux-efi@vger.kernel.org S: Maintained @@ -4321,6 +4321,13 @@ F: include/linux/fmc*.h F: include/linux/ipmi-fru.h K: fmc_d.*register +FPGA MANAGER FRAMEWORK +M: Alan Tull <atull@opensource.altera.com> +S: Maintained +F: drivers/fpga/ +F: include/linux/fpga/fpga-mgr.h +W: http://www.rocketboards.org + FPU EMULATOR M: Bill Metzenthen <billm@melbpc.org.au> W: http://floatingpoint.sourceforge.net/emulator/index.html @@ -5515,6 +5522,12 @@ F: Documentation/networking/README.ipw2100 F: Documentation/networking/README.ipw2200 F: drivers/net/wireless/ipw2x00/ +INTEL(R) TRACE HUB +M: Alexander Shishkin <alexander.shishkin@linux.intel.com> +S: Supported +F: Documentation/trace/intel_th.txt +F: drivers/hwtracing/intel_th/ + INTEL(R) TRUSTED EXECUTION TECHNOLOGY (TXT) M: Richard L Maliszewski <richard.l.maliszewski@intel.com> M: Gang Wei <gang.wei@intel.com> @@ -5957,7 +5970,7 @@ F: virt/kvm/ KERNEL VIRTUAL MACHINE (KVM) FOR AMD-V M: Joerg Roedel <joro@8bytes.org> L: kvm@vger.kernel.org -W: http://kvm.qumranet.com +W: http://www.linux-kvm.org/ S: Maintained F: arch/x86/include/asm/svm.h F: arch/x86/kvm/svm.c @@ -5965,7 +5978,7 @@ F: arch/x86/kvm/svm.c KERNEL VIRTUAL MACHINE (KVM) FOR POWERPC M: Alexander Graf <agraf@suse.com> L: kvm-ppc@vger.kernel.org -W: http://kvm.qumranet.com +W: http://www.linux-kvm.org/ T: git git://github.com/agraf/linux-2.6.git S: Supported F: arch/powerpc/include/asm/kvm* @@ -9109,6 +9122,14 @@ S: Maintained F: include/linux/mmc/dw_mmc.h F: drivers/mmc/host/dw_mmc* +SYSTEM TRACE MODULE CLASS +M: Alexander Shishkin <alexander.shishkin@linux.intel.com> +S: Maintained +F: Documentation/trace/stm.txt +F: drivers/hwtracing/stm/ +F: include/linux/stm.h +F: include/uapi/linux/stm.h + THUNDERBOLT DRIVER M: Andreas Noever <andreas.noever@gmail.com> S: Maintained @@ -9914,7 +9935,6 @@ S: Maintained F: drivers/staging/lustre STAGING - NVIDIA COMPLIANT EMBEDDED CONTROLLER INTERFACE (nvec) -M: Julian Andres Klode <jak@jak-linux.org> M: Marc Dietrich <marvin24@gmx.de> L: ac100@lists.launchpad.net (moderated for non-subscribers) L: linux-tegra@vger.kernel.org @@ -11378,15 +11398,6 @@ W: http://oops.ghostprotocols.net:81/blog S: Maintained F: drivers/net/wireless/wl3501* -WM97XX TOUCHSCREEN DRIVERS -M: Mark Brown <broonie@kernel.org> -M: Liam Girdwood <lrg@slimlogic.co.uk> -L: linux-input@vger.kernel.org -W: https://github.com/CirrusLogic/linux-drivers/wiki -S: Supported -F: drivers/input/touchscreen/*wm97* -F: include/linux/wm97xx.h - WOLFSON MICROELECTRONICS DRIVERS L: patches@opensource.wolfsonmicro.com T: git https://github.com/CirrusLogic/linux-drivers.git @@ -1,8 +1,8 @@ VERSION = 4 PATCHLEVEL = 3 SUBLEVEL = 0 -EXTRAVERSION = -rc3 -NAME = Hurr durr I'ma sheep +EXTRAVERSION = -rc5 +NAME = Blurry Fish Butt # *DOCUMENTATION* # To see a list of typical targets execute "make help" diff --git a/arch/alpha/include/asm/word-at-a-time.h b/arch/alpha/include/asm/word-at-a-time.h index 6b340d0f1521..902e6ab00a06 100644 --- a/arch/alpha/include/asm/word-at-a-time.h +++ b/arch/alpha/include/asm/word-at-a-time.h @@ -52,4 +52,6 @@ static inline unsigned long find_zero(unsigned long bits) #endif } +#define zero_bytemask(mask) ((2ul << (find_zero(mask) * 8)) - 1) + #endif /* _ASM_WORD_AT_A_TIME_H */ diff --git a/arch/arc/include/asm/Kbuild b/arch/arc/include/asm/Kbuild index 7611b10a2d23..0b10ef2a4372 100644 --- a/arch/arc/include/asm/Kbuild +++ b/arch/arc/include/asm/Kbuild @@ -48,4 +48,5 @@ generic-y += types.h generic-y += ucontext.h generic-y += user.h generic-y += vga.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/arm/boot/dts/Makefile b/arch/arm/boot/dts/Makefile index 233159d2eaab..bb8fa023d574 100644 --- a/arch/arm/boot/dts/Makefile +++ b/arch/arm/boot/dts/Makefile @@ -578,7 +578,7 @@ dtb-$(CONFIG_MACH_SUN4I) += \ sun4i-a10-hackberry.dtb \ sun4i-a10-hyundai-a7hd.dtb \ sun4i-a10-inet97fv2.dtb \ - sun4i-a10-itead-iteaduino-plus.dts \ + sun4i-a10-itead-iteaduino-plus.dtb \ sun4i-a10-jesurun-q5.dtb \ sun4i-a10-marsboard.dtb \ sun4i-a10-mini-xplus.dtb \ diff --git a/arch/arm/boot/dts/exynos4412.dtsi b/arch/arm/boot/dts/exynos4412.dtsi index ca0e3c15977f..294cfe40388d 100644 --- a/arch/arm/boot/dts/exynos4412.dtsi +++ b/arch/arm/boot/dts/exynos4412.dtsi @@ -98,6 +98,7 @@ opp-hz = /bits/ 64 <800000000>; opp-microvolt = <1000000>; clock-latency-ns = <200000>; + opp-suspend; }; opp07 { opp-hz = /bits/ 64 <900000000>; diff --git a/arch/arm/boot/dts/exynos5250-smdk5250.dts b/arch/arm/boot/dts/exynos5250-smdk5250.dts index 15aea760c1da..c625e71217aa 100644 --- a/arch/arm/boot/dts/exynos5250-smdk5250.dts +++ b/arch/arm/boot/dts/exynos5250-smdk5250.dts @@ -197,6 +197,7 @@ regulator-name = "P1.8V_LDO_OUT10"; regulator-min-microvolt = <1800000>; regulator-max-microvolt = <1800000>; + regulator-always-on; }; ldo11_reg: LDO11 { diff --git a/arch/arm/boot/dts/exynos5420.dtsi b/arch/arm/boot/dts/exynos5420.dtsi index df9aee92ecf4..1b3d6c769a3c 100644 --- a/arch/arm/boot/dts/exynos5420.dtsi +++ b/arch/arm/boot/dts/exynos5420.dtsi @@ -1117,7 +1117,7 @@ interrupt-parent = <&combiner>; interrupts = <3 0>; clock-names = "sysmmu", "master"; - clocks = <&clock CLK_SMMU_FIMD1M0>, <&clock CLK_FIMD1>; + clocks = <&clock CLK_SMMU_FIMD1M1>, <&clock CLK_FIMD1>; power-domains = <&disp_pd>; #iommu-cells = <0>; }; diff --git a/arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi b/arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi index 79ffdfe712aa..3b43e57845ae 100644 --- a/arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi +++ b/arch/arm/boot/dts/exynos5422-odroidxu3-common.dtsi @@ -472,7 +472,6 @@ */ pinctrl-0 = <&pwm0_out &pwm1_out &pwm2_out &pwm3_out>; pinctrl-names = "default"; - samsung,pwm-outputs = <0>; status = "okay"; }; diff --git a/arch/arm/boot/dts/imx53-qsrb.dts b/arch/arm/boot/dts/imx53-qsrb.dts index 66e47de5e826..96d7eede412e 100644 --- a/arch/arm/boot/dts/imx53-qsrb.dts +++ b/arch/arm/boot/dts/imx53-qsrb.dts @@ -36,7 +36,7 @@ pinctrl-0 = <&pinctrl_pmic>; reg = <0x08>; interrupt-parent = <&gpio5>; - interrupts = <23 0x8>; + interrupts = <23 IRQ_TYPE_LEVEL_HIGH>; regulators { sw1_reg: sw1a { regulator-name = "SW1"; diff --git a/arch/arm/boot/dts/imx53.dtsi b/arch/arm/boot/dts/imx53.dtsi index c3e3ca9362fb..cd170376eaca 100644 --- a/arch/arm/boot/dts/imx53.dtsi +++ b/arch/arm/boot/dts/imx53.dtsi @@ -15,6 +15,7 @@ #include <dt-bindings/clock/imx5-clock.h> #include <dt-bindings/gpio/gpio.h> #include <dt-bindings/input/input.h> +#include <dt-bindings/interrupt-controller/irq.h> / { aliases { diff --git a/arch/arm/boot/dts/imx6qdl-rex.dtsi b/arch/arm/boot/dts/imx6qdl-rex.dtsi index 3373fd958e95..a50356243888 100644 --- a/arch/arm/boot/dts/imx6qdl-rex.dtsi +++ b/arch/arm/boot/dts/imx6qdl-rex.dtsi @@ -35,7 +35,6 @@ compatible = "regulator-fixed"; reg = <1>; pinctrl-names = "default"; - pinctrl-0 = <&pinctrl_usbh1>; regulator-name = "usbh1_vbus"; regulator-min-microvolt = <5000000>; regulator-max-microvolt = <5000000>; @@ -47,7 +46,6 @@ compatible = "regulator-fixed"; reg = <2>; pinctrl-names = "default"; - pinctrl-0 = <&pinctrl_usbotg>; regulator-name = "usb_otg_vbus"; regulator-min-microvolt = <5000000>; regulator-max-microvolt = <5000000>; diff --git a/arch/arm/boot/dts/r8a7790.dtsi b/arch/arm/boot/dts/r8a7790.dtsi index a0b2a79cbfbd..4624d0f2a754 100644 --- a/arch/arm/boot/dts/r8a7790.dtsi +++ b/arch/arm/boot/dts/r8a7790.dtsi @@ -1627,6 +1627,7 @@ "mix.0", "mix.1", "dvc.0", "dvc.1", "clk_a", "clk_b", "clk_c", "clk_i"; + power-domains = <&cpg_clocks>; status = "disabled"; diff --git a/arch/arm/boot/dts/r8a7791.dtsi b/arch/arm/boot/dts/r8a7791.dtsi index 831525dd39a6..1666c8a6b143 100644 --- a/arch/arm/boot/dts/r8a7791.dtsi +++ b/arch/arm/boot/dts/r8a7791.dtsi @@ -1677,6 +1677,7 @@ "mix.0", "mix.1", "dvc.0", "dvc.1", "clk_a", "clk_b", "clk_c", "clk_i"; + power-domains = <&cpg_clocks>; status = "disabled"; diff --git a/arch/arm/boot/dts/sun7i-a20.dtsi b/arch/arm/boot/dts/sun7i-a20.dtsi index 2bebaa286f9a..391230c3dc93 100644 --- a/arch/arm/boot/dts/sun7i-a20.dtsi +++ b/arch/arm/boot/dts/sun7i-a20.dtsi @@ -107,7 +107,7 @@ 720000 1200000 528000 1100000 312000 1000000 - 144000 900000 + 144000 1000000 >; #cooling-cells = <2>; cooling-min-level = <0>; diff --git a/arch/arm/mach-exynos/mcpm-exynos.c b/arch/arm/mach-exynos/mcpm-exynos.c index 9bdf54795f05..56978199c479 100644 --- a/arch/arm/mach-exynos/mcpm-exynos.c +++ b/arch/arm/mach-exynos/mcpm-exynos.c @@ -20,6 +20,7 @@ #include <asm/cputype.h> #include <asm/cp15.h> #include <asm/mcpm.h> +#include <asm/smp_plat.h> #include "regs-pmu.h" #include "common.h" @@ -70,7 +71,31 @@ static int exynos_cpu_powerup(unsigned int cpu, unsigned int cluster) cluster >= EXYNOS5420_NR_CLUSTERS) return -EINVAL; - exynos_cpu_power_up(cpunr); + if (!exynos_cpu_power_state(cpunr)) { + exynos_cpu_power_up(cpunr); + + /* + * This assumes the cluster number of the big cores(Cortex A15) + * is 0 and the Little cores(Cortex A7) is 1. + * When the system was booted from the Little core, + * they should be reset during power up cpu. + */ + if (cluster && + cluster == MPIDR_AFFINITY_LEVEL(cpu_logical_map(0), 1)) { + /* + * Before we reset the Little cores, we should wait + * the SPARE2 register is set to 1 because the init + * codes of the iROM will set the register after + * initialization. + */ + while (!pmu_raw_readl(S5P_PMU_SPARE2)) + udelay(10); + + pmu_raw_writel(EXYNOS5420_KFC_CORE_RESET(cpu), + EXYNOS_SWRESET); + } + } + return 0; } diff --git a/arch/arm/mach-exynos/regs-pmu.h b/arch/arm/mach-exynos/regs-pmu.h index b7614333d296..fba9068ed260 100644 --- a/arch/arm/mach-exynos/regs-pmu.h +++ b/arch/arm/mach-exynos/regs-pmu.h @@ -513,6 +513,12 @@ static inline unsigned int exynos_pmu_cpunr(unsigned int mpidr) #define SPREAD_ENABLE 0xF #define SPREAD_USE_STANDWFI 0xF +#define EXYNOS5420_KFC_CORE_RESET0 BIT(8) +#define EXYNOS5420_KFC_ETM_RESET0 BIT(20) + +#define EXYNOS5420_KFC_CORE_RESET(_nr) \ + ((EXYNOS5420_KFC_CORE_RESET0 | EXYNOS5420_KFC_ETM_RESET0) << (_nr)) + #define EXYNOS5420_BB_CON1 0x0784 #define EXYNOS5420_BB_SEL_EN BIT(31) #define EXYNOS5420_BB_PMOS_EN BIT(7) diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h index b0329be95cb1..26b066690593 100644 --- a/arch/arm64/include/asm/pgtable.h +++ b/arch/arm64/include/asm/pgtable.h @@ -79,7 +79,7 @@ extern void __pgd_error(const char *file, int line, unsigned long val); #define PAGE_S2 __pgprot(PROT_DEFAULT | PTE_S2_MEMATTR(MT_S2_NORMAL) | PTE_S2_RDONLY) #define PAGE_S2_DEVICE __pgprot(PROT_DEFAULT | PTE_S2_MEMATTR(MT_S2_DEVICE_nGnRE) | PTE_S2_RDONLY | PTE_UXN) -#define PAGE_NONE __pgprot(((_PAGE_DEFAULT) & ~PTE_TYPE_MASK) | PTE_PROT_NONE | PTE_PXN | PTE_UXN) +#define PAGE_NONE __pgprot(((_PAGE_DEFAULT) & ~PTE_VALID) | PTE_PROT_NONE | PTE_PXN | PTE_UXN) #define PAGE_SHARED __pgprot(_PAGE_DEFAULT | PTE_USER | PTE_NG | PTE_PXN | PTE_UXN | PTE_WRITE) #define PAGE_SHARED_EXEC __pgprot(_PAGE_DEFAULT | PTE_USER | PTE_NG | PTE_PXN | PTE_WRITE) #define PAGE_COPY __pgprot(_PAGE_DEFAULT | PTE_USER | PTE_NG | PTE_PXN | PTE_UXN) @@ -496,7 +496,7 @@ static inline pud_t *pud_offset(pgd_t *pgd, unsigned long addr) static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) { const pteval_t mask = PTE_USER | PTE_PXN | PTE_UXN | PTE_RDONLY | - PTE_PROT_NONE | PTE_WRITE | PTE_TYPE_MASK; + PTE_PROT_NONE | PTE_VALID | PTE_WRITE; /* preserve the hardware dirty information */ if (pte_hw_dirty(pte)) pte = pte_mkdirty(pte); diff --git a/arch/arm64/kernel/debug-monitors.c b/arch/arm64/kernel/debug-monitors.c index cebf78661a55..253021ef2769 100644 --- a/arch/arm64/kernel/debug-monitors.c +++ b/arch/arm64/kernel/debug-monitors.c @@ -201,7 +201,7 @@ void unregister_step_hook(struct step_hook *hook) } /* - * Call registered single step handers + * Call registered single step handlers * There is no Syndrome info to check for determining the handler. * So we call all the registered handlers, until the right handler is * found which returns zero. @@ -271,20 +271,21 @@ static int single_step_handler(unsigned long addr, unsigned int esr, * Use reader/writer locks instead of plain spinlock. */ static LIST_HEAD(break_hook); -static DEFINE_RWLOCK(break_hook_lock); +static DEFINE_SPINLOCK(break_hook_lock); void register_break_hook(struct break_hook *hook) { - write_lock(&break_hook_lock); - list_add(&hook->node, &break_hook); - write_unlock(&break_hook_lock); + spin_lock(&break_hook_lock); + list_add_rcu(&hook->node, &break_hook); + spin_unlock(&break_hook_lock); } void unregister_break_hook(struct break_hook *hook) { - write_lock(&break_hook_lock); - list_del(&hook->node); - write_unlock(&break_hook_lock); + spin_lock(&break_hook_lock); + list_del_rcu(&hook->node); + spin_unlock(&break_hook_lock); + synchronize_rcu(); } static int call_break_hook(struct pt_regs *regs, unsigned int esr) @@ -292,11 +293,11 @@ static int call_break_hook(struct pt_regs *regs, unsigned int esr) struct break_hook *hook; int (*fn)(struct pt_regs *regs, unsigned int esr) = NULL; - read_lock(&break_hook_lock); - list_for_each_entry(hook, &break_hook, node) + rcu_read_lock(); + list_for_each_entry_rcu(hook, &break_hook, node) if ((esr & hook->esr_mask) == hook->esr_val) fn = hook->fn; - read_unlock(&break_hook_lock); + rcu_read_unlock(); return fn ? fn(regs, esr) : DBG_HOOK_ERROR; } diff --git a/arch/arm64/kernel/efi.c b/arch/arm64/kernel/efi.c index e8ca6eaedd02..13671a9cf016 100644 --- a/arch/arm64/kernel/efi.c +++ b/arch/arm64/kernel/efi.c @@ -258,7 +258,8 @@ static bool __init efi_virtmap_init(void) */ if (!is_normal_ram(md)) prot = __pgprot(PROT_DEVICE_nGnRE); - else if (md->type == EFI_RUNTIME_SERVICES_CODE) + else if (md->type == EFI_RUNTIME_SERVICES_CODE || + !PAGE_ALIGNED(md->phys_addr)) prot = PAGE_KERNEL_EXEC; else prot = PAGE_KERNEL; diff --git a/arch/arm64/kernel/entry-ftrace.S b/arch/arm64/kernel/entry-ftrace.S index 08cafc518b9a..0f03a8fe2314 100644 --- a/arch/arm64/kernel/entry-ftrace.S +++ b/arch/arm64/kernel/entry-ftrace.S @@ -178,6 +178,24 @@ ENTRY(ftrace_stub) ENDPROC(ftrace_stub) #ifdef CONFIG_FUNCTION_GRAPH_TRACER + /* save return value regs*/ + .macro save_return_regs + sub sp, sp, #64 + stp x0, x1, [sp] + stp x2, x3, [sp, #16] + stp x4, x5, [sp, #32] + stp x6, x7, [sp, #48] + .endm + + /* restore return value regs*/ + .macro restore_return_regs + ldp x0, x1, [sp] + ldp x2, x3, [sp, #16] + ldp x4, x5, [sp, #32] + ldp x6, x7, [sp, #48] + add sp, sp, #64 + .endm + /* * void ftrace_graph_caller(void) * @@ -204,11 +222,11 @@ ENDPROC(ftrace_graph_caller) * only when CONFIG_HAVE_FUNCTION_GRAPH_FP_TEST is enabled. */ ENTRY(return_to_handler) - str x0, [sp, #-16]! + save_return_regs mov x0, x29 // parent's fp bl ftrace_return_to_handler// addr = ftrace_return_to_hander(fp); mov x30, x0 // restore the original return address - ldr x0, [sp], #16 + restore_return_regs ret END(return_to_handler) #endif /* CONFIG_FUNCTION_GRAPH_TRACER */ diff --git a/arch/arm64/kernel/insn.c b/arch/arm64/kernel/insn.c index f341866aa810..c08b9ad6f429 100644 --- a/arch/arm64/kernel/insn.c +++ b/arch/arm64/kernel/insn.c @@ -85,7 +85,7 @@ bool aarch64_insn_is_branch_imm(u32 insn) aarch64_insn_is_bcond(insn)); } -static DEFINE_SPINLOCK(patch_lock); +static DEFINE_RAW_SPINLOCK(patch_lock); static void __kprobes *patch_map(void *addr, int fixmap) { @@ -131,13 +131,13 @@ static int __kprobes __aarch64_insn_write(void *addr, u32 insn) unsigned long flags = 0; int ret; - spin_lock_irqsave(&patch_lock, flags); + raw_spin_lock_irqsave(&patch_lock, flags); waddr = patch_map(addr, FIX_TEXT_POKE0); ret = probe_kernel_write(waddr, &insn, AARCH64_INSN_SIZE); patch_unmap(FIX_TEXT_POKE0); - spin_unlock_irqrestore(&patch_lock, flags); + raw_spin_unlock_irqrestore(&patch_lock, flags); return ret; } diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c index 6bab21f84a9f..232247945b1c 100644 --- a/arch/arm64/kernel/setup.c +++ b/arch/arm64/kernel/setup.c @@ -364,6 +364,8 @@ static void __init relocate_initrd(void) to_free = ram_end - orig_start; size = orig_end - orig_start; + if (!size) + return; /* initrd needs to be relocated completely inside linear mapping */ new_start = memblock_find_in_range(0, PFN_PHYS(max_pfn), diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c index aba9ead1384c..9fadf6d7039b 100644 --- a/arch/arm64/mm/fault.c +++ b/arch/arm64/mm/fault.c @@ -287,6 +287,7 @@ retry: * starvation. */ mm_flags &= ~FAULT_FLAG_ALLOW_RETRY; + mm_flags |= FAULT_FLAG_TRIED; goto retry; } } diff --git a/arch/avr32/include/asm/Kbuild b/arch/avr32/include/asm/Kbuild index f61f2dd67464..241b9b9729d8 100644 --- a/arch/avr32/include/asm/Kbuild +++ b/arch/avr32/include/asm/Kbuild @@ -20,4 +20,5 @@ generic-y += sections.h generic-y += topology.h generic-y += trace_clock.h generic-y += vga.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/blackfin/include/asm/Kbuild b/arch/blackfin/include/asm/Kbuild index 61cd1e786a14..91d49c0a3118 100644 --- a/arch/blackfin/include/asm/Kbuild +++ b/arch/blackfin/include/asm/Kbuild @@ -46,4 +46,5 @@ generic-y += types.h generic-y += ucontext.h generic-y += unaligned.h generic-y += user.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/c6x/include/asm/Kbuild b/arch/c6x/include/asm/Kbuild index f17c4dc6050c..945544ec603e 100644 --- a/arch/c6x/include/asm/Kbuild +++ b/arch/c6x/include/asm/Kbuild @@ -59,4 +59,5 @@ generic-y += types.h generic-y += ucontext.h generic-y += user.h generic-y += vga.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/cris/include/asm/Kbuild b/arch/cris/include/asm/Kbuild index b7f68192d15b..1778805f6380 100644 --- a/arch/cris/include/asm/Kbuild +++ b/arch/cris/include/asm/Kbuild @@ -43,4 +43,5 @@ generic-y += topology.h generic-y += trace_clock.h generic-y += types.h generic-y += vga.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/frv/include/asm/Kbuild b/arch/frv/include/asm/Kbuild index 8e47b832cc76..1fa084cf1a43 100644 --- a/arch/frv/include/asm/Kbuild +++ b/arch/frv/include/asm/Kbuild @@ -7,3 +7,4 @@ generic-y += mcs_spinlock.h generic-y += mm-arch-hooks.h generic-y += preempt.h generic-y += trace_clock.h +generic-y += word-at-a-time.h diff --git a/arch/h8300/include/asm/Kbuild b/arch/h8300/include/asm/Kbuild index 70e6ae1e7006..373cb23301e3 100644 --- a/arch/h8300/include/asm/Kbuild +++ b/arch/h8300/include/asm/Kbuild @@ -73,4 +73,5 @@ generic-y += uaccess.h generic-y += ucontext.h generic-y += unaligned.h generic-y += vga.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/hexagon/include/asm/Kbuild b/arch/hexagon/include/asm/Kbuild index daee37bd0999..db8ddabc6bd2 100644 --- a/arch/hexagon/include/asm/Kbuild +++ b/arch/hexagon/include/asm/Kbuild @@ -58,4 +58,5 @@ generic-y += types.h generic-y += ucontext.h generic-y += unaligned.h generic-y += vga.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/ia64/include/asm/Kbuild b/arch/ia64/include/asm/Kbuild index 9de3ba12f6b9..502a91d8dbbd 100644 --- a/arch/ia64/include/asm/Kbuild +++ b/arch/ia64/include/asm/Kbuild @@ -8,3 +8,4 @@ generic-y += mm-arch-hooks.h generic-y += preempt.h generic-y += trace_clock.h generic-y += vtime.h +generic-y += word-at-a-time.h diff --git a/arch/m32r/include/asm/Kbuild b/arch/m32r/include/asm/Kbuild index e0eb704ca1fa..fd104bd221ce 100644 --- a/arch/m32r/include/asm/Kbuild +++ b/arch/m32r/include/asm/Kbuild @@ -9,3 +9,4 @@ generic-y += module.h generic-y += preempt.h generic-y += sections.h generic-y += trace_clock.h +generic-y += word-at-a-time.h diff --git a/arch/m68k/configs/amiga_defconfig b/arch/m68k/configs/amiga_defconfig index 0b6b40d37b95..5b4ec541ba7c 100644 --- a/arch/m68k/configs/amiga_defconfig +++ b/arch/m68k/configs/amiga_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -57,7 +58,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -67,10 +67,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -179,6 +181,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -206,6 +209,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -271,6 +275,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -370,6 +375,7 @@ CONFIG_ZORRO8390=y # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_SMSC is not set # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -537,6 +543,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/apollo_defconfig b/arch/m68k/configs/apollo_defconfig index eeb3a8991fc4..6e5198e2c124 100644 --- a/arch/m68k/configs/apollo_defconfig +++ b/arch/m68k/configs/apollo_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -55,7 +56,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -65,10 +65,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -177,6 +179,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -204,6 +207,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -269,6 +273,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -344,6 +349,7 @@ CONFIG_VETH=m # CONFIG_NET_VENDOR_SAMSUNG is not set # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -495,6 +501,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/atari_defconfig b/arch/m68k/configs/atari_defconfig index 3a7006654ce9..f75600b0ca23 100644 --- a/arch/m68k/configs/atari_defconfig +++ b/arch/m68k/configs/atari_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -55,7 +56,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -65,10 +65,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -177,6 +179,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -204,6 +207,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -269,6 +273,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -355,6 +360,7 @@ CONFIG_NE2000=y # CONFIG_NET_VENDOR_SEEQ is not set CONFIG_SMC91X=y # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -517,6 +523,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/bvme6000_defconfig b/arch/m68k/configs/bvme6000_defconfig index 0586b323a673..a42d91c389a6 100644 --- a/arch/m68k/configs/bvme6000_defconfig +++ b/arch/m68k/configs/bvme6000_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -53,7 +54,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -63,10 +63,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -175,6 +177,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -202,6 +205,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -267,6 +271,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -343,6 +348,7 @@ CONFIG_BVME6000_NET=y # CONFIG_NET_VENDOR_SAMSUNG is not set # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -488,6 +494,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/hp300_defconfig b/arch/m68k/configs/hp300_defconfig index ad1dbce07aa4..77f4a11083e9 100644 --- a/arch/m68k/configs/hp300_defconfig +++ b/arch/m68k/configs/hp300_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -55,7 +56,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -65,10 +65,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -177,6 +179,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -204,6 +207,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -269,6 +273,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -345,6 +350,7 @@ CONFIG_HPLANCE=y # CONFIG_NET_VENDOR_SAMSUNG is not set # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -497,6 +503,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/mac_defconfig b/arch/m68k/configs/mac_defconfig index b44acacaecf4..5a329f77329b 100644 --- a/arch/m68k/configs/mac_defconfig +++ b/arch/m68k/configs/mac_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -54,7 +55,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -64,10 +64,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -176,6 +178,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -203,6 +206,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -271,6 +275,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -364,6 +369,7 @@ CONFIG_MAC8390=y # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_SMSC is not set # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -519,6 +525,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/multi_defconfig b/arch/m68k/configs/multi_defconfig index 8afca3753db1..83c80d2030ec 100644 --- a/arch/m68k/configs/multi_defconfig +++ b/arch/m68k/configs/multi_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -64,7 +65,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -74,10 +74,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -186,6 +188,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -213,6 +216,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -281,6 +285,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -410,6 +415,7 @@ CONFIG_ZORRO8390=y # CONFIG_NET_VENDOR_SEEQ is not set CONFIG_SMC91X=y # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PLIP=m @@ -599,6 +605,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/mvme147_defconfig b/arch/m68k/configs/mvme147_defconfig index ef00875994d9..6cb42c3bf5a2 100644 --- a/arch/m68k/configs/mvme147_defconfig +++ b/arch/m68k/configs/mvme147_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -52,7 +53,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -62,10 +62,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -174,6 +176,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -201,6 +204,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -266,6 +270,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -343,6 +348,7 @@ CONFIG_MVME147_NET=y # CONFIG_NET_VENDOR_SAMSUNG is not set # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -488,6 +494,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/mvme16x_defconfig b/arch/m68k/configs/mvme16x_defconfig index 387c2bd90ff1..c7508c30330c 100644 --- a/arch/m68k/configs/mvme16x_defconfig +++ b/arch/m68k/configs/mvme16x_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -53,7 +54,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -63,10 +63,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -175,6 +177,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -202,6 +205,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -267,6 +271,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -343,6 +348,7 @@ CONFIG_MVME16x_NET=y # CONFIG_NET_VENDOR_SAMSUNG is not set # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -488,6 +494,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/q40_defconfig b/arch/m68k/configs/q40_defconfig index 35355c1bc714..64b71664a303 100644 --- a/arch/m68k/configs/q40_defconfig +++ b/arch/m68k/configs/q40_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -53,7 +54,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -63,10 +63,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -175,6 +177,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -202,6 +205,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -267,6 +271,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -354,6 +359,7 @@ CONFIG_NE2000=y # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_SMSC is not set # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PLIP=m @@ -510,6 +516,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/configs/sun3_defconfig b/arch/m68k/configs/sun3_defconfig index 8442d267b877..9a4cab78a2ea 100644 --- a/arch/m68k/configs/sun3_defconfig +++ b/arch/m68k/configs/sun3_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -50,7 +51,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -60,10 +60,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -172,6 +174,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -199,6 +202,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -264,6 +268,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -341,6 +346,7 @@ CONFIG_SUN3_82586=y # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_STMICRO is not set # CONFIG_NET_VENDOR_SUN is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -489,6 +495,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m CONFIG_CRYPTO_MANAGER=y diff --git a/arch/m68k/configs/sun3x_defconfig b/arch/m68k/configs/sun3x_defconfig index 0e1b542e1555..1a2eaac13dbd 100644 --- a/arch/m68k/configs/sun3x_defconfig +++ b/arch/m68k/configs/sun3x_defconfig @@ -10,6 +10,7 @@ CONFIG_LOG_BUF_SHIFT=16 # CONFIG_PID_NS is not set # CONFIG_NET_NS is not set CONFIG_BLK_DEV_INITRD=y +CONFIG_USERFAULTFD=y CONFIG_SLAB=y CONFIG_MODULES=y CONFIG_MODULE_UNLOAD=y @@ -50,7 +51,6 @@ CONFIG_NET_IPGRE_DEMUX=m CONFIG_NET_IPGRE=m CONFIG_NET_IPVTI=m CONFIG_NET_FOU_IP_TUNNELS=y -CONFIG_GENEVE_CORE=m CONFIG_INET_AH=m CONFIG_INET_ESP=m CONFIG_INET_IPCOMP=m @@ -60,10 +60,12 @@ CONFIG_INET_XFRM_MODE_BEET=m # CONFIG_INET_LRO is not set CONFIG_INET_DIAG=m CONFIG_INET_UDP_DIAG=m +CONFIG_IPV6=m CONFIG_IPV6_ROUTER_PREF=y CONFIG_INET6_AH=m CONFIG_INET6_ESP=m CONFIG_INET6_IPCOMP=m +CONFIG_IPV6_ILA=m CONFIG_IPV6_VTI=m CONFIG_IPV6_GRE=m CONFIG_NETFILTER=y @@ -172,6 +174,7 @@ CONFIG_IP_SET_HASH_NETIFACE=m CONFIG_IP_SET_LIST_SET=m CONFIG_NF_CONNTRACK_IPV4=m CONFIG_NFT_CHAIN_ROUTE_IPV4=m +CONFIG_NFT_DUP_IPV4=m CONFIG_NF_TABLES_ARP=m CONFIG_NF_LOG_ARP=m CONFIG_NFT_CHAIN_NAT_IPV4=m @@ -199,6 +202,7 @@ CONFIG_IP_NF_ARPFILTER=m CONFIG_IP_NF_ARP_MANGLE=m CONFIG_NF_CONNTRACK_IPV6=m CONFIG_NFT_CHAIN_ROUTE_IPV6=m +CONFIG_NFT_DUP_IPV6=m CONFIG_NFT_CHAIN_NAT_IPV6=m CONFIG_NFT_MASQ_IPV6=m CONFIG_NFT_REDIR_IPV6=m @@ -264,6 +268,7 @@ CONFIG_NETLINK_DIAG=m CONFIG_MPLS=y CONFIG_NET_MPLS_GSO=m CONFIG_MPLS_ROUTING=m +CONFIG_MPLS_IPTUNNEL=m # CONFIG_WIRELESS is not set # CONFIG_UEVENT_HELPER is not set CONFIG_DEVTMPFS=y @@ -341,6 +346,7 @@ CONFIG_SUN3LANCE=y # CONFIG_NET_VENDOR_SAMSUNG is not set # CONFIG_NET_VENDOR_SEEQ is not set # CONFIG_NET_VENDOR_STMICRO is not set +# CONFIG_NET_VENDOR_SYNOPSYS is not set # CONFIG_NET_VENDOR_VIA is not set # CONFIG_NET_VENDOR_WIZNET is not set CONFIG_PPP=m @@ -489,6 +495,7 @@ CONFIG_TEST_USER_COPY=m CONFIG_TEST_BPF=m CONFIG_TEST_FIRMWARE=m CONFIG_TEST_UDELAY=m +CONFIG_TEST_STATIC_KEYS=m CONFIG_EARLY_PRINTK=y CONFIG_ENCRYPTED_KEYS=m CONFIG_CRYPTO_RSA=m diff --git a/arch/m68k/include/asm/linkage.h b/arch/m68k/include/asm/linkage.h index 5a822bb790f7..066e74f666ae 100644 --- a/arch/m68k/include/asm/linkage.h +++ b/arch/m68k/include/asm/linkage.h @@ -4,4 +4,34 @@ #define __ALIGN .align 4 #define __ALIGN_STR ".align 4" +/* + * Make sure the compiler doesn't do anything stupid with the + * arguments on the stack - they are owned by the *caller*, not + * the callee. This just fools gcc into not spilling into them, + * and keeps it from doing tailcall recursion and/or using the + * stack slots for temporaries, since they are live and "used" + * all the way to the end of the function. + */ +#define asmlinkage_protect(n, ret, args...) \ + __asmlinkage_protect##n(ret, ##args) +#define __asmlinkage_protect_n(ret, args...) \ + __asm__ __volatile__ ("" : "=r" (ret) : "0" (ret), ##args) +#define __asmlinkage_protect0(ret) \ + __asmlinkage_protect_n(ret) +#define __asmlinkage_protect1(ret, arg1) \ + __asmlinkage_protect_n(ret, "m" (arg1)) +#define __asmlinkage_protect2(ret, arg1, arg2) \ + __asmlinkage_protect_n(ret, "m" (arg1), "m" (arg2)) +#define __asmlinkage_protect3(ret, arg1, arg2, arg3) \ + __asmlinkage_protect_n(ret, "m" (arg1), "m" (arg2), "m" (arg3)) +#define __asmlinkage_protect4(ret, arg1, arg2, arg3, arg4) \ + __asmlinkage_protect_n(ret, "m" (arg1), "m" (arg2), "m" (arg3), \ + "m" (arg4)) +#define __asmlinkage_protect5(ret, arg1, arg2, arg3, arg4, arg5) \ + __asmlinkage_protect_n(ret, "m" (arg1), "m" (arg2), "m" (arg3), \ + "m" (arg4), "m" (arg5)) +#define __asmlinkage_protect6(ret, arg1, arg2, arg3, arg4, arg5, arg6) \ + __asmlinkage_protect_n(ret, "m" (arg1), "m" (arg2), "m" (arg3), \ + "m" (arg4), "m" (arg5), "m" (arg6)) + #endif diff --git a/arch/m68k/include/asm/unistd.h b/arch/m68k/include/asm/unistd.h index 244e0dbe45db..0793a7f17417 100644 --- a/arch/m68k/include/asm/unistd.h +++ b/arch/m68k/include/asm/unistd.h @@ -4,7 +4,7 @@ #include <uapi/asm/unistd.h> -#define NR_syscalls 356 +#define NR_syscalls 375 #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_OLD_STAT diff --git a/arch/m68k/include/uapi/asm/unistd.h b/arch/m68k/include/uapi/asm/unistd.h index 61fb6cb9d2ae..5e6fae6c275f 100644 --- a/arch/m68k/include/uapi/asm/unistd.h +++ b/arch/m68k/include/uapi/asm/unistd.h @@ -361,5 +361,24 @@ #define __NR_memfd_create 353 #define __NR_bpf 354 #define __NR_execveat 355 +#define __NR_socket 356 +#define __NR_socketpair 357 +#define __NR_bind 358 +#define __NR_connect 359 +#define __NR_listen 360 +#define __NR_accept4 361 +#define __NR_getsockopt 362 +#define __NR_setsockopt 363 +#define __NR_getsockname 364 +#define __NR_getpeername 365 +#define __NR_sendto 366 +#define __NR_sendmsg 367 +#define __NR_recvfrom 368 +#define __NR_recvmsg 369 +#define __NR_shutdown 370 +#define __NR_recvmmsg 371 +#define __NR_sendmmsg 372 +#define __NR_userfaultfd 373 +#define __NR_membarrier 374 #endif /* _UAPI_ASM_M68K_UNISTD_H_ */ diff --git a/arch/m68k/kernel/syscalltable.S b/arch/m68k/kernel/syscalltable.S index a0ec4303f2c8..5dd0e80042f5 100644 --- a/arch/m68k/kernel/syscalltable.S +++ b/arch/m68k/kernel/syscalltable.S @@ -376,4 +376,22 @@ ENTRY(sys_call_table) .long sys_memfd_create .long sys_bpf .long sys_execveat /* 355 */ - + .long sys_socket + .long sys_socketpair + .long sys_bind + .long sys_connect + .long sys_listen /* 360 */ + .long sys_accept4 + .long sys_getsockopt + .long sys_setsockopt + .long sys_getsockname + .long sys_getpeername /* 365 */ + .long sys_sendto + .long sys_sendmsg + .long sys_recvfrom + .long sys_recvmsg + .long sys_shutdown /* 370 */ + .long sys_recvmmsg + .long sys_sendmmsg + .long sys_userfaultfd + .long sys_membarrier diff --git a/arch/metag/include/asm/Kbuild b/arch/metag/include/asm/Kbuild index df31353fd200..29acb89daaaa 100644 --- a/arch/metag/include/asm/Kbuild +++ b/arch/metag/include/asm/Kbuild @@ -54,4 +54,5 @@ generic-y += ucontext.h generic-y += unaligned.h generic-y += user.h generic-y += vga.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/microblaze/include/asm/Kbuild b/arch/microblaze/include/asm/Kbuild index 2f222f355c4b..b0ae88c9fed9 100644 --- a/arch/microblaze/include/asm/Kbuild +++ b/arch/microblaze/include/asm/Kbuild @@ -10,3 +10,4 @@ generic-y += mm-arch-hooks.h generic-y += preempt.h generic-y += syscalls.h generic-y += trace_clock.h +generic-y += word-at-a-time.h diff --git a/arch/mips/ath79/irq.c b/arch/mips/ath79/irq.c index 15ecb4831e12..eeb3953ed8ac 100644 --- a/arch/mips/ath79/irq.c +++ b/arch/mips/ath79/irq.c @@ -293,8 +293,26 @@ static int __init ath79_misc_intc_of_init( return 0; } -IRQCHIP_DECLARE(ath79_misc_intc, "qca,ar7100-misc-intc", - ath79_misc_intc_of_init); + +static int __init ar7100_misc_intc_of_init( + struct device_node *node, struct device_node *parent) +{ + ath79_misc_irq_chip.irq_mask_ack = ar71xx_misc_irq_mask; + return ath79_misc_intc_of_init(node, parent); +} + +IRQCHIP_DECLARE(ar7100_misc_intc, "qca,ar7100-misc-intc", + ar7100_misc_intc_of_init); + +static int __init ar7240_misc_intc_of_init( + struct device_node *node, struct device_node *parent) +{ + ath79_misc_irq_chip.irq_ack = ar724x_misc_irq_ack; + return ath79_misc_intc_of_init(node, parent); +} + +IRQCHIP_DECLARE(ar7240_misc_intc, "qca,ar7240-misc-intc", + ar7240_misc_intc_of_init); static int __init ar79_cpu_intc_of_init( struct device_node *node, struct device_node *parent) diff --git a/arch/mips/cavium-octeon/setup.c b/arch/mips/cavium-octeon/setup.c index 89a628455bc2..bd634259eab9 100644 --- a/arch/mips/cavium-octeon/setup.c +++ b/arch/mips/cavium-octeon/setup.c @@ -933,7 +933,7 @@ void __init plat_mem_setup(void) while ((boot_mem_map.nr_map < BOOT_MEM_MAP_MAX) && (total < MAX_MEMORY)) { memory = cvmx_bootmem_phy_alloc(mem_alloc_size, - __pa_symbol(&__init_end), -1, + __pa_symbol(&_end), -1, 0x100000, CVMX_BOOTMEM_FLAG_NO_LOCKING); if (memory >= 0) { diff --git a/arch/mips/include/asm/Kbuild b/arch/mips/include/asm/Kbuild index 40ec4ca3f946..c7fe4d01e79c 100644 --- a/arch/mips/include/asm/Kbuild +++ b/arch/mips/include/asm/Kbuild @@ -17,4 +17,5 @@ generic-y += segment.h generic-y += serial.h generic-y += trace_clock.h generic-y += user.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/mips/include/asm/cpu-features.h b/arch/mips/include/asm/cpu-features.h index 9801ac982655..fe67f12ac239 100644 --- a/arch/mips/include/asm/cpu-features.h +++ b/arch/mips/include/asm/cpu-features.h @@ -20,6 +20,9 @@ #ifndef cpu_has_tlb #define cpu_has_tlb (cpu_data[0].options & MIPS_CPU_TLB) #endif +#ifndef cpu_has_ftlb +#define cpu_has_ftlb (cpu_data[0].options & MIPS_CPU_FTLB) +#endif #ifndef cpu_has_tlbinv #define cpu_has_tlbinv (cpu_data[0].options & MIPS_CPU_TLBINV) #endif diff --git a/arch/mips/include/asm/cpu.h b/arch/mips/include/asm/cpu.h index cd89e9855775..82ad15f11049 100644 --- a/arch/mips/include/asm/cpu.h +++ b/arch/mips/include/asm/cpu.h @@ -385,6 +385,7 @@ enum cpu_type_enum { #define MIPS_CPU_CDMM 0x4000000000ull /* CPU has Common Device Memory Map */ #define MIPS_CPU_BP_GHIST 0x8000000000ull /* R12K+ Branch Prediction Global History */ #define MIPS_CPU_SP 0x10000000000ull /* Small (1KB) page support */ +#define MIPS_CPU_FTLB 0x20000000000ull /* CPU has Fixed-page-size TLB */ /* * CPU ASE encodings diff --git a/arch/mips/include/asm/io.h b/arch/mips/include/asm/io.h index 9e777cd42b67..d10fd80dbb7e 100644 --- a/arch/mips/include/asm/io.h +++ b/arch/mips/include/asm/io.h @@ -256,6 +256,7 @@ static inline void __iomem * __ioremap_mode(phys_addr_t offset, unsigned long si */ #define ioremap_nocache(offset, size) \ __ioremap_mode((offset), (size), _CACHE_UNCACHED) +#define ioremap_uc ioremap_nocache /* * ioremap_cachable - map bus memory into CPU space diff --git a/arch/mips/include/asm/maar.h b/arch/mips/include/asm/maar.h index b02891f9caaf..21d9607c80d7 100644 --- a/arch/mips/include/asm/maar.h +++ b/arch/mips/include/asm/maar.h @@ -66,6 +66,15 @@ static inline void write_maar_pair(unsigned idx, phys_addr_t lower, } /** + * maar_init() - initialise MAARs + * + * Performs initialisation of MAARs for the current CPU, making use of the + * platforms implementation of platform_maar_init where necessary and + * duplicating the setup it provides on secondary CPUs. + */ +extern void maar_init(void); + +/** * struct maar_config - MAAR configuration data * @lower: The lowest address that the MAAR pair will affect. Must be * aligned to a 2^16 byte boundary. diff --git a/arch/mips/include/asm/mips-cm.h b/arch/mips/include/asm/mips-cm.h index d75b75e78ebb..1f1927ab4269 100644 --- a/arch/mips/include/asm/mips-cm.h +++ b/arch/mips/include/asm/mips-cm.h @@ -194,6 +194,7 @@ BUILD_CM_RW(reg3_mask, MIPS_CM_GCB_OFS + 0xc8) BUILD_CM_R_(gic_status, MIPS_CM_GCB_OFS + 0xd0) BUILD_CM_R_(cpc_status, MIPS_CM_GCB_OFS + 0xf0) BUILD_CM_RW(l2_config, MIPS_CM_GCB_OFS + 0x130) +BUILD_CM_RW(sys_config2, MIPS_CM_GCB_OFS + 0x150) /* Core Local & Core Other register accessor functions */ BUILD_CM_Cx_RW(reset_release, 0x00) @@ -316,6 +317,10 @@ BUILD_CM_Cx_R_(tcid_8_priority, 0x80) #define CM_GCR_L2_CONFIG_ASSOC_SHF 0 #define CM_GCR_L2_CONFIG_ASSOC_MSK (_ULCAST_(0xff) << 0) +/* GCR_SYS_CONFIG2 register fields */ +#define CM_GCR_SYS_CONFIG2_MAXVPW_SHF 0 +#define CM_GCR_SYS_CONFIG2_MAXVPW_MSK (_ULCAST_(0xf) << 0) + /* GCR_Cx_COHERENCE register fields */ #define CM_GCR_Cx_COHERENCE_COHDOMAINEN_SHF 0 #define CM_GCR_Cx_COHERENCE_COHDOMAINEN_MSK (_ULCAST_(0xff) << 0) @@ -405,4 +410,38 @@ static inline int mips_cm_revision(void) return read_gcr_rev(); } +/** + * mips_cm_max_vp_width() - return the width in bits of VP indices + * + * Return: the width, in bits, of VP indices in fields that combine core & VP + * indices. + */ +static inline unsigned int mips_cm_max_vp_width(void) +{ + extern int smp_num_siblings; + + if (mips_cm_revision() >= CM_REV_CM3) + return read_gcr_sys_config2() & CM_GCR_SYS_CONFIG2_MAXVPW_MSK; + + return smp_num_siblings; +} + +/** + * mips_cm_vp_id() - calculate the hardware VP ID for a CPU + * @cpu: the CPU whose VP ID to calculate + * + * Hardware such as the GIC uses identifiers for VPs which may not match the + * CPU numbers used by Linux. This function calculates the hardware VP + * identifier corresponding to a given CPU. + * + * Return: the VP ID for the CPU. + */ +static inline unsigned int mips_cm_vp_id(unsigned int cpu) +{ + unsigned int core = cpu_data[cpu].core; + unsigned int vp = cpu_vpe_id(&cpu_data[cpu]); + + return (core * mips_cm_max_vp_width()) + vp; +} + #endif /* __MIPS_ASM_MIPS_CM_H__ */ diff --git a/arch/mips/include/asm/mipsregs.h b/arch/mips/include/asm/mipsregs.h index d3cd8eac81e3..c64781cf649f 100644 --- a/arch/mips/include/asm/mipsregs.h +++ b/arch/mips/include/asm/mipsregs.h @@ -487,6 +487,8 @@ /* Bits specific to the MIPS32/64 PRA. */ #define MIPS_CONF_MT (_ULCAST_(7) << 7) +#define MIPS_CONF_MT_TLB (_ULCAST_(1) << 7) +#define MIPS_CONF_MT_FTLB (_ULCAST_(4) << 7) #define MIPS_CONF_AR (_ULCAST_(7) << 10) #define MIPS_CONF_AT (_ULCAST_(3) << 13) #define MIPS_CONF_M (_ULCAST_(1) << 31) diff --git a/arch/mips/include/uapi/asm/swab.h b/arch/mips/include/uapi/asm/swab.h index c4ddc4f0d2dc..23cd9b118c9e 100644 --- a/arch/mips/include/uapi/asm/swab.h +++ b/arch/mips/include/uapi/asm/swab.h @@ -13,16 +13,15 @@ #define __SWAB_64_THRU_32__ -#if (defined(__mips_isa_rev) && (__mips_isa_rev >= 2)) || \ - defined(_MIPS_ARCH_LOONGSON3A) +#if !defined(__mips16) && \ + ((defined(__mips_isa_rev) && (__mips_isa_rev >= 2)) || \ + defined(_MIPS_ARCH_LOONGSON3A)) -static inline __attribute__((nomips16)) __attribute_const__ - __u16 __arch_swab16(__u16 x) +static inline __attribute_const__ __u16 __arch_swab16(__u16 x) { __asm__( " .set push \n" " .set arch=mips32r2 \n" - " .set nomips16 \n" " wsbh %0, %1 \n" " .set pop \n" : "=r" (x) @@ -32,13 +31,11 @@ static inline __attribute__((nomips16)) __attribute_const__ } #define __arch_swab16 __arch_swab16 -static inline __attribute__((nomips16)) __attribute_const__ - __u32 __arch_swab32(__u32 x) +static inline __attribute_const__ __u32 __arch_swab32(__u32 x) { __asm__( " .set push \n" " .set arch=mips32r2 \n" - " .set nomips16 \n" " wsbh %0, %1 \n" " rotr %0, %0, 16 \n" " .set pop \n" @@ -54,13 +51,11 @@ static inline __attribute__((nomips16)) __attribute_const__ * 64-bit kernel on r2 CPUs. */ #ifdef __mips64 -static inline __attribute__((nomips16)) __attribute_const__ - __u64 __arch_swab64(__u64 x) +static inline __attribute_const__ __u64 __arch_swab64(__u64 x) { __asm__( " .set push \n" " .set arch=mips64r2 \n" - " .set nomips16 \n" " dsbh %0, %1 \n" " dshd %0, %0 \n" " .set pop \n" @@ -71,5 +66,5 @@ static inline __attribute__((nomips16)) __attribute_const__ } #define __arch_swab64 __arch_swab64 #endif /* __mips64 */ -#endif /* MIPS R2 or newer or Loongson 3A */ +#endif /* (not __mips16) and (MIPS R2 or newer or Loongson 3A) */ #endif /* _ASM_SWAB_H */ diff --git a/arch/mips/include/uapi/asm/unistd.h b/arch/mips/include/uapi/asm/unistd.h index c03088f9f514..cfabadb135d9 100644 --- a/arch/mips/include/uapi/asm/unistd.h +++ b/arch/mips/include/uapi/asm/unistd.h @@ -377,16 +377,18 @@ #define __NR_memfd_create (__NR_Linux + 354) #define __NR_bpf (__NR_Linux + 355) #define __NR_execveat (__NR_Linux + 356) +#define __NR_userfaultfd (__NR_Linux + 357) +#define __NR_membarrier (__NR_Linux + 358) /* * Offset of the last Linux o32 flavoured syscall */ -#define __NR_Linux_syscalls 356 +#define __NR_Linux_syscalls 358 #endif /* _MIPS_SIM == _MIPS_SIM_ABI32 */ #define __NR_O32_Linux 4000 -#define __NR_O32_Linux_syscalls 356 +#define __NR_O32_Linux_syscalls 358 #if _MIPS_SIM == _MIPS_SIM_ABI64 @@ -711,16 +713,18 @@ #define __NR_memfd_create (__NR_Linux + 314) #define __NR_bpf (__NR_Linux + 315) #define __NR_execveat (__NR_Linux + 316) +#define __NR_userfaultfd (__NR_Linux + 317) +#define __NR_membarrier (__NR_Linux + 318) /* * Offset of the last Linux 64-bit flavoured syscall */ -#define __NR_Linux_syscalls 316 +#define __NR_Linux_syscalls 318 #endif /* _MIPS_SIM == _MIPS_SIM_ABI64 */ #define __NR_64_Linux 5000 -#define __NR_64_Linux_syscalls 316 +#define __NR_64_Linux_syscalls 318 #if _MIPS_SIM == _MIPS_SIM_NABI32 @@ -1049,15 +1053,17 @@ #define __NR_memfd_create (__NR_Linux + 318) #define __NR_bpf (__NR_Linux + 319) #define __NR_execveat (__NR_Linux + 320) +#define __NR_userfaultfd (__NR_Linux + 321) +#define __NR_membarrier (__NR_Linux + 322) /* * Offset of the last N32 flavoured syscall */ -#define __NR_Linux_syscalls 320 +#define __NR_Linux_syscalls 322 #endif /* _MIPS_SIM == _MIPS_SIM_NABI32 */ #define __NR_N32_Linux 6000 -#define __NR_N32_Linux_syscalls 320 +#define __NR_N32_Linux_syscalls 322 #endif /* _UAPI_ASM_UNISTD_H */ diff --git a/arch/mips/jz4740/board-qi_lb60.c b/arch/mips/jz4740/board-qi_lb60.c index 4e62bf85d0b0..459cb017306c 100644 --- a/arch/mips/jz4740/board-qi_lb60.c +++ b/arch/mips/jz4740/board-qi_lb60.c @@ -26,6 +26,7 @@ #include <linux/power/jz4740-battery.h> #include <linux/power/gpio-charger.h> +#include <asm/mach-jz4740/gpio.h> #include <asm/mach-jz4740/jz4740_fb.h> #include <asm/mach-jz4740/jz4740_mmc.h> #include <asm/mach-jz4740/jz4740_nand.h> diff --git a/arch/mips/jz4740/gpio.c b/arch/mips/jz4740/gpio.c index a74e181058b0..8c6d76c9b2d6 100644 --- a/arch/mips/jz4740/gpio.c +++ b/arch/mips/jz4740/gpio.c @@ -28,6 +28,7 @@ #include <linux/seq_file.h> #include <asm/mach-jz4740/base.h> +#include <asm/mach-jz4740/gpio.h> #define JZ4740_GPIO_BASE_A (32*0) #define JZ4740_GPIO_BASE_B (32*1) diff --git a/arch/mips/kernel/cps-vec.S b/arch/mips/kernel/cps-vec.S index 9f71c06aebf6..209ded16806b 100644 --- a/arch/mips/kernel/cps-vec.S +++ b/arch/mips/kernel/cps-vec.S @@ -39,6 +39,7 @@ mfc0 \dest, CP0_CONFIG, 3 andi \dest, \dest, MIPS_CONF3_MT beqz \dest, \nomt + nop .endm .section .text.cps-vec @@ -223,10 +224,9 @@ LEAF(excep_ejtag) END(excep_ejtag) LEAF(mips_cps_core_init) -#ifdef CONFIG_MIPS_MT +#ifdef CONFIG_MIPS_MT_SMP /* Check that the core implements the MT ASE */ has_mt t0, 3f - nop .set push .set mips64r2 @@ -310,8 +310,9 @@ LEAF(mips_cps_boot_vpes) PTR_ADDU t0, t0, t1 /* Calculate this VPEs ID. If the core doesn't support MT use 0 */ + li t9, 0 +#ifdef CONFIG_MIPS_MT_SMP has_mt ta2, 1f - li t9, 0 /* Find the number of VPEs present in the core */ mfc0 t1, CP0_MVPCONF0 @@ -330,6 +331,7 @@ LEAF(mips_cps_boot_vpes) /* Retrieve the VPE ID from EBase.CPUNum */ mfc0 t9, $15, 1 and t9, t9, t1 +#endif 1: /* Calculate a pointer to this VPEs struct vpe_boot_config */ li t1, VPEBOOTCFG_SIZE @@ -337,7 +339,7 @@ LEAF(mips_cps_boot_vpes) PTR_L ta3, COREBOOTCFG_VPECONFIG(t0) PTR_ADDU v0, v0, ta3 -#ifdef CONFIG_MIPS_MT +#ifdef CONFIG_MIPS_MT_SMP /* If the core doesn't support MT then return */ bnez ta2, 1f @@ -451,7 +453,7 @@ LEAF(mips_cps_boot_vpes) 2: .set pop -#endif /* CONFIG_MIPS_MT */ +#endif /* CONFIG_MIPS_MT_SMP */ /* Return */ jr ra diff --git a/arch/mips/kernel/cpu-probe.c b/arch/mips/kernel/cpu-probe.c index 571a8e6ea5bd..09a51d091941 100644 --- a/arch/mips/kernel/cpu-probe.c +++ b/arch/mips/kernel/cpu-probe.c @@ -410,16 +410,18 @@ static int set_ftlb_enable(struct cpuinfo_mips *c, int enable) static inline unsigned int decode_config0(struct cpuinfo_mips *c) { unsigned int config0; - int isa; + int isa, mt; config0 = read_c0_config(); /* * Look for Standard TLB or Dual VTLB and FTLB */ - if ((((config0 & MIPS_CONF_MT) >> 7) == 1) || - (((config0 & MIPS_CONF_MT) >> 7) == 4)) + mt = config0 & MIPS_CONF_MT; + if (mt == MIPS_CONF_MT_TLB) c->options |= MIPS_CPU_TLB; + else if (mt == MIPS_CONF_MT_FTLB) + c->options |= MIPS_CPU_TLB | MIPS_CPU_FTLB; isa = (config0 & MIPS_CONF_AT) >> 13; switch (isa) { @@ -559,15 +561,18 @@ static inline unsigned int decode_config4(struct cpuinfo_mips *c) if (cpu_has_tlb) { if (((config4 & MIPS_CONF4_IE) >> 29) == 2) c->options |= MIPS_CPU_TLBINV; + /* - * This is a bit ugly. R6 has dropped that field from - * config4 and the only valid configuration is VTLB+FTLB so - * set a good value for mmuextdef for that case. + * R6 has dropped the MMUExtDef field from config4. + * On R6 the fields always describe the FTLB, and only if it is + * present according to Config.MT. */ - if (cpu_has_mips_r6) + if (!cpu_has_mips_r6) + mmuextdef = config4 & MIPS_CONF4_MMUEXTDEF; + else if (cpu_has_ftlb) mmuextdef = MIPS_CONF4_MMUEXTDEF_VTLBSIZEEXT; else - mmuextdef = config4 & MIPS_CONF4_MMUEXTDEF; + mmuextdef = 0; switch (mmuextdef) { case MIPS_CONF4_MMUEXTDEF_MMUSIZEEXT: diff --git a/arch/mips/kernel/octeon_switch.S b/arch/mips/kernel/octeon_switch.S index 423ae83af1fb..3375745b9198 100644 --- a/arch/mips/kernel/octeon_switch.S +++ b/arch/mips/kernel/octeon_switch.S @@ -18,7 +18,7 @@ .set pop /* * task_struct *resume(task_struct *prev, task_struct *next, - * struct thread_info *next_ti, int usedfpu) + * struct thread_info *next_ti) */ .align 7 LEAF(resume) @@ -28,30 +28,6 @@ cpu_save_nonscratch a0 LONG_S ra, THREAD_REG31(a0) - /* - * check if we need to save FPU registers - */ - .set push - .set noreorder - beqz a3, 1f - PTR_L t3, TASK_THREAD_INFO(a0) - .set pop - - /* - * clear saved user stack CU1 bit - */ - LONG_L t0, ST_OFF(t3) - li t1, ~ST0_CU1 - and t0, t0, t1 - LONG_S t0, ST_OFF(t3) - - .set push - .set arch=mips64r2 - fpu_save_double a0 t0 t1 # c0_status passed in t0 - # clobbers t1 - .set pop -1: - #if CONFIG_CAVIUM_OCTEON_CVMSEG_SIZE > 0 /* Check if we need to store CVMSEG state */ dmfc0 t0, $11,7 /* CvmMemCtl */ diff --git a/arch/mips/kernel/r2300_switch.S b/arch/mips/kernel/r2300_switch.S index 5087a4b72e6b..ac27ef7d4d0e 100644 --- a/arch/mips/kernel/r2300_switch.S +++ b/arch/mips/kernel/r2300_switch.S @@ -31,18 +31,8 @@ #define ST_OFF (_THREAD_SIZE - 32 - PT_SIZE + PT_STATUS) /* - * FPU context is saved iff the process has used it's FPU in the current - * time slice as indicated by TIF_USEDFPU. In any case, the CU1 bit for user - * space STATUS register should be 0, so that a process *always* starts its - * userland with FPU disabled after each context switch. - * - * FPU will be enabled as soon as the process accesses FPU again, through - * do_cpu() trap. - */ - -/* * task_struct *resume(task_struct *prev, task_struct *next, - * struct thread_info *next_ti, int usedfpu) + * struct thread_info *next_ti) */ LEAF(resume) mfc0 t1, CP0_STATUS @@ -50,22 +40,6 @@ LEAF(resume) cpu_save_nonscratch a0 sw ra, THREAD_REG31(a0) - beqz a3, 1f - - PTR_L t3, TASK_THREAD_INFO(a0) - - /* - * clear saved user stack CU1 bit - */ - lw t0, ST_OFF(t3) - li t1, ~ST0_CU1 - and t0, t0, t1 - sw t0, ST_OFF(t3) - - fpu_save_single a0, t0 # clobbers t0 - -1: - #if defined(CONFIG_CC_STACKPROTECTOR) && !defined(CONFIG_SMP) PTR_LA t8, __stack_chk_guard LONG_L t9, TASK_STACK_CANARY(a1) diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S index 4cc13508d967..65a74e4f0f45 100644 --- a/arch/mips/kernel/scall32-o32.S +++ b/arch/mips/kernel/scall32-o32.S @@ -36,16 +36,8 @@ NESTED(handle_sys, PT_SIZE, sp) lw t1, PT_EPC(sp) # skip syscall on return subu v0, v0, __NR_O32_Linux # check syscall number - sltiu t0, v0, __NR_O32_Linux_syscalls + 1 addiu t1, 4 # skip to next instruction sw t1, PT_EPC(sp) - beqz t0, illegal_syscall - - sll t0, v0, 2 - la t1, sys_call_table - addu t1, t0 - lw t2, (t1) # syscall routine - beqz t2, illegal_syscall sw a3, PT_R26(sp) # save a3 for syscall restarting @@ -96,6 +88,16 @@ loads_done: li t1, _TIF_WORK_SYSCALL_ENTRY and t0, t1 bnez t0, syscall_trace_entry # -> yes +syscall_common: + sltiu t0, v0, __NR_O32_Linux_syscalls + 1 + beqz t0, illegal_syscall + + sll t0, v0, 2 + la t1, sys_call_table + addu t1, t0 + lw t2, (t1) # syscall routine + + beqz t2, illegal_syscall jalr t2 # Do The Real Thing (TM) @@ -116,7 +118,7 @@ o32_syscall_exit: syscall_trace_entry: SAVE_STATIC - move s0, t2 + move s0, v0 move a0, sp /* @@ -129,27 +131,18 @@ syscall_trace_entry: 1: jal syscall_trace_enter - bltz v0, 2f # seccomp failed? Skip syscall + bltz v0, 1f # seccomp failed? Skip syscall + + move v0, s0 # restore syscall - move t0, s0 RESTORE_STATIC lw a0, PT_R4(sp) # Restore argument registers lw a1, PT_R5(sp) lw a2, PT_R6(sp) lw a3, PT_R7(sp) - jalr t0 - - li t0, -EMAXERRNO - 1 # error? - sltu t0, t0, v0 - sw t0, PT_R7(sp) # set error flag - beqz t0, 1f - - lw t1, PT_R2(sp) # syscall number - negu v0 # error - sw t1, PT_R0(sp) # save it for syscall restarting -1: sw v0, PT_R2(sp) # result + j syscall_common -2: j syscall_exit +1: j syscall_exit /* ------------------------------------------------------------------------ */ @@ -599,3 +592,5 @@ EXPORT(sys_call_table) PTR sys_memfd_create PTR sys_bpf /* 4355 */ PTR sys_execveat + PTR sys_userfaultfd + PTR sys_membarrier diff --git a/arch/mips/kernel/scall64-64.S b/arch/mips/kernel/scall64-64.S index a6f6b762c47a..e732981cf99f 100644 --- a/arch/mips/kernel/scall64-64.S +++ b/arch/mips/kernel/scall64-64.S @@ -39,18 +39,11 @@ NESTED(handle_sys64, PT_SIZE, sp) .set at #endif - dsubu t0, v0, __NR_64_Linux # check syscall number - sltiu t0, t0, __NR_64_Linux_syscalls + 1 #if !defined(CONFIG_MIPS32_O32) && !defined(CONFIG_MIPS32_N32) ld t1, PT_EPC(sp) # skip syscall on return daddiu t1, 4 # skip to next instruction sd t1, PT_EPC(sp) #endif - beqz t0, illegal_syscall - - dsll t0, v0, 3 # offset into table - ld t2, (sys_call_table - (__NR_64_Linux * 8))(t0) - # syscall routine sd a3, PT_R26(sp) # save a3 for syscall restarting @@ -59,6 +52,17 @@ NESTED(handle_sys64, PT_SIZE, sp) and t0, t1, t0 bnez t0, syscall_trace_entry +syscall_common: + dsubu t2, v0, __NR_64_Linux + sltiu t0, t2, __NR_64_Linux_syscalls + 1 + beqz t0, illegal_syscall + + dsll t0, t2, 3 # offset into table + dla t2, sys_call_table + daddu t0, t2, t0 + ld t2, (t0) # syscall routine + beqz t2, illegal_syscall + jalr t2 # Do The Real Thing (TM) li t0, -EMAXERRNO - 1 # error? @@ -78,14 +82,14 @@ n64_syscall_exit: syscall_trace_entry: SAVE_STATIC - move s0, t2 + move s0, v0 move a0, sp move a1, v0 jal syscall_trace_enter - bltz v0, 2f # seccomp failed? Skip syscall + bltz v0, 1f # seccomp failed? Skip syscall - move t0, s0 + move v0, s0 RESTORE_STATIC ld a0, PT_R4(sp) # Restore argument registers ld a1, PT_R5(sp) @@ -93,19 +97,9 @@ syscall_trace_entry: ld a3, PT_R7(sp) ld a4, PT_R8(sp) ld a5, PT_R9(sp) - jalr t0 - - li t0, -EMAXERRNO - 1 # error? - sltu t0, t0, v0 - sd t0, PT_R7(sp) # set error flag - beqz t0, 1f - - ld t1, PT_R2(sp) # syscall number - dnegu v0 # error - sd t1, PT_R0(sp) # save it for syscall restarting -1: sd v0, PT_R2(sp) # result + j syscall_common -2: j syscall_exit +1: j syscall_exit illegal_syscall: /* This also isn't a 64-bit syscall, throw an error. */ @@ -436,4 +430,6 @@ EXPORT(sys_call_table) PTR sys_memfd_create PTR sys_bpf /* 5315 */ PTR sys_execveat + PTR sys_userfaultfd + PTR sys_membarrier .size sys_call_table,.-sys_call_table diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S index 4b2010654c46..c79484397584 100644 --- a/arch/mips/kernel/scall64-n32.S +++ b/arch/mips/kernel/scall64-n32.S @@ -52,6 +52,7 @@ NESTED(handle_sysn32, PT_SIZE, sp) and t0, t1, t0 bnez t0, n32_syscall_trace_entry +syscall_common: jalr t2 # Do The Real Thing (TM) li t0, -EMAXERRNO - 1 # error? @@ -75,9 +76,9 @@ n32_syscall_trace_entry: move a1, v0 jal syscall_trace_enter - bltz v0, 2f # seccomp failed? Skip syscall + bltz v0, 1f # seccomp failed? Skip syscall - move t0, s0 + move t2, s0 RESTORE_STATIC ld a0, PT_R4(sp) # Restore argument registers ld a1, PT_R5(sp) @@ -85,19 +86,9 @@ n32_syscall_trace_entry: ld a3, PT_R7(sp) ld a4, PT_R8(sp) ld a5, PT_R9(sp) - jalr t0 + j syscall_common - li t0, -EMAXERRNO - 1 # error? - sltu t0, t0, v0 - sd t0, PT_R7(sp) # set error flag - beqz t0, 1f - - ld t1, PT_R2(sp) # syscall number - dnegu v0 # error - sd t1, PT_R0(sp) # save it for syscall restarting -1: sd v0, PT_R2(sp) # result - -2: j syscall_exit +1: j syscall_exit not_n32_scall: /* This is not an n32 compatibility syscall, pass it on to @@ -429,4 +420,6 @@ EXPORT(sysn32_call_table) PTR sys_memfd_create PTR sys_bpf PTR compat_sys_execveat /* 6320 */ + PTR sys_userfaultfd + PTR sys_membarrier .size sysn32_call_table,.-sysn32_call_table diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S index f543ff4feef9..6369cfd390c6 100644 --- a/arch/mips/kernel/scall64-o32.S +++ b/arch/mips/kernel/scall64-o32.S @@ -87,6 +87,7 @@ loads_done: and t0, t1, t0 bnez t0, trace_a_syscall +syscall_common: jalr t2 # Do The Real Thing (TM) li t0, -EMAXERRNO - 1 # error? @@ -130,9 +131,9 @@ trace_a_syscall: 1: jal syscall_trace_enter - bltz v0, 2f # seccomp failed? Skip syscall + bltz v0, 1f # seccomp failed? Skip syscall - move t0, s0 + move t2, s0 RESTORE_STATIC ld a0, PT_R4(sp) # Restore argument registers ld a1, PT_R5(sp) @@ -142,19 +143,9 @@ trace_a_syscall: ld a5, PT_R9(sp) ld a6, PT_R10(sp) ld a7, PT_R11(sp) # For indirect syscalls - jalr t0 + j syscall_common - li t0, -EMAXERRNO - 1 # error? - sltu t0, t0, v0 - sd t0, PT_R7(sp) # set error flag - beqz t0, 1f - - ld t1, PT_R2(sp) # syscall number - dnegu v0 # error - sd t1, PT_R0(sp) # save it for syscall restarting -1: sd v0, PT_R2(sp) # result - -2: j syscall_exit +1: j syscall_exit /* ------------------------------------------------------------------------ */ @@ -584,4 +575,6 @@ EXPORT(sys32_call_table) PTR sys_memfd_create PTR sys_bpf /* 4355 */ PTR compat_sys_execveat + PTR sys_userfaultfd + PTR sys_membarrier .size sys32_call_table,.-sys32_call_table diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c index 35b8316002f8..479515109e5b 100644 --- a/arch/mips/kernel/setup.c +++ b/arch/mips/kernel/setup.c @@ -338,7 +338,7 @@ static void __init bootmem_init(void) if (end <= reserved_end) continue; #ifdef CONFIG_BLK_DEV_INITRD - /* mapstart should be after initrd_end */ + /* Skip zones before initrd and initrd itself */ if (initrd_end && end <= (unsigned long)PFN_UP(__pa(initrd_end))) continue; #endif @@ -371,6 +371,14 @@ static void __init bootmem_init(void) max_low_pfn = PFN_DOWN(HIGHMEM_START); } +#ifdef CONFIG_BLK_DEV_INITRD + /* + * mapstart should be after initrd_end + */ + if (initrd_end) + mapstart = max(mapstart, (unsigned long)PFN_UP(__pa(initrd_end))); +#endif + /* * Initialize the boot-time allocator with low memory only. */ diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c index a31896c33716..bd4385a8e6e8 100644 --- a/arch/mips/kernel/smp.c +++ b/arch/mips/kernel/smp.c @@ -42,6 +42,7 @@ #include <asm/mmu_context.h> #include <asm/time.h> #include <asm/setup.h> +#include <asm/maar.h> cpumask_t cpu_callin_map; /* Bitmask of started secondaries */ @@ -157,6 +158,7 @@ asmlinkage void start_secondary(void) mips_clockevent_init(); mp_ops->init_secondary(); cpu_report(); + maar_init(); /* * XXX parity protection should be folded in here when it's converted diff --git a/arch/mips/loongson64/common/env.c b/arch/mips/loongson64/common/env.c index f6c44dd332e2..d6d07ad56180 100644 --- a/arch/mips/loongson64/common/env.c +++ b/arch/mips/loongson64/common/env.c @@ -64,6 +64,9 @@ void __init prom_init_env(void) } if (memsize == 0) memsize = 256; + + loongson_sysconf.nr_uarts = 1; + pr_info("memsize=%u, highmemsize=%u\n", memsize, highmemsize); #else struct boot_params *boot_p; diff --git a/arch/mips/mm/dma-default.c b/arch/mips/mm/dma-default.c index a914dc1cb6d1..d8117be729a2 100644 --- a/arch/mips/mm/dma-default.c +++ b/arch/mips/mm/dma-default.c @@ -100,7 +100,7 @@ static gfp_t massage_gfp_flags(const struct device *dev, gfp_t gfp) else #endif #if defined(CONFIG_ZONE_DMA) && !defined(CONFIG_ZONE_DMA32) - if (dev->coherent_dma_mask < DMA_BIT_MASK(64)) + if (dev->coherent_dma_mask < DMA_BIT_MASK(sizeof(phys_addr_t) * 8)) dma_flag = __GFP_DMA; else #endif diff --git a/arch/mips/mm/init.c b/arch/mips/mm/init.c index 66d0f49c5bec..8770e619185e 100644 --- a/arch/mips/mm/init.c +++ b/arch/mips/mm/init.c @@ -44,6 +44,7 @@ #include <asm/pgalloc.h> #include <asm/tlb.h> #include <asm/fixmap.h> +#include <asm/maar.h> /* * We have up to 8 empty zeroed pages so we can map one of the right colour @@ -252,6 +253,119 @@ void __init fixrange_init(unsigned long start, unsigned long end, #endif } +unsigned __weak platform_maar_init(unsigned num_pairs) +{ + struct maar_config cfg[BOOT_MEM_MAP_MAX]; + unsigned i, num_configured, num_cfg = 0; + phys_addr_t skip; + + for (i = 0; i < boot_mem_map.nr_map; i++) { + switch (boot_mem_map.map[i].type) { + case BOOT_MEM_RAM: + case BOOT_MEM_INIT_RAM: + break; + default: + continue; + } + + skip = 0x10000 - (boot_mem_map.map[i].addr & 0xffff); + + cfg[num_cfg].lower = boot_mem_map.map[i].addr; + cfg[num_cfg].lower += skip; + + cfg[num_cfg].upper = cfg[num_cfg].lower; + cfg[num_cfg].upper += boot_mem_map.map[i].size - 1; + cfg[num_cfg].upper -= skip; + + cfg[num_cfg].attrs = MIPS_MAAR_S; + num_cfg++; + } + + num_configured = maar_config(cfg, num_cfg, num_pairs); + if (num_configured < num_cfg) + pr_warn("Not enough MAAR pairs (%u) for all bootmem regions (%u)\n", + num_pairs, num_cfg); + + return num_configured; +} + +void maar_init(void) +{ + unsigned num_maars, used, i; + phys_addr_t lower, upper, attr; + static struct { + struct maar_config cfgs[3]; + unsigned used; + } recorded = { { { 0 } }, 0 }; + + if (!cpu_has_maar) + return; + + /* Detect the number of MAARs */ + write_c0_maari(~0); + back_to_back_c0_hazard(); + num_maars = read_c0_maari() + 1; + + /* MAARs should be in pairs */ + WARN_ON(num_maars % 2); + + /* Set MAARs using values we recorded already */ + if (recorded.used) { + used = maar_config(recorded.cfgs, recorded.used, num_maars / 2); + BUG_ON(used != recorded.used); + } else { + /* Configure the required MAARs */ + used = platform_maar_init(num_maars / 2); + } + + /* Disable any further MAARs */ + for (i = (used * 2); i < num_maars; i++) { + write_c0_maari(i); + back_to_back_c0_hazard(); + write_c0_maar(0); + back_to_back_c0_hazard(); + } + + if (recorded.used) + return; + + pr_info("MAAR configuration:\n"); + for (i = 0; i < num_maars; i += 2) { + write_c0_maari(i); + back_to_back_c0_hazard(); + upper = read_c0_maar(); + + write_c0_maari(i + 1); + back_to_back_c0_hazard(); + lower = read_c0_maar(); + + attr = lower & upper; + lower = (lower & MIPS_MAAR_ADDR) << 4; + upper = ((upper & MIPS_MAAR_ADDR) << 4) | 0xffff; + + pr_info(" [%d]: ", i / 2); + if (!(attr & MIPS_MAAR_V)) { + pr_cont("disabled\n"); + continue; + } + + pr_cont("%pa-%pa", &lower, &upper); + + if (attr & MIPS_MAAR_S) + pr_cont(" speculate"); + + pr_cont("\n"); + + /* Record the setup for use on secondary CPUs */ + if (used <= ARRAY_SIZE(recorded.cfgs)) { + recorded.cfgs[recorded.used].lower = lower; + recorded.cfgs[recorded.used].upper = upper; + recorded.cfgs[recorded.used].attrs = attr; + recorded.used++; + } + } +} + #ifndef CONFIG_NEED_MULTIPLE_NODES int page_is_ram(unsigned long pagenr) { @@ -334,69 +448,6 @@ static inline void mem_init_free_highmem(void) #endif } -unsigned __weak platform_maar_init(unsigned num_pairs) -{ - struct maar_config cfg[BOOT_MEM_MAP_MAX]; - unsigned i, num_configured, num_cfg = 0; - phys_addr_t skip; - - for (i = 0; i < boot_mem_map.nr_map; i++) { - switch (boot_mem_map.map[i].type) { - case BOOT_MEM_RAM: - case BOOT_MEM_INIT_RAM: - break; - default: - continue; - } - - skip = 0x10000 - (boot_mem_map.map[i].addr & 0xffff); - - cfg[num_cfg].lower = boot_mem_map.map[i].addr; - cfg[num_cfg].lower += skip; - - cfg[num_cfg].upper = cfg[num_cfg].lower; - cfg[num_cfg].upper += boot_mem_map.map[i].size - 1; - cfg[num_cfg].upper -= skip; - - cfg[num_cfg].attrs = MIPS_MAAR_S; - num_cfg++; - } - - num_configured = maar_config(cfg, num_cfg, num_pairs); - if (num_configured < num_cfg) - pr_warn("Not enough MAAR pairs (%u) for all bootmem regions (%u)\n", - num_pairs, num_cfg); - - return num_configured; -} - -static void maar_init(void) -{ - unsigned num_maars, used, i; - - if (!cpu_has_maar) - return; - - /* Detect the number of MAARs */ - write_c0_maari(~0); - back_to_back_c0_hazard(); - num_maars = read_c0_maari() + 1; - - /* MAARs should be in pairs */ - WARN_ON(num_maars % 2); - - /* Configure the required MAARs */ - used = platform_maar_init(num_maars / 2); - - /* Disable any further MAARs */ - for (i = (used * 2); i < num_maars; i++) { - write_c0_maari(i); - back_to_back_c0_hazard(); - write_c0_maar(0); - back_to_back_c0_hazard(); - } -} - void __init mem_init(void) { #ifdef CONFIG_HIGHMEM diff --git a/arch/mips/net/bpf_jit_asm.S b/arch/mips/net/bpf_jit_asm.S index e92726099be0..5d2e0c8d29c0 100644 --- a/arch/mips/net/bpf_jit_asm.S +++ b/arch/mips/net/bpf_jit_asm.S @@ -57,15 +57,28 @@ LEAF(sk_load_word) is_offset_negative(word) - .globl sk_load_word_positive -sk_load_word_positive: +FEXPORT(sk_load_word_positive) is_offset_in_header(4, word) /* Offset within header boundaries */ PTR_ADDU t1, $r_skb_data, offset + .set reorder lw $r_A, 0(t1) + .set noreorder #ifdef CONFIG_CPU_LITTLE_ENDIAN +# if defined(__mips_isa_rev) && (__mips_isa_rev >= 2) wsbh t0, $r_A rotr $r_A, t0, 16 +# else + sll t0, $r_A, 24 + srl t1, $r_A, 24 + srl t2, $r_A, 8 + or t0, t0, t1 + andi t2, t2, 0xff00 + andi t1, $r_A, 0xff00 + or t0, t0, t2 + sll t1, t1, 8 + or $r_A, t0, t1 +# endif #endif jr $r_ra move $r_ret, zero @@ -73,15 +86,24 @@ sk_load_word_positive: LEAF(sk_load_half) is_offset_negative(half) - .globl sk_load_half_positive -sk_load_half_positive: +FEXPORT(sk_load_half_positive) is_offset_in_header(2, half) /* Offset within header boundaries */ PTR_ADDU t1, $r_skb_data, offset + .set reorder lh $r_A, 0(t1) + .set noreorder #ifdef CONFIG_CPU_LITTLE_ENDIAN +# if defined(__mips_isa_rev) && (__mips_isa_rev >= 2) wsbh t0, $r_A seh $r_A, t0 +# else + sll t0, $r_A, 24 + andi t1, $r_A, 0xff00 + sra t0, t0, 16 + srl t1, t1, 8 + or $r_A, t0, t1 +# endif #endif jr $r_ra move $r_ret, zero @@ -89,8 +111,7 @@ sk_load_half_positive: LEAF(sk_load_byte) is_offset_negative(byte) - .globl sk_load_byte_positive -sk_load_byte_positive: +FEXPORT(sk_load_byte_positive) is_offset_in_header(1, byte) /* Offset within header boundaries */ PTR_ADDU t1, $r_skb_data, offset @@ -148,23 +169,47 @@ sk_load_byte_positive: NESTED(bpf_slow_path_word, (6 * SZREG), $r_sp) bpf_slow_path_common(4) #ifdef CONFIG_CPU_LITTLE_ENDIAN +# if defined(__mips_isa_rev) && (__mips_isa_rev >= 2) wsbh t0, $r_s0 jr $r_ra rotr $r_A, t0, 16 -#endif +# else + sll t0, $r_s0, 24 + srl t1, $r_s0, 24 + srl t2, $r_s0, 8 + or t0, t0, t1 + andi t2, t2, 0xff00 + andi t1, $r_s0, 0xff00 + or t0, t0, t2 + sll t1, t1, 8 + jr $r_ra + or $r_A, t0, t1 +# endif +#else jr $r_ra - move $r_A, $r_s0 + move $r_A, $r_s0 +#endif END(bpf_slow_path_word) NESTED(bpf_slow_path_half, (6 * SZREG), $r_sp) bpf_slow_path_common(2) #ifdef CONFIG_CPU_LITTLE_ENDIAN +# if defined(__mips_isa_rev) && (__mips_isa_rev >= 2) jr $r_ra wsbh $r_A, $r_s0 -#endif +# else + sll t0, $r_s0, 8 + andi t1, $r_s0, 0xff00 + andi t0, t0, 0xff00 + srl t1, t1, 8 + jr $r_ra + or $r_A, t0, t1 +# endif +#else jr $r_ra move $r_A, $r_s0 +#endif END(bpf_slow_path_half) diff --git a/arch/mn10300/include/asm/Kbuild b/arch/mn10300/include/asm/Kbuild index 6edb9ee6128e..1c8dd0f5cd5d 100644 --- a/arch/mn10300/include/asm/Kbuild +++ b/arch/mn10300/include/asm/Kbuild @@ -9,3 +9,4 @@ generic-y += mm-arch-hooks.h generic-y += preempt.h generic-y += sections.h generic-y += trace_clock.h +generic-y += word-at-a-time.h diff --git a/arch/nios2/include/asm/Kbuild b/arch/nios2/include/asm/Kbuild index 914864eb5a25..d63330e88379 100644 --- a/arch/nios2/include/asm/Kbuild +++ b/arch/nios2/include/asm/Kbuild @@ -61,4 +61,5 @@ generic-y += types.h generic-y += unaligned.h generic-y += user.h generic-y += vga.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/powerpc/include/asm/word-at-a-time.h b/arch/powerpc/include/asm/word-at-a-time.h index 5b3a903adae6..e4396a7d0f7c 100644 --- a/arch/powerpc/include/asm/word-at-a-time.h +++ b/arch/powerpc/include/asm/word-at-a-time.h @@ -40,6 +40,11 @@ static inline bool has_zero(unsigned long val, unsigned long *data, const struct return (val + c->high_bits) & ~rhs; } +static inline unsigned long zero_bytemask(unsigned long mask) +{ + return ~1ul << __fls(mask); +} + #else #ifdef CONFIG_64BIT diff --git a/arch/s390/boot/compressed/Makefile b/arch/s390/boot/compressed/Makefile index d4788111c161..fac6ac9790fa 100644 --- a/arch/s390/boot/compressed/Makefile +++ b/arch/s390/boot/compressed/Makefile @@ -10,7 +10,7 @@ targets += misc.o piggy.o sizes.h head.o KBUILD_CFLAGS := -m64 -D__KERNEL__ $(LINUX_INCLUDE) -O2 KBUILD_CFLAGS += -DDISABLE_BRANCH_PROFILING -KBUILD_CFLAGS += $(cflags-y) -fno-delete-null-pointer-checks +KBUILD_CFLAGS += $(cflags-y) -fno-delete-null-pointer-checks -msoft-float KBUILD_CFLAGS += $(call cc-option,-mpacked-stack) KBUILD_CFLAGS += $(call cc-option,-ffreestanding) diff --git a/arch/s390/configs/default_defconfig b/arch/s390/configs/default_defconfig index 0c98f1508542..ed7da281df66 100644 --- a/arch/s390/configs/default_defconfig +++ b/arch/s390/configs/default_defconfig @@ -381,7 +381,7 @@ CONFIG_ISCSI_TCP=m CONFIG_SCSI_DEBUG=m CONFIG_ZFCP=y CONFIG_SCSI_VIRTIO=m -CONFIG_SCSI_DH=m +CONFIG_SCSI_DH=y CONFIG_SCSI_DH_RDAC=m CONFIG_SCSI_DH_HP_SW=m CONFIG_SCSI_DH_EMC=m diff --git a/arch/s390/configs/gcov_defconfig b/arch/s390/configs/gcov_defconfig index 82083e1fbdc4..9858b14cde1e 100644 --- a/arch/s390/configs/gcov_defconfig +++ b/arch/s390/configs/gcov_defconfig @@ -377,7 +377,7 @@ CONFIG_ISCSI_TCP=m CONFIG_SCSI_DEBUG=m CONFIG_ZFCP=y CONFIG_SCSI_VIRTIO=m -CONFIG_SCSI_DH=m +CONFIG_SCSI_DH=y CONFIG_SCSI_DH_RDAC=m CONFIG_SCSI_DH_HP_SW=m CONFIG_SCSI_DH_EMC=m diff --git a/arch/s390/configs/performance_defconfig b/arch/s390/configs/performance_defconfig index c05c9e0821e3..7f14f80717d4 100644 --- a/arch/s390/configs/performance_defconfig +++ b/arch/s390/configs/performance_defconfig @@ -377,7 +377,7 @@ CONFIG_ISCSI_TCP=m CONFIG_SCSI_DEBUG=m CONFIG_ZFCP=y CONFIG_SCSI_VIRTIO=m -CONFIG_SCSI_DH=m +CONFIG_SCSI_DH=y CONFIG_SCSI_DH_RDAC=m CONFIG_SCSI_DH_HP_SW=m CONFIG_SCSI_DH_EMC=m diff --git a/arch/s390/include/asm/Kbuild b/arch/s390/include/asm/Kbuild index 5ad26dd94d77..9043d2e1e2ae 100644 --- a/arch/s390/include/asm/Kbuild +++ b/arch/s390/include/asm/Kbuild @@ -6,3 +6,4 @@ generic-y += mcs_spinlock.h generic-y += mm-arch-hooks.h generic-y += preempt.h generic-y += trace_clock.h +generic-y += word-at-a-time.h diff --git a/arch/s390/include/asm/numa.h b/arch/s390/include/asm/numa.h index 2a0efc63b9e5..dc19ee0c92aa 100644 --- a/arch/s390/include/asm/numa.h +++ b/arch/s390/include/asm/numa.h @@ -19,7 +19,7 @@ int numa_pfn_to_nid(unsigned long pfn); int __node_distance(int a, int b); void numa_update_cpu_topology(void); -extern cpumask_var_t node_to_cpumask_map[MAX_NUMNODES]; +extern cpumask_t node_to_cpumask_map[MAX_NUMNODES]; extern int numa_debug_enabled; #else diff --git a/arch/s390/include/asm/topology.h b/arch/s390/include/asm/topology.h index 27ebde643933..94fc55fc72ce 100644 --- a/arch/s390/include/asm/topology.h +++ b/arch/s390/include/asm/topology.h @@ -68,7 +68,7 @@ static inline int cpu_to_node(int cpu) #define cpumask_of_node cpumask_of_node static inline const struct cpumask *cpumask_of_node(int node) { - return node_to_cpumask_map[node]; + return &node_to_cpumask_map[node]; } /* diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c index 48c9af7a7683..3aeeb1b562c0 100644 --- a/arch/s390/kernel/asm-offsets.c +++ b/arch/s390/kernel/asm-offsets.c @@ -176,6 +176,7 @@ int main(void) DEFINE(__LC_PASTE, offsetof(struct _lowcore, paste)); DEFINE(__LC_FP_CREG_SAVE_AREA, offsetof(struct _lowcore, fpt_creg_save_area)); DEFINE(__LC_LAST_BREAK, offsetof(struct _lowcore, breaking_event_addr)); + DEFINE(__LC_PERCPU_OFFSET, offsetof(struct _lowcore, percpu_offset)); DEFINE(__LC_VDSO_PER_CPU, offsetof(struct _lowcore, vdso_per_cpu_data)); DEFINE(__LC_GMAP, offsetof(struct _lowcore, gmap)); DEFINE(__LC_PGM_TDB, offsetof(struct _lowcore, pgm_tdb)); diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index 09b039d7983d..582fe44ab07c 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -733,6 +733,14 @@ ENTRY(psw_idle) stg %r3,__SF_EMPTY(%r15) larl %r1,.Lpsw_idle_lpsw+4 stg %r1,__SF_EMPTY+8(%r15) +#ifdef CONFIG_SMP + larl %r1,smp_cpu_mtid + llgf %r1,0(%r1) + ltgr %r1,%r1 + jz .Lpsw_idle_stcctm + .insn rsy,0xeb0000000017,%r1,5,__SF_EMPTY+16(%r15) +.Lpsw_idle_stcctm: +#endif STCK __CLOCK_IDLE_ENTER(%r2) stpt __TIMER_IDLE_ENTER(%r2) .Lpsw_idle_lpsw: @@ -1159,7 +1167,27 @@ cleanup_critical: jhe 1f mvc __CLOCK_IDLE_ENTER(8,%r2),__CLOCK_IDLE_EXIT(%r2) mvc __TIMER_IDLE_ENTER(8,%r2),__TIMER_IDLE_EXIT(%r2) -1: # account system time going idle +1: # calculate idle cycles +#ifdef CONFIG_SMP + clg %r9,BASED(.Lcleanup_idle_insn) + jl 3f + larl %r1,smp_cpu_mtid + llgf %r1,0(%r1) + ltgr %r1,%r1 + jz 3f + .insn rsy,0xeb0000000017,%r1,5,__SF_EMPTY+80(%r15) + larl %r3,mt_cycles + ag %r3,__LC_PERCPU_OFFSET + la %r4,__SF_EMPTY+16(%r15) +2: lg %r0,0(%r3) + slg %r0,0(%r4) + alg %r0,64(%r4) + stg %r0,0(%r3) + la %r3,8(%r3) + la %r4,8(%r4) + brct %r1,2b +#endif +3: # account system time going idle lg %r9,__LC_STEAL_TIMER alg %r9,__CLOCK_IDLE_ENTER(%r2) slg %r9,__LC_LAST_UPDATE_CLOCK diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c index c8653435c70d..dafc44f519c3 100644 --- a/arch/s390/kernel/vtime.c +++ b/arch/s390/kernel/vtime.c @@ -25,7 +25,7 @@ static DEFINE_SPINLOCK(virt_timer_lock); static atomic64_t virt_timer_current; static atomic64_t virt_timer_elapsed; -static DEFINE_PER_CPU(u64, mt_cycles[32]); +DEFINE_PER_CPU(u64, mt_cycles[8]); static DEFINE_PER_CPU(u64, mt_scaling_mult) = { 1 }; static DEFINE_PER_CPU(u64, mt_scaling_div) = { 1 }; static DEFINE_PER_CPU(u64, mt_scaling_jiffies); @@ -60,6 +60,34 @@ static inline int virt_timer_forward(u64 elapsed) return elapsed >= atomic64_read(&virt_timer_current); } +static void update_mt_scaling(void) +{ + u64 cycles_new[8], *cycles_old; + u64 delta, fac, mult, div; + int i; + + stcctm5(smp_cpu_mtid + 1, cycles_new); + cycles_old = this_cpu_ptr(mt_cycles); + fac = 1; + mult = div = 0; + for (i = 0; i <= smp_cpu_mtid; i++) { + delta = cycles_new[i] - cycles_old[i]; + div += delta; + mult *= i + 1; + mult += delta * fac; + fac *= i + 1; + } + div *= fac; + if (div > 0) { + /* Update scaling factor */ + __this_cpu_write(mt_scaling_mult, mult); + __this_cpu_write(mt_scaling_div, div); + memcpy(cycles_old, cycles_new, + sizeof(u64) * (smp_cpu_mtid + 1)); + } + __this_cpu_write(mt_scaling_jiffies, jiffies_64); +} + /* * Update process times based on virtual cpu times stored by entry.S * to the lowcore fields user_timer, system_timer & steal_clock. @@ -69,7 +97,6 @@ static int do_account_vtime(struct task_struct *tsk, int hardirq_offset) struct thread_info *ti = task_thread_info(tsk); u64 timer, clock, user, system, steal; u64 user_scaled, system_scaled; - int i; timer = S390_lowcore.last_update_timer; clock = S390_lowcore.last_update_clock; @@ -85,34 +112,10 @@ static int do_account_vtime(struct task_struct *tsk, int hardirq_offset) S390_lowcore.system_timer += timer - S390_lowcore.last_update_timer; S390_lowcore.steal_timer += S390_lowcore.last_update_clock - clock; - /* Do MT utilization calculation */ + /* Update MT utilization calculation */ if (smp_cpu_mtid && - time_after64(jiffies_64, __this_cpu_read(mt_scaling_jiffies))) { - u64 cycles_new[32], *cycles_old; - u64 delta, fac, mult, div; - - cycles_old = this_cpu_ptr(mt_cycles); - if (stcctm5(smp_cpu_mtid + 1, cycles_new) < 2) { - fac = 1; - mult = div = 0; - for (i = 0; i <= smp_cpu_mtid; i++) { - delta = cycles_new[i] - cycles_old[i]; - div += delta; - mult *= i + 1; - mult += delta * fac; - fac *= i + 1; - } - div *= fac; - if (div > 0) { - /* Update scaling factor */ - __this_cpu_write(mt_scaling_mult, mult); - __this_cpu_write(mt_scaling_div, div); - memcpy(cycles_old, cycles_new, - sizeof(u64) * (smp_cpu_mtid + 1)); - } - } - __this_cpu_write(mt_scaling_jiffies, jiffies_64); - } + time_after64(jiffies_64, this_cpu_read(mt_scaling_jiffies))) + update_mt_scaling(); user = S390_lowcore.user_timer - ti->user_timer; S390_lowcore.steal_timer -= user; @@ -181,6 +184,11 @@ void vtime_account_irq_enter(struct task_struct *tsk) S390_lowcore.last_update_timer = get_vtimer(); S390_lowcore.system_timer += timer - S390_lowcore.last_update_timer; + /* Update MT utilization calculation */ + if (smp_cpu_mtid && + time_after64(jiffies_64, this_cpu_read(mt_scaling_jiffies))) + update_mt_scaling(); + system = S390_lowcore.system_timer - ti->system_timer; S390_lowcore.steal_timer -= system; ti->system_timer = S390_lowcore.system_timer; diff --git a/arch/s390/numa/mode_emu.c b/arch/s390/numa/mode_emu.c index 7de4e2f780d7..30b2698a28e2 100644 --- a/arch/s390/numa/mode_emu.c +++ b/arch/s390/numa/mode_emu.c @@ -368,7 +368,7 @@ static void topology_add_core(struct toptree *core) cpumask_copy(&top->thread_mask, &core->mask); cpumask_copy(&top->core_mask, &core_mc(core)->mask); cpumask_copy(&top->book_mask, &core_book(core)->mask); - cpumask_set_cpu(cpu, node_to_cpumask_map[core_node(core)->id]); + cpumask_set_cpu(cpu, &node_to_cpumask_map[core_node(core)->id]); top->node_id = core_node(core)->id; } } @@ -383,7 +383,7 @@ static void toptree_to_topology(struct toptree *numa) /* Clear all node masks */ for (i = 0; i < MAX_NUMNODES; i++) - cpumask_clear(node_to_cpumask_map[i]); + cpumask_clear(&node_to_cpumask_map[i]); /* Rebuild all masks */ toptree_for_each(core, numa, CORE) diff --git a/arch/s390/numa/numa.c b/arch/s390/numa/numa.c index 09b1d2355bd9..43f32ce60aa3 100644 --- a/arch/s390/numa/numa.c +++ b/arch/s390/numa/numa.c @@ -23,7 +23,7 @@ pg_data_t *node_data[MAX_NUMNODES]; EXPORT_SYMBOL(node_data); -cpumask_var_t node_to_cpumask_map[MAX_NUMNODES]; +cpumask_t node_to_cpumask_map[MAX_NUMNODES]; EXPORT_SYMBOL(node_to_cpumask_map); const struct numa_mode numa_mode_plain = { @@ -144,7 +144,7 @@ void __init numa_setup(void) static int __init numa_init_early(void) { /* Attach all possible CPUs to node 0 for now. */ - cpumask_copy(node_to_cpumask_map[0], cpu_possible_mask); + cpumask_copy(&node_to_cpumask_map[0], cpu_possible_mask); return 0; } early_initcall(numa_init_early); diff --git a/arch/score/include/asm/Kbuild b/arch/score/include/asm/Kbuild index 92ffe397b893..a05218ff3fe4 100644 --- a/arch/score/include/asm/Kbuild +++ b/arch/score/include/asm/Kbuild @@ -13,3 +13,4 @@ generic-y += sections.h generic-y += trace_clock.h generic-y += xor.h generic-y += serial.h +generic-y += word-at-a-time.h diff --git a/arch/tile/gxio/mpipe.c b/arch/tile/gxio/mpipe.c index ee186e13dfe6..f102048d9c0e 100644 --- a/arch/tile/gxio/mpipe.c +++ b/arch/tile/gxio/mpipe.c @@ -19,6 +19,7 @@ #include <linux/errno.h> #include <linux/io.h> #include <linux/module.h> +#include <linux/string.h> #include <gxio/iorpc_globals.h> #include <gxio/iorpc_mpipe.h> @@ -29,32 +30,6 @@ /* HACK: Avoid pointless "shadow" warnings. */ #define link link_shadow -/** - * strscpy - Copy a C-string into a sized buffer, but only if it fits - * @dest: Where to copy the string to - * @src: Where to copy the string from - * @size: size of destination buffer - * - * Use this routine to avoid copying too-long strings. - * The routine returns the total number of bytes copied - * (including the trailing NUL) or zero if the buffer wasn't - * big enough. To ensure that programmers pay attention - * to the return code, the destination has a single NUL - * written at the front (if size is non-zero) when the - * buffer is not big enough. - */ -static size_t strscpy(char *dest, const char *src, size_t size) -{ - size_t len = strnlen(src, size) + 1; - if (len > size) { - if (size) - dest[0] = '\0'; - return 0; - } - memcpy(dest, src, len); - return len; -} - int gxio_mpipe_init(gxio_mpipe_context_t *context, unsigned int mpipe_index) { char file[32]; @@ -540,7 +515,7 @@ int gxio_mpipe_link_instance(const char *link_name) if (!context) return GXIO_ERR_NO_DEVICE; - if (strscpy(name.name, link_name, sizeof(name.name)) == 0) + if (strscpy(name.name, link_name, sizeof(name.name)) < 0) return GXIO_ERR_NO_DEVICE; return gxio_mpipe_info_instance_aux(context, name); @@ -559,7 +534,7 @@ int gxio_mpipe_link_enumerate_mac(int idx, char *link_name, uint8_t *link_mac) rv = gxio_mpipe_info_enumerate_aux(context, idx, &name, &mac); if (rv >= 0) { - if (strscpy(link_name, name.name, sizeof(name.name)) == 0) + if (strscpy(link_name, name.name, sizeof(name.name)) < 0) return GXIO_ERR_INVAL_MEMORY_SIZE; memcpy(link_mac, mac.mac, sizeof(mac.mac)); } @@ -576,7 +551,7 @@ int gxio_mpipe_link_open(gxio_mpipe_link_t *link, _gxio_mpipe_link_name_t name; int rv; - if (strscpy(name.name, link_name, sizeof(name.name)) == 0) + if (strscpy(name.name, link_name, sizeof(name.name)) < 0) return GXIO_ERR_NO_DEVICE; rv = gxio_mpipe_link_open_aux(context, name, flags); diff --git a/arch/tile/include/asm/word-at-a-time.h b/arch/tile/include/asm/word-at-a-time.h index 9e5ce0d7b292..b66a693c2c34 100644 --- a/arch/tile/include/asm/word-at-a-time.h +++ b/arch/tile/include/asm/word-at-a-time.h @@ -6,7 +6,7 @@ struct word_at_a_time { /* unused */ }; #define WORD_AT_A_TIME_CONSTANTS {} -/* Generate 0x01 byte values for non-zero bytes using a SIMD instruction. */ +/* Generate 0x01 byte values for zero bytes using a SIMD instruction. */ static inline unsigned long has_zero(unsigned long val, unsigned long *data, const struct word_at_a_time *c) { @@ -33,4 +33,10 @@ static inline long find_zero(unsigned long mask) #endif } +#ifdef __BIG_ENDIAN +#define zero_bytemask(mask) (~1ul << (63 - __builtin_clzl(mask))) +#else +#define zero_bytemask(mask) ((2ul << __builtin_ctzl(mask)) - 1) +#endif + #endif /* _ASM_WORD_AT_A_TIME_H */ diff --git a/arch/tile/kernel/usb.c b/arch/tile/kernel/usb.c index f0da5a237e94..9f1e05e12255 100644 --- a/arch/tile/kernel/usb.c +++ b/arch/tile/kernel/usb.c @@ -22,6 +22,7 @@ #include <linux/platform_device.h> #include <linux/usb/tilegx.h> #include <linux/init.h> +#include <linux/module.h> #include <linux/types.h> static u64 ehci_dmamask = DMA_BIT_MASK(32); diff --git a/arch/um/include/asm/Kbuild b/arch/um/include/asm/Kbuild index 149ec55f9c46..904f3ebf4220 100644 --- a/arch/um/include/asm/Kbuild +++ b/arch/um/include/asm/Kbuild @@ -25,4 +25,5 @@ generic-y += preempt.h generic-y += switch_to.h generic-y += topology.h generic-y += trace_clock.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/unicore32/include/asm/Kbuild b/arch/unicore32/include/asm/Kbuild index 1fc7a286dc6f..256c45b3ae34 100644 --- a/arch/unicore32/include/asm/Kbuild +++ b/arch/unicore32/include/asm/Kbuild @@ -62,4 +62,5 @@ generic-y += ucontext.h generic-y += unaligned.h generic-y += user.h generic-y += vga.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 328c8352480c..96d058a87100 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -1308,6 +1308,7 @@ config HIGHMEM config X86_PAE bool "PAE (Physical Address Extension) Support" depends on X86_32 && !HIGHMEM4G + select SWIOTLB ---help--- PAE is required for NX support, and furthermore enables larger swapspace support for non-overcommit purposes. It diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index e6cf2ad350d1..9727b3b48bd1 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -193,7 +193,7 @@ #define X86_FEATURE_HW_PSTATE ( 7*32+ 8) /* AMD HW-PState */ #define X86_FEATURE_PROC_FEEDBACK ( 7*32+ 9) /* AMD ProcFeedbackInterface */ #define X86_FEATURE_HWP ( 7*32+ 10) /* "hwp" Intel HWP */ -#define X86_FEATURE_HWP_NOITFY ( 7*32+ 11) /* Intel HWP_NOTIFY */ +#define X86_FEATURE_HWP_NOTIFY ( 7*32+ 11) /* Intel HWP_NOTIFY */ #define X86_FEATURE_HWP_ACT_WINDOW ( 7*32+ 12) /* Intel HWP_ACT_WINDOW */ #define X86_FEATURE_HWP_EPP ( 7*32+13) /* Intel HWP_EPP */ #define X86_FEATURE_HWP_PKG_REQ ( 7*32+14) /* Intel HWP_PKG_REQ */ diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index ab5f1d447ef9..ae68be92f755 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -86,6 +86,7 @@ extern u64 asmlinkage efi_call(void *fp, ...); extern void __iomem *__init efi_ioremap(unsigned long addr, unsigned long size, u32 type, u64 attribute); +#ifdef CONFIG_KASAN /* * CONFIG_KASAN may redefine memset to __memset. __memset function is present * only in kernel binary. Since the EFI stub linked into a separate binary it @@ -95,6 +96,7 @@ extern void __iomem *__init efi_ioremap(unsigned long addr, unsigned long size, #undef memcpy #undef memset #undef memmove +#endif #endif /* CONFIG_X86_32 */ diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index b98b471a3b7e..b8c14bb7fc8f 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -141,6 +141,8 @@ #define DEBUGCTLMSR_BTS_OFF_USR (1UL << 10) #define DEBUGCTLMSR_FREEZE_LBRS_ON_PMI (1UL << 11) +#define MSR_PEBS_FRONTEND 0x000003f7 + #define MSR_IA32_POWER_CTL 0x000001fc #define MSR_IA32_MC0_CTL 0x00000400 diff --git a/arch/x86/include/asm/pvclock-abi.h b/arch/x86/include/asm/pvclock-abi.h index 655e07a48f6c..67f08230103a 100644 --- a/arch/x86/include/asm/pvclock-abi.h +++ b/arch/x86/include/asm/pvclock-abi.h @@ -41,6 +41,7 @@ struct pvclock_wall_clock { #define PVCLOCK_TSC_STABLE_BIT (1 << 0) #define PVCLOCK_GUEST_STOPPED (1 << 1) +/* PVCLOCK_COUNTS_FROM_ZERO broke ABI and can't be used anymore. */ #define PVCLOCK_COUNTS_FROM_ZERO (1 << 2) #endif /* __ASSEMBLY__ */ #endif /* _ASM_X86_PVCLOCK_ABI_H */ diff --git a/arch/x86/include/asm/xen/hypercall.h b/arch/x86/include/asm/xen/hypercall.h index 83aea8055119..4c20dd333412 100644 --- a/arch/x86/include/asm/xen/hypercall.h +++ b/arch/x86/include/asm/xen/hypercall.h @@ -336,10 +336,10 @@ HYPERVISOR_update_descriptor(u64 ma, u64 desc) return _hypercall4(int, update_descriptor, ma, ma>>32, desc, desc>>32); } -static inline int +static inline long HYPERVISOR_memory_op(unsigned int cmd, void *arg) { - return _hypercall2(int, memory_op, cmd, arg); + return _hypercall2(long, memory_op, cmd, arg); } static inline int diff --git a/arch/x86/include/uapi/asm/bitsperlong.h b/arch/x86/include/uapi/asm/bitsperlong.h index b0ae1c4dc791..217909b4d6f5 100644 --- a/arch/x86/include/uapi/asm/bitsperlong.h +++ b/arch/x86/include/uapi/asm/bitsperlong.h @@ -1,7 +1,7 @@ #ifndef __ASM_X86_BITSPERLONG_H #define __ASM_X86_BITSPERLONG_H -#ifdef __x86_64__ +#if defined(__x86_64__) && !defined(__ILP32__) # define __BITS_PER_LONG 64 #else # define __BITS_PER_LONG 32 diff --git a/arch/x86/kernel/cpu/mshyperv.c b/arch/x86/kernel/cpu/mshyperv.c index 381c8b9b3a33..20e242ea1bc4 100644 --- a/arch/x86/kernel/cpu/mshyperv.c +++ b/arch/x86/kernel/cpu/mshyperv.c @@ -34,11 +34,10 @@ struct ms_hyperv_info ms_hyperv; EXPORT_SYMBOL_GPL(ms_hyperv); -static void (*hv_kexec_handler)(void); -static void (*hv_crash_handler)(struct pt_regs *regs); - #if IS_ENABLED(CONFIG_HYPERV) static void (*vmbus_handler)(void); +static void (*hv_kexec_handler)(void); +static void (*hv_crash_handler)(struct pt_regs *regs); void hyperv_vector_handler(struct pt_regs *regs) { @@ -96,8 +95,8 @@ void hv_remove_crash_handler(void) hv_crash_handler = NULL; } EXPORT_SYMBOL_GPL(hv_remove_crash_handler); -#endif +#ifdef CONFIG_KEXEC_CORE static void hv_machine_shutdown(void) { if (kexec_in_progress && hv_kexec_handler) @@ -111,7 +110,8 @@ static void hv_machine_crash_shutdown(struct pt_regs *regs) hv_crash_handler(regs); native_machine_crash_shutdown(regs); } - +#endif /* CONFIG_KEXEC_CORE */ +#endif /* CONFIG_HYPERV */ static uint32_t __init ms_hyperv_platform(void) { @@ -186,8 +186,10 @@ static void __init ms_hyperv_init_platform(void) no_timer_check = 1; #endif +#if IS_ENABLED(CONFIG_HYPERV) && defined(CONFIG_KEXEC_CORE) machine_ops.shutdown = hv_machine_shutdown; machine_ops.crash_shutdown = hv_machine_crash_shutdown; +#endif mark_tsc_unstable("running on Hyper-V"); } diff --git a/arch/x86/kernel/cpu/perf_event.h b/arch/x86/kernel/cpu/perf_event.h index 5edf6d868fc1..165be83a7fa4 100644 --- a/arch/x86/kernel/cpu/perf_event.h +++ b/arch/x86/kernel/cpu/perf_event.h @@ -47,6 +47,7 @@ enum extra_reg_type { EXTRA_REG_RSP_1 = 1, /* offcore_response_1 */ EXTRA_REG_LBR = 2, /* lbr_select */ EXTRA_REG_LDLAT = 3, /* ld_lat_threshold */ + EXTRA_REG_FE = 4, /* fe_* */ EXTRA_REG_MAX /* number of entries needed */ }; diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index 3fefebfbdf4b..f63360be2238 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -205,6 +205,11 @@ static struct extra_reg intel_skl_extra_regs[] __read_mostly = { INTEL_UEVENT_EXTRA_REG(0x01b7, MSR_OFFCORE_RSP_0, 0x3fffff8fffull, RSP_0), INTEL_UEVENT_EXTRA_REG(0x01bb, MSR_OFFCORE_RSP_1, 0x3fffff8fffull, RSP_1), INTEL_UEVENT_PEBS_LDLAT_EXTRA_REG(0x01cd), + /* + * Note the low 8 bits eventsel code is not a continuous field, containing + * some #GPing bits. These are masked out. + */ + INTEL_UEVENT_EXTRA_REG(0x01c6, MSR_PEBS_FRONTEND, 0x7fff17, FE), EVENT_EXTRA_END }; @@ -250,7 +255,7 @@ struct event_constraint intel_bdw_event_constraints[] = { FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ FIXED_EVENT_CONSTRAINT(0x0300, 2), /* CPU_CLK_UNHALTED.REF */ INTEL_UEVENT_CONSTRAINT(0x148, 0x4), /* L1D_PEND_MISS.PENDING */ - INTEL_EVENT_CONSTRAINT(0xa3, 0x4), /* CYCLE_ACTIVITY.* */ + INTEL_UEVENT_CONSTRAINT(0x8a3, 0x4), /* CYCLE_ACTIVITY.CYCLES_L1D_MISS */ EVENT_CONSTRAINT_END }; @@ -2891,6 +2896,8 @@ PMU_FORMAT_ATTR(offcore_rsp, "config1:0-63"); PMU_FORMAT_ATTR(ldlat, "config1:0-15"); +PMU_FORMAT_ATTR(frontend, "config1:0-23"); + static struct attribute *intel_arch3_formats_attr[] = { &format_attr_event.attr, &format_attr_umask.attr, @@ -2907,6 +2914,11 @@ static struct attribute *intel_arch3_formats_attr[] = { NULL, }; +static struct attribute *skl_format_attr[] = { + &format_attr_frontend.attr, + NULL, +}; + static __initconst const struct x86_pmu core_pmu = { .name = "core", .handle_irq = x86_pmu_handle_irq, @@ -3516,7 +3528,8 @@ __init int intel_pmu_init(void) x86_pmu.hw_config = hsw_hw_config; x86_pmu.get_event_constraints = hsw_get_event_constraints; - x86_pmu.cpu_events = hsw_events_attrs; + x86_pmu.format_attrs = merge_attr(intel_arch3_formats_attr, + skl_format_attr); WARN_ON(!x86_pmu.format_attrs); x86_pmu.cpu_events = hsw_events_attrs; pr_cont("Skylake events, "); diff --git a/arch/x86/kernel/cpu/perf_event_msr.c b/arch/x86/kernel/cpu/perf_event_msr.c index 086b12eae794..f32ac13934f2 100644 --- a/arch/x86/kernel/cpu/perf_event_msr.c +++ b/arch/x86/kernel/cpu/perf_event_msr.c @@ -10,12 +10,12 @@ enum perf_msr_id { PERF_MSR_EVENT_MAX, }; -bool test_aperfmperf(int idx) +static bool test_aperfmperf(int idx) { return boot_cpu_has(X86_FEATURE_APERFMPERF); } -bool test_intel(int idx) +static bool test_intel(int idx) { if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL || boot_cpu_data.x86 != 6) diff --git a/arch/x86/kernel/cpu/scattered.c b/arch/x86/kernel/cpu/scattered.c index 3d423a101fae..608fb26c7254 100644 --- a/arch/x86/kernel/cpu/scattered.c +++ b/arch/x86/kernel/cpu/scattered.c @@ -37,7 +37,7 @@ void init_scattered_cpuid_features(struct cpuinfo_x86 *c) { X86_FEATURE_PLN, CR_EAX, 4, 0x00000006, 0 }, { X86_FEATURE_PTS, CR_EAX, 6, 0x00000006, 0 }, { X86_FEATURE_HWP, CR_EAX, 7, 0x00000006, 0 }, - { X86_FEATURE_HWP_NOITFY, CR_EAX, 8, 0x00000006, 0 }, + { X86_FEATURE_HWP_NOTIFY, CR_EAX, 8, 0x00000006, 0 }, { X86_FEATURE_HWP_ACT_WINDOW, CR_EAX, 9, 0x00000006, 0 }, { X86_FEATURE_HWP_EPP, CR_EAX,10, 0x00000006, 0 }, { X86_FEATURE_HWP_PKG_REQ, CR_EAX,11, 0x00000006, 0 }, diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c index e068d6683dba..74ca2fe7a0b3 100644 --- a/arch/x86/kernel/crash.c +++ b/arch/x86/kernel/crash.c @@ -185,10 +185,9 @@ void native_machine_crash_shutdown(struct pt_regs *regs) } #ifdef CONFIG_KEXEC_FILE -static int get_nr_ram_ranges_callback(unsigned long start_pfn, - unsigned long nr_pfn, void *arg) +static int get_nr_ram_ranges_callback(u64 start, u64 end, void *arg) { - int *nr_ranges = arg; + unsigned int *nr_ranges = arg; (*nr_ranges)++; return 0; @@ -214,7 +213,7 @@ static void fill_up_crash_elf_data(struct crash_elf_data *ced, ced->image = image; - walk_system_ram_range(0, -1, &nr_ranges, + walk_system_ram_res(0, -1, &nr_ranges, get_nr_ram_ranges_callback); ced->max_nr_ranges = nr_ranges; diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 6d0e62ae8516..39e585a554b7 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -506,3 +506,58 @@ unsigned long arch_randomize_brk(struct mm_struct *mm) return randomize_range(mm->brk, range_end, 0) ? : mm->brk; } +/* + * Called from fs/proc with a reference on @p to find the function + * which called into schedule(). This needs to be done carefully + * because the task might wake up and we might look at a stack + * changing under us. + */ +unsigned long get_wchan(struct task_struct *p) +{ + unsigned long start, bottom, top, sp, fp, ip; + int count = 0; + + if (!p || p == current || p->state == TASK_RUNNING) + return 0; + + start = (unsigned long)task_stack_page(p); + if (!start) + return 0; + + /* + * Layout of the stack page: + * + * ----------- topmax = start + THREAD_SIZE - sizeof(unsigned long) + * PADDING + * ----------- top = topmax - TOP_OF_KERNEL_STACK_PADDING + * stack + * ----------- bottom = start + sizeof(thread_info) + * thread_info + * ----------- start + * + * The tasks stack pointer points at the location where the + * framepointer is stored. The data on the stack is: + * ... IP FP ... IP FP + * + * We need to read FP and IP, so we need to adjust the upper + * bound by another unsigned long. + */ + top = start + THREAD_SIZE - TOP_OF_KERNEL_STACK_PADDING; + top -= 2 * sizeof(unsigned long); + bottom = start + sizeof(struct thread_info); + + sp = READ_ONCE(p->thread.sp); + if (sp < bottom || sp > top) + return 0; + + fp = READ_ONCE(*(unsigned long *)sp); + do { + if (fp < bottom || fp > top) + return 0; + ip = READ_ONCE(*(unsigned long *)(fp + sizeof(unsigned long))); + if (!in_sched_functions(ip)) + return ip; + fp = READ_ONCE(*(unsigned long *)fp); + } while (count++ < 16 && p->state != TASK_RUNNING); + return 0; +} diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index c13df2c735f8..737527b40e5b 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -324,31 +324,3 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) return prev_p; } - -#define top_esp (THREAD_SIZE - sizeof(unsigned long)) -#define top_ebp (THREAD_SIZE - 2*sizeof(unsigned long)) - -unsigned long get_wchan(struct task_struct *p) -{ - unsigned long bp, sp, ip; - unsigned long stack_page; - int count = 0; - if (!p || p == current || p->state == TASK_RUNNING) - return 0; - stack_page = (unsigned long)task_stack_page(p); - sp = p->thread.sp; - if (!stack_page || sp < stack_page || sp > top_esp+stack_page) - return 0; - /* include/asm-i386/system.h:switch_to() pushes bp last. */ - bp = *(unsigned long *) sp; - do { - if (bp < stack_page || bp > top_ebp+stack_page) - return 0; - ip = *(unsigned long *) (bp+4); - if (!in_sched_functions(ip)) - return ip; - bp = *(unsigned long *) bp; - } while (count++ < 16); - return 0; -} - diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 3c1bbcf12924..b35921a670b2 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -499,30 +499,6 @@ void set_personality_ia32(bool x32) } EXPORT_SYMBOL_GPL(set_personality_ia32); -unsigned long get_wchan(struct task_struct *p) -{ - unsigned long stack; - u64 fp, ip; - int count = 0; - - if (!p || p == current || p->state == TASK_RUNNING) - return 0; - stack = (unsigned long)task_stack_page(p); - if (p->thread.sp < stack || p->thread.sp >= stack+THREAD_SIZE) - return 0; - fp = *(u64 *)(p->thread.sp); - do { - if (fp < (unsigned long)stack || - fp >= (unsigned long)stack+THREAD_SIZE) - return 0; - ip = *(u64 *)(fp+8); - if (!in_sched_functions(ip)) - return ip; - fp = *(u64 *)fp; - } while (count++ < 16); - return 0; -} - long do_arch_prctl(struct task_struct *task, int code, unsigned long addr) { int ret = 0; diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 94b7d15db3fc..2f9ed1ff0632 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -514,7 +514,7 @@ static void skip_emulated_instruction(struct kvm_vcpu *vcpu) struct vcpu_svm *svm = to_svm(vcpu); if (svm->vmcb->control.next_rip != 0) { - WARN_ON(!static_cpu_has(X86_FEATURE_NRIPS)); + WARN_ON_ONCE(!static_cpu_has(X86_FEATURE_NRIPS)); svm->next_rip = svm->vmcb->control.next_rip; } @@ -866,64 +866,6 @@ static void svm_disable_lbrv(struct vcpu_svm *svm) set_msr_interception(msrpm, MSR_IA32_LASTINTTOIP, 0, 0); } -#define MTRR_TYPE_UC_MINUS 7 -#define MTRR2PROTVAL_INVALID 0xff - -static u8 mtrr2protval[8]; - -static u8 fallback_mtrr_type(int mtrr) -{ - /* - * WT and WP aren't always available in the host PAT. Treat - * them as UC and UC- respectively. Everything else should be - * there. - */ - switch (mtrr) - { - case MTRR_TYPE_WRTHROUGH: - return MTRR_TYPE_UNCACHABLE; - case MTRR_TYPE_WRPROT: - return MTRR_TYPE_UC_MINUS; - default: - BUG(); - } -} - -static void build_mtrr2protval(void) -{ - int i; - u64 pat; - - for (i = 0; i < 8; i++) - mtrr2protval[i] = MTRR2PROTVAL_INVALID; - - /* Ignore the invalid MTRR types. */ - mtrr2protval[2] = 0; - mtrr2protval[3] = 0; - - /* - * Use host PAT value to figure out the mapping from guest MTRR - * values to nested page table PAT/PCD/PWT values. We do not - * want to change the host PAT value every time we enter the - * guest. - */ - rdmsrl(MSR_IA32_CR_PAT, pat); - for (i = 0; i < 8; i++) { - u8 mtrr = pat >> (8 * i); - - if (mtrr2protval[mtrr] == MTRR2PROTVAL_INVALID) - mtrr2protval[mtrr] = __cm_idx2pte(i); - } - - for (i = 0; i < 8; i++) { - if (mtrr2protval[i] == MTRR2PROTVAL_INVALID) { - u8 fallback = fallback_mtrr_type(i); - mtrr2protval[i] = mtrr2protval[fallback]; - BUG_ON(mtrr2protval[i] == MTRR2PROTVAL_INVALID); - } - } -} - static __init int svm_hardware_setup(void) { int cpu; @@ -990,7 +932,6 @@ static __init int svm_hardware_setup(void) } else kvm_disable_tdp(); - build_mtrr2protval(); return 0; err: @@ -1145,43 +1086,6 @@ static u64 svm_compute_tsc_offset(struct kvm_vcpu *vcpu, u64 target_tsc) return target_tsc - tsc; } -static void svm_set_guest_pat(struct vcpu_svm *svm, u64 *g_pat) -{ - struct kvm_vcpu *vcpu = &svm->vcpu; - - /* Unlike Intel, AMD takes the guest's CR0.CD into account. - * - * AMD does not have IPAT. To emulate it for the case of guests - * with no assigned devices, just set everything to WB. If guests - * have assigned devices, however, we cannot force WB for RAM - * pages only, so use the guest PAT directly. - */ - if (!kvm_arch_has_assigned_device(vcpu->kvm)) - *g_pat = 0x0606060606060606; - else - *g_pat = vcpu->arch.pat; -} - -static u64 svm_get_mt_mask(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio) -{ - u8 mtrr; - - /* - * 1. MMIO: trust guest MTRR, so same as item 3. - * 2. No passthrough: always map as WB, and force guest PAT to WB as well - * 3. Passthrough: can't guarantee the result, try to trust guest. - */ - if (!is_mmio && !kvm_arch_has_assigned_device(vcpu->kvm)) - return 0; - - if (!kvm_check_has_quirk(vcpu->kvm, KVM_X86_QUIRK_CD_NW_CLEARED) && - kvm_read_cr0(vcpu) & X86_CR0_CD) - return _PAGE_NOCACHE; - - mtrr = kvm_mtrr_get_guest_memory_type(vcpu, gfn); - return mtrr2protval[mtrr]; -} - static void init_vmcb(struct vcpu_svm *svm, bool init_event) { struct vmcb_control_area *control = &svm->vmcb->control; @@ -1278,7 +1182,6 @@ static void init_vmcb(struct vcpu_svm *svm, bool init_event) clr_cr_intercept(svm, INTERCEPT_CR3_READ); clr_cr_intercept(svm, INTERCEPT_CR3_WRITE); save->g_pat = svm->vcpu.arch.pat; - svm_set_guest_pat(svm, &save->g_pat); save->cr3 = 0; save->cr4 = 0; } @@ -1673,10 +1576,13 @@ static void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) if (!vcpu->fpu_active) cr0 |= X86_CR0_TS; - - /* These are emulated via page tables. */ - cr0 &= ~(X86_CR0_CD | X86_CR0_NW); - + /* + * re-enable caching here because the QEMU bios + * does not do it - this results in some delay at + * reboot + */ + if (kvm_check_has_quirk(vcpu->kvm, KVM_X86_QUIRK_CD_NW_CLEARED)) + cr0 &= ~(X86_CR0_CD | X86_CR0_NW); svm->vmcb->save.cr0 = cr0; mark_dirty(svm->vmcb, VMCB_CR); update_cr0_intercept(svm); @@ -3351,16 +3257,6 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr) case MSR_VM_IGNNE: vcpu_unimpl(vcpu, "unimplemented wrmsr: 0x%x data 0x%llx\n", ecx, data); break; - case MSR_IA32_CR_PAT: - if (npt_enabled) { - if (!kvm_mtrr_valid(vcpu, MSR_IA32_CR_PAT, data)) - return 1; - vcpu->arch.pat = data; - svm_set_guest_pat(svm, &svm->vmcb->save.g_pat); - mark_dirty(svm->vmcb, VMCB_NPT); - break; - } - /* fall through */ default: return kvm_set_msr_common(vcpu, msr); } @@ -4195,6 +4091,11 @@ static bool svm_has_high_real_mode_segbase(void) return true; } +static u64 svm_get_mt_mask(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio) +{ + return 0; +} + static void svm_cpuid_update(struct kvm_vcpu *vcpu) { } diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 64076740251e..06ef4908ba61 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -8617,17 +8617,22 @@ static u64 vmx_get_mt_mask(struct kvm_vcpu *vcpu, gfn_t gfn, bool is_mmio) u64 ipat = 0; /* For VT-d and EPT combination - * 1. MMIO: guest may want to apply WC, trust it. + * 1. MMIO: always map as UC * 2. EPT with VT-d: * a. VT-d without snooping control feature: can't guarantee the - * result, try to trust guest. So the same as item 1. + * result, try to trust guest. * b. VT-d with snooping control feature: snooping control feature of * VT-d engine can guarantee the cache correctness. Just set it * to WB to keep consistent with host. So the same as item 3. * 3. EPT without VT-d: always map as WB and set IPAT=1 to keep * consistent with host MTRR */ - if (!is_mmio && !kvm_arch_has_noncoherent_dma(vcpu->kvm)) { + if (is_mmio) { + cache = MTRR_TYPE_UNCACHABLE; + goto exit; + } + + if (!kvm_arch_has_noncoherent_dma(vcpu->kvm)) { ipat = VMX_EPT_IPAT_BIT; cache = MTRR_TYPE_WRBACK; goto exit; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 991466bf8dee..92511d4b7236 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -1708,8 +1708,6 @@ static int kvm_guest_time_update(struct kvm_vcpu *v) vcpu->pvclock_set_guest_stopped_request = false; } - pvclock_flags |= PVCLOCK_COUNTS_FROM_ZERO; - /* If the host uses TSC clocksource, then it is stable */ if (use_master_clock) pvclock_flags |= PVCLOCK_TSC_STABLE_BIT; @@ -2007,8 +2005,6 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info) &vcpu->requests); ka->boot_vcpu_runs_old_kvmclock = tmp; - - ka->kvmclock_offset = -get_kernel_ns(); } vcpu->arch.time = data; diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 30564e2752d3..df48430c279b 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -1132,7 +1132,7 @@ void mark_rodata_ro(void) * has been zapped already via cleanup_highmem(). */ all_end = roundup((unsigned long)_brk_end, PMD_SIZE); - set_memory_nx(rodata_start, (all_end - rodata_start) >> PAGE_SHIFT); + set_memory_nx(text_end, (all_end - text_end) >> PAGE_SHIFT); rodata_test(); diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c index 1db84c0758b7..6a28ded74211 100644 --- a/arch/x86/platform/efi/efi.c +++ b/arch/x86/platform/efi/efi.c @@ -705,6 +705,70 @@ out: } /* + * Iterate the EFI memory map in reverse order because the regions + * will be mapped top-down. The end result is the same as if we had + * mapped things forward, but doesn't require us to change the + * existing implementation of efi_map_region(). + */ +static inline void *efi_map_next_entry_reverse(void *entry) +{ + /* Initial call */ + if (!entry) + return memmap.map_end - memmap.desc_size; + + entry -= memmap.desc_size; + if (entry < memmap.map) + return NULL; + + return entry; +} + +/* + * efi_map_next_entry - Return the next EFI memory map descriptor + * @entry: Previous EFI memory map descriptor + * + * This is a helper function to iterate over the EFI memory map, which + * we do in different orders depending on the current configuration. + * + * To begin traversing the memory map @entry must be %NULL. + * + * Returns %NULL when we reach the end of the memory map. + */ +static void *efi_map_next_entry(void *entry) +{ + if (!efi_enabled(EFI_OLD_MEMMAP) && efi_enabled(EFI_64BIT)) { + /* + * Starting in UEFI v2.5 the EFI_PROPERTIES_TABLE + * config table feature requires us to map all entries + * in the same order as they appear in the EFI memory + * map. That is to say, entry N must have a lower + * virtual address than entry N+1. This is because the + * firmware toolchain leaves relative references in + * the code/data sections, which are split and become + * separate EFI memory regions. Mapping things + * out-of-order leads to the firmware accessing + * unmapped addresses. + * + * Since we need to map things this way whether or not + * the kernel actually makes use of + * EFI_PROPERTIES_TABLE, let's just switch to this + * scheme by default for 64-bit. + */ + return efi_map_next_entry_reverse(entry); + } + + /* Initial call */ + if (!entry) + return memmap.map; + + entry += memmap.desc_size; + if (entry >= memmap.map_end) + return NULL; + + return entry; +} + +/* * Map the efi memory ranges of the runtime services and update new_mmap with * virtual addresses. */ @@ -714,7 +778,8 @@ static void * __init efi_map_regions(int *count, int *pg_shift) unsigned long left = 0; efi_memory_desc_t *md; - for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) { + p = NULL; + while ((p = efi_map_next_entry(p))) { md = p; if (!(md->attribute & EFI_MEMORY_RUNTIME)) { #ifdef CONFIG_X86_64 diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 30d12afe52ed..993b7a71386d 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -33,6 +33,10 @@ #include <linux/memblock.h> #include <linux/edd.h> +#ifdef CONFIG_KEXEC_CORE +#include <linux/kexec.h> +#endif + #include <xen/xen.h> #include <xen/events.h> #include <xen/interface/xen.h> @@ -1077,6 +1081,7 @@ static int xen_write_msr_safe(unsigned int msr, unsigned low, unsigned high) /* Fast syscall setup is all done in hypercalls, so these are all ignored. Stub them out here to stop Xen console noise. */ + break; default: if (!pmu_msr_write(msr, low, high, &ret)) @@ -1807,6 +1812,21 @@ static struct notifier_block xen_hvm_cpu_notifier = { .notifier_call = xen_hvm_cpu_notify, }; +#ifdef CONFIG_KEXEC_CORE +static void xen_hvm_shutdown(void) +{ + native_machine_shutdown(); + if (kexec_in_progress) + xen_reboot(SHUTDOWN_soft_reset); +} + +static void xen_hvm_crash_shutdown(struct pt_regs *regs) +{ + native_machine_crash_shutdown(regs); + xen_reboot(SHUTDOWN_soft_reset); +} +#endif + static void __init xen_hvm_guest_init(void) { if (xen_pv_domain()) @@ -1826,6 +1846,10 @@ static void __init xen_hvm_guest_init(void) x86_init.irqs.intr_init = xen_init_IRQ; xen_hvm_init_time_ops(); xen_hvm_init_mmu_ops(); +#ifdef CONFIG_KEXEC_CORE + machine_ops.shutdown = xen_hvm_shutdown; + machine_ops.crash_shutdown = xen_hvm_crash_shutdown; +#endif } #endif diff --git a/arch/x86/xen/p2m.c b/arch/x86/xen/p2m.c index bfc08b13044b..660b3cfef234 100644 --- a/arch/x86/xen/p2m.c +++ b/arch/x86/xen/p2m.c @@ -112,6 +112,15 @@ static unsigned long *p2m_identity; static pte_t *p2m_missing_pte; static pte_t *p2m_identity_pte; +/* + * Hint at last populated PFN. + * + * Used to set HYPERVISOR_shared_info->arch.max_pfn so the toolstack + * can avoid scanning the whole P2M (which may be sized to account for + * hotplugged memory). + */ +static unsigned long xen_p2m_last_pfn; + static inline unsigned p2m_top_index(unsigned long pfn) { BUG_ON(pfn >= MAX_P2M_PFN); @@ -270,7 +279,7 @@ void xen_setup_mfn_list_list(void) else HYPERVISOR_shared_info->arch.pfn_to_mfn_frame_list_list = virt_to_mfn(p2m_top_mfn); - HYPERVISOR_shared_info->arch.max_pfn = xen_max_p2m_pfn; + HYPERVISOR_shared_info->arch.max_pfn = xen_p2m_last_pfn; HYPERVISOR_shared_info->arch.p2m_generation = 0; HYPERVISOR_shared_info->arch.p2m_vaddr = (unsigned long)xen_p2m_addr; HYPERVISOR_shared_info->arch.p2m_cr3 = @@ -406,6 +415,8 @@ void __init xen_vmalloc_p2m_tree(void) static struct vm_struct vm; unsigned long p2m_limit; + xen_p2m_last_pfn = xen_max_p2m_pfn; + p2m_limit = (phys_addr_t)P2M_LIMIT * 1024 * 1024 * 1024 / PAGE_SIZE; vm.flags = VM_ALLOC; vm.size = ALIGN(sizeof(unsigned long) * max(xen_max_p2m_pfn, p2m_limit), @@ -608,6 +619,12 @@ static bool alloc_p2m(unsigned long pfn) free_p2m_page(p2m); } + /* Expanded the p2m? */ + if (pfn > xen_p2m_last_pfn) { + xen_p2m_last_pfn = pfn; + HYPERVISOR_shared_info->arch.max_pfn = xen_p2m_last_pfn; + } + return true; } diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c index f5ef6746d47a..1c30e4ab1022 100644 --- a/arch/x86/xen/setup.c +++ b/arch/x86/xen/setup.c @@ -548,7 +548,7 @@ static unsigned long __init xen_get_max_pages(void) { unsigned long max_pages, limit; domid_t domid = DOMID_SELF; - int ret; + long ret; limit = xen_get_pages_limit(); max_pages = limit; @@ -798,7 +798,7 @@ char * __init xen_memory_setup(void) xen_ignore_unusable(); /* Make sure the Xen-supplied memory map is well-ordered. */ - sanitize_e820_map(xen_e820_map, xen_e820_map_entries, + sanitize_e820_map(xen_e820_map, ARRAY_SIZE(xen_e820_map), &xen_e820_map_entries); max_pages = xen_get_max_pages(); diff --git a/arch/xtensa/include/asm/Kbuild b/arch/xtensa/include/asm/Kbuild index 63c223dff5f1..b56855a1382a 100644 --- a/arch/xtensa/include/asm/Kbuild +++ b/arch/xtensa/include/asm/Kbuild @@ -28,4 +28,5 @@ generic-y += statfs.h generic-y += termios.h generic-y += topology.h generic-y += trace_clock.h +generic-y += word-at-a-time.h generic-y += xor.h diff --git a/block/blk-mq-cpumap.c b/block/blk-mq-cpumap.c index 1e28ddb656b8..8764c241e5bb 100644 --- a/block/blk-mq-cpumap.c +++ b/block/blk-mq-cpumap.c @@ -31,7 +31,8 @@ static int get_first_sibling(unsigned int cpu) return cpu; } -int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues) +int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues, + const struct cpumask *online_mask) { unsigned int i, nr_cpus, nr_uniq_cpus, queue, first_sibling; cpumask_var_t cpus; @@ -41,7 +42,7 @@ int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues) cpumask_clear(cpus); nr_cpus = nr_uniq_cpus = 0; - for_each_online_cpu(i) { + for_each_cpu(i, online_mask) { nr_cpus++; first_sibling = get_first_sibling(i); if (!cpumask_test_cpu(first_sibling, cpus)) @@ -51,7 +52,7 @@ int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues) queue = 0; for_each_possible_cpu(i) { - if (!cpu_online(i)) { + if (!cpumask_test_cpu(i, online_mask)) { map[i] = 0; continue; } @@ -95,7 +96,7 @@ unsigned int *blk_mq_make_queue_map(struct blk_mq_tag_set *set) if (!map) return NULL; - if (!blk_mq_update_queue_map(map, set->nr_hw_queues)) + if (!blk_mq_update_queue_map(map, set->nr_hw_queues, cpu_online_mask)) return map; kfree(map); diff --git a/block/blk-mq-sysfs.c b/block/blk-mq-sysfs.c index 279c5d674edf..788fffd9b409 100644 --- a/block/blk-mq-sysfs.c +++ b/block/blk-mq-sysfs.c @@ -229,8 +229,6 @@ static ssize_t blk_mq_hw_sysfs_cpus_show(struct blk_mq_hw_ctx *hctx, char *page) unsigned int i, first = 1; ssize_t ret = 0; - blk_mq_disable_hotplug(); - for_each_cpu(i, hctx->cpumask) { if (first) ret += sprintf(ret + page, "%u", i); @@ -240,8 +238,6 @@ static ssize_t blk_mq_hw_sysfs_cpus_show(struct blk_mq_hw_ctx *hctx, char *page) first = 0; } - blk_mq_enable_hotplug(); - ret += sprintf(ret + page, "\n"); return ret; } @@ -343,7 +339,7 @@ static void blk_mq_unregister_hctx(struct blk_mq_hw_ctx *hctx) struct blk_mq_ctx *ctx; int i; - if (!hctx->nr_ctx || !(hctx->flags & BLK_MQ_F_SYSFS_UP)) + if (!hctx->nr_ctx) return; hctx_for_each_ctx(hctx, ctx, i) @@ -358,7 +354,7 @@ static int blk_mq_register_hctx(struct blk_mq_hw_ctx *hctx) struct blk_mq_ctx *ctx; int i, ret; - if (!hctx->nr_ctx || !(hctx->flags & BLK_MQ_F_SYSFS_UP)) + if (!hctx->nr_ctx) return 0; ret = kobject_add(&hctx->kobj, &q->mq_kobj, "%u", hctx->queue_num); @@ -381,6 +377,8 @@ void blk_mq_unregister_disk(struct gendisk *disk) struct blk_mq_ctx *ctx; int i, j; + blk_mq_disable_hotplug(); + queue_for_each_hw_ctx(q, hctx, i) { blk_mq_unregister_hctx(hctx); @@ -395,6 +393,9 @@ void blk_mq_unregister_disk(struct gendisk *disk) kobject_put(&q->mq_kobj); kobject_put(&disk_to_dev(disk)->kobj); + + q->mq_sysfs_init_done = false; + blk_mq_enable_hotplug(); } static void blk_mq_sysfs_init(struct request_queue *q) @@ -425,27 +426,30 @@ int blk_mq_register_disk(struct gendisk *disk) struct blk_mq_hw_ctx *hctx; int ret, i; + blk_mq_disable_hotplug(); + blk_mq_sysfs_init(q); ret = kobject_add(&q->mq_kobj, kobject_get(&dev->kobj), "%s", "mq"); if (ret < 0) - return ret; + goto out; kobject_uevent(&q->mq_kobj, KOBJ_ADD); queue_for_each_hw_ctx(q, hctx, i) { - hctx->flags |= BLK_MQ_F_SYSFS_UP; ret = blk_mq_register_hctx(hctx); if (ret) break; } - if (ret) { + if (ret) blk_mq_unregister_disk(disk); - return ret; - } + else + q->mq_sysfs_init_done = true; +out: + blk_mq_enable_hotplug(); - return 0; + return ret; } EXPORT_SYMBOL_GPL(blk_mq_register_disk); @@ -454,6 +458,9 @@ void blk_mq_sysfs_unregister(struct request_queue *q) struct blk_mq_hw_ctx *hctx; int i; + if (!q->mq_sysfs_init_done) + return; + queue_for_each_hw_ctx(q, hctx, i) blk_mq_unregister_hctx(hctx); } @@ -463,6 +470,9 @@ int blk_mq_sysfs_register(struct request_queue *q) struct blk_mq_hw_ctx *hctx; int i, ret = 0; + if (!q->mq_sysfs_init_done) + return ret; + queue_for_each_hw_ctx(q, hctx, i) { ret = blk_mq_register_hctx(hctx); if (ret) diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c index 9115c6d59948..ed96474d75cb 100644 --- a/block/blk-mq-tag.c +++ b/block/blk-mq-tag.c @@ -471,17 +471,30 @@ void blk_mq_all_tag_busy_iter(struct blk_mq_tags *tags, busy_tag_iter_fn *fn, } EXPORT_SYMBOL(blk_mq_all_tag_busy_iter); -void blk_mq_tag_busy_iter(struct blk_mq_hw_ctx *hctx, busy_iter_fn *fn, +void blk_mq_queue_tag_busy_iter(struct request_queue *q, busy_iter_fn *fn, void *priv) { - struct blk_mq_tags *tags = hctx->tags; + struct blk_mq_hw_ctx *hctx; + int i; + + + queue_for_each_hw_ctx(q, hctx, i) { + struct blk_mq_tags *tags = hctx->tags; + + /* + * If not software queues are currently mapped to this + * hardware queue, there's nothing to check + */ + if (!blk_mq_hw_queue_mapped(hctx)) + continue; + + if (tags->nr_reserved_tags) + bt_for_each(hctx, &tags->breserved_tags, 0, fn, priv, true); + bt_for_each(hctx, &tags->bitmap_tags, tags->nr_reserved_tags, fn, priv, + false); + } - if (tags->nr_reserved_tags) - bt_for_each(hctx, &tags->breserved_tags, 0, fn, priv, true); - bt_for_each(hctx, &tags->bitmap_tags, tags->nr_reserved_tags, fn, priv, - false); } -EXPORT_SYMBOL(blk_mq_tag_busy_iter); static unsigned int bt_unused_tags(struct blk_mq_bitmap_tags *bt) { diff --git a/block/blk-mq-tag.h b/block/blk-mq-tag.h index 9eb2cf4f01cb..d468a79f2c4a 100644 --- a/block/blk-mq-tag.h +++ b/block/blk-mq-tag.h @@ -58,6 +58,8 @@ extern ssize_t blk_mq_tag_sysfs_show(struct blk_mq_tags *tags, char *page); extern void blk_mq_tag_init_last_tag(struct blk_mq_tags *tags, unsigned int *last_tag); extern int blk_mq_tag_update_depth(struct blk_mq_tags *tags, unsigned int depth); extern void blk_mq_tag_wakeup_all(struct blk_mq_tags *tags, bool); +void blk_mq_queue_tag_busy_iter(struct request_queue *q, busy_iter_fn *fn, + void *priv); enum { BLK_MQ_TAG_CACHE_MIN = 1, diff --git a/block/blk-mq.c b/block/blk-mq.c index f2d67b4047a0..7785ae96267a 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -393,14 +393,16 @@ void __blk_mq_complete_request(struct request *rq) * Ends all I/O on a request. It does not handle partial completions. * The actual completion happens out-of-order, through a IPI handler. **/ -void blk_mq_complete_request(struct request *rq) +void blk_mq_complete_request(struct request *rq, int error) { struct request_queue *q = rq->q; if (unlikely(blk_should_fake_timeout(q))) return; - if (!blk_mark_rq_complete(rq)) + if (!blk_mark_rq_complete(rq)) { + rq->errors = error; __blk_mq_complete_request(rq); + } } EXPORT_SYMBOL(blk_mq_complete_request); @@ -616,10 +618,8 @@ static void blk_mq_check_expired(struct blk_mq_hw_ctx *hctx, * If a request wasn't started before the queue was * marked dying, kill it here or it'll go unnoticed. */ - if (unlikely(blk_queue_dying(rq->q))) { - rq->errors = -EIO; - blk_mq_complete_request(rq); - } + if (unlikely(blk_queue_dying(rq->q))) + blk_mq_complete_request(rq, -EIO); return; } if (rq->cmd_flags & REQ_NO_TIMEOUT) @@ -641,24 +641,16 @@ static void blk_mq_rq_timer(unsigned long priv) .next = 0, .next_set = 0, }; - struct blk_mq_hw_ctx *hctx; int i; - queue_for_each_hw_ctx(q, hctx, i) { - /* - * If not software queues are currently mapped to this - * hardware queue, there's nothing to check - */ - if (!blk_mq_hw_queue_mapped(hctx)) - continue; - - blk_mq_tag_busy_iter(hctx, blk_mq_check_expired, &data); - } + blk_mq_queue_tag_busy_iter(q, blk_mq_check_expired, &data); if (data.next_set) { data.next = blk_rq_timeout(round_jiffies_up(data.next)); mod_timer(&q->timeout, data.next); } else { + struct blk_mq_hw_ctx *hctx; + queue_for_each_hw_ctx(q, hctx, i) { /* the hctx may be unmapped, so check it here */ if (blk_mq_hw_queue_mapped(hctx)) @@ -1789,13 +1781,19 @@ static void blk_mq_init_cpu_queues(struct request_queue *q, } } -static void blk_mq_map_swqueue(struct request_queue *q) +static void blk_mq_map_swqueue(struct request_queue *q, + const struct cpumask *online_mask) { unsigned int i; struct blk_mq_hw_ctx *hctx; struct blk_mq_ctx *ctx; struct blk_mq_tag_set *set = q->tag_set; + /* + * Avoid others reading imcomplete hctx->cpumask through sysfs + */ + mutex_lock(&q->sysfs_lock); + queue_for_each_hw_ctx(q, hctx, i) { cpumask_clear(hctx->cpumask); hctx->nr_ctx = 0; @@ -1806,16 +1804,17 @@ static void blk_mq_map_swqueue(struct request_queue *q) */ queue_for_each_ctx(q, ctx, i) { /* If the cpu isn't online, the cpu is mapped to first hctx */ - if (!cpu_online(i)) + if (!cpumask_test_cpu(i, online_mask)) continue; hctx = q->mq_ops->map_queue(q, i); cpumask_set_cpu(i, hctx->cpumask); - cpumask_set_cpu(i, hctx->tags->cpumask); ctx->index_hw = hctx->nr_ctx; hctx->ctxs[hctx->nr_ctx++] = ctx; } + mutex_unlock(&q->sysfs_lock); + queue_for_each_hw_ctx(q, hctx, i) { struct blk_mq_ctxmap *map = &hctx->ctx_map; @@ -1851,6 +1850,14 @@ static void blk_mq_map_swqueue(struct request_queue *q) hctx->next_cpu = cpumask_first(hctx->cpumask); hctx->next_cpu_batch = BLK_MQ_CPU_WORK_BATCH; } + + queue_for_each_ctx(q, ctx, i) { + if (!cpumask_test_cpu(i, online_mask)) + continue; + + hctx = q->mq_ops->map_queue(q, i); + cpumask_set_cpu(i, hctx->tags->cpumask); + } } static void blk_mq_update_tag_set_depth(struct blk_mq_tag_set *set) @@ -1918,6 +1925,9 @@ void blk_mq_release(struct request_queue *q) kfree(hctx); } + kfree(q->mq_map); + q->mq_map = NULL; + kfree(q->queue_hw_ctx); /* ctx kobj stays in queue_ctx */ @@ -2027,13 +2037,15 @@ struct request_queue *blk_mq_init_allocated_queue(struct blk_mq_tag_set *set, if (blk_mq_init_hw_queues(q, set)) goto err_hctxs; + get_online_cpus(); mutex_lock(&all_q_mutex); - list_add_tail(&q->all_q_node, &all_q_list); - mutex_unlock(&all_q_mutex); + list_add_tail(&q->all_q_node, &all_q_list); blk_mq_add_queue_tag_set(set, q); + blk_mq_map_swqueue(q, cpu_online_mask); - blk_mq_map_swqueue(q); + mutex_unlock(&all_q_mutex); + put_online_cpus(); return q; @@ -2057,30 +2069,27 @@ void blk_mq_free_queue(struct request_queue *q) { struct blk_mq_tag_set *set = q->tag_set; + mutex_lock(&all_q_mutex); + list_del_init(&q->all_q_node); + mutex_unlock(&all_q_mutex); + blk_mq_del_queue_tag_set(q); blk_mq_exit_hw_queues(q, set, set->nr_hw_queues); blk_mq_free_hw_queues(q, set); percpu_ref_exit(&q->mq_usage_counter); - - kfree(q->mq_map); - - q->mq_map = NULL; - - mutex_lock(&all_q_mutex); - list_del_init(&q->all_q_node); - mutex_unlock(&all_q_mutex); } /* Basically redo blk_mq_init_queue with queue frozen */ -static void blk_mq_queue_reinit(struct request_queue *q) +static void blk_mq_queue_reinit(struct request_queue *q, + const struct cpumask *online_mask) { WARN_ON_ONCE(!atomic_read(&q->mq_freeze_depth)); blk_mq_sysfs_unregister(q); - blk_mq_update_queue_map(q->mq_map, q->nr_hw_queues); + blk_mq_update_queue_map(q->mq_map, q->nr_hw_queues, online_mask); /* * redo blk_mq_init_cpu_queues and blk_mq_init_hw_queues. FIXME: maybe @@ -2088,7 +2097,7 @@ static void blk_mq_queue_reinit(struct request_queue *q) * involves free and re-allocate memory, worthy doing?) */ - blk_mq_map_swqueue(q); + blk_mq_map_swqueue(q, online_mask); blk_mq_sysfs_register(q); } @@ -2097,16 +2106,43 @@ static int blk_mq_queue_reinit_notify(struct notifier_block *nb, unsigned long action, void *hcpu) { struct request_queue *q; + int cpu = (unsigned long)hcpu; + /* + * New online cpumask which is going to be set in this hotplug event. + * Declare this cpumasks as global as cpu-hotplug operation is invoked + * one-by-one and dynamically allocating this could result in a failure. + */ + static struct cpumask online_new; /* - * Before new mappings are established, hotadded cpu might already - * start handling requests. This doesn't break anything as we map - * offline CPUs to first hardware queue. We will re-init the queue - * below to get optimal settings. + * Before hotadded cpu starts handling requests, new mappings must + * be established. Otherwise, these requests in hw queue might + * never be dispatched. + * + * For example, there is a single hw queue (hctx) and two CPU queues + * (ctx0 for CPU0, and ctx1 for CPU1). + * + * Now CPU1 is just onlined and a request is inserted into + * ctx1->rq_list and set bit0 in pending bitmap as ctx1->index_hw is + * still zero. + * + * And then while running hw queue, flush_busy_ctxs() finds bit0 is + * set in pending bitmap and tries to retrieve requests in + * hctx->ctxs[0]->rq_list. But htx->ctxs[0] is a pointer to ctx0, + * so the request in ctx1->rq_list is ignored. */ - if (action != CPU_DEAD && action != CPU_DEAD_FROZEN && - action != CPU_ONLINE && action != CPU_ONLINE_FROZEN) + switch (action & ~CPU_TASKS_FROZEN) { + case CPU_DEAD: + case CPU_UP_CANCELED: + cpumask_copy(&online_new, cpu_online_mask); + break; + case CPU_UP_PREPARE: + cpumask_copy(&online_new, cpu_online_mask); + cpumask_set_cpu(cpu, &online_new); + break; + default: return NOTIFY_OK; + } mutex_lock(&all_q_mutex); @@ -2130,7 +2166,7 @@ static int blk_mq_queue_reinit_notify(struct notifier_block *nb, } list_for_each_entry(q, &all_q_list, all_q_node) - blk_mq_queue_reinit(q); + blk_mq_queue_reinit(q, &online_new); list_for_each_entry(q, &all_q_list, all_q_node) blk_mq_unfreeze_queue(q); diff --git a/block/blk-mq.h b/block/blk-mq.h index 6a48c4c0d8a2..f4fea7964910 100644 --- a/block/blk-mq.h +++ b/block/blk-mq.h @@ -51,7 +51,8 @@ void blk_mq_disable_hotplug(void); * CPU -> queue mappings */ extern unsigned int *blk_mq_make_queue_map(struct blk_mq_tag_set *set); -extern int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues); +extern int blk_mq_update_queue_map(unsigned int *map, unsigned int nr_queues, + const struct cpumask *online_mask); extern int blk_mq_hw_queue_to_node(unsigned int *map, unsigned int); /* diff --git a/crypto/asymmetric_keys/x509_public_key.c b/crypto/asymmetric_keys/x509_public_key.c index 6d88dd15c98d..197096632412 100644 --- a/crypto/asymmetric_keys/x509_public_key.c +++ b/crypto/asymmetric_keys/x509_public_key.c @@ -332,10 +332,6 @@ static int x509_key_preparse(struct key_preparsed_payload *prep) srlen = cert->raw_serial_size; q = cert->raw_serial; } - if (srlen > 1 && *q == 0) { - srlen--; - q++; - } ret = -ENOMEM; desc = kmalloc(sulen + 2 + srlen * 2 + 1, GFP_KERNEL); diff --git a/drivers/Kconfig b/drivers/Kconfig index 46b4a8e0f859..5a89e409ad18 100644 --- a/drivers/Kconfig +++ b/drivers/Kconfig @@ -188,4 +188,10 @@ source "drivers/nvdimm/Kconfig" source "drivers/nvmem/Kconfig" +source "drivers/hwtracing/stm/Kconfig" + +source "drivers/hwtracing/intel_th/Kconfig" + +source "drivers/fpga/Kconfig" + endmenu diff --git a/drivers/Makefile b/drivers/Makefile index b250b36b54f2..7064bf476c2a 100644 --- a/drivers/Makefile +++ b/drivers/Makefile @@ -165,5 +165,8 @@ obj-$(CONFIG_PERF_EVENTS) += perf/ obj-$(CONFIG_RAS) += ras/ obj-$(CONFIG_THUNDERBOLT) += thunderbolt/ obj-$(CONFIG_CORESIGHT) += hwtracing/coresight/ +obj-y += hwtracing/intel_th/ +obj-$(CONFIG_STM) += hwtracing/stm/ obj-$(CONFIG_ANDROID) += android/ obj-$(CONFIG_NVMEM) += nvmem/ +obj-$(CONFIG_FPGA) += fpga/ diff --git a/drivers/acpi/ec.c b/drivers/acpi/ec.c index 2614a839c60d..42c66b64c12c 100644 --- a/drivers/acpi/ec.c +++ b/drivers/acpi/ec.c @@ -1044,8 +1044,10 @@ static int acpi_ec_query(struct acpi_ec *ec, u8 *data) goto err_exit; mutex_lock(&ec->mutex); + result = -ENODATA; list_for_each_entry(handler, &ec->list, node) { if (value == handler->query_bit) { + result = 0; q->handler = acpi_ec_get_query_handler(handler); ec_dbg_evt("Query(0x%02x) scheduled", q->handler->query_bit); diff --git a/drivers/acpi/pci_irq.c b/drivers/acpi/pci_irq.c index 6da0f9beab19..c9336751e5e3 100644 --- a/drivers/acpi/pci_irq.c +++ b/drivers/acpi/pci_irq.c @@ -372,6 +372,7 @@ static int acpi_isa_register_gsi(struct pci_dev *dev) /* Interrupt Line values above 0xF are forbidden */ if (dev->irq > 0 && (dev->irq <= 0xF) && + acpi_isa_irq_available(dev->irq) && (acpi_isa_irq_to_gsi(dev->irq, &dev_gsi) == 0)) { dev_warn(&dev->dev, "PCI INT %c: no GSI - using ISA IRQ %d\n", pin_name(dev->pin), dev->irq); diff --git a/drivers/acpi/pci_link.c b/drivers/acpi/pci_link.c index 3b4ea98e3ea0..7c8408b946ca 100644 --- a/drivers/acpi/pci_link.c +++ b/drivers/acpi/pci_link.c @@ -498,8 +498,7 @@ int __init acpi_irq_penalty_init(void) PIRQ_PENALTY_PCI_POSSIBLE; } } - /* Add a penalty for the SCI */ - acpi_irq_penalty[acpi_gbl_FADT.sci_interrupt] += PIRQ_PENALTY_PCI_USING; + return 0; } @@ -553,6 +552,13 @@ static int acpi_pci_link_allocate(struct acpi_pci_link *link) irq = link->irq.possible[i]; } } + if (acpi_irq_penalty[irq] >= PIRQ_PENALTY_ISA_ALWAYS) { + printk(KERN_ERR PREFIX "No IRQ available for %s [%s]. " + "Try pci=noacpi or acpi=off\n", + acpi_device_name(link->device), + acpi_device_bid(link->device)); + return -ENODEV; + } /* Attempt to enable the link device at this IRQ. */ if (acpi_pci_link_set(link, irq)) { @@ -821,6 +827,12 @@ void acpi_penalize_isa_irq(int irq, int active) } } +bool acpi_isa_irq_available(int irq) +{ + return irq >= 0 && (irq >= ARRAY_SIZE(acpi_irq_penalty) || + acpi_irq_penalty[irq] < PIRQ_PENALTY_ISA_ALWAYS); +} + /* * Penalize IRQ used by ACPI SCI. If ACPI SCI pin attributes conflict with * PCI IRQ attributes, mark ACPI SCI as ISA_ALWAYS so it won't be use for diff --git a/drivers/base/power/opp.c b/drivers/base/power/opp.c index 28cd75c535b0..7ae7cd990fbf 100644 --- a/drivers/base/power/opp.c +++ b/drivers/base/power/opp.c @@ -892,10 +892,17 @@ static int opp_get_microvolt(struct dev_pm_opp *opp, struct device *dev) u32 microvolt[3] = {0}; int count, ret; - count = of_property_count_u32_elems(opp->np, "opp-microvolt"); - if (!count) + /* Missing property isn't a problem, but an invalid entry is */ + if (!of_find_property(opp->np, "opp-microvolt", NULL)) return 0; + count = of_property_count_u32_elems(opp->np, "opp-microvolt"); + if (count < 0) { + dev_err(dev, "%s: Invalid opp-microvolt property (%d)\n", + __func__, count); + return count; + } + /* There can be one or three elements here */ if (count != 1 && count != 3) { dev_err(dev, "%s: Invalid number of elements in opp-microvolt property (%d)\n", @@ -1063,7 +1070,7 @@ EXPORT_SYMBOL_GPL(dev_pm_opp_add); * share a common logic which is isolated here. * * Return: -EINVAL for bad pointers, -ENOMEM if no memory available for the - * copy operation, returns 0 if no modifcation was done OR modification was + * copy operation, returns 0 if no modification was done OR modification was * successful. * * Locking: The internal device_opp and opp structures are RCU protected. @@ -1151,7 +1158,7 @@ unlock: * mutex locking or synchronize_rcu() blocking calls cannot be used. * * Return: -EINVAL for bad pointers, -ENOMEM if no memory available for the - * copy operation, returns 0 if no modifcation was done OR modification was + * copy operation, returns 0 if no modification was done OR modification was * successful. */ int dev_pm_opp_enable(struct device *dev, unsigned long freq) @@ -1177,7 +1184,7 @@ EXPORT_SYMBOL_GPL(dev_pm_opp_enable); * mutex locking or synchronize_rcu() blocking calls cannot be used. * * Return: -EINVAL for bad pointers, -ENOMEM if no memory available for the - * copy operation, returns 0 if no modifcation was done OR modification was + * copy operation, returns 0 if no modification was done OR modification was * successful. */ int dev_pm_opp_disable(struct device *dev, unsigned long freq) diff --git a/drivers/base/regmap/regmap-debugfs.c b/drivers/base/regmap/regmap-debugfs.c index f42f2bac6466..4c55cfbad19e 100644 --- a/drivers/base/regmap/regmap-debugfs.c +++ b/drivers/base/regmap/regmap-debugfs.c @@ -32,8 +32,7 @@ static DEFINE_MUTEX(regmap_debugfs_early_lock); /* Calculate the length of a fixed format */ static size_t regmap_calc_reg_len(int max_val, char *buf, size_t buf_size) { - snprintf(buf, buf_size, "%x", max_val); - return strlen(buf); + return snprintf(NULL, 0, "%x", max_val); } static ssize_t regmap_name_read_file(struct file *file, @@ -432,7 +431,7 @@ static ssize_t regmap_access_read_file(struct file *file, /* If we're in the region the user is trying to read */ if (p >= *ppos) { /* ...but not beyond it */ - if (buf_pos >= count - 1 - tot_len) + if (buf_pos + tot_len + 1 >= count) break; /* Format the register */ diff --git a/drivers/block/loop.c b/drivers/block/loop.c index f9889b6bc02c..674f800a3b57 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -1486,17 +1486,16 @@ static void loop_handle_cmd(struct loop_cmd *cmd) { const bool write = cmd->rq->cmd_flags & REQ_WRITE; struct loop_device *lo = cmd->rq->q->queuedata; - int ret = -EIO; + int ret = 0; - if (write && (lo->lo_flags & LO_FLAGS_READ_ONLY)) + if (write && (lo->lo_flags & LO_FLAGS_READ_ONLY)) { + ret = -EIO; goto failed; + } ret = do_req_filebacked(lo, cmd->rq); - failed: - if (ret) - cmd->rq->errors = -EIO; - blk_mq_complete_request(cmd->rq); + blk_mq_complete_request(cmd->rq, ret ? -EIO : 0); } static void loop_queue_write_work(struct work_struct *work) diff --git a/drivers/block/null_blk.c b/drivers/block/null_blk.c index a295b98c6bae..1c9e4fe5aa44 100644 --- a/drivers/block/null_blk.c +++ b/drivers/block/null_blk.c @@ -289,7 +289,7 @@ static inline void null_handle_cmd(struct nullb_cmd *cmd) case NULL_IRQ_SOFTIRQ: switch (queue_mode) { case NULL_Q_MQ: - blk_mq_complete_request(cmd->rq); + blk_mq_complete_request(cmd->rq, cmd->rq->errors); break; case NULL_Q_RQ: blk_complete_request(cmd->rq); diff --git a/drivers/block/nvme-core.c b/drivers/block/nvme-core.c index b97fc3fe0916..6f04771f1019 100644 --- a/drivers/block/nvme-core.c +++ b/drivers/block/nvme-core.c @@ -618,16 +618,15 @@ static void req_completion(struct nvme_queue *nvmeq, void *ctx, spin_unlock_irqrestore(req->q->queue_lock, flags); return; } + if (req->cmd_type == REQ_TYPE_DRV_PRIV) { if (cmd_rq->ctx == CMD_CTX_CANCELLED) - req->errors = -EINTR; - else - req->errors = status; + status = -EINTR; } else { - req->errors = nvme_error_status(status); + status = nvme_error_status(status); } - } else - req->errors = 0; + } + if (req->cmd_type == REQ_TYPE_DRV_PRIV) { u32 result = le32_to_cpup(&cqe->result); req->special = (void *)(uintptr_t)result; @@ -650,7 +649,7 @@ static void req_completion(struct nvme_queue *nvmeq, void *ctx, } nvme_free_iod(nvmeq->dev, iod); - blk_mq_complete_request(req); + blk_mq_complete_request(req, status); } /* length is in bytes. gfp flags indicates whether we may sleep. */ @@ -863,8 +862,7 @@ static int nvme_queue_rq(struct blk_mq_hw_ctx *hctx, if (ns && ns->ms && !blk_integrity_rq(req)) { if (!(ns->pi_type && ns->ms == 8) && req->cmd_type != REQ_TYPE_DRV_PRIV) { - req->errors = -EFAULT; - blk_mq_complete_request(req); + blk_mq_complete_request(req, -EFAULT); return BLK_MQ_RQ_QUEUE_OK; } } @@ -2439,6 +2437,22 @@ static void nvme_scan_namespaces(struct nvme_dev *dev, unsigned nn) list_sort(NULL, &dev->namespaces, ns_cmp); } +static void nvme_set_irq_hints(struct nvme_dev *dev) +{ + struct nvme_queue *nvmeq; + int i; + + for (i = 0; i < dev->online_queues; i++) { + nvmeq = dev->queues[i]; + + if (!nvmeq->tags || !(*nvmeq->tags)) + continue; + + irq_set_affinity_hint(dev->entry[nvmeq->cq_vector].vector, + blk_mq_tags_cpumask(*nvmeq->tags)); + } +} + static void nvme_dev_scan(struct work_struct *work) { struct nvme_dev *dev = container_of(work, struct nvme_dev, scan_work); @@ -2450,6 +2464,7 @@ static void nvme_dev_scan(struct work_struct *work) return; nvme_scan_namespaces(dev, le32_to_cpup(&ctrl->nn)); kfree(ctrl); + nvme_set_irq_hints(dev); } /* @@ -2953,22 +2968,6 @@ static const struct file_operations nvme_dev_fops = { .compat_ioctl = nvme_dev_ioctl, }; -static void nvme_set_irq_hints(struct nvme_dev *dev) -{ - struct nvme_queue *nvmeq; - int i; - - for (i = 0; i < dev->online_queues; i++) { - nvmeq = dev->queues[i]; - - if (!nvmeq->tags || !(*nvmeq->tags)) - continue; - - irq_set_affinity_hint(dev->entry[nvmeq->cq_vector].vector, - blk_mq_tags_cpumask(*nvmeq->tags)); - } -} - static int nvme_dev_start(struct nvme_dev *dev) { int result; @@ -3010,8 +3009,6 @@ static int nvme_dev_start(struct nvme_dev *dev) if (result) goto free_tags; - nvme_set_irq_hints(dev); - dev->event_limit = 1; return result; @@ -3062,7 +3059,6 @@ static int nvme_dev_resume(struct nvme_dev *dev) } else { nvme_unfreeze_queues(dev); nvme_dev_add(dev); - nvme_set_irq_hints(dev); } return 0; } diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c index e93899cc6f60..6ca35495a5be 100644 --- a/drivers/block/virtio_blk.c +++ b/drivers/block/virtio_blk.c @@ -144,7 +144,7 @@ static void virtblk_done(struct virtqueue *vq) do { virtqueue_disable_cb(vq); while ((vbr = virtqueue_get_buf(vblk->vqs[qid].vq, &len)) != NULL) { - blk_mq_complete_request(vbr->req); + blk_mq_complete_request(vbr->req, vbr->req->errors); req_done = true; } if (unlikely(virtqueue_is_broken(vq))) diff --git a/drivers/block/xen-blkback/xenbus.c b/drivers/block/xen-blkback/xenbus.c index deb3f001791f..767657565de6 100644 --- a/drivers/block/xen-blkback/xenbus.c +++ b/drivers/block/xen-blkback/xenbus.c @@ -212,6 +212,9 @@ static int xen_blkif_map(struct xen_blkif *blkif, grant_ref_t *gref, static int xen_blkif_disconnect(struct xen_blkif *blkif) { + struct pending_req *req, *n; + int i = 0, j; + if (blkif->xenblkd) { kthread_stop(blkif->xenblkd); wake_up(&blkif->shutdown_wq); @@ -238,13 +241,28 @@ static int xen_blkif_disconnect(struct xen_blkif *blkif) /* Remove all persistent grants and the cache of ballooned pages. */ xen_blkbk_free_caches(blkif); + /* Check that there is no request in use */ + list_for_each_entry_safe(req, n, &blkif->pending_free, free_list) { + list_del(&req->free_list); + + for (j = 0; j < MAX_INDIRECT_SEGMENTS; j++) + kfree(req->segments[j]); + + for (j = 0; j < MAX_INDIRECT_PAGES; j++) + kfree(req->indirect_pages[j]); + + kfree(req); + i++; + } + + WARN_ON(i != (XEN_BLKIF_REQS_PER_PAGE * blkif->nr_ring_pages)); + blkif->nr_ring_pages = 0; + return 0; } static void xen_blkif_free(struct xen_blkif *blkif) { - struct pending_req *req, *n; - int i = 0, j; xen_blkif_disconnect(blkif); xen_vbd_free(&blkif->vbd); @@ -257,22 +275,6 @@ static void xen_blkif_free(struct xen_blkif *blkif) BUG_ON(!list_empty(&blkif->free_pages)); BUG_ON(!RB_EMPTY_ROOT(&blkif->persistent_gnts)); - /* Check that there is no request in use */ - list_for_each_entry_safe(req, n, &blkif->pending_free, free_list) { - list_del(&req->free_list); - - for (j = 0; j < MAX_INDIRECT_SEGMENTS; j++) - kfree(req->segments[j]); - - for (j = 0; j < MAX_INDIRECT_PAGES; j++) - kfree(req->indirect_pages[j]); - - kfree(req); - i++; - } - - WARN_ON(i != (XEN_BLKIF_REQS_PER_PAGE * blkif->nr_ring_pages)); - kmem_cache_free(xen_blkif_cachep, blkif); } diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index 0823a96902f8..611170896b8c 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -1142,6 +1142,7 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id) RING_IDX i, rp; unsigned long flags; struct blkfront_info *info = (struct blkfront_info *)dev_id; + int error; spin_lock_irqsave(&info->io_lock, flags); @@ -1182,37 +1183,37 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id) continue; } - req->errors = (bret->status == BLKIF_RSP_OKAY) ? 0 : -EIO; + error = (bret->status == BLKIF_RSP_OKAY) ? 0 : -EIO; switch (bret->operation) { case BLKIF_OP_DISCARD: if (unlikely(bret->status == BLKIF_RSP_EOPNOTSUPP)) { struct request_queue *rq = info->rq; printk(KERN_WARNING "blkfront: %s: %s op failed\n", info->gd->disk_name, op_name(bret->operation)); - req->errors = -EOPNOTSUPP; + error = -EOPNOTSUPP; info->feature_discard = 0; info->feature_secdiscard = 0; queue_flag_clear(QUEUE_FLAG_DISCARD, rq); queue_flag_clear(QUEUE_FLAG_SECDISCARD, rq); } - blk_mq_complete_request(req); + blk_mq_complete_request(req, error); break; case BLKIF_OP_FLUSH_DISKCACHE: case BLKIF_OP_WRITE_BARRIER: if (unlikely(bret->status == BLKIF_RSP_EOPNOTSUPP)) { printk(KERN_WARNING "blkfront: %s: %s op failed\n", info->gd->disk_name, op_name(bret->operation)); - req->errors = -EOPNOTSUPP; + error = -EOPNOTSUPP; } if (unlikely(bret->status == BLKIF_RSP_ERROR && info->shadow[id].req.u.rw.nr_segments == 0)) { printk(KERN_WARNING "blkfront: %s: empty %s op failed\n", info->gd->disk_name, op_name(bret->operation)); - req->errors = -EOPNOTSUPP; + error = -EOPNOTSUPP; } - if (unlikely(req->errors)) { - if (req->errors == -EOPNOTSUPP) - req->errors = 0; + if (unlikely(error)) { + if (error == -EOPNOTSUPP) + error = 0; info->feature_flush = 0; xlvbd_flush(info); } @@ -1223,7 +1224,7 @@ static irqreturn_t blkif_interrupt(int irq, void *dev_id) dev_dbg(&info->xbdev->dev, "Bad return from blkdev data " "request: %x\n", bret->status); - blk_mq_complete_request(req); + blk_mq_complete_request(req, error); break; default: BUG(); diff --git a/drivers/bus/Kconfig b/drivers/bus/Kconfig index 1a82f3a17681..0ebca8ba7bc4 100644 --- a/drivers/bus/Kconfig +++ b/drivers/bus/Kconfig @@ -36,7 +36,6 @@ config ARM_CCI400_PORT_CTRL config ARM_CCI500_PMU bool "ARM CCI500 PMU support" - default y depends on (ARM && CPU_V7) || ARM64 depends on PERF_EVENTS select ARM_CCI_PMU diff --git a/drivers/char/efirtc.c b/drivers/char/efirtc.c index e39e7402e623..dc62568b7dde 100644 --- a/drivers/char/efirtc.c +++ b/drivers/char/efirtc.c @@ -30,7 +30,6 @@ #include <linux/types.h> #include <linux/errno.h> #include <linux/miscdevice.h> -#include <linux/module.h> #include <linux/init.h> #include <linux/rtc.h> #include <linux/proc_fs.h> @@ -395,14 +394,8 @@ efi_rtc_init(void) } return 0; } +device_initcall(efi_rtc_init); -static void __exit -efi_rtc_exit(void) -{ - /* not yet used */ -} - -module_init(efi_rtc_init); -module_exit(efi_rtc_exit); - +/* MODULE_LICENSE("GPL"); +*/ diff --git a/drivers/char/hpet.c b/drivers/char/hpet.c index 5c0baa9ffc64..240b6cf1d97c 100644 --- a/drivers/char/hpet.c +++ b/drivers/char/hpet.c @@ -12,7 +12,6 @@ */ #include <linux/interrupt.h> -#include <linux/module.h> #include <linux/kernel.h> #include <linux/types.h> #include <linux/miscdevice.h> @@ -1043,24 +1042,16 @@ static int hpet_acpi_add(struct acpi_device *device) return hpet_alloc(&data); } -static int hpet_acpi_remove(struct acpi_device *device) -{ - /* XXX need to unregister clocksource, dealloc mem, etc */ - return -EINVAL; -} - static const struct acpi_device_id hpet_device_ids[] = { {"PNP0103", 0}, {"", 0}, }; -MODULE_DEVICE_TABLE(acpi, hpet_device_ids); static struct acpi_driver hpet_acpi_driver = { .name = "hpet", .ids = hpet_device_ids, .ops = { .add = hpet_acpi_add, - .remove = hpet_acpi_remove, }, }; @@ -1086,19 +1077,9 @@ static int __init hpet_init(void) return 0; } +device_initcall(hpet_init); -static void __exit hpet_exit(void) -{ - acpi_bus_unregister_driver(&hpet_acpi_driver); - - if (sysctl_header) - unregister_sysctl_table(sysctl_header); - misc_deregister(&hpet_misc); - - return; -} - -module_init(hpet_init); -module_exit(hpet_exit); +/* MODULE_AUTHOR("Bob Picco <Robert.Picco@hp.com>"); MODULE_LICENSE("GPL"); +*/ diff --git a/drivers/char/snsc.c b/drivers/char/snsc.c index 8a80ead8d316..94006f9c2e43 100644 --- a/drivers/char/snsc.c +++ b/drivers/char/snsc.c @@ -19,7 +19,7 @@ #include <linux/sched.h> #include <linux/device.h> #include <linux/poll.h> -#include <linux/module.h> +#include <linux/init.h> #include <linux/slab.h> #include <linux/mutex.h> #include <asm/sn/io.h> @@ -461,5 +461,4 @@ scdrv_init(void) } return 0; } - -module_init(scdrv_init); +device_initcall(scdrv_init); diff --git a/drivers/clk/samsung/clk-cpu.c b/drivers/clk/samsung/clk-cpu.c index 7c1e1f58e2da..2fe37f708dc7 100644 --- a/drivers/clk/samsung/clk-cpu.c +++ b/drivers/clk/samsung/clk-cpu.c @@ -164,7 +164,7 @@ static int exynos_cpuclk_pre_rate_change(struct clk_notifier_data *ndata, * the values for DIV_COPY and DIV_HPM dividers need not be set. */ div0 = cfg_data->div0; - if (test_bit(CLK_CPU_HAS_DIV1, &cpuclk->flags)) { + if (cpuclk->flags & CLK_CPU_HAS_DIV1) { div1 = cfg_data->div1; if (readl(base + E4210_SRC_CPU) & E4210_MUX_HPM_MASK) div1 = readl(base + E4210_DIV_CPU1) & @@ -185,7 +185,7 @@ static int exynos_cpuclk_pre_rate_change(struct clk_notifier_data *ndata, alt_div = DIV_ROUND_UP(alt_prate, tmp_rate) - 1; WARN_ON(alt_div >= MAX_DIV); - if (test_bit(CLK_CPU_NEEDS_DEBUG_ALT_DIV, &cpuclk->flags)) { + if (cpuclk->flags & CLK_CPU_NEEDS_DEBUG_ALT_DIV) { /* * In Exynos4210, ATB clock parent is also mout_core. So * ATB clock also needs to be mantained at safe speed. @@ -206,7 +206,7 @@ static int exynos_cpuclk_pre_rate_change(struct clk_notifier_data *ndata, writel(div0, base + E4210_DIV_CPU0); wait_until_divider_stable(base + E4210_DIV_STAT_CPU0, DIV_MASK_ALL); - if (test_bit(CLK_CPU_HAS_DIV1, &cpuclk->flags)) { + if (cpuclk->flags & CLK_CPU_HAS_DIV1) { writel(div1, base + E4210_DIV_CPU1); wait_until_divider_stable(base + E4210_DIV_STAT_CPU1, DIV_MASK_ALL); @@ -225,7 +225,7 @@ static int exynos_cpuclk_post_rate_change(struct clk_notifier_data *ndata, unsigned long mux_reg; /* find out the divider values to use for clock data */ - if (test_bit(CLK_CPU_NEEDS_DEBUG_ALT_DIV, &cpuclk->flags)) { + if (cpuclk->flags & CLK_CPU_NEEDS_DEBUG_ALT_DIV) { while ((cfg_data->prate * 1000) != ndata->new_rate) { if (cfg_data->prate == 0) return -EINVAL; @@ -240,7 +240,7 @@ static int exynos_cpuclk_post_rate_change(struct clk_notifier_data *ndata, writel(mux_reg & ~(1 << 16), base + E4210_SRC_CPU); wait_until_mux_stable(base + E4210_STAT_CPU, 16, 1); - if (test_bit(CLK_CPU_NEEDS_DEBUG_ALT_DIV, &cpuclk->flags)) { + if (cpuclk->flags & CLK_CPU_NEEDS_DEBUG_ALT_DIV) { div |= (cfg_data->div0 & E4210_DIV0_ATB_MASK); div_mask |= E4210_DIV0_ATB_MASK; } diff --git a/drivers/clk/ti/clk-3xxx.c b/drivers/clk/ti/clk-3xxx.c index 676ee8f6d813..8831e1a05367 100644 --- a/drivers/clk/ti/clk-3xxx.c +++ b/drivers/clk/ti/clk-3xxx.c @@ -374,7 +374,6 @@ static struct ti_dt_clk omap3xxx_clks[] = { DT_CLK(NULL, "gpio2_ick", "gpio2_ick"), DT_CLK(NULL, "wdt3_ick", "wdt3_ick"), DT_CLK(NULL, "uart3_ick", "uart3_ick"), - DT_CLK(NULL, "uart4_ick", "uart4_ick"), DT_CLK(NULL, "gpt9_ick", "gpt9_ick"), DT_CLK(NULL, "gpt8_ick", "gpt8_ick"), DT_CLK(NULL, "gpt7_ick", "gpt7_ick"), @@ -519,6 +518,7 @@ static struct ti_dt_clk am35xx_clks[] = { static struct ti_dt_clk omap36xx_clks[] = { DT_CLK(NULL, "omap_192m_alwon_fck", "omap_192m_alwon_fck"), DT_CLK(NULL, "uart4_fck", "uart4_fck"), + DT_CLK(NULL, "uart4_ick", "uart4_ick"), { .node_name = NULL }, }; diff --git a/drivers/clk/ti/clk-7xx.c b/drivers/clk/ti/clk-7xx.c index 9b5b289e6334..a911d7de3377 100644 --- a/drivers/clk/ti/clk-7xx.c +++ b/drivers/clk/ti/clk-7xx.c @@ -18,7 +18,6 @@ #include "clock.h" -#define DRA7_DPLL_ABE_DEFFREQ 180633600 #define DRA7_DPLL_GMAC_DEFFREQ 1000000000 #define DRA7_DPLL_USB_DEFFREQ 960000000 @@ -313,27 +312,12 @@ static struct ti_dt_clk dra7xx_clks[] = { int __init dra7xx_dt_clk_init(void) { int rc; - struct clk *abe_dpll_mux, *sys_clkin2, *dpll_ck, *hdcp_ck; + struct clk *dpll_ck, *hdcp_ck; ti_dt_clocks_register(dra7xx_clks); omap2_clk_disable_autoidle_all(); - abe_dpll_mux = clk_get_sys(NULL, "abe_dpll_sys_clk_mux"); - sys_clkin2 = clk_get_sys(NULL, "sys_clkin2"); - dpll_ck = clk_get_sys(NULL, "dpll_abe_ck"); - - rc = clk_set_parent(abe_dpll_mux, sys_clkin2); - if (!rc) - rc = clk_set_rate(dpll_ck, DRA7_DPLL_ABE_DEFFREQ); - if (rc) - pr_err("%s: failed to configure ABE DPLL!\n", __func__); - - dpll_ck = clk_get_sys(NULL, "dpll_abe_m2x2_ck"); - rc = clk_set_rate(dpll_ck, DRA7_DPLL_ABE_DEFFREQ * 2); - if (rc) - pr_err("%s: failed to configure ABE DPLL m2x2!\n", __func__); - dpll_ck = clk_get_sys(NULL, "dpll_gmac_ck"); rc = clk_set_rate(dpll_ck, DRA7_DPLL_GMAC_DEFFREQ); if (rc) diff --git a/drivers/clk/ti/clkt_dflt.c b/drivers/clk/ti/clkt_dflt.c index 90d7d8a21c49..1ddc288fce4e 100644 --- a/drivers/clk/ti/clkt_dflt.c +++ b/drivers/clk/ti/clkt_dflt.c @@ -222,7 +222,7 @@ int omap2_dflt_clk_enable(struct clk_hw *hw) } } - if (unlikely(!clk->enable_reg)) { + if (unlikely(IS_ERR(clk->enable_reg))) { pr_err("%s: %s missing enable_reg\n", __func__, clk_hw_get_name(hw)); ret = -EINVAL; @@ -264,7 +264,7 @@ void omap2_dflt_clk_disable(struct clk_hw *hw) u32 v; clk = to_clk_hw_omap(hw); - if (!clk->enable_reg) { + if (IS_ERR(clk->enable_reg)) { /* * 'independent' here refers to a clock which is not * controlled by its parent. diff --git a/drivers/clocksource/rockchip_timer.c b/drivers/clocksource/rockchip_timer.c index bb2c2b050964..d3c1742ded1a 100644 --- a/drivers/clocksource/rockchip_timer.c +++ b/drivers/clocksource/rockchip_timer.c @@ -148,7 +148,7 @@ static void __init rk_timer_init(struct device_node *np) bc_timer.freq = clk_get_rate(timer_clk); irq = irq_of_parse_and_map(np, 0); - if (irq == NO_IRQ) { + if (!irq) { pr_err("Failed to map interrupts for '%s'\n", TIMER_NAME); return; } diff --git a/drivers/clocksource/timer-keystone.c b/drivers/clocksource/timer-keystone.c index edacf3902e10..1cea08cf603e 100644 --- a/drivers/clocksource/timer-keystone.c +++ b/drivers/clocksource/timer-keystone.c @@ -152,7 +152,7 @@ static void __init keystone_timer_init(struct device_node *np) int irq, error; irq = irq_of_parse_and_map(np, 0); - if (irq == NO_IRQ) { + if (!irq) { pr_err("%s: failed to map interrupts\n", __func__); return; } diff --git a/drivers/cpufreq/acpi-cpufreq.c b/drivers/cpufreq/acpi-cpufreq.c index 798277227de7..cec1ee2d2f74 100644 --- a/drivers/cpufreq/acpi-cpufreq.c +++ b/drivers/cpufreq/acpi-cpufreq.c @@ -149,6 +149,9 @@ static ssize_t show_freqdomain_cpus(struct cpufreq_policy *policy, char *buf) { struct acpi_cpufreq_data *data = policy->driver_data; + if (unlikely(!data)) + return -ENODEV; + return cpufreq_show_cpus(data->freqdomain_cpus, buf); } diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index ef5ed9470de9..25c4c15103a0 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -1436,8 +1436,10 @@ static void cpufreq_offline_finish(unsigned int cpu) * since this is a core component, and is essential for the * subsequent light-weight ->init() to succeed. */ - if (cpufreq_driver->exit) + if (cpufreq_driver->exit) { cpufreq_driver->exit(policy); + policy->freq_table = NULL; + } } /** diff --git a/drivers/devfreq/devfreq.c b/drivers/devfreq/devfreq.c index 3927ed9fdbd5..ca848cc6a8fd 100644 --- a/drivers/devfreq/devfreq.c +++ b/drivers/devfreq/devfreq.c @@ -492,7 +492,7 @@ struct devfreq *devfreq_add_device(struct device *dev, if (err) { put_device(&devfreq->dev); mutex_unlock(&devfreq->lock); - goto err_dev; + goto err_out; } mutex_unlock(&devfreq->lock); @@ -518,7 +518,6 @@ struct devfreq *devfreq_add_device(struct device *dev, err_init: list_del(&devfreq->node); device_unregister(&devfreq->dev); -err_dev: kfree(devfreq); err_out: return ERR_PTR(err); @@ -795,8 +794,10 @@ static ssize_t governor_store(struct device *dev, struct device_attribute *attr, ret = PTR_ERR(governor); goto out; } - if (df->governor == governor) + if (df->governor == governor) { + ret = 0; goto out; + } if (df->governor) { ret = df->governor->event_handler(df, DEVFREQ_GOV_STOP, NULL); diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c index a165b4bfd330..dd24375b76dd 100644 --- a/drivers/dma/at_xdmac.c +++ b/drivers/dma/at_xdmac.c @@ -455,6 +455,15 @@ static struct at_xdmac_desc *at_xdmac_alloc_desc(struct dma_chan *chan, return desc; } +void at_xdmac_init_used_desc(struct at_xdmac_desc *desc) +{ + memset(&desc->lld, 0, sizeof(desc->lld)); + INIT_LIST_HEAD(&desc->descs_list); + desc->direction = DMA_TRANS_NONE; + desc->xfer_size = 0; + desc->active_xfer = false; +} + /* Call must be protected by lock. */ static struct at_xdmac_desc *at_xdmac_get_desc(struct at_xdmac_chan *atchan) { @@ -466,7 +475,7 @@ static struct at_xdmac_desc *at_xdmac_get_desc(struct at_xdmac_chan *atchan) desc = list_first_entry(&atchan->free_descs_list, struct at_xdmac_desc, desc_node); list_del(&desc->desc_node); - desc->active_xfer = false; + at_xdmac_init_used_desc(desc); } return desc; @@ -875,14 +884,14 @@ at_xdmac_interleaved_queue_desc(struct dma_chan *chan, if (xt->src_inc) { if (xt->src_sgl) - chan_cc |= AT_XDMAC_CC_SAM_UBS_DS_AM; + chan_cc |= AT_XDMAC_CC_SAM_UBS_AM; else chan_cc |= AT_XDMAC_CC_SAM_INCREMENTED_AM; } if (xt->dst_inc) { if (xt->dst_sgl) - chan_cc |= AT_XDMAC_CC_DAM_UBS_DS_AM; + chan_cc |= AT_XDMAC_CC_DAM_UBS_AM; else chan_cc |= AT_XDMAC_CC_DAM_INCREMENTED_AM; } diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c index 3ff284c8e3d5..09479d4be4db 100644 --- a/drivers/dma/dmaengine.c +++ b/drivers/dma/dmaengine.c @@ -554,10 +554,18 @@ struct dma_chan *dma_get_slave_channel(struct dma_chan *chan) mutex_lock(&dma_list_mutex); if (chan->client_count == 0) { + struct dma_device *device = chan->device; + + dma_cap_set(DMA_PRIVATE, device->cap_mask); + device->privatecnt++; err = dma_chan_get(chan); - if (err) + if (err) { pr_debug("%s: failed to get %s: (%d)\n", __func__, dma_chan_name(chan), err); + chan = NULL; + if (--device->privatecnt == 0) + dma_cap_clear(DMA_PRIVATE, device->cap_mask); + } } else chan = NULL; diff --git a/drivers/dma/dw/core.c b/drivers/dma/dw/core.c index cf1c87fa1edd..bedce038c6e2 100644 --- a/drivers/dma/dw/core.c +++ b/drivers/dma/dw/core.c @@ -1591,7 +1591,6 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata) INIT_LIST_HEAD(&dw->dma.channels); for (i = 0; i < nr_channels; i++) { struct dw_dma_chan *dwc = &dw->chan[i]; - int r = nr_channels - i - 1; dwc->chan.device = &dw->dma; dma_cookie_init(&dwc->chan); @@ -1603,7 +1602,7 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata) /* 7 is highest priority & 0 is lowest. */ if (pdata->chan_priority == CHAN_PRIORITY_ASCENDING) - dwc->priority = r; + dwc->priority = nr_channels - i - 1; else dwc->priority = i; @@ -1622,6 +1621,7 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata) /* Hardware configuration */ if (autocfg) { unsigned int dwc_params; + unsigned int r = DW_DMA_MAX_NR_CHANNELS - i - 1; void __iomem *addr = chip->regs + r * sizeof(u32); dwc_params = dma_read_byaddr(addr, DWC_PARAMS); diff --git a/drivers/dma/idma64.c b/drivers/dma/idma64.c index 18c14e1f1414..48d6d9e94f67 100644 --- a/drivers/dma/idma64.c +++ b/drivers/dma/idma64.c @@ -355,23 +355,23 @@ static size_t idma64_active_desc_size(struct idma64_chan *idma64c) struct idma64_desc *desc = idma64c->desc; struct idma64_hw_desc *hw; size_t bytes = desc->length; - u64 llp; - u32 ctlhi; + u64 llp = channel_readq(idma64c, LLP); + u32 ctlhi = channel_readl(idma64c, CTL_HI); unsigned int i = 0; - llp = channel_readq(idma64c, LLP); do { hw = &desc->hw[i]; - } while ((hw->llp != llp) && (++i < desc->ndesc)); + if (hw->llp == llp) + break; + bytes -= hw->len; + } while (++i < desc->ndesc); if (!i) return bytes; - do { - bytes -= desc->hw[--i].len; - } while (i); + /* The current chunk is not fully transfered yet */ + bytes += desc->hw[--i].len; - ctlhi = channel_readl(idma64c, CTL_HI); return bytes - IDMA64C_CTLH_BLOCK_TS(ctlhi); } diff --git a/drivers/dma/mic_x100_dma.c b/drivers/dma/mic_x100_dma.c index 74d9db05a5ad..068e920ecb68 100644 --- a/drivers/dma/mic_x100_dma.c +++ b/drivers/dma/mic_x100_dma.c @@ -193,8 +193,16 @@ static void mic_dma_prog_intr(struct mic_dma_chan *ch) static int mic_dma_do_dma(struct mic_dma_chan *ch, int flags, dma_addr_t src, dma_addr_t dst, size_t len) { - if (-ENOMEM == mic_dma_prog_memcpy_desc(ch, src, dst, len)) + if (len && -ENOMEM == mic_dma_prog_memcpy_desc(ch, src, dst, len)) { return -ENOMEM; + } else { + /* 3 is the maximum number of status descriptors */ + int ret = mic_dma_avail_desc_ring_space(ch, 3); + + if (ret < 0) + return ret; + } + /* Above mic_dma_prog_memcpy_desc() makes sure we have enough space */ if (flags & DMA_PREP_FENCE) { mic_dma_prep_status_desc(&ch->desc_ring[ch->head], 0, @@ -270,6 +278,33 @@ allocate_tx(struct mic_dma_chan *ch) return tx; } +/* Program a status descriptor with dst as address and value to be written */ +static struct dma_async_tx_descriptor * +mic_dma_prep_status_lock(struct dma_chan *ch, dma_addr_t dst, u64 src_val, + unsigned long flags) +{ + struct mic_dma_chan *mic_ch = to_mic_dma_chan(ch); + int result; + + spin_lock(&mic_ch->prep_lock); + result = mic_dma_avail_desc_ring_space(mic_ch, 4); + if (result < 0) + goto error; + mic_dma_prep_status_desc(&mic_ch->desc_ring[mic_ch->head], src_val, dst, + false); + mic_dma_hw_ring_inc_head(mic_ch); + result = mic_dma_do_dma(mic_ch, flags, 0, 0, 0); + if (result < 0) + goto error; + + return allocate_tx(mic_ch); +error: + dev_err(mic_dma_ch_to_device(mic_ch), + "Error enqueueing dma status descriptor, error=%d\n", result); + spin_unlock(&mic_ch->prep_lock); + return NULL; +} + /* * Prepare a memcpy descriptor to be added to the ring. * Note that the temporary descriptor adds an extra overhead of copying the @@ -587,6 +622,8 @@ static int mic_dma_register_dma_device(struct mic_dma_device *mic_dma_dev, mic_dma_free_chan_resources; mic_dma_dev->dma_dev.device_tx_status = mic_dma_tx_status; mic_dma_dev->dma_dev.device_prep_dma_memcpy = mic_dma_prep_memcpy_lock; + mic_dma_dev->dma_dev.device_prep_dma_imm_data = + mic_dma_prep_status_lock; mic_dma_dev->dma_dev.device_prep_dma_interrupt = mic_dma_prep_interrupt_lock; mic_dma_dev->dma_dev.device_issue_pending = mic_dma_issue_pending; diff --git a/drivers/dma/pxa_dma.c b/drivers/dma/pxa_dma.c index 5cb61ce01036..fc4156afa070 100644 --- a/drivers/dma/pxa_dma.c +++ b/drivers/dma/pxa_dma.c @@ -473,8 +473,10 @@ static void pxad_free_phy(struct pxad_chan *chan) return; /* clear the channel mapping in DRCMR */ - reg = pxad_drcmr(chan->drcmr); - writel_relaxed(0, chan->phy->base + reg); + if (chan->drcmr <= DRCMR_CHLNUM) { + reg = pxad_drcmr(chan->drcmr); + writel_relaxed(0, chan->phy->base + reg); + } spin_lock_irqsave(&pdev->phy_lock, flags); for (i = 0; i < 32; i++) @@ -516,8 +518,10 @@ static void phy_enable(struct pxad_phy *phy, bool misaligned) "%s(); phy=%p(%d) misaligned=%d\n", __func__, phy, phy->idx, misaligned); - reg = pxad_drcmr(phy->vchan->drcmr); - writel_relaxed(DRCMR_MAPVLD | phy->idx, phy->base + reg); + if (phy->vchan->drcmr <= DRCMR_CHLNUM) { + reg = pxad_drcmr(phy->vchan->drcmr); + writel_relaxed(DRCMR_MAPVLD | phy->idx, phy->base + reg); + } dalgn = phy_readl_relaxed(phy, DALGN); if (misaligned) @@ -887,6 +891,7 @@ pxad_tx_prep(struct virt_dma_chan *vc, struct virt_dma_desc *vd, struct dma_async_tx_descriptor *tx; struct pxad_chan *chan = container_of(vc, struct pxad_chan, vc); + INIT_LIST_HEAD(&vd->node); tx = vchan_tx_prep(vc, vd, tx_flags); tx->tx_submit = pxad_tx_submit; dev_dbg(&chan->vc.chan.dev->device, @@ -910,14 +915,18 @@ static void pxad_get_config(struct pxad_chan *chan, width = chan->cfg.src_addr_width; dev_addr = chan->cfg.src_addr; *dev_src = dev_addr; - *dcmd |= PXA_DCMD_INCTRGADDR | PXA_DCMD_FLOWSRC; + *dcmd |= PXA_DCMD_INCTRGADDR; + if (chan->drcmr <= DRCMR_CHLNUM) + *dcmd |= PXA_DCMD_FLOWSRC; } if (dir == DMA_MEM_TO_DEV) { maxburst = chan->cfg.dst_maxburst; width = chan->cfg.dst_addr_width; dev_addr = chan->cfg.dst_addr; *dev_dst = dev_addr; - *dcmd |= PXA_DCMD_INCSRCADDR | PXA_DCMD_FLOWTRG; + *dcmd |= PXA_DCMD_INCSRCADDR; + if (chan->drcmr <= DRCMR_CHLNUM) + *dcmd |= PXA_DCMD_FLOWTRG; } if (dir == DMA_MEM_TO_MEM) *dcmd |= PXA_DCMD_BURST32 | PXA_DCMD_INCTRGADDR | @@ -1177,6 +1186,16 @@ static unsigned int pxad_residue(struct pxad_chan *chan, else curr = phy_readl_relaxed(chan->phy, DTADR); + /* + * curr has to be actually read before checking descriptor + * completion, so that a curr inside a status updater + * descriptor implies the following test returns true, and + * preventing reordering of curr load and the test. + */ + rmb(); + if (is_desc_completed(vd)) + goto out; + for (i = 0; i < sw_desc->nb_desc - 1; i++) { hw_desc = sw_desc->hw_desc[i]; if (sw_desc->hw_desc[0]->dcmd & PXA_DCMD_INCSRCADDR) diff --git a/drivers/dma/sun4i-dma.c b/drivers/dma/sun4i-dma.c index a1a500d96ff2..1661d518224a 100644 --- a/drivers/dma/sun4i-dma.c +++ b/drivers/dma/sun4i-dma.c @@ -599,13 +599,13 @@ get_next_cyclic_promise(struct sun4i_dma_contract *contract) static void sun4i_dma_free_contract(struct virt_dma_desc *vd) { struct sun4i_dma_contract *contract = to_sun4i_dma_contract(vd); - struct sun4i_dma_promise *promise; + struct sun4i_dma_promise *promise, *tmp; /* Free all the demands and completed demands */ - list_for_each_entry(promise, &contract->demands, list) + list_for_each_entry_safe(promise, tmp, &contract->demands, list) kfree(promise); - list_for_each_entry(promise, &contract->completed_demands, list) + list_for_each_entry_safe(promise, tmp, &contract->completed_demands, list) kfree(promise); kfree(contract); diff --git a/drivers/dma/xgene-dma.c b/drivers/dma/xgene-dma.c index b23e8d52d126..8d57b1b12e41 100644 --- a/drivers/dma/xgene-dma.c +++ b/drivers/dma/xgene-dma.c @@ -59,7 +59,6 @@ #define XGENE_DMA_RING_MEM_RAM_SHUTDOWN 0xD070 #define XGENE_DMA_RING_BLK_MEM_RDY 0xD074 #define XGENE_DMA_RING_BLK_MEM_RDY_VAL 0xFFFFFFFF -#define XGENE_DMA_RING_DESC_CNT(v) (((v) & 0x0001FFFE) >> 1) #define XGENE_DMA_RING_ID_GET(owner, num) (((owner) << 6) | (num)) #define XGENE_DMA_RING_DST_ID(v) ((1 << 10) | (v)) #define XGENE_DMA_RING_CMD_OFFSET 0x2C @@ -379,14 +378,6 @@ static u8 xgene_dma_encode_xor_flyby(u32 src_cnt) return flyby_type[src_cnt]; } -static u32 xgene_dma_ring_desc_cnt(struct xgene_dma_ring *ring) -{ - u32 __iomem *cmd_base = ring->cmd_base; - u32 ring_state = ioread32(&cmd_base[1]); - - return XGENE_DMA_RING_DESC_CNT(ring_state); -} - static void xgene_dma_set_src_buffer(__le64 *ext8, size_t *len, dma_addr_t *paddr) { @@ -659,15 +650,12 @@ static void xgene_dma_clean_running_descriptor(struct xgene_dma_chan *chan, dma_pool_free(chan->desc_pool, desc, desc->tx.phys); } -static int xgene_chan_xfer_request(struct xgene_dma_ring *ring, - struct xgene_dma_desc_sw *desc_sw) +static void xgene_chan_xfer_request(struct xgene_dma_chan *chan, + struct xgene_dma_desc_sw *desc_sw) { + struct xgene_dma_ring *ring = &chan->tx_ring; struct xgene_dma_desc_hw *desc_hw; - /* Check if can push more descriptor to hw for execution */ - if (xgene_dma_ring_desc_cnt(ring) > (ring->slots - 2)) - return -EBUSY; - /* Get hw descriptor from DMA tx ring */ desc_hw = &ring->desc_hw[ring->head]; @@ -694,11 +682,13 @@ static int xgene_chan_xfer_request(struct xgene_dma_ring *ring, memcpy(desc_hw, &desc_sw->desc2, sizeof(*desc_hw)); } + /* Increment the pending transaction count */ + chan->pending += ((desc_sw->flags & + XGENE_DMA_FLAG_64B_DESC) ? 2 : 1); + /* Notify the hw that we have descriptor ready for execution */ iowrite32((desc_sw->flags & XGENE_DMA_FLAG_64B_DESC) ? 2 : 1, ring->cmd); - - return 0; } /** @@ -710,7 +700,6 @@ static int xgene_chan_xfer_request(struct xgene_dma_ring *ring, static void xgene_chan_xfer_ld_pending(struct xgene_dma_chan *chan) { struct xgene_dma_desc_sw *desc_sw, *_desc_sw; - int ret; /* * If the list of pending descriptors is empty, then we @@ -735,18 +724,13 @@ static void xgene_chan_xfer_ld_pending(struct xgene_dma_chan *chan) if (chan->pending >= chan->max_outstanding) return; - ret = xgene_chan_xfer_request(&chan->tx_ring, desc_sw); - if (ret) - return; + xgene_chan_xfer_request(chan, desc_sw); /* * Delete this element from ld pending queue and append it to * ld running queue */ list_move_tail(&desc_sw->node, &chan->ld_running); - - /* Increment the pending transaction count */ - chan->pending++; } } @@ -821,7 +805,8 @@ static void xgene_dma_cleanup_descriptors(struct xgene_dma_chan *chan) * Decrement the pending transaction count * as we have processed one */ - chan->pending--; + chan->pending -= ((desc_sw->flags & + XGENE_DMA_FLAG_64B_DESC) ? 2 : 1); /* * Delete this node from ld running queue and append it to @@ -1421,15 +1406,18 @@ static int xgene_dma_create_ring_one(struct xgene_dma_chan *chan, struct xgene_dma_ring *ring, enum xgene_dma_ring_cfgsize cfgsize) { + int ret; + /* Setup DMA ring descriptor variables */ ring->pdma = chan->pdma; ring->cfgsize = cfgsize; ring->num = chan->pdma->ring_num++; ring->id = XGENE_DMA_RING_ID_GET(ring->owner, ring->buf_num); - ring->size = xgene_dma_get_ring_size(chan, cfgsize); - if (ring->size <= 0) - return ring->size; + ret = xgene_dma_get_ring_size(chan, cfgsize); + if (ret <= 0) + return ret; + ring->size = ret; /* Allocate memory for DMA ring descriptor */ ring->desc_vaddr = dma_zalloc_coherent(chan->dev, ring->size, @@ -1482,7 +1470,7 @@ static int xgene_dma_create_chan_rings(struct xgene_dma_chan *chan) tx_ring->id, tx_ring->num, tx_ring->desc_vaddr); /* Set the max outstanding request possible to this channel */ - chan->max_outstanding = rx_ring->slots; + chan->max_outstanding = tx_ring->slots; return ret; } diff --git a/drivers/dma/zx296702_dma.c b/drivers/dma/zx296702_dma.c index 39915a6b7986..c017fcd8e07c 100644 --- a/drivers/dma/zx296702_dma.c +++ b/drivers/dma/zx296702_dma.c @@ -739,7 +739,7 @@ static struct dma_chan *zx_of_dma_simple_xlate(struct of_phandle_args *dma_spec, struct dma_chan *chan; struct zx_dma_chan *c; - if (request > d->dma_requests) + if (request >= d->dma_requests) return NULL; chan = dma_get_any_slave_channel(&d->slave); diff --git a/drivers/firmware/efi/libstub/arm-stub.c b/drivers/firmware/efi/libstub/arm-stub.c index e29560e6b40b..950c87f5d279 100644 --- a/drivers/firmware/efi/libstub/arm-stub.c +++ b/drivers/firmware/efi/libstub/arm-stub.c @@ -13,6 +13,7 @@ */ #include <linux/efi.h> +#include <linux/sort.h> #include <asm/efi.h> #include "efistub.h" @@ -305,6 +306,44 @@ fail: */ #define EFI_RT_VIRTUAL_BASE 0x40000000 +static int cmp_mem_desc(const void *l, const void *r) +{ + const efi_memory_desc_t *left = l, *right = r; + + return (left->phys_addr > right->phys_addr) ? 1 : -1; +} + +/* + * Returns whether region @left ends exactly where region @right starts, + * or false if either argument is NULL. + */ +static bool regions_are_adjacent(efi_memory_desc_t *left, + efi_memory_desc_t *right) +{ + u64 left_end; + + if (left == NULL || right == NULL) + return false; + + left_end = left->phys_addr + left->num_pages * EFI_PAGE_SIZE; + + return left_end == right->phys_addr; +} + +/* + * Returns whether region @left and region @right have compatible memory type + * mapping attributes, and are both EFI_MEMORY_RUNTIME regions. + */ +static bool regions_have_compatible_memory_type_attrs(efi_memory_desc_t *left, + efi_memory_desc_t *right) +{ + static const u64 mem_type_mask = EFI_MEMORY_WB | EFI_MEMORY_WT | + EFI_MEMORY_WC | EFI_MEMORY_UC | + EFI_MEMORY_RUNTIME; + + return ((left->attribute ^ right->attribute) & mem_type_mask) == 0; +} + /* * efi_get_virtmap() - create a virtual mapping for the EFI memory map * @@ -317,33 +356,52 @@ void efi_get_virtmap(efi_memory_desc_t *memory_map, unsigned long map_size, int *count) { u64 efi_virt_base = EFI_RT_VIRTUAL_BASE; - efi_memory_desc_t *out = runtime_map; + efi_memory_desc_t *in, *prev = NULL, *out = runtime_map; int l; - for (l = 0; l < map_size; l += desc_size) { - efi_memory_desc_t *in = (void *)memory_map + l; + /* + * To work around potential issues with the Properties Table feature + * introduced in UEFI 2.5, which may split PE/COFF executable images + * in memory into several RuntimeServicesCode and RuntimeServicesData + * regions, we need to preserve the relative offsets between adjacent + * EFI_MEMORY_RUNTIME regions with the same memory type attributes. + * The easiest way to find adjacent regions is to sort the memory map + * before traversing it. + */ + sort(memory_map, map_size / desc_size, desc_size, cmp_mem_desc, NULL); + + for (l = 0; l < map_size; l += desc_size, prev = in) { u64 paddr, size; + in = (void *)memory_map + l; if (!(in->attribute & EFI_MEMORY_RUNTIME)) continue; + paddr = in->phys_addr; + size = in->num_pages * EFI_PAGE_SIZE; + /* * Make the mapping compatible with 64k pages: this allows * a 4k page size kernel to kexec a 64k page size kernel and * vice versa. */ - paddr = round_down(in->phys_addr, SZ_64K); - size = round_up(in->num_pages * EFI_PAGE_SIZE + - in->phys_addr - paddr, SZ_64K); - - /* - * Avoid wasting memory on PTEs by choosing a virtual base that - * is compatible with section mappings if this region has the - * appropriate size and physical alignment. (Sections are 2 MB - * on 4k granule kernels) - */ - if (IS_ALIGNED(in->phys_addr, SZ_2M) && size >= SZ_2M) - efi_virt_base = round_up(efi_virt_base, SZ_2M); + if (!regions_are_adjacent(prev, in) || + !regions_have_compatible_memory_type_attrs(prev, in)) { + + paddr = round_down(in->phys_addr, SZ_64K); + size += in->phys_addr - paddr; + + /* + * Avoid wasting memory on PTEs by choosing a virtual + * base that is compatible with section mappings if this + * region has the appropriate size and physical + * alignment. (Sections are 2 MB on 4k granule kernels) + */ + if (IS_ALIGNED(in->phys_addr, SZ_2M) && size >= SZ_2M) + efi_virt_base = round_up(efi_virt_base, SZ_2M); + else + efi_virt_base = round_up(efi_virt_base, SZ_64K); + } in->virt_addr = efi_virt_base + in->phys_addr - paddr; efi_virt_base += size; diff --git a/drivers/fpga/Kconfig b/drivers/fpga/Kconfig new file mode 100644 index 000000000000..dfc1f1ec093b --- /dev/null +++ b/drivers/fpga/Kconfig @@ -0,0 +1,24 @@ +# +# FPGA framework configuration +# + +menu "FPGA Configuration Support" + +config FPGA + tristate "FPGA Configuration Framework" + help + Say Y here if you want support for configuring FPGAs from the + kernel. The FPGA framework adds a FPGA manager class and FPGA + manager drivers. + +if FPGA + +config FPGA_MGR_SOCFPGA + tristate "Altera SOCFPGA FPGA Manager" + depends on ARCH_SOCFPGA + help + FPGA manager driver support for Altera SOCFPGA. + +endif # FPGA + +endmenu diff --git a/drivers/fpga/Makefile b/drivers/fpga/Makefile new file mode 100644 index 000000000000..ba6c5c597a4d --- /dev/null +++ b/drivers/fpga/Makefile @@ -0,0 +1,9 @@ +# +# Makefile for the fpga framework and fpga manager drivers. +# + +# Core FPGA Manager Framework +obj-$(CONFIG_FPGA) += fpga-mgr.o + +# FPGA Manager Drivers +obj-$(CONFIG_FPGA_MGR_SOCFPGA) += socfpga.o diff --git a/drivers/fpga/fpga-mgr.c b/drivers/fpga/fpga-mgr.c new file mode 100644 index 000000000000..25261636687c --- /dev/null +++ b/drivers/fpga/fpga-mgr.c @@ -0,0 +1,382 @@ +/* + * FPGA Manager Core + * + * Copyright (C) 2013-2015 Altera Corporation + * + * With code from the mailing list: + * Copyright (C) 2013 Xilinx, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see <http://www.gnu.org/licenses/>. + */ +#include <linux/firmware.h> +#include <linux/fpga/fpga-mgr.h> +#include <linux/idr.h> +#include <linux/module.h> +#include <linux/of.h> +#include <linux/mutex.h> +#include <linux/slab.h> + +static DEFINE_IDA(fpga_mgr_ida); +static struct class *fpga_mgr_class; + +/** + * fpga_mgr_buf_load - load fpga from image in buffer + * @mgr: fpga manager + * @flags: flags setting fpga confuration modes + * @buf: buffer contain fpga image + * @count: byte count of buf + * + * Step the low level fpga manager through the device-specific steps of getting + * an FPGA ready to be configured, writing the image to it, then doing whatever + * post-configuration steps necessary. + * + * Return: 0 on success, negative error code otherwise. + */ +int fpga_mgr_buf_load(struct fpga_manager *mgr, u32 flags, const char *buf, + size_t count) +{ + struct device *dev = &mgr->dev; + int ret; + + if (!mgr) + return -ENODEV; + + /* + * Call the low level driver's write_init function. This will do the + * device-specific things to get the FPGA into the state where it is + * ready to receive an FPGA image. + */ + mgr->state = FPGA_MGR_STATE_WRITE_INIT; + ret = mgr->mops->write_init(mgr, flags, buf, count); + if (ret) { + dev_err(dev, "Error preparing FPGA for writing\n"); + mgr->state = FPGA_MGR_STATE_WRITE_INIT_ERR; + return ret; + } + + /* + * Write the FPGA image to the FPGA. + */ + mgr->state = FPGA_MGR_STATE_WRITE; + ret = mgr->mops->write(mgr, buf, count); + if (ret) { + dev_err(dev, "Error while writing image data to FPGA\n"); + mgr->state = FPGA_MGR_STATE_WRITE_ERR; + return ret; + } + + /* + * After all the FPGA image has been written, do the device specific + * steps to finish and set the FPGA into operating mode. + */ + mgr->state = FPGA_MGR_STATE_WRITE_COMPLETE; + ret = mgr->mops->write_complete(mgr, flags); + if (ret) { + dev_err(dev, "Error after writing image data to FPGA\n"); + mgr->state = FPGA_MGR_STATE_WRITE_COMPLETE_ERR; + return ret; + } + mgr->state = FPGA_MGR_STATE_OPERATING; + + return 0; +} +EXPORT_SYMBOL_GPL(fpga_mgr_buf_load); + +/** + * fpga_mgr_firmware_load - request firmware and load to fpga + * @mgr: fpga manager + * @flags: flags setting fpga confuration modes + * @image_name: name of image file on the firmware search path + * + * Request an FPGA image using the firmware class, then write out to the FPGA. + * Update the state before each step to provide info on what step failed if + * there is a failure. + * + * Return: 0 on success, negative error code otherwise. + */ +int fpga_mgr_firmware_load(struct fpga_manager *mgr, u32 flags, + const char *image_name) +{ + struct device *dev = &mgr->dev; + const struct firmware *fw; + int ret; + + if (!mgr) + return -ENODEV; + + dev_info(dev, "writing %s to %s\n", image_name, mgr->name); + + mgr->state = FPGA_MGR_STATE_FIRMWARE_REQ; + + ret = request_firmware(&fw, image_name, dev); + if (ret) { + mgr->state = FPGA_MGR_STATE_FIRMWARE_REQ_ERR; + dev_err(dev, "Error requesting firmware %s\n", image_name); + return ret; + } + + ret = fpga_mgr_buf_load(mgr, flags, fw->data, fw->size); + if (ret) + return ret; + + release_firmware(fw); + + return 0; +} +EXPORT_SYMBOL_GPL(fpga_mgr_firmware_load); + +static const char * const state_str[] = { + [FPGA_MGR_STATE_UNKNOWN] = "unknown", + [FPGA_MGR_STATE_POWER_OFF] = "power off", + [FPGA_MGR_STATE_POWER_UP] = "power up", + [FPGA_MGR_STATE_RESET] = "reset", + + /* requesting FPGA image from firmware */ + [FPGA_MGR_STATE_FIRMWARE_REQ] = "firmware request", + [FPGA_MGR_STATE_FIRMWARE_REQ_ERR] = "firmware request error", + + /* Preparing FPGA to receive image */ + [FPGA_MGR_STATE_WRITE_INIT] = "write init", + [FPGA_MGR_STATE_WRITE_INIT_ERR] = "write init error", + + /* Writing image to FPGA */ + [FPGA_MGR_STATE_WRITE] = "write", + [FPGA_MGR_STATE_WRITE_ERR] = "write error", + + /* Finishing configuration after image has been written */ + [FPGA_MGR_STATE_WRITE_COMPLETE] = "write complete", + [FPGA_MGR_STATE_WRITE_COMPLETE_ERR] = "write complete error", + + /* FPGA reports to be in normal operating mode */ + [FPGA_MGR_STATE_OPERATING] = "operating", +}; + +static ssize_t name_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct fpga_manager *mgr = to_fpga_manager(dev); + + return sprintf(buf, "%s\n", mgr->name); +} + +static ssize_t state_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct fpga_manager *mgr = to_fpga_manager(dev); + + return sprintf(buf, "%s\n", state_str[mgr->state]); +} + +static DEVICE_ATTR_RO(name); +static DEVICE_ATTR_RO(state); + +static struct attribute *fpga_mgr_attrs[] = { + &dev_attr_name.attr, + &dev_attr_state.attr, + NULL, +}; +ATTRIBUTE_GROUPS(fpga_mgr); + +static int fpga_mgr_of_node_match(struct device *dev, const void *data) +{ + return dev->of_node == data; +} + +/** + * of_fpga_mgr_get - get an exclusive reference to a fpga mgr + * @node: device node + * + * Given a device node, get an exclusive reference to a fpga mgr. + * + * Return: fpga manager struct or IS_ERR() condition containing error code. + */ +struct fpga_manager *of_fpga_mgr_get(struct device_node *node) +{ + struct fpga_manager *mgr; + struct device *dev; + + if (!node) + return ERR_PTR(-EINVAL); + + dev = class_find_device(fpga_mgr_class, NULL, node, + fpga_mgr_of_node_match); + if (!dev) + return ERR_PTR(-ENODEV); + + mgr = to_fpga_manager(dev); + put_device(dev); + if (!mgr) + return ERR_PTR(-ENODEV); + + /* Get exclusive use of fpga manager */ + if (!mutex_trylock(&mgr->ref_mutex)) + return ERR_PTR(-EBUSY); + + if (!try_module_get(THIS_MODULE)) { + mutex_unlock(&mgr->ref_mutex); + return ERR_PTR(-ENODEV); + } + + return mgr; +} +EXPORT_SYMBOL_GPL(of_fpga_mgr_get); + +/** + * fpga_mgr_put - release a reference to a fpga manager + * @mgr: fpga manager structure + */ +void fpga_mgr_put(struct fpga_manager *mgr) +{ + if (mgr) { + module_put(THIS_MODULE); + mutex_unlock(&mgr->ref_mutex); + } +} +EXPORT_SYMBOL_GPL(fpga_mgr_put); + +/** + * fpga_mgr_register - register a low level fpga manager driver + * @dev: fpga manager device from pdev + * @name: fpga manager name + * @mops: pointer to structure of fpga manager ops + * @priv: fpga manager private data + * + * Return: 0 on success, negative error code otherwise. + */ +int fpga_mgr_register(struct device *dev, const char *name, + const struct fpga_manager_ops *mops, + void *priv) +{ + struct fpga_manager *mgr; + const char *dt_label; + int id, ret; + + if (!mops || !mops->write_init || !mops->write || + !mops->write_complete || !mops->state) { + dev_err(dev, "Attempt to register without fpga_manager_ops\n"); + return -EINVAL; + } + + if (!name || !strlen(name)) { + dev_err(dev, "Attempt to register with no name!\n"); + return -EINVAL; + } + + mgr = kzalloc(sizeof(*mgr), GFP_KERNEL); + if (!mgr) + return -ENOMEM; + + id = ida_simple_get(&fpga_mgr_ida, 0, 0, GFP_KERNEL); + if (id < 0) { + ret = id; + goto error_kfree; + } + + mutex_init(&mgr->ref_mutex); + + mgr->name = name; + mgr->mops = mops; + mgr->priv = priv; + + /* + * Initialize framework state by requesting low level driver read state + * from device. FPGA may be in reset mode or may have been programmed + * by bootloader or EEPROM. + */ + mgr->state = mgr->mops->state(mgr); + + device_initialize(&mgr->dev); + mgr->dev.class = fpga_mgr_class; + mgr->dev.parent = dev; + mgr->dev.of_node = dev->of_node; + mgr->dev.id = id; + dev_set_drvdata(dev, mgr); + + dt_label = of_get_property(mgr->dev.of_node, "label", NULL); + if (dt_label) + ret = dev_set_name(&mgr->dev, "%s", dt_label); + else + ret = dev_set_name(&mgr->dev, "fpga%d", id); + + ret = device_add(&mgr->dev); + if (ret) + goto error_device; + + dev_info(&mgr->dev, "%s registered\n", mgr->name); + + return 0; + +error_device: + ida_simple_remove(&fpga_mgr_ida, id); +error_kfree: + kfree(mgr); + + return ret; +} +EXPORT_SYMBOL_GPL(fpga_mgr_register); + +/** + * fpga_mgr_unregister - unregister a low level fpga manager driver + * @dev: fpga manager device from pdev + */ +void fpga_mgr_unregister(struct device *dev) +{ + struct fpga_manager *mgr = dev_get_drvdata(dev); + + dev_info(&mgr->dev, "%s %s\n", __func__, mgr->name); + + /* + * If the low level driver provides a method for putting fpga into + * a desired state upon unregister, do it. + */ + if (mgr->mops->fpga_remove) + mgr->mops->fpga_remove(mgr); + + device_unregister(&mgr->dev); +} +EXPORT_SYMBOL_GPL(fpga_mgr_unregister); + +static void fpga_mgr_dev_release(struct device *dev) +{ + struct fpga_manager *mgr = to_fpga_manager(dev); + + ida_simple_remove(&fpga_mgr_ida, mgr->dev.id); + kfree(mgr); +} + +static int __init fpga_mgr_class_init(void) +{ + pr_info("FPGA manager framework\n"); + + fpga_mgr_class = class_create(THIS_MODULE, "fpga_manager"); + if (IS_ERR(fpga_mgr_class)) + return PTR_ERR(fpga_mgr_class); + + fpga_mgr_class->dev_groups = fpga_mgr_groups; + fpga_mgr_class->dev_release = fpga_mgr_dev_release; + + return 0; +} + +static void __exit fpga_mgr_class_exit(void) +{ + class_destroy(fpga_mgr_class); + ida_destroy(&fpga_mgr_ida); +} + +MODULE_AUTHOR("Alan Tull <atull@opensource.altera.com>"); +MODULE_DESCRIPTION("FPGA manager framework"); +MODULE_LICENSE("GPL v2"); + +subsys_initcall(fpga_mgr_class_init); +module_exit(fpga_mgr_class_exit); diff --git a/drivers/fpga/socfpga.c b/drivers/fpga/socfpga.c new file mode 100644 index 000000000000..706b80d25266 --- /dev/null +++ b/drivers/fpga/socfpga.c @@ -0,0 +1,616 @@ +/* + * FPGA Manager Driver for Altera SOCFPGA + * + * Copyright (C) 2013-2015 Altera Corporation + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see <http://www.gnu.org/licenses/>. + */ +#include <linux/completion.h> +#include <linux/delay.h> +#include <linux/fpga/fpga-mgr.h> +#include <linux/interrupt.h> +#include <linux/io.h> +#include <linux/module.h> +#include <linux/of_address.h> +#include <linux/of_irq.h> +#include <linux/pm.h> + +/* Register offsets */ +#define SOCFPGA_FPGMGR_STAT_OFST 0x0 +#define SOCFPGA_FPGMGR_CTL_OFST 0x4 +#define SOCFPGA_FPGMGR_DCLKCNT_OFST 0x8 +#define SOCFPGA_FPGMGR_DCLKSTAT_OFST 0xc +#define SOCFPGA_FPGMGR_GPIO_INTEN_OFST 0x830 +#define SOCFPGA_FPGMGR_GPIO_INTMSK_OFST 0x834 +#define SOCFPGA_FPGMGR_GPIO_INTTYPE_LEVEL_OFST 0x838 +#define SOCFPGA_FPGMGR_GPIO_INT_POL_OFST 0x83c +#define SOCFPGA_FPGMGR_GPIO_INTSTAT_OFST 0x840 +#define SOCFPGA_FPGMGR_GPIO_RAW_INTSTAT_OFST 0x844 +#define SOCFPGA_FPGMGR_GPIO_PORTA_EOI_OFST 0x84c +#define SOCFPGA_FPGMGR_GPIO_EXT_PORTA_OFST 0x850 + +/* Register bit defines */ +/* SOCFPGA_FPGMGR_STAT register mode field values */ +#define SOCFPGA_FPGMGR_STAT_POWER_UP 0x0 /*ramping*/ +#define SOCFPGA_FPGMGR_STAT_RESET 0x1 +#define SOCFPGA_FPGMGR_STAT_CFG 0x2 +#define SOCFPGA_FPGMGR_STAT_INIT 0x3 +#define SOCFPGA_FPGMGR_STAT_USER_MODE 0x4 +#define SOCFPGA_FPGMGR_STAT_UNKNOWN 0x5 +#define SOCFPGA_FPGMGR_STAT_STATE_MASK 0x7 +/* This is a flag value that doesn't really happen in this register field */ +#define SOCFPGA_FPGMGR_STAT_POWER_OFF 0x0 + +#define MSEL_PP16_FAST_NOAES_NODC 0x0 +#define MSEL_PP16_FAST_AES_NODC 0x1 +#define MSEL_PP16_FAST_AESOPT_DC 0x2 +#define MSEL_PP16_SLOW_NOAES_NODC 0x4 +#define MSEL_PP16_SLOW_AES_NODC 0x5 +#define MSEL_PP16_SLOW_AESOPT_DC 0x6 +#define MSEL_PP32_FAST_NOAES_NODC 0x8 +#define MSEL_PP32_FAST_AES_NODC 0x9 +#define MSEL_PP32_FAST_AESOPT_DC 0xa +#define MSEL_PP32_SLOW_NOAES_NODC 0xc +#define MSEL_PP32_SLOW_AES_NODC 0xd +#define MSEL_PP32_SLOW_AESOPT_DC 0xe +#define SOCFPGA_FPGMGR_STAT_MSEL_MASK 0x000000f8 +#define SOCFPGA_FPGMGR_STAT_MSEL_SHIFT 3 + +/* SOCFPGA_FPGMGR_CTL register */ +#define SOCFPGA_FPGMGR_CTL_EN 0x00000001 +#define SOCFPGA_FPGMGR_CTL_NCE 0x00000002 +#define SOCFPGA_FPGMGR_CTL_NCFGPULL 0x00000004 + +#define CDRATIO_X1 0x00000000 +#define CDRATIO_X2 0x00000040 +#define CDRATIO_X4 0x00000080 +#define CDRATIO_X8 0x000000c0 +#define SOCFPGA_FPGMGR_CTL_CDRATIO_MASK 0x000000c0 + +#define SOCFPGA_FPGMGR_CTL_AXICFGEN 0x00000100 + +#define CFGWDTH_16 0x00000000 +#define CFGWDTH_32 0x00000200 +#define SOCFPGA_FPGMGR_CTL_CFGWDTH_MASK 0x00000200 + +/* SOCFPGA_FPGMGR_DCLKSTAT register */ +#define SOCFPGA_FPGMGR_DCLKSTAT_DCNTDONE_E_DONE 0x1 + +/* SOCFPGA_FPGMGR_GPIO_* registers share the same bit positions */ +#define SOCFPGA_FPGMGR_MON_NSTATUS 0x0001 +#define SOCFPGA_FPGMGR_MON_CONF_DONE 0x0002 +#define SOCFPGA_FPGMGR_MON_INIT_DONE 0x0004 +#define SOCFPGA_FPGMGR_MON_CRC_ERROR 0x0008 +#define SOCFPGA_FPGMGR_MON_CVP_CONF_DONE 0x0010 +#define SOCFPGA_FPGMGR_MON_PR_READY 0x0020 +#define SOCFPGA_FPGMGR_MON_PR_ERROR 0x0040 +#define SOCFPGA_FPGMGR_MON_PR_DONE 0x0080 +#define SOCFPGA_FPGMGR_MON_NCONFIG_PIN 0x0100 +#define SOCFPGA_FPGMGR_MON_NSTATUS_PIN 0x0200 +#define SOCFPGA_FPGMGR_MON_CONF_DONE_PIN 0x0400 +#define SOCFPGA_FPGMGR_MON_FPGA_POWER_ON 0x0800 +#define SOCFPGA_FPGMGR_MON_STATUS_MASK 0x0fff + +#define SOCFPGA_FPGMGR_NUM_SUPPLIES 3 +#define SOCFPGA_RESUME_TIMEOUT 3 + +/* In power-up order. Reverse for power-down. */ +static const char *supply_names[SOCFPGA_FPGMGR_NUM_SUPPLIES] __maybe_unused = { + "FPGA-1.5V", + "FPGA-1.1V", + "FPGA-2.5V", +}; + +struct socfpga_fpga_priv { + void __iomem *fpga_base_addr; + void __iomem *fpga_data_addr; + struct completion status_complete; + int irq; +}; + +struct cfgmgr_mode { + /* Values to set in the CTRL register */ + u32 ctrl; + + /* flag that this table entry is a valid mode */ + bool valid; +}; + +/* For SOCFPGA_FPGMGR_STAT_MSEL field */ +static struct cfgmgr_mode cfgmgr_modes[] = { + [MSEL_PP16_FAST_NOAES_NODC] = { CFGWDTH_16 | CDRATIO_X1, 1 }, + [MSEL_PP16_FAST_AES_NODC] = { CFGWDTH_16 | CDRATIO_X2, 1 }, + [MSEL_PP16_FAST_AESOPT_DC] = { CFGWDTH_16 | CDRATIO_X4, 1 }, + [MSEL_PP16_SLOW_NOAES_NODC] = { CFGWDTH_16 | CDRATIO_X1, 1 }, + [MSEL_PP16_SLOW_AES_NODC] = { CFGWDTH_16 | CDRATIO_X2, 1 }, + [MSEL_PP16_SLOW_AESOPT_DC] = { CFGWDTH_16 | CDRATIO_X4, 1 }, + [MSEL_PP32_FAST_NOAES_NODC] = { CFGWDTH_32 | CDRATIO_X1, 1 }, + [MSEL_PP32_FAST_AES_NODC] = { CFGWDTH_32 | CDRATIO_X4, 1 }, + [MSEL_PP32_FAST_AESOPT_DC] = { CFGWDTH_32 | CDRATIO_X8, 1 }, + [MSEL_PP32_SLOW_NOAES_NODC] = { CFGWDTH_32 | CDRATIO_X1, 1 }, + [MSEL_PP32_SLOW_AES_NODC] = { CFGWDTH_32 | CDRATIO_X4, 1 }, + [MSEL_PP32_SLOW_AESOPT_DC] = { CFGWDTH_32 | CDRATIO_X8, 1 }, +}; + +static u32 socfpga_fpga_readl(struct socfpga_fpga_priv *priv, u32 reg_offset) +{ + return readl(priv->fpga_base_addr + reg_offset); +} + +static void socfpga_fpga_writel(struct socfpga_fpga_priv *priv, u32 reg_offset, + u32 value) +{ + writel(value, priv->fpga_base_addr + reg_offset); +} + +static u32 socfpga_fpga_raw_readl(struct socfpga_fpga_priv *priv, + u32 reg_offset) +{ + return __raw_readl(priv->fpga_base_addr + reg_offset); +} + +static void socfpga_fpga_raw_writel(struct socfpga_fpga_priv *priv, + u32 reg_offset, u32 value) +{ + __raw_writel(value, priv->fpga_base_addr + reg_offset); +} + +static void socfpga_fpga_data_writel(struct socfpga_fpga_priv *priv, u32 value) +{ + writel(value, priv->fpga_data_addr); +} + +static inline void socfpga_fpga_set_bitsl(struct socfpga_fpga_priv *priv, + u32 offset, u32 bits) +{ + u32 val; + + val = socfpga_fpga_readl(priv, offset); + val |= bits; + socfpga_fpga_writel(priv, offset, val); +} + +static inline void socfpga_fpga_clr_bitsl(struct socfpga_fpga_priv *priv, + u32 offset, u32 bits) +{ + u32 val; + + val = socfpga_fpga_readl(priv, offset); + val &= ~bits; + socfpga_fpga_writel(priv, offset, val); +} + +static u32 socfpga_fpga_mon_status_get(struct socfpga_fpga_priv *priv) +{ + return socfpga_fpga_readl(priv, SOCFPGA_FPGMGR_GPIO_EXT_PORTA_OFST) & + SOCFPGA_FPGMGR_MON_STATUS_MASK; +} + +static u32 socfpga_fpga_state_get(struct socfpga_fpga_priv *priv) +{ + u32 status = socfpga_fpga_mon_status_get(priv); + + if ((status & SOCFPGA_FPGMGR_MON_FPGA_POWER_ON) == 0) + return SOCFPGA_FPGMGR_STAT_POWER_OFF; + + return socfpga_fpga_readl(priv, SOCFPGA_FPGMGR_STAT_OFST) & + SOCFPGA_FPGMGR_STAT_STATE_MASK; +} + +static void socfpga_fpga_clear_done_status(struct socfpga_fpga_priv *priv) +{ + socfpga_fpga_writel(priv, SOCFPGA_FPGMGR_DCLKSTAT_OFST, + SOCFPGA_FPGMGR_DCLKSTAT_DCNTDONE_E_DONE); +} + +/* + * Set the DCLKCNT, wait for DCLKSTAT to report the count completed, and clear + * the complete status. + */ +static int socfpga_fpga_dclk_set_and_wait_clear(struct socfpga_fpga_priv *priv, + u32 count) +{ + int timeout = 2; + u32 done; + + /* Clear any existing DONE status. */ + if (socfpga_fpga_readl(priv, SOCFPGA_FPGMGR_DCLKSTAT_OFST)) + socfpga_fpga_clear_done_status(priv); + + /* Issue the DCLK count. */ + socfpga_fpga_writel(priv, SOCFPGA_FPGMGR_DCLKCNT_OFST, count); + + /* Poll DCLKSTAT to see if it completed in the timeout period. */ + do { + done = socfpga_fpga_readl(priv, SOCFPGA_FPGMGR_DCLKSTAT_OFST); + if (done == SOCFPGA_FPGMGR_DCLKSTAT_DCNTDONE_E_DONE) { + socfpga_fpga_clear_done_status(priv); + return 0; + } + udelay(1); + } while (timeout--); + + return -ETIMEDOUT; +} + +static int socfpga_fpga_wait_for_state(struct socfpga_fpga_priv *priv, + u32 state) +{ + int timeout = 2; + + /* + * HW doesn't support an interrupt for changes in state, so poll to see + * if it matches the requested state within the timeout period. + */ + do { + if ((socfpga_fpga_state_get(priv) & state) != 0) + return 0; + msleep(20); + } while (timeout--); + + return -ETIMEDOUT; +} + +static void socfpga_fpga_enable_irqs(struct socfpga_fpga_priv *priv, u32 irqs) +{ + /* set irqs to level sensitive */ + socfpga_fpga_writel(priv, SOCFPGA_FPGMGR_GPIO_INTTYPE_LEVEL_OFST, 0); + + /* set interrupt polarity */ + socfpga_fpga_writel(priv, SOCFPGA_FPGMGR_GPIO_INT_POL_OFST, irqs); + + /* clear irqs */ + socfpga_fpga_writel(priv, SOCFPGA_FPGMGR_GPIO_PORTA_EOI_OFST, irqs); + + /* unmask interrupts */ + socfpga_fpga_writel(priv, SOCFPGA_FPGMGR_GPIO_INTMSK_OFST, 0); + + /* enable interrupts */ + socfpga_fpga_writel(priv, SOCFPGA_FPGMGR_GPIO_INTEN_OFST, irqs); +} + +static void socfpga_fpga_disable_irqs(struct socfpga_fpga_priv *priv) +{ + socfpga_fpga_writel(priv, SOCFPGA_FPGMGR_GPIO_INTEN_OFST, 0); +} + +static irqreturn_t socfpga_fpga_isr(int irq, void *dev_id) +{ + struct socfpga_fpga_priv *priv = dev_id; + u32 irqs, st; + bool conf_done, nstatus; + + /* clear irqs */ + irqs = socfpga_fpga_raw_readl(priv, SOCFPGA_FPGMGR_GPIO_INTSTAT_OFST); + + socfpga_fpga_raw_writel(priv, SOCFPGA_FPGMGR_GPIO_PORTA_EOI_OFST, irqs); + + st = socfpga_fpga_raw_readl(priv, SOCFPGA_FPGMGR_GPIO_EXT_PORTA_OFST); + conf_done = (st & SOCFPGA_FPGMGR_MON_CONF_DONE) != 0; + nstatus = (st & SOCFPGA_FPGMGR_MON_NSTATUS) != 0; + + /* success */ + if (conf_done && nstatus) { + /* disable irqs */ + socfpga_fpga_raw_writel(priv, + SOCFPGA_FPGMGR_GPIO_INTEN_OFST, 0); + complete(&priv->status_complete); + } + + return IRQ_HANDLED; +} + +static int socfpga_fpga_wait_for_config_done(struct socfpga_fpga_priv *priv) +{ + int timeout, ret = 0; + + socfpga_fpga_disable_irqs(priv); + init_completion(&priv->status_complete); + socfpga_fpga_enable_irqs(priv, SOCFPGA_FPGMGR_MON_CONF_DONE); + + timeout = wait_for_completion_interruptible_timeout( + &priv->status_complete, + msecs_to_jiffies(10)); + if (timeout == 0) + ret = -ETIMEDOUT; + + socfpga_fpga_disable_irqs(priv); + return ret; +} + +static int socfpga_fpga_cfg_mode_get(struct socfpga_fpga_priv *priv) +{ + u32 msel; + + msel = socfpga_fpga_readl(priv, SOCFPGA_FPGMGR_STAT_OFST); + msel &= SOCFPGA_FPGMGR_STAT_MSEL_MASK; + msel >>= SOCFPGA_FPGMGR_STAT_MSEL_SHIFT; + + /* Check that this MSEL setting is supported */ + if ((msel >= ARRAY_SIZE(cfgmgr_modes)) || !cfgmgr_modes[msel].valid) + return -EINVAL; + + return msel; +} + +static int socfpga_fpga_cfg_mode_set(struct socfpga_fpga_priv *priv) +{ + u32 ctrl_reg; + int mode; + + /* get value from MSEL pins */ + mode = socfpga_fpga_cfg_mode_get(priv); + if (mode < 0) + return mode; + + /* Adjust CTRL for the CDRATIO */ + ctrl_reg = socfpga_fpga_readl(priv, SOCFPGA_FPGMGR_CTL_OFST); + ctrl_reg &= ~SOCFPGA_FPGMGR_CTL_CDRATIO_MASK; + ctrl_reg &= ~SOCFPGA_FPGMGR_CTL_CFGWDTH_MASK; + ctrl_reg |= cfgmgr_modes[mode].ctrl; + + /* Set NCE to 0. */ + ctrl_reg &= ~SOCFPGA_FPGMGR_CTL_NCE; + socfpga_fpga_writel(priv, SOCFPGA_FPGMGR_CTL_OFST, ctrl_reg); + + return 0; +} + +static int socfpga_fpga_reset(struct fpga_manager *mgr) +{ + struct socfpga_fpga_priv *priv = mgr->priv; + u32 ctrl_reg, status; + int ret; + + /* + * Step 1: + * - Set CTRL.CFGWDTH, CTRL.CDRATIO to match cfg mode + * - Set CTRL.NCE to 0 + */ + ret = socfpga_fpga_cfg_mode_set(priv); + if (ret) + return ret; + + /* Step 2: Set CTRL.EN to 1 */ + socfpga_fpga_set_bitsl(priv, SOCFPGA_FPGMGR_CTL_OFST, + SOCFPGA_FPGMGR_CTL_EN); + + /* Step 3: Set CTRL.NCONFIGPULL to 1 to put FPGA in reset */ + ctrl_reg = socfpga_fpga_readl(priv, SOCFPGA_FPGMGR_CTL_OFST); + ctrl_reg |= SOCFPGA_FPGMGR_CTL_NCFGPULL; + socfpga_fpga_writel(priv, SOCFPGA_FPGMGR_CTL_OFST, ctrl_reg); + + /* Step 4: Wait for STATUS.MODE to report FPGA is in reset phase */ + status = socfpga_fpga_wait_for_state(priv, SOCFPGA_FPGMGR_STAT_RESET); + + /* Step 5: Set CONTROL.NCONFIGPULL to 0 to release FPGA from reset */ + ctrl_reg &= ~SOCFPGA_FPGMGR_CTL_NCFGPULL; + socfpga_fpga_writel(priv, SOCFPGA_FPGMGR_CTL_OFST, ctrl_reg); + + /* Timeout waiting for reset */ + if (status) + return -ETIMEDOUT; + + return 0; +} + +/* + * Prepare the FPGA to receive the configuration data. + */ +static int socfpga_fpga_ops_configure_init(struct fpga_manager *mgr, u32 flags, + const char *buf, size_t count) +{ + struct socfpga_fpga_priv *priv = mgr->priv; + int ret; + + if (flags & FPGA_MGR_PARTIAL_RECONFIG) { + dev_err(&mgr->dev, "Partial reconfiguration not supported.\n"); + return -EINVAL; + } + /* Steps 1 - 5: Reset the FPGA */ + ret = socfpga_fpga_reset(mgr); + if (ret) + return ret; + + /* Step 6: Wait for FPGA to enter configuration phase */ + if (socfpga_fpga_wait_for_state(priv, SOCFPGA_FPGMGR_STAT_CFG)) + return -ETIMEDOUT; + + /* Step 7: Clear nSTATUS interrupt */ + socfpga_fpga_writel(priv, SOCFPGA_FPGMGR_GPIO_PORTA_EOI_OFST, + SOCFPGA_FPGMGR_MON_NSTATUS); + + /* Step 8: Set CTRL.AXICFGEN to 1 to enable transfer of config data */ + socfpga_fpga_set_bitsl(priv, SOCFPGA_FPGMGR_CTL_OFST, + SOCFPGA_FPGMGR_CTL_AXICFGEN); + + return 0; +} + +/* + * Step 9: write data to the FPGA data register + */ +static int socfpga_fpga_ops_configure_write(struct fpga_manager *mgr, + const char *buf, size_t count) +{ + struct socfpga_fpga_priv *priv = mgr->priv; + u32 *buffer_32 = (u32 *)buf; + size_t i = 0; + + if (count <= 0) + return -EINVAL; + + /* Write out the complete 32-bit chunks. */ + while (count >= sizeof(u32)) { + socfpga_fpga_data_writel(priv, buffer_32[i++]); + count -= sizeof(u32); + } + + /* Write out remaining non 32-bit chunks. */ + switch (count) { + case 3: + socfpga_fpga_data_writel(priv, buffer_32[i++] & 0x00ffffff); + break; + case 2: + socfpga_fpga_data_writel(priv, buffer_32[i++] & 0x0000ffff); + break; + case 1: + socfpga_fpga_data_writel(priv, buffer_32[i++] & 0x000000ff); + break; + case 0: + break; + default: + /* This will never happen. */ + return -EFAULT; + } + + return 0; +} + +static int socfpga_fpga_ops_configure_complete(struct fpga_manager *mgr, + u32 flags) +{ + struct socfpga_fpga_priv *priv = mgr->priv; + u32 status; + + /* + * Step 10: + * - Observe CONF_DONE and nSTATUS (active low) + * - if CONF_DONE = 1 and nSTATUS = 1, configuration was successful + * - if CONF_DONE = 0 and nSTATUS = 0, configuration failed + */ + status = socfpga_fpga_wait_for_config_done(priv); + if (status) + return status; + + /* Step 11: Clear CTRL.AXICFGEN to disable transfer of config data */ + socfpga_fpga_clr_bitsl(priv, SOCFPGA_FPGMGR_CTL_OFST, + SOCFPGA_FPGMGR_CTL_AXICFGEN); + + /* + * Step 12: + * - Write 4 to DCLKCNT + * - Wait for STATUS.DCNTDONE = 1 + * - Clear W1C bit in STATUS.DCNTDONE + */ + if (socfpga_fpga_dclk_set_and_wait_clear(priv, 4)) + return -ETIMEDOUT; + + /* Step 13: Wait for STATUS.MODE to report USER MODE */ + if (socfpga_fpga_wait_for_state(priv, SOCFPGA_FPGMGR_STAT_USER_MODE)) + return -ETIMEDOUT; + + /* Step 14: Set CTRL.EN to 0 */ + socfpga_fpga_clr_bitsl(priv, SOCFPGA_FPGMGR_CTL_OFST, + SOCFPGA_FPGMGR_CTL_EN); + + return 0; +} + +/* Translate state register values to FPGA framework state */ +static const enum fpga_mgr_states socfpga_state_to_framework_state[] = { + [SOCFPGA_FPGMGR_STAT_POWER_OFF] = FPGA_MGR_STATE_POWER_OFF, + [SOCFPGA_FPGMGR_STAT_RESET] = FPGA_MGR_STATE_RESET, + [SOCFPGA_FPGMGR_STAT_CFG] = FPGA_MGR_STATE_WRITE_INIT, + [SOCFPGA_FPGMGR_STAT_INIT] = FPGA_MGR_STATE_WRITE_INIT, + [SOCFPGA_FPGMGR_STAT_USER_MODE] = FPGA_MGR_STATE_OPERATING, + [SOCFPGA_FPGMGR_STAT_UNKNOWN] = FPGA_MGR_STATE_UNKNOWN, +}; + +static enum fpga_mgr_states socfpga_fpga_ops_state(struct fpga_manager *mgr) +{ + struct socfpga_fpga_priv *priv = mgr->priv; + enum fpga_mgr_states ret; + u32 state; + + state = socfpga_fpga_state_get(priv); + + if (state < ARRAY_SIZE(socfpga_state_to_framework_state)) + ret = socfpga_state_to_framework_state[state]; + else + ret = FPGA_MGR_STATE_UNKNOWN; + + return ret; +} + +static const struct fpga_manager_ops socfpga_fpga_ops = { + .state = socfpga_fpga_ops_state, + .write_init = socfpga_fpga_ops_configure_init, + .write = socfpga_fpga_ops_configure_write, + .write_complete = socfpga_fpga_ops_configure_complete, +}; + +static int socfpga_fpga_probe(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct socfpga_fpga_priv *priv; + struct resource *res; + int ret; + + priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL); + if (!priv) + return -ENOMEM; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + priv->fpga_base_addr = devm_ioremap_resource(dev, res); + if (IS_ERR(priv->fpga_base_addr)) + return PTR_ERR(priv->fpga_base_addr); + + res = platform_get_resource(pdev, IORESOURCE_MEM, 1); + priv->fpga_data_addr = devm_ioremap_resource(dev, res); + if (IS_ERR(priv->fpga_data_addr)) + return PTR_ERR(priv->fpga_data_addr); + + priv->irq = platform_get_irq(pdev, 0); + if (priv->irq < 0) + return priv->irq; + + ret = devm_request_irq(dev, priv->irq, socfpga_fpga_isr, 0, + dev_name(dev), priv); + if (IS_ERR_VALUE(ret)) + return ret; + + return fpga_mgr_register(dev, "Altera SOCFPGA FPGA Manager", + &socfpga_fpga_ops, priv); +} + +static int socfpga_fpga_remove(struct platform_device *pdev) +{ + fpga_mgr_unregister(&pdev->dev); + + return 0; +} + +#ifdef CONFIG_OF +static const struct of_device_id socfpga_fpga_of_match[] = { + { .compatible = "altr,socfpga-fpga-mgr", }, + {}, +}; + +MODULE_DEVICE_TABLE(of, socfpga_fpga_of_match); +#endif + +static struct platform_driver socfpga_fpga_driver = { + .probe = socfpga_fpga_probe, + .remove = socfpga_fpga_remove, + .driver = { + .name = "socfpga_fpga_manager", + .of_match_table = of_match_ptr(socfpga_fpga_of_match), + }, +}; + +module_platform_driver(socfpga_fpga_driver); + +MODULE_AUTHOR("Alan Tull <atull@opensource.altera.com>"); +MODULE_DESCRIPTION("Altera SOCFPGA FPGA Manager"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c index 1c3fc99c5465..8e995148f56e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c @@ -208,44 +208,6 @@ static int amdgpu_cgs_alloc_gpu_mem(void *cgs_device, return ret; } -static int amdgpu_cgs_import_gpu_mem(void *cgs_device, int dmabuf_fd, - cgs_handle_t *handle) -{ - CGS_FUNC_ADEV; - int r; - uint32_t dma_handle; - struct drm_gem_object *obj; - struct amdgpu_bo *bo; - struct drm_device *dev = adev->ddev; - struct drm_file *file_priv = NULL, *priv; - - mutex_lock(&dev->struct_mutex); - list_for_each_entry(priv, &dev->filelist, lhead) { - rcu_read_lock(); - if (priv->pid == get_pid(task_pid(current))) - file_priv = priv; - rcu_read_unlock(); - if (file_priv) - break; - } - mutex_unlock(&dev->struct_mutex); - r = dev->driver->prime_fd_to_handle(dev, - file_priv, dmabuf_fd, - &dma_handle); - spin_lock(&file_priv->table_lock); - - /* Check if we currently have a reference on the object */ - obj = idr_find(&file_priv->object_idr, dma_handle); - if (obj == NULL) { - spin_unlock(&file_priv->table_lock); - return -EINVAL; - } - spin_unlock(&file_priv->table_lock); - bo = gem_to_amdgpu_bo(obj); - *handle = (cgs_handle_t)bo; - return 0; -} - static int amdgpu_cgs_free_gpu_mem(void *cgs_device, cgs_handle_t handle) { struct amdgpu_bo *obj = (struct amdgpu_bo *)handle; @@ -810,7 +772,6 @@ static const struct cgs_ops amdgpu_cgs_ops = { }; static const struct cgs_os_ops amdgpu_cgs_os_ops = { - amdgpu_cgs_import_gpu_mem, amdgpu_cgs_add_irq_source, amdgpu_cgs_irq_get, amdgpu_cgs_irq_put diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index 749420f1ea6f..cb3c274edb0a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -156,7 +156,8 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data) uint64_t *chunk_array_user; uint64_t *chunk_array; struct amdgpu_fpriv *fpriv = p->filp->driver_priv; - unsigned size, i; + unsigned size; + int i; int ret; if (cs->in.num_chunks == 0) diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c b/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c index cd6edc40c9cd..1e0bba29e167 100644 --- a/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c +++ b/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c @@ -1279,8 +1279,7 @@ amdgpu_atombios_encoder_setup_dig(struct drm_encoder *encoder, int action) amdgpu_atombios_encoder_setup_dig_encoder(encoder, ATOM_ENCODER_CMD_DP_VIDEO_ON, 0); } if (amdgpu_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) - amdgpu_atombios_encoder_setup_dig_transmitter(encoder, - ATOM_TRANSMITTER_ACTION_LCD_BLON, 0, 0); + amdgpu_atombios_encoder_set_backlight_level(amdgpu_encoder, dig->backlight_level); if (ext_encoder) amdgpu_atombios_encoder_setup_external_encoder(encoder, ext_encoder, ATOM_ENABLE); } else { diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c index 774528ab8704..fab5471d25d7 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c @@ -1262,6 +1262,12 @@ static int gmc_v7_0_process_interrupt(struct amdgpu_device *adev, addr = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_ADDR); status = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_STATUS); mc_client = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_MCCLIENT); + /* reset addr and status */ + WREG32_P(mmVM_CONTEXT1_CNTL2, 1, ~1); + + if (!addr && !status) + return 0; + dev_err(adev->dev, "GPU fault detected: %d 0x%08x\n", entry->src_id, entry->src_data); dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n", @@ -1269,8 +1275,6 @@ static int gmc_v7_0_process_interrupt(struct amdgpu_device *adev, dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n", status); gmc_v7_0_vm_decode_fault(adev, status, addr, mc_client); - /* reset addr and status */ - WREG32_P(mmVM_CONTEXT1_CNTL2, 1, ~1); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index 9a07742620d0..7bc9e9fcf3d2 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -1262,6 +1262,12 @@ static int gmc_v8_0_process_interrupt(struct amdgpu_device *adev, addr = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_ADDR); status = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_STATUS); mc_client = RREG32(mmVM_CONTEXT1_PROTECTION_FAULT_MCCLIENT); + /* reset addr and status */ + WREG32_P(mmVM_CONTEXT1_CNTL2, 1, ~1); + + if (!addr && !status) + return 0; + dev_err(adev->dev, "GPU fault detected: %d 0x%08x\n", entry->src_id, entry->src_data); dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_ADDR 0x%08X\n", @@ -1269,8 +1275,6 @@ static int gmc_v8_0_process_interrupt(struct amdgpu_device *adev, dev_err(adev->dev, " VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x%08X\n", status); gmc_v8_0_vm_decode_fault(adev, status, addr, mc_client); - /* reset addr and status */ - WREG32_P(mmVM_CONTEXT1_CNTL2, 1, ~1); return 0; } diff --git a/drivers/gpu/drm/amd/include/cgs_linux.h b/drivers/gpu/drm/amd/include/cgs_linux.h index 488642f08267..3b47ae313e36 100644 --- a/drivers/gpu/drm/amd/include/cgs_linux.h +++ b/drivers/gpu/drm/amd/include/cgs_linux.h @@ -27,19 +27,6 @@ #include "cgs_common.h" /** - * cgs_import_gpu_mem() - Import dmabuf handle - * @cgs_device: opaque device handle - * @dmabuf_fd: DMABuf file descriptor - * @handle: memory handle (output) - * - * Must be called in the process context that dmabuf_fd belongs to. - * - * Return: 0 on success, -errno otherwise - */ -typedef int (*cgs_import_gpu_mem_t)(void *cgs_device, int dmabuf_fd, - cgs_handle_t *handle); - -/** * cgs_irq_source_set_func() - Callback for enabling/disabling interrupt sources * @private_data: private data provided to cgs_add_irq_source * @src_id: interrupt source ID @@ -114,16 +101,12 @@ typedef int (*cgs_irq_get_t)(void *cgs_device, unsigned src_id, unsigned type); typedef int (*cgs_irq_put_t)(void *cgs_device, unsigned src_id, unsigned type); struct cgs_os_ops { - cgs_import_gpu_mem_t import_gpu_mem; - /* IRQ handling */ cgs_add_irq_source_t add_irq_source; cgs_irq_get_t irq_get; cgs_irq_put_t irq_put; }; -#define cgs_import_gpu_mem(dev,dmabuf_fd,handle) \ - CGS_OS_CALL(import_gpu_mem,dev,dmabuf_fd,handle) #define cgs_add_irq_source(dev,src_id,num_types,set,handler,private_data) \ CGS_OS_CALL(add_irq_source,dev,src_id,num_types,set,handler, \ private_data) diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c index e23df5fd3836..bf27a07dbce3 100644 --- a/drivers/gpu/drm/drm_dp_mst_topology.c +++ b/drivers/gpu/drm/drm_dp_mst_topology.c @@ -53,8 +53,8 @@ static int drm_dp_send_dpcd_write(struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_port *port, int offset, int size, u8 *bytes); -static int drm_dp_send_link_address(struct drm_dp_mst_topology_mgr *mgr, - struct drm_dp_mst_branch *mstb); +static void drm_dp_send_link_address(struct drm_dp_mst_topology_mgr *mgr, + struct drm_dp_mst_branch *mstb); static int drm_dp_send_enum_path_resources(struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_branch *mstb, struct drm_dp_mst_port *port); @@ -804,8 +804,6 @@ static void drm_dp_destroy_mst_branch_device(struct kref *kref) struct drm_dp_mst_port *port, *tmp; bool wake_tx = false; - cancel_work_sync(&mstb->mgr->work); - /* * destroy all ports - don't need lock * as there are no more references to the mst branch @@ -863,29 +861,33 @@ static void drm_dp_destroy_port(struct kref *kref) { struct drm_dp_mst_port *port = container_of(kref, struct drm_dp_mst_port, kref); struct drm_dp_mst_topology_mgr *mgr = port->mgr; + if (!port->input) { port->vcpi.num_slots = 0; kfree(port->cached_edid); - /* we can't destroy the connector here, as - we might be holding the mode_config.mutex - from an EDID retrieval */ + /* + * The only time we don't have a connector + * on an output port is if the connector init + * fails. + */ if (port->connector) { + /* we can't destroy the connector here, as + * we might be holding the mode_config.mutex + * from an EDID retrieval */ + mutex_lock(&mgr->destroy_connector_lock); list_add(&port->next, &mgr->destroy_connector_list); mutex_unlock(&mgr->destroy_connector_lock); schedule_work(&mgr->destroy_connector_work); return; } + /* no need to clean up vcpi + * as if we have no connector we never setup a vcpi */ drm_dp_port_teardown_pdt(port, port->pdt); - - if (!port->input && port->vcpi.vcpi > 0) - drm_dp_mst_put_payload_id(mgr, port->vcpi.vcpi); } kfree(port); - - (*mgr->cbs->hotplug)(mgr); } static void drm_dp_put_port(struct drm_dp_mst_port *port) @@ -1027,8 +1029,8 @@ static void drm_dp_check_port_guid(struct drm_dp_mst_branch *mstb, } } -static void build_mst_prop_path(struct drm_dp_mst_port *port, - struct drm_dp_mst_branch *mstb, +static void build_mst_prop_path(const struct drm_dp_mst_branch *mstb, + int pnum, char *proppath, size_t proppath_size) { @@ -1041,7 +1043,7 @@ static void build_mst_prop_path(struct drm_dp_mst_port *port, snprintf(temp, sizeof(temp), "-%d", port_num); strlcat(proppath, temp, proppath_size); } - snprintf(temp, sizeof(temp), "-%d", port->port_num); + snprintf(temp, sizeof(temp), "-%d", pnum); strlcat(proppath, temp, proppath_size); } @@ -1105,22 +1107,32 @@ static void drm_dp_add_port(struct drm_dp_mst_branch *mstb, drm_dp_port_teardown_pdt(port, old_pdt); ret = drm_dp_port_setup_pdt(port); - if (ret == true) { + if (ret == true) drm_dp_send_link_address(mstb->mgr, port->mstb); - port->mstb->link_address_sent = true; - } } if (created && !port->input) { char proppath[255]; - build_mst_prop_path(port, mstb, proppath, sizeof(proppath)); - port->connector = (*mstb->mgr->cbs->add_connector)(mstb->mgr, port, proppath); - if (port->port_num >= 8) { + build_mst_prop_path(mstb, port->port_num, proppath, sizeof(proppath)); + port->connector = (*mstb->mgr->cbs->add_connector)(mstb->mgr, port, proppath); + if (!port->connector) { + /* remove it from the port list */ + mutex_lock(&mstb->mgr->lock); + list_del(&port->next); + mutex_unlock(&mstb->mgr->lock); + /* drop port list reference */ + drm_dp_put_port(port); + goto out; + } + if (port->port_num >= DP_MST_LOGICAL_PORT_0) { port->cached_edid = drm_get_edid(port->connector, &port->aux.ddc); + drm_mode_connector_set_tile_property(port->connector); } + (*mstb->mgr->cbs->register_connector)(port->connector); } +out: /* put reference to this port */ drm_dp_put_port(port); } @@ -1202,10 +1214,9 @@ static void drm_dp_check_and_send_link_address(struct drm_dp_mst_topology_mgr *m { struct drm_dp_mst_port *port; struct drm_dp_mst_branch *mstb_child; - if (!mstb->link_address_sent) { + if (!mstb->link_address_sent) drm_dp_send_link_address(mgr, mstb); - mstb->link_address_sent = true; - } + list_for_each_entry(port, &mstb->ports, next) { if (port->input) continue; @@ -1458,8 +1469,8 @@ static void drm_dp_queue_down_tx(struct drm_dp_mst_topology_mgr *mgr, mutex_unlock(&mgr->qlock); } -static int drm_dp_send_link_address(struct drm_dp_mst_topology_mgr *mgr, - struct drm_dp_mst_branch *mstb) +static void drm_dp_send_link_address(struct drm_dp_mst_topology_mgr *mgr, + struct drm_dp_mst_branch *mstb) { int len; struct drm_dp_sideband_msg_tx *txmsg; @@ -1467,11 +1478,12 @@ static int drm_dp_send_link_address(struct drm_dp_mst_topology_mgr *mgr, txmsg = kzalloc(sizeof(*txmsg), GFP_KERNEL); if (!txmsg) - return -ENOMEM; + return; txmsg->dst = mstb; len = build_link_address(txmsg); + mstb->link_address_sent = true; drm_dp_queue_down_tx(mgr, txmsg); ret = drm_dp_mst_wait_tx_reply(mstb, txmsg); @@ -1499,11 +1511,12 @@ static int drm_dp_send_link_address(struct drm_dp_mst_topology_mgr *mgr, } (*mgr->cbs->hotplug)(mgr); } - } else + } else { + mstb->link_address_sent = false; DRM_DEBUG_KMS("link address failed %d\n", ret); + } kfree(txmsg); - return 0; } static int drm_dp_send_enum_path_resources(struct drm_dp_mst_topology_mgr *mgr, @@ -1978,6 +1991,8 @@ void drm_dp_mst_topology_mgr_suspend(struct drm_dp_mst_topology_mgr *mgr) drm_dp_dpcd_writeb(mgr->aux, DP_MSTM_CTRL, DP_MST_EN | DP_UPSTREAM_IS_SRC); mutex_unlock(&mgr->lock); + flush_work(&mgr->work); + flush_work(&mgr->destroy_connector_work); } EXPORT_SYMBOL(drm_dp_mst_topology_mgr_suspend); @@ -2263,10 +2278,10 @@ struct edid *drm_dp_mst_get_edid(struct drm_connector *connector, struct drm_dp_ if (port->cached_edid) edid = drm_edid_duplicate(port->cached_edid); - else + else { edid = drm_get_edid(connector, &port->aux.ddc); - - drm_mode_connector_set_tile_property(connector); + drm_mode_connector_set_tile_property(connector); + } drm_dp_put_port(port); return edid; } @@ -2671,7 +2686,7 @@ static void drm_dp_destroy_connector_work(struct work_struct *work) { struct drm_dp_mst_topology_mgr *mgr = container_of(work, struct drm_dp_mst_topology_mgr, destroy_connector_work); struct drm_dp_mst_port *port; - + bool send_hotplug = false; /* * Not a regular list traverse as we have to drop the destroy * connector lock before destroying the connector, to avoid AB->BA @@ -2694,7 +2709,10 @@ static void drm_dp_destroy_connector_work(struct work_struct *work) if (!port->input && port->vcpi.vcpi > 0) drm_dp_mst_put_payload_id(mgr, port->vcpi.vcpi); kfree(port); + send_hotplug = true; } + if (send_hotplug) + (*mgr->cbs->hotplug)(mgr); } /** @@ -2747,6 +2765,7 @@ EXPORT_SYMBOL(drm_dp_mst_topology_mgr_init); */ void drm_dp_mst_topology_mgr_destroy(struct drm_dp_mst_topology_mgr *mgr) { + flush_work(&mgr->work); flush_work(&mgr->destroy_connector_work); mutex_lock(&mgr->payload_lock); kfree(mgr->payloads); diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c index 418d299f3b12..ca08c472311b 100644 --- a/drivers/gpu/drm/drm_fb_helper.c +++ b/drivers/gpu/drm/drm_fb_helper.c @@ -345,7 +345,11 @@ static bool restore_fbdev_mode(struct drm_fb_helper *fb_helper) struct drm_crtc *crtc = mode_set->crtc; int ret; - if (crtc->funcs->cursor_set) { + if (crtc->funcs->cursor_set2) { + ret = crtc->funcs->cursor_set2(crtc, NULL, 0, 0, 0, 0, 0); + if (ret) + error = true; + } else if (crtc->funcs->cursor_set) { ret = crtc->funcs->cursor_set(crtc, NULL, 0, 0, 0); if (ret) error = true; diff --git a/drivers/gpu/drm/drm_probe_helper.c b/drivers/gpu/drm/drm_probe_helper.c index d734780b31c0..a18164f2f6d2 100644 --- a/drivers/gpu/drm/drm_probe_helper.c +++ b/drivers/gpu/drm/drm_probe_helper.c @@ -94,7 +94,18 @@ static int drm_helper_probe_add_cmdline_mode(struct drm_connector *connector) } #define DRM_OUTPUT_POLL_PERIOD (10*HZ) -static void __drm_kms_helper_poll_enable(struct drm_device *dev) +/** + * drm_kms_helper_poll_enable_locked - re-enable output polling. + * @dev: drm_device + * + * This function re-enables the output polling work without + * locking the mode_config mutex. + * + * This is like drm_kms_helper_poll_enable() however it is to be + * called from a context where the mode_config mutex is locked + * already. + */ +void drm_kms_helper_poll_enable_locked(struct drm_device *dev) { bool poll = false; struct drm_connector *connector; @@ -113,6 +124,8 @@ static void __drm_kms_helper_poll_enable(struct drm_device *dev) if (poll) schedule_delayed_work(&dev->mode_config.output_poll_work, DRM_OUTPUT_POLL_PERIOD); } +EXPORT_SYMBOL(drm_kms_helper_poll_enable_locked); + static int drm_helper_probe_single_connector_modes_merge_bits(struct drm_connector *connector, uint32_t maxX, uint32_t maxY, bool merge_type_bits) @@ -174,7 +187,7 @@ static int drm_helper_probe_single_connector_modes_merge_bits(struct drm_connect /* Re-enable polling in case the global poll config changed. */ if (drm_kms_helper_poll != dev->mode_config.poll_running) - __drm_kms_helper_poll_enable(dev); + drm_kms_helper_poll_enable_locked(dev); dev->mode_config.poll_running = drm_kms_helper_poll; @@ -428,7 +441,7 @@ EXPORT_SYMBOL(drm_kms_helper_poll_disable); void drm_kms_helper_poll_enable(struct drm_device *dev) { mutex_lock(&dev->mode_config.mutex); - __drm_kms_helper_poll_enable(dev); + drm_kms_helper_poll_enable_locked(dev); mutex_unlock(&dev->mode_config.mutex); } EXPORT_SYMBOL(drm_kms_helper_poll_enable); diff --git a/drivers/gpu/drm/exynos/exynos7_drm_decon.c b/drivers/gpu/drm/exynos/exynos7_drm_decon.c index cbdb78ef3bac..e6cbaca821a4 100644 --- a/drivers/gpu/drm/exynos/exynos7_drm_decon.c +++ b/drivers/gpu/drm/exynos/exynos7_drm_decon.c @@ -37,7 +37,6 @@ * DECON stands for Display and Enhancement controller. */ -#define DECON_DEFAULT_FRAMERATE 60 #define MIN_FB_WIDTH_FOR_16WORD_BURST 128 #define WINDOWS_NR 2 @@ -165,16 +164,6 @@ static u32 decon_calc_clkdiv(struct decon_context *ctx, return (clkdiv < 0x100) ? clkdiv : 0xff; } -static bool decon_mode_fixup(struct exynos_drm_crtc *crtc, - const struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode) -{ - if (adjusted_mode->vrefresh == 0) - adjusted_mode->vrefresh = DECON_DEFAULT_FRAMERATE; - - return true; -} - static void decon_commit(struct exynos_drm_crtc *crtc) { struct decon_context *ctx = crtc->ctx; @@ -637,7 +626,6 @@ static void decon_disable(struct exynos_drm_crtc *crtc) static const struct exynos_drm_crtc_ops decon_crtc_ops = { .enable = decon_enable, .disable = decon_disable, - .mode_fixup = decon_mode_fixup, .commit = decon_commit, .enable_vblank = decon_enable_vblank, .disable_vblank = decon_disable_vblank, diff --git a/drivers/gpu/drm/exynos/exynos_dp_core.c b/drivers/gpu/drm/exynos/exynos_dp_core.c index d66ade0efac8..124fb9a56f02 100644 --- a/drivers/gpu/drm/exynos/exynos_dp_core.c +++ b/drivers/gpu/drm/exynos/exynos_dp_core.c @@ -1383,28 +1383,6 @@ static int exynos_dp_remove(struct platform_device *pdev) return 0; } -#ifdef CONFIG_PM_SLEEP -static int exynos_dp_suspend(struct device *dev) -{ - struct exynos_dp_device *dp = dev_get_drvdata(dev); - - exynos_dp_disable(&dp->encoder); - return 0; -} - -static int exynos_dp_resume(struct device *dev) -{ - struct exynos_dp_device *dp = dev_get_drvdata(dev); - - exynos_dp_enable(&dp->encoder); - return 0; -} -#endif - -static const struct dev_pm_ops exynos_dp_pm_ops = { - SET_SYSTEM_SLEEP_PM_OPS(exynos_dp_suspend, exynos_dp_resume) -}; - static const struct of_device_id exynos_dp_match[] = { { .compatible = "samsung,exynos5-dp" }, {}, @@ -1417,7 +1395,6 @@ struct platform_driver dp_driver = { .driver = { .name = "exynos-dp", .owner = THIS_MODULE, - .pm = &exynos_dp_pm_ops, .of_match_table = exynos_dp_match, }, }; diff --git a/drivers/gpu/drm/exynos/exynos_drm_core.c b/drivers/gpu/drm/exynos/exynos_drm_core.c index c68a6a2a9b57..7f55ba6771c6 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_core.c +++ b/drivers/gpu/drm/exynos/exynos_drm_core.c @@ -28,7 +28,6 @@ int exynos_drm_subdrv_register(struct exynos_drm_subdrv *subdrv) return 0; } -EXPORT_SYMBOL_GPL(exynos_drm_subdrv_register); int exynos_drm_subdrv_unregister(struct exynos_drm_subdrv *subdrv) { @@ -39,7 +38,6 @@ int exynos_drm_subdrv_unregister(struct exynos_drm_subdrv *subdrv) return 0; } -EXPORT_SYMBOL_GPL(exynos_drm_subdrv_unregister); int exynos_drm_device_subdrv_probe(struct drm_device *dev) { @@ -69,7 +67,6 @@ int exynos_drm_device_subdrv_probe(struct drm_device *dev) return 0; } -EXPORT_SYMBOL_GPL(exynos_drm_device_subdrv_probe); int exynos_drm_device_subdrv_remove(struct drm_device *dev) { @@ -87,7 +84,6 @@ int exynos_drm_device_subdrv_remove(struct drm_device *dev) return 0; } -EXPORT_SYMBOL_GPL(exynos_drm_device_subdrv_remove); int exynos_drm_subdrv_open(struct drm_device *dev, struct drm_file *file) { @@ -111,7 +107,6 @@ err: } return ret; } -EXPORT_SYMBOL_GPL(exynos_drm_subdrv_open); void exynos_drm_subdrv_close(struct drm_device *dev, struct drm_file *file) { @@ -122,4 +117,3 @@ void exynos_drm_subdrv_close(struct drm_device *dev, struct drm_file *file) subdrv->close(dev, subdrv->dev, file); } } -EXPORT_SYMBOL_GPL(exynos_drm_subdrv_close); diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c index 0872aa2f450f..ed28823d3b35 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c @@ -41,20 +41,6 @@ static void exynos_drm_crtc_disable(struct drm_crtc *crtc) exynos_crtc->ops->disable(exynos_crtc); } -static bool -exynos_drm_crtc_mode_fixup(struct drm_crtc *crtc, - const struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode) -{ - struct exynos_drm_crtc *exynos_crtc = to_exynos_crtc(crtc); - - if (exynos_crtc->ops->mode_fixup) - return exynos_crtc->ops->mode_fixup(exynos_crtc, mode, - adjusted_mode); - - return true; -} - static void exynos_drm_crtc_mode_set_nofb(struct drm_crtc *crtc) { @@ -99,7 +85,6 @@ static void exynos_crtc_atomic_flush(struct drm_crtc *crtc, static struct drm_crtc_helper_funcs exynos_crtc_helper_funcs = { .enable = exynos_drm_crtc_enable, .disable = exynos_drm_crtc_disable, - .mode_fixup = exynos_drm_crtc_mode_fixup, .mode_set_nofb = exynos_drm_crtc_mode_set_nofb, .atomic_begin = exynos_crtc_atomic_begin, .atomic_flush = exynos_crtc_atomic_flush, diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c index 831d2e4cacf9..ae9e6b2d3758 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.c +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c @@ -304,6 +304,7 @@ int exynos_atomic_commit(struct drm_device *dev, struct drm_atomic_state *state, return 0; } +#ifdef CONFIG_PM_SLEEP static int exynos_drm_suspend(struct drm_device *dev, pm_message_t state) { struct drm_connector *connector; @@ -340,6 +341,7 @@ static int exynos_drm_resume(struct drm_device *dev) return 0; } +#endif static int exynos_drm_open(struct drm_device *dev, struct drm_file *file) { diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h index b7ba21dfb696..6c717ba672db 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.h +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h @@ -82,7 +82,6 @@ struct exynos_drm_plane { * * @enable: enable the device * @disable: disable the device - * @mode_fixup: fix mode data before applying it * @commit: set current hw specific display mode to hw. * @enable_vblank: specific driver callback for enabling vblank interrupt. * @disable_vblank: specific driver callback for disabling vblank interrupt. @@ -103,9 +102,6 @@ struct exynos_drm_crtc; struct exynos_drm_crtc_ops { void (*enable)(struct exynos_drm_crtc *crtc); void (*disable)(struct exynos_drm_crtc *crtc); - bool (*mode_fixup)(struct exynos_drm_crtc *crtc, - const struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode); void (*commit)(struct exynos_drm_crtc *crtc); int (*enable_vblank)(struct exynos_drm_crtc *crtc); void (*disable_vblank)(struct exynos_drm_crtc *crtc); diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimc.c b/drivers/gpu/drm/exynos/exynos_drm_fimc.c index 2a652359af64..dd3a5e6d58c8 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimc.c @@ -1206,23 +1206,6 @@ static struct exynos_drm_ipp_ops fimc_dst_ops = { .set_addr = fimc_dst_set_addr, }; -static int fimc_clk_ctrl(struct fimc_context *ctx, bool enable) -{ - DRM_DEBUG_KMS("enable[%d]\n", enable); - - if (enable) { - clk_prepare_enable(ctx->clocks[FIMC_CLK_GATE]); - clk_prepare_enable(ctx->clocks[FIMC_CLK_WB_A]); - ctx->suspended = false; - } else { - clk_disable_unprepare(ctx->clocks[FIMC_CLK_GATE]); - clk_disable_unprepare(ctx->clocks[FIMC_CLK_WB_A]); - ctx->suspended = true; - } - - return 0; -} - static irqreturn_t fimc_irq_handler(int irq, void *dev_id) { struct fimc_context *ctx = dev_id; @@ -1780,6 +1763,24 @@ static int fimc_remove(struct platform_device *pdev) return 0; } +#ifdef CONFIG_PM +static int fimc_clk_ctrl(struct fimc_context *ctx, bool enable) +{ + DRM_DEBUG_KMS("enable[%d]\n", enable); + + if (enable) { + clk_prepare_enable(ctx->clocks[FIMC_CLK_GATE]); + clk_prepare_enable(ctx->clocks[FIMC_CLK_WB_A]); + ctx->suspended = false; + } else { + clk_disable_unprepare(ctx->clocks[FIMC_CLK_GATE]); + clk_disable_unprepare(ctx->clocks[FIMC_CLK_WB_A]); + ctx->suspended = true; + } + + return 0; +} + #ifdef CONFIG_PM_SLEEP static int fimc_suspend(struct device *dev) { @@ -1806,7 +1807,6 @@ static int fimc_resume(struct device *dev) } #endif -#ifdef CONFIG_PM static int fimc_runtime_suspend(struct device *dev) { struct fimc_context *ctx = get_fimc_context(dev); diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c index 750a9e6b9e8d..3d1aba67758b 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c @@ -41,7 +41,6 @@ * CPU Interface. */ -#define FIMD_DEFAULT_FRAMERATE 60 #define MIN_FB_WIDTH_FOR_16WORD_BURST 128 /* position control register for hardware window 0, 2 ~ 4.*/ @@ -377,16 +376,6 @@ static u32 fimd_calc_clkdiv(struct fimd_context *ctx, return (clkdiv < 0x100) ? clkdiv : 0xff; } -static bool fimd_mode_fixup(struct exynos_drm_crtc *crtc, - const struct drm_display_mode *mode, - struct drm_display_mode *adjusted_mode) -{ - if (adjusted_mode->vrefresh == 0) - adjusted_mode->vrefresh = FIMD_DEFAULT_FRAMERATE; - - return true; -} - static void fimd_commit(struct exynos_drm_crtc *crtc) { struct fimd_context *ctx = crtc->ctx; @@ -882,13 +871,12 @@ static void fimd_dp_clock_enable(struct exynos_drm_crtc *crtc, bool enable) return; val = enable ? DP_MIE_CLK_DP_ENABLE : DP_MIE_CLK_DISABLE; - writel(DP_MIE_CLK_DP_ENABLE, ctx->regs + DP_MIE_CLKCON); + writel(val, ctx->regs + DP_MIE_CLKCON); } static const struct exynos_drm_crtc_ops fimd_crtc_ops = { .enable = fimd_enable, .disable = fimd_disable, - .mode_fixup = fimd_mode_fixup, .commit = fimd_commit, .enable_vblank = fimd_enable_vblank, .disable_vblank = fimd_disable_vblank, diff --git a/drivers/gpu/drm/exynos/exynos_drm_g2d.c b/drivers/gpu/drm/exynos/exynos_drm_g2d.c index 3734c34aed16..c17efdb238a6 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_g2d.c +++ b/drivers/gpu/drm/exynos/exynos_drm_g2d.c @@ -1059,7 +1059,6 @@ int exynos_g2d_get_ver_ioctl(struct drm_device *drm_dev, void *data, return 0; } -EXPORT_SYMBOL_GPL(exynos_g2d_get_ver_ioctl); int exynos_g2d_set_cmdlist_ioctl(struct drm_device *drm_dev, void *data, struct drm_file *file) @@ -1230,7 +1229,6 @@ err: g2d_put_cmdlist(g2d, node); return ret; } -EXPORT_SYMBOL_GPL(exynos_g2d_set_cmdlist_ioctl); int exynos_g2d_exec_ioctl(struct drm_device *drm_dev, void *data, struct drm_file *file) @@ -1293,7 +1291,6 @@ int exynos_g2d_exec_ioctl(struct drm_device *drm_dev, void *data, out: return 0; } -EXPORT_SYMBOL_GPL(exynos_g2d_exec_ioctl); static int g2d_subdrv_probe(struct drm_device *drm_dev, struct device *dev) { diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c index f12fbc36b120..407afedb6003 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c @@ -56,39 +56,35 @@ static int exynos_drm_alloc_buf(struct exynos_drm_gem_obj *obj) nr_pages = obj->size >> PAGE_SHIFT; if (!is_drm_iommu_supported(dev)) { - dma_addr_t start_addr; - unsigned int i = 0; - obj->pages = drm_calloc_large(nr_pages, sizeof(struct page *)); if (!obj->pages) { DRM_ERROR("failed to allocate pages.\n"); return -ENOMEM; } + } - obj->cookie = dma_alloc_attrs(dev->dev, - obj->size, - &obj->dma_addr, GFP_KERNEL, - &obj->dma_attrs); - if (!obj->cookie) { - DRM_ERROR("failed to allocate buffer.\n"); + obj->cookie = dma_alloc_attrs(dev->dev, obj->size, &obj->dma_addr, + GFP_KERNEL, &obj->dma_attrs); + if (!obj->cookie) { + DRM_ERROR("failed to allocate buffer.\n"); + if (obj->pages) drm_free_large(obj->pages); - return -ENOMEM; - } + return -ENOMEM; + } + + if (obj->pages) { + dma_addr_t start_addr; + unsigned int i = 0; start_addr = obj->dma_addr; while (i < nr_pages) { - obj->pages[i] = phys_to_page(start_addr); + obj->pages[i] = pfn_to_page(dma_to_pfn(dev->dev, + start_addr)); start_addr += PAGE_SIZE; i++; } } else { - obj->pages = dma_alloc_attrs(dev->dev, obj->size, - &obj->dma_addr, GFP_KERNEL, - &obj->dma_attrs); - if (!obj->pages) { - DRM_ERROR("failed to allocate buffer.\n"); - return -ENOMEM; - } + obj->pages = obj->cookie; } DRM_DEBUG_KMS("dma_addr(0x%lx), size(0x%lx)\n", @@ -110,15 +106,11 @@ static void exynos_drm_free_buf(struct exynos_drm_gem_obj *obj) DRM_DEBUG_KMS("dma_addr(0x%lx), size(0x%lx)\n", (unsigned long)obj->dma_addr, obj->size); - if (!is_drm_iommu_supported(dev)) { - dma_free_attrs(dev->dev, obj->size, obj->cookie, - (dma_addr_t)obj->dma_addr, &obj->dma_attrs); - drm_free_large(obj->pages); - } else - dma_free_attrs(dev->dev, obj->size, obj->pages, - (dma_addr_t)obj->dma_addr, &obj->dma_attrs); + dma_free_attrs(dev->dev, obj->size, obj->cookie, + (dma_addr_t)obj->dma_addr, &obj->dma_attrs); - obj->dma_addr = (dma_addr_t)NULL; + if (!is_drm_iommu_supported(dev)) + drm_free_large(obj->pages); } static int exynos_drm_gem_handle_create(struct drm_gem_object *obj, @@ -156,18 +148,14 @@ void exynos_drm_gem_destroy(struct exynos_drm_gem_obj *exynos_gem_obj) * once dmabuf's refcount becomes 0. */ if (obj->import_attach) - goto out; - - exynos_drm_free_buf(exynos_gem_obj); - -out: - drm_gem_free_mmap_offset(obj); + drm_prime_gem_destroy(obj, exynos_gem_obj->sgt); + else + exynos_drm_free_buf(exynos_gem_obj); /* release file pointer to gem object. */ drm_gem_object_release(obj); kfree(exynos_gem_obj); - exynos_gem_obj = NULL; } unsigned long exynos_drm_gem_get_size(struct drm_device *dev, @@ -190,8 +178,7 @@ unsigned long exynos_drm_gem_get_size(struct drm_device *dev, return exynos_gem_obj->size; } - -struct exynos_drm_gem_obj *exynos_drm_gem_init(struct drm_device *dev, +static struct exynos_drm_gem_obj *exynos_drm_gem_init(struct drm_device *dev, unsigned long size) { struct exynos_drm_gem_obj *exynos_gem_obj; @@ -212,6 +199,13 @@ struct exynos_drm_gem_obj *exynos_drm_gem_init(struct drm_device *dev, return ERR_PTR(ret); } + ret = drm_gem_create_mmap_offset(obj); + if (ret < 0) { + drm_gem_object_release(obj); + kfree(exynos_gem_obj); + return ERR_PTR(ret); + } + DRM_DEBUG_KMS("created file object = 0x%x\n", (unsigned int)obj->filp); return exynos_gem_obj; @@ -313,7 +307,7 @@ void exynos_drm_gem_put_dma_addr(struct drm_device *dev, drm_gem_object_unreference_unlocked(obj); } -int exynos_drm_gem_mmap_buffer(struct exynos_drm_gem_obj *exynos_gem_obj, +static int exynos_drm_gem_mmap_buffer(struct exynos_drm_gem_obj *exynos_gem_obj, struct vm_area_struct *vma) { struct drm_device *drm_dev = exynos_gem_obj->base.dev; @@ -342,7 +336,8 @@ int exynos_drm_gem_mmap_buffer(struct exynos_drm_gem_obj *exynos_gem_obj, int exynos_drm_gem_get_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv) -{ struct exynos_drm_gem_obj *exynos_gem_obj; +{ + struct exynos_drm_gem_obj *exynos_gem_obj; struct drm_exynos_gem_info *args = data; struct drm_gem_object *obj; @@ -402,6 +397,7 @@ int exynos_drm_gem_dumb_create(struct drm_file *file_priv, struct drm_mode_create_dumb *args) { struct exynos_drm_gem_obj *exynos_gem_obj; + unsigned int flags; int ret; /* @@ -413,16 +409,12 @@ int exynos_drm_gem_dumb_create(struct drm_file *file_priv, args->pitch = args->width * ((args->bpp + 7) / 8); args->size = args->pitch * args->height; - if (is_drm_iommu_supported(dev)) { - exynos_gem_obj = exynos_drm_gem_create(dev, - EXYNOS_BO_NONCONTIG | EXYNOS_BO_WC, - args->size); - } else { - exynos_gem_obj = exynos_drm_gem_create(dev, - EXYNOS_BO_CONTIG | EXYNOS_BO_WC, - args->size); - } + if (is_drm_iommu_supported(dev)) + flags = EXYNOS_BO_NONCONTIG | EXYNOS_BO_WC; + else + flags = EXYNOS_BO_CONTIG | EXYNOS_BO_WC; + exynos_gem_obj = exynos_drm_gem_create(dev, flags, args->size); if (IS_ERR(exynos_gem_obj)) { dev_warn(dev->dev, "FB allocation failed.\n"); return PTR_ERR(exynos_gem_obj); @@ -460,14 +452,9 @@ int exynos_drm_gem_dumb_map_offset(struct drm_file *file_priv, goto unlock; } - ret = drm_gem_create_mmap_offset(obj); - if (ret) - goto out; - *offset = drm_vma_node_offset_addr(&obj->vma_node); DRM_DEBUG_KMS("offset = 0x%lx\n", (unsigned long)*offset); -out: drm_gem_object_unreference(obj); unlock: mutex_unlock(&dev->struct_mutex); @@ -543,7 +530,6 @@ int exynos_drm_gem_mmap(struct file *filp, struct vm_area_struct *vma) err_close_vm: drm_gem_vm_close(vma); - drm_gem_free_mmap_offset(obj); return ret; } @@ -588,6 +574,8 @@ exynos_drm_gem_prime_import_sg_table(struct drm_device *dev, if (ret < 0) goto err_free_large; + exynos_gem_obj->sgt = sgt; + if (sgt->nents == 1) { /* always physically continuous memory if sgt->nents is 1. */ exynos_gem_obj->flags |= EXYNOS_BO_CONTIG; diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.h b/drivers/gpu/drm/exynos/exynos_drm_gem.h index cd62f8410d1e..b62d1007c0e0 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.h +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.h @@ -39,6 +39,7 @@ * - this address could be physical address without IOMMU and * device address with IOMMU. * @pages: Array of backing pages. + * @sgt: Imported sg_table. * * P.S. this object would be transferred to user as kms_bo.handle so * user can access the buffer through kms_bo.handle. @@ -52,6 +53,7 @@ struct exynos_drm_gem_obj { dma_addr_t dma_addr; struct dma_attrs dma_attrs; struct page **pages; + struct sg_table *sgt; }; struct page **exynos_gem_get_pages(struct drm_gem_object *obj, gfp_t gfpmask); @@ -59,10 +61,6 @@ struct page **exynos_gem_get_pages(struct drm_gem_object *obj, gfp_t gfpmask); /* destroy a buffer with gem object */ void exynos_drm_gem_destroy(struct exynos_drm_gem_obj *exynos_gem_obj); -/* create a private gem object and initialize it. */ -struct exynos_drm_gem_obj *exynos_drm_gem_init(struct drm_device *dev, - unsigned long size); - /* create a new buffer with gem object */ struct exynos_drm_gem_obj *exynos_drm_gem_create(struct drm_device *dev, unsigned int flags, diff --git a/drivers/gpu/drm/exynos/exynos_drm_rotator.c b/drivers/gpu/drm/exynos/exynos_drm_rotator.c index 425e70625388..2f5c118f4c8e 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_rotator.c +++ b/drivers/gpu/drm/exynos/exynos_drm_rotator.c @@ -786,6 +786,7 @@ static int rotator_remove(struct platform_device *pdev) return 0; } +#ifdef CONFIG_PM static int rotator_clk_crtl(struct rot_context *rot, bool enable) { if (enable) { @@ -822,7 +823,6 @@ static int rotator_resume(struct device *dev) } #endif -#ifdef CONFIG_PM static int rotator_runtime_suspend(struct device *dev) { struct rot_context *rot = dev_get_drvdata(dev); diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c index 3e4be5a3becd..6ade06888432 100644 --- a/drivers/gpu/drm/i915/intel_dp_mst.c +++ b/drivers/gpu/drm/i915/intel_dp_mst.c @@ -462,11 +462,17 @@ static struct drm_connector *intel_dp_add_mst_connector(struct drm_dp_mst_topolo drm_object_attach_property(&connector->base, dev->mode_config.tile_property, 0); drm_mode_connector_set_path_property(connector, pathprop); + return connector; +} + +static void intel_dp_register_mst_connector(struct drm_connector *connector) +{ + struct intel_connector *intel_connector = to_intel_connector(connector); + struct drm_device *dev = connector->dev; drm_modeset_lock_all(dev); intel_connector_add_to_fbdev(intel_connector); drm_modeset_unlock_all(dev); drm_connector_register(&intel_connector->base); - return connector; } static void intel_dp_destroy_mst_connector(struct drm_dp_mst_topology_mgr *mgr, @@ -512,6 +518,7 @@ static void intel_dp_mst_hotplug(struct drm_dp_mst_topology_mgr *mgr) static struct drm_dp_mst_topology_cbs mst_cbs = { .add_connector = intel_dp_add_mst_connector, + .register_connector = intel_dp_register_mst_connector, .destroy_connector = intel_dp_destroy_mst_connector, .hotplug = intel_dp_mst_hotplug, }; diff --git a/drivers/gpu/drm/i915/intel_hotplug.c b/drivers/gpu/drm/i915/intel_hotplug.c index 53c0173a39fe..b17785719598 100644 --- a/drivers/gpu/drm/i915/intel_hotplug.c +++ b/drivers/gpu/drm/i915/intel_hotplug.c @@ -180,7 +180,7 @@ static void intel_hpd_irq_storm_disable(struct drm_i915_private *dev_priv) /* Enable polling and queue hotplug re-enabling. */ if (hpd_disabled) { - drm_kms_helper_poll_enable(dev); + drm_kms_helper_poll_enable_locked(dev); mod_delayed_work(system_wq, &dev_priv->hotplug.reenable_work, msecs_to_jiffies(HPD_STORM_REENABLE_DELAY)); } diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index 72e0edd7bbde..7412caedcf7f 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -484,18 +484,18 @@ void intel_lrc_irq_handler(struct intel_engine_cs *ring) status_pointer = I915_READ(RING_CONTEXT_STATUS_PTR(ring)); read_pointer = ring->next_context_status_buffer; - write_pointer = status_pointer & 0x07; + write_pointer = status_pointer & GEN8_CSB_PTR_MASK; if (read_pointer > write_pointer) - write_pointer += 6; + write_pointer += GEN8_CSB_ENTRIES; spin_lock(&ring->execlist_lock); while (read_pointer < write_pointer) { read_pointer++; status = I915_READ(RING_CONTEXT_STATUS_BUF(ring) + - (read_pointer % 6) * 8); + (read_pointer % GEN8_CSB_ENTRIES) * 8); status_id = I915_READ(RING_CONTEXT_STATUS_BUF(ring) + - (read_pointer % 6) * 8 + 4); + (read_pointer % GEN8_CSB_ENTRIES) * 8 + 4); if (status & GEN8_CTX_STATUS_IDLE_ACTIVE) continue; @@ -521,10 +521,12 @@ void intel_lrc_irq_handler(struct intel_engine_cs *ring) spin_unlock(&ring->execlist_lock); WARN(submit_contexts > 2, "More than two context complete events?\n"); - ring->next_context_status_buffer = write_pointer % 6; + ring->next_context_status_buffer = write_pointer % GEN8_CSB_ENTRIES; I915_WRITE(RING_CONTEXT_STATUS_PTR(ring), - _MASKED_FIELD(0x07 << 8, ((u32)ring->next_context_status_buffer & 0x07) << 8)); + _MASKED_FIELD(GEN8_CSB_PTR_MASK << 8, + ((u32)ring->next_context_status_buffer & + GEN8_CSB_PTR_MASK) << 8)); } static int execlists_context_queue(struct drm_i915_gem_request *request) @@ -1422,6 +1424,7 @@ static int gen8_init_common_ring(struct intel_engine_cs *ring) { struct drm_device *dev = ring->dev; struct drm_i915_private *dev_priv = dev->dev_private; + u8 next_context_status_buffer_hw; I915_WRITE_IMR(ring, ~(ring->irq_enable_mask | ring->irq_keep_mask)); I915_WRITE(RING_HWSTAM(ring->mmio_base), 0xffffffff); @@ -1436,7 +1439,29 @@ static int gen8_init_common_ring(struct intel_engine_cs *ring) _MASKED_BIT_DISABLE(GFX_REPLAY_MODE) | _MASKED_BIT_ENABLE(GFX_RUN_LIST_ENABLE)); POSTING_READ(RING_MODE_GEN7(ring)); - ring->next_context_status_buffer = 0; + + /* + * Instead of resetting the Context Status Buffer (CSB) read pointer to + * zero, we need to read the write pointer from hardware and use its + * value because "this register is power context save restored". + * Effectively, these states have been observed: + * + * | Suspend-to-idle (freeze) | Suspend-to-RAM (mem) | + * BDW | CSB regs not reset | CSB regs reset | + * CHT | CSB regs not reset | CSB regs not reset | + */ + next_context_status_buffer_hw = (I915_READ(RING_CONTEXT_STATUS_PTR(ring)) + & GEN8_CSB_PTR_MASK); + + /* + * When the CSB registers are reset (also after power-up / gpu reset), + * CSB write pointer is set to all 1's, which is not valid, use '5' in + * this special case, so the first element read is CSB[0]. + */ + if (next_context_status_buffer_hw == GEN8_CSB_PTR_MASK) + next_context_status_buffer_hw = (GEN8_CSB_ENTRIES - 1); + + ring->next_context_status_buffer = next_context_status_buffer_hw; DRM_DEBUG_DRIVER("Execlists enabled for %s\n", ring->name); memset(&ring->hangcheck, 0, sizeof(ring->hangcheck)); diff --git a/drivers/gpu/drm/i915/intel_lrc.h b/drivers/gpu/drm/i915/intel_lrc.h index 64f89f9982a2..3c63bb32ad81 100644 --- a/drivers/gpu/drm/i915/intel_lrc.h +++ b/drivers/gpu/drm/i915/intel_lrc.h @@ -25,6 +25,8 @@ #define _INTEL_LRC_H_ #define GEN8_LR_CONTEXT_ALIGN 4096 +#define GEN8_CSB_ENTRIES 6 +#define GEN8_CSB_PTR_MASK 0x07 /* Execlists regs */ #define RING_ELSP(ring) ((ring)->mmio_base+0x230) diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c index af7fdb3bd663..7401cf90b0db 100644 --- a/drivers/gpu/drm/i915/intel_runtime_pm.c +++ b/drivers/gpu/drm/i915/intel_runtime_pm.c @@ -246,7 +246,8 @@ static void skl_power_well_post_enable(struct drm_i915_private *dev_priv, } if (power_well->data == SKL_DISP_PW_1) { - intel_prepare_ddi(dev); + if (!dev_priv->power_domains.initializing) + intel_prepare_ddi(dev); gen8_irq_power_well_post_enable(dev_priv, 1 << PIPE_A); } } diff --git a/drivers/gpu/drm/qxl/qxl_display.c b/drivers/gpu/drm/qxl/qxl_display.c index dd845f82cc24..4649bd2ed340 100644 --- a/drivers/gpu/drm/qxl/qxl_display.c +++ b/drivers/gpu/drm/qxl/qxl_display.c @@ -618,7 +618,7 @@ static int qxl_crtc_mode_set(struct drm_crtc *crtc, adjusted_mode->hdisplay, adjusted_mode->vdisplay); - if (qcrtc->index == 0) + if (bo->is_primary == false) recreate_primary = true; if (bo->surf.stride * bo->surf.height > qdev->vram_size) { diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c index c3872598b85a..65adb9c72377 100644 --- a/drivers/gpu/drm/radeon/atombios_encoders.c +++ b/drivers/gpu/drm/radeon/atombios_encoders.c @@ -1624,8 +1624,9 @@ radeon_atom_encoder_dpms_avivo(struct drm_encoder *encoder, int mode) } else atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) { - args.ucAction = ATOM_LCD_BLON; - atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; + + atombios_set_backlight_level(radeon_encoder, dig->backlight_level); } break; case DRM_MODE_DPMS_STANDBY: @@ -1706,8 +1707,7 @@ radeon_atom_encoder_dpms_dig(struct drm_encoder *encoder, int mode) atombios_dig_encoder_setup(encoder, ATOM_ENCODER_CMD_DP_VIDEO_ON, 0); } if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) - atombios_dig_transmitter_setup(encoder, - ATOM_TRANSMITTER_ACTION_LCD_BLON, 0, 0); + atombios_set_backlight_level(radeon_encoder, dig->backlight_level); if (ext_encoder) atombios_external_encoder_setup(encoder, ext_encoder, ATOM_ENABLE); break; diff --git a/drivers/gpu/drm/radeon/radeon_dp_mst.c b/drivers/gpu/drm/radeon/radeon_dp_mst.c index 5e09c061847f..6cddae44fa6e 100644 --- a/drivers/gpu/drm/radeon/radeon_dp_mst.c +++ b/drivers/gpu/drm/radeon/radeon_dp_mst.c @@ -265,7 +265,6 @@ static struct drm_connector *radeon_dp_add_mst_connector(struct drm_dp_mst_topol { struct radeon_connector *master = container_of(mgr, struct radeon_connector, mst_mgr); struct drm_device *dev = master->base.dev; - struct radeon_device *rdev = dev->dev_private; struct radeon_connector *radeon_connector; struct drm_connector *connector; @@ -286,12 +285,19 @@ static struct drm_connector *radeon_dp_add_mst_connector(struct drm_dp_mst_topol drm_object_attach_property(&connector->base, dev->mode_config.path_property, 0); drm_mode_connector_set_path_property(connector, pathprop); + return connector; +} + +static void radeon_dp_register_mst_connector(struct drm_connector *connector) +{ + struct drm_device *dev = connector->dev; + struct radeon_device *rdev = dev->dev_private; + drm_modeset_lock_all(dev); radeon_fb_add_connector(rdev, connector); drm_modeset_unlock_all(dev); drm_connector_register(connector); - return connector; } static void radeon_dp_destroy_mst_connector(struct drm_dp_mst_topology_mgr *mgr, @@ -324,6 +330,7 @@ static void radeon_dp_mst_hotplug(struct drm_dp_mst_topology_mgr *mgr) struct drm_dp_mst_topology_cbs mst_cbs = { .add_connector = radeon_dp_add_mst_connector, + .register_connector = radeon_dp_register_mst_connector, .destroy_connector = radeon_dp_destroy_mst_connector, .hotplug = radeon_dp_mst_hotplug, }; diff --git a/drivers/gpu/drm/radeon/radeon_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c index 7214858ffcea..1aa657fe31cb 100644 --- a/drivers/gpu/drm/radeon/radeon_fb.c +++ b/drivers/gpu/drm/radeon/radeon_fb.c @@ -48,40 +48,10 @@ struct radeon_fbdev { struct radeon_device *rdev; }; -/** - * radeon_fb_helper_set_par - Hide cursor on CRTCs used by fbdev. - * - * @info: fbdev info - * - * This function hides the cursor on all CRTCs used by fbdev. - */ -static int radeon_fb_helper_set_par(struct fb_info *info) -{ - int ret; - - ret = drm_fb_helper_set_par(info); - - /* XXX: with universal plane support fbdev will automatically disable - * all non-primary planes (including the cursor) - */ - if (ret == 0) { - struct drm_fb_helper *fb_helper = info->par; - int i; - - for (i = 0; i < fb_helper->crtc_count; i++) { - struct drm_crtc *crtc = fb_helper->crtc_info[i].mode_set.crtc; - - radeon_crtc_cursor_set2(crtc, NULL, 0, 0, 0, 0, 0); - } - } - - return ret; -} - static struct fb_ops radeonfb_ops = { .owner = THIS_MODULE, .fb_check_var = drm_fb_helper_check_var, - .fb_set_par = radeon_fb_helper_set_par, + .fb_set_par = drm_fb_helper_set_par, .fb_fillrect = drm_fb_helper_cfb_fillrect, .fb_copyarea = drm_fb_helper_cfb_copyarea, .fb_imageblit = drm_fb_helper_cfb_imageblit, diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c index 5ae8f921da2a..8a76821177a6 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_cmdbuf.c @@ -681,6 +681,14 @@ static bool vmw_cmdbuf_try_alloc(struct vmw_cmdbuf_man *man, 0, 0, DRM_MM_SEARCH_DEFAULT, DRM_MM_CREATE_DEFAULT); + if (ret) { + (void) vmw_cmdbuf_man_process(man); + ret = drm_mm_insert_node_generic(&man->mm, info->node, + info->page_size, 0, 0, + DRM_MM_SEARCH_DEFAULT, + DRM_MM_CREATE_DEFAULT); + } + spin_unlock_bh(&man->lock); info->done = !ret; diff --git a/drivers/hwmon/abx500.c b/drivers/hwmon/abx500.c index 6cb89c0ebab6..1fd46859ed29 100644 --- a/drivers/hwmon/abx500.c +++ b/drivers/hwmon/abx500.c @@ -470,6 +470,7 @@ static const struct of_device_id abx500_temp_match[] = { { .compatible = "stericsson,abx500-temp" }, {}, }; +MODULE_DEVICE_TABLE(of, abx500_temp_match); #endif static struct platform_driver abx500_temp_driver = { diff --git a/drivers/hwmon/gpio-fan.c b/drivers/hwmon/gpio-fan.c index a3dae6d0082a..82de3deeb18a 100644 --- a/drivers/hwmon/gpio-fan.c +++ b/drivers/hwmon/gpio-fan.c @@ -539,6 +539,7 @@ static const struct of_device_id of_gpio_fan_match[] = { { .compatible = "gpio-fan", }, {}, }; +MODULE_DEVICE_TABLE(of, of_gpio_fan_match); #endif /* CONFIG_OF_GPIO */ static int gpio_fan_probe(struct platform_device *pdev) diff --git a/drivers/hwmon/pwm-fan.c b/drivers/hwmon/pwm-fan.c index 2d9a712699ff..3e23003f78b0 100644 --- a/drivers/hwmon/pwm-fan.c +++ b/drivers/hwmon/pwm-fan.c @@ -323,6 +323,7 @@ static const struct of_device_id of_pwm_fan_match[] = { { .compatible = "pwm-fan", }, {}, }; +MODULE_DEVICE_TABLE(of, of_pwm_fan_match); static struct platform_driver pwm_fan_driver = { .probe = pwm_fan_probe, diff --git a/drivers/hwtracing/coresight/coresight-etm3x.c b/drivers/hwtracing/coresight/coresight-etm3x.c index bf2476ed9356..d630b7ece735 100644 --- a/drivers/hwtracing/coresight/coresight-etm3x.c +++ b/drivers/hwtracing/coresight/coresight-etm3x.c @@ -191,7 +191,8 @@ static void etm_set_prog(struct etm_drvdata *drvdata) isb(); if (coresight_timeout_etm(drvdata, ETMSR, ETMSR_PROG_BIT, 1)) { dev_err(drvdata->dev, - "timeout observed when probing at offset %#x\n", ETMSR); + "%s: timeout observed when probing at offset %#x\n", + __func__, ETMSR); } } @@ -209,7 +210,8 @@ static void etm_clr_prog(struct etm_drvdata *drvdata) isb(); if (coresight_timeout_etm(drvdata, ETMSR, ETMSR_PROG_BIT, 0)) { dev_err(drvdata->dev, - "timeout observed when probing at offset %#x\n", ETMSR); + "%s: timeout observed when probing at offset %#x\n", + __func__, ETMSR); } } @@ -313,14 +315,6 @@ static void etm_enable_hw(void *info) dev_dbg(drvdata->dev, "cpu: %d enable smp call done\n", drvdata->cpu); } -static int etm_trace_id_simple(struct etm_drvdata *drvdata) -{ - if (!drvdata->enable) - return drvdata->traceid; - - return (etm_readl(drvdata, ETMTRACEIDR) & ETM_TRACEID_MASK); -} - static int etm_trace_id(struct coresight_device *csdev) { struct etm_drvdata *drvdata = dev_get_drvdata(csdev->dev.parent); @@ -1506,44 +1500,17 @@ static ssize_t timestamp_event_store(struct device *dev, } static DEVICE_ATTR_RW(timestamp_event); -static ssize_t status_show(struct device *dev, - struct device_attribute *attr, char *buf) +static ssize_t cpu_show(struct device *dev, + struct device_attribute *attr, char *buf) { - int ret; - unsigned long flags; + int val; struct etm_drvdata *drvdata = dev_get_drvdata(dev->parent); - pm_runtime_get_sync(drvdata->dev); - spin_lock_irqsave(&drvdata->spinlock, flags); - - CS_UNLOCK(drvdata->base); - ret = sprintf(buf, - "ETMCCR: 0x%08x\n" - "ETMCCER: 0x%08x\n" - "ETMSCR: 0x%08x\n" - "ETMIDR: 0x%08x\n" - "ETMCR: 0x%08x\n" - "ETMTRACEIDR: 0x%08x\n" - "Enable event: 0x%08x\n" - "Enable start/stop: 0x%08x\n" - "Enable control: CR1 0x%08x CR2 0x%08x\n" - "CPU affinity: %d\n", - drvdata->etmccr, drvdata->etmccer, - etm_readl(drvdata, ETMSCR), etm_readl(drvdata, ETMIDR), - etm_readl(drvdata, ETMCR), etm_trace_id_simple(drvdata), - etm_readl(drvdata, ETMTEEVR), - etm_readl(drvdata, ETMTSSCR), - etm_readl(drvdata, ETMTECR1), - etm_readl(drvdata, ETMTECR2), - drvdata->cpu); - CS_LOCK(drvdata->base); - - spin_unlock_irqrestore(&drvdata->spinlock, flags); - pm_runtime_put(drvdata->dev); + val = drvdata->cpu; + return scnprintf(buf, PAGE_SIZE, "%d\n", val); - return ret; } -static DEVICE_ATTR_RO(status); +static DEVICE_ATTR_RO(cpu); static ssize_t traceid_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -1619,11 +1586,61 @@ static struct attribute *coresight_etm_attrs[] = { &dev_attr_ctxid_mask.attr, &dev_attr_sync_freq.attr, &dev_attr_timestamp_event.attr, - &dev_attr_status.attr, &dev_attr_traceid.attr, + &dev_attr_cpu.attr, + NULL, +}; + +#define coresight_simple_func(name, offset) \ +static ssize_t name##_show(struct device *_dev, \ + struct device_attribute *attr, char *buf) \ +{ \ + struct etm_drvdata *drvdata = dev_get_drvdata(_dev->parent); \ + return scnprintf(buf, PAGE_SIZE, "0x%x\n", \ + readl_relaxed(drvdata->base + offset)); \ +} \ +DEVICE_ATTR_RO(name) + +coresight_simple_func(etmccr, ETMCCR); +coresight_simple_func(etmccer, ETMCCER); +coresight_simple_func(etmscr, ETMSCR); +coresight_simple_func(etmidr, ETMIDR); +coresight_simple_func(etmcr, ETMCR); +coresight_simple_func(etmtraceidr, ETMTRACEIDR); +coresight_simple_func(etmteevr, ETMTEEVR); +coresight_simple_func(etmtssvr, ETMTSSCR); +coresight_simple_func(etmtecr1, ETMTECR1); +coresight_simple_func(etmtecr2, ETMTECR2); + +static struct attribute *coresight_etm_mgmt_attrs[] = { + &dev_attr_etmccr.attr, + &dev_attr_etmccer.attr, + &dev_attr_etmscr.attr, + &dev_attr_etmidr.attr, + &dev_attr_etmcr.attr, + &dev_attr_etmtraceidr.attr, + &dev_attr_etmteevr.attr, + &dev_attr_etmtssvr.attr, + &dev_attr_etmtecr1.attr, + &dev_attr_etmtecr2.attr, + NULL, +}; + +static const struct attribute_group coresight_etm_group = { + .attrs = coresight_etm_attrs, +}; + + +static const struct attribute_group coresight_etm_mgmt_group = { + .attrs = coresight_etm_mgmt_attrs, + .name = "mgmt", +}; + +static const struct attribute_group *coresight_etm_groups[] = { + &coresight_etm_group, + &coresight_etm_mgmt_group, NULL, }; -ATTRIBUTE_GROUPS(coresight_etm); static int etm_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) diff --git a/drivers/hwtracing/coresight/coresight-etm4x.c b/drivers/hwtracing/coresight/coresight-etm4x.c index 254a81a4e6f4..a6707642bb23 100644 --- a/drivers/hwtracing/coresight/coresight-etm4x.c +++ b/drivers/hwtracing/coresight/coresight-etm4x.c @@ -136,7 +136,9 @@ static void etm4_enable_hw(void *info) writel_relaxed(drvdata->cntr_val[i], drvdata->base + TRCCNTVRn(i)); } - for (i = 0; i < drvdata->nr_resource; i++) + + /* Resource selector pair 0 is always implemented and reserved */ + for (i = 2; i < drvdata->nr_resource * 2; i++) writel_relaxed(drvdata->res_ctrl[i], drvdata->base + TRCRSCTLRn(i)); @@ -489,8 +491,9 @@ static ssize_t reset_store(struct device *dev, drvdata->cntr_val[i] = 0x0; } - drvdata->res_idx = 0x0; - for (i = 0; i < drvdata->nr_resource; i++) + /* Resource selector pair 0 is always implemented and reserved */ + drvdata->res_idx = 0x2; + for (i = 2; i < drvdata->nr_resource * 2; i++) drvdata->res_ctrl[i] = 0x0; for (i = 0; i < drvdata->nr_ss_cmp; i++) { @@ -1732,7 +1735,7 @@ static ssize_t res_idx_store(struct device *dev, if (kstrtoul(buf, 16, &val)) return -EINVAL; /* Resource selector pair 0 is always implemented and reserved */ - if ((val == 0) || (val >= drvdata->nr_resource)) + if (val < 2 || val >= drvdata->nr_resource * 2) return -EINVAL; /* @@ -2416,8 +2419,13 @@ static void etm4_init_arch_data(void *info) drvdata->nr_addr_cmp = BMVAL(etmidr4, 0, 3); /* NUMPC, bits[15:12] number of PE comparator inputs for tracing */ drvdata->nr_pe_cmp = BMVAL(etmidr4, 12, 15); - /* NUMRSPAIR, bits[19:16] the number of resource pairs for tracing */ - drvdata->nr_resource = BMVAL(etmidr4, 16, 19); + /* + * NUMRSPAIR, bits[19:16] + * The number of resource pairs conveyed by the HW starts at 0, i.e a + * value of 0x0 indicate 1 resource pair, 0x1 indicate two and so on. + * As such add 1 to the value of NUMRSPAIR for a better representation. + */ + drvdata->nr_resource = BMVAL(etmidr4, 16, 19) + 1; /* * NUMSSCC, bits[23:20] the number of single-shot * comparator control for tracing @@ -2504,6 +2512,8 @@ static void etm4_init_default_data(struct etmv4_drvdata *drvdata) drvdata->cntr_val[i] = 0x0; } + /* Resource selector pair 0 is always implemented and reserved */ + drvdata->res_idx = 0x2; for (i = 2; i < drvdata->nr_resource * 2; i++) drvdata->res_ctrl[i] = 0x0; diff --git a/drivers/hwtracing/coresight/coresight.c b/drivers/hwtracing/coresight/coresight.c index 894531d315b8..e25492137d8b 100644 --- a/drivers/hwtracing/coresight/coresight.c +++ b/drivers/hwtracing/coresight/coresight.c @@ -240,6 +240,11 @@ static int coresight_enable_path(struct list_head *path) int ret = 0; struct coresight_device *cd; + /* + * At this point we have a full @path, from source to sink. The + * sink is the first entry and the source the last one. Go through + * all the components and enable them one by one. + */ list_for_each_entry(cd, path, path_link) { if (cd == list_first_entry(path, struct coresight_device, path_link)) { diff --git a/drivers/hwtracing/intel_th/Kconfig b/drivers/hwtracing/intel_th/Kconfig new file mode 100644 index 000000000000..b7a9073d968b --- /dev/null +++ b/drivers/hwtracing/intel_th/Kconfig @@ -0,0 +1,72 @@ +config INTEL_TH + tristate "Intel(R) Trace Hub controller" + help + Intel(R) Trace Hub (TH) is a set of hardware blocks (subdevices) that + produce, switch and output trace data from multiple hardware and + software sources over several types of trace output ports encoded + in System Trace Protocol (MIPI STPv2) and is intended to perform + full system debugging. + + This option enables intel_th bus and common code used by TH + subdevices to interact with each other and hardware and for + platform glue layers to drive Intel TH devices. + + Say Y here to enable Intel(R) Trace Hub controller support. + +if INTEL_TH + +config INTEL_TH_PCI + tristate "Intel(R) Trace Hub PCI controller" + depends on PCI + help + Intel(R) Trace Hub may exist as a PCI device. This option enables + support glue layer for PCI-based Intel TH. + + Say Y here to enable PCI Intel TH support. + +config INTEL_TH_GTH + tristate "Intel(R) Trace Hub Global Trace Hub" + help + Global Trace Hub (GTH) is the central component of the + Intel TH infrastructure and acts as a switch for source + and output devices. This driver is required for other + Intel TH subdevices to initialize. + + Say Y here to enable GTH subdevice of Intel(R) Trace Hub. + +config INTEL_TH_STH + tristate "Intel(R) Trace Hub Software Trace Hub support" + depends on STM + help + Software Trace Hub (STH) enables trace data from software + trace sources to be sent out via Intel(R) Trace Hub. It + uses stm class device to interface with its sources. + + Say Y here to enable STH subdevice of Intel(R) Trace Hub. + +config INTEL_TH_MSU + tristate "Intel(R) Trace Hub Memory Storage Unit" + help + Memory Storage Unit (MSU) trace output device enables + storing STP traces to system memory. It supports single + and multiblock modes of operation and provides read() + and mmap() access to the collected data. + + Say Y here to enable MSU output device for Intel TH. + +config INTEL_TH_PTI + tristate "Intel(R) Trace Hub PTI output" + help + Parallel Trace Interface unit (PTI) is a trace output device + of Intel TH architecture that facilitates STP trace output via + a PTI port. + + Say Y to enable PTI output of Intel TH data. + +config INTEL_TH_DEBUG + bool "Intel(R) Trace Hub debugging" + depends on DEBUG_FS + help + Say Y here to enable debugging. + +endif diff --git a/drivers/hwtracing/intel_th/Makefile b/drivers/hwtracing/intel_th/Makefile new file mode 100644 index 000000000000..81d42fe918f7 --- /dev/null +++ b/drivers/hwtracing/intel_th/Makefile @@ -0,0 +1,18 @@ +obj-$(CONFIG_INTEL_TH) += intel_th.o +intel_th-y := core.o +intel_th-$(CONFIG_INTEL_TH_DEBUG) += debug.o + +obj-$(CONFIG_INTEL_TH_PCI) += intel_th_pci.o +intel_th_pci-y := pci.o + +obj-$(CONFIG_INTEL_TH_GTH) += intel_th_gth.o +intel_th_gth-y := gth.o + +obj-$(CONFIG_INTEL_TH_STH) += intel_th_sth.o +intel_th_sth-y := sth.o + +obj-$(CONFIG_INTEL_TH_MSU) += intel_th_msu.o +intel_th_msu-y := msu.o + +obj-$(CONFIG_INTEL_TH_PTI) += intel_th_pti.o +intel_th_pti-y := pti.o diff --git a/drivers/hwtracing/intel_th/core.c b/drivers/hwtracing/intel_th/core.c new file mode 100644 index 000000000000..165d3001c301 --- /dev/null +++ b/drivers/hwtracing/intel_th/core.c @@ -0,0 +1,692 @@ +/* + * Intel(R) Trace Hub driver core + * + * Copyright (C) 2014-2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/types.h> +#include <linux/module.h> +#include <linux/device.h> +#include <linux/sysfs.h> +#include <linux/kdev_t.h> +#include <linux/debugfs.h> +#include <linux/idr.h> +#include <linux/pci.h> +#include <linux/dma-mapping.h> + +#include "intel_th.h" +#include "debug.h" + +static DEFINE_IDA(intel_th_ida); + +static int intel_th_match(struct device *dev, struct device_driver *driver) +{ + struct intel_th_driver *thdrv = to_intel_th_driver(driver); + struct intel_th_device *thdev = to_intel_th_device(dev); + + if (thdev->type == INTEL_TH_SWITCH && + (!thdrv->enable || !thdrv->disable)) + return 0; + + return !strcmp(thdev->name, driver->name); +} + +static int intel_th_child_remove(struct device *dev, void *data) +{ + device_release_driver(dev); + + return 0; +} + +static int intel_th_probe(struct device *dev) +{ + struct intel_th_driver *thdrv = to_intel_th_driver(dev->driver); + struct intel_th_device *thdev = to_intel_th_device(dev); + struct intel_th_driver *hubdrv; + struct intel_th_device *hub = NULL; + int ret; + + if (thdev->type == INTEL_TH_SWITCH) + hub = thdev; + else if (dev->parent) + hub = to_intel_th_device(dev->parent); + + if (!hub || !hub->dev.driver) + return -EPROBE_DEFER; + + hubdrv = to_intel_th_driver(hub->dev.driver); + + ret = thdrv->probe(to_intel_th_device(dev)); + if (ret) + return ret; + + if (thdev->type == INTEL_TH_OUTPUT && + !intel_th_output_assigned(thdev)) + ret = hubdrv->assign(hub, thdev); + + return ret; +} + +static int intel_th_remove(struct device *dev) +{ + struct intel_th_driver *thdrv = to_intel_th_driver(dev->driver); + struct intel_th_device *thdev = to_intel_th_device(dev); + struct intel_th_device *hub = to_intel_th_device(dev->parent); + int err; + + if (thdev->type == INTEL_TH_SWITCH) { + err = device_for_each_child(dev, thdev, intel_th_child_remove); + if (err) + return err; + } + + thdrv->remove(thdev); + + if (intel_th_output_assigned(thdev)) { + struct intel_th_driver *hubdrv = + to_intel_th_driver(dev->parent->driver); + + if (hub->dev.driver) + hubdrv->unassign(hub, thdev); + } + + return 0; +} + +static struct bus_type intel_th_bus = { + .name = "intel_th", + .dev_attrs = NULL, + .match = intel_th_match, + .probe = intel_th_probe, + .remove = intel_th_remove, +}; + +static void intel_th_device_free(struct intel_th_device *thdev); + +static void intel_th_device_release(struct device *dev) +{ + intel_th_device_free(to_intel_th_device(dev)); +} + +static struct device_type intel_th_source_device_type = { + .name = "intel_th_source_device", + .release = intel_th_device_release, +}; + +static char *intel_th_output_devnode(struct device *dev, umode_t *mode, + kuid_t *uid, kgid_t *gid) +{ + struct intel_th_device *thdev = to_intel_th_device(dev); + char *node; + + if (thdev->id >= 0) + node = kasprintf(GFP_KERNEL, "intel_th%d/%s%d", 0, thdev->name, + thdev->id); + else + node = kasprintf(GFP_KERNEL, "intel_th%d/%s", 0, thdev->name); + + return node; +} + +static ssize_t port_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct intel_th_device *thdev = to_intel_th_device(dev); + + if (thdev->output.port >= 0) + return scnprintf(buf, PAGE_SIZE, "%u\n", thdev->output.port); + + return scnprintf(buf, PAGE_SIZE, "unassigned\n"); +} + +static DEVICE_ATTR_RO(port); + +static int intel_th_output_activate(struct intel_th_device *thdev) +{ + struct intel_th_driver *thdrv = to_intel_th_driver(thdev->dev.driver); + + if (thdrv->activate) + return thdrv->activate(thdev); + + intel_th_trace_enable(thdev); + + return 0; +} + +static void intel_th_output_deactivate(struct intel_th_device *thdev) +{ + struct intel_th_driver *thdrv = to_intel_th_driver(thdev->dev.driver); + + if (thdrv->deactivate) + thdrv->deactivate(thdev); + else + intel_th_trace_disable(thdev); +} + +static ssize_t active_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct intel_th_device *thdev = to_intel_th_device(dev); + + return scnprintf(buf, PAGE_SIZE, "%d\n", thdev->output.active); +} + +static ssize_t active_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t size) +{ + struct intel_th_device *thdev = to_intel_th_device(dev); + unsigned long val; + int ret; + + ret = kstrtoul(buf, 10, &val); + if (ret) + return ret; + + if (!!val != thdev->output.active) { + if (val) + ret = intel_th_output_activate(thdev); + else + intel_th_output_deactivate(thdev); + } + + return ret ? ret : size; +} + +static DEVICE_ATTR_RW(active); + +static struct attribute *intel_th_output_attrs[] = { + &dev_attr_port.attr, + &dev_attr_active.attr, + NULL, +}; + +ATTRIBUTE_GROUPS(intel_th_output); + +static struct device_type intel_th_output_device_type = { + .name = "intel_th_output_device", + .groups = intel_th_output_groups, + .release = intel_th_device_release, + .devnode = intel_th_output_devnode, +}; + +static struct device_type intel_th_switch_device_type = { + .name = "intel_th_switch_device", + .release = intel_th_device_release, +}; + +static struct device_type *intel_th_device_type[] = { + [INTEL_TH_SOURCE] = &intel_th_source_device_type, + [INTEL_TH_OUTPUT] = &intel_th_output_device_type, + [INTEL_TH_SWITCH] = &intel_th_switch_device_type, +}; + +int intel_th_driver_register(struct intel_th_driver *thdrv) +{ + if (!thdrv->probe || !thdrv->remove) + return -EINVAL; + + thdrv->driver.bus = &intel_th_bus; + + return driver_register(&thdrv->driver); +} +EXPORT_SYMBOL_GPL(intel_th_driver_register); + +void intel_th_driver_unregister(struct intel_th_driver *thdrv) +{ + driver_unregister(&thdrv->driver); +} +EXPORT_SYMBOL_GPL(intel_th_driver_unregister); + +static struct intel_th_device * +intel_th_device_alloc(struct intel_th *th, unsigned int type, const char *name, + int id) +{ + struct device *parent; + struct intel_th_device *thdev; + + if (type == INTEL_TH_SWITCH) + parent = th->dev; + else + parent = &th->hub->dev; + + thdev = kzalloc(sizeof(*thdev) + strlen(name) + 1, GFP_KERNEL); + if (!thdev) + return NULL; + + thdev->id = id; + thdev->type = type; + + strcpy(thdev->name, name); + device_initialize(&thdev->dev); + thdev->dev.bus = &intel_th_bus; + thdev->dev.type = intel_th_device_type[type]; + thdev->dev.parent = parent; + thdev->dev.dma_mask = parent->dma_mask; + thdev->dev.dma_parms = parent->dma_parms; + dma_set_coherent_mask(&thdev->dev, parent->coherent_dma_mask); + if (id >= 0) + dev_set_name(&thdev->dev, "%d-%s%d", th->id, name, id); + else + dev_set_name(&thdev->dev, "%d-%s", th->id, name); + + return thdev; +} + +static int intel_th_device_add_resources(struct intel_th_device *thdev, + struct resource *res, int nres) +{ + struct resource *r; + + r = kmemdup(res, sizeof(*res) * nres, GFP_KERNEL); + if (!r) + return -ENOMEM; + + thdev->resource = r; + thdev->num_resources = nres; + + return 0; +} + +static void intel_th_device_remove(struct intel_th_device *thdev) +{ + device_del(&thdev->dev); + put_device(&thdev->dev); +} + +static void intel_th_device_free(struct intel_th_device *thdev) +{ + kfree(thdev->resource); + kfree(thdev); +} + +/* + * Intel(R) Trace Hub subdevices + */ +static struct intel_th_subdevice { + const char *name; + struct resource res[3]; + unsigned nres; + unsigned type; + unsigned otype; + int id; +} intel_th_subdevices[TH_SUBDEVICE_MAX] = { + { + .nres = 1, + .res = { + { + .start = REG_GTH_OFFSET, + .end = REG_GTH_OFFSET + REG_GTH_LENGTH - 1, + .flags = IORESOURCE_MEM, + }, + }, + .name = "gth", + .type = INTEL_TH_SWITCH, + .id = -1, + }, + { + .nres = 2, + .res = { + { + .start = REG_MSU_OFFSET, + .end = REG_MSU_OFFSET + REG_MSU_LENGTH - 1, + .flags = IORESOURCE_MEM, + }, + { + .start = BUF_MSU_OFFSET, + .end = BUF_MSU_OFFSET + BUF_MSU_LENGTH - 1, + .flags = IORESOURCE_MEM, + }, + }, + .name = "msc", + .id = 0, + .type = INTEL_TH_OUTPUT, + .otype = GTH_MSU, + }, + { + .nres = 2, + .res = { + { + .start = REG_MSU_OFFSET, + .end = REG_MSU_OFFSET + REG_MSU_LENGTH - 1, + .flags = IORESOURCE_MEM, + }, + { + .start = BUF_MSU_OFFSET, + .end = BUF_MSU_OFFSET + BUF_MSU_LENGTH - 1, + .flags = IORESOURCE_MEM, + }, + }, + .name = "msc", + .id = 1, + .type = INTEL_TH_OUTPUT, + .otype = GTH_MSU, + }, + { + .nres = 2, + .res = { + { + .start = REG_STH_OFFSET, + .end = REG_STH_OFFSET + REG_STH_LENGTH - 1, + .flags = IORESOURCE_MEM, + }, + { + .start = TH_MMIO_SW, + .end = 0, + .flags = IORESOURCE_MEM, + }, + }, + .id = -1, + .name = "sth", + .type = INTEL_TH_SOURCE, + }, + { + .nres = 1, + .res = { + { + .start = REG_PTI_OFFSET, + .end = REG_PTI_OFFSET + REG_PTI_LENGTH - 1, + .flags = IORESOURCE_MEM, + }, + }, + .id = -1, + .name = "pti", + .type = INTEL_TH_OUTPUT, + .otype = GTH_PTI, + }, + { + .nres = 1, + .res = { + { + .start = REG_DCIH_OFFSET, + .end = REG_DCIH_OFFSET + REG_DCIH_LENGTH - 1, + .flags = IORESOURCE_MEM, + }, + }, + .id = -1, + .name = "dcih", + .type = INTEL_TH_OUTPUT, + }, +}; + +static int intel_th_populate(struct intel_th *th, struct resource *devres, + unsigned int ndevres, int irq) +{ + struct resource res[3]; + unsigned int req = 0; + int i, err; + + /* create devices for each intel_th_subdevice */ + for (i = 0; i < ARRAY_SIZE(intel_th_subdevices); i++) { + struct intel_th_subdevice *subdev = &intel_th_subdevices[i]; + struct intel_th_device *thdev; + int r; + + thdev = intel_th_device_alloc(th, subdev->type, subdev->name, + subdev->id); + if (!thdev) { + err = -ENOMEM; + goto kill_subdevs; + } + + memcpy(res, subdev->res, + sizeof(struct resource) * subdev->nres); + + for (r = 0; r < subdev->nres; r++) { + int bar = TH_MMIO_CONFIG; + + /* + * Take .end == 0 to mean 'take the whole bar', + * .start then tells us which bar it is. Default to + * TH_MMIO_CONFIG. + */ + if (!res[r].end && res[r].flags == IORESOURCE_MEM) { + bar = res[r].start; + res[r].start = 0; + res[r].end = resource_size(&devres[bar]) - 1; + } + + if (res[r].flags & IORESOURCE_MEM) { + res[r].start += devres[bar].start; + res[r].end += devres[bar].start; + + dev_dbg(th->dev, "%s:%d @ %pR\n", + subdev->name, r, &res[r]); + } else if (res[r].flags & IORESOURCE_IRQ) { + res[r].start = irq; + } + } + + err = intel_th_device_add_resources(thdev, res, subdev->nres); + if (err) { + put_device(&thdev->dev); + goto kill_subdevs; + } + + if (subdev->type == INTEL_TH_OUTPUT) { + thdev->dev.devt = MKDEV(th->major, i); + thdev->output.type = subdev->otype; + thdev->output.port = -1; + } + + err = device_add(&thdev->dev); + if (err) { + put_device(&thdev->dev); + goto kill_subdevs; + } + + /* need switch driver to be loaded to enumerate the rest */ + if (subdev->type == INTEL_TH_SWITCH && !req) { + th->hub = thdev; + err = request_module("intel_th_%s", subdev->name); + if (!err) + req++; + } + + th->thdev[i] = thdev; + } + + return 0; + +kill_subdevs: + for (i-- ; i >= 0; i--) + intel_th_device_remove(th->thdev[i]); + + return err; +} + +static int match_devt(struct device *dev, void *data) +{ + dev_t devt = (dev_t)(unsigned long)data; + + return dev->devt == devt; +} + +static int intel_th_output_open(struct inode *inode, struct file *file) +{ + const struct file_operations *fops; + struct intel_th_driver *thdrv; + struct device *dev; + int err; + + dev = bus_find_device(&intel_th_bus, NULL, + (void *)(unsigned long)inode->i_rdev, + match_devt); + if (!dev || !dev->driver) + return -ENODEV; + + thdrv = to_intel_th_driver(dev->driver); + fops = fops_get(thdrv->fops); + if (!fops) + return -ENODEV; + + replace_fops(file, fops); + + file->private_data = to_intel_th_device(dev); + + if (file->f_op->open) { + err = file->f_op->open(inode, file); + return err; + } + + return 0; +} + +static const struct file_operations intel_th_output_fops = { + .open = intel_th_output_open, + .llseek = noop_llseek, +}; + +/** + * intel_th_alloc() - allocate a new Intel TH device and its subdevices + * @dev: parent device + * @devres: parent's resources + * @ndevres: number of resources + * @irq: irq number + */ +struct intel_th * +intel_th_alloc(struct device *dev, struct resource *devres, + unsigned int ndevres, int irq) +{ + struct intel_th *th; + int err; + + th = kzalloc(sizeof(*th), GFP_KERNEL); + if (!th) + return ERR_PTR(-ENOMEM); + + th->id = ida_simple_get(&intel_th_ida, 0, 0, GFP_KERNEL); + if (th->id < 0) { + err = th->id; + goto err_alloc; + } + + th->major = __register_chrdev(0, 0, TH_POSSIBLE_OUTPUTS, + "intel_th/output", &intel_th_output_fops); + if (th->major < 0) { + err = th->major; + goto err_ida; + } + th->dev = dev; + + err = intel_th_populate(th, devres, ndevres, irq); + if (err) + goto err_chrdev; + + return th; + +err_chrdev: + __unregister_chrdev(th->major, 0, TH_POSSIBLE_OUTPUTS, + "intel_th/output"); + +err_ida: + ida_simple_remove(&intel_th_ida, th->id); + +err_alloc: + kfree(th); + + return ERR_PTR(err); +} +EXPORT_SYMBOL_GPL(intel_th_alloc); + +void intel_th_free(struct intel_th *th) +{ + int i; + + for (i = 0; i < TH_SUBDEVICE_MAX; i++) + if (th->thdev[i] != th->hub) + intel_th_device_remove(th->thdev[i]); + + intel_th_device_remove(th->hub); + + __unregister_chrdev(th->major, 0, TH_POSSIBLE_OUTPUTS, + "intel_th/output"); + + ida_simple_remove(&intel_th_ida, th->id); + + kfree(th); +} +EXPORT_SYMBOL_GPL(intel_th_free); + +/** + * intel_th_trace_enable() - enable tracing for an output device + * @thdev: output device that requests tracing be enabled + */ +int intel_th_trace_enable(struct intel_th_device *thdev) +{ + struct intel_th_device *hub = to_intel_th_device(thdev->dev.parent); + struct intel_th_driver *hubdrv = to_intel_th_driver(hub->dev.driver); + + if (WARN_ON_ONCE(hub->type != INTEL_TH_SWITCH)) + return -EINVAL; + + if (WARN_ON_ONCE(thdev->type != INTEL_TH_OUTPUT)) + return -EINVAL; + + hubdrv->enable(hub, &thdev->output); + + return 0; +} +EXPORT_SYMBOL_GPL(intel_th_trace_enable); + +/** + * intel_th_trace_disable() - disable tracing for an output device + * @thdev: output device that requests tracing be disabled + */ +int intel_th_trace_disable(struct intel_th_device *thdev) +{ + struct intel_th_device *hub = to_intel_th_device(thdev->dev.parent); + struct intel_th_driver *hubdrv = to_intel_th_driver(hub->dev.driver); + + WARN_ON_ONCE(hub->type != INTEL_TH_SWITCH); + if (WARN_ON_ONCE(thdev->type != INTEL_TH_OUTPUT)) + return -EINVAL; + + hubdrv->disable(hub, &thdev->output); + + return 0; +} +EXPORT_SYMBOL_GPL(intel_th_trace_disable); + +int intel_th_set_output(struct intel_th_device *thdev, + unsigned int master) +{ + struct intel_th_device *hub = to_intel_th_device(thdev->dev.parent); + struct intel_th_driver *hubdrv = to_intel_th_driver(hub->dev.driver); + + if (!hubdrv->set_output) + return -ENOTSUPP; + + return hubdrv->set_output(hub, master); +} +EXPORT_SYMBOL_GPL(intel_th_set_output); + +static int __init intel_th_init(void) +{ + intel_th_debug_init(); + + return bus_register(&intel_th_bus); +} +subsys_initcall(intel_th_init); + +static void __exit intel_th_exit(void) +{ + intel_th_debug_done(); + + bus_unregister(&intel_th_bus); +} +module_exit(intel_th_exit); + +MODULE_LICENSE("GPL v2"); +MODULE_DESCRIPTION("Intel(R) Trace Hub controller driver"); +MODULE_AUTHOR("Alexander Shishkin <alexander.shishkin@linux.intel.com>"); diff --git a/drivers/hwtracing/intel_th/debug.c b/drivers/hwtracing/intel_th/debug.c new file mode 100644 index 000000000000..788a1f0a97ad --- /dev/null +++ b/drivers/hwtracing/intel_th/debug.c @@ -0,0 +1,36 @@ +/* + * Intel(R) Trace Hub driver debugging + * + * Copyright (C) 2014-2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ + +#include <linux/types.h> +#include <linux/device.h> +#include <linux/debugfs.h> + +#include "intel_th.h" +#include "debug.h" + +struct dentry *intel_th_dbg; + +void intel_th_debug_init(void) +{ + intel_th_dbg = debugfs_create_dir("intel_th", NULL); + if (IS_ERR(intel_th_dbg)) + intel_th_dbg = NULL; +} + +void intel_th_debug_done(void) +{ + debugfs_remove(intel_th_dbg); + intel_th_dbg = NULL; +} diff --git a/drivers/hwtracing/intel_th/debug.h b/drivers/hwtracing/intel_th/debug.h new file mode 100644 index 000000000000..88311bad3ba4 --- /dev/null +++ b/drivers/hwtracing/intel_th/debug.h @@ -0,0 +1,34 @@ +/* + * Intel(R) Trace Hub driver debugging + * + * Copyright (C) 2014-2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ + +#ifndef __INTEL_TH_DEBUG_H__ +#define __INTEL_TH_DEBUG_H__ + +#ifdef CONFIG_INTEL_TH_DEBUG +extern struct dentry *intel_th_dbg; + +void intel_th_debug_init(void); +void intel_th_debug_done(void); +#else +static inline void intel_th_debug_init(void) +{ +} + +static inline void intel_th_debug_done(void) +{ +} +#endif + +#endif /* __INTEL_TH_DEBUG_H__ */ diff --git a/drivers/hwtracing/intel_th/gth.c b/drivers/hwtracing/intel_th/gth.c new file mode 100644 index 000000000000..673d272ee400 --- /dev/null +++ b/drivers/hwtracing/intel_th/gth.c @@ -0,0 +1,706 @@ +/* + * Intel(R) Trace Hub Global Trace Hub + * + * Copyright (C) 2014-2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/types.h> +#include <linux/module.h> +#include <linux/device.h> +#include <linux/io.h> +#include <linux/mm.h> +#include <linux/slab.h> +#include <linux/bitmap.h> + +#include "intel_th.h" +#include "gth.h" + +struct gth_device; + +/** + * struct gth_output - GTH view on an output port + * @gth: backlink to the GTH device + * @output: link to output device's output descriptor + * @index: output port number + * @port_type: one of GTH_* port type values + * @master: bitmap of masters configured for this output + */ +struct gth_output { + struct gth_device *gth; + struct intel_th_output *output; + unsigned int index; + unsigned int port_type; + DECLARE_BITMAP(master, TH_CONFIGURABLE_MASTERS + 1); +}; + +/** + * struct gth_device - GTH device + * @dev: driver core's device + * @base: register window base address + * @output_group: attributes describing output ports + * @master_group: attributes describing master assignments + * @output: output ports + * @master: master/output port assignments + * @gth_lock: serializes accesses to GTH bits + */ +struct gth_device { + struct device *dev; + void __iomem *base; + + struct attribute_group output_group; + struct attribute_group master_group; + struct gth_output output[TH_POSSIBLE_OUTPUTS]; + signed char master[TH_CONFIGURABLE_MASTERS + 1]; + spinlock_t gth_lock; +}; + +static void gth_output_set(struct gth_device *gth, int port, + unsigned int config) +{ + unsigned long reg = port & 4 ? REG_GTH_GTHOPT1 : REG_GTH_GTHOPT0; + u32 val; + int shift = (port & 3) * 8; + + val = ioread32(gth->base + reg); + val &= ~(0xff << shift); + val |= config << shift; + iowrite32(val, gth->base + reg); +} + +static unsigned int gth_output_get(struct gth_device *gth, int port) +{ + unsigned long reg = port & 4 ? REG_GTH_GTHOPT1 : REG_GTH_GTHOPT0; + u32 val; + int shift = (port & 3) * 8; + + val = ioread32(gth->base + reg); + val &= 0xff << shift; + val >>= shift; + + return val; +} + +static void gth_smcfreq_set(struct gth_device *gth, int port, + unsigned int freq) +{ + unsigned long reg = REG_GTH_SMCR0 + ((port / 2) * 4); + int shift = (port & 1) * 16; + u32 val; + + val = ioread32(gth->base + reg); + val &= ~(0xffff << shift); + val |= freq << shift; + iowrite32(val, gth->base + reg); +} + +static unsigned int gth_smcfreq_get(struct gth_device *gth, int port) +{ + unsigned long reg = REG_GTH_SMCR0 + ((port / 2) * 4); + int shift = (port & 1) * 16; + u32 val; + + val = ioread32(gth->base + reg); + val &= 0xffff << shift; + val >>= shift; + + return val; +} + +/* + * "masters" attribute group + */ + +struct master_attribute { + struct device_attribute attr; + struct gth_device *gth; + unsigned int master; +}; + +static void +gth_master_set(struct gth_device *gth, unsigned int master, int port) +{ + unsigned int reg = REG_GTH_SWDEST0 + ((master >> 1) & ~3u); + unsigned int shift = (master & 0x7) * 4; + u32 val; + + if (master >= 256) { + reg = REG_GTH_GSWTDEST; + shift = 0; + } + + val = ioread32(gth->base + reg); + val &= ~(0xf << shift); + if (port >= 0) + val |= (0x8 | port) << shift; + iowrite32(val, gth->base + reg); +} + +/*static int gth_master_get(struct gth_device *gth, unsigned int master) +{ + unsigned int reg = REG_GTH_SWDEST0 + ((master >> 1) & ~3u); + unsigned int shift = (master & 0x7) * 4; + u32 val; + + if (master >= 256) { + reg = REG_GTH_GSWTDEST; + shift = 0; + } + + val = ioread32(gth->base + reg); + val &= (0xf << shift); + val >>= shift; + + return val ? val & 0x7 : -1; + }*/ + +static ssize_t master_attr_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct master_attribute *ma = + container_of(attr, struct master_attribute, attr); + struct gth_device *gth = ma->gth; + size_t count; + int port; + + spin_lock(>h->gth_lock); + port = gth->master[ma->master]; + spin_unlock(>h->gth_lock); + + if (port >= 0) + count = snprintf(buf, PAGE_SIZE, "%x\n", port); + else + count = snprintf(buf, PAGE_SIZE, "disabled\n"); + + return count; +} + +static ssize_t master_attr_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct master_attribute *ma = + container_of(attr, struct master_attribute, attr); + struct gth_device *gth = ma->gth; + int old_port, port; + + if (kstrtoint(buf, 10, &port) < 0) + return -EINVAL; + + if (port >= TH_POSSIBLE_OUTPUTS || port < -1) + return -EINVAL; + + spin_lock(>h->gth_lock); + + /* disconnect from the previous output port, if any */ + old_port = gth->master[ma->master]; + if (old_port >= 0) { + gth->master[ma->master] = -1; + clear_bit(ma->master, gth->output[old_port].master); + if (gth->output[old_port].output->active) + gth_master_set(gth, ma->master, -1); + } + + /* connect to the new output port, if any */ + if (port >= 0) { + /* check if there's a driver for this port */ + if (!gth->output[port].output) { + count = -ENODEV; + goto unlock; + } + + set_bit(ma->master, gth->output[port].master); + + /* if the port is active, program this setting */ + if (gth->output[port].output->active) + gth_master_set(gth, ma->master, port); + } + + gth->master[ma->master] = port; + +unlock: + spin_unlock(>h->gth_lock); + + return count; +} + +struct output_attribute { + struct device_attribute attr; + struct gth_device *gth; + unsigned int port; + unsigned int parm; +}; + +#define OUTPUT_PARM(_name, _mask, _r, _w, _what) \ + [TH_OUTPUT_PARM(_name)] = { .name = __stringify(_name), \ + .get = gth_ ## _what ## _get, \ + .set = gth_ ## _what ## _set, \ + .mask = (_mask), \ + .readable = (_r), \ + .writable = (_w) } + +static const struct output_parm { + const char *name; + unsigned int (*get)(struct gth_device *gth, int port); + void (*set)(struct gth_device *gth, int port, + unsigned int val); + unsigned int mask; + unsigned int readable : 1, + writable : 1; +} output_parms[] = { + OUTPUT_PARM(port, 0x7, 1, 0, output), + OUTPUT_PARM(null, BIT(3), 1, 1, output), + OUTPUT_PARM(drop, BIT(4), 1, 1, output), + OUTPUT_PARM(reset, BIT(5), 1, 0, output), + OUTPUT_PARM(flush, BIT(7), 0, 1, output), + OUTPUT_PARM(smcfreq, 0xffff, 1, 1, smcfreq), +}; + +static void +gth_output_parm_set(struct gth_device *gth, int port, unsigned int parm, + unsigned int val) +{ + unsigned int config = output_parms[parm].get(gth, port); + unsigned int mask = output_parms[parm].mask; + unsigned int shift = __ffs(mask); + + config &= ~mask; + config |= (val << shift) & mask; + output_parms[parm].set(gth, port, config); +} + +static unsigned int +gth_output_parm_get(struct gth_device *gth, int port, unsigned int parm) +{ + unsigned int config = output_parms[parm].get(gth, port); + unsigned int mask = output_parms[parm].mask; + unsigned int shift = __ffs(mask); + + config &= mask; + config >>= shift; + return config; +} + +/* + * Reset outputs and sources + */ +static int intel_th_gth_reset(struct gth_device *gth) +{ + u32 scratchpad; + int port, i; + + scratchpad = ioread32(gth->base + REG_GTH_SCRPD0); + if (scratchpad & SCRPD_DEBUGGER_IN_USE) + return -EBUSY; + + /* output ports */ + for (port = 0; port < 8; port++) { + if (gth_output_parm_get(gth, port, TH_OUTPUT_PARM(port)) == + GTH_NONE) + continue; + + gth_output_set(gth, port, 0); + gth_smcfreq_set(gth, port, 16); + } + /* disable overrides */ + iowrite32(0, gth->base + REG_GTH_DESTOVR); + + /* masters swdest_0~31 and gswdest */ + for (i = 0; i < 33; i++) + iowrite32(0, gth->base + REG_GTH_SWDEST0 + i * 4); + + /* sources */ + iowrite32(0, gth->base + REG_GTH_SCR); + iowrite32(0xfc, gth->base + REG_GTH_SCR2); + + return 0; +} + +/* + * "outputs" attribute group + */ + +static ssize_t output_attr_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct output_attribute *oa = + container_of(attr, struct output_attribute, attr); + struct gth_device *gth = oa->gth; + size_t count; + + spin_lock(>h->gth_lock); + count = snprintf(buf, PAGE_SIZE, "%x\n", + gth_output_parm_get(gth, oa->port, oa->parm)); + spin_unlock(>h->gth_lock); + + return count; +} + +static ssize_t output_attr_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct output_attribute *oa = + container_of(attr, struct output_attribute, attr); + struct gth_device *gth = oa->gth; + unsigned int config; + + if (kstrtouint(buf, 16, &config) < 0) + return -EINVAL; + + spin_lock(>h->gth_lock); + gth_output_parm_set(gth, oa->port, oa->parm, config); + spin_unlock(>h->gth_lock); + + return count; +} + +static int intel_th_master_attributes(struct gth_device *gth) +{ + struct master_attribute *master_attrs; + struct attribute **attrs; + int i, nattrs = TH_CONFIGURABLE_MASTERS + 2; + + attrs = devm_kcalloc(gth->dev, nattrs, sizeof(void *), GFP_KERNEL); + if (!attrs) + return -ENOMEM; + + master_attrs = devm_kcalloc(gth->dev, nattrs, + sizeof(struct master_attribute), + GFP_KERNEL); + if (!master_attrs) + return -ENOMEM; + + for (i = 0; i < TH_CONFIGURABLE_MASTERS + 1; i++) { + char *name; + + name = devm_kasprintf(gth->dev, GFP_KERNEL, "%d%s", i, + i == TH_CONFIGURABLE_MASTERS ? "+" : ""); + if (!name) + return -ENOMEM; + + master_attrs[i].attr.attr.name = name; + master_attrs[i].attr.attr.mode = S_IRUGO | S_IWUSR; + master_attrs[i].attr.show = master_attr_show; + master_attrs[i].attr.store = master_attr_store; + + sysfs_attr_init(&master_attrs[i].attr.attr); + attrs[i] = &master_attrs[i].attr.attr; + + master_attrs[i].gth = gth; + master_attrs[i].master = i; + } + + gth->master_group.name = "masters"; + gth->master_group.attrs = attrs; + + return sysfs_create_group(>h->dev->kobj, >h->master_group); +} + +static int intel_th_output_attributes(struct gth_device *gth) +{ + struct output_attribute *out_attrs; + struct attribute **attrs; + int i, j, nouts = TH_POSSIBLE_OUTPUTS; + int nparms = ARRAY_SIZE(output_parms); + int nattrs = nouts * nparms + 1; + + attrs = devm_kcalloc(gth->dev, nattrs, sizeof(void *), GFP_KERNEL); + if (!attrs) + return -ENOMEM; + + out_attrs = devm_kcalloc(gth->dev, nattrs, + sizeof(struct output_attribute), + GFP_KERNEL); + if (!out_attrs) + return -ENOMEM; + + for (i = 0; i < nouts; i++) { + for (j = 0; j < nparms; j++) { + unsigned int idx = i * nparms + j; + char *name; + + name = devm_kasprintf(gth->dev, GFP_KERNEL, "%d_%s", i, + output_parms[j].name); + if (!name) + return -ENOMEM; + + out_attrs[idx].attr.attr.name = name; + + if (output_parms[j].readable) { + out_attrs[idx].attr.attr.mode |= S_IRUGO; + out_attrs[idx].attr.show = output_attr_show; + } + + if (output_parms[j].writable) { + out_attrs[idx].attr.attr.mode |= S_IWUSR; + out_attrs[idx].attr.store = output_attr_store; + } + + sysfs_attr_init(&out_attrs[idx].attr.attr); + attrs[idx] = &out_attrs[idx].attr.attr; + + out_attrs[idx].gth = gth; + out_attrs[idx].port = i; + out_attrs[idx].parm = j; + } + } + + gth->output_group.name = "outputs"; + gth->output_group.attrs = attrs; + + return sysfs_create_group(>h->dev->kobj, >h->output_group); +} + +/** + * intel_th_gth_disable() - enable tracing to an output device + * @thdev: GTH device + * @output: output device's descriptor + * + * This will deconfigure all masters set to output to this device, + * disable tracing using force storeEn off signal and wait for the + * "pipeline empty" bit for corresponding output port. + */ +static void intel_th_gth_disable(struct intel_th_device *thdev, + struct intel_th_output *output) +{ + struct gth_device *gth = dev_get_drvdata(&thdev->dev); + unsigned long count; + int master; + u32 reg; + + spin_lock(>h->gth_lock); + output->active = false; + + for_each_set_bit(master, gth->output[output->port].master, + TH_CONFIGURABLE_MASTERS) { + gth_master_set(gth, master, -1); + } + spin_unlock(>h->gth_lock); + + iowrite32(0, gth->base + REG_GTH_SCR); + iowrite32(0xfd, gth->base + REG_GTH_SCR2); + + /* wait on pipeline empty for the given port */ + for (reg = 0, count = GTH_PLE_WAITLOOP_DEPTH; + count && !(reg & BIT(output->port)); count--) { + reg = ioread32(gth->base + REG_GTH_STAT); + cpu_relax(); + } + + /* clear force capture done for next captures */ + iowrite32(0xfc, gth->base + REG_GTH_SCR2); + + if (!count) + dev_dbg(&thdev->dev, "timeout waiting for GTH[%d] PLE\n", + output->port); +} + +/** + * intel_th_gth_enable() - enable tracing to an output device + * @thdev: GTH device + * @output: output device's descriptor + * + * This will configure all masters set to output to this device and + * enable tracing using force storeEn signal. + */ +static void intel_th_gth_enable(struct intel_th_device *thdev, + struct intel_th_output *output) +{ + struct gth_device *gth = dev_get_drvdata(&thdev->dev); + u32 scr = 0xfc0000; + int master; + + spin_lock(>h->gth_lock); + for_each_set_bit(master, gth->output[output->port].master, + TH_CONFIGURABLE_MASTERS + 1) { + gth_master_set(gth, master, output->port); + } + + if (output->multiblock) + scr |= 0xff; + + output->active = true; + spin_unlock(>h->gth_lock); + + iowrite32(scr, gth->base + REG_GTH_SCR); + iowrite32(0, gth->base + REG_GTH_SCR2); +} + +/** + * intel_th_gth_assign() - assign output device to a GTH output port + * @thdev: GTH device + * @othdev: output device + * + * This will match a given output device parameters against present + * output ports on the GTH and fill out relevant bits in output device's + * descriptor. + * + * Return: 0 on success, -errno on error. + */ +static int intel_th_gth_assign(struct intel_th_device *thdev, + struct intel_th_device *othdev) +{ + struct gth_device *gth = dev_get_drvdata(&thdev->dev); + int i, id; + + if (othdev->type != INTEL_TH_OUTPUT) + return -EINVAL; + + for (i = 0, id = 0; i < TH_POSSIBLE_OUTPUTS; i++) { + if (gth->output[i].port_type != othdev->output.type) + continue; + + if (othdev->id == -1 || othdev->id == id) + goto found; + + id++; + } + + return -ENOENT; + +found: + spin_lock(>h->gth_lock); + othdev->output.port = i; + othdev->output.active = false; + gth->output[i].output = &othdev->output; + spin_unlock(>h->gth_lock); + + return 0; +} + +/** + * intel_th_gth_unassign() - deassociate an output device from its output port + * @thdev: GTH device + * @othdev: output device + */ +static void intel_th_gth_unassign(struct intel_th_device *thdev, + struct intel_th_device *othdev) +{ + struct gth_device *gth = dev_get_drvdata(&thdev->dev); + int port = othdev->output.port; + + spin_lock(>h->gth_lock); + othdev->output.port = -1; + othdev->output.active = false; + gth->output[port].output = NULL; + spin_unlock(>h->gth_lock); +} + +static int +intel_th_gth_set_output(struct intel_th_device *thdev, unsigned int master) +{ + struct gth_device *gth = dev_get_drvdata(&thdev->dev); + int port = 0; /* FIXME: make default output configurable */ + + /* + * everything above TH_CONFIGURABLE_MASTERS is controlled by the + * same register + */ + if (master > TH_CONFIGURABLE_MASTERS) + master = TH_CONFIGURABLE_MASTERS; + + spin_lock(>h->gth_lock); + if (gth->master[master] == -1) { + set_bit(master, gth->output[port].master); + gth->master[master] = port; + } + spin_unlock(>h->gth_lock); + + return 0; +} + +static int intel_th_gth_probe(struct intel_th_device *thdev) +{ + struct device *dev = &thdev->dev; + struct gth_device *gth; + struct resource *res; + void __iomem *base; + int i, ret; + + res = intel_th_device_get_resource(thdev, IORESOURCE_MEM, 0); + if (!res) + return -ENODEV; + + base = devm_ioremap(dev, res->start, resource_size(res)); + if (IS_ERR(base)) + return PTR_ERR(base); + + gth = devm_kzalloc(dev, sizeof(*gth), GFP_KERNEL); + if (!gth) + return -ENOMEM; + + gth->dev = dev; + gth->base = base; + spin_lock_init(>h->gth_lock); + + ret = intel_th_gth_reset(gth); + if (ret) + return ret; + + for (i = 0; i < TH_CONFIGURABLE_MASTERS + 1; i++) + gth->master[i] = -1; + + for (i = 0; i < TH_POSSIBLE_OUTPUTS; i++) { + gth->output[i].gth = gth; + gth->output[i].index = i; + gth->output[i].port_type = + gth_output_parm_get(gth, i, TH_OUTPUT_PARM(port)); + } + + if (intel_th_output_attributes(gth) || + intel_th_master_attributes(gth)) { + pr_warn("Can't initialize sysfs attributes\n"); + + if (gth->output_group.attrs) + sysfs_remove_group(>h->dev->kobj, >h->output_group); + return -ENOMEM; + } + + dev_set_drvdata(dev, gth); + + return 0; +} + +static void intel_th_gth_remove(struct intel_th_device *thdev) +{ + struct gth_device *gth = dev_get_drvdata(&thdev->dev); + + sysfs_remove_group(>h->dev->kobj, >h->output_group); + sysfs_remove_group(>h->dev->kobj, >h->master_group); +} + +static struct intel_th_driver intel_th_gth_driver = { + .probe = intel_th_gth_probe, + .remove = intel_th_gth_remove, + .assign = intel_th_gth_assign, + .unassign = intel_th_gth_unassign, + .set_output = intel_th_gth_set_output, + .enable = intel_th_gth_enable, + .disable = intel_th_gth_disable, + .driver = { + .name = "gth", + .owner = THIS_MODULE, + }, +}; + +module_driver(intel_th_gth_driver, + intel_th_driver_register, + intel_th_driver_unregister); + +MODULE_ALIAS("intel_th_switch"); +MODULE_LICENSE("GPL v2"); +MODULE_DESCRIPTION("Intel(R) Trace Hub Global Trace Hub driver"); +MODULE_AUTHOR("Alexander Shishkin <alexander.shishkin@linux.intel.com>"); diff --git a/drivers/hwtracing/intel_th/gth.h b/drivers/hwtracing/intel_th/gth.h new file mode 100644 index 000000000000..3b714b7a61db --- /dev/null +++ b/drivers/hwtracing/intel_th/gth.h @@ -0,0 +1,66 @@ +/* + * Intel(R) Trace Hub Global Trace Hub (GTH) data structures + * + * Copyright (C) 2014-2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ + +#ifndef __INTEL_TH_GTH_H__ +#define __INTEL_TH_GTH_H__ + +/* Map output port parameter bits to symbolic names */ +#define TH_OUTPUT_PARM(name) \ + TH_OUTPUT_ ## name + +enum intel_th_output_parm { + /* output port type */ + TH_OUTPUT_PARM(port), + /* generate NULL packet */ + TH_OUTPUT_PARM(null), + /* packet drop */ + TH_OUTPUT_PARM(drop), + /* port in reset state */ + TH_OUTPUT_PARM(reset), + /* flush out data */ + TH_OUTPUT_PARM(flush), + /* mainenance packet frequency */ + TH_OUTPUT_PARM(smcfreq), +}; + +/* + * Register offsets + */ +enum { + REG_GTH_GTHOPT0 = 0x00, /* Output ports 0..3 config */ + REG_GTH_GTHOPT1 = 0x04, /* Output ports 4..7 config */ + REG_GTH_SWDEST0 = 0x08, /* Switching destination masters 0..7 */ + REG_GTH_GSWTDEST = 0x88, /* Global sw trace destination */ + REG_GTH_SMCR0 = 0x9c, /* STP mainenance for ports 0/1 */ + REG_GTH_SMCR1 = 0xa0, /* STP mainenance for ports 2/3 */ + REG_GTH_SMCR2 = 0xa4, /* STP mainenance for ports 4/5 */ + REG_GTH_SMCR3 = 0xa8, /* STP mainenance for ports 6/7 */ + REG_GTH_SCR = 0xc8, /* Source control (storeEn override) */ + REG_GTH_STAT = 0xd4, /* GTH status */ + REG_GTH_SCR2 = 0xd8, /* Source control (force storeEn off) */ + REG_GTH_DESTOVR = 0xdc, /* Destination override */ + REG_GTH_SCRPD0 = 0xe0, /* ScratchPad[0] */ + REG_GTH_SCRPD1 = 0xe4, /* ScratchPad[1] */ + REG_GTH_SCRPD2 = 0xe8, /* ScratchPad[2] */ + REG_GTH_SCRPD3 = 0xec, /* ScratchPad[3] */ +}; + +/* Externall debugger is using Intel TH */ +#define SCRPD_DEBUGGER_IN_USE BIT(24) + +/* waiting for Pipeline Empty bit(s) to assert for GTH */ +#define GTH_PLE_WAITLOOP_DEPTH 10000 + +#endif /* __INTEL_TH_GTH_H__ */ diff --git a/drivers/hwtracing/intel_th/intel_th.h b/drivers/hwtracing/intel_th/intel_th.h new file mode 100644 index 000000000000..57fd72b20fae --- /dev/null +++ b/drivers/hwtracing/intel_th/intel_th.h @@ -0,0 +1,244 @@ +/* + * Intel(R) Trace Hub data structures + * + * Copyright (C) 2014-2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ + +#ifndef __INTEL_TH_H__ +#define __INTEL_TH_H__ + +/* intel_th_device device types */ +enum { + /* Devices that generate trace data */ + INTEL_TH_SOURCE = 0, + /* Output ports (MSC, PTI) */ + INTEL_TH_OUTPUT, + /* Switch, the Global Trace Hub (GTH) */ + INTEL_TH_SWITCH, +}; + +/** + * struct intel_th_output - descriptor INTEL_TH_OUTPUT type devices + * @port: output port number, assigned by the switch + * @type: GTH_{MSU,CTP,PTI} + * @multiblock: true for multiblock output configuration + * @active: true when this output is enabled + * + * Output port descriptor, used by switch driver to tell which output + * port this output device corresponds to. Filled in at output device's + * probe time by switch::assign(). Passed from output device driver to + * switch related code to enable/disable its port. + */ +struct intel_th_output { + int port; + unsigned int type; + bool multiblock; + bool active; +}; + +/** + * struct intel_th_device - device on the intel_th bus + * @dev: device + * @resource: array of resources available to this device + * @num_resources: number of resources in @resource array + * @type: INTEL_TH_{SOURCE,OUTPUT,SWITCH} + * @id: device instance or -1 + * @output: output descriptor for INTEL_TH_OUTPUT devices + * @name: device name to match the driver + */ +struct intel_th_device { + struct device dev; + struct resource *resource; + unsigned int num_resources; + unsigned int type; + int id; + + /* INTEL_TH_OUTPUT specific */ + struct intel_th_output output; + + char name[]; +}; + +#define to_intel_th_device(_d) \ + container_of((_d), struct intel_th_device, dev) + +/** + * intel_th_device_get_resource() - obtain @num'th resource of type @type + * @thdev: the device to search the resource for + * @type: resource type + * @num: number of the resource + */ +static inline struct resource * +intel_th_device_get_resource(struct intel_th_device *thdev, unsigned int type, + unsigned int num) +{ + int i; + + for (i = 0; i < thdev->num_resources; i++) + if (resource_type(&thdev->resource[i]) == type && !num--) + return &thdev->resource[i]; + + return NULL; +} + +/** + * intel_th_output_assigned() - if an output device is assigned to a switch port + * @thdev: the output device + * + * Return: true if the device is INTEL_TH_OUTPUT *and* is assigned a port + */ +static inline bool +intel_th_output_assigned(struct intel_th_device *thdev) +{ + return thdev->type == INTEL_TH_OUTPUT && + thdev->output.port >= 0; +} + +/** + * struct intel_th_driver - driver for an intel_th_device device + * @driver: generic driver + * @probe: probe method + * @remove: remove method + * @assign: match a given output type device against available outputs + * @unassign: deassociate an output type device from an output port + * @enable: enable tracing for a given output device + * @disable: disable tracing for a given output device + * @fops: file operations for device nodes + * + * Callbacks @probe and @remove are required for all device types. + * Switch device driver needs to fill in @assign, @enable and @disable + * callbacks. + */ +struct intel_th_driver { + struct device_driver driver; + int (*probe)(struct intel_th_device *thdev); + void (*remove)(struct intel_th_device *thdev); + /* switch (GTH) ops */ + int (*assign)(struct intel_th_device *thdev, + struct intel_th_device *othdev); + void (*unassign)(struct intel_th_device *thdev, + struct intel_th_device *othdev); + void (*enable)(struct intel_th_device *thdev, + struct intel_th_output *output); + void (*disable)(struct intel_th_device *thdev, + struct intel_th_output *output); + /* output ops */ + void (*irq)(struct intel_th_device *thdev); + int (*activate)(struct intel_th_device *thdev); + void (*deactivate)(struct intel_th_device *thdev); + /* file_operations for those who want a device node */ + const struct file_operations *fops; + + /* source ops */ + int (*set_output)(struct intel_th_device *thdev, + unsigned int master); +}; + +#define to_intel_th_driver(_d) \ + container_of((_d), struct intel_th_driver, driver) + +static inline struct intel_th_device * +to_intel_th_hub(struct intel_th_device *thdev) +{ + struct device *parent = thdev->dev.parent; + + if (!parent) + return NULL; + + return to_intel_th_device(parent); +} + +struct intel_th * +intel_th_alloc(struct device *dev, struct resource *devres, + unsigned int ndevres, int irq); +void intel_th_free(struct intel_th *th); + +int intel_th_driver_register(struct intel_th_driver *thdrv); +void intel_th_driver_unregister(struct intel_th_driver *thdrv); + +int intel_th_trace_enable(struct intel_th_device *thdev); +int intel_th_trace_disable(struct intel_th_device *thdev); +int intel_th_set_output(struct intel_th_device *thdev, + unsigned int master); + +enum { + TH_MMIO_CONFIG = 0, + TH_MMIO_SW = 2, + TH_MMIO_END, +}; + +#define TH_SUBDEVICE_MAX 6 +#define TH_POSSIBLE_OUTPUTS 8 +#define TH_CONFIGURABLE_MASTERS 256 +#define TH_MSC_MAX 2 + +/** + * struct intel_th - Intel TH controller + * @dev: driver core's device + * @thdev: subdevices + * @hub: "switch" subdevice (GTH) + * @id: this Intel TH controller's device ID in the system + * @major: device node major for output devices + */ +struct intel_th { + struct device *dev; + + struct intel_th_device *thdev[TH_SUBDEVICE_MAX]; + struct intel_th_device *hub; + + int id; + int major; +#ifdef CONFIG_INTEL_TH_DEBUG + struct dentry *dbg; +#endif +}; + +/* + * Register windows + */ +enum { + /* Global Trace Hub (GTH) */ + REG_GTH_OFFSET = 0x0000, + REG_GTH_LENGTH = 0x2000, + + /* Software Trace Hub (STH) [0x4000..0x4fff] */ + REG_STH_OFFSET = 0x4000, + REG_STH_LENGTH = 0x2000, + + /* Memory Storage Unit (MSU) [0xa0000..0xa1fff] */ + REG_MSU_OFFSET = 0xa0000, + REG_MSU_LENGTH = 0x02000, + + /* Internal MSU trace buffer [0x80000..0x9ffff] */ + BUF_MSU_OFFSET = 0x80000, + BUF_MSU_LENGTH = 0x20000, + + /* PTI output == same window as GTH */ + REG_PTI_OFFSET = REG_GTH_OFFSET, + REG_PTI_LENGTH = REG_GTH_LENGTH, + + /* DCI Handler (DCIH) == some window as MSU */ + REG_DCIH_OFFSET = REG_MSU_OFFSET, + REG_DCIH_LENGTH = REG_MSU_LENGTH, +}; + +/* + * GTH, output ports configuration + */ +enum { + GTH_NONE = 0, + GTH_MSU, /* memory/usb */ + GTH_CTP, /* Common Trace Port */ + GTH_PTI = 4, /* MIPI-PTI */ +}; + +#endif diff --git a/drivers/hwtracing/intel_th/msu.c b/drivers/hwtracing/intel_th/msu.c new file mode 100644 index 000000000000..80a12384ed20 --- /dev/null +++ b/drivers/hwtracing/intel_th/msu.c @@ -0,0 +1,1509 @@ +/* + * Intel(R) Trace Hub Memory Storage Unit + * + * Copyright (C) 2014-2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/types.h> +#include <linux/module.h> +#include <linux/device.h> +#include <linux/uaccess.h> +#include <linux/sizes.h> +#include <linux/printk.h> +#include <linux/slab.h> +#include <linux/mm.h> +#include <linux/fs.h> +#include <linux/io.h> +#include <linux/dma-mapping.h> + +#include <asm/cacheflush.h> + +#include "intel_th.h" +#include "msu.h" + +#define msc_dev(x) (&(x)->thdev->dev) + +/** + * struct msc_block - multiblock mode block descriptor + * @bdesc: pointer to hardware descriptor (beginning of the block) + * @addr: physical address of the block + */ +struct msc_block { + struct msc_block_desc *bdesc; + dma_addr_t addr; +}; + +/** + * struct msc_window - multiblock mode window descriptor + * @entry: window list linkage (msc::win_list) + * @pgoff: page offset into the buffer that this window starts at + * @nr_blocks: number of blocks (pages) in this window + * @block: array of block descriptors + */ +struct msc_window { + struct list_head entry; + unsigned long pgoff; + unsigned int nr_blocks; + struct msc *msc; + struct msc_block block[0]; +}; + +/** + * struct msc_iter - iterator for msc buffer + * @entry: msc::iter_list linkage + * @msc: pointer to the MSC device + * @start_win: oldest window + * @win: current window + * @offset: current logical offset into the buffer + * @start_block: oldest block in the window + * @block: block number in the window + * @block_off: offset into current block + * @wrap_count: block wrapping handling + * @eof: end of buffer reached + */ +struct msc_iter { + struct list_head entry; + struct msc *msc; + struct msc_window *start_win; + struct msc_window *win; + unsigned long offset; + int start_block; + int block; + unsigned int block_off; + unsigned int wrap_count; + unsigned int eof; +}; + +/** + * struct msc - MSC device representation + * @reg_base: register window base address + * @thdev: intel_th_device pointer + * @win_list: list of windows in multiblock mode + * @nr_pages: total number of pages allocated for this buffer + * @single_sz: amount of data in single mode + * @single_wrap: single mode wrap occurred + * @base: buffer's base pointer + * @base_addr: buffer's base address + * @user_count: number of users of the buffer + * @mmap_count: number of mappings + * @buf_mutex: mutex to serialize access to buffer-related bits + + * @enabled: MSC is enabled + * @wrap: wrapping is enabled + * @mode: MSC operating mode + * @burst_len: write burst length + * @index: number of this MSC in the MSU + */ +struct msc { + void __iomem *reg_base; + struct intel_th_device *thdev; + + struct list_head win_list; + unsigned long nr_pages; + unsigned long single_sz; + unsigned int single_wrap : 1; + void *base; + dma_addr_t base_addr; + + /* <0: no buffer, 0: no users, >0: active users */ + atomic_t user_count; + + atomic_t mmap_count; + struct mutex buf_mutex; + + struct mutex iter_mutex; + struct list_head iter_list; + + /* config */ + unsigned int enabled : 1, + wrap : 1; + unsigned int mode; + unsigned int burst_len; + unsigned int index; +}; + +static inline bool msc_block_is_empty(struct msc_block_desc *bdesc) +{ + /* header hasn't been written */ + if (!bdesc->valid_dw) + return true; + + /* valid_dw includes the header */ + if (!msc_data_sz(bdesc)) + return true; + + return false; +} + +/** + * msc_oldest_window() - locate the window with oldest data + * @msc: MSC device + * + * This should only be used in multiblock mode. Caller should hold the + * msc::user_count reference. + * + * Return: the oldest window with valid data + */ +static struct msc_window *msc_oldest_window(struct msc *msc) +{ + struct msc_window *win; + u32 reg = ioread32(msc->reg_base + REG_MSU_MSC0NWSA); + unsigned long win_addr = (unsigned long)reg << PAGE_SHIFT; + unsigned int found = 0; + + if (list_empty(&msc->win_list)) + return NULL; + + /* + * we might need a radix tree for this, depending on how + * many windows a typical user would allocate; ideally it's + * something like 2, in which case we're good + */ + list_for_each_entry(win, &msc->win_list, entry) { + if (win->block[0].addr == win_addr) + found++; + + /* skip the empty ones */ + if (msc_block_is_empty(win->block[0].bdesc)) + continue; + + if (found) + return win; + } + + return list_entry(msc->win_list.next, struct msc_window, entry); +} + +/** + * msc_win_oldest_block() - locate the oldest block in a given window + * @win: window to look at + * + * Return: index of the block with the oldest data + */ +static unsigned int msc_win_oldest_block(struct msc_window *win) +{ + unsigned int blk; + struct msc_block_desc *bdesc = win->block[0].bdesc; + + /* without wrapping, first block is the oldest */ + if (!msc_block_wrapped(bdesc)) + return 0; + + /* + * with wrapping, last written block contains both the newest and the + * oldest data for this window. + */ + for (blk = 0; blk < win->nr_blocks; blk++) { + bdesc = win->block[blk].bdesc; + + if (msc_block_last_written(bdesc)) + return blk; + } + + return 0; +} + +/** + * msc_is_last_win() - check if a window is the last one for a given MSC + * @win: window + * Return: true if @win is the last window in MSC's multiblock buffer + */ +static inline bool msc_is_last_win(struct msc_window *win) +{ + return win->entry.next == &win->msc->win_list; +} + +/** + * msc_next_window() - return next window in the multiblock buffer + * @win: current window + * + * Return: window following the current one + */ +static struct msc_window *msc_next_window(struct msc_window *win) +{ + if (msc_is_last_win(win)) + return list_entry(win->msc->win_list.next, struct msc_window, + entry); + + return list_entry(win->entry.next, struct msc_window, entry); +} + +static struct msc_block_desc *msc_iter_bdesc(struct msc_iter *iter) +{ + return iter->win->block[iter->block].bdesc; +} + +static void msc_iter_init(struct msc_iter *iter) +{ + memset(iter, 0, sizeof(*iter)); + iter->start_block = -1; + iter->block = -1; +} + +static struct msc_iter *msc_iter_install(struct msc *msc) +{ + struct msc_iter *iter; + + iter = kzalloc(sizeof(*iter), GFP_KERNEL); + if (!iter) + return NULL; + + msc_iter_init(iter); + iter->msc = msc; + + mutex_lock(&msc->iter_mutex); + list_add_tail(&iter->entry, &msc->iter_list); + mutex_unlock(&msc->iter_mutex); + + return iter; +} + +static void msc_iter_remove(struct msc_iter *iter, struct msc *msc) +{ + mutex_lock(&msc->iter_mutex); + list_del(&iter->entry); + mutex_unlock(&msc->iter_mutex); + + kfree(iter); +} + +static void msc_iter_block_start(struct msc_iter *iter) +{ + if (iter->start_block != -1) + return; + + iter->start_block = msc_win_oldest_block(iter->win); + iter->block = iter->start_block; + iter->wrap_count = 0; + + /* + * start with the block with oldest data; if data has wrapped + * in this window, it should be in this block + */ + if (msc_block_wrapped(msc_iter_bdesc(iter))) + iter->wrap_count = 2; + +} + +static int msc_iter_win_start(struct msc_iter *iter, struct msc *msc) +{ + /* already started, nothing to do */ + if (iter->start_win) + return 0; + + iter->start_win = msc_oldest_window(msc); + if (!iter->start_win) + return -EINVAL; + + iter->win = iter->start_win; + iter->start_block = -1; + + msc_iter_block_start(iter); + + return 0; +} + +static int msc_iter_win_advance(struct msc_iter *iter) +{ + iter->win = msc_next_window(iter->win); + iter->start_block = -1; + + if (iter->win == iter->start_win) { + iter->eof++; + return 1; + } + + msc_iter_block_start(iter); + + return 0; +} + +static int msc_iter_block_advance(struct msc_iter *iter) +{ + iter->block_off = 0; + + /* wrapping */ + if (iter->wrap_count && iter->block == iter->start_block) { + iter->wrap_count--; + if (!iter->wrap_count) + /* copied newest data from the wrapped block */ + return msc_iter_win_advance(iter); + } + + /* no wrapping, check for last written block */ + if (!iter->wrap_count && msc_block_last_written(msc_iter_bdesc(iter))) + /* copied newest data for the window */ + return msc_iter_win_advance(iter); + + /* block advance */ + if (++iter->block == iter->win->nr_blocks) + iter->block = 0; + + /* no wrapping, sanity check in case there is no last written block */ + if (!iter->wrap_count && iter->block == iter->start_block) + return msc_iter_win_advance(iter); + + return 0; +} + +/** + * msc_buffer_iterate() - go through multiblock buffer's data + * @iter: iterator structure + * @size: amount of data to scan + * @data: callback's private data + * @fn: iterator callback + * + * This will start at the window which will be written to next (containing + * the oldest data) and work its way to the current window, calling @fn + * for each chunk of data as it goes. + * + * Caller should have msc::user_count reference to make sure the buffer + * doesn't disappear from under us. + * + * Return: amount of data actually scanned. + */ +static ssize_t +msc_buffer_iterate(struct msc_iter *iter, size_t size, void *data, + unsigned long (*fn)(void *, void *, size_t)) +{ + struct msc *msc = iter->msc; + size_t len = size; + unsigned int advance; + + if (iter->eof) + return 0; + + /* start with the oldest window */ + if (msc_iter_win_start(iter, msc)) + return 0; + + do { + unsigned long data_bytes = msc_data_sz(msc_iter_bdesc(iter)); + void *src = (void *)msc_iter_bdesc(iter) + MSC_BDESC; + size_t tocopy = data_bytes, copied = 0; + size_t remaining = 0; + + advance = 1; + + /* + * If block wrapping happened, we need to visit the last block + * twice, because it contains both the oldest and the newest + * data in this window. + * + * First time (wrap_count==2), in the very beginning, to collect + * the oldest data, which is in the range + * (data_bytes..DATA_IN_PAGE). + * + * Second time (wrap_count==1), it's just like any other block, + * containing data in the range of [MSC_BDESC..data_bytes]. + */ + if (iter->block == iter->start_block && iter->wrap_count) { + tocopy = DATA_IN_PAGE - data_bytes; + src += data_bytes; + } + + if (!tocopy) + goto next_block; + + tocopy -= iter->block_off; + src += iter->block_off; + + if (len < tocopy) { + tocopy = len; + advance = 0; + } + + remaining = fn(data, src, tocopy); + + if (remaining) + advance = 0; + + copied = tocopy - remaining; + len -= copied; + iter->block_off += copied; + iter->offset += copied; + + if (!advance) + break; + +next_block: + if (msc_iter_block_advance(iter)) + break; + + } while (len); + + return size - len; +} + +/** + * msc_buffer_clear_hw_header() - clear hw header for multiblock + * @msc: MSC device + */ +static void msc_buffer_clear_hw_header(struct msc *msc) +{ + struct msc_window *win; + + mutex_lock(&msc->buf_mutex); + list_for_each_entry(win, &msc->win_list, entry) { + unsigned int blk; + size_t hw_sz = sizeof(struct msc_block_desc) - + offsetof(struct msc_block_desc, hw_tag); + + for (blk = 0; blk < win->nr_blocks; blk++) { + struct msc_block_desc *bdesc = win->block[blk].bdesc; + + memset(&bdesc->hw_tag, 0, hw_sz); + } + } + mutex_unlock(&msc->buf_mutex); +} + +/** + * msc_configure() - set up MSC hardware + * @msc: the MSC device to configure + * + * Program storage mode, wrapping, burst length and trace buffer address + * into a given MSC. If msc::enabled is set, enable the trace, too. + */ +static int msc_configure(struct msc *msc) +{ + u32 reg; + + if (msc->mode > MSC_MODE_MULTI) + return -ENOTSUPP; + + if (msc->mode == MSC_MODE_MULTI) + msc_buffer_clear_hw_header(msc); + + reg = msc->base_addr >> PAGE_SHIFT; + iowrite32(reg, msc->reg_base + REG_MSU_MSC0BAR); + + if (msc->mode == MSC_MODE_SINGLE) { + reg = msc->nr_pages; + iowrite32(reg, msc->reg_base + REG_MSU_MSC0SIZE); + } + + reg = ioread32(msc->reg_base + REG_MSU_MSC0CTL); + reg &= ~(MSC_MODE | MSC_WRAPEN | MSC_EN | MSC_RD_HDR_OVRD); + + reg |= msc->mode << __ffs(MSC_MODE); + reg |= msc->burst_len << __ffs(MSC_LEN); + /*if (msc->mode == MSC_MODE_MULTI) + reg |= MSC_RD_HDR_OVRD; */ + if (msc->wrap) + reg |= MSC_WRAPEN; + if (msc->enabled) + reg |= MSC_EN; + + iowrite32(reg, msc->reg_base + REG_MSU_MSC0CTL); + + if (msc->enabled) { + msc->thdev->output.multiblock = msc->mode == MSC_MODE_MULTI; + intel_th_trace_enable(msc->thdev); + } + + return 0; +} + +/** + * msc_disable() - disable MSC hardware + * @msc: MSC device to disable + * + * If @msc is enabled, disable tracing on the switch and then disable MSC + * storage. + */ +static void msc_disable(struct msc *msc) +{ + unsigned long count; + u32 reg; + + if (!msc->enabled) + return; + + intel_th_trace_disable(msc->thdev); + + for (reg = 0, count = MSC_PLE_WAITLOOP_DEPTH; + count && !(reg & MSCSTS_PLE); count--) { + reg = ioread32(msc->reg_base + REG_MSU_MSC0STS); + cpu_relax(); + } + + if (!count) + dev_dbg(msc_dev(msc), "timeout waiting for MSC0 PLE\n"); + + if (msc->mode == MSC_MODE_SINGLE) { + msc->single_wrap = !!(reg & MSCSTS_WRAPSTAT); + + reg = ioread32(msc->reg_base + REG_MSU_MSC0MWP); + msc->single_sz = reg & ((msc->nr_pages << PAGE_SHIFT) - 1); + dev_dbg(msc_dev(msc), "MSCnMWP: %08x/%08lx, wrap: %d\n", + reg, msc->single_sz, msc->single_wrap); + } + + reg = ioread32(msc->reg_base + REG_MSU_MSC0CTL); + reg &= ~MSC_EN; + iowrite32(reg, msc->reg_base + REG_MSU_MSC0CTL); + msc->enabled = 0; + + iowrite32(0, msc->reg_base + REG_MSU_MSC0BAR); + iowrite32(0, msc->reg_base + REG_MSU_MSC0SIZE); + + dev_dbg(msc_dev(msc), "MSCnNWSA: %08x\n", + ioread32(msc->reg_base + REG_MSU_MSC0NWSA)); + + reg = ioread32(msc->reg_base + REG_MSU_MSC0STS); + dev_dbg(msc_dev(msc), "MSCnSTS: %08x\n", reg); +} + +static int intel_th_msc_activate(struct intel_th_device *thdev) +{ + struct msc *msc = dev_get_drvdata(&thdev->dev); + int ret = 0; + + if (!atomic_inc_unless_negative(&msc->user_count)) + return -ENODEV; + + mutex_lock(&msc->iter_mutex); + if (!list_empty(&msc->iter_list)) + ret = -EBUSY; + mutex_unlock(&msc->iter_mutex); + + if (ret) { + atomic_dec(&msc->user_count); + return ret; + } + + msc->enabled = 1; + + return msc_configure(msc); +} + +static void intel_th_msc_deactivate(struct intel_th_device *thdev) +{ + struct msc *msc = dev_get_drvdata(&thdev->dev); + + msc_disable(msc); + + atomic_dec(&msc->user_count); +} + +/** + * msc_buffer_contig_alloc() - allocate a contiguous buffer for SINGLE mode + * @msc: MSC device + * @size: allocation size in bytes + * + * This modifies msc::base, which requires msc::buf_mutex to serialize, so the + * caller is expected to hold it. + * + * Return: 0 on success, -errno otherwise. + */ +static int msc_buffer_contig_alloc(struct msc *msc, unsigned long size) +{ + unsigned int order = get_order(size); + struct page *page; + + if (!size) + return 0; + + page = alloc_pages(GFP_KERNEL | __GFP_ZERO, order); + if (!page) + return -ENOMEM; + + split_page(page, order); + msc->nr_pages = size >> PAGE_SHIFT; + msc->base = page_address(page); + msc->base_addr = page_to_phys(page); + + return 0; +} + +/** + * msc_buffer_contig_free() - free a contiguous buffer + * @msc: MSC configured in SINGLE mode + */ +static void msc_buffer_contig_free(struct msc *msc) +{ + unsigned long off; + + for (off = 0; off < msc->nr_pages << PAGE_SHIFT; off += PAGE_SIZE) { + struct page *page = virt_to_page(msc->base + off); + + page->mapping = NULL; + __free_page(page); + } + + msc->nr_pages = 0; +} + +/** + * msc_buffer_contig_get_page() - find a page at a given offset + * @msc: MSC configured in SINGLE mode + * @pgoff: page offset + * + * Return: page, if @pgoff is within the range, NULL otherwise. + */ +static struct page *msc_buffer_contig_get_page(struct msc *msc, + unsigned long pgoff) +{ + if (pgoff >= msc->nr_pages) + return NULL; + + return virt_to_page(msc->base + (pgoff << PAGE_SHIFT)); +} + +/** + * msc_buffer_win_alloc() - alloc a window for a multiblock mode + * @msc: MSC device + * @nr_blocks: number of pages in this window + * + * This modifies msc::win_list and msc::base, which requires msc::buf_mutex + * to serialize, so the caller is expected to hold it. + * + * Return: 0 on success, -errno otherwise. + */ +static int msc_buffer_win_alloc(struct msc *msc, unsigned int nr_blocks) +{ + struct msc_window *win; + unsigned long size = PAGE_SIZE; + int i, ret = -ENOMEM; + + if (!nr_blocks) + return 0; + + win = kzalloc(offsetof(struct msc_window, block[nr_blocks]), + GFP_KERNEL); + if (!win) + return -ENOMEM; + + if (!list_empty(&msc->win_list)) { + struct msc_window *prev = list_entry(msc->win_list.prev, + struct msc_window, entry); + + win->pgoff = prev->pgoff + prev->nr_blocks; + } + + for (i = 0; i < nr_blocks; i++) { + win->block[i].bdesc = dma_alloc_coherent(msc_dev(msc), size, + &win->block[i].addr, + GFP_KERNEL); + +#ifdef CONFIG_X86 + /* Set the page as uncached */ + set_memory_uc((unsigned long)win->block[i].bdesc, 1); +#endif + + if (!win->block[i].bdesc) + goto err_nomem; + } + + win->msc = msc; + win->nr_blocks = nr_blocks; + + if (list_empty(&msc->win_list)) { + msc->base = win->block[0].bdesc; + msc->base_addr = win->block[0].addr; + } + + list_add_tail(&win->entry, &msc->win_list); + msc->nr_pages += nr_blocks; + + return 0; + +err_nomem: + for (i--; i >= 0; i--) { +#ifdef CONFIG_X86 + /* Reset the page to write-back before releasing */ + set_memory_wb((unsigned long)win->block[i].bdesc, 1); +#endif + dma_free_coherent(msc_dev(msc), size, win->block[i].bdesc, + win->block[i].addr); + } + kfree(win); + + return ret; +} + +/** + * msc_buffer_win_free() - free a window from MSC's window list + * @msc: MSC device + * @win: window to free + * + * This modifies msc::win_list and msc::base, which requires msc::buf_mutex + * to serialize, so the caller is expected to hold it. + */ +static void msc_buffer_win_free(struct msc *msc, struct msc_window *win) +{ + int i; + + msc->nr_pages -= win->nr_blocks; + + list_del(&win->entry); + if (list_empty(&msc->win_list)) { + msc->base = NULL; + msc->base_addr = 0; + } + + for (i = 0; i < win->nr_blocks; i++) { + struct page *page = virt_to_page(win->block[i].bdesc); + + page->mapping = NULL; +#ifdef CONFIG_X86 + /* Reset the page to write-back before releasing */ + set_memory_wb((unsigned long)win->block[i].bdesc, 1); +#endif + dma_free_coherent(msc_dev(win->msc), PAGE_SIZE, + win->block[i].bdesc, win->block[i].addr); + } + + kfree(win); +} + +/** + * msc_buffer_relink() - set up block descriptors for multiblock mode + * @msc: MSC device + * + * This traverses msc::win_list, which requires msc::buf_mutex to serialize, + * so the caller is expected to hold it. + */ +static void msc_buffer_relink(struct msc *msc) +{ + struct msc_window *win, *next_win; + + /* call with msc::mutex locked */ + list_for_each_entry(win, &msc->win_list, entry) { + unsigned int blk; + u32 sw_tag = 0; + + /* + * Last window's next_win should point to the first window + * and MSC_SW_TAG_LASTWIN should be set. + */ + if (msc_is_last_win(win)) { + sw_tag |= MSC_SW_TAG_LASTWIN; + next_win = list_entry(msc->win_list.next, + struct msc_window, entry); + } else { + next_win = list_entry(win->entry.next, + struct msc_window, entry); + } + + for (blk = 0; blk < win->nr_blocks; blk++) { + struct msc_block_desc *bdesc = win->block[blk].bdesc; + + memset(bdesc, 0, sizeof(*bdesc)); + + bdesc->next_win = next_win->block[0].addr >> PAGE_SHIFT; + + /* + * Similarly to last window, last block should point + * to the first one. + */ + if (blk == win->nr_blocks - 1) { + sw_tag |= MSC_SW_TAG_LASTBLK; + bdesc->next_blk = + win->block[0].addr >> PAGE_SHIFT; + } else { + bdesc->next_blk = + win->block[blk + 1].addr >> PAGE_SHIFT; + } + + bdesc->sw_tag = sw_tag; + bdesc->block_sz = PAGE_SIZE / 64; + } + } + + /* + * Make the above writes globally visible before tracing is + * enabled to make sure hardware sees them coherently. + */ + wmb(); +} + +static void msc_buffer_multi_free(struct msc *msc) +{ + struct msc_window *win, *iter; + + list_for_each_entry_safe(win, iter, &msc->win_list, entry) + msc_buffer_win_free(msc, win); +} + +static int msc_buffer_multi_alloc(struct msc *msc, unsigned long *nr_pages, + unsigned int nr_wins) +{ + int ret, i; + + for (i = 0; i < nr_wins; i++) { + ret = msc_buffer_win_alloc(msc, nr_pages[i]); + if (ret) { + msc_buffer_multi_free(msc); + return ret; + } + } + + msc_buffer_relink(msc); + + return 0; +} + +/** + * msc_buffer_free() - free buffers for MSC + * @msc: MSC device + * + * Free MSC's storage buffers. + * + * This modifies msc::win_list and msc::base, which requires msc::buf_mutex to + * serialize, so the caller is expected to hold it. + */ +static void msc_buffer_free(struct msc *msc) +{ + if (msc->mode == MSC_MODE_SINGLE) + msc_buffer_contig_free(msc); + else if (msc->mode == MSC_MODE_MULTI) + msc_buffer_multi_free(msc); +} + +/** + * msc_buffer_alloc() - allocate a buffer for MSC + * @msc: MSC device + * @size: allocation size in bytes + * + * Allocate a storage buffer for MSC, depending on the msc::mode, it will be + * either done via msc_buffer_contig_alloc() for SINGLE operation mode or + * msc_buffer_win_alloc() for multiblock operation. The latter allocates one + * window per invocation, so in multiblock mode this can be called multiple + * times for the same MSC to allocate multiple windows. + * + * This modifies msc::win_list and msc::base, which requires msc::buf_mutex + * to serialize, so the caller is expected to hold it. + * + * Return: 0 on success, -errno otherwise. + */ +static int msc_buffer_alloc(struct msc *msc, unsigned long *nr_pages, + unsigned int nr_wins) +{ + int ret; + + /* -1: buffer not allocated */ + if (atomic_read(&msc->user_count) != -1) + return -EBUSY; + + if (msc->mode == MSC_MODE_SINGLE) { + if (nr_wins != 1) + return -EINVAL; + + ret = msc_buffer_contig_alloc(msc, nr_pages[0] << PAGE_SHIFT); + } else if (msc->mode == MSC_MODE_MULTI) { + ret = msc_buffer_multi_alloc(msc, nr_pages, nr_wins); + } else { + ret = -ENOTSUPP; + } + + if (!ret) { + /* allocation should be visible before the counter goes to 0 */ + smp_mb__before_atomic(); + + if (WARN_ON_ONCE(atomic_cmpxchg(&msc->user_count, -1, 0) != -1)) + return -EINVAL; + } + + return ret; +} + +/** + * msc_buffer_unlocked_free_unless_used() - free a buffer unless it's in use + * @msc: MSC device + * + * This will free MSC buffer unless it is in use or there is no allocated + * buffer. + * Caller needs to hold msc::buf_mutex. + * + * Return: 0 on successful deallocation or if there was no buffer to + * deallocate, -EBUSY if there are active users. + */ +static int msc_buffer_unlocked_free_unless_used(struct msc *msc) +{ + int count, ret = 0; + + count = atomic_cmpxchg(&msc->user_count, 0, -1); + + /* > 0: buffer is allocated and has users */ + if (count > 0) + ret = -EBUSY; + /* 0: buffer is allocated, no users */ + else if (!count) + msc_buffer_free(msc); + /* < 0: no buffer, nothing to do */ + + return ret; +} + +/** + * msc_buffer_free_unless_used() - free a buffer unless it's in use + * @msc: MSC device + * + * This is a locked version of msc_buffer_unlocked_free_unless_used(). + */ +static int msc_buffer_free_unless_used(struct msc *msc) +{ + int ret; + + mutex_lock(&msc->buf_mutex); + ret = msc_buffer_unlocked_free_unless_used(msc); + mutex_unlock(&msc->buf_mutex); + + return ret; +} + +/** + * msc_buffer_get_page() - get MSC buffer page at a given offset + * @msc: MSC device + * @pgoff: page offset into the storage buffer + * + * This traverses msc::win_list, so holding msc::buf_mutex is expected from + * the caller. + * + * Return: page if @pgoff corresponds to a valid buffer page or NULL. + */ +static struct page *msc_buffer_get_page(struct msc *msc, unsigned long pgoff) +{ + struct msc_window *win; + + if (msc->mode == MSC_MODE_SINGLE) + return msc_buffer_contig_get_page(msc, pgoff); + + list_for_each_entry(win, &msc->win_list, entry) + if (pgoff >= win->pgoff && pgoff < win->pgoff + win->nr_blocks) + goto found; + + return NULL; + +found: + pgoff -= win->pgoff; + return virt_to_page(win->block[pgoff].bdesc); +} + +/** + * struct msc_win_to_user_struct - data for copy_to_user() callback + * @buf: userspace buffer to copy data to + * @offset: running offset + */ +struct msc_win_to_user_struct { + char __user *buf; + unsigned long offset; +}; + +/** + * msc_win_to_user() - iterator for msc_buffer_iterate() to copy data to user + * @data: callback's private data + * @src: source buffer + * @len: amount of data to copy from the source buffer + */ +static unsigned long msc_win_to_user(void *data, void *src, size_t len) +{ + struct msc_win_to_user_struct *u = data; + unsigned long ret; + + ret = copy_to_user(u->buf + u->offset, src, len); + u->offset += len - ret; + + return ret; +} + + +/* + * file operations' callbacks + */ + +static int intel_th_msc_open(struct inode *inode, struct file *file) +{ + struct intel_th_device *thdev = file->private_data; + struct msc *msc = dev_get_drvdata(&thdev->dev); + struct msc_iter *iter; + + if (!capable(CAP_SYS_RAWIO)) + return -EPERM; + + iter = msc_iter_install(msc); + if (!iter) + return -ENOMEM; + + file->private_data = iter; + + return nonseekable_open(inode, file); +} + +static int intel_th_msc_release(struct inode *inode, struct file *file) +{ + struct msc_iter *iter = file->private_data; + struct msc *msc = iter->msc; + + msc_iter_remove(iter, msc); + + return 0; +} + +static ssize_t +msc_single_to_user(struct msc *msc, char __user *buf, loff_t off, size_t len) +{ + unsigned long size = msc->nr_pages << PAGE_SHIFT, rem = len; + unsigned long start = off, tocopy = 0; + + if (msc->single_wrap) { + start += msc->single_sz; + if (start < size) { + tocopy = min(rem, size - start); + if (copy_to_user(buf, msc->base + start, tocopy)) + return -EFAULT; + + buf += tocopy; + rem -= tocopy; + start += tocopy; + } + + start &= size - 1; + if (rem) { + tocopy = min(rem, msc->single_sz - start); + if (copy_to_user(buf, msc->base + start, tocopy)) + return -EFAULT; + + rem -= tocopy; + } + + return len - rem; + } + + if (copy_to_user(buf, msc->base + start, rem)) + return -EFAULT; + + return len; +} + +static ssize_t intel_th_msc_read(struct file *file, char __user *buf, + size_t len, loff_t *ppos) +{ + struct msc_iter *iter = file->private_data; + struct msc *msc = iter->msc; + size_t size; + loff_t off = *ppos; + ssize_t ret = 0; + + if (!atomic_inc_unless_negative(&msc->user_count)) + return 0; + + if (msc->enabled) { + ret = -EBUSY; + goto put_count; + } + + if (msc->mode == MSC_MODE_SINGLE && !msc->single_wrap) + size = msc->single_sz; + else + size = msc->nr_pages << PAGE_SHIFT; + + if (!size) + return 0; + + if (off >= size) { + len = 0; + goto put_count; + } + if (off + len >= size) + len = size - off; + + if (msc->mode == MSC_MODE_SINGLE) { + ret = msc_single_to_user(msc, buf, off, len); + if (ret >= 0) + *ppos += ret; + } else if (msc->mode == MSC_MODE_MULTI) { + struct msc_win_to_user_struct u = { + .buf = buf, + .offset = 0, + }; + + ret = msc_buffer_iterate(iter, len, &u, msc_win_to_user); + if (ret >= 0) + *ppos = iter->offset; + } else { + ret = -ENOTSUPP; + } + +put_count: + atomic_dec(&msc->user_count); + + return ret; +} + +/* + * vm operations callbacks (vm_ops) + */ + +static void msc_mmap_open(struct vm_area_struct *vma) +{ + struct msc_iter *iter = vma->vm_file->private_data; + struct msc *msc = iter->msc; + + atomic_inc(&msc->mmap_count); +} + +static void msc_mmap_close(struct vm_area_struct *vma) +{ + struct msc_iter *iter = vma->vm_file->private_data; + struct msc *msc = iter->msc; + unsigned long pg; + + if (!atomic_dec_and_mutex_lock(&msc->mmap_count, &msc->buf_mutex)) + return; + + /* drop page _counts */ + for (pg = 0; pg < msc->nr_pages; pg++) { + struct page *page = msc_buffer_get_page(msc, pg); + + if (WARN_ON_ONCE(!page)) + continue; + + if (page->mapping) + page->mapping = NULL; + } + + /* last mapping -- drop user_count */ + atomic_dec(&msc->user_count); + mutex_unlock(&msc->buf_mutex); +} + +static int msc_mmap_fault(struct vm_area_struct *vma, struct vm_fault *vmf) +{ + struct msc_iter *iter = vma->vm_file->private_data; + struct msc *msc = iter->msc; + + vmf->page = msc_buffer_get_page(msc, vmf->pgoff); + if (!vmf->page) + return VM_FAULT_SIGBUS; + + get_page(vmf->page); + vmf->page->mapping = vma->vm_file->f_mapping; + vmf->page->index = vmf->pgoff; + + return 0; +} + +static const struct vm_operations_struct msc_mmap_ops = { + .open = msc_mmap_open, + .close = msc_mmap_close, + .fault = msc_mmap_fault, +}; + +static int intel_th_msc_mmap(struct file *file, struct vm_area_struct *vma) +{ + unsigned long size = vma->vm_end - vma->vm_start; + struct msc_iter *iter = vma->vm_file->private_data; + struct msc *msc = iter->msc; + int ret = -EINVAL; + + if (!size || offset_in_page(size)) + return -EINVAL; + + if (vma->vm_pgoff) + return -EINVAL; + + /* grab user_count once per mmap; drop in msc_mmap_close() */ + if (!atomic_inc_unless_negative(&msc->user_count)) + return -EINVAL; + + if (msc->mode != MSC_MODE_SINGLE && + msc->mode != MSC_MODE_MULTI) + goto out; + + if (size >> PAGE_SHIFT != msc->nr_pages) + goto out; + + atomic_set(&msc->mmap_count, 1); + ret = 0; + +out: + if (ret) + atomic_dec(&msc->user_count); + + vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); + vma->vm_flags |= VM_DONTEXPAND | VM_DONTCOPY; + vma->vm_ops = &msc_mmap_ops; + return ret; +} + +static const struct file_operations intel_th_msc_fops = { + .open = intel_th_msc_open, + .release = intel_th_msc_release, + .read = intel_th_msc_read, + .mmap = intel_th_msc_mmap, + .llseek = no_llseek, +}; + +static int intel_th_msc_init(struct msc *msc) +{ + atomic_set(&msc->user_count, -1); + + msc->mode = MSC_MODE_MULTI; + mutex_init(&msc->buf_mutex); + INIT_LIST_HEAD(&msc->win_list); + + mutex_init(&msc->iter_mutex); + INIT_LIST_HEAD(&msc->iter_list); + + msc->burst_len = + (ioread32(msc->reg_base + REG_MSU_MSC0CTL) & MSC_LEN) >> + __ffs(MSC_LEN); + + return 0; +} + +static const char * const msc_mode[] = { + [MSC_MODE_SINGLE] = "single", + [MSC_MODE_MULTI] = "multi", + [MSC_MODE_EXI] = "ExI", + [MSC_MODE_DEBUG] = "debug", +}; + +static ssize_t +wrap_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct msc *msc = dev_get_drvdata(dev); + + return scnprintf(buf, PAGE_SIZE, "%d\n", msc->wrap); +} + +static ssize_t +wrap_store(struct device *dev, struct device_attribute *attr, const char *buf, + size_t size) +{ + struct msc *msc = dev_get_drvdata(dev); + unsigned long val; + int ret; + + ret = kstrtoul(buf, 10, &val); + if (ret) + return ret; + + msc->wrap = !!val; + + return size; +} + +static DEVICE_ATTR_RW(wrap); + +static ssize_t +mode_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct msc *msc = dev_get_drvdata(dev); + + return scnprintf(buf, PAGE_SIZE, "%s\n", msc_mode[msc->mode]); +} + +static ssize_t +mode_store(struct device *dev, struct device_attribute *attr, const char *buf, + size_t size) +{ + struct msc *msc = dev_get_drvdata(dev); + size_t len = size; + char *cp; + int i, ret; + + if (!capable(CAP_SYS_RAWIO)) + return -EPERM; + + cp = memchr(buf, '\n', len); + if (cp) + len = cp - buf; + + for (i = 0; i < ARRAY_SIZE(msc_mode); i++) + if (!strncmp(msc_mode[i], buf, len)) + goto found; + + return -EINVAL; + +found: + mutex_lock(&msc->buf_mutex); + ret = msc_buffer_unlocked_free_unless_used(msc); + if (!ret) + msc->mode = i; + mutex_unlock(&msc->buf_mutex); + + return ret ? ret : size; +} + +static DEVICE_ATTR_RW(mode); + +static ssize_t +nr_pages_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct msc *msc = dev_get_drvdata(dev); + struct msc_window *win; + size_t count = 0; + + mutex_lock(&msc->buf_mutex); + + if (msc->mode == MSC_MODE_SINGLE) + count = scnprintf(buf, PAGE_SIZE, "%ld\n", msc->nr_pages); + else if (msc->mode == MSC_MODE_MULTI) { + list_for_each_entry(win, &msc->win_list, entry) { + count += scnprintf(buf + count, PAGE_SIZE - count, + "%d%c", win->nr_blocks, + msc_is_last_win(win) ? '\n' : ','); + } + } else { + count = scnprintf(buf, PAGE_SIZE, "unsupported\n"); + } + + mutex_unlock(&msc->buf_mutex); + + return count; +} + +static ssize_t +nr_pages_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t size) +{ + struct msc *msc = dev_get_drvdata(dev); + unsigned long val, *win = NULL, *rewin; + size_t len = size; + const char *p = buf; + char *end, *s; + int ret, nr_wins = 0; + + if (!capable(CAP_SYS_RAWIO)) + return -EPERM; + + ret = msc_buffer_free_unless_used(msc); + if (ret) + return ret; + + /* scan the comma-separated list of allocation sizes */ + end = memchr(buf, '\n', len); + if (end) + len = end - buf; + + do { + end = memchr(p, ',', len); + s = kstrndup(p, end ? end - p : len, GFP_KERNEL); + ret = kstrtoul(s, 10, &val); + kfree(s); + + if (ret || !val) + goto free_win; + + if (nr_wins && msc->mode == MSC_MODE_SINGLE) { + ret = -EINVAL; + goto free_win; + } + + nr_wins++; + rewin = krealloc(win, sizeof(*win) * nr_wins, GFP_KERNEL); + if (!rewin) { + kfree(win); + return -ENOMEM; + } + + win = rewin; + win[nr_wins - 1] = val; + + if (!end) + break; + + len -= end - p; + p = end + 1; + } while (len); + + mutex_lock(&msc->buf_mutex); + ret = msc_buffer_alloc(msc, win, nr_wins); + mutex_unlock(&msc->buf_mutex); + +free_win: + kfree(win); + + return ret ? ret : size; +} + +static DEVICE_ATTR_RW(nr_pages); + +static struct attribute *msc_output_attrs[] = { + &dev_attr_wrap.attr, + &dev_attr_mode.attr, + &dev_attr_nr_pages.attr, + NULL, +}; + +static struct attribute_group msc_output_group = { + .attrs = msc_output_attrs, +}; + +static int intel_th_msc_probe(struct intel_th_device *thdev) +{ + struct device *dev = &thdev->dev; + struct resource *res; + struct msc *msc; + void __iomem *base; + int err; + + res = intel_th_device_get_resource(thdev, IORESOURCE_MEM, 0); + if (!res) + return -ENODEV; + + base = devm_ioremap(dev, res->start, resource_size(res)); + if (IS_ERR(base)) + return PTR_ERR(base); + + msc = devm_kzalloc(dev, sizeof(*msc), GFP_KERNEL); + if (!msc) + return -ENOMEM; + + msc->index = thdev->id; + + msc->thdev = thdev; + msc->reg_base = base + msc->index * 0x100; + + err = intel_th_msc_init(msc); + if (err) + return err; + + err = sysfs_create_group(&dev->kobj, &msc_output_group); + if (err) + return err; + + dev_set_drvdata(dev, msc); + + return 0; +} + +static void intel_th_msc_remove(struct intel_th_device *thdev) +{ + sysfs_remove_group(&thdev->dev.kobj, &msc_output_group); +} + +static struct intel_th_driver intel_th_msc_driver = { + .probe = intel_th_msc_probe, + .remove = intel_th_msc_remove, + .activate = intel_th_msc_activate, + .deactivate = intel_th_msc_deactivate, + .fops = &intel_th_msc_fops, + .driver = { + .name = "msc", + .owner = THIS_MODULE, + }, +}; + +module_driver(intel_th_msc_driver, + intel_th_driver_register, + intel_th_driver_unregister); + +MODULE_LICENSE("GPL v2"); +MODULE_DESCRIPTION("Intel(R) Trace Hub Memory Storage Unit driver"); +MODULE_AUTHOR("Alexander Shishkin <alexander.shishkin@linux.intel.com>"); diff --git a/drivers/hwtracing/intel_th/msu.h b/drivers/hwtracing/intel_th/msu.h new file mode 100644 index 000000000000..9b710e4aa98a --- /dev/null +++ b/drivers/hwtracing/intel_th/msu.h @@ -0,0 +1,116 @@ +/* + * Intel(R) Trace Hub Memory Storage Unit (MSU) data structures + * + * Copyright (C) 2014-2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ + +#ifndef __INTEL_TH_MSU_H__ +#define __INTEL_TH_MSU_H__ + +enum { + REG_MSU_MSUPARAMS = 0x0000, + REG_MSU_MSUSTS = 0x0008, + REG_MSU_MSC0CTL = 0x0100, /* MSC0 control */ + REG_MSU_MSC0STS = 0x0104, /* MSC0 status */ + REG_MSU_MSC0BAR = 0x0108, /* MSC0 output base address */ + REG_MSU_MSC0SIZE = 0x010c, /* MSC0 output size */ + REG_MSU_MSC0MWP = 0x0110, /* MSC0 write pointer */ + REG_MSU_MSC0NWSA = 0x011c, /* MSC0 next window start address */ + + REG_MSU_MSC1CTL = 0x0200, /* MSC1 control */ + REG_MSU_MSC1STS = 0x0204, /* MSC1 status */ + REG_MSU_MSC1BAR = 0x0208, /* MSC1 output base address */ + REG_MSU_MSC1SIZE = 0x020c, /* MSC1 output size */ + REG_MSU_MSC1MWP = 0x0210, /* MSC1 write pointer */ + REG_MSU_MSC1NWSA = 0x021c, /* MSC1 next window start address */ +}; + +/* MSUSTS bits */ +#define MSUSTS_MSU_INT BIT(0) + +/* MSCnCTL bits */ +#define MSC_EN BIT(0) +#define MSC_WRAPEN BIT(1) +#define MSC_RD_HDR_OVRD BIT(2) +#define MSC_MODE (BIT(4) | BIT(5)) +#define MSC_LEN (BIT(8) | BIT(9) | BIT(10)) + +/* MSC operating modes (MSC_MODE) */ +enum { + MSC_MODE_SINGLE = 0, + MSC_MODE_MULTI, + MSC_MODE_EXI, + MSC_MODE_DEBUG, +}; + +/* MSCnSTS bits */ +#define MSCSTS_WRAPSTAT BIT(1) /* Wrap occurred */ +#define MSCSTS_PLE BIT(2) /* Pipeline Empty */ + +/* + * Multiblock/multiwindow block descriptor + */ +struct msc_block_desc { + u32 sw_tag; + u32 block_sz; + u32 next_blk; + u32 next_win; + u32 res0[4]; + u32 hw_tag; + u32 valid_dw; + u32 ts_low; + u32 ts_high; + u32 res1[4]; +} __packed; + +#define MSC_BDESC sizeof(struct msc_block_desc) +#define DATA_IN_PAGE (PAGE_SIZE - MSC_BDESC) + +/* MSC multiblock sw tag bits */ +#define MSC_SW_TAG_LASTBLK BIT(0) +#define MSC_SW_TAG_LASTWIN BIT(1) + +/* MSC multiblock hw tag bits */ +#define MSC_HW_TAG_TRIGGER BIT(0) +#define MSC_HW_TAG_BLOCKWRAP BIT(1) +#define MSC_HW_TAG_WINWRAP BIT(2) +#define MSC_HW_TAG_ENDBIT BIT(3) + +static inline unsigned long msc_data_sz(struct msc_block_desc *bdesc) +{ + if (!bdesc->valid_dw) + return 0; + + return bdesc->valid_dw * 4 - MSC_BDESC; +} + +static inline bool msc_block_wrapped(struct msc_block_desc *bdesc) +{ + if (bdesc->hw_tag & MSC_HW_TAG_BLOCKWRAP) + return true; + + return false; +} + +static inline bool msc_block_last_written(struct msc_block_desc *bdesc) +{ + if ((bdesc->hw_tag & MSC_HW_TAG_ENDBIT) || + (msc_data_sz(bdesc) != DATA_IN_PAGE)) + return true; + + return false; +} + +/* waiting for Pipeline Empty bit(s) to assert for MSC */ +#define MSC_PLE_WAITLOOP_DEPTH 10000 + +#endif /* __INTEL_TH_MSU_H__ */ diff --git a/drivers/hwtracing/intel_th/pci.c b/drivers/hwtracing/intel_th/pci.c new file mode 100644 index 000000000000..641e87936064 --- /dev/null +++ b/drivers/hwtracing/intel_th/pci.c @@ -0,0 +1,86 @@ +/* + * Intel(R) Trace Hub pci driver + * + * Copyright (C) 2014-2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/types.h> +#include <linux/module.h> +#include <linux/device.h> +#include <linux/sysfs.h> +#include <linux/pci.h> + +#include "intel_th.h" + +#define DRIVER_NAME "intel_th_pci" + +#define BAR_MASK (BIT(TH_MMIO_CONFIG) | BIT(TH_MMIO_SW)) + +static int intel_th_pci_probe(struct pci_dev *pdev, + const struct pci_device_id *id) +{ + struct intel_th *th; + int err; + + err = pcim_enable_device(pdev); + if (err) + return err; + + err = pcim_iomap_regions_request_all(pdev, BAR_MASK, DRIVER_NAME); + if (err) + return err; + + th = intel_th_alloc(&pdev->dev, pdev->resource, + DEVICE_COUNT_RESOURCE, pdev->irq); + if (IS_ERR(th)) + return PTR_ERR(th); + + pci_set_drvdata(pdev, th); + + return 0; +} + +static void intel_th_pci_remove(struct pci_dev *pdev) +{ + struct intel_th *th = pci_get_drvdata(pdev); + + intel_th_free(th); +} + +static const struct pci_device_id intel_th_pci_id_table[] = { + { + PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x9d26), + .driver_data = (kernel_ulong_t)0, + }, + { + PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xa126), + .driver_data = (kernel_ulong_t)0, + }, + { 0 }, +}; + +MODULE_DEVICE_TABLE(pci, intel_th_pci_id_table); + +static struct pci_driver intel_th_pci_driver = { + .name = DRIVER_NAME, + .id_table = intel_th_pci_id_table, + .probe = intel_th_pci_probe, + .remove = intel_th_pci_remove, +}; + +module_pci_driver(intel_th_pci_driver); + +MODULE_LICENSE("GPL v2"); +MODULE_DESCRIPTION("Intel(R) Trace Hub PCI controller driver"); +MODULE_AUTHOR("Alexander Shishkin <alexander.shishkin@intel.com>"); diff --git a/drivers/hwtracing/intel_th/pti.c b/drivers/hwtracing/intel_th/pti.c new file mode 100644 index 000000000000..1e3bbc89825c --- /dev/null +++ b/drivers/hwtracing/intel_th/pti.c @@ -0,0 +1,252 @@ +/* + * Intel(R) Trace Hub PTI output driver + * + * Copyright (C) 2014-2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/types.h> +#include <linux/module.h> +#include <linux/device.h> +#include <linux/sizes.h> +#include <linux/printk.h> +#include <linux/slab.h> +#include <linux/mm.h> +#include <linux/io.h> + +#include "intel_th.h" +#include "pti.h" + +struct pti_device { + void __iomem *base; + struct intel_th_device *thdev; + unsigned int mode; + unsigned int freeclk; + unsigned int clkdiv; + unsigned int patgen; +}; + +/* map PTI widths to MODE settings of PTI_CTL register */ +static const unsigned int pti_mode[] = { + 0, 4, 8, 0, 12, 0, 0, 0, 16, 0, 0, 0, 0, 0, 0, 0, +}; + +static int pti_width_mode(unsigned int width) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(pti_mode); i++) + if (pti_mode[i] == width) + return i; + + return -EINVAL; +} + +static ssize_t mode_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct pti_device *pti = dev_get_drvdata(dev); + + return scnprintf(buf, PAGE_SIZE, "%d\n", pti_mode[pti->mode]); +} + +static ssize_t mode_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t size) +{ + struct pti_device *pti = dev_get_drvdata(dev); + unsigned long val; + int ret; + + ret = kstrtoul(buf, 10, &val); + if (ret) + return ret; + + ret = pti_width_mode(val); + if (ret < 0) + return ret; + + pti->mode = ret; + + return size; +} + +static DEVICE_ATTR_RW(mode); + +static ssize_t +freerunning_clock_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct pti_device *pti = dev_get_drvdata(dev); + + return scnprintf(buf, PAGE_SIZE, "%d\n", pti->freeclk); +} + +static ssize_t +freerunning_clock_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t size) +{ + struct pti_device *pti = dev_get_drvdata(dev); + unsigned long val; + int ret; + + ret = kstrtoul(buf, 10, &val); + if (ret) + return ret; + + pti->freeclk = !!val; + + return size; +} + +static DEVICE_ATTR_RW(freerunning_clock); + +static ssize_t +clock_divider_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct pti_device *pti = dev_get_drvdata(dev); + + return scnprintf(buf, PAGE_SIZE, "%d\n", 1u << pti->clkdiv); +} + +static ssize_t +clock_divider_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t size) +{ + struct pti_device *pti = dev_get_drvdata(dev); + unsigned long val; + int ret; + + ret = kstrtoul(buf, 10, &val); + if (ret) + return ret; + + if (!is_power_of_2(val) || val > 8 || !val) + return -EINVAL; + + pti->clkdiv = val; + + return size; +} + +static DEVICE_ATTR_RW(clock_divider); + +static struct attribute *pti_output_attrs[] = { + &dev_attr_mode.attr, + &dev_attr_freerunning_clock.attr, + &dev_attr_clock_divider.attr, + NULL, +}; + +static struct attribute_group pti_output_group = { + .attrs = pti_output_attrs, +}; + +static int intel_th_pti_activate(struct intel_th_device *thdev) +{ + struct pti_device *pti = dev_get_drvdata(&thdev->dev); + u32 ctl = PTI_EN; + + if (pti->patgen) + ctl |= pti->patgen << __ffs(PTI_PATGENMODE); + if (pti->freeclk) + ctl |= PTI_FCEN; + ctl |= pti->mode << __ffs(PTI_MODE); + ctl |= pti->clkdiv << __ffs(PTI_CLKDIV); + + iowrite32(ctl, pti->base + REG_PTI_CTL); + + intel_th_trace_enable(thdev); + + return 0; +} + +static void intel_th_pti_deactivate(struct intel_th_device *thdev) +{ + struct pti_device *pti = dev_get_drvdata(&thdev->dev); + + intel_th_trace_disable(thdev); + + iowrite32(0, pti->base + REG_PTI_CTL); +} + +static void read_hw_config(struct pti_device *pti) +{ + u32 ctl = ioread32(pti->base + REG_PTI_CTL); + + pti->mode = (ctl & PTI_MODE) >> __ffs(PTI_MODE); + pti->clkdiv = (ctl & PTI_CLKDIV) >> __ffs(PTI_CLKDIV); + pti->freeclk = !!(ctl & PTI_FCEN); + + if (!pti_mode[pti->mode]) + pti->mode = pti_width_mode(4); + if (!pti->clkdiv) + pti->clkdiv = 1; +} + +static int intel_th_pti_probe(struct intel_th_device *thdev) +{ + struct device *dev = &thdev->dev; + struct resource *res; + struct pti_device *pti; + void __iomem *base; + int ret; + + res = intel_th_device_get_resource(thdev, IORESOURCE_MEM, 0); + if (!res) + return -ENODEV; + + base = devm_ioremap(dev, res->start, resource_size(res)); + if (IS_ERR(base)) + return PTR_ERR(base); + + pti = devm_kzalloc(dev, sizeof(*pti), GFP_KERNEL); + if (!pti) + return -ENOMEM; + + pti->thdev = thdev; + pti->base = base; + + read_hw_config(pti); + + ret = sysfs_create_group(&dev->kobj, &pti_output_group); + if (ret) + return ret; + + dev_set_drvdata(dev, pti); + + return 0; +} + +static void intel_th_pti_remove(struct intel_th_device *thdev) +{ +} + +static struct intel_th_driver intel_th_pti_driver = { + .probe = intel_th_pti_probe, + .remove = intel_th_pti_remove, + .activate = intel_th_pti_activate, + .deactivate = intel_th_pti_deactivate, + .driver = { + .name = "pti", + .owner = THIS_MODULE, + }, +}; + +module_driver(intel_th_pti_driver, + intel_th_driver_register, + intel_th_driver_unregister); + +MODULE_LICENSE("GPL v2"); +MODULE_DESCRIPTION("Intel(R) Trace Hub PTI output driver"); +MODULE_AUTHOR("Alexander Shishkin <alexander.shishkin@linux.intel.com>"); diff --git a/drivers/hwtracing/intel_th/pti.h b/drivers/hwtracing/intel_th/pti.h new file mode 100644 index 000000000000..20883f5628cf --- /dev/null +++ b/drivers/hwtracing/intel_th/pti.h @@ -0,0 +1,29 @@ +/* + * Intel(R) Trace Hub PTI output data structures + * + * Copyright (C) 2014-2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ + +#ifndef __INTEL_TH_STH_H__ +#define __INTEL_TH_STH_H__ + +enum { + REG_PTI_CTL = 0x1c00, +}; + +#define PTI_EN BIT(0) +#define PTI_FCEN BIT(1) +#define PTI_MODE 0xf0 +#define PTI_CLKDIV 0x000f0000 +#define PTI_PATGENMODE 0x00f00000 + +#endif /* __INTEL_TH_STH_H__ */ diff --git a/drivers/hwtracing/intel_th/sth.c b/drivers/hwtracing/intel_th/sth.c new file mode 100644 index 000000000000..e488fccbfdec --- /dev/null +++ b/drivers/hwtracing/intel_th/sth.c @@ -0,0 +1,259 @@ +/* + * Intel(R) Trace Hub Software Trace Hub support + * + * Copyright (C) 2014-2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/types.h> +#include <linux/module.h> +#include <linux/device.h> +#include <linux/io.h> +#include <linux/mm.h> +#include <linux/slab.h> +#include <linux/stm.h> + +#include "intel_th.h" +#include "sth.h" + +struct sth_device { + void __iomem *base; + void __iomem *channels; + phys_addr_t channels_phys; + struct device *dev; + struct stm_data stm; + unsigned int sw_nmasters; +}; + +static struct intel_th_channel __iomem * +sth_channel(struct sth_device *sth, unsigned int master, unsigned int channel) +{ + struct intel_th_channel __iomem *sw_map = sth->channels; + + return &sw_map[(master - sth->stm.sw_start) * sth->stm.sw_nchannels + + channel]; +} + +static void sth_iowrite(void __iomem *dest, const unsigned char *payload, + unsigned int size) +{ + switch (size) { +#ifdef CONFIG_64BIT + case 8: + writeq_relaxed(*(u64 *)payload, dest); + break; +#endif + case 4: + writel_relaxed(*(u32 *)payload, dest); + break; + case 2: + writew_relaxed(*(u16 *)payload, dest); + break; + case 1: + writeb_relaxed(*(u8 *)payload, dest); + break; + default: + break; + } +} + +static ssize_t sth_stm_packet(struct stm_data *stm_data, unsigned int master, + unsigned int channel, unsigned int packet, + unsigned int flags, unsigned int size, + const unsigned char *payload) +{ + struct sth_device *sth = container_of(stm_data, struct sth_device, stm); + struct intel_th_channel __iomem *out = + sth_channel(sth, master, channel); + u64 __iomem *outp = &out->Dn; + unsigned long reg = REG_STH_TRIG; + +#ifndef CONFIG_64BIT + if (size > 4) + size = 4; +#endif + + size = rounddown_pow_of_two(size); + + switch (packet) { + /* Global packets (GERR, XSYNC, TRIG) are sent with register writes */ + case STP_PACKET_GERR: + reg += 4; + case STP_PACKET_XSYNC: + reg += 8; + case STP_PACKET_TRIG: + if (flags & STP_PACKET_TIMESTAMPED) + reg += 4; + iowrite8(*payload, sth->base + reg); + break; + + case STP_PACKET_MERR: + sth_iowrite(&out->MERR, payload, size); + break; + + case STP_PACKET_FLAG: + if (flags & STP_PACKET_TIMESTAMPED) + outp = (u64 __iomem *)&out->FLAG_TS; + else + outp = (u64 __iomem *)&out->FLAG; + + size = 1; + sth_iowrite(outp, payload, size); + break; + + case STP_PACKET_USER: + if (flags & STP_PACKET_TIMESTAMPED) + outp = &out->USER_TS; + else + outp = &out->USER; + sth_iowrite(outp, payload, size); + break; + + case STP_PACKET_DATA: + outp = &out->Dn; + + if (flags & STP_PACKET_TIMESTAMPED) + outp += 2; + if (flags & STP_PACKET_MARKED) + outp++; + + sth_iowrite(outp, payload, size); + break; + } + + return size; +} + +static phys_addr_t +sth_stm_mmio_addr(struct stm_data *stm_data, unsigned int master, + unsigned int channel, unsigned int nr_chans) +{ + struct sth_device *sth = container_of(stm_data, struct sth_device, stm); + phys_addr_t addr; + + master -= sth->stm.sw_start; + addr = sth->channels_phys + (master * sth->stm.sw_nchannels + channel) * + sizeof(struct intel_th_channel); + + if (offset_in_page(addr) || + offset_in_page(nr_chans * sizeof(struct intel_th_channel))) + return 0; + + return addr; +} + +static int sth_stm_link(struct stm_data *stm_data, unsigned int master, + unsigned int channel) +{ + struct sth_device *sth = container_of(stm_data, struct sth_device, stm); + + intel_th_set_output(to_intel_th_device(sth->dev), master); + + return 0; +} + +static int intel_th_sw_init(struct sth_device *sth) +{ + u32 reg; + + reg = ioread32(sth->base + REG_STH_STHCAP1); + sth->stm.sw_nchannels = reg & 0xff; + + reg = ioread32(sth->base + REG_STH_STHCAP0); + sth->stm.sw_start = reg & 0xffff; + sth->stm.sw_end = reg >> 16; + + sth->sw_nmasters = sth->stm.sw_end - sth->stm.sw_start; + dev_dbg(sth->dev, "sw_start: %x sw_end: %x masters: %x nchannels: %x\n", + sth->stm.sw_start, sth->stm.sw_end, sth->sw_nmasters, + sth->stm.sw_nchannels); + + return 0; +} + +static int intel_th_sth_probe(struct intel_th_device *thdev) +{ + struct device *dev = &thdev->dev; + struct sth_device *sth; + struct resource *res; + void __iomem *base, *channels; + int err; + + res = intel_th_device_get_resource(thdev, IORESOURCE_MEM, 0); + if (!res) + return -ENODEV; + + base = devm_ioremap(dev, res->start, resource_size(res)); + if (IS_ERR(base)) + return PTR_ERR(base); + + res = intel_th_device_get_resource(thdev, IORESOURCE_MEM, 1); + if (!res) + return -ENODEV; + + channels = devm_ioremap(dev, res->start, resource_size(res)); + if (IS_ERR(channels)) + return PTR_ERR(channels); + + sth = devm_kzalloc(dev, sizeof(*sth), GFP_KERNEL); + if (!sth) + return -ENOMEM; + + sth->dev = dev; + sth->base = base; + sth->channels = channels; + sth->channels_phys = res->start; + sth->stm.name = dev_name(dev); + sth->stm.packet = sth_stm_packet; + sth->stm.mmio_addr = sth_stm_mmio_addr; + sth->stm.sw_mmiosz = sizeof(struct intel_th_channel); + sth->stm.link = sth_stm_link; + + err = intel_th_sw_init(sth); + if (err) + return err; + + err = stm_register_device(dev, &sth->stm, THIS_MODULE); + if (err) { + dev_err(dev, "stm_register_device failed\n"); + return err; + } + + dev_set_drvdata(dev, sth); + + return 0; +} + +static void intel_th_sth_remove(struct intel_th_device *thdev) +{ + struct sth_device *sth = dev_get_drvdata(&thdev->dev); + + stm_unregister_device(&sth->stm); +} + +static struct intel_th_driver intel_th_sth_driver = { + .probe = intel_th_sth_probe, + .remove = intel_th_sth_remove, + .driver = { + .name = "sth", + .owner = THIS_MODULE, + }, +}; + +module_driver(intel_th_sth_driver, + intel_th_driver_register, + intel_th_driver_unregister); + +MODULE_LICENSE("GPL v2"); +MODULE_DESCRIPTION("Intel(R) Trace Hub Software Trace Hub driver"); +MODULE_AUTHOR("Alexander Shishkin <alexander.shishkin@intel.com>"); diff --git a/drivers/hwtracing/intel_th/sth.h b/drivers/hwtracing/intel_th/sth.h new file mode 100644 index 000000000000..f1390cd4f2ed --- /dev/null +++ b/drivers/hwtracing/intel_th/sth.h @@ -0,0 +1,42 @@ +/* + * Intel(R) Trace Hub Software Trace Hub (STH) data structures + * + * Copyright (C) 2014-2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ + +#ifndef __INTEL_TH_STH_H__ +#define __INTEL_TH_STH_H__ + +enum { + REG_STH_STHCAP0 = 0x0000, /* capabilities pt1 */ + REG_STH_STHCAP1 = 0x0004, /* capabilities pt2 */ + REG_STH_TRIG = 0x0008, /* TRIG packet payload */ + REG_STH_TRIG_TS = 0x000c, /* TRIG_TS packet payload */ + REG_STH_XSYNC = 0x0010, /* XSYNC packet payload */ + REG_STH_XSYNC_TS = 0x0014, /* XSYNC_TS packet payload */ + REG_STH_GERR = 0x0018, /* GERR packet payload */ +}; + +struct intel_th_channel { + u64 Dn; + u64 DnM; + u64 DnTS; + u64 DnMTS; + u64 USER; + u64 USER_TS; + u32 FLAG; + u32 FLAG_TS; + u32 MERR; + u32 __unused; +} __packed; + +#endif /* __INTEL_TH_STH_H__ */ diff --git a/drivers/hwtracing/stm/Kconfig b/drivers/hwtracing/stm/Kconfig new file mode 100644 index 000000000000..5a59a28cc3da --- /dev/null +++ b/drivers/hwtracing/stm/Kconfig @@ -0,0 +1,25 @@ +config STM + tristate "System Trace Module devices" + help + A System Trace Module (STM) is a device exporting data in System + Trace Protocol (STP) format as defined by MIPI STP standards. + Examples of such devices are Intel(R) Trace Hub and Coresight STM. + + Say Y here to enable System Trace Module device support. + +config STM_DUMMY + tristate "Dummy STM driver" + help + This is a simple dummy device that pretends to be an stm device + and discards your data. Use for stm class testing. + + If you don't know what this is, say N. + +config STM_SOURCE_CONSOLE + tristate "Kernel console over STM devices" + help + This is a kernel space trace source that sends kernel log + messages to trace hosts over STM devices. + + If you want to send kernel console messages over STM devices, + say Y. diff --git a/drivers/hwtracing/stm/Makefile b/drivers/hwtracing/stm/Makefile new file mode 100644 index 000000000000..f9312c38dd7a --- /dev/null +++ b/drivers/hwtracing/stm/Makefile @@ -0,0 +1,9 @@ +obj-$(CONFIG_STM) += stm_core.o + +stm_core-y := core.o policy.o + +obj-$(CONFIG_STM_DUMMY) += dummy_stm.o + +obj-$(CONFIG_STM_SOURCE_CONSOLE) += stm_console.o + +stm_console-y := console.o diff --git a/drivers/hwtracing/stm/console.c b/drivers/hwtracing/stm/console.c new file mode 100644 index 000000000000..c9d9a8d2ff52 --- /dev/null +++ b/drivers/hwtracing/stm/console.c @@ -0,0 +1,80 @@ +/* + * Simple kernel console driver for STM devices + * Copyright (c) 2014, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * STM console will send kernel messages over STM devices to a trace host. + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/console.h> +#include <linux/slab.h> +#include <linux/stm.h> + +static int stm_console_link(struct stm_source_data *data); +static void stm_console_unlink(struct stm_source_data *data); + +static struct stm_console { + struct stm_source_data data; + struct console console; +} stm_console = { + .data = { + .name = "console", + .nr_chans = 1, + .link = stm_console_link, + .unlink = stm_console_unlink, + }, +}; + +static void +stm_console_write(struct console *con, const char *buf, unsigned len) +{ + struct stm_console *sc = container_of(con, struct stm_console, console); + + stm_source_write(&sc->data, 0, buf, len); +} + +static int stm_console_link(struct stm_source_data *data) +{ + struct stm_console *sc = container_of(data, struct stm_console, data); + + strcpy(sc->console.name, "stm_console"); + sc->console.write = stm_console_write; + sc->console.flags = CON_ENABLED | CON_PRINTBUFFER; + register_console(&sc->console); + + return 0; +} + +static void stm_console_unlink(struct stm_source_data *data) +{ + struct stm_console *sc = container_of(data, struct stm_console, data); + + unregister_console(&sc->console); +} + +static int stm_console_init(void) +{ + return stm_source_register_device(NULL, &stm_console.data); +} + +static void stm_console_exit(void) +{ + stm_source_unregister_device(&stm_console.data); +} + +module_init(stm_console_init); +module_exit(stm_console_exit); + +MODULE_LICENSE("GPL v2"); +MODULE_DESCRIPTION("stm_console driver"); +MODULE_AUTHOR("Alexander Shishkin <alexander.shishkin@linux.intel.com>"); diff --git a/drivers/hwtracing/stm/core.c b/drivers/hwtracing/stm/core.c new file mode 100644 index 000000000000..b6445d9e5453 --- /dev/null +++ b/drivers/hwtracing/stm/core.c @@ -0,0 +1,1032 @@ +/* + * System Trace Module (STM) infrastructure + * Copyright (c) 2014, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * STM class implements generic infrastructure for System Trace Module devices + * as defined in MIPI STPv2 specification. + */ + +#include <linux/uaccess.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/device.h> +#include <linux/compat.h> +#include <linux/kdev_t.h> +#include <linux/srcu.h> +#include <linux/slab.h> +#include <linux/stm.h> +#include <linux/fs.h> +#include <linux/mm.h> +#include "stm.h" + +#include <uapi/linux/stm.h> + +static unsigned int stm_core_up; + +/* + * The SRCU here makes sure that STM device doesn't disappear from under a + * stm_source_write() caller, which may want to have as little overhead as + * possible. + */ +static struct srcu_struct stm_source_srcu; + +static ssize_t masters_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct stm_device *stm = to_stm_device(dev); + int ret; + + ret = sprintf(buf, "%u %u\n", stm->data->sw_start, stm->data->sw_end); + + return ret; +} + +static DEVICE_ATTR_RO(masters); + +static ssize_t channels_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct stm_device *stm = to_stm_device(dev); + int ret; + + ret = sprintf(buf, "%u\n", stm->data->sw_nchannels); + + return ret; +} + +static DEVICE_ATTR_RO(channels); + +static struct attribute *stm_attrs[] = { + &dev_attr_masters.attr, + &dev_attr_channels.attr, + NULL, +}; + +ATTRIBUTE_GROUPS(stm); + +static struct class stm_class = { + .name = "stm", + .dev_groups = stm_groups, +}; + +static int stm_dev_match(struct device *dev, const void *data) +{ + const char *name = data; + + return sysfs_streq(name, dev_name(dev)); +} + +/** + * stm_find_device() - find stm device by name + * @buf: character buffer containing the name + * + * This is called when either policy gets assigned to an stm device or an + * stm_source device gets linked to an stm device. + * + * This grabs device's reference (get_device()) and module reference, both + * of which the calling path needs to make sure to drop with stm_put_device(). + * + * Return: stm device pointer or null if lookup failed. + */ +struct stm_device *stm_find_device(const char *buf) +{ + struct stm_device *stm; + struct device *dev; + + if (!stm_core_up) + return NULL; + + dev = class_find_device(&stm_class, NULL, buf, stm_dev_match); + if (!dev) + return NULL; + + stm = to_stm_device(dev); + if (!try_module_get(stm->owner)) { + put_device(dev); + return NULL; + } + + return stm; +} + +/** + * stm_put_device() - drop references on the stm device + * @stm: stm device, previously acquired by stm_find_device() + * + * This drops the module reference and device reference taken by + * stm_find_device(). + */ +void stm_put_device(struct stm_device *stm) +{ + module_put(stm->owner); + put_device(&stm->dev); +} + +/* + * Internally we only care about software-writable masters here, that is the + * ones in the range [stm_data->sw_start..stm_data..sw_end], however we need + * original master numbers to be visible externally, since they are the ones + * that will appear in the STP stream. Thus, the internal bookkeeping uses + * $master - stm_data->sw_start to reference master descriptors and such. + */ + +#define __stm_master(_s, _m) \ + ((_s)->masters[(_m) - (_s)->data->sw_start]) + +static inline struct stp_master * +stm_master(struct stm_device *stm, unsigned int idx) +{ + if (idx < stm->data->sw_start || idx > stm->data->sw_end) + return NULL; + + return __stm_master(stm, idx); +} + +static int stp_master_alloc(struct stm_device *stm, unsigned int idx) +{ + struct stp_master *master; + size_t size; + + size = ALIGN(stm->data->sw_nchannels, 8) / 8; + size += sizeof(struct stp_master); + master = kzalloc(size, GFP_ATOMIC); + if (!master) + return -ENOMEM; + + master->nr_free = stm->data->sw_nchannels; + __stm_master(stm, idx) = master; + + return 0; +} + +static void stp_master_free(struct stm_device *stm, unsigned int idx) +{ + struct stp_master *master = stm_master(stm, idx); + + if (!master) + return; + + __stm_master(stm, idx) = NULL; + kfree(master); +} + +static void stm_output_claim(struct stm_device *stm, struct stm_output *output) +{ + struct stp_master *master = stm_master(stm, output->master); + + if (WARN_ON_ONCE(master->nr_free < output->nr_chans)) + return; + + bitmap_allocate_region(&master->chan_map[0], output->channel, + ilog2(output->nr_chans)); + + master->nr_free -= output->nr_chans; +} + +static void +stm_output_disclaim(struct stm_device *stm, struct stm_output *output) +{ + struct stp_master *master = stm_master(stm, output->master); + + bitmap_release_region(&master->chan_map[0], output->channel, + ilog2(output->nr_chans)); + + output->nr_chans = 0; + master->nr_free += output->nr_chans; +} + +/* + * This is like bitmap_find_free_region(), except it can ignore @start bits + * at the beginning. + */ +static int find_free_channels(unsigned long *bitmap, unsigned int start, + unsigned int end, unsigned int width) +{ + unsigned int pos; + int i; + + for (pos = start; pos < end + 1; pos = ALIGN(pos, width)) { + pos = find_next_zero_bit(bitmap, end + 1, pos); + if (pos + width > end + 1) + break; + + if (pos & (width - 1)) + continue; + + for (i = 1; i < width && !test_bit(pos + i, bitmap); i++) + ; + if (i == width) + return pos; + } + + return -1; +} + +static unsigned int +stm_find_master_chan(struct stm_device *stm, unsigned int width, + unsigned int *mstart, unsigned int mend, + unsigned int *cstart, unsigned int cend) +{ + struct stp_master *master; + unsigned int midx; + int pos, err; + + for (midx = *mstart; midx <= mend; midx++) { + if (!stm_master(stm, midx)) { + err = stp_master_alloc(stm, midx); + if (err) + return err; + } + + master = stm_master(stm, midx); + + if (!master->nr_free) + continue; + + pos = find_free_channels(master->chan_map, *cstart, cend, + width); + if (pos < 0) + continue; + + *mstart = midx; + *cstart = pos; + return 0; + } + + return -ENOSPC; +} + +static int stm_output_assign(struct stm_device *stm, unsigned int width, + struct stp_policy_node *policy_node, + struct stm_output *output) +{ + unsigned int midx, cidx, mend, cend; + int ret = -EINVAL; + + if (width > stm->data->sw_nchannels) + return -EINVAL; + + if (policy_node) { + stp_policy_node_get_ranges(policy_node, + &midx, &mend, &cidx, &cend); + } else { + midx = stm->data->sw_start; + cidx = 0; + mend = stm->data->sw_end; + cend = stm->data->sw_nchannels - 1; + } + + spin_lock(&stm->mc_lock); + /* output is already assigned -- shouldn't happen */ + if (WARN_ON_ONCE(output->nr_chans)) + goto unlock; + + ret = stm_find_master_chan(stm, width, &midx, mend, &cidx, cend); + if (ret) + goto unlock; + + output->master = midx; + output->channel = cidx; + output->nr_chans = width; + stm_output_claim(stm, output); + dev_dbg(&stm->dev, "assigned %u:%u (+%u)\n", midx, cidx, width); + + ret = 0; +unlock: + spin_unlock(&stm->mc_lock); + + return ret; +} + +static void stm_output_free(struct stm_device *stm, struct stm_output *output) +{ + spin_lock(&stm->mc_lock); + if (output->nr_chans) + stm_output_disclaim(stm, output); + spin_unlock(&stm->mc_lock); +} + +static int major_match(struct device *dev, const void *data) +{ + unsigned int major = *(unsigned int *)data; + + return MAJOR(dev->devt) == major; +} + +static int stm_char_open(struct inode *inode, struct file *file) +{ + struct stm_file *stmf; + struct device *dev; + unsigned int major = imajor(inode); + int err = -ENODEV; + + dev = class_find_device(&stm_class, NULL, &major, major_match); + if (!dev) + return -ENODEV; + + stmf = kzalloc(sizeof(*stmf), GFP_KERNEL); + if (!stmf) + return -ENOMEM; + + stmf->stm = to_stm_device(dev); + + if (!try_module_get(stmf->stm->owner)) + goto err_free; + + file->private_data = stmf; + + return nonseekable_open(inode, file); + +err_free: + kfree(stmf); + + return err; +} + +static int stm_char_release(struct inode *inode, struct file *file) +{ + struct stm_file *stmf = file->private_data; + + stm_output_free(stmf->stm, &stmf->output); + stm_put_device(stmf->stm); + kfree(stmf); + + return 0; +} + +static int stm_file_assign(struct stm_file *stmf, char *id, unsigned int width) +{ + struct stm_device *stm = stmf->stm; + int ret; + + stmf->policy_node = stp_policy_node_lookup(stm, id); + + ret = stm_output_assign(stm, width, stmf->policy_node, &stmf->output); + + if (stmf->policy_node) + stp_policy_node_put(stmf->policy_node); + + return ret; +} + +static void stm_write(struct stm_data *data, unsigned int master, + unsigned int channel, const char *buf, size_t count) +{ + unsigned int flags = STP_PACKET_TIMESTAMPED; + const unsigned char *p = buf, nil = 0; + size_t pos; + ssize_t sz; + + for (pos = 0, p = buf; count > pos; pos += sz, p += sz) { + sz = min_t(unsigned int, count - pos, 8); + sz = data->packet(data, master, channel, STP_PACKET_DATA, flags, + sz, p); + flags = 0; + } + + data->packet(data, master, channel, STP_PACKET_FLAG, 0, 0, &nil); +} + +static ssize_t stm_char_write(struct file *file, const char __user *buf, + size_t count, loff_t *ppos) +{ + struct stm_file *stmf = file->private_data; + struct stm_device *stm = stmf->stm; + char *kbuf; + int err; + + /* + * if no m/c have been assigned to this writer up to this + * point, use "default" policy entry + */ + if (!stmf->output.nr_chans) { + err = stm_file_assign(stmf, "default", 1); + /* + * EBUSY means that somebody else just assigned this + * output, which is just fine for write() + */ + if (err && err != -EBUSY) + return err; + } + + kbuf = kmalloc(count + 1, GFP_KERNEL); + if (!kbuf) + return -ENOMEM; + + err = copy_from_user(kbuf, buf, count); + if (err) { + kfree(kbuf); + return -EFAULT; + } + + stm_write(stm->data, stmf->output.master, stmf->output.channel, kbuf, + count); + + kfree(kbuf); + + return count; +} + +static int stm_char_mmap(struct file *file, struct vm_area_struct *vma) +{ + struct stm_file *stmf = file->private_data; + struct stm_device *stm = stmf->stm; + unsigned long size, phys; + + if (!stm->data->mmio_addr) + return -EOPNOTSUPP; + + if (vma->vm_pgoff) + return -EINVAL; + + size = vma->vm_end - vma->vm_start; + + if (stmf->output.nr_chans * stm->data->sw_mmiosz != size) + return -EINVAL; + + phys = stm->data->mmio_addr(stm->data, stmf->output.master, + stmf->output.channel, + stmf->output.nr_chans); + + if (!phys) + return -EINVAL; + + vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); + vma->vm_flags |= VM_IO | VM_DONTEXPAND | VM_DONTDUMP; + vm_iomap_memory(vma, phys, size); + + return 0; +} + +static int stm_char_policy_set_ioctl(struct stm_file *stmf, void __user *arg) +{ + struct stm_device *stm = stmf->stm; + struct stp_policy_id *id; + int ret = -EINVAL; + u32 size; + + if (stmf->output.nr_chans) + return -EBUSY; + + if (copy_from_user(&size, arg, sizeof(size))) + return -EFAULT; + + if (size >= PATH_MAX + sizeof(*id)) + return -EINVAL; + + /* + * size + 1 to make sure the .id string at the bottom is terminated, + * which is also why memdup_user() is not useful here + */ + id = kzalloc(size + 1, GFP_KERNEL); + if (!id) + return -ENOMEM; + + if (copy_from_user(id, arg, size)) { + ret = -EFAULT; + goto err_free; + } + + if (id->__reserved_0 || id->__reserved_1) + goto err_free; + + if (id->width < 1 || + id->width > PAGE_SIZE / stm->data->sw_mmiosz) + goto err_free; + + ret = stm_file_assign(stmf, id->id, id->width); + if (ret) + goto err_free; + + ret = 0; + + if (stm->data->link) + ret = stm->data->link(stm->data, stmf->output.master, + stmf->output.channel); + + if (ret) { + stm_output_free(stmf->stm, &stmf->output); + stm_put_device(stmf->stm); + } + +err_free: + kfree(id); + + return ret; +} + +static int stm_char_policy_get_ioctl(struct stm_file *stmf, void __user *arg) +{ + struct stp_policy_id id = { + .size = sizeof(id), + .master = stmf->output.master, + .channel = stmf->output.channel, + .width = stmf->output.nr_chans, + .__reserved_0 = 0, + .__reserved_1 = 0, + }; + + return copy_to_user(arg, &id, id.size) ? -EFAULT : 0; +} + +static long +stm_char_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +{ + struct stm_file *stmf = file->private_data; + struct stm_data *stm_data = stmf->stm->data; + int err = -ENOTTY; + u64 options; + + switch (cmd) { + case STP_POLICY_ID_SET: + err = stm_char_policy_set_ioctl(stmf, (void __user *)arg); + if (err) + return err; + + return stm_char_policy_get_ioctl(stmf, (void __user *)arg); + + case STP_POLICY_ID_GET: + return stm_char_policy_get_ioctl(stmf, (void __user *)arg); + + case STP_SET_OPTIONS: + if (copy_from_user(&options, (u64 __user *)arg, sizeof(u64))) + return -EFAULT; + + if (stm_data->set_options) + err = stm_data->set_options(stm_data, + stmf->output.master, + stmf->output.channel, + stmf->output.nr_chans, + options); + + break; + default: + break; + } + + return err; +} + +#ifdef CONFIG_COMPAT +static long +stm_char_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +{ + return stm_char_ioctl(file, cmd, (unsigned long)compat_ptr(arg)); +} +#else +#define stm_char_compat_ioctl NULL +#endif + +static const struct file_operations stm_fops = { + .open = stm_char_open, + .release = stm_char_release, + .write = stm_char_write, + .mmap = stm_char_mmap, + .unlocked_ioctl = stm_char_ioctl, + .compat_ioctl = stm_char_compat_ioctl, + .llseek = no_llseek, +}; + +static void stm_device_release(struct device *dev) +{ + struct stm_device *stm = to_stm_device(dev); + + kfree(stm); +} + +int stm_register_device(struct device *parent, struct stm_data *stm_data, + struct module *owner) +{ + struct stm_device *stm; + unsigned int nmasters; + int err = -ENOMEM; + + if (!stm_core_up) + return -EPROBE_DEFER; + + if (!stm_data->packet || !stm_data->sw_nchannels) + return -EINVAL; + + nmasters = stm_data->sw_end - stm_data->sw_start; + stm = kzalloc(sizeof(*stm) + nmasters * sizeof(void *), GFP_KERNEL); + if (!stm) + return -ENOMEM; + + stm->major = register_chrdev(0, stm_data->name, &stm_fops); + if (stm->major < 0) + goto err_free; + + device_initialize(&stm->dev); + stm->dev.devt = MKDEV(stm->major, 0); + stm->dev.class = &stm_class; + stm->dev.parent = parent; + stm->dev.release = stm_device_release; + + err = kobject_set_name(&stm->dev.kobj, "%s", stm_data->name); + if (err) + goto err_device; + + err = device_add(&stm->dev); + if (err) + goto err_device; + + spin_lock_init(&stm->link_lock); + INIT_LIST_HEAD(&stm->link_list); + + spin_lock_init(&stm->mc_lock); + mutex_init(&stm->policy_mutex); + stm->sw_nmasters = nmasters; + stm->owner = owner; + stm->data = stm_data; + stm_data->stm = stm; + + return 0; + +err_device: + put_device(&stm->dev); +err_free: + kfree(stm); + + return err; +} +EXPORT_SYMBOL_GPL(stm_register_device); + +static void __stm_source_link_drop(struct stm_source_device *src, + struct stm_device *stm); + +void stm_unregister_device(struct stm_data *stm_data) +{ + struct stm_device *stm = stm_data->stm; + struct stm_source_device *src, *iter; + int i; + + spin_lock(&stm->link_lock); + list_for_each_entry_safe(src, iter, &stm->link_list, link_entry) { + __stm_source_link_drop(src, stm); + } + spin_unlock(&stm->link_lock); + + synchronize_srcu(&stm_source_srcu); + + unregister_chrdev(stm->major, stm_data->name); + + mutex_lock(&stm->policy_mutex); + if (stm->policy) + stp_policy_unbind(stm->policy); + mutex_unlock(&stm->policy_mutex); + + for (i = 0; i < stm->sw_nmasters; i++) + stp_master_free(stm, i); + + device_unregister(&stm->dev); + stm_data->stm = NULL; +} +EXPORT_SYMBOL_GPL(stm_unregister_device); + +/** + * stm_source_link_add() - connect an stm_source device to an stm device + * @src: stm_source device + * @stm: stm device + * + * This function establishes a link from stm_source to an stm device so that + * the former can send out trace data to the latter. + * + * Return: 0 on success, -errno otherwise. + */ +static int stm_source_link_add(struct stm_source_device *src, + struct stm_device *stm) +{ + char *id; + int err; + + spin_lock(&stm->link_lock); + spin_lock(&src->link_lock); + + /* src->link is dereferenced under stm_source_srcu but not the list */ + rcu_assign_pointer(src->link, stm); + list_add_tail(&src->link_entry, &stm->link_list); + + spin_unlock(&src->link_lock); + spin_unlock(&stm->link_lock); + + id = kstrdup(src->data->name, GFP_KERNEL); + if (id) { + src->policy_node = + stp_policy_node_lookup(stm, id); + + kfree(id); + } + + err = stm_output_assign(stm, src->data->nr_chans, + src->policy_node, &src->output); + + if (src->policy_node) + stp_policy_node_put(src->policy_node); + + if (err) + goto fail_detach; + + /* this is to notify the STM device that a new link has been made */ + if (stm->data->link) + err = stm->data->link(stm->data, src->output.master, + src->output.channel); + + if (err) + goto fail_free_output; + + /* this is to let the source carry out all necessary preparations */ + if (src->data->link) + src->data->link(src->data); + + return 0; + +fail_free_output: + stm_output_free(stm, &src->output); + stm_put_device(stm); + +fail_detach: + spin_lock(&stm->link_lock); + spin_lock(&src->link_lock); + + rcu_assign_pointer(src->link, NULL); + list_del_init(&src->link_entry); + + spin_unlock(&src->link_lock); + spin_unlock(&stm->link_lock); + + return err; +} + +/** + * __stm_source_link_drop() - detach stm_source from an stm device + * @src: stm_source device + * @stm: stm device + * + * If @stm is @src::link, disconnect them from one another and put the + * reference on the @stm device. + * + * Caller must hold stm::link_lock. + */ +static void __stm_source_link_drop(struct stm_source_device *src, + struct stm_device *stm) +{ + struct stm_device *link; + + spin_lock(&src->link_lock); + link = srcu_dereference_check(src->link, &stm_source_srcu, 1); + if (WARN_ON_ONCE(link != stm)) { + spin_unlock(&src->link_lock); + return; + } + + stm_output_free(link, &src->output); + /* caller must hold stm::link_lock */ + list_del_init(&src->link_entry); + /* matches stm_find_device() from stm_source_link_store() */ + stm_put_device(link); + rcu_assign_pointer(src->link, NULL); + + spin_unlock(&src->link_lock); +} + +/** + * stm_source_link_drop() - detach stm_source from its stm device + * @src: stm_source device + * + * Unlinking means disconnecting from source's STM device; after this + * writes will be unsuccessful until it is linked to a new STM device. + * + * This will happen on "stm_source_link" sysfs attribute write to undo + * the existing link (if any), or on linked STM device's de-registration. + */ +static void stm_source_link_drop(struct stm_source_device *src) +{ + struct stm_device *stm; + int idx; + + idx = srcu_read_lock(&stm_source_srcu); + stm = srcu_dereference(src->link, &stm_source_srcu); + + if (stm) { + if (src->data->unlink) + src->data->unlink(src->data); + + spin_lock(&stm->link_lock); + __stm_source_link_drop(src, stm); + spin_unlock(&stm->link_lock); + } + + srcu_read_unlock(&stm_source_srcu, idx); +} + +static ssize_t stm_source_link_show(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + struct stm_source_device *src = to_stm_source_device(dev); + struct stm_device *stm; + int idx, ret; + + idx = srcu_read_lock(&stm_source_srcu); + stm = srcu_dereference(src->link, &stm_source_srcu); + ret = sprintf(buf, "%s\n", + stm ? dev_name(&stm->dev) : "<none>"); + srcu_read_unlock(&stm_source_srcu, idx); + + return ret; +} + +static ssize_t stm_source_link_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct stm_source_device *src = to_stm_source_device(dev); + struct stm_device *link; + int err; + + stm_source_link_drop(src); + + link = stm_find_device(buf); + if (!link) + return -EINVAL; + + err = stm_source_link_add(src, link); + if (err) + stm_put_device(link); + + return err ? : count; +} + +static DEVICE_ATTR_RW(stm_source_link); + +static struct attribute *stm_source_attrs[] = { + &dev_attr_stm_source_link.attr, + NULL, +}; + +ATTRIBUTE_GROUPS(stm_source); + +static struct class stm_source_class = { + .name = "stm_source", + .dev_groups = stm_source_groups, +}; + +static void stm_source_device_release(struct device *dev) +{ + struct stm_source_device *src = to_stm_source_device(dev); + + kfree(src); +} + +/** + * stm_source_register_device() - register an stm_source device + * @parent: parent device + * @data: device description structure + * + * This will create a device of stm_source class that can write + * data to an stm device once linked. + * + * Return: 0 on success, -errno otherwise. + */ +int stm_source_register_device(struct device *parent, + struct stm_source_data *data) +{ + struct stm_source_device *src; + int err; + + if (!stm_core_up) + return -EPROBE_DEFER; + + src = kzalloc(sizeof(*src), GFP_KERNEL); + if (!src) + return -ENOMEM; + + device_initialize(&src->dev); + src->dev.class = &stm_source_class; + src->dev.parent = parent; + src->dev.release = stm_source_device_release; + + err = kobject_set_name(&src->dev.kobj, "%s", data->name); + if (err) + goto err; + + err = device_add(&src->dev); + if (err) + goto err; + + spin_lock_init(&src->link_lock); + INIT_LIST_HEAD(&src->link_entry); + src->data = data; + data->src = src; + + return 0; + +err: + put_device(&src->dev); + kfree(src); + + return err; +} +EXPORT_SYMBOL_GPL(stm_source_register_device); + +/** + * stm_source_unregister_device() - unregister an stm_source device + * @data: device description that was used to register the device + * + * This will remove a previously created stm_source device from the system. + */ +void stm_source_unregister_device(struct stm_source_data *data) +{ + struct stm_source_device *src = data->src; + + stm_source_link_drop(src); + + device_destroy(&stm_source_class, src->dev.devt); +} +EXPORT_SYMBOL_GPL(stm_source_unregister_device); + +int stm_source_write(struct stm_source_data *data, unsigned int chan, + const char *buf, size_t count) +{ + struct stm_source_device *src = data->src; + struct stm_device *stm; + int idx; + + if (!src->output.nr_chans) + return -ENODEV; + + if (chan >= src->output.nr_chans) + return -EINVAL; + + idx = srcu_read_lock(&stm_source_srcu); + + stm = srcu_dereference(src->link, &stm_source_srcu); + if (stm) + stm_write(stm->data, src->output.master, + src->output.channel + chan, + buf, count); + else + count = -ENODEV; + + srcu_read_unlock(&stm_source_srcu, idx); + + return count; +} +EXPORT_SYMBOL_GPL(stm_source_write); + +static int __init stm_core_init(void) +{ + int err; + + err = class_register(&stm_class); + if (err) + return err; + + err = class_register(&stm_source_class); + if (err) + goto err_stm; + + err = stp_configfs_init(); + if (err) + goto err_src; + + init_srcu_struct(&stm_source_srcu); + + stm_core_up++; + + return 0; + +err_src: + class_unregister(&stm_source_class); +err_stm: + class_unregister(&stm_class); + + return err; +} + +module_init(stm_core_init); + +static void __exit stm_core_exit(void) +{ + cleanup_srcu_struct(&stm_source_srcu); + class_unregister(&stm_source_class); + class_unregister(&stm_class); + stp_configfs_exit(); +} + +module_exit(stm_core_exit); + +MODULE_LICENSE("GPL v2"); +MODULE_DESCRIPTION("System Trace Module device class"); +MODULE_AUTHOR("Alexander Shishkin <alexander.shishkin@linux.intel.com>"); diff --git a/drivers/hwtracing/stm/dummy_stm.c b/drivers/hwtracing/stm/dummy_stm.c new file mode 100644 index 000000000000..3709bef0b21f --- /dev/null +++ b/drivers/hwtracing/stm/dummy_stm.c @@ -0,0 +1,66 @@ +/* + * A dummy STM device for stm/stm_source class testing. + * Copyright (c) 2014, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * STM class implements generic infrastructure for System Trace Module devices + * as defined in MIPI STPv2 specification. + */ + +#undef DEBUG +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/stm.h> + +static ssize_t +dummy_stm_packet(struct stm_data *stm_data, unsigned int master, + unsigned int channel, unsigned int packet, unsigned int flags, + unsigned int size, const unsigned char *payload) +{ +#ifdef DEBUG + u64 pl = 0; + + if (payload) + pl = *(u64 *)payload; + + if (size < 8) + pl &= (1ull << (size * 8)) - 1; + trace_printk("[%u:%u] [pkt: %x/%x] (%llx)\n", master, channel, + packet, size, pl); +#endif + return size; +} + +static struct stm_data dummy_stm = { + .name = "dummy_stm", + .sw_start = 0x0000, + .sw_end = 0xffff, + .sw_nchannels = 0xffff, + .packet = dummy_stm_packet, +}; + +static int dummy_stm_init(void) +{ + return stm_register_device(NULL, &dummy_stm, THIS_MODULE); +} + +static void dummy_stm_exit(void) +{ + stm_unregister_device(&dummy_stm); +} + +module_init(dummy_stm_init); +module_exit(dummy_stm_exit); + +MODULE_LICENSE("GPL v2"); +MODULE_DESCRIPTION("dummy_stm device"); +MODULE_AUTHOR("Alexander Shishkin <alexander.shishkin@linux.intel.com>"); diff --git a/drivers/hwtracing/stm/policy.c b/drivers/hwtracing/stm/policy.c new file mode 100644 index 000000000000..6498a9dbb7bd --- /dev/null +++ b/drivers/hwtracing/stm/policy.c @@ -0,0 +1,529 @@ +/* + * System Trace Module (STM) master/channel allocation policy management + * Copyright (c) 2014, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * A master/channel allocation policy allows mapping string identifiers to + * master and channel ranges, where allocation can be done. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/types.h> +#include <linux/module.h> +#include <linux/device.h> +#include <linux/configfs.h> +#include <linux/slab.h> +#include <linux/stm.h> +#include "stm.h" + +/* + * STP Master/Channel allocation policy configfs layout. + */ + +struct stp_policy { + struct config_group group; + struct stm_device *stm; +}; + +struct stp_policy_node { + struct config_group group; + struct stp_policy *policy; + unsigned int first_master; + unsigned int last_master; + unsigned int first_channel; + unsigned int last_channel; +}; + +static struct configfs_subsystem stp_policy_subsys; + +void stp_policy_node_get_ranges(struct stp_policy_node *policy_node, + unsigned int *mstart, unsigned int *mend, + unsigned int *cstart, unsigned int *cend) +{ + *mstart = policy_node->first_master; + *mend = policy_node->last_master; + *cstart = policy_node->first_channel; + *cend = policy_node->last_channel; +} + +static inline char *stp_policy_node_name(struct stp_policy_node *policy_node) +{ + return policy_node->group.cg_item.ci_name ? : "<none>"; +} + +static inline struct stp_policy *to_stp_policy(struct config_item *item) +{ + return item ? + container_of(to_config_group(item), struct stp_policy, group) : + NULL; +} + +static inline struct stp_policy_node * +to_stp_policy_node(struct config_item *item) +{ + return item ? + container_of(to_config_group(item), struct stp_policy_node, + group) : + NULL; +} + +static ssize_t stp_policy_node_masters_show(struct stp_policy_node *policy_node, + char *page) +{ + ssize_t count; + + count = sprintf(page, "%u %u\n", policy_node->first_master, + policy_node->last_master); + + return count; +} + +static ssize_t +stp_policy_node_masters_store(struct stp_policy_node *policy_node, + const char *page, size_t count) +{ + unsigned int first, last; + struct stm_device *stm; + char *p = (char *)page; + ssize_t ret = -ENODEV; + + if (sscanf(p, "%u %u", &first, &last) != 2) + return -EINVAL; + + mutex_lock(&stp_policy_subsys.su_mutex); + stm = policy_node->policy->stm; + if (!stm) + goto unlock; + + /* must be within [sw_start..sw_end], which is an inclusive range */ + if (first > INT_MAX || last > INT_MAX || first > last || + first < stm->data->sw_start || + last > stm->data->sw_end) { + ret = -ERANGE; + goto unlock; + } + + ret = count; + policy_node->first_master = first; + policy_node->last_master = last; + +unlock: + mutex_unlock(&stp_policy_subsys.su_mutex); + + return ret; +} + +static ssize_t +stp_policy_node_channels_show(struct stp_policy_node *policy_node, char *page) +{ + ssize_t count; + + count = sprintf(page, "%u %u\n", policy_node->first_channel, + policy_node->last_channel); + + return count; +} + +static ssize_t +stp_policy_node_channels_store(struct stp_policy_node *policy_node, + const char *page, size_t count) +{ + unsigned int first, last; + struct stm_device *stm; + char *p = (char *)page; + ssize_t ret = -ENODEV; + + if (sscanf(p, "%u %u", &first, &last) != 2) + return -EINVAL; + + mutex_lock(&stp_policy_subsys.su_mutex); + stm = policy_node->policy->stm; + if (!stm) + goto unlock; + + if (first > INT_MAX || last > INT_MAX || first > last || + last >= stm->data->sw_nchannels) { + ret = -ERANGE; + goto unlock; + } + + ret = count; + policy_node->first_channel = first; + policy_node->last_channel = last; + +unlock: + mutex_unlock(&stp_policy_subsys.su_mutex); + + return ret; +} + +static void stp_policy_node_release(struct config_item *item) +{ + kfree(to_stp_policy_node(item)); +} + +struct stp_policy_node_attribute { + struct configfs_attribute attr; + ssize_t (*show)(struct stp_policy_node *, char *); + ssize_t (*store)(struct stp_policy_node *, const char *, size_t); +}; + +static ssize_t stp_policy_node_attr_show(struct config_item *item, + struct configfs_attribute *attr, + char *page) +{ + struct stp_policy_node *policy_node = to_stp_policy_node(item); + struct stp_policy_node_attribute *pn_attr = + container_of(attr, struct stp_policy_node_attribute, attr); + ssize_t count = 0; + + if (pn_attr->show) + count = pn_attr->show(policy_node, page); + + return count; +} + +static ssize_t stp_policy_node_attr_store(struct config_item *item, + struct configfs_attribute *attr, + const char *page, size_t len) +{ + struct stp_policy_node *policy_node = to_stp_policy_node(item); + struct stp_policy_node_attribute *pn_attr = + container_of(attr, struct stp_policy_node_attribute, attr); + ssize_t count = -EINVAL; + + if (pn_attr->store) + count = pn_attr->store(policy_node, page, len); + + return count; +} + +static struct configfs_item_operations stp_policy_node_item_ops = { + .release = stp_policy_node_release, + .show_attribute = stp_policy_node_attr_show, + .store_attribute = stp_policy_node_attr_store, +}; + +static struct stp_policy_node_attribute stp_policy_node_attr_range = { + .attr = { + .ca_owner = THIS_MODULE, + .ca_name = "masters", + .ca_mode = S_IRUGO | S_IWUSR, + }, + .show = stp_policy_node_masters_show, + .store = stp_policy_node_masters_store, +}; + +static struct stp_policy_node_attribute stp_policy_node_attr_channels = { + .attr = { + .ca_owner = THIS_MODULE, + .ca_name = "channels", + .ca_mode = S_IRUGO | S_IWUSR, + }, + .show = stp_policy_node_channels_show, + .store = stp_policy_node_channels_store, +}; + +static struct configfs_attribute *stp_policy_node_attrs[] = { + &stp_policy_node_attr_range.attr, + &stp_policy_node_attr_channels.attr, + NULL, +}; + +static struct config_item_type stp_policy_type; +static struct config_item_type stp_policy_node_type; + +static struct config_group * +stp_policy_node_make(struct config_group *group, const char *name) +{ + struct stp_policy_node *policy_node, *parent_node; + struct stp_policy *policy; + + if (group->cg_item.ci_type == &stp_policy_type) { + policy = container_of(group, struct stp_policy, group); + } else { + parent_node = container_of(group, struct stp_policy_node, + group); + policy = parent_node->policy; + } + + if (!policy->stm) + return ERR_PTR(-ENODEV); + + policy_node = kzalloc(sizeof(struct stp_policy_node), GFP_KERNEL); + if (!policy_node) + return ERR_PTR(-ENOMEM); + + config_group_init_type_name(&policy_node->group, name, + &stp_policy_node_type); + + policy_node->policy = policy; + + /* default values for the attributes */ + policy_node->first_master = policy->stm->data->sw_start; + policy_node->last_master = policy->stm->data->sw_end; + policy_node->first_channel = 0; + policy_node->last_channel = policy->stm->data->sw_nchannels - 1; + + return &policy_node->group; +} + +static void +stp_policy_node_drop(struct config_group *group, struct config_item *item) +{ + config_item_put(item); +} + +static struct configfs_group_operations stp_policy_node_group_ops = { + .make_group = stp_policy_node_make, + .drop_item = stp_policy_node_drop, +}; + +static struct config_item_type stp_policy_node_type = { + .ct_item_ops = &stp_policy_node_item_ops, + .ct_group_ops = &stp_policy_node_group_ops, + .ct_attrs = stp_policy_node_attrs, + .ct_owner = THIS_MODULE, +}; + +/* + * Root group: policies. + */ +static struct configfs_attribute stp_policy_attr_device = { + .ca_owner = THIS_MODULE, + .ca_name = "device", + .ca_mode = S_IRUGO, +}; + +static struct configfs_attribute *stp_policy_attrs[] = { + &stp_policy_attr_device, + NULL, +}; + +static ssize_t stp_policy_attr_show(struct config_item *item, + struct configfs_attribute *attr, + char *page) +{ + struct stp_policy *policy = to_stp_policy(item); + ssize_t count; + + count = sprintf(page, "%s\n", + (policy && policy->stm) ? + policy->stm->data->name : + "<none>"); + + return count; +} + +void stp_policy_unbind(struct stp_policy *policy) +{ + struct stm_device *stm = policy->stm; + + if (WARN_ON_ONCE(!policy->stm)) + return; + + mutex_lock(&stm->policy_mutex); + stm->policy = NULL; + mutex_unlock(&stm->policy_mutex); + + policy->stm = NULL; + + stm_put_device(stm); +} + +static void stp_policy_release(struct config_item *item) +{ + struct stp_policy *policy = to_stp_policy(item); + + stp_policy_unbind(policy); + kfree(policy); +} + +static struct configfs_item_operations stp_policy_item_ops = { + .release = stp_policy_release, + .show_attribute = stp_policy_attr_show, +}; + +static struct configfs_group_operations stp_policy_group_ops = { + .make_group = stp_policy_node_make, +}; + +static struct config_item_type stp_policy_type = { + .ct_item_ops = &stp_policy_item_ops, + .ct_group_ops = &stp_policy_group_ops, + .ct_attrs = stp_policy_attrs, + .ct_owner = THIS_MODULE, +}; + +static struct config_group * +stp_policies_make(struct config_group *group, const char *name) +{ + struct config_group *ret; + struct stm_device *stm; + char *devname, *p; + + devname = kasprintf(GFP_KERNEL, "%s", name); + if (!devname) + return ERR_PTR(-ENOMEM); + + /* + * node must look like <device_name>.<policy_name>, where + * <device_name> is the name of an existing stm device and + * <policy_name> is an arbitrary string + */ + p = strchr(devname, '.'); + if (!p) { + kfree(devname); + return ERR_PTR(-EINVAL); + } + + *p++ = '\0'; + + stm = stm_find_device(devname); + kfree(devname); + + if (!stm) + return ERR_PTR(-ENODEV); + + mutex_lock(&stm->policy_mutex); + if (stm->policy) { + ret = ERR_PTR(-EBUSY); + goto unlock_policy; + } + + stm->policy = kzalloc(sizeof(*stm->policy), GFP_KERNEL); + if (!stm->policy) { + ret = ERR_PTR(-ENOMEM); + goto unlock_policy; + } + + config_group_init_type_name(&stm->policy->group, name, + &stp_policy_type); + stm->policy->stm = stm; + + ret = &stm->policy->group; + +unlock_policy: + mutex_unlock(&stm->policy_mutex); + + if (IS_ERR(ret)) + stm_put_device(stm); + + return ret; +} + +static struct configfs_group_operations stp_policies_group_ops = { + .make_group = stp_policies_make, +}; + +static struct config_item_type stp_policies_type = { + .ct_group_ops = &stp_policies_group_ops, + .ct_owner = THIS_MODULE, +}; + +static struct configfs_subsystem stp_policy_subsys = { + .su_group = { + .cg_item = { + .ci_namebuf = "stp-policy", + .ci_type = &stp_policies_type, + }, + }, +}; + +/* + * Lock the policy mutex from the outside + */ +static struct stp_policy_node * +__stp_policy_node_lookup(struct stp_policy *policy, char *s) +{ + struct stp_policy_node *policy_node, *ret; + struct list_head *head = &policy->group.cg_children; + struct config_item *item; + char *start, *end = s; + + if (list_empty(head)) + return NULL; + + /* return the first entry if everything else fails */ + item = list_entry(head->next, struct config_item, ci_entry); + ret = to_stp_policy_node(item); + +next: + for (;;) { + start = strsep(&end, "/"); + if (!start) + break; + + if (!*start) + continue; + + list_for_each_entry(item, head, ci_entry) { + policy_node = to_stp_policy_node(item); + + if (!strcmp(start, + policy_node->group.cg_item.ci_name)) { + ret = policy_node; + + if (!end) + goto out; + + head = &policy_node->group.cg_children; + goto next; + } + } + break; + } + +out: + return ret; +} + + +struct stp_policy_node * +stp_policy_node_lookup(struct stm_device *stm, char *s) +{ + struct stp_policy_node *policy_node = NULL; + + mutex_lock(&stp_policy_subsys.su_mutex); + + mutex_lock(&stm->policy_mutex); + if (stm->policy) + policy_node = __stp_policy_node_lookup(stm->policy, s); + mutex_unlock(&stm->policy_mutex); + + if (policy_node) + config_item_get(&policy_node->group.cg_item); + mutex_unlock(&stp_policy_subsys.su_mutex); + + return policy_node; +} + +void stp_policy_node_put(struct stp_policy_node *policy_node) +{ + config_item_put(&policy_node->group.cg_item); +} + +int __init stp_configfs_init(void) +{ + int err; + + config_group_init(&stp_policy_subsys.su_group); + mutex_init(&stp_policy_subsys.su_mutex); + err = configfs_register_subsystem(&stp_policy_subsys); + + return err; +} + +void __exit stp_configfs_exit(void) +{ + configfs_unregister_subsystem(&stp_policy_subsys); +} diff --git a/drivers/hwtracing/stm/stm.h b/drivers/hwtracing/stm/stm.h new file mode 100644 index 000000000000..95ece0292c99 --- /dev/null +++ b/drivers/hwtracing/stm/stm.h @@ -0,0 +1,87 @@ +/* + * System Trace Module (STM) infrastructure + * Copyright (c) 2014, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * STM class implements generic infrastructure for System Trace Module devices + * as defined in MIPI STPv2 specification. + */ + +#ifndef _STM_STM_H_ +#define _STM_STM_H_ + +struct stp_policy; +struct stp_policy_node; + +struct stp_policy_node * +stp_policy_node_lookup(struct stm_device *stm, char *s); +void stp_policy_node_put(struct stp_policy_node *policy_node); +void stp_policy_unbind(struct stp_policy *policy); + +void stp_policy_node_get_ranges(struct stp_policy_node *policy_node, + unsigned int *mstart, unsigned int *mend, + unsigned int *cstart, unsigned int *cend); +int stp_configfs_init(void); +void stp_configfs_exit(void); + +struct stp_master { + unsigned int nr_free; + unsigned long chan_map[0]; +}; + +struct stm_device { + struct device dev; + struct module *owner; + struct stp_policy *policy; + struct mutex policy_mutex; + int major; + unsigned int sw_nmasters; + struct stm_data *data; + spinlock_t link_lock; + struct list_head link_list; + /* master allocation */ + spinlock_t mc_lock; + struct stp_master *masters[0]; +}; + +#define to_stm_device(_d) \ + container_of((_d), struct stm_device, dev) + +struct stm_output { + unsigned int master; + unsigned int channel; + unsigned int nr_chans; +}; + +struct stm_file { + struct stm_device *stm; + struct stp_policy_node *policy_node; + struct stm_output output; +}; + +struct stm_device *stm_find_device(const char *name); +void stm_put_device(struct stm_device *stm); + +struct stm_source_device { + struct device dev; + struct stm_source_data *data; + spinlock_t link_lock; + struct stm_device __rcu *link; + struct list_head link_entry; + /* one output per stm_source device */ + struct stp_policy_node *policy_node; + struct stm_output output; +}; + +#define to_stm_source_device(_d) \ + container_of((_d), struct stm_source_device, dev) + +#endif /* _STM_STM_H_ */ diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c index 3a3738fe016b..cd4510a63375 100644 --- a/drivers/idle/intel_idle.c +++ b/drivers/idle/intel_idle.c @@ -620,7 +620,7 @@ static struct cpuidle_state skl_cstates[] = { .name = "C6-SKL", .desc = "MWAIT 0x20", .flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TLB_FLUSHED, - .exit_latency = 75, + .exit_latency = 85, .target_residency = 200, .enter = &intel_idle, .enter_freeze = intel_idle_freeze, }, @@ -636,11 +636,19 @@ static struct cpuidle_state skl_cstates[] = { .name = "C8-SKL", .desc = "MWAIT 0x40", .flags = MWAIT2flg(0x40) | CPUIDLE_FLAG_TLB_FLUSHED, - .exit_latency = 174, + .exit_latency = 200, .target_residency = 800, .enter = &intel_idle, .enter_freeze = intel_idle_freeze, }, { + .name = "C9-SKL", + .desc = "MWAIT 0x50", + .flags = MWAIT2flg(0x50) | CPUIDLE_FLAG_TLB_FLUSHED, + .exit_latency = 480, + .target_residency = 5000, + .enter = &intel_idle, + .enter_freeze = intel_idle_freeze, }, + { .name = "C10-SKL", .desc = "MWAIT 0x60", .flags = MWAIT2flg(0x60) | CPUIDLE_FLAG_TLB_FLUSHED, diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index 41d6911e244e..f1ccd40beae9 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -245,7 +245,6 @@ static int mlx5_ib_query_device(struct ib_device *ibdev, props->device_cap_flags |= IB_DEVICE_BAD_QKEY_CNTR; if (MLX5_CAP_GEN(mdev, apm)) props->device_cap_flags |= IB_DEVICE_AUTO_PATH_MIG; - props->device_cap_flags |= IB_DEVICE_LOCAL_DMA_LKEY; if (MLX5_CAP_GEN(mdev, xrc)) props->device_cap_flags |= IB_DEVICE_XRC; props->device_cap_flags |= IB_DEVICE_MEM_MGT_EXTENSIONS; @@ -795,53 +794,6 @@ static int mlx5_ib_mmap(struct ib_ucontext *ibcontext, struct vm_area_struct *vm return 0; } -static int alloc_pa_mkey(struct mlx5_ib_dev *dev, u32 *key, u32 pdn) -{ - struct mlx5_create_mkey_mbox_in *in; - struct mlx5_mkey_seg *seg; - struct mlx5_core_mr mr; - int err; - - in = kzalloc(sizeof(*in), GFP_KERNEL); - if (!in) - return -ENOMEM; - - seg = &in->seg; - seg->flags = MLX5_PERM_LOCAL_READ | MLX5_ACCESS_MODE_PA; - seg->flags_pd = cpu_to_be32(pdn | MLX5_MKEY_LEN64); - seg->qpn_mkey7_0 = cpu_to_be32(0xffffff << 8); - seg->start_addr = 0; - - err = mlx5_core_create_mkey(dev->mdev, &mr, in, sizeof(*in), - NULL, NULL, NULL); - if (err) { - mlx5_ib_warn(dev, "failed to create mkey, %d\n", err); - goto err_in; - } - - kfree(in); - *key = mr.key; - - return 0; - -err_in: - kfree(in); - - return err; -} - -static void free_pa_mkey(struct mlx5_ib_dev *dev, u32 key) -{ - struct mlx5_core_mr mr; - int err; - - memset(&mr, 0, sizeof(mr)); - mr.key = key; - err = mlx5_core_destroy_mkey(dev->mdev, &mr); - if (err) - mlx5_ib_warn(dev, "failed to destroy mkey 0x%x\n", key); -} - static struct ib_pd *mlx5_ib_alloc_pd(struct ib_device *ibdev, struct ib_ucontext *context, struct ib_udata *udata) @@ -867,13 +819,6 @@ static struct ib_pd *mlx5_ib_alloc_pd(struct ib_device *ibdev, kfree(pd); return ERR_PTR(-EFAULT); } - } else { - err = alloc_pa_mkey(to_mdev(ibdev), &pd->pa_lkey, pd->pdn); - if (err) { - mlx5_core_dealloc_pd(to_mdev(ibdev)->mdev, pd->pdn); - kfree(pd); - return ERR_PTR(err); - } } return &pd->ibpd; @@ -884,9 +829,6 @@ static int mlx5_ib_dealloc_pd(struct ib_pd *pd) struct mlx5_ib_dev *mdev = to_mdev(pd->device); struct mlx5_ib_pd *mpd = to_mpd(pd); - if (!pd->uobject) - free_pa_mkey(mdev, mpd->pa_lkey); - mlx5_core_dealloc_pd(mdev->mdev, mpd->pdn); kfree(mpd); @@ -1245,18 +1187,10 @@ static int create_dev_resources(struct mlx5_ib_resources *devr) struct ib_srq_init_attr attr; struct mlx5_ib_dev *dev; struct ib_cq_init_attr cq_attr = {.cqe = 1}; - u32 rsvd_lkey; int ret = 0; dev = container_of(devr, struct mlx5_ib_dev, devr); - ret = mlx5_core_query_special_context(dev->mdev, &rsvd_lkey); - if (ret) { - pr_err("Failed to query special context %d\n", ret); - return ret; - } - dev->ib_dev.local_dma_lkey = rsvd_lkey; - devr->p0 = mlx5_ib_alloc_pd(&dev->ib_dev, NULL, NULL); if (IS_ERR(devr->p0)) { ret = PTR_ERR(devr->p0); @@ -1418,6 +1352,7 @@ static void *mlx5_ib_add(struct mlx5_core_dev *mdev) strlcpy(dev->ib_dev.name, "mlx5_%d", IB_DEVICE_NAME_MAX); dev->ib_dev.owner = THIS_MODULE; dev->ib_dev.node_type = RDMA_NODE_IB_CA; + dev->ib_dev.local_dma_lkey = 0 /* not supported for now */; dev->num_ports = MLX5_CAP_GEN(mdev, num_ports); dev->ib_dev.phys_port_cnt = dev->num_ports; dev->ib_dev.num_comp_vectors = diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h index bb8cda79e881..22123b79d550 100644 --- a/drivers/infiniband/hw/mlx5/mlx5_ib.h +++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h @@ -103,7 +103,6 @@ static inline struct mlx5_ib_ucontext *to_mucontext(struct ib_ucontext *ibuconte struct mlx5_ib_pd { struct ib_pd ibpd; u32 pdn; - u32 pa_lkey; }; /* Use macros here so that don't have to duplicate @@ -213,7 +212,6 @@ struct mlx5_ib_qp { int uuarn; int create_type; - u32 pa_lkey; /* Store signature errors */ bool signature_en; diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index c745c6c5e10d..6f521a3418e8 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -925,8 +925,6 @@ static int create_qp_common(struct mlx5_ib_dev *dev, struct ib_pd *pd, err = create_kernel_qp(dev, init_attr, qp, &in, &inlen); if (err) mlx5_ib_dbg(dev, "err %d\n", err); - else - qp->pa_lkey = to_mpd(pd)->pa_lkey; } if (err) @@ -2045,7 +2043,7 @@ static void set_frwr_pages(struct mlx5_wqe_data_seg *dseg, mfrpl->mapped_page_list[i] = cpu_to_be64(page_list[i] | perm); dseg->addr = cpu_to_be64(mfrpl->map); dseg->byte_count = cpu_to_be32(ALIGN(sizeof(u64) * wr->wr.fast_reg.page_list_len, 64)); - dseg->lkey = cpu_to_be32(pd->pa_lkey); + dseg->lkey = cpu_to_be32(pd->ibpd.local_dma_lkey); } static __be32 send_ieth(struct ib_send_wr *wr) diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h index ca2873698d75..4cd5428a2399 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib.h +++ b/drivers/infiniband/ulp/ipoib/ipoib.h @@ -80,7 +80,7 @@ enum { IPOIB_NUM_WC = 4, IPOIB_MAX_PATH_REC_QUEUE = 3, - IPOIB_MAX_MCAST_QUEUE = 3, + IPOIB_MAX_MCAST_QUEUE = 64, IPOIB_FLAG_OPER_UP = 0, IPOIB_FLAG_INITIALIZED = 1, @@ -548,6 +548,8 @@ void ipoib_path_iter_read(struct ipoib_path_iter *iter, int ipoib_mcast_attach(struct net_device *dev, u16 mlid, union ib_gid *mgid, int set_qkey); +int ipoib_mcast_leave(struct net_device *dev, struct ipoib_mcast *mcast); +struct ipoib_mcast *__ipoib_mcast_find(struct net_device *dev, void *mgid); int ipoib_init_qp(struct net_device *dev); int ipoib_transport_dev_init(struct net_device *dev, struct ib_device *ca); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index 36536ce5a3e2..f74316e679d2 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -1149,6 +1149,9 @@ static void __ipoib_reap_neigh(struct ipoib_dev_priv *priv) unsigned long dt; unsigned long flags; int i; + LIST_HEAD(remove_list); + struct ipoib_mcast *mcast, *tmcast; + struct net_device *dev = priv->dev; if (test_bit(IPOIB_STOP_NEIGH_GC, &priv->flags)) return; @@ -1176,6 +1179,19 @@ static void __ipoib_reap_neigh(struct ipoib_dev_priv *priv) lockdep_is_held(&priv->lock))) != NULL) { /* was the neigh idle for two GC periods */ if (time_after(neigh_obsolete, neigh->alive)) { + u8 *mgid = neigh->daddr + 4; + + /* Is this multicast ? */ + if (*mgid == 0xff) { + mcast = __ipoib_mcast_find(dev, mgid); + + if (mcast && test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags)) { + list_del(&mcast->list); + rb_erase(&mcast->rb_node, &priv->multicast_tree); + list_add_tail(&mcast->list, &remove_list); + } + } + rcu_assign_pointer(*np, rcu_dereference_protected(neigh->hnext, lockdep_is_held(&priv->lock))); @@ -1191,6 +1207,8 @@ static void __ipoib_reap_neigh(struct ipoib_dev_priv *priv) out_unlock: spin_unlock_irqrestore(&priv->lock, flags); + list_for_each_entry_safe(mcast, tmcast, &remove_list, list) + ipoib_mcast_leave(dev, mcast); } static void ipoib_reap_neigh(struct work_struct *work) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c index 09a1748f9d13..136cbefe00f8 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c @@ -153,7 +153,7 @@ static struct ipoib_mcast *ipoib_mcast_alloc(struct net_device *dev, return mcast; } -static struct ipoib_mcast *__ipoib_mcast_find(struct net_device *dev, void *mgid) +struct ipoib_mcast *__ipoib_mcast_find(struct net_device *dev, void *mgid) { struct ipoib_dev_priv *priv = netdev_priv(dev); struct rb_node *n = priv->multicast_tree.rb_node; @@ -508,17 +508,19 @@ static void ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast) rec.hop_limit = priv->broadcast->mcmember.hop_limit; /* - * Historically Linux IPoIB has never properly supported SEND - * ONLY join. It emulated it by not providing all the required - * attributes, which is enough to prevent group creation and - * detect if there are full members or not. A major problem - * with supporting SEND ONLY is detecting when the group is - * auto-destroyed as IPoIB will cache the MLID.. + * Send-only IB Multicast joins do not work at the core + * IB layer yet, so we can't use them here. However, + * we are emulating an Ethernet multicast send, which + * does not require a multicast subscription and will + * still send properly. The most appropriate thing to + * do is to create the group if it doesn't exist as that + * most closely emulates the behavior, from a user space + * application perspecitive, of Ethernet multicast + * operation. For now, we do a full join, maybe later + * when the core IB layers support send only joins we + * will use them. */ -#if 1 - if (test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags)) - comp_mask &= ~IB_SA_MCMEMBER_REC_TRAFFIC_CLASS; -#else +#if 0 if (test_bit(IPOIB_MCAST_FLAG_SENDONLY, &mcast->flags)) rec.join_state = 4; #endif @@ -675,7 +677,7 @@ int ipoib_mcast_stop_thread(struct net_device *dev) return 0; } -static int ipoib_mcast_leave(struct net_device *dev, struct ipoib_mcast *mcast) +int ipoib_mcast_leave(struct net_device *dev, struct ipoib_mcast *mcast) { struct ipoib_dev_priv *priv = netdev_priv(dev); int ret = 0; diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c index 1ace5d83a4d7..f58ff96b6cbb 100644 --- a/drivers/infiniband/ulp/iser/iscsi_iser.c +++ b/drivers/infiniband/ulp/iser/iscsi_iser.c @@ -97,6 +97,11 @@ unsigned int iser_max_sectors = ISER_DEF_MAX_SECTORS; module_param_named(max_sectors, iser_max_sectors, uint, S_IRUGO | S_IWUSR); MODULE_PARM_DESC(max_sectors, "Max number of sectors in a single scsi command (default:1024"); +bool iser_always_reg = true; +module_param_named(always_register, iser_always_reg, bool, S_IRUGO); +MODULE_PARM_DESC(always_register, + "Always register memory, even for continuous memory regions (default:true)"); + bool iser_pi_enable = false; module_param_named(pi_enable, iser_pi_enable, bool, S_IRUGO); MODULE_PARM_DESC(pi_enable, "Enable T10-PI offload support (default:disabled)"); diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.h b/drivers/infiniband/ulp/iser/iscsi_iser.h index 86f6583485ef..a5edd6ede692 100644 --- a/drivers/infiniband/ulp/iser/iscsi_iser.h +++ b/drivers/infiniband/ulp/iser/iscsi_iser.h @@ -611,6 +611,7 @@ extern int iser_debug_level; extern bool iser_pi_enable; extern int iser_pi_guard; extern unsigned int iser_max_sectors; +extern bool iser_always_reg; int iser_assign_reg_ops(struct iser_device *device); diff --git a/drivers/infiniband/ulp/iser/iser_memory.c b/drivers/infiniband/ulp/iser/iser_memory.c index 2493cc748db8..4c46d67d37a1 100644 --- a/drivers/infiniband/ulp/iser/iser_memory.c +++ b/drivers/infiniband/ulp/iser/iser_memory.c @@ -803,11 +803,12 @@ static int iser_reg_prot_sg(struct iscsi_iser_task *task, struct iser_data_buf *mem, struct iser_fr_desc *desc, + bool use_dma_key, struct iser_mem_reg *reg) { struct iser_device *device = task->iser_conn->ib_conn.device; - if (mem->dma_nents == 1) + if (use_dma_key) return iser_reg_dma(device, mem, reg); return device->reg_ops->reg_mem(task, mem, &desc->pi_ctx->rsc, reg); @@ -817,11 +818,12 @@ static int iser_reg_data_sg(struct iscsi_iser_task *task, struct iser_data_buf *mem, struct iser_fr_desc *desc, + bool use_dma_key, struct iser_mem_reg *reg) { struct iser_device *device = task->iser_conn->ib_conn.device; - if (mem->dma_nents == 1) + if (use_dma_key) return iser_reg_dma(device, mem, reg); return device->reg_ops->reg_mem(task, mem, &desc->rsc, reg); @@ -836,14 +838,17 @@ int iser_reg_rdma_mem(struct iscsi_iser_task *task, struct iser_mem_reg *reg = &task->rdma_reg[dir]; struct iser_mem_reg *data_reg; struct iser_fr_desc *desc = NULL; + bool use_dma_key; int err; err = iser_handle_unaligned_buf(task, mem, dir); if (unlikely(err)) return err; - if (mem->dma_nents != 1 || - scsi_get_prot_op(task->sc) != SCSI_PROT_NORMAL) { + use_dma_key = (mem->dma_nents == 1 && !iser_always_reg && + scsi_get_prot_op(task->sc) == SCSI_PROT_NORMAL); + + if (!use_dma_key) { desc = device->reg_ops->reg_desc_get(ib_conn); reg->mem_h = desc; } @@ -853,7 +858,7 @@ int iser_reg_rdma_mem(struct iscsi_iser_task *task, else data_reg = &task->desc.data_reg; - err = iser_reg_data_sg(task, mem, desc, data_reg); + err = iser_reg_data_sg(task, mem, desc, use_dma_key, data_reg); if (unlikely(err)) goto err_reg; @@ -866,7 +871,8 @@ int iser_reg_rdma_mem(struct iscsi_iser_task *task, if (unlikely(err)) goto err_reg; - err = iser_reg_prot_sg(task, mem, desc, prot_reg); + err = iser_reg_prot_sg(task, mem, desc, + use_dma_key, prot_reg); if (unlikely(err)) goto err_reg; } diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c b/drivers/infiniband/ulp/iser/iser_verbs.c index ae70cc1463ac..85132d867bc8 100644 --- a/drivers/infiniband/ulp/iser/iser_verbs.c +++ b/drivers/infiniband/ulp/iser/iser_verbs.c @@ -133,11 +133,15 @@ static int iser_create_device_ib_res(struct iser_device *device) (unsigned long)comp); } - device->mr = ib_get_dma_mr(device->pd, IB_ACCESS_LOCAL_WRITE | - IB_ACCESS_REMOTE_WRITE | - IB_ACCESS_REMOTE_READ); - if (IS_ERR(device->mr)) - goto dma_mr_err; + if (!iser_always_reg) { + int access = IB_ACCESS_LOCAL_WRITE | + IB_ACCESS_REMOTE_WRITE | + IB_ACCESS_REMOTE_READ; + + device->mr = ib_get_dma_mr(device->pd, access); + if (IS_ERR(device->mr)) + goto dma_mr_err; + } INIT_IB_EVENT_HANDLER(&device->event_handler, device->ib_device, iser_event_handler); @@ -147,7 +151,8 @@ static int iser_create_device_ib_res(struct iser_device *device) return 0; handler_err: - ib_dereg_mr(device->mr); + if (device->mr) + ib_dereg_mr(device->mr); dma_mr_err: for (i = 0; i < device->comps_used; i++) tasklet_kill(&device->comps[i].tasklet); @@ -173,7 +178,6 @@ comps_err: static void iser_free_device_ib_res(struct iser_device *device) { int i; - BUG_ON(device->mr == NULL); for (i = 0; i < device->comps_used; i++) { struct iser_comp *comp = &device->comps[i]; @@ -184,7 +188,8 @@ static void iser_free_device_ib_res(struct iser_device *device) } (void)ib_unregister_event_handler(&device->event_handler); - (void)ib_dereg_mr(device->mr); + if (device->mr) + (void)ib_dereg_mr(device->mr); ib_dealloc_pd(device->pd); kfree(device->comps); diff --git a/drivers/input/joystick/Kconfig b/drivers/input/joystick/Kconfig index 56eb471b5576..4215b5382092 100644 --- a/drivers/input/joystick/Kconfig +++ b/drivers/input/joystick/Kconfig @@ -196,6 +196,7 @@ config JOYSTICK_TWIDJOY config JOYSTICK_ZHENHUA tristate "5-byte Zhenhua RC transmitter" select SERIO + select BITREVERSE help Say Y here if you have a Zhen Hua PPM-4CH transmitter which is supplied with a ready to fly micro electric indoor helicopters diff --git a/drivers/input/joystick/walkera0701.c b/drivers/input/joystick/walkera0701.c index b76ac580703c..a8bc2fe170dd 100644 --- a/drivers/input/joystick/walkera0701.c +++ b/drivers/input/joystick/walkera0701.c @@ -150,7 +150,7 @@ static void walkera0701_irq_handler(void *handler_data) if (w->counter == 24) { /* full frame */ walkera0701_parse_frame(w); w->counter = NO_SYNC; - if (abs(pulse_time - SYNC_PULSE) < RESERVE) /* new frame sync */ + if (abs64(pulse_time - SYNC_PULSE) < RESERVE) /* new frame sync */ w->counter = 0; } else { if ((pulse_time > (ANALOG_MIN_PULSE - RESERVE) @@ -161,7 +161,7 @@ static void walkera0701_irq_handler(void *handler_data) } else w->counter = NO_SYNC; } - } else if (abs(pulse_time - SYNC_PULSE - BIN0_PULSE) < + } else if (abs64(pulse_time - SYNC_PULSE - BIN0_PULSE) < RESERVE + BIN1_PULSE - BIN0_PULSE) /* frame sync .. */ w->counter = 0; diff --git a/drivers/input/keyboard/omap4-keypad.c b/drivers/input/keyboard/omap4-keypad.c index b052afec9a11..6639b2b8528a 100644 --- a/drivers/input/keyboard/omap4-keypad.c +++ b/drivers/input/keyboard/omap4-keypad.c @@ -266,7 +266,7 @@ static int omap4_keypad_probe(struct platform_device *pdev) error = omap4_keypad_parse_dt(&pdev->dev, keypad_data); if (error) - return error; + goto err_free_keypad; res = request_mem_region(res->start, resource_size(res), pdev->name); if (!res) { diff --git a/drivers/input/misc/pm8941-pwrkey.c b/drivers/input/misc/pm8941-pwrkey.c index 867db8a91372..e317b75357a0 100644 --- a/drivers/input/misc/pm8941-pwrkey.c +++ b/drivers/input/misc/pm8941-pwrkey.c @@ -93,7 +93,7 @@ static int pm8941_reboot_notify(struct notifier_block *nb, default: reset_type = PON_PS_HOLD_TYPE_HARD_RESET; break; - }; + } error = regmap_update_bits(pwrkey->regmap, pwrkey->baseaddr + PON_PS_HOLD_RST_CTL, diff --git a/drivers/input/misc/uinput.c b/drivers/input/misc/uinput.c index 345df9b03aed..5adbcedcb81c 100644 --- a/drivers/input/misc/uinput.c +++ b/drivers/input/misc/uinput.c @@ -414,7 +414,7 @@ static int uinput_setup_device(struct uinput_device *udev, dev->id.product = user_dev->id.product; dev->id.version = user_dev->id.version; - for_each_set_bit(i, dev->absbit, ABS_CNT) { + for (i = 0; i < ABS_CNT; i++) { input_abs_set_max(dev, i, user_dev->absmax[i]); input_abs_set_min(dev, i, user_dev->absmin[i]); input_abs_set_fuzz(dev, i, user_dev->absfuzz[i]); diff --git a/drivers/input/mouse/elan_i2c.h b/drivers/input/mouse/elan_i2c.h index 73670f2aebfd..c0ec26118732 100644 --- a/drivers/input/mouse/elan_i2c.h +++ b/drivers/input/mouse/elan_i2c.h @@ -60,7 +60,7 @@ struct elan_transport_ops { int (*get_sm_version)(struct i2c_client *client, u8* ic_type, u8 *version); int (*get_checksum)(struct i2c_client *client, bool iap, u16 *csum); - int (*get_product_id)(struct i2c_client *client, u8 *id); + int (*get_product_id)(struct i2c_client *client, u16 *id); int (*get_max)(struct i2c_client *client, unsigned int *max_x, unsigned int *max_y); diff --git a/drivers/input/mouse/elan_i2c_core.c b/drivers/input/mouse/elan_i2c_core.c index fa945304b9a5..5e1665bbaa0b 100644 --- a/drivers/input/mouse/elan_i2c_core.c +++ b/drivers/input/mouse/elan_i2c_core.c @@ -40,7 +40,7 @@ #include "elan_i2c.h" #define DRIVER_NAME "elan_i2c" -#define ELAN_DRIVER_VERSION "1.6.0" +#define ELAN_DRIVER_VERSION "1.6.1" #define ETP_MAX_PRESSURE 255 #define ETP_FWIDTH_REDUCE 90 #define ETP_FINGER_WIDTH 15 @@ -76,7 +76,7 @@ struct elan_tp_data { unsigned int x_res; unsigned int y_res; - u8 product_id; + u16 product_id; u8 fw_version; u8 sm_version; u8 iap_version; @@ -98,15 +98,25 @@ static int elan_get_fwinfo(u8 iap_version, u16 *validpage_count, u16 *signature_address) { switch (iap_version) { + case 0x00: + case 0x06: case 0x08: *validpage_count = 512; break; + case 0x03: + case 0x07: case 0x09: + case 0x0A: + case 0x0B: + case 0x0C: *validpage_count = 768; break; case 0x0D: *validpage_count = 896; break; + case 0x0E: + *validpage_count = 640; + break; default: /* unknown ic type clear value */ *validpage_count = 0; @@ -266,11 +276,10 @@ static int elan_query_device_info(struct elan_tp_data *data) error = elan_get_fwinfo(data->iap_version, &data->fw_validpage_count, &data->fw_signature_address); - if (error) { - dev_err(&data->client->dev, - "unknown iap version %d\n", data->iap_version); - return error; - } + if (error) + dev_warn(&data->client->dev, + "unexpected iap version %#04x (ic type: %#04x), firmware update will not work\n", + data->iap_version, data->ic_type); return 0; } @@ -486,6 +495,9 @@ static ssize_t elan_sysfs_update_fw(struct device *dev, const u8 *fw_signature; static const u8 signature[] = {0xAA, 0x55, 0xCC, 0x33, 0xFF, 0xFF}; + if (data->fw_validpage_count == 0) + return -EINVAL; + /* Look for a firmware with the product id appended. */ fw_name = kasprintf(GFP_KERNEL, ETP_FW_NAME, data->product_id); if (!fw_name) { diff --git a/drivers/input/mouse/elan_i2c_i2c.c b/drivers/input/mouse/elan_i2c_i2c.c index 683c840c9dd7..a679e56c44cd 100644 --- a/drivers/input/mouse/elan_i2c_i2c.c +++ b/drivers/input/mouse/elan_i2c_i2c.c @@ -276,7 +276,7 @@ static int elan_i2c_get_sm_version(struct i2c_client *client, return 0; } -static int elan_i2c_get_product_id(struct i2c_client *client, u8 *id) +static int elan_i2c_get_product_id(struct i2c_client *client, u16 *id) { int error; u8 val[3]; @@ -287,7 +287,7 @@ static int elan_i2c_get_product_id(struct i2c_client *client, u8 *id) return error; } - *id = val[0]; + *id = le16_to_cpup((__le16 *)val); return 0; } diff --git a/drivers/input/mouse/elan_i2c_smbus.c b/drivers/input/mouse/elan_i2c_smbus.c index ff36a366b2aa..cb6aecbc1dc2 100644 --- a/drivers/input/mouse/elan_i2c_smbus.c +++ b/drivers/input/mouse/elan_i2c_smbus.c @@ -183,7 +183,7 @@ static int elan_smbus_get_sm_version(struct i2c_client *client, return 0; } -static int elan_smbus_get_product_id(struct i2c_client *client, u8 *id) +static int elan_smbus_get_product_id(struct i2c_client *client, u16 *id) { int error; u8 val[3]; @@ -195,7 +195,7 @@ static int elan_smbus_get_product_id(struct i2c_client *client, u8 *id) return error; } - *id = val[1]; + *id = be16_to_cpup((__be16 *)val); return 0; } diff --git a/drivers/input/mouse/synaptics.c b/drivers/input/mouse/synaptics.c index 994ae7886156..6025eb430c0a 100644 --- a/drivers/input/mouse/synaptics.c +++ b/drivers/input/mouse/synaptics.c @@ -519,18 +519,14 @@ static int synaptics_set_mode(struct psmouse *psmouse) struct synaptics_data *priv = psmouse->private; priv->mode = 0; - - if (priv->absolute_mode) { + if (priv->absolute_mode) priv->mode |= SYN_BIT_ABSOLUTE_MODE; - if (SYN_CAP_EXTENDED(priv->capabilities)) - priv->mode |= SYN_BIT_W_MODE; - } - - if (!SYN_MODE_WMODE(priv->mode) && priv->disable_gesture) + if (priv->disable_gesture) priv->mode |= SYN_BIT_DISABLE_GESTURE; - if (psmouse->rate >= 80) priv->mode |= SYN_BIT_HIGH_RATE; + if (SYN_CAP_EXTENDED(priv->capabilities)) + priv->mode |= SYN_BIT_W_MODE; if (synaptics_mode_cmd(psmouse, priv->mode)) return -1; diff --git a/drivers/input/serio/libps2.c b/drivers/input/serio/libps2.c index 75516996db20..316f2c897101 100644 --- a/drivers/input/serio/libps2.c +++ b/drivers/input/serio/libps2.c @@ -212,12 +212,17 @@ int __ps2_command(struct ps2dev *ps2dev, unsigned char *param, int command) * time before the ACK arrives. */ if (ps2_sendbyte(ps2dev, command & 0xff, - command == PS2_CMD_RESET_BAT ? 1000 : 200)) - goto out; + command == PS2_CMD_RESET_BAT ? 1000 : 200)) { + serio_pause_rx(ps2dev->serio); + goto out_reset_flags; + } - for (i = 0; i < send; i++) - if (ps2_sendbyte(ps2dev, param[i], 200)) - goto out; + for (i = 0; i < send; i++) { + if (ps2_sendbyte(ps2dev, param[i], 200)) { + serio_pause_rx(ps2dev->serio); + goto out_reset_flags; + } + } /* * The reset command takes a long time to execute. @@ -234,17 +239,18 @@ int __ps2_command(struct ps2dev *ps2dev, unsigned char *param, int command) !(ps2dev->flags & PS2_FLAG_CMD), timeout); } + serio_pause_rx(ps2dev->serio); + if (param) for (i = 0; i < receive; i++) param[i] = ps2dev->cmdbuf[(receive - 1) - i]; if (ps2dev->cmdcnt && (command != PS2_CMD_RESET_BAT || ps2dev->cmdcnt != 1)) - goto out; + goto out_reset_flags; rc = 0; - out: - serio_pause_rx(ps2dev->serio); + out_reset_flags: ps2dev->flags = 0; serio_continue_rx(ps2dev->serio); diff --git a/drivers/input/serio/parkbd.c b/drivers/input/serio/parkbd.c index 26b45936f9fd..1e8cd6f1fe9e 100644 --- a/drivers/input/serio/parkbd.c +++ b/drivers/input/serio/parkbd.c @@ -194,6 +194,7 @@ static int __init parkbd_init(void) parkbd_port = parkbd_allocate_serio(); if (!parkbd_port) { parport_release(parkbd_dev); + parport_unregister_device(parkbd_dev); return -ENOMEM; } diff --git a/drivers/input/touchscreen/imx6ul_tsc.c b/drivers/input/touchscreen/imx6ul_tsc.c index ff0b75813daa..8275267eac25 100644 --- a/drivers/input/touchscreen/imx6ul_tsc.c +++ b/drivers/input/touchscreen/imx6ul_tsc.c @@ -94,7 +94,7 @@ struct imx6ul_tsc { * TSC module need ADC to get the measure value. So * before config TSC, we should initialize ADC module. */ -static void imx6ul_adc_init(struct imx6ul_tsc *tsc) +static int imx6ul_adc_init(struct imx6ul_tsc *tsc) { int adc_hc = 0; int adc_gc; @@ -122,17 +122,23 @@ static void imx6ul_adc_init(struct imx6ul_tsc *tsc) timeout = wait_for_completion_timeout (&tsc->completion, ADC_TIMEOUT); - if (timeout == 0) + if (timeout == 0) { dev_err(tsc->dev, "Timeout for adc calibration\n"); + return -ETIMEDOUT; + } adc_gs = readl(tsc->adc_regs + REG_ADC_GS); - if (adc_gs & ADC_CALF) + if (adc_gs & ADC_CALF) { dev_err(tsc->dev, "ADC calibration failed\n"); + return -EINVAL; + } /* TSC need the ADC work in hardware trigger */ adc_cfg = readl(tsc->adc_regs + REG_ADC_CFG); adc_cfg |= ADC_HARDWARE_TRIGGER; writel(adc_cfg, tsc->adc_regs + REG_ADC_CFG); + + return 0; } /* @@ -188,11 +194,17 @@ static void imx6ul_tsc_set(struct imx6ul_tsc *tsc) writel(start, tsc->tsc_regs + REG_TSC_FLOW_CONTROL); } -static void imx6ul_tsc_init(struct imx6ul_tsc *tsc) +static int imx6ul_tsc_init(struct imx6ul_tsc *tsc) { - imx6ul_adc_init(tsc); + int err; + + err = imx6ul_adc_init(tsc); + if (err) + return err; imx6ul_tsc_channel_config(tsc); imx6ul_tsc_set(tsc); + + return 0; } static void imx6ul_tsc_disable(struct imx6ul_tsc *tsc) @@ -311,9 +323,7 @@ static int imx6ul_tsc_open(struct input_dev *input_dev) return err; } - imx6ul_tsc_init(tsc); - - return 0; + return imx6ul_tsc_init(tsc); } static void imx6ul_tsc_close(struct input_dev *input_dev) @@ -337,7 +347,7 @@ static int imx6ul_tsc_probe(struct platform_device *pdev) int tsc_irq; int adc_irq; - tsc = devm_kzalloc(&pdev->dev, sizeof(struct imx6ul_tsc), GFP_KERNEL); + tsc = devm_kzalloc(&pdev->dev, sizeof(*tsc), GFP_KERNEL); if (!tsc) return -ENOMEM; @@ -345,7 +355,7 @@ static int imx6ul_tsc_probe(struct platform_device *pdev) if (!input_dev) return -ENOMEM; - input_dev->name = "iMX6UL TouchScreen Controller"; + input_dev->name = "iMX6UL Touchscreen Controller"; input_dev->id.bustype = BUS_HOST; input_dev->open = imx6ul_tsc_open; @@ -406,7 +416,7 @@ static int imx6ul_tsc_probe(struct platform_device *pdev) } adc_irq = platform_get_irq(pdev, 1); - if (adc_irq <= 0) { + if (adc_irq < 0) { dev_err(&pdev->dev, "no adc irq resource?\n"); return adc_irq; } @@ -491,7 +501,7 @@ static int __maybe_unused imx6ul_tsc_resume(struct device *dev) goto out; } - imx6ul_tsc_init(tsc); + retval = imx6ul_tsc_init(tsc); } out: diff --git a/drivers/input/touchscreen/mms114.c b/drivers/input/touchscreen/mms114.c index 7cce87650fc8..1fafc9f57af6 100644 --- a/drivers/input/touchscreen/mms114.c +++ b/drivers/input/touchscreen/mms114.c @@ -394,12 +394,12 @@ static struct mms114_platform_data *mms114_parse_dt(struct device *dev) if (of_property_read_u32(np, "x-size", &pdata->x_size)) { dev_err(dev, "failed to get x-size property\n"); return NULL; - }; + } if (of_property_read_u32(np, "y-size", &pdata->y_size)) { dev_err(dev, "failed to get y-size property\n"); return NULL; - }; + } of_property_read_u32(np, "contact-threshold", &pdata->contact_threshold); diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig index 4664c2a96c67..d9da766719c8 100644 --- a/drivers/iommu/Kconfig +++ b/drivers/iommu/Kconfig @@ -43,7 +43,7 @@ config IOMMU_IO_PGTABLE_LPAE_SELFTEST endmenu config IOMMU_IOVA - bool + tristate config OF_IOMMU def_bool y diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c index 2d7349a3ee14..041bc1810a86 100644 --- a/drivers/iommu/intel-iommu.c +++ b/drivers/iommu/intel-iommu.c @@ -3215,6 +3215,8 @@ static struct iova *intel_alloc_iova(struct device *dev, /* Restrict dma_mask to the width that the iommu can handle */ dma_mask = min_t(uint64_t, DOMAIN_MAX_ADDR(domain->gaw), dma_mask); + /* Ensure we reserve the whole size-aligned region */ + nrpages = __roundup_pow_of_two(nrpages); if (!dmar_forcedac && dma_mask > DMA_BIT_MASK(32)) { /* @@ -3711,7 +3713,7 @@ static inline int iommu_devinfo_cache_init(void) static int __init iommu_init_mempool(void) { int ret; - ret = iommu_iova_cache_init(); + ret = iova_cache_get(); if (ret) return ret; @@ -3725,7 +3727,7 @@ static int __init iommu_init_mempool(void) kmem_cache_destroy(iommu_domain_cache); domain_error: - iommu_iova_cache_destroy(); + iova_cache_put(); return -ENOMEM; } @@ -3734,7 +3736,7 @@ static void __init iommu_exit_mempool(void) { kmem_cache_destroy(iommu_devinfo_cache); kmem_cache_destroy(iommu_domain_cache); - iommu_iova_cache_destroy(); + iova_cache_put(); } static void quirk_ioat_snb_local_iommu(struct pci_dev *pdev) diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c index b7c3d923f3e1..fa0adef32bd6 100644 --- a/drivers/iommu/iova.c +++ b/drivers/iommu/iova.c @@ -18,42 +18,9 @@ */ #include <linux/iova.h> +#include <linux/module.h> #include <linux/slab.h> -static struct kmem_cache *iommu_iova_cache; - -int iommu_iova_cache_init(void) -{ - int ret = 0; - - iommu_iova_cache = kmem_cache_create("iommu_iova", - sizeof(struct iova), - 0, - SLAB_HWCACHE_ALIGN, - NULL); - if (!iommu_iova_cache) { - pr_err("Couldn't create iova cache\n"); - ret = -ENOMEM; - } - - return ret; -} - -void iommu_iova_cache_destroy(void) -{ - kmem_cache_destroy(iommu_iova_cache); -} - -struct iova *alloc_iova_mem(void) -{ - return kmem_cache_alloc(iommu_iova_cache, GFP_ATOMIC); -} - -void free_iova_mem(struct iova *iova) -{ - kmem_cache_free(iommu_iova_cache, iova); -} - void init_iova_domain(struct iova_domain *iovad, unsigned long granule, unsigned long start_pfn, unsigned long pfn_32bit) @@ -72,6 +39,7 @@ init_iova_domain(struct iova_domain *iovad, unsigned long granule, iovad->start_pfn = start_pfn; iovad->dma_32bit_pfn = pfn_32bit; } +EXPORT_SYMBOL_GPL(init_iova_domain); static struct rb_node * __get_cached_rbnode(struct iova_domain *iovad, unsigned long *limit_pfn) @@ -120,19 +88,14 @@ __cached_rbnode_delete_update(struct iova_domain *iovad, struct iova *free) } } -/* Computes the padding size required, to make the - * the start address naturally aligned on its size +/* + * Computes the padding size required, to make the start address + * naturally aligned on the power-of-two order of its size */ -static int -iova_get_pad_size(int size, unsigned int limit_pfn) +static unsigned int +iova_get_pad_size(unsigned int size, unsigned int limit_pfn) { - unsigned int pad_size = 0; - unsigned int order = ilog2(size); - - if (order) - pad_size = (limit_pfn + 1) % (1 << order); - - return pad_size; + return (limit_pfn + 1 - size) & (__roundup_pow_of_two(size) - 1); } static int __alloc_and_insert_iova_range(struct iova_domain *iovad, @@ -242,6 +205,57 @@ iova_insert_rbtree(struct rb_root *root, struct iova *iova) rb_insert_color(&iova->node, root); } +static struct kmem_cache *iova_cache; +static unsigned int iova_cache_users; +static DEFINE_MUTEX(iova_cache_mutex); + +struct iova *alloc_iova_mem(void) +{ + return kmem_cache_alloc(iova_cache, GFP_ATOMIC); +} +EXPORT_SYMBOL(alloc_iova_mem); + +void free_iova_mem(struct iova *iova) +{ + kmem_cache_free(iova_cache, iova); +} +EXPORT_SYMBOL(free_iova_mem); + +int iova_cache_get(void) +{ + mutex_lock(&iova_cache_mutex); + if (!iova_cache_users) { + iova_cache = kmem_cache_create( + "iommu_iova", sizeof(struct iova), 0, + SLAB_HWCACHE_ALIGN, NULL); + if (!iova_cache) { + mutex_unlock(&iova_cache_mutex); + printk(KERN_ERR "Couldn't create iova cache\n"); + return -ENOMEM; + } + } + + iova_cache_users++; + mutex_unlock(&iova_cache_mutex); + + return 0; +} +EXPORT_SYMBOL_GPL(iova_cache_get); + +void iova_cache_put(void) +{ + mutex_lock(&iova_cache_mutex); + if (WARN_ON(!iova_cache_users)) { + mutex_unlock(&iova_cache_mutex); + return; + } + iova_cache_users--; + if (!iova_cache_users) + kmem_cache_destroy(iova_cache); + mutex_unlock(&iova_cache_mutex); +} +EXPORT_SYMBOL_GPL(iova_cache_put); + /** * alloc_iova - allocates an iova * @iovad: - iova domain in question @@ -265,12 +279,6 @@ alloc_iova(struct iova_domain *iovad, unsigned long size, if (!new_iova) return NULL; - /* If size aligned is set then round the size to - * to next power of two. - */ - if (size_aligned) - size = __roundup_pow_of_two(size); - ret = __alloc_and_insert_iova_range(iovad, size, limit_pfn, new_iova, size_aligned); @@ -281,6 +289,7 @@ alloc_iova(struct iova_domain *iovad, unsigned long size, return new_iova; } +EXPORT_SYMBOL_GPL(alloc_iova); /** * find_iova - find's an iova for a given pfn @@ -321,6 +330,7 @@ struct iova *find_iova(struct iova_domain *iovad, unsigned long pfn) spin_unlock_irqrestore(&iovad->iova_rbtree_lock, flags); return NULL; } +EXPORT_SYMBOL_GPL(find_iova); /** * __free_iova - frees the given iova @@ -339,6 +349,7 @@ __free_iova(struct iova_domain *iovad, struct iova *iova) spin_unlock_irqrestore(&iovad->iova_rbtree_lock, flags); free_iova_mem(iova); } +EXPORT_SYMBOL_GPL(__free_iova); /** * free_iova - finds and frees the iova for a given pfn @@ -356,6 +367,7 @@ free_iova(struct iova_domain *iovad, unsigned long pfn) __free_iova(iovad, iova); } +EXPORT_SYMBOL_GPL(free_iova); /** * put_iova_domain - destroys the iova doamin @@ -378,6 +390,7 @@ void put_iova_domain(struct iova_domain *iovad) } spin_unlock_irqrestore(&iovad->iova_rbtree_lock, flags); } +EXPORT_SYMBOL_GPL(put_iova_domain); static int __is_range_overlap(struct rb_node *node, @@ -467,6 +480,7 @@ finish: spin_unlock_irqrestore(&iovad->iova_rbtree_lock, flags); return iova; } +EXPORT_SYMBOL_GPL(reserve_iova); /** * copy_reserved_iova - copies the reserved between domains @@ -493,6 +507,7 @@ copy_reserved_iova(struct iova_domain *from, struct iova_domain *to) } spin_unlock_irqrestore(&from->iova_rbtree_lock, flags); } +EXPORT_SYMBOL_GPL(copy_reserved_iova); struct iova * split_and_remove_iova(struct iova_domain *iovad, struct iova *iova, @@ -534,3 +549,6 @@ error: free_iova_mem(prev); return NULL; } + +MODULE_AUTHOR("Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>"); +MODULE_LICENSE("GPL"); diff --git a/drivers/irqchip/irq-gic-v3-its-pci-msi.c b/drivers/irqchip/irq-gic-v3-its-pci-msi.c index cf351c637464..a7c8c9ffbafd 100644 --- a/drivers/irqchip/irq-gic-v3-its-pci-msi.c +++ b/drivers/irqchip/irq-gic-v3-its-pci-msi.c @@ -62,7 +62,7 @@ static int its_get_pci_alias(struct pci_dev *pdev, u16 alias, void *data) dev_alias->dev_id = alias; if (pdev != dev_alias->pdev) - dev_alias->count += its_pci_msi_vec_count(dev_alias->pdev); + dev_alias->count += its_pci_msi_vec_count(pdev); return 0; } diff --git a/drivers/irqchip/irq-gic-v3-its.c b/drivers/irqchip/irq-gic-v3-its.c index ac7ae2b3cb83..25ceae9f7348 100644 --- a/drivers/irqchip/irq-gic-v3-its.c +++ b/drivers/irqchip/irq-gic-v3-its.c @@ -719,6 +719,9 @@ static unsigned long *its_lpi_alloc_chunks(int nr_irqs, int *base, int *nr_ids) out: spin_unlock(&lpi_lock); + if (!bitmap) + *base = *nr_ids = 0; + return bitmap; } diff --git a/drivers/irqchip/irq-mips-gic.c b/drivers/irqchip/irq-mips-gic.c index af2f16bb8a94..aeaa061f0dbf 100644 --- a/drivers/irqchip/irq-mips-gic.c +++ b/drivers/irqchip/irq-mips-gic.c @@ -320,6 +320,14 @@ static void gic_handle_shared_int(bool chained) intrmask[i] = gic_read(intrmask_reg); pending_reg += gic_reg_step; intrmask_reg += gic_reg_step; + + if (!config_enabled(CONFIG_64BIT) || mips_cm_is64) + continue; + + pending[i] |= (u64)gic_read(pending_reg) << 32; + intrmask[i] |= (u64)gic_read(intrmask_reg) << 32; + pending_reg += gic_reg_step; + intrmask_reg += gic_reg_step; } bitmap_and(pending, pending, intrmask, gic_shared_intrs); @@ -426,7 +434,7 @@ static int gic_set_affinity(struct irq_data *d, const struct cpumask *cpumask, spin_lock_irqsave(&gic_lock, flags); /* Re-route this IRQ */ - gic_map_to_vpe(irq, cpumask_first(&tmp)); + gic_map_to_vpe(irq, mips_cm_vp_id(cpumask_first(&tmp))); /* Update the pcpu_masks */ for (i = 0; i < NR_CPUS; i++) @@ -599,7 +607,7 @@ static __init void gic_ipi_init_one(unsigned int intr, int cpu, GIC_SHARED_TO_HWIRQ(intr)); int i; - gic_map_to_vpe(intr, cpu); + gic_map_to_vpe(intr, mips_cm_vp_id(cpu)); for (i = 0; i < NR_CPUS; i++) clear_bit(intr, pcpu_masks[i].pcpu_mask); set_bit(intr, pcpu_masks[cpu].pcpu_mask); diff --git a/drivers/mcb/mcb-pci.c b/drivers/mcb/mcb-pci.c index de36237d7c6b..051645498b53 100644 --- a/drivers/mcb/mcb-pci.c +++ b/drivers/mcb/mcb-pci.c @@ -74,7 +74,7 @@ static int mcb_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) ret = -ENOTSUPP; dev_err(&pdev->dev, "IO mapped PCI devices are not supported\n"); - goto out_release; + goto out_iounmap; } pci_set_drvdata(pdev, priv); @@ -89,7 +89,7 @@ static int mcb_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) ret = chameleon_parse_cells(priv->bus, priv->mapbase, priv->base); if (ret < 0) - goto out_iounmap; + goto out_mcb_bus; num_cells = ret; dev_dbg(&pdev->dev, "Found %d cells\n", num_cells); @@ -98,6 +98,8 @@ static int mcb_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id) return 0; +out_mcb_bus: + mcb_release_bus(priv->bus); out_iounmap: iounmap(priv->base); out_release: diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index e51de52eeb94..48b5890c28e3 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -1997,7 +1997,8 @@ int bitmap_resize(struct bitmap *bitmap, sector_t blocks, if (bitmap->mddev->bitmap_info.offset || bitmap->mddev->bitmap_info.file) ret = bitmap_storage_alloc(&store, chunks, !bitmap->mddev->bitmap_info.external, - bitmap->cluster_slot); + mddev_is_clustered(bitmap->mddev) + ? bitmap->cluster_slot : 0); if (ret) goto err; diff --git a/drivers/md/dm-cache-policy-cleaner.c b/drivers/md/dm-cache-policy-cleaner.c index 240c9f0e85e7..8a096456579b 100644 --- a/drivers/md/dm-cache-policy-cleaner.c +++ b/drivers/md/dm-cache-policy-cleaner.c @@ -436,7 +436,7 @@ static struct dm_cache_policy *wb_create(dm_cblock_t cache_size, static struct dm_cache_policy_type wb_policy_type = { .name = "cleaner", .version = {1, 0, 0}, - .hint_size = 0, + .hint_size = 4, .owner = THIS_MODULE, .create = wb_create }; diff --git a/drivers/md/dm-exception-store.c b/drivers/md/dm-exception-store.c index ebaa4f803eec..192bb8beeb6b 100644 --- a/drivers/md/dm-exception-store.c +++ b/drivers/md/dm-exception-store.c @@ -203,7 +203,7 @@ int dm_exception_store_create(struct dm_target *ti, int argc, char **argv, return -EINVAL; } - tmp_store = kmalloc(sizeof(*tmp_store), GFP_KERNEL); + tmp_store = kzalloc(sizeof(*tmp_store), GFP_KERNEL); if (!tmp_store) { ti->error = "Exception store allocation failed"; return -ENOMEM; @@ -215,7 +215,7 @@ int dm_exception_store_create(struct dm_target *ti, int argc, char **argv, else if (persistent == 'N') type = get_type("N"); else { - ti->error = "Persistent flag is not P or N"; + ti->error = "Exception store type is not P or N"; r = -EINVAL; goto bad_type; } @@ -233,7 +233,7 @@ int dm_exception_store_create(struct dm_target *ti, int argc, char **argv, if (r) goto bad; - r = type->ctr(tmp_store, 0, NULL); + r = type->ctr(tmp_store, (strlen(argv[0]) > 1 ? &argv[0][1] : NULL)); if (r) { ti->error = "Exception store type constructor failed"; goto bad; diff --git a/drivers/md/dm-exception-store.h b/drivers/md/dm-exception-store.h index 0b2536247cf5..fae34e7a0b1e 100644 --- a/drivers/md/dm-exception-store.h +++ b/drivers/md/dm-exception-store.h @@ -42,8 +42,7 @@ struct dm_exception_store_type { const char *name; struct module *module; - int (*ctr) (struct dm_exception_store *store, - unsigned argc, char **argv); + int (*ctr) (struct dm_exception_store *store, char *options); /* * Destroys this object when you've finished with it. @@ -123,6 +122,8 @@ struct dm_exception_store { unsigned chunk_shift; void *context; + + bool userspace_supports_overflow; }; /* diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c index 97e165183e79..a0901214aef5 100644 --- a/drivers/md/dm-raid.c +++ b/drivers/md/dm-raid.c @@ -329,8 +329,7 @@ static int validate_region_size(struct raid_set *rs, unsigned long region_size) */ if (min_region_size > (1 << 13)) { /* If not a power of 2, make it the next power of 2 */ - if (min_region_size & (min_region_size - 1)) - region_size = 1 << fls(region_size); + region_size = roundup_pow_of_two(min_region_size); DMINFO("Choosing default region size of %lu sectors", region_size); } else { diff --git a/drivers/md/dm-snap-persistent.c b/drivers/md/dm-snap-persistent.c index bf71583296f7..aeacad9be51d 100644 --- a/drivers/md/dm-snap-persistent.c +++ b/drivers/md/dm-snap-persistent.c @@ -7,6 +7,7 @@ #include "dm-exception-store.h" +#include <linux/ctype.h> #include <linux/mm.h> #include <linux/pagemap.h> #include <linux/vmalloc.h> @@ -843,8 +844,7 @@ static void persistent_drop_snapshot(struct dm_exception_store *store) DMWARN("write header failed"); } -static int persistent_ctr(struct dm_exception_store *store, - unsigned argc, char **argv) +static int persistent_ctr(struct dm_exception_store *store, char *options) { struct pstore *ps; @@ -873,6 +873,16 @@ static int persistent_ctr(struct dm_exception_store *store, return -ENOMEM; } + if (options) { + char overflow = toupper(options[0]); + if (overflow == 'O') + store->userspace_supports_overflow = true; + else { + DMERR("Unsupported persistent store option: %s", options); + return -EINVAL; + } + } + store->context = ps; return 0; @@ -888,7 +898,8 @@ static unsigned persistent_status(struct dm_exception_store *store, case STATUSTYPE_INFO: break; case STATUSTYPE_TABLE: - DMEMIT(" P %llu", (unsigned long long)store->chunk_size); + DMEMIT(" %s %llu", store->userspace_supports_overflow ? "PO" : "P", + (unsigned long long)store->chunk_size); } return sz; diff --git a/drivers/md/dm-snap-transient.c b/drivers/md/dm-snap-transient.c index 1ce9a2586e41..9b7c8c8049d6 100644 --- a/drivers/md/dm-snap-transient.c +++ b/drivers/md/dm-snap-transient.c @@ -70,8 +70,7 @@ static void transient_usage(struct dm_exception_store *store, *metadata_sectors = 0; } -static int transient_ctr(struct dm_exception_store *store, - unsigned argc, char **argv) +static int transient_ctr(struct dm_exception_store *store, char *options) { struct transient_c *tc; diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c index c0bcd6516dfe..c06b74e91cd6 100644 --- a/drivers/md/dm-snap.c +++ b/drivers/md/dm-snap.c @@ -1098,7 +1098,7 @@ static void stop_merge(struct dm_snapshot *s) } /* - * Construct a snapshot mapping: <origin_dev> <COW-dev> <p/n> <chunk-size> + * Construct a snapshot mapping: <origin_dev> <COW-dev> <p|po|n> <chunk-size> */ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv) { @@ -1302,6 +1302,7 @@ static void __handover_exceptions(struct dm_snapshot *snap_src, u.store_swap = snap_dest->store; snap_dest->store = snap_src->store; + snap_dest->store->userspace_supports_overflow = u.store_swap->userspace_supports_overflow; snap_src->store = u.store_swap; snap_dest->store->snap = snap_dest; @@ -1739,8 +1740,11 @@ static int snapshot_map(struct dm_target *ti, struct bio *bio) pe = __find_pending_exception(s, pe, chunk); if (!pe) { - s->snapshot_overflowed = 1; - DMERR("Snapshot overflowed: Unable to allocate exception."); + if (s->store->userspace_supports_overflow) { + s->snapshot_overflowed = 1; + DMERR("Snapshot overflowed: Unable to allocate exception."); + } else + __invalidate_snapshot(s, -ENOMEM); r = -EIO; goto out_unlock; } @@ -2365,7 +2369,7 @@ static struct target_type origin_target = { static struct target_type snapshot_target = { .name = "snapshot", - .version = {1, 14, 0}, + .version = {1, 15, 0}, .module = THIS_MODULE, .ctr = snapshot_ctr, .dtr = snapshot_dtr, @@ -2379,7 +2383,7 @@ static struct target_type snapshot_target = { static struct target_type merge_target = { .name = dm_snapshot_merge_target_name, - .version = {1, 3, 0}, + .version = {1, 4, 0}, .module = THIS_MODULE, .ctr = snapshot_ctr, .dtr = snapshot_dtr, diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 6264781dc69a..1b5c6047e4f1 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -1001,6 +1001,7 @@ static void end_clone_bio(struct bio *clone) struct dm_rq_target_io *tio = info->tio; struct bio *bio = info->orig; unsigned int nr_bytes = info->orig->bi_iter.bi_size; + int error = clone->bi_error; bio_put(clone); @@ -1011,13 +1012,13 @@ static void end_clone_bio(struct bio *clone) * the remainder. */ return; - else if (bio->bi_error) { + else if (error) { /* * Don't notice the error to the upper layer yet. * The error handling decision is made by the target driver, * when the request is completed. */ - tio->error = bio->bi_error; + tio->error = error; return; } @@ -2837,8 +2838,6 @@ static void __dm_destroy(struct mapped_device *md, bool wait) might_sleep(); - map = dm_get_live_table(md, &srcu_idx); - spin_lock(&_minor_lock); idr_replace(&_minor_idr, MINOR_ALLOCED, MINOR(disk_devt(dm_disk(md)))); set_bit(DMF_FREEING, &md->flags); @@ -2852,14 +2851,14 @@ static void __dm_destroy(struct mapped_device *md, bool wait) * do not race with internal suspend. */ mutex_lock(&md->suspend_lock); + map = dm_get_live_table(md, &srcu_idx); if (!dm_suspended_md(md)) { dm_table_presuspend_targets(map); dm_table_postsuspend_targets(map); } - mutex_unlock(&md->suspend_lock); - /* dm_put_live_table must be before msleep, otherwise deadlock is possible */ dm_put_live_table(md, srcu_idx); + mutex_unlock(&md->suspend_lock); /* * Rare, but there may be I/O requests still going to complete, diff --git a/drivers/md/md.c b/drivers/md/md.c index 4f5ecbe94ccb..c702de18207a 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -5409,9 +5409,13 @@ static int md_set_readonly(struct mddev *mddev, struct block_device *bdev) * which will now never happen */ wake_up_process(mddev->sync_thread->tsk); + if (mddev->external && test_bit(MD_CHANGE_PENDING, &mddev->flags)) + return -EBUSY; mddev_unlock(mddev); wait_event(resync_wait, !test_bit(MD_RECOVERY_RUNNING, &mddev->recovery)); + wait_event(mddev->sb_wait, + !test_bit(MD_CHANGE_PENDING, &mddev->flags)); mddev_lock_nointr(mddev); mutex_lock(&mddev->open_mutex); @@ -8160,6 +8164,7 @@ void md_check_recovery(struct mddev *mddev) md_reap_sync_thread(mddev); clear_bit(MD_RECOVERY_RECOVER, &mddev->recovery); clear_bit(MD_RECOVERY_NEEDED, &mddev->recovery); + clear_bit(MD_CHANGE_PENDING, &mddev->flags); goto unlock; } diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c index d222522c52e0..d132f06afdd1 100644 --- a/drivers/md/multipath.c +++ b/drivers/md/multipath.c @@ -470,8 +470,7 @@ static int multipath_run (struct mddev *mddev) return 0; out_free_conf: - if (conf->pool) - mempool_destroy(conf->pool); + mempool_destroy(conf->pool); kfree(conf->multipaths); kfree(conf); mddev->private = NULL; diff --git a/drivers/md/raid0.c b/drivers/md/raid0.c index 63e619b2f44e..f8e5db0cb5aa 100644 --- a/drivers/md/raid0.c +++ b/drivers/md/raid0.c @@ -376,12 +376,6 @@ static int raid0_run(struct mddev *mddev) struct md_rdev *rdev; bool discard_supported = false; - rdev_for_each(rdev, mddev) { - disk_stack_limits(mddev->gendisk, rdev->bdev, - rdev->data_offset << 9); - if (blk_queue_discard(bdev_get_queue(rdev->bdev))) - discard_supported = true; - } blk_queue_max_hw_sectors(mddev->queue, mddev->chunk_sectors); blk_queue_max_write_same_sectors(mddev->queue, mddev->chunk_sectors); blk_queue_max_discard_sectors(mddev->queue, mddev->chunk_sectors); @@ -390,6 +384,12 @@ static int raid0_run(struct mddev *mddev) blk_queue_io_opt(mddev->queue, (mddev->chunk_sectors << 9) * mddev->raid_disks); + rdev_for_each(rdev, mddev) { + disk_stack_limits(mddev->gendisk, rdev->bdev, + rdev->data_offset << 9); + if (blk_queue_discard(bdev_get_queue(rdev->bdev))) + discard_supported = true; + } if (!discard_supported) queue_flag_clear_unlocked(QUEUE_FLAG_DISCARD, mddev->queue); else diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index 4517f06c41ba..ddd8a5f572aa 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -881,8 +881,7 @@ static sector_t wait_barrier(struct r1conf *conf, struct bio *bio) } if (bio && bio_data_dir(bio) == WRITE) { - if (bio->bi_iter.bi_sector >= - conf->mddev->curr_resync_completed) { + if (bio->bi_iter.bi_sector >= conf->next_resync) { if (conf->start_next_window == MaxSector) conf->start_next_window = conf->next_resync + @@ -1516,7 +1515,7 @@ static void close_sync(struct r1conf *conf) conf->r1buf_pool = NULL; spin_lock_irq(&conf->resync_lock); - conf->next_resync = 0; + conf->next_resync = MaxSector - 2 * NEXT_NORMALIO_DISTANCE; conf->start_next_window = MaxSector; conf->current_window_requests += conf->next_window_requests; @@ -2383,8 +2382,8 @@ static void raid1d(struct md_thread *thread) } spin_unlock_irqrestore(&conf->device_lock, flags); while (!list_empty(&tmp)) { - r1_bio = list_first_entry(&conf->bio_end_io_list, - struct r1bio, retry_list); + r1_bio = list_first_entry(&tmp, struct r1bio, + retry_list); list_del(&r1_bio->retry_list); raid_end_bio_io(r1_bio); } @@ -2843,8 +2842,7 @@ static struct r1conf *setup_conf(struct mddev *mddev) abort: if (conf) { - if (conf->r1bio_pool) - mempool_destroy(conf->r1bio_pool); + mempool_destroy(conf->r1bio_pool); kfree(conf->mirrors); safe_put_page(conf->tmppage); kfree(conf->poolinfo); @@ -2946,8 +2944,7 @@ static void raid1_free(struct mddev *mddev, void *priv) { struct r1conf *conf = priv; - if (conf->r1bio_pool) - mempool_destroy(conf->r1bio_pool); + mempool_destroy(conf->r1bio_pool); kfree(conf->mirrors); safe_put_page(conf->tmppage); kfree(conf->poolinfo); diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index 0fc33eb88855..9f69dc526f8c 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -2688,8 +2688,8 @@ static void raid10d(struct md_thread *thread) } spin_unlock_irqrestore(&conf->device_lock, flags); while (!list_empty(&tmp)) { - r10_bio = list_first_entry(&conf->bio_end_io_list, - struct r10bio, retry_list); + r10_bio = list_first_entry(&tmp, struct r10bio, + retry_list); list_del(&r10_bio->retry_list); raid_end_bio_io(r10_bio); } @@ -3486,8 +3486,7 @@ static struct r10conf *setup_conf(struct mddev *mddev) printk(KERN_ERR "md/raid10:%s: couldn't allocate memory.\n", mdname(mddev)); if (conf) { - if (conf->r10bio_pool) - mempool_destroy(conf->r10bio_pool); + mempool_destroy(conf->r10bio_pool); kfree(conf->mirrors); safe_put_page(conf->tmppage); kfree(conf); @@ -3682,8 +3681,7 @@ static int run(struct mddev *mddev) out_free_conf: md_unregister_thread(&mddev->thread); - if (conf->r10bio_pool) - mempool_destroy(conf->r10bio_pool); + mempool_destroy(conf->r10bio_pool); safe_put_page(conf->tmppage); kfree(conf->mirrors); kfree(conf); @@ -3696,8 +3694,7 @@ static void raid10_free(struct mddev *mddev, void *priv) { struct r10conf *conf = priv; - if (conf->r10bio_pool) - mempool_destroy(conf->r10bio_pool); + mempool_destroy(conf->r10bio_pool); safe_put_page(conf->tmppage); kfree(conf->mirrors); kfree(conf->mirrors_old); diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 15ef2c641b2b..49bb8d3ff9be 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -2271,8 +2271,7 @@ static void shrink_stripes(struct r5conf *conf) drop_one_stripe(conf)) ; - if (conf->slab_cache) - kmem_cache_destroy(conf->slab_cache); + kmem_cache_destroy(conf->slab_cache); conf->slab_cache = NULL; } @@ -3150,6 +3149,8 @@ handle_failed_stripe(struct r5conf *conf, struct stripe_head *sh, spin_unlock_irq(&sh->stripe_lock); if (test_and_clear_bit(R5_Overlap, &sh->dev[i].flags)) wake_up(&conf->wait_for_overlap); + if (bi) + s->to_read--; while (bi && bi->bi_iter.bi_sector < sh->dev[i].sector + STRIPE_SECTORS) { struct bio *nextbi = @@ -3169,6 +3170,8 @@ handle_failed_stripe(struct r5conf *conf, struct stripe_head *sh, */ clear_bit(R5_LOCKED, &sh->dev[i].flags); } + s->to_write = 0; + s->written = 0; if (test_and_clear_bit(STRIPE_FULL_WRITE, &sh->state)) if (atomic_dec_and_test(&conf->pending_full_writes)) @@ -3300,7 +3303,7 @@ static int need_this_block(struct stripe_head *sh, struct stripe_head_state *s, */ return 0; - for (i = 0; i < s->failed; i++) { + for (i = 0; i < s->failed && i < 2; i++) { if (fdev[i]->towrite && !test_bit(R5_UPTODATE, &fdev[i]->flags) && !test_bit(R5_OVERWRITE, &fdev[i]->flags)) @@ -3324,7 +3327,7 @@ static int need_this_block(struct stripe_head *sh, struct stripe_head_state *s, sh->sector < sh->raid_conf->mddev->recovery_cp) /* reconstruct-write isn't being forced */ return 0; - for (i = 0; i < s->failed; i++) { + for (i = 0; i < s->failed && i < 2; i++) { if (s->failed_num[i] != sh->pd_idx && s->failed_num[i] != sh->qd_idx && !test_bit(R5_UPTODATE, &fdev[i]->flags) && diff --git a/drivers/memory/fsl-corenet-cf.c b/drivers/memory/fsl-corenet-cf.c index d708ded5457b..662d050243be 100644 --- a/drivers/memory/fsl-corenet-cf.c +++ b/drivers/memory/fsl-corenet-cf.c @@ -61,6 +61,7 @@ static const struct of_device_id ccf_matches[] = { }, {} }; +MODULE_DEVICE_TABLE(of, ccf_matches); struct ccf_err_regs { u32 errdet; /* 0x00 Error Detect Register */ diff --git a/drivers/memory/ti-aemif.c b/drivers/memory/ti-aemif.c index ca7d97a9a9ba..a579a0f25840 100644 --- a/drivers/memory/ti-aemif.c +++ b/drivers/memory/ti-aemif.c @@ -324,6 +324,7 @@ static const struct of_device_id aemif_of_match[] = { { .compatible = "ti,da850-aemif", }, {}, }; +MODULE_DEVICE_TABLE(of, aemif_of_match); static int aemif_probe(struct platform_device *pdev) { diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig index ccccc2943f2f..22892c701c63 100644 --- a/drivers/misc/Kconfig +++ b/drivers/misc/Kconfig @@ -414,7 +414,7 @@ config TI_DAC7512 config VMWARE_BALLOON tristate "VMware Balloon Driver" - depends on X86 && HYPERVISOR_GUEST + depends on VMWARE_VMCI && X86 && HYPERVISOR_GUEST help This is VMware physical memory management driver which acts like a "balloon" that can be inflated to reclaim physical pages diff --git a/drivers/misc/ad525x_dpot-i2c.c b/drivers/misc/ad525x_dpot-i2c.c index d11187d36ddd..4f832002d116 100644 --- a/drivers/misc/ad525x_dpot-i2c.c +++ b/drivers/misc/ad525x_dpot-i2c.c @@ -117,4 +117,3 @@ module_i2c_driver(ad_dpot_i2c_driver); MODULE_AUTHOR("Michael Hennerich <hennerich@blackfin.uclinux.org>"); MODULE_DESCRIPTION("digital potentiometer I2C bus driver"); MODULE_LICENSE("GPL"); -MODULE_ALIAS("i2c:ad_dpot"); diff --git a/drivers/misc/genwqe/card_base.h b/drivers/misc/genwqe/card_base.h index e7353449874b..cb851c14ca4b 100644 --- a/drivers/misc/genwqe/card_base.h +++ b/drivers/misc/genwqe/card_base.h @@ -514,7 +514,7 @@ int __genwqe_execute_ddcb(struct genwqe_dev *cd, /** * __genwqe_execute_raw_ddcb() - Execute DDCB request without addr translation * - * This version will not do address translation or any modifcation of + * This version will not do address translation or any modification of * the DDCB data. It is used e.g. for the MoveFlash DDCB which is * entirely prepared by the driver itself. That means the appropriate * DMA addresses are already in the DDCB and do not need any diff --git a/drivers/misc/genwqe/card_ddcb.c b/drivers/misc/genwqe/card_ddcb.c index 6d51e5f08664..353ee0cc733d 100644 --- a/drivers/misc/genwqe/card_ddcb.c +++ b/drivers/misc/genwqe/card_ddcb.c @@ -203,7 +203,7 @@ struct genwqe_ddcb_cmd *ddcb_requ_alloc(void) { struct ddcb_requ *req; - req = kzalloc(sizeof(*req), GFP_ATOMIC); + req = kzalloc(sizeof(*req), GFP_KERNEL); if (!req) return NULL; diff --git a/drivers/misc/genwqe/card_dev.c b/drivers/misc/genwqe/card_dev.c index 70e62d6a3231..7f1b282d7d96 100644 --- a/drivers/misc/genwqe/card_dev.c +++ b/drivers/misc/genwqe/card_dev.c @@ -449,7 +449,7 @@ static int genwqe_mmap(struct file *filp, struct vm_area_struct *vma) if (get_order(vsize) > MAX_ORDER) return -ENOMEM; - dma_map = kzalloc(sizeof(struct dma_mapping), GFP_ATOMIC); + dma_map = kzalloc(sizeof(struct dma_mapping), GFP_KERNEL); if (dma_map == NULL) return -ENOMEM; @@ -785,7 +785,7 @@ static int genwqe_pin_mem(struct genwqe_file *cfile, struct genwqe_mem *m) map_addr = (m->addr & PAGE_MASK); map_size = round_up(m->size + (m->addr & ~PAGE_MASK), PAGE_SIZE); - dma_map = kzalloc(sizeof(struct dma_mapping), GFP_ATOMIC); + dma_map = kzalloc(sizeof(struct dma_mapping), GFP_KERNEL); if (dma_map == NULL) return -ENOMEM; diff --git a/drivers/misc/genwqe/card_utils.c b/drivers/misc/genwqe/card_utils.c index 1ca94e6fa8fb..222367cc8c81 100644 --- a/drivers/misc/genwqe/card_utils.c +++ b/drivers/misc/genwqe/card_utils.c @@ -220,7 +220,8 @@ void *__genwqe_alloc_consistent(struct genwqe_dev *cd, size_t size, if (get_order(size) > MAX_ORDER) return NULL; - return pci_alloc_consistent(cd->pci_dev, size, dma_handle); + return dma_alloc_coherent(&cd->pci_dev->dev, size, dma_handle, + GFP_KERNEL); } void __genwqe_free_consistent(struct genwqe_dev *cd, size_t size, @@ -229,7 +230,7 @@ void __genwqe_free_consistent(struct genwqe_dev *cd, size_t size, if (vaddr == NULL) return; - pci_free_consistent(cd->pci_dev, size, vaddr, dma_handle); + dma_free_coherent(&cd->pci_dev->dev, size, vaddr, dma_handle); } static void genwqe_unmap_pages(struct genwqe_dev *cd, dma_addr_t *dma_list, diff --git a/drivers/misc/hpilo.c b/drivers/misc/hpilo.c index b83e3ca12a41..d6a901cd4222 100644 --- a/drivers/misc/hpilo.c +++ b/drivers/misc/hpilo.c @@ -2,7 +2,7 @@ * Driver for the HP iLO management processor. * * Copyright (C) 2008 Hewlett-Packard Development Company, L.P. - * David Altobelli <david.altobelli@hp.com> + * David Altobelli <david.altobelli@hpe.com> * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -902,11 +902,11 @@ static void __exit ilo_exit(void) MODULE_VERSION("1.4.1"); MODULE_ALIAS(ILO_NAME); MODULE_DESCRIPTION(ILO_NAME); -MODULE_AUTHOR("David Altobelli <david.altobelli@hp.com>"); +MODULE_AUTHOR("David Altobelli <david.altobelli@hpe.com>"); MODULE_LICENSE("GPL v2"); module_param(max_ccb, uint, 0444); -MODULE_PARM_DESC(max_ccb, "Maximum number of HP iLO channels to attach (16)"); +MODULE_PARM_DESC(max_ccb, "Maximum number of HP iLO channels to attach (8-24)(default=16)"); module_init(ilo_init); module_exit(ilo_exit); diff --git a/drivers/misc/kgdbts.c b/drivers/misc/kgdbts.c index 9a60bd4d3c49..99635dd9dbac 100644 --- a/drivers/misc/kgdbts.c +++ b/drivers/misc/kgdbts.c @@ -1112,6 +1112,7 @@ static int __init init_kgdbts(void) return configure_kgdbts(); } +device_initcall(init_kgdbts); static int kgdbts_get_char(void) { @@ -1180,10 +1181,9 @@ static struct kgdb_io kgdbts_io_ops = { .post_exception = kgdbts_post_exp_handler, }; -module_init(init_kgdbts); +/* + * not really modular, but the easiest way to keep compat with existing + * bootargs behaviour is to continue using module_param here. + */ module_param_call(kgdbts, param_set_kgdbts_var, param_get_string, &kps, 0644); MODULE_PARM_DESC(kgdbts, "<A|V1|V2>[F#|S#][N#]"); -MODULE_DESCRIPTION("KGDB Test Suite"); -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Wind River Systems, Inc."); - diff --git a/drivers/misc/mei/bus-fixup.c b/drivers/misc/mei/bus-fixup.c index 3e536ca85f7d..020de5919c21 100644 --- a/drivers/misc/mei/bus-fixup.c +++ b/drivers/misc/mei/bus-fixup.c @@ -285,11 +285,11 @@ static struct mei_fixup { }; /** - * mei_cl_dev_fixup - run fixup handlers + * mei_cldev_fixup - run fixup handlers * * @cldev: me client device */ -void mei_cl_dev_fixup(struct mei_cl_device *cldev) +void mei_cl_bus_dev_fixup(struct mei_cl_device *cldev) { struct mei_fixup *f; const uuid_le *uuid = mei_me_cl_uuid(cldev->me_cl); diff --git a/drivers/misc/mei/bus.c b/drivers/misc/mei/bus.c index eef1c6b46ad8..832085207a7f 100644 --- a/drivers/misc/mei/bus.c +++ b/drivers/misc/mei/bus.c @@ -165,7 +165,7 @@ out: } /** - * mei_cl_send - me device send (write) + * mei_cldev_send - me device send (write) * * @cldev: me client device * @buf: buffer to send @@ -173,7 +173,7 @@ out: * * Return: written size in bytes or < 0 on error */ -ssize_t mei_cl_send(struct mei_cl_device *cldev, u8 *buf, size_t length) +ssize_t mei_cldev_send(struct mei_cl_device *cldev, u8 *buf, size_t length) { struct mei_cl *cl = cldev->cl; @@ -182,10 +182,10 @@ ssize_t mei_cl_send(struct mei_cl_device *cldev, u8 *buf, size_t length) return __mei_cl_send(cl, buf, length, 1); } -EXPORT_SYMBOL_GPL(mei_cl_send); +EXPORT_SYMBOL_GPL(mei_cldev_send); /** - * mei_cl_recv - client receive (read) + * mei_cldev_recv - client receive (read) * * @cldev: me client device * @buf: buffer to send @@ -193,7 +193,7 @@ EXPORT_SYMBOL_GPL(mei_cl_send); * * Return: read size in bytes of < 0 on error */ -ssize_t mei_cl_recv(struct mei_cl_device *cldev, u8 *buf, size_t length) +ssize_t mei_cldev_recv(struct mei_cl_device *cldev, u8 *buf, size_t length) { struct mei_cl *cl = cldev->cl; @@ -202,15 +202,15 @@ ssize_t mei_cl_recv(struct mei_cl_device *cldev, u8 *buf, size_t length) return __mei_cl_recv(cl, buf, length); } -EXPORT_SYMBOL_GPL(mei_cl_recv); +EXPORT_SYMBOL_GPL(mei_cldev_recv); /** - * mei_bus_event_work - dispatch rx event for a bus device + * mei_cl_bus_event_work - dispatch rx event for a bus device * and schedule new work * * @work: work */ -static void mei_bus_event_work(struct work_struct *work) +static void mei_cl_bus_event_work(struct work_struct *work) { struct mei_cl_device *cldev; @@ -272,7 +272,7 @@ void mei_cl_bus_rx_event(struct mei_cl *cl) } /** - * mei_cl_register_event_cb - register event callback + * mei_cldev_register_event_cb - register event callback * * @cldev: me client devices * @event_cb: callback function @@ -283,9 +283,9 @@ void mei_cl_bus_rx_event(struct mei_cl *cl) * -EALREADY if an callback is already registered * <0 on other errors */ -int mei_cl_register_event_cb(struct mei_cl_device *cldev, - unsigned long events_mask, - mei_cl_event_cb_t event_cb, void *context) +int mei_cldev_register_event_cb(struct mei_cl_device *cldev, + unsigned long events_mask, + mei_cldev_event_cb_t event_cb, void *context) { int ret; @@ -296,7 +296,7 @@ int mei_cl_register_event_cb(struct mei_cl_device *cldev, cldev->events_mask = events_mask; cldev->event_cb = event_cb; cldev->event_context = context; - INIT_WORK(&cldev->event_work, mei_bus_event_work); + INIT_WORK(&cldev->event_work, mei_cl_bus_event_work); if (cldev->events_mask & BIT(MEI_CL_EVENT_RX)) { ret = mei_cl_read_start(cldev->cl, 0, NULL); @@ -314,42 +314,81 @@ int mei_cl_register_event_cb(struct mei_cl_device *cldev, return 0; } -EXPORT_SYMBOL_GPL(mei_cl_register_event_cb); +EXPORT_SYMBOL_GPL(mei_cldev_register_event_cb); /** - * mei_cl_get_drvdata - driver data getter + * mei_cldev_get_drvdata - driver data getter * * @cldev: mei client device * * Return: driver private data */ -void *mei_cl_get_drvdata(const struct mei_cl_device *cldev) +void *mei_cldev_get_drvdata(const struct mei_cl_device *cldev) { return dev_get_drvdata(&cldev->dev); } -EXPORT_SYMBOL_GPL(mei_cl_get_drvdata); +EXPORT_SYMBOL_GPL(mei_cldev_get_drvdata); /** - * mei_cl_set_drvdata - driver data setter + * mei_cldev_set_drvdata - driver data setter * * @cldev: mei client device * @data: data to store */ -void mei_cl_set_drvdata(struct mei_cl_device *cldev, void *data) +void mei_cldev_set_drvdata(struct mei_cl_device *cldev, void *data) { dev_set_drvdata(&cldev->dev, data); } -EXPORT_SYMBOL_GPL(mei_cl_set_drvdata); +EXPORT_SYMBOL_GPL(mei_cldev_set_drvdata); /** - * mei_cl_enable_device - enable me client device + * mei_cldev_uuid - return uuid of the underlying me client + * + * @cldev: mei client device + * + * Return: me client uuid + */ +const uuid_le *mei_cldev_uuid(const struct mei_cl_device *cldev) +{ + return mei_me_cl_uuid(cldev->me_cl); +} +EXPORT_SYMBOL_GPL(mei_cldev_uuid); + +/** + * mei_cldev_ver - return protocol version of the underlying me client + * + * @cldev: mei client device + * + * Return: me client protocol version + */ +u8 mei_cldev_ver(const struct mei_cl_device *cldev) +{ + return mei_me_cl_ver(cldev->me_cl); +} +EXPORT_SYMBOL_GPL(mei_cldev_ver); + +/** + * mei_cldev_enabled - check whether the device is enabled + * + * @cldev: mei client device + * + * Return: true if me client is initialized and connected + */ +bool mei_cldev_enabled(struct mei_cl_device *cldev) +{ + return cldev->cl && mei_cl_is_connected(cldev->cl); +} +EXPORT_SYMBOL_GPL(mei_cldev_enabled); + +/** + * mei_cldev_enable_device - enable me client device * create connection with me client * * @cldev: me client device * * Return: 0 on success and < 0 on error */ -int mei_cl_enable_device(struct mei_cl_device *cldev) +int mei_cldev_enable(struct mei_cl_device *cldev) { struct mei_device *bus = cldev->bus; struct mei_cl *cl; @@ -389,17 +428,17 @@ out: return ret; } -EXPORT_SYMBOL_GPL(mei_cl_enable_device); +EXPORT_SYMBOL_GPL(mei_cldev_enable); /** - * mei_cl_disable_device - disable me client device + * mei_cldev_disable - disable me client device * disconnect form the me client * * @cldev: me client device * * Return: 0 on success and < 0 on error */ -int mei_cl_disable_device(struct mei_cl_device *cldev) +int mei_cldev_disable(struct mei_cl_device *cldev) { struct mei_device *bus; struct mei_cl *cl; @@ -437,7 +476,7 @@ out: mutex_unlock(&bus->device_lock); return err; } -EXPORT_SYMBOL_GPL(mei_cl_disable_device); +EXPORT_SYMBOL_GPL(mei_cldev_disable); /** * mei_cl_device_find - find matching entry in the driver id table @@ -453,17 +492,26 @@ struct mei_cl_device_id *mei_cl_device_find(struct mei_cl_device *cldev, { const struct mei_cl_device_id *id; const uuid_le *uuid; + u8 version; + bool match; uuid = mei_me_cl_uuid(cldev->me_cl); + version = mei_me_cl_ver(cldev->me_cl); id = cldrv->id_table; while (uuid_le_cmp(NULL_UUID_LE, id->uuid)) { if (!uuid_le_cmp(*uuid, id->uuid)) { + match = true; - if (!cldev->name[0]) - return id; + if (cldev->name[0]) + if (strncmp(cldev->name, id->name, + sizeof(id->name))) + match = false; - if (!strncmp(cldev->name, id->name, sizeof(id->name))) + if (id->version != MEI_CL_VERSION_ANY) + if (id->version != version) + match = false; + if (match) return id; } @@ -590,6 +638,19 @@ static ssize_t uuid_show(struct device *dev, struct device_attribute *a, } static DEVICE_ATTR_RO(uuid); +static ssize_t version_show(struct device *dev, struct device_attribute *a, + char *buf) +{ + struct mei_cl_device *cldev = to_mei_cl_device(dev); + u8 version = mei_me_cl_ver(cldev->me_cl); + size_t len; + + len = snprintf(buf, PAGE_SIZE, "%02X", version); + + return (len >= PAGE_SIZE) ? (PAGE_SIZE - 1) : len; +} +static DEVICE_ATTR_RO(version); + static ssize_t modalias_show(struct device *dev, struct device_attribute *a, char *buf) { @@ -597,20 +658,19 @@ static ssize_t modalias_show(struct device *dev, struct device_attribute *a, const uuid_le *uuid = mei_me_cl_uuid(cldev->me_cl); size_t len; - len = snprintf(buf, PAGE_SIZE, "mei:%s:" MEI_CL_UUID_FMT ":", - cldev->name, MEI_CL_UUID_ARGS(uuid->b)); - + len = snprintf(buf, PAGE_SIZE, "mei:%s:%pUl:", cldev->name, uuid); return (len >= PAGE_SIZE) ? (PAGE_SIZE - 1) : len; } static DEVICE_ATTR_RO(modalias); -static struct attribute *mei_cl_dev_attrs[] = { +static struct attribute *mei_cldev_attrs[] = { &dev_attr_name.attr, &dev_attr_uuid.attr, + &dev_attr_version.attr, &dev_attr_modalias.attr, NULL, }; -ATTRIBUTE_GROUPS(mei_cl_dev); +ATTRIBUTE_GROUPS(mei_cldev); /** * mei_cl_device_uevent - me client bus uevent handler @@ -624,6 +684,10 @@ static int mei_cl_device_uevent(struct device *dev, struct kobj_uevent_env *env) { struct mei_cl_device *cldev = to_mei_cl_device(dev); const uuid_le *uuid = mei_me_cl_uuid(cldev->me_cl); + u8 version = mei_me_cl_ver(cldev->me_cl); + + if (add_uevent_var(env, "MEI_CL_VERSION=%d", version)) + return -ENOMEM; if (add_uevent_var(env, "MEI_CL_UUID=%pUl", uuid)) return -ENOMEM; @@ -631,8 +695,8 @@ static int mei_cl_device_uevent(struct device *dev, struct kobj_uevent_env *env) if (add_uevent_var(env, "MEI_CL_NAME=%s", cldev->name)) return -ENOMEM; - if (add_uevent_var(env, "MODALIAS=mei:%s:" MEI_CL_UUID_FMT ":", - cldev->name, MEI_CL_UUID_ARGS(uuid->b))) + if (add_uevent_var(env, "MODALIAS=mei:%s:%pUl:%02X:", + cldev->name, uuid, version)) return -ENOMEM; return 0; @@ -640,7 +704,7 @@ static int mei_cl_device_uevent(struct device *dev, struct kobj_uevent_env *env) static struct bus_type mei_cl_bus_type = { .name = "mei", - .dev_groups = mei_cl_dev_groups, + .dev_groups = mei_cldev_groups, .match = mei_cl_device_match, .probe = mei_cl_device_probe, .remove = mei_cl_device_remove, @@ -661,7 +725,7 @@ static void mei_dev_bus_put(struct mei_device *bus) put_device(bus->dev); } -static void mei_cl_dev_release(struct device *dev) +static void mei_cl_bus_dev_release(struct device *dev) { struct mei_cl_device *cldev = to_mei_cl_device(dev); @@ -674,19 +738,19 @@ static void mei_cl_dev_release(struct device *dev) } static struct device_type mei_cl_device_type = { - .release = mei_cl_dev_release, + .release = mei_cl_bus_dev_release, }; /** - * mei_cl_dev_alloc - initialize and allocate mei client device + * mei_cl_bus_dev_alloc - initialize and allocate mei client device * * @bus: mei device * @me_cl: me client * * Return: allocated device structur or NULL on allocation failure */ -static struct mei_cl_device *mei_cl_dev_alloc(struct mei_device *bus, - struct mei_me_client *me_cl) +static struct mei_cl_device *mei_cl_bus_dev_alloc(struct mei_device *bus, + struct mei_me_client *me_cl) { struct mei_cl_device *cldev; @@ -715,15 +779,17 @@ static struct mei_cl_device *mei_cl_dev_alloc(struct mei_device *bus, * * Return: true if the device is eligible for enumeration */ -static bool mei_cl_dev_setup(struct mei_device *bus, - struct mei_cl_device *cldev) +static bool mei_cl_bus_dev_setup(struct mei_device *bus, + struct mei_cl_device *cldev) { cldev->do_match = 1; - mei_cl_dev_fixup(cldev); + mei_cl_bus_dev_fixup(cldev); if (cldev->do_match) - dev_set_name(&cldev->dev, "mei:%s:%pUl", - cldev->name, mei_me_cl_uuid(cldev->me_cl)); + dev_set_name(&cldev->dev, "mei:%s:%pUl:%02X", + cldev->name, + mei_me_cl_uuid(cldev->me_cl), + mei_me_cl_ver(cldev->me_cl)); return cldev->do_match == 1; } @@ -739,7 +805,9 @@ static int mei_cl_bus_dev_add(struct mei_cl_device *cldev) { int ret; - dev_dbg(cldev->bus->dev, "adding %pUL\n", mei_me_cl_uuid(cldev->me_cl)); + dev_dbg(cldev->bus->dev, "adding %pUL:%02X\n", + mei_me_cl_uuid(cldev->me_cl), + mei_me_cl_ver(cldev->me_cl)); ret = device_add(&cldev->dev); if (!ret) cldev->is_added = 1; @@ -804,13 +872,14 @@ void mei_cl_bus_remove_devices(struct mei_device *bus) /** - * mei_cl_dev_init - allocate and initializes an mei client devices + * mei_cl_bus_dev_init - allocate and initializes an mei client devices * based on me client * * @bus: mei device * @me_cl: me client */ -static void mei_cl_dev_init(struct mei_device *bus, struct mei_me_client *me_cl) +static void mei_cl_bus_dev_init(struct mei_device *bus, + struct mei_me_client *me_cl) { struct mei_cl_device *cldev; @@ -819,7 +888,7 @@ static void mei_cl_dev_init(struct mei_device *bus, struct mei_me_client *me_cl) if (me_cl->bus_added) return; - cldev = mei_cl_dev_alloc(bus, me_cl); + cldev = mei_cl_bus_dev_alloc(bus, me_cl); if (!cldev) return; @@ -843,7 +912,7 @@ void mei_cl_bus_rescan(struct mei_device *bus) down_read(&bus->me_clients_rwsem); list_for_each_entry(me_cl, &bus->me_clients, list) - mei_cl_dev_init(bus, me_cl); + mei_cl_bus_dev_init(bus, me_cl); up_read(&bus->me_clients_rwsem); mutex_lock(&bus->cl_bus_lock); @@ -857,7 +926,7 @@ void mei_cl_bus_rescan(struct mei_device *bus) if (cldev->is_added) continue; - if (mei_cl_dev_setup(bus, cldev)) + if (mei_cl_bus_dev_setup(bus, cldev)) mei_cl_bus_dev_add(cldev); else { list_del_init(&cldev->bus_list); @@ -869,7 +938,8 @@ void mei_cl_bus_rescan(struct mei_device *bus) dev_dbg(bus->dev, "rescan end"); } -int __mei_cl_driver_register(struct mei_cl_driver *cldrv, struct module *owner) +int __mei_cldev_driver_register(struct mei_cl_driver *cldrv, + struct module *owner) { int err; @@ -885,15 +955,15 @@ int __mei_cl_driver_register(struct mei_cl_driver *cldrv, struct module *owner) return 0; } -EXPORT_SYMBOL_GPL(__mei_cl_driver_register); +EXPORT_SYMBOL_GPL(__mei_cldev_driver_register); -void mei_cl_driver_unregister(struct mei_cl_driver *cldrv) +void mei_cldev_driver_unregister(struct mei_cl_driver *cldrv) { driver_unregister(&cldrv->driver); pr_debug("mei: driver [%s] unregistered\n", cldrv->driver.name); } -EXPORT_SYMBOL_GPL(mei_cl_driver_unregister); +EXPORT_SYMBOL_GPL(mei_cldev_driver_unregister); int __init mei_cl_bus_init(void) diff --git a/drivers/misc/mei/client.h b/drivers/misc/mei/client.h index 1c7cad07d731..04e1aa39243f 100644 --- a/drivers/misc/mei/client.h +++ b/drivers/misc/mei/client.h @@ -68,6 +68,18 @@ static inline const uuid_le *mei_me_cl_uuid(const struct mei_me_client *me_cl) return &me_cl->props.protocol_name; } +/** + * mei_me_cl_ver - return me client protocol version + * + * @me_cl: me client + * + * Return: me client protocol version + */ +static inline u8 mei_me_cl_ver(const struct mei_me_client *me_cl) +{ + return me_cl->props.protocol_version; +} + /* * MEI IO Functions */ diff --git a/drivers/misc/mei/hbm.c b/drivers/misc/mei/hbm.c index 8eec887c8f70..6d7c188fb65c 100644 --- a/drivers/misc/mei/hbm.c +++ b/drivers/misc/mei/hbm.c @@ -1209,7 +1209,7 @@ int mei_hbm_dispatch(struct mei_device *dev, struct mei_msg_hdr *hdr) * after the host receives the enum_resp * message clients may be added or removed */ - if (dev->hbm_state <= MEI_HBM_ENUM_CLIENTS && + if (dev->hbm_state <= MEI_HBM_ENUM_CLIENTS || dev->hbm_state >= MEI_HBM_STOPPED) { dev_err(dev->dev, "hbm: add client: state mismatch, [%d, %d]\n", dev->dev_state, dev->hbm_state); diff --git a/drivers/misc/mei/hw-me.c b/drivers/misc/mei/hw-me.c index 65511d39d89b..25b1997a62cb 100644 --- a/drivers/misc/mei/hw-me.c +++ b/drivers/misc/mei/hw-me.c @@ -150,7 +150,7 @@ static inline u32 mei_me_d0i3c_read(const struct mei_device *dev) u32 reg; reg = mei_me_reg_read(to_me_hw(dev), H_D0I3C); - trace_mei_reg_read(dev->dev, "H_D0I3C", H_CSR, reg); + trace_mei_reg_read(dev->dev, "H_D0I3C", H_D0I3C, reg); return reg; } @@ -163,7 +163,7 @@ static inline u32 mei_me_d0i3c_read(const struct mei_device *dev) */ static inline void mei_me_d0i3c_write(struct mei_device *dev, u32 reg) { - trace_mei_reg_write(dev->dev, "H_D0I3C", H_CSR, reg); + trace_mei_reg_write(dev->dev, "H_D0I3C", H_D0I3C, reg); mei_me_reg_write(to_me_hw(dev), H_D0I3C, reg); } diff --git a/drivers/misc/mei/mei_dev.h b/drivers/misc/mei/mei_dev.h index e25ee16c658e..0f87dffa6be1 100644 --- a/drivers/misc/mei/mei_dev.h +++ b/drivers/misc/mei/mei_dev.h @@ -340,7 +340,7 @@ struct mei_hw_ops { /* MEI bus API*/ void mei_cl_bus_rescan(struct mei_device *bus); -void mei_cl_dev_fixup(struct mei_cl_device *dev); +void mei_cl_bus_dev_fixup(struct mei_cl_device *dev); ssize_t __mei_cl_send(struct mei_cl *cl, u8 *buf, size_t length, bool blocking); ssize_t __mei_cl_recv(struct mei_cl *cl, u8 *buf, size_t length); diff --git a/drivers/misc/mic/Kconfig b/drivers/misc/mic/Kconfig index e9f2f56c370d..40677df7f996 100644 --- a/drivers/misc/mic/Kconfig +++ b/drivers/misc/mic/Kconfig @@ -36,7 +36,7 @@ comment "Intel MIC Host Driver" config INTEL_MIC_HOST tristate "Intel MIC Host Driver" - depends on 64BIT && PCI && X86 && INTEL_MIC_BUS && SCIF_BUS + depends on 64BIT && PCI && X86 && INTEL_MIC_BUS && SCIF_BUS && MIC_COSM select VHOST_RING help This enables Host Driver support for the Intel Many Integrated @@ -56,7 +56,7 @@ comment "Intel MIC Card Driver" config INTEL_MIC_CARD tristate "Intel MIC Card Driver" - depends on 64BIT && X86 && INTEL_MIC_BUS && SCIF_BUS + depends on 64BIT && X86 && INTEL_MIC_BUS && SCIF_BUS && MIC_COSM select VIRTIO help This enables card driver support for the Intel Many Integrated @@ -74,7 +74,8 @@ comment "SCIF Driver" config SCIF tristate "SCIF Driver" - depends on 64BIT && PCI && X86 && SCIF_BUS + depends on 64BIT && PCI && X86 && SCIF_BUS && IOMMU_SUPPORT + select IOMMU_IOVA help This enables SCIF Driver support for the Intel Many Integrated Core (MIC) family of PCIe form factor coprocessor devices that @@ -88,3 +89,21 @@ config SCIF More information about the Intel MIC family as well as the Linux OS and tools for MIC to use with this driver are available from <http://software.intel.com/en-us/mic-developer>. + +comment "Intel MIC Coprocessor State Management (COSM) Drivers" + +config MIC_COSM + tristate "Intel MIC Coprocessor State Management (COSM) Drivers" + depends on 64BIT && PCI && X86 && SCIF + help + This enables COSM driver support for the Intel Many + Integrated Core (MIC) family of PCIe form factor coprocessor + devices. COSM drivers implement functions such as boot, + shutdown, reset and reboot of MIC devices. + + If you are building a host kernel with an Intel MIC device then + say M (recommended) or Y, else say N. If unsure say N. + + More information about the Intel MIC family as well as the Linux + OS and tools for MIC to use with this driver are available from + <http://software.intel.com/en-us/mic-developer>. diff --git a/drivers/misc/mic/Makefile b/drivers/misc/mic/Makefile index a74042c58649..e288a1106738 100644 --- a/drivers/misc/mic/Makefile +++ b/drivers/misc/mic/Makefile @@ -6,3 +6,5 @@ obj-$(CONFIG_INTEL_MIC_HOST) += host/ obj-$(CONFIG_INTEL_MIC_CARD) += card/ obj-y += bus/ obj-$(CONFIG_SCIF) += scif/ +obj-$(CONFIG_MIC_COSM) += cosm/ +obj-$(CONFIG_MIC_COSM) += cosm_client/ diff --git a/drivers/misc/mic/bus/Makefile b/drivers/misc/mic/bus/Makefile index 1ed37e234c96..761842b0d0bb 100644 --- a/drivers/misc/mic/bus/Makefile +++ b/drivers/misc/mic/bus/Makefile @@ -4,3 +4,4 @@ # obj-$(CONFIG_INTEL_MIC_BUS) += mic_bus.o obj-$(CONFIG_SCIF_BUS) += scif_bus.o +obj-$(CONFIG_MIC_COSM) += cosm_bus.o diff --git a/drivers/misc/mic/bus/cosm_bus.c b/drivers/misc/mic/bus/cosm_bus.c new file mode 100644 index 000000000000..1e36830f1a4e --- /dev/null +++ b/drivers/misc/mic/bus/cosm_bus.c @@ -0,0 +1,141 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC COSM Bus Driver + */ +#include <linux/slab.h> +#include <linux/module.h> +#include <linux/idr.h> +#include "cosm_bus.h" + +/* Unique numbering for cosm devices. */ +static DEFINE_IDA(cosm_index_ida); + +static int cosm_dev_probe(struct device *d) +{ + struct cosm_device *dev = dev_to_cosm(d); + struct cosm_driver *drv = drv_to_cosm(dev->dev.driver); + + return drv->probe(dev); +} + +static int cosm_dev_remove(struct device *d) +{ + struct cosm_device *dev = dev_to_cosm(d); + struct cosm_driver *drv = drv_to_cosm(dev->dev.driver); + + drv->remove(dev); + return 0; +} + +static struct bus_type cosm_bus = { + .name = "cosm_bus", + .probe = cosm_dev_probe, + .remove = cosm_dev_remove, +}; + +int cosm_register_driver(struct cosm_driver *driver) +{ + driver->driver.bus = &cosm_bus; + return driver_register(&driver->driver); +} +EXPORT_SYMBOL_GPL(cosm_register_driver); + +void cosm_unregister_driver(struct cosm_driver *driver) +{ + driver_unregister(&driver->driver); +} +EXPORT_SYMBOL_GPL(cosm_unregister_driver); + +static inline void cosm_release_dev(struct device *d) +{ + struct cosm_device *cdev = dev_to_cosm(d); + + kfree(cdev); +} + +struct cosm_device * +cosm_register_device(struct device *pdev, struct cosm_hw_ops *hw_ops) +{ + struct cosm_device *cdev; + int ret; + + cdev = kzalloc(sizeof(*cdev), GFP_KERNEL); + if (!cdev) + return ERR_PTR(-ENOMEM); + + cdev->dev.parent = pdev; + cdev->dev.release = cosm_release_dev; + cdev->hw_ops = hw_ops; + dev_set_drvdata(&cdev->dev, cdev); + cdev->dev.bus = &cosm_bus; + + /* Assign a unique device index and hence name */ + ret = ida_simple_get(&cosm_index_ida, 0, 0, GFP_KERNEL); + if (ret < 0) + goto free_cdev; + + cdev->index = ret; + cdev->dev.id = ret; + dev_set_name(&cdev->dev, "cosm-dev%u", cdev->index); + + ret = device_register(&cdev->dev); + if (ret) + goto ida_remove; + return cdev; +ida_remove: + ida_simple_remove(&cosm_index_ida, cdev->index); +free_cdev: + kfree(cdev); + return ERR_PTR(ret); +} +EXPORT_SYMBOL_GPL(cosm_register_device); + +void cosm_unregister_device(struct cosm_device *dev) +{ + int index = dev->index; /* save for after device release */ + + device_unregister(&dev->dev); + ida_simple_remove(&cosm_index_ida, index); +} +EXPORT_SYMBOL_GPL(cosm_unregister_device); + +struct cosm_device *cosm_find_cdev_by_id(int id) +{ + struct device *dev = subsys_find_device_by_id(&cosm_bus, id, NULL); + + return dev ? container_of(dev, struct cosm_device, dev) : NULL; +} +EXPORT_SYMBOL_GPL(cosm_find_cdev_by_id); + +static int __init cosm_init(void) +{ + return bus_register(&cosm_bus); +} + +static void __exit cosm_exit(void) +{ + bus_unregister(&cosm_bus); + ida_destroy(&cosm_index_ida); +} + +core_initcall(cosm_init); +module_exit(cosm_exit); + +MODULE_AUTHOR("Intel Corporation"); +MODULE_DESCRIPTION("Intel(R) MIC card OS state management bus driver"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/misc/mic/bus/cosm_bus.h b/drivers/misc/mic/bus/cosm_bus.h new file mode 100644 index 000000000000..f7c57f266916 --- /dev/null +++ b/drivers/misc/mic/bus/cosm_bus.h @@ -0,0 +1,134 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC COSM Bus Driver + */ +#ifndef _COSM_BUS_H_ +#define _COSM_BUS_H_ + +#include <linux/scif.h> +#include <linux/mic_common.h> +#include "../common/mic_dev.h" + +/** + * cosm_device - representation of a cosm device + * + * @attr_group: Pointer to list of sysfs attribute groups. + * @sdev: Device for sysfs entries. + * @state: MIC state. + * @shutdown_status: MIC status reported by card for shutdown/crashes. + * @shutdown_status_int: Internal shutdown status maintained by the driver + * @cosm_mutex: Mutex for synchronizing access to data structures. + * @reset_trigger_work: Work for triggering reset requests. + * @scif_work: Work for handling per device SCIF connections + * @cmdline: Kernel command line. + * @firmware: Firmware file name. + * @ramdisk: Ramdisk file name. + * @bootmode: Boot mode i.e. "linux" or "elf" for flash updates. + * @log_buf_addr: Log buffer address for MIC. + * @log_buf_len: Log buffer length address for MIC. + * @state_sysfs: Sysfs dirent for notifying ring 3 about MIC state changes. + * @hw_ops: the hardware bus ops for this device. + * @dev: underlying device. + * @index: unique position on the cosm bus + * @dbg_dir: debug fs directory + * @newepd: new endpoint from scif accept to be assigned to this cdev + * @epd: SCIF endpoint for this cdev + * @heartbeat_watchdog_enable: if heartbeat watchdog is enabled for this cdev + * @sysfs_heartbeat_enable: sysfs setting for disabling heartbeat notification + */ +struct cosm_device { + const struct attribute_group **attr_group; + struct device *sdev; + u8 state; + u8 shutdown_status; + u8 shutdown_status_int; + struct mutex cosm_mutex; + struct work_struct reset_trigger_work; + struct work_struct scif_work; + char *cmdline; + char *firmware; + char *ramdisk; + char *bootmode; + void *log_buf_addr; + int *log_buf_len; + struct kernfs_node *state_sysfs; + struct cosm_hw_ops *hw_ops; + struct device dev; + int index; + struct dentry *dbg_dir; + scif_epd_t newepd; + scif_epd_t epd; + bool heartbeat_watchdog_enable; + bool sysfs_heartbeat_enable; +}; + +/** + * cosm_driver - operations for a cosm driver + * + * @driver: underlying device driver (populate name and owner). + * @probe: the function to call when a device is found. Returns 0 or -errno. + * @remove: the function to call when a device is removed. + */ +struct cosm_driver { + struct device_driver driver; + int (*probe)(struct cosm_device *dev); + void (*remove)(struct cosm_device *dev); +}; + +/** + * cosm_hw_ops - cosm bus ops + * + * @reset: trigger MIC reset + * @force_reset: force MIC reset + * @post_reset: inform MIC reset is complete + * @ready: is MIC ready for OS download + * @start: boot MIC + * @stop: prepare MIC for reset + * @family: return MIC HW family string + * @stepping: return MIC HW stepping string + * @aper: return MIC PCIe aperture + */ +struct cosm_hw_ops { + void (*reset)(struct cosm_device *cdev); + void (*force_reset)(struct cosm_device *cdev); + void (*post_reset)(struct cosm_device *cdev, enum mic_states state); + bool (*ready)(struct cosm_device *cdev); + int (*start)(struct cosm_device *cdev, int id); + void (*stop)(struct cosm_device *cdev, bool force); + ssize_t (*family)(struct cosm_device *cdev, char *buf); + ssize_t (*stepping)(struct cosm_device *cdev, char *buf); + struct mic_mw *(*aper)(struct cosm_device *cdev); +}; + +struct cosm_device * +cosm_register_device(struct device *pdev, struct cosm_hw_ops *hw_ops); +void cosm_unregister_device(struct cosm_device *dev); +int cosm_register_driver(struct cosm_driver *drv); +void cosm_unregister_driver(struct cosm_driver *drv); +struct cosm_device *cosm_find_cdev_by_id(int id); + +static inline struct cosm_device *dev_to_cosm(struct device *dev) +{ + return container_of(dev, struct cosm_device, dev); +} + +static inline struct cosm_driver *drv_to_cosm(struct device_driver *drv) +{ + return container_of(drv, struct cosm_driver, driver); +} +#endif /* _COSM_BUS_H */ diff --git a/drivers/misc/mic/bus/mic_bus.c b/drivers/misc/mic/bus/mic_bus.c index 961ae90aae47..c64955d8cbc1 100644 --- a/drivers/misc/mic/bus/mic_bus.c +++ b/drivers/misc/mic/bus/mic_bus.c @@ -25,9 +25,6 @@ #include <linux/idr.h> #include <linux/mic_bus.h> -/* Unique numbering for mbus devices. */ -static DEFINE_IDA(mbus_index_ida); - static ssize_t device_show(struct device *d, struct device_attribute *attr, char *buf) { @@ -147,7 +144,8 @@ static void mbus_release_dev(struct device *d) struct mbus_device * mbus_register_device(struct device *pdev, int id, struct dma_map_ops *dma_ops, - struct mbus_hw_ops *hw_ops, void __iomem *mmio_va) + struct mbus_hw_ops *hw_ops, int index, + void __iomem *mmio_va) { int ret; struct mbus_device *mbdev; @@ -166,13 +164,7 @@ mbus_register_device(struct device *pdev, int id, struct dma_map_ops *dma_ops, mbdev->dev.release = mbus_release_dev; mbdev->hw_ops = hw_ops; mbdev->dev.bus = &mic_bus; - - /* Assign a unique device index and hence name. */ - ret = ida_simple_get(&mbus_index_ida, 0, 0, GFP_KERNEL); - if (ret < 0) - goto free_mbdev; - - mbdev->index = ret; + mbdev->index = index; dev_set_name(&mbdev->dev, "mbus-dev%u", mbdev->index); /* * device_register() causes the bus infrastructure to look for a @@ -180,10 +172,8 @@ mbus_register_device(struct device *pdev, int id, struct dma_map_ops *dma_ops, */ ret = device_register(&mbdev->dev); if (ret) - goto ida_remove; + goto free_mbdev; return mbdev; -ida_remove: - ida_simple_remove(&mbus_index_ida, mbdev->index); free_mbdev: kfree(mbdev); return ERR_PTR(ret); @@ -192,10 +182,7 @@ EXPORT_SYMBOL_GPL(mbus_register_device); void mbus_unregister_device(struct mbus_device *mbdev) { - int index = mbdev->index; /* save for after device release */ - device_unregister(&mbdev->dev); - ida_simple_remove(&mbus_index_ida, index); } EXPORT_SYMBOL_GPL(mbus_unregister_device); @@ -207,7 +194,6 @@ static int __init mbus_init(void) static void __exit mbus_exit(void) { bus_unregister(&mic_bus); - ida_destroy(&mbus_index_ida); } core_initcall(mbus_init); diff --git a/drivers/misc/mic/bus/scif_bus.c b/drivers/misc/mic/bus/scif_bus.c index 2da7ceed015d..fd2702143022 100644 --- a/drivers/misc/mic/bus/scif_bus.c +++ b/drivers/misc/mic/bus/scif_bus.c @@ -28,7 +28,6 @@ static ssize_t device_show(struct device *d, return sprintf(buf, "0x%04x\n", dev->id.device); } - static DEVICE_ATTR_RO(device); static ssize_t vendor_show(struct device *d, @@ -38,7 +37,6 @@ static ssize_t vendor_show(struct device *d, return sprintf(buf, "0x%04x\n", dev->id.vendor); } - static DEVICE_ATTR_RO(vendor); static ssize_t modalias_show(struct device *d, @@ -49,7 +47,6 @@ static ssize_t modalias_show(struct device *d, return sprintf(buf, "scif:d%08Xv%08X\n", dev->id.device, dev->id.vendor); } - static DEVICE_ATTR_RO(modalias); static struct attribute *scif_dev_attrs[] = { @@ -144,7 +141,8 @@ struct scif_hw_dev * scif_register_device(struct device *pdev, int id, struct dma_map_ops *dma_ops, struct scif_hw_ops *hw_ops, u8 dnode, u8 snode, struct mic_mw *mmio, struct mic_mw *aper, void *dp, - void __iomem *rdp, struct dma_chan **chan, int num_chan) + void __iomem *rdp, struct dma_chan **chan, int num_chan, + bool card_rel_da) { int ret; struct scif_hw_dev *sdev; @@ -171,6 +169,7 @@ scif_register_device(struct device *pdev, int id, struct dma_map_ops *dma_ops, dma_set_mask(&sdev->dev, DMA_BIT_MASK(64)); sdev->dma_ch = chan; sdev->num_dma_ch = num_chan; + sdev->card_rel_da = card_rel_da; dev_set_name(&sdev->dev, "scif-dev%u", sdev->dnode); /* * device_register() causes the bus infrastructure to look for a diff --git a/drivers/misc/mic/bus/scif_bus.h b/drivers/misc/mic/bus/scif_bus.h index 335a228a8236..94f29ac608b6 100644 --- a/drivers/misc/mic/bus/scif_bus.h +++ b/drivers/misc/mic/bus/scif_bus.h @@ -46,6 +46,8 @@ struct scif_hw_dev_id { * @rdp - Remote device page * @dma_ch - Array of DMA channels * @num_dma_ch - Number of DMA channels available + * @card_rel_da - Set to true if DMA addresses programmed in the DMA engine + * are relative to the card point of view */ struct scif_hw_dev { struct scif_hw_ops *hw_ops; @@ -59,6 +61,7 @@ struct scif_hw_dev { void __iomem *rdp; struct dma_chan **dma_ch; int num_dma_ch; + bool card_rel_da; }; /** @@ -114,7 +117,8 @@ scif_register_device(struct device *pdev, int id, struct scif_hw_ops *hw_ops, u8 dnode, u8 snode, struct mic_mw *mmio, struct mic_mw *aper, void *dp, void __iomem *rdp, - struct dma_chan **chan, int num_chan); + struct dma_chan **chan, int num_chan, + bool card_rel_da); void scif_unregister_device(struct scif_hw_dev *sdev); static inline struct scif_hw_dev *dev_to_scif(struct device *dev) diff --git a/drivers/misc/mic/card/mic_device.c b/drivers/misc/mic/card/mic_device.c index 6338908b2252..d0edaf7e0cd5 100644 --- a/drivers/misc/mic/card/mic_device.c +++ b/drivers/misc/mic/card/mic_device.c @@ -37,71 +37,6 @@ #include "mic_virtio.h" static struct mic_driver *g_drv; -static struct mic_irq *shutdown_cookie; - -static void mic_notify_host(u8 state) -{ - struct mic_driver *mdrv = g_drv; - struct mic_bootparam __iomem *bootparam = mdrv->dp; - - iowrite8(state, &bootparam->shutdown_status); - dev_dbg(mdrv->dev, "%s %d system_state %d\n", - __func__, __LINE__, state); - mic_send_intr(&mdrv->mdev, ioread8(&bootparam->c2h_shutdown_db)); -} - -static int mic_panic_event(struct notifier_block *this, unsigned long event, - void *ptr) -{ - struct mic_driver *mdrv = g_drv; - struct mic_bootparam __iomem *bootparam = mdrv->dp; - - iowrite8(-1, &bootparam->h2c_config_db); - iowrite8(-1, &bootparam->h2c_shutdown_db); - mic_notify_host(MIC_CRASHED); - return NOTIFY_DONE; -} - -static struct notifier_block mic_panic = { - .notifier_call = mic_panic_event, -}; - -static irqreturn_t mic_shutdown_isr(int irq, void *data) -{ - struct mic_driver *mdrv = g_drv; - struct mic_bootparam __iomem *bootparam = mdrv->dp; - - mic_ack_interrupt(&g_drv->mdev); - if (ioread8(&bootparam->shutdown_card)) - orderly_poweroff(true); - return IRQ_HANDLED; -} - -static int mic_shutdown_init(void) -{ - int rc = 0; - struct mic_driver *mdrv = g_drv; - struct mic_bootparam __iomem *bootparam = mdrv->dp; - int shutdown_db; - - shutdown_db = mic_next_card_db(); - shutdown_cookie = mic_request_card_irq(mic_shutdown_isr, NULL, - "Shutdown", mdrv, shutdown_db); - if (IS_ERR(shutdown_cookie)) - rc = PTR_ERR(shutdown_cookie); - else - iowrite8(shutdown_db, &bootparam->h2c_shutdown_db); - return rc; -} - -static void mic_shutdown_uninit(void) -{ - struct mic_driver *mdrv = g_drv; - struct mic_bootparam __iomem *bootparam = mdrv->dp; - - iowrite8(-1, &bootparam->h2c_shutdown_db); - mic_free_card_irq(shutdown_cookie, mdrv); -} static int __init mic_dp_init(void) { @@ -359,11 +294,7 @@ int __init mic_driver_init(struct mic_driver *mdrv) u8 node_id; g_drv = mdrv; - /* - * Unloading the card module is not supported. The MIC card module - * handles fundamental operations like host/card initiated shutdowns - * and informing the host about card crashes and cannot be unloaded. - */ + /* Unloading the card module is not supported. */ if (!try_module_get(mdrv->dev->driver->owner)) { rc = -ENODEV; goto done; @@ -374,12 +305,9 @@ int __init mic_driver_init(struct mic_driver *mdrv) rc = mic_init_irq(); if (rc) goto dp_uninit; - rc = mic_shutdown_init(); - if (rc) - goto irq_uninit; if (!mic_request_dma_chans(mdrv)) { rc = -ENODEV; - goto shutdown_uninit; + goto irq_uninit; } rc = mic_devices_init(mdrv); if (rc) @@ -390,21 +318,18 @@ int __init mic_driver_init(struct mic_driver *mdrv) NULL, &scif_hw_ops, 0, node_id, &mdrv->mdev.mmio, NULL, NULL, mdrv->dp, mdrv->dma_ch, - mdrv->num_dma_ch); + mdrv->num_dma_ch, true); if (IS_ERR(mdrv->scdev)) { rc = PTR_ERR(mdrv->scdev); goto device_uninit; } mic_create_card_debug_dir(mdrv); - atomic_notifier_chain_register(&panic_notifier_list, &mic_panic); done: return rc; device_uninit: mic_devices_uninit(mdrv); dma_free: mic_free_dma_chans(mdrv); -shutdown_uninit: - mic_shutdown_uninit(); irq_uninit: mic_uninit_irq(); dp_uninit: @@ -425,13 +350,6 @@ void mic_driver_uninit(struct mic_driver *mdrv) scif_unregister_device(mdrv->scdev); mic_devices_uninit(mdrv); mic_free_dma_chans(mdrv); - /* - * Inform the host about the shutdown status i.e. poweroff/restart etc. - * The module cannot be unloaded so the only code path to call - * mic_devices_uninit(..) is the shutdown callback. - */ - mic_notify_host(system_state); - mic_shutdown_uninit(); mic_uninit_irq(); mic_dp_uninit(); module_put(mdrv->dev->driver->owner); diff --git a/drivers/misc/mic/card/mic_x100.c b/drivers/misc/mic/card/mic_x100.c index 77fd41781c2e..b2958ce2368c 100644 --- a/drivers/misc/mic/card/mic_x100.c +++ b/drivers/misc/mic/card/mic_x100.c @@ -261,7 +261,7 @@ static int __init mic_probe(struct platform_device *pdev) mic_hw_intr_init(mdrv); platform_set_drvdata(pdev, mdrv); mdrv->dma_mbdev = mbus_register_device(mdrv->dev, MBUS_DEV_DMA_MIC, - NULL, &mbus_hw_ops, + NULL, &mbus_hw_ops, 0, mdrv->mdev.mmio.va); if (IS_ERR(mdrv->dma_mbdev)) { rc = PTR_ERR(mdrv->dma_mbdev); diff --git a/drivers/misc/mic/common/mic_dev.h b/drivers/misc/mic/common/mic_dev.h index 0b58c46045dc..50776772ebdf 100644 --- a/drivers/misc/mic/common/mic_dev.h +++ b/drivers/misc/mic/common/mic_dev.h @@ -21,6 +21,19 @@ #ifndef __MIC_DEV_H__ #define __MIC_DEV_H__ +/* The maximum number of MIC devices supported in a single host system. */ +#define MIC_MAX_NUM_DEVS 128 + +/** + * enum mic_hw_family - The hardware family to which a device belongs. + */ +enum mic_hw_family { + MIC_FAMILY_X100 = 0, + MIC_FAMILY_X200, + MIC_FAMILY_UNKNOWN, + MIC_FAMILY_LAST +}; + /** * struct mic_mw - MIC memory window * diff --git a/drivers/misc/mic/cosm/Makefile b/drivers/misc/mic/cosm/Makefile new file mode 100644 index 000000000000..b85d4d49df46 --- /dev/null +++ b/drivers/misc/mic/cosm/Makefile @@ -0,0 +1,10 @@ +# +# Makefile - Intel MIC Coprocessor State Management (COSM) Driver +# Copyright(c) 2015, Intel Corporation. +# +obj-$(CONFIG_MIC_COSM) += mic_cosm.o + +mic_cosm-objs := cosm_main.o +mic_cosm-objs += cosm_debugfs.o +mic_cosm-objs += cosm_sysfs.o +mic_cosm-objs += cosm_scif_server.o diff --git a/drivers/misc/mic/cosm/cosm_debugfs.c b/drivers/misc/mic/cosm/cosm_debugfs.c new file mode 100644 index 000000000000..30d953dff4fa --- /dev/null +++ b/drivers/misc/mic/cosm/cosm_debugfs.c @@ -0,0 +1,155 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Coprocessor State Management (COSM) Driver + * + */ + +#include <linux/debugfs.h> +#include <linux/slab.h> +#include "cosm_main.h" + +/* Debugfs parent dir */ +static struct dentry *cosm_dbg; + +/** + * cosm_log_buf_show - Display MIC kernel log buffer + * + * log_buf addr/len is read from System.map by user space + * and populated in sysfs entries. + */ +static int cosm_log_buf_show(struct seq_file *s, void *unused) +{ + void __iomem *log_buf_va; + int __iomem *log_buf_len_va; + struct cosm_device *cdev = s->private; + void *kva; + int size; + u64 aper_offset; + + if (!cdev || !cdev->log_buf_addr || !cdev->log_buf_len) + goto done; + + mutex_lock(&cdev->cosm_mutex); + switch (cdev->state) { + case MIC_BOOTING: + case MIC_ONLINE: + case MIC_SHUTTING_DOWN: + break; + default: + goto unlock; + } + + /* + * Card kernel will never be relocated and any kernel text/data mapping + * can be translated to phys address by subtracting __START_KERNEL_map. + */ + aper_offset = (u64)cdev->log_buf_len - __START_KERNEL_map; + log_buf_len_va = cdev->hw_ops->aper(cdev)->va + aper_offset; + aper_offset = (u64)cdev->log_buf_addr - __START_KERNEL_map; + log_buf_va = cdev->hw_ops->aper(cdev)->va + aper_offset; + + size = ioread32(log_buf_len_va); + kva = kmalloc(size, GFP_KERNEL); + if (!kva) + goto unlock; + + memcpy_fromio(kva, log_buf_va, size); + seq_write(s, kva, size); + kfree(kva); +unlock: + mutex_unlock(&cdev->cosm_mutex); +done: + return 0; +} + +static int cosm_log_buf_open(struct inode *inode, struct file *file) +{ + return single_open(file, cosm_log_buf_show, inode->i_private); +} + +static const struct file_operations log_buf_ops = { + .owner = THIS_MODULE, + .open = cosm_log_buf_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release +}; + +/** + * cosm_force_reset_show - Force MIC reset + * + * Invokes the force_reset COSM bus op instead of the standard reset + * op in case a force reset of the MIC device is required + */ +static int cosm_force_reset_show(struct seq_file *s, void *pos) +{ + struct cosm_device *cdev = s->private; + + cosm_stop(cdev, true); + return 0; +} + +static int cosm_force_reset_debug_open(struct inode *inode, struct file *file) +{ + return single_open(file, cosm_force_reset_show, inode->i_private); +} + +static const struct file_operations force_reset_ops = { + .owner = THIS_MODULE, + .open = cosm_force_reset_debug_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release +}; + +void cosm_create_debug_dir(struct cosm_device *cdev) +{ + char name[16]; + + if (!cosm_dbg) + return; + + scnprintf(name, sizeof(name), "mic%d", cdev->index); + cdev->dbg_dir = debugfs_create_dir(name, cosm_dbg); + if (!cdev->dbg_dir) + return; + + debugfs_create_file("log_buf", 0444, cdev->dbg_dir, cdev, &log_buf_ops); + debugfs_create_file("force_reset", 0444, cdev->dbg_dir, cdev, + &force_reset_ops); +} + +void cosm_delete_debug_dir(struct cosm_device *cdev) +{ + if (!cdev->dbg_dir) + return; + + debugfs_remove_recursive(cdev->dbg_dir); +} + +void cosm_init_debugfs(void) +{ + cosm_dbg = debugfs_create_dir(KBUILD_MODNAME, NULL); + if (!cosm_dbg) + pr_err("can't create debugfs dir\n"); +} + +void cosm_exit_debugfs(void) +{ + debugfs_remove(cosm_dbg); +} diff --git a/drivers/misc/mic/cosm/cosm_main.c b/drivers/misc/mic/cosm/cosm_main.c new file mode 100644 index 000000000000..4b4b356c797d --- /dev/null +++ b/drivers/misc/mic/cosm/cosm_main.c @@ -0,0 +1,388 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Coprocessor State Management (COSM) Driver + * + */ + +#include <linux/module.h> +#include <linux/delay.h> +#include <linux/idr.h> +#include <linux/slab.h> +#include <linux/cred.h> +#include "cosm_main.h" + +static const char cosm_driver_name[] = "mic"; + +/* COSM ID allocator */ +static struct ida g_cosm_ida; +/* Class of MIC devices for sysfs accessibility. */ +static struct class *g_cosm_class; +/* Number of MIC devices */ +static atomic_t g_num_dev; + +/** + * cosm_hw_reset - Issue a HW reset for the MIC device + * @cdev: pointer to cosm_device instance + */ +static void cosm_hw_reset(struct cosm_device *cdev, bool force) +{ + int i; + +#define MIC_RESET_TO (45) + if (force && cdev->hw_ops->force_reset) + cdev->hw_ops->force_reset(cdev); + else + cdev->hw_ops->reset(cdev); + + for (i = 0; i < MIC_RESET_TO; i++) { + if (cdev->hw_ops->ready(cdev)) { + cosm_set_state(cdev, MIC_READY); + return; + } + /* + * Resets typically take 10s of seconds to complete. + * Since an MMIO read is required to check if the + * firmware is ready or not, a 1 second delay works nicely. + */ + msleep(1000); + } + cosm_set_state(cdev, MIC_RESET_FAILED); +} + +/** + * cosm_start - Start the MIC + * @cdev: pointer to cosm_device instance + * + * This function prepares an MIC for boot and initiates boot. + * RETURNS: An appropriate -ERRNO error value on error, or 0 for success. + */ +int cosm_start(struct cosm_device *cdev) +{ + const struct cred *orig_cred; + struct cred *override_cred; + int rc; + + mutex_lock(&cdev->cosm_mutex); + if (!cdev->bootmode) { + dev_err(&cdev->dev, "%s %d bootmode not set\n", + __func__, __LINE__); + rc = -EINVAL; + goto unlock_ret; + } +retry: + if (cdev->state != MIC_READY) { + dev_err(&cdev->dev, "%s %d MIC state not READY\n", + __func__, __LINE__); + rc = -EINVAL; + goto unlock_ret; + } + if (!cdev->hw_ops->ready(cdev)) { + cosm_hw_reset(cdev, false); + /* + * The state will either be MIC_READY if the reset succeeded + * or MIC_RESET_FAILED if the firmware reset failed. + */ + goto retry; + } + + /* + * Set credentials to root to allow non-root user to download initramsfs + * with 600 permissions + */ + override_cred = prepare_creds(); + if (!override_cred) { + dev_err(&cdev->dev, "%s %d prepare_creds failed\n", + __func__, __LINE__); + rc = -ENOMEM; + goto unlock_ret; + } + override_cred->fsuid = GLOBAL_ROOT_UID; + orig_cred = override_creds(override_cred); + + rc = cdev->hw_ops->start(cdev, cdev->index); + + revert_creds(orig_cred); + put_cred(override_cred); + if (rc) + goto unlock_ret; + + /* + * If linux is being booted, card is treated 'online' only + * when the scif interface in the card is up. If anything else + * is booted, we set card to 'online' immediately. + */ + if (!strcmp(cdev->bootmode, "linux")) + cosm_set_state(cdev, MIC_BOOTING); + else + cosm_set_state(cdev, MIC_ONLINE); +unlock_ret: + mutex_unlock(&cdev->cosm_mutex); + if (rc) + dev_err(&cdev->dev, "cosm_start failed rc %d\n", rc); + return rc; +} + +/** + * cosm_stop - Prepare the MIC for reset and trigger reset + * @cdev: pointer to cosm_device instance + * @force: force a MIC to reset even if it is already reset and ready. + * + * RETURNS: None + */ +void cosm_stop(struct cosm_device *cdev, bool force) +{ + mutex_lock(&cdev->cosm_mutex); + if (cdev->state != MIC_READY || force) { + /* + * Don't call hw_ops if they have been called previously. + * stop(..) calls device_unregister and will crash the system if + * called multiple times. + */ + bool call_hw_ops = cdev->state != MIC_RESET_FAILED && + cdev->state != MIC_READY; + + if (cdev->state != MIC_RESETTING) + cosm_set_state(cdev, MIC_RESETTING); + cdev->heartbeat_watchdog_enable = false; + if (call_hw_ops) + cdev->hw_ops->stop(cdev, force); + cosm_hw_reset(cdev, force); + cosm_set_shutdown_status(cdev, MIC_NOP); + if (call_hw_ops && cdev->hw_ops->post_reset) + cdev->hw_ops->post_reset(cdev, cdev->state); + } + mutex_unlock(&cdev->cosm_mutex); + flush_work(&cdev->scif_work); +} + +/** + * cosm_reset_trigger_work - Trigger MIC reset + * @work: The work structure + * + * This work is scheduled whenever the host wants to reset the MIC. + */ +static void cosm_reset_trigger_work(struct work_struct *work) +{ + struct cosm_device *cdev = container_of(work, struct cosm_device, + reset_trigger_work); + cosm_stop(cdev, false); +} + +/** + * cosm_reset - Schedule MIC reset + * @cdev: pointer to cosm_device instance + * + * RETURNS: An -EINVAL if the card is already READY or 0 for success. + */ +int cosm_reset(struct cosm_device *cdev) +{ + int rc = 0; + + mutex_lock(&cdev->cosm_mutex); + if (cdev->state != MIC_READY) { + cosm_set_state(cdev, MIC_RESETTING); + schedule_work(&cdev->reset_trigger_work); + } else { + dev_err(&cdev->dev, "%s %d MIC is READY\n", __func__, __LINE__); + rc = -EINVAL; + } + mutex_unlock(&cdev->cosm_mutex); + return rc; +} + +/** + * cosm_shutdown - Initiate MIC shutdown. + * @cdev: pointer to cosm_device instance + * + * RETURNS: None + */ +int cosm_shutdown(struct cosm_device *cdev) +{ + struct cosm_msg msg = { .id = COSM_MSG_SHUTDOWN }; + int rc = 0; + + mutex_lock(&cdev->cosm_mutex); + if (cdev->state != MIC_ONLINE) { + rc = -EINVAL; + dev_err(&cdev->dev, "%s %d skipping shutdown in state: %s\n", + __func__, __LINE__, cosm_state_string[cdev->state]); + goto err; + } + + if (!cdev->epd) { + rc = -ENOTCONN; + dev_err(&cdev->dev, "%s %d scif endpoint not connected rc %d\n", + __func__, __LINE__, rc); + goto err; + } + + rc = scif_send(cdev->epd, &msg, sizeof(msg), SCIF_SEND_BLOCK); + if (rc < 0) { + dev_err(&cdev->dev, "%s %d scif_send failed rc %d\n", + __func__, __LINE__, rc); + goto err; + } + cdev->heartbeat_watchdog_enable = false; + cosm_set_state(cdev, MIC_SHUTTING_DOWN); + rc = 0; +err: + mutex_unlock(&cdev->cosm_mutex); + return rc; +} + +static int cosm_driver_probe(struct cosm_device *cdev) +{ + int rc; + + /* Initialize SCIF server at first probe */ + if (atomic_add_return(1, &g_num_dev) == 1) { + rc = cosm_scif_init(); + if (rc) + goto scif_exit; + } + mutex_init(&cdev->cosm_mutex); + INIT_WORK(&cdev->reset_trigger_work, cosm_reset_trigger_work); + INIT_WORK(&cdev->scif_work, cosm_scif_work); + cdev->sysfs_heartbeat_enable = true; + cosm_sysfs_init(cdev); + cdev->sdev = device_create_with_groups(g_cosm_class, cdev->dev.parent, + MKDEV(0, cdev->index), cdev, cdev->attr_group, + "mic%d", cdev->index); + if (IS_ERR(cdev->sdev)) { + rc = PTR_ERR(cdev->sdev); + dev_err(&cdev->dev, "device_create_with_groups failed rc %d\n", + rc); + goto scif_exit; + } + + cdev->state_sysfs = sysfs_get_dirent(cdev->sdev->kobj.sd, + "state"); + if (!cdev->state_sysfs) { + rc = -ENODEV; + dev_err(&cdev->dev, "sysfs_get_dirent failed rc %d\n", rc); + goto destroy_device; + } + cosm_create_debug_dir(cdev); + return 0; +destroy_device: + device_destroy(g_cosm_class, MKDEV(0, cdev->index)); +scif_exit: + if (atomic_dec_and_test(&g_num_dev)) + cosm_scif_exit(); + return rc; +} + +static void cosm_driver_remove(struct cosm_device *cdev) +{ + cosm_delete_debug_dir(cdev); + sysfs_put(cdev->state_sysfs); + device_destroy(g_cosm_class, MKDEV(0, cdev->index)); + flush_work(&cdev->reset_trigger_work); + cosm_stop(cdev, false); + if (atomic_dec_and_test(&g_num_dev)) + cosm_scif_exit(); + + /* These sysfs entries might have allocated */ + kfree(cdev->cmdline); + kfree(cdev->firmware); + kfree(cdev->ramdisk); + kfree(cdev->bootmode); +} + +static int cosm_suspend(struct device *dev) +{ + struct cosm_device *cdev = dev_to_cosm(dev); + + mutex_lock(&cdev->cosm_mutex); + switch (cdev->state) { + /** + * Suspend/freeze hooks in userspace have already shutdown the card. + * Card should be 'ready' in most cases. It is however possible that + * some userspace application initiated a boot. In those cases, we + * simply reset the card. + */ + case MIC_ONLINE: + case MIC_BOOTING: + case MIC_SHUTTING_DOWN: + mutex_unlock(&cdev->cosm_mutex); + cosm_stop(cdev, false); + break; + default: + mutex_unlock(&cdev->cosm_mutex); + break; + } + return 0; +} + +static const struct dev_pm_ops cosm_pm_ops = { + .suspend = cosm_suspend, + .freeze = cosm_suspend +}; + +static struct cosm_driver cosm_driver = { + .driver = { + .name = KBUILD_MODNAME, + .owner = THIS_MODULE, + .pm = &cosm_pm_ops, + }, + .probe = cosm_driver_probe, + .remove = cosm_driver_remove +}; + +static int __init cosm_init(void) +{ + int ret; + + cosm_init_debugfs(); + + g_cosm_class = class_create(THIS_MODULE, cosm_driver_name); + if (IS_ERR(g_cosm_class)) { + ret = PTR_ERR(g_cosm_class); + pr_err("class_create failed ret %d\n", ret); + goto cleanup_debugfs; + } + + ida_init(&g_cosm_ida); + ret = cosm_register_driver(&cosm_driver); + if (ret) { + pr_err("cosm_register_driver failed ret %d\n", ret); + goto ida_destroy; + } + return 0; +ida_destroy: + ida_destroy(&g_cosm_ida); + class_destroy(g_cosm_class); +cleanup_debugfs: + cosm_exit_debugfs(); + return ret; +} + +static void __exit cosm_exit(void) +{ + cosm_unregister_driver(&cosm_driver); + ida_destroy(&g_cosm_ida); + class_destroy(g_cosm_class); + cosm_exit_debugfs(); +} + +module_init(cosm_init); +module_exit(cosm_exit); + +MODULE_AUTHOR("Intel Corporation"); +MODULE_DESCRIPTION("Intel(R) MIC Coprocessor State Management (COSM) Driver"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/misc/mic/cosm/cosm_main.h b/drivers/misc/mic/cosm/cosm_main.h new file mode 100644 index 000000000000..f01156fca881 --- /dev/null +++ b/drivers/misc/mic/cosm/cosm_main.h @@ -0,0 +1,70 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Coprocessor State Management (COSM) Driver + * + */ +#ifndef _COSM_COSM_H_ +#define _COSM_COSM_H_ + +#include <linux/scif.h> +#include "../bus/cosm_bus.h" + +#define COSM_HEARTBEAT_SEND_SEC 30 +#define SCIF_COSM_LISTEN_PORT 201 + +/** + * enum COSM msg id's + * @COSM_MSG_SHUTDOWN: host->card trigger shutdown + * @COSM_MSG_SYNC_TIME: host->card send host time to card to sync time + * @COSM_MSG_HEARTBEAT: card->host heartbeat + * @COSM_MSG_SHUTDOWN_STATUS: card->host with shutdown status as payload + */ +enum cosm_msg_id { + COSM_MSG_SHUTDOWN, + COSM_MSG_SYNC_TIME, + COSM_MSG_HEARTBEAT, + COSM_MSG_SHUTDOWN_STATUS, +}; + +struct cosm_msg { + u64 id; + union { + u64 shutdown_status; + struct timespec64 timespec; + }; +}; + +extern const char * const cosm_state_string[]; +extern const char * const cosm_shutdown_status_string[]; + +void cosm_sysfs_init(struct cosm_device *cdev); +int cosm_start(struct cosm_device *cdev); +void cosm_stop(struct cosm_device *cdev, bool force); +int cosm_reset(struct cosm_device *cdev); +int cosm_shutdown(struct cosm_device *cdev); +void cosm_set_state(struct cosm_device *cdev, u8 state); +void cosm_set_shutdown_status(struct cosm_device *cdev, u8 status); +void cosm_init_debugfs(void); +void cosm_exit_debugfs(void); +void cosm_create_debug_dir(struct cosm_device *cdev); +void cosm_delete_debug_dir(struct cosm_device *cdev); +int cosm_scif_init(void); +void cosm_scif_exit(void); +void cosm_scif_work(struct work_struct *work); + +#endif diff --git a/drivers/misc/mic/cosm/cosm_scif_server.c b/drivers/misc/mic/cosm/cosm_scif_server.c new file mode 100644 index 000000000000..5696df4326b5 --- /dev/null +++ b/drivers/misc/mic/cosm/cosm_scif_server.c @@ -0,0 +1,405 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Coprocessor State Management (COSM) Driver + * + */ +#include <linux/kthread.h> +#include "cosm_main.h" + +/* + * The COSM driver uses SCIF to communicate between the management node and the + * MIC cards. SCIF is used to (a) Send a shutdown command to the card (b) + * receive a shutdown status back from the card upon completion of shutdown and + * (c) receive periodic heartbeat messages from the card used to deduce if the + * card has crashed. + * + * A COSM server consisting of a SCIF listening endpoint waits for incoming + * connections from the card. Upon acceptance of the connection, a separate + * work-item is scheduled to handle SCIF message processing for that card. The + * life-time of this work-item is therefore the time from which the connection + * from a card is accepted to the time at which the connection is closed. A new + * work-item starts each time the card boots and is alive till the card (a) + * shuts down (b) is reset (c) crashes (d) cosm_client driver on the card is + * unloaded. + * + * From the point of view of COSM interactions with SCIF during card + * shutdown, reset and crash are as follows: + * + * Card shutdown + * ------------- + * 1. COSM client on the card invokes orderly_poweroff() in response to SHUTDOWN + * message from the host. + * 2. Card driver shutdown callback invokes scif_unregister_device(..) resulting + * in scif_remove(..) getting called on the card + * 3. scif_remove -> scif_stop -> scif_handle_remove_node -> + * scif_peer_unregister_device -> device_unregister for the host peer device + * 4. During device_unregister remove(..) method of cosm_client is invoked which + * closes the COSM SCIF endpoint on the card. This results in a SCIF_DISCNCT + * message being sent to host SCIF. SCIF_DISCNCT message processing on the + * host SCIF sets the host COSM SCIF endpoint state to DISCONNECTED and wakes + * up the host COSM thread blocked in scif_poll(..) resulting in + * scif_poll(..) returning POLLHUP. + * 5. On the card, scif_peer_release_dev is next called which results in an + * SCIF_EXIT message being sent to the host and after receiving the + * SCIF_EXIT_ACK from the host the peer device teardown on the card is + * complete. + * 6. As part of the SCIF_EXIT message processing on the host, host sends a + * SCIF_REMOVE_NODE to itself corresponding to the card being removed. This + * starts a similar SCIF peer device teardown sequence on the host + * corresponding to the card being shut down. + * + * Card reset + * ---------- + * The case of interest here is when the card has not been previously shut down + * since most of the steps below are skipped in that case: + + * 1. cosm_stop(..) invokes hw_ops->stop(..) method of the base PCIe driver + * which unregisters the SCIF HW device resulting in scif_remove(..) being + * called on the host. + * 2. scif_remove(..) calls scif_disconnect_node(..) which results in a + * SCIF_EXIT message being sent to the card. + * 3. The card executes scif_stop() as part of SCIF_EXIT message + * processing. This results in the COSM endpoint on the card being closed and + * the SCIF host peer device on the card getting unregistered similar to + * steps 3, 4 and 5 for the card shutdown case above. scif_poll(..) on the + * host returns POLLHUP as a result. + * 4. On the host, card peer device unregister and SCIF HW remove(..) also + * subsequently complete. + * + * Card crash + * ---------- + * If a reset is issued after the card has crashed, there is no SCIF_DISCNT + * message from the card which would result in scif_poll(..) returning + * POLLHUP. In this case when the host SCIF driver sends a SCIF_REMOVE_NODE + * message to itself resulting in the card SCIF peer device being unregistered, + * this results in a scif_peer_release_dev -> scif_cleanup_scifdev-> + * scif_invalidate_ep call sequence which sets the endpoint state to + * DISCONNECTED and results in scif_poll(..) returning POLLHUP. + */ + +#define COSM_SCIF_BACKLOG 16 +#define COSM_HEARTBEAT_CHECK_DELTA_SEC 10 +#define COSM_HEARTBEAT_TIMEOUT_SEC \ + (COSM_HEARTBEAT_SEND_SEC + COSM_HEARTBEAT_CHECK_DELTA_SEC) +#define COSM_HEARTBEAT_TIMEOUT_MSEC (COSM_HEARTBEAT_TIMEOUT_SEC * MSEC_PER_SEC) + +static struct task_struct *server_thread; +static scif_epd_t listen_epd; + +/* Publish MIC card's shutdown status to user space MIC daemon */ +static void cosm_update_mic_status(struct cosm_device *cdev) +{ + if (cdev->shutdown_status_int != MIC_NOP) { + cosm_set_shutdown_status(cdev, cdev->shutdown_status_int); + cdev->shutdown_status_int = MIC_NOP; + } +} + +/* Store MIC card's shutdown status internally when it is received */ +static void cosm_shutdown_status_int(struct cosm_device *cdev, + enum mic_status shutdown_status) +{ + switch (shutdown_status) { + case MIC_HALTED: + case MIC_POWER_OFF: + case MIC_RESTART: + case MIC_CRASHED: + break; + default: + dev_err(&cdev->dev, "%s %d Unexpected shutdown_status %d\n", + __func__, __LINE__, shutdown_status); + return; + }; + cdev->shutdown_status_int = shutdown_status; + cdev->heartbeat_watchdog_enable = false; + + if (cdev->state != MIC_SHUTTING_DOWN) + cosm_set_state(cdev, MIC_SHUTTING_DOWN); +} + +/* Non-blocking recv. Read and process all available messages */ +static void cosm_scif_recv(struct cosm_device *cdev) +{ + struct cosm_msg msg; + int rc; + + while (1) { + rc = scif_recv(cdev->epd, &msg, sizeof(msg), 0); + if (!rc) { + break; + } else if (rc < 0) { + dev_dbg(&cdev->dev, "%s: %d rc %d\n", + __func__, __LINE__, rc); + break; + } + dev_dbg(&cdev->dev, "%s: %d rc %d id 0x%llx\n", + __func__, __LINE__, rc, msg.id); + + switch (msg.id) { + case COSM_MSG_SHUTDOWN_STATUS: + cosm_shutdown_status_int(cdev, msg.shutdown_status); + break; + case COSM_MSG_HEARTBEAT: + /* Nothing to do, heartbeat only unblocks scif_poll */ + break; + default: + dev_err(&cdev->dev, "%s: %d unknown msg.id %lld\n", + __func__, __LINE__, msg.id); + break; + } + } +} + +/* Publish crashed status for this MIC card */ +static void cosm_set_crashed(struct cosm_device *cdev) +{ + dev_err(&cdev->dev, "node alive timeout\n"); + cosm_shutdown_status_int(cdev, MIC_CRASHED); + cosm_update_mic_status(cdev); +} + +/* Send host time to the MIC card to sync system time between host and MIC */ +static void cosm_send_time(struct cosm_device *cdev) +{ + struct cosm_msg msg = { .id = COSM_MSG_SYNC_TIME }; + int rc; + + getnstimeofday64(&msg.timespec); + rc = scif_send(cdev->epd, &msg, sizeof(msg), SCIF_SEND_BLOCK); + if (rc < 0) + dev_err(&cdev->dev, "%s %d scif_send failed rc %d\n", + __func__, __LINE__, rc); +} + +/* + * Close this cosm_device's endpoint after its peer endpoint on the card has + * been closed. In all cases except MIC card crash POLLHUP on the host is + * triggered by the client's endpoint being closed. + */ +static void cosm_scif_close(struct cosm_device *cdev) +{ + /* + * Because SHUTDOWN_STATUS message is sent by the MIC cards in the + * reboot notifier when shutdown is still not complete, we notify mpssd + * to reset the card when SCIF endpoint is closed. + */ + cosm_update_mic_status(cdev); + scif_close(cdev->epd); + cdev->epd = NULL; + dev_dbg(&cdev->dev, "%s %d\n", __func__, __LINE__); +} + +/* + * Set card state to ONLINE when a new SCIF connection from a MIC card is + * received. Normally the state is BOOTING when the connection comes in, but can + * be ONLINE if cosm_client driver on the card was unloaded and then reloaded. + */ +static int cosm_set_online(struct cosm_device *cdev) +{ + int rc = 0; + + if (MIC_BOOTING == cdev->state || MIC_ONLINE == cdev->state) { + cdev->heartbeat_watchdog_enable = cdev->sysfs_heartbeat_enable; + cdev->epd = cdev->newepd; + if (cdev->state == MIC_BOOTING) + cosm_set_state(cdev, MIC_ONLINE); + cosm_send_time(cdev); + dev_dbg(&cdev->dev, "%s %d\n", __func__, __LINE__); + } else { + dev_warn(&cdev->dev, "%s %d not going online in state: %s\n", + __func__, __LINE__, cosm_state_string[cdev->state]); + rc = -EINVAL; + } + /* Drop reference acquired by bus_find_device in the server thread */ + put_device(&cdev->dev); + return rc; +} + +/* + * Work function for handling work for a SCIF connection from a particular MIC + * card. It first sets the card state to ONLINE and then calls scif_poll to + * block on activity such as incoming messages on the SCIF endpoint. When the + * endpoint is closed, the work function exits, completing its life cycle, from + * MIC card boot to card shutdown/reset/crash. + */ +void cosm_scif_work(struct work_struct *work) +{ + struct cosm_device *cdev = container_of(work, struct cosm_device, + scif_work); + struct scif_pollepd pollepd; + int rc; + + mutex_lock(&cdev->cosm_mutex); + if (cosm_set_online(cdev)) + goto exit; + + while (1) { + pollepd.epd = cdev->epd; + pollepd.events = POLLIN; + + /* Drop the mutex before blocking in scif_poll(..) */ + mutex_unlock(&cdev->cosm_mutex); + /* poll(..) with timeout on our endpoint */ + rc = scif_poll(&pollepd, 1, COSM_HEARTBEAT_TIMEOUT_MSEC); + mutex_lock(&cdev->cosm_mutex); + if (rc < 0) { + dev_err(&cdev->dev, "%s %d scif_poll rc %d\n", + __func__, __LINE__, rc); + continue; + } + + /* There is a message from the card */ + if (pollepd.revents & POLLIN) + cosm_scif_recv(cdev); + + /* The peer endpoint is closed or this endpoint disconnected */ + if (pollepd.revents & POLLHUP) { + cosm_scif_close(cdev); + break; + } + + /* Did we timeout from poll? */ + if (!rc && cdev->heartbeat_watchdog_enable) + cosm_set_crashed(cdev); + } +exit: + dev_dbg(&cdev->dev, "%s %d exiting\n", __func__, __LINE__); + mutex_unlock(&cdev->cosm_mutex); +} + +/* + * COSM SCIF server thread function. Accepts incoming SCIF connections from MIC + * cards, finds the correct cosm_device to associate that connection with and + * schedules individual work items for each MIC card. + */ +static int cosm_scif_server(void *unused) +{ + struct cosm_device *cdev; + scif_epd_t newepd; + struct scif_port_id port_id; + int rc; + + allow_signal(SIGKILL); + + while (!kthread_should_stop()) { + rc = scif_accept(listen_epd, &port_id, &newepd, + SCIF_ACCEPT_SYNC); + if (rc < 0) { + if (-ERESTARTSYS != rc) + pr_err("%s %d rc %d\n", __func__, __LINE__, rc); + continue; + } + + /* + * Associate the incoming connection with a particular + * cosm_device, COSM device ID == SCIF node ID - 1 + */ + cdev = cosm_find_cdev_by_id(port_id.node - 1); + if (!cdev) + continue; + cdev->newepd = newepd; + schedule_work(&cdev->scif_work); + } + + pr_debug("%s %d Server thread stopped\n", __func__, __LINE__); + return 0; +} + +static int cosm_scif_listen(void) +{ + int rc; + + listen_epd = scif_open(); + if (!listen_epd) { + pr_err("%s %d scif_open failed\n", __func__, __LINE__); + return -ENOMEM; + } + + rc = scif_bind(listen_epd, SCIF_COSM_LISTEN_PORT); + if (rc < 0) { + pr_err("%s %d scif_bind failed rc %d\n", + __func__, __LINE__, rc); + goto err; + } + + rc = scif_listen(listen_epd, COSM_SCIF_BACKLOG); + if (rc < 0) { + pr_err("%s %d scif_listen rc %d\n", __func__, __LINE__, rc); + goto err; + } + pr_debug("%s %d listen_epd set up\n", __func__, __LINE__); + return 0; +err: + scif_close(listen_epd); + listen_epd = NULL; + return rc; +} + +static void cosm_scif_listen_exit(void) +{ + pr_debug("%s %d closing listen_epd\n", __func__, __LINE__); + if (listen_epd) { + scif_close(listen_epd); + listen_epd = NULL; + } +} + +/* + * Create a listening SCIF endpoint and a server kthread which accepts incoming + * SCIF connections from MIC cards + */ +int cosm_scif_init(void) +{ + int rc = cosm_scif_listen(); + + if (rc) { + pr_err("%s %d cosm_scif_listen rc %d\n", + __func__, __LINE__, rc); + goto err; + } + + server_thread = kthread_run(cosm_scif_server, NULL, "cosm_server"); + if (IS_ERR(server_thread)) { + rc = PTR_ERR(server_thread); + pr_err("%s %d kthread_run rc %d\n", __func__, __LINE__, rc); + goto listen_exit; + } + return 0; +listen_exit: + cosm_scif_listen_exit(); +err: + return rc; +} + +/* Stop the running server thread and close the listening SCIF endpoint */ +void cosm_scif_exit(void) +{ + int rc; + + if (!IS_ERR_OR_NULL(server_thread)) { + rc = send_sig(SIGKILL, server_thread, 0); + if (rc) { + pr_err("%s %d send_sig rc %d\n", + __func__, __LINE__, rc); + return; + } + kthread_stop(server_thread); + } + + cosm_scif_listen_exit(); +} diff --git a/drivers/misc/mic/cosm/cosm_sysfs.c b/drivers/misc/mic/cosm/cosm_sysfs.c new file mode 100644 index 000000000000..29d6863b6e59 --- /dev/null +++ b/drivers/misc/mic/cosm/cosm_sysfs.c @@ -0,0 +1,461 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC Coprocessor State Management (COSM) Driver + * + */ +#include <linux/slab.h> +#include "cosm_main.h" + +/* + * A state-to-string lookup table, for exposing a human readable state + * via sysfs. Always keep in sync with enum cosm_states + */ +const char * const cosm_state_string[] = { + [MIC_READY] = "ready", + [MIC_BOOTING] = "booting", + [MIC_ONLINE] = "online", + [MIC_SHUTTING_DOWN] = "shutting_down", + [MIC_RESETTING] = "resetting", + [MIC_RESET_FAILED] = "reset_failed", +}; + +/* + * A shutdown-status-to-string lookup table, for exposing a human + * readable state via sysfs. Always keep in sync with enum cosm_shutdown_status + */ +const char * const cosm_shutdown_status_string[] = { + [MIC_NOP] = "nop", + [MIC_CRASHED] = "crashed", + [MIC_HALTED] = "halted", + [MIC_POWER_OFF] = "poweroff", + [MIC_RESTART] = "restart", +}; + +void cosm_set_shutdown_status(struct cosm_device *cdev, u8 shutdown_status) +{ + dev_dbg(&cdev->dev, "Shutdown Status %s -> %s\n", + cosm_shutdown_status_string[cdev->shutdown_status], + cosm_shutdown_status_string[shutdown_status]); + cdev->shutdown_status = shutdown_status; +} + +void cosm_set_state(struct cosm_device *cdev, u8 state) +{ + dev_dbg(&cdev->dev, "State %s -> %s\n", + cosm_state_string[cdev->state], + cosm_state_string[state]); + cdev->state = state; + sysfs_notify_dirent(cdev->state_sysfs); +} + +static ssize_t +family_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + + if (!cdev) + return -EINVAL; + + return cdev->hw_ops->family(cdev, buf); +} +static DEVICE_ATTR_RO(family); + +static ssize_t +stepping_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + + if (!cdev) + return -EINVAL; + + return cdev->hw_ops->stepping(cdev, buf); +} +static DEVICE_ATTR_RO(stepping); + +static ssize_t +state_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + + if (!cdev || cdev->state >= MIC_LAST) + return -EINVAL; + + return scnprintf(buf, PAGE_SIZE, "%s\n", + cosm_state_string[cdev->state]); +} + +static ssize_t +state_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + int rc; + + if (!cdev) + return -EINVAL; + + if (sysfs_streq(buf, "boot")) { + rc = cosm_start(cdev); + goto done; + } + if (sysfs_streq(buf, "reset")) { + rc = cosm_reset(cdev); + goto done; + } + + if (sysfs_streq(buf, "shutdown")) { + rc = cosm_shutdown(cdev); + goto done; + } + rc = -EINVAL; +done: + if (rc) + count = rc; + return count; +} +static DEVICE_ATTR_RW(state); + +static ssize_t shutdown_status_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + + if (!cdev || cdev->shutdown_status >= MIC_STATUS_LAST) + return -EINVAL; + + return scnprintf(buf, PAGE_SIZE, "%s\n", + cosm_shutdown_status_string[cdev->shutdown_status]); +} +static DEVICE_ATTR_RO(shutdown_status); + +static ssize_t +heartbeat_enable_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + + if (!cdev) + return -EINVAL; + + return scnprintf(buf, PAGE_SIZE, "%d\n", cdev->sysfs_heartbeat_enable); +} + +static ssize_t +heartbeat_enable_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + int enable; + int ret; + + if (!cdev) + return -EINVAL; + + mutex_lock(&cdev->cosm_mutex); + ret = kstrtoint(buf, 10, &enable); + if (ret) + goto unlock; + + cdev->sysfs_heartbeat_enable = enable; + /* if state is not online, cdev->heartbeat_watchdog_enable is 0 */ + if (cdev->state == MIC_ONLINE) + cdev->heartbeat_watchdog_enable = enable; + ret = count; +unlock: + mutex_unlock(&cdev->cosm_mutex); + return ret; +} +static DEVICE_ATTR_RW(heartbeat_enable); + +static ssize_t +cmdline_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + char *cmdline; + + if (!cdev) + return -EINVAL; + + cmdline = cdev->cmdline; + + if (cmdline) + return scnprintf(buf, PAGE_SIZE, "%s\n", cmdline); + return 0; +} + +static ssize_t +cmdline_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + + if (!cdev) + return -EINVAL; + + mutex_lock(&cdev->cosm_mutex); + kfree(cdev->cmdline); + + cdev->cmdline = kmalloc(count + 1, GFP_KERNEL); + if (!cdev->cmdline) { + count = -ENOMEM; + goto unlock; + } + + strncpy(cdev->cmdline, buf, count); + + if (cdev->cmdline[count - 1] == '\n') + cdev->cmdline[count - 1] = '\0'; + else + cdev->cmdline[count] = '\0'; +unlock: + mutex_unlock(&cdev->cosm_mutex); + return count; +} +static DEVICE_ATTR_RW(cmdline); + +static ssize_t +firmware_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + char *firmware; + + if (!cdev) + return -EINVAL; + + firmware = cdev->firmware; + + if (firmware) + return scnprintf(buf, PAGE_SIZE, "%s\n", firmware); + return 0; +} + +static ssize_t +firmware_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + + if (!cdev) + return -EINVAL; + + mutex_lock(&cdev->cosm_mutex); + kfree(cdev->firmware); + + cdev->firmware = kmalloc(count + 1, GFP_KERNEL); + if (!cdev->firmware) { + count = -ENOMEM; + goto unlock; + } + strncpy(cdev->firmware, buf, count); + + if (cdev->firmware[count - 1] == '\n') + cdev->firmware[count - 1] = '\0'; + else + cdev->firmware[count] = '\0'; +unlock: + mutex_unlock(&cdev->cosm_mutex); + return count; +} +static DEVICE_ATTR_RW(firmware); + +static ssize_t +ramdisk_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + char *ramdisk; + + if (!cdev) + return -EINVAL; + + ramdisk = cdev->ramdisk; + + if (ramdisk) + return scnprintf(buf, PAGE_SIZE, "%s\n", ramdisk); + return 0; +} + +static ssize_t +ramdisk_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + + if (!cdev) + return -EINVAL; + + mutex_lock(&cdev->cosm_mutex); + kfree(cdev->ramdisk); + + cdev->ramdisk = kmalloc(count + 1, GFP_KERNEL); + if (!cdev->ramdisk) { + count = -ENOMEM; + goto unlock; + } + + strncpy(cdev->ramdisk, buf, count); + + if (cdev->ramdisk[count - 1] == '\n') + cdev->ramdisk[count - 1] = '\0'; + else + cdev->ramdisk[count] = '\0'; +unlock: + mutex_unlock(&cdev->cosm_mutex); + return count; +} +static DEVICE_ATTR_RW(ramdisk); + +static ssize_t +bootmode_show(struct device *dev, struct device_attribute *attr, char *buf) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + char *bootmode; + + if (!cdev) + return -EINVAL; + + bootmode = cdev->bootmode; + + if (bootmode) + return scnprintf(buf, PAGE_SIZE, "%s\n", bootmode); + return 0; +} + +static ssize_t +bootmode_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + + if (!cdev) + return -EINVAL; + + if (!sysfs_streq(buf, "linux") && !sysfs_streq(buf, "flash")) + return -EINVAL; + + mutex_lock(&cdev->cosm_mutex); + kfree(cdev->bootmode); + + cdev->bootmode = kmalloc(count + 1, GFP_KERNEL); + if (!cdev->bootmode) { + count = -ENOMEM; + goto unlock; + } + + strncpy(cdev->bootmode, buf, count); + + if (cdev->bootmode[count - 1] == '\n') + cdev->bootmode[count - 1] = '\0'; + else + cdev->bootmode[count] = '\0'; +unlock: + mutex_unlock(&cdev->cosm_mutex); + return count; +} +static DEVICE_ATTR_RW(bootmode); + +static ssize_t +log_buf_addr_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + + if (!cdev) + return -EINVAL; + + return scnprintf(buf, PAGE_SIZE, "%p\n", cdev->log_buf_addr); +} + +static ssize_t +log_buf_addr_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + int ret; + unsigned long addr; + + if (!cdev) + return -EINVAL; + + ret = kstrtoul(buf, 16, &addr); + if (ret) + goto exit; + + cdev->log_buf_addr = (void *)addr; + ret = count; +exit: + return ret; +} +static DEVICE_ATTR_RW(log_buf_addr); + +static ssize_t +log_buf_len_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + + if (!cdev) + return -EINVAL; + + return scnprintf(buf, PAGE_SIZE, "%p\n", cdev->log_buf_len); +} + +static ssize_t +log_buf_len_store(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct cosm_device *cdev = dev_get_drvdata(dev); + int ret; + unsigned long addr; + + if (!cdev) + return -EINVAL; + + ret = kstrtoul(buf, 16, &addr); + if (ret) + goto exit; + + cdev->log_buf_len = (int *)addr; + ret = count; +exit: + return ret; +} +static DEVICE_ATTR_RW(log_buf_len); + +static struct attribute *cosm_default_attrs[] = { + &dev_attr_family.attr, + &dev_attr_stepping.attr, + &dev_attr_state.attr, + &dev_attr_shutdown_status.attr, + &dev_attr_heartbeat_enable.attr, + &dev_attr_cmdline.attr, + &dev_attr_firmware.attr, + &dev_attr_ramdisk.attr, + &dev_attr_bootmode.attr, + &dev_attr_log_buf_addr.attr, + &dev_attr_log_buf_len.attr, + + NULL +}; + +ATTRIBUTE_GROUPS(cosm_default); + +void cosm_sysfs_init(struct cosm_device *cdev) +{ + cdev->attr_group = cosm_default_groups; +} diff --git a/drivers/misc/mic/cosm_client/Makefile b/drivers/misc/mic/cosm_client/Makefile new file mode 100644 index 000000000000..6f751a519a09 --- /dev/null +++ b/drivers/misc/mic/cosm_client/Makefile @@ -0,0 +1,7 @@ +# +# Makefile - Intel MIC COSM Client Driver +# Copyright(c) 2015, Intel Corporation. +# +obj-$(CONFIG_MIC_COSM) += cosm_client.o + +cosm_client-objs += cosm_scif_client.o diff --git a/drivers/misc/mic/cosm_client/cosm_scif_client.c b/drivers/misc/mic/cosm_client/cosm_scif_client.c new file mode 100644 index 000000000000..03e98bf1ac15 --- /dev/null +++ b/drivers/misc/mic/cosm_client/cosm_scif_client.c @@ -0,0 +1,275 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * The full GNU General Public License is included in this distribution in + * the file called "COPYING". + * + * Intel MIC COSM Client Driver + * + */ +#include <linux/module.h> +#include <linux/delay.h> +#include <linux/reboot.h> +#include <linux/kthread.h> +#include "../cosm/cosm_main.h" + +#define COSM_SCIF_MAX_RETRIES 10 +#define COSM_HEARTBEAT_SEND_MSEC (COSM_HEARTBEAT_SEND_SEC * MSEC_PER_SEC) + +static struct task_struct *client_thread; +static scif_epd_t client_epd; +static struct scif_peer_dev *client_spdev; + +/* + * Reboot notifier: receives shutdown status from the OS and communicates it + * back to the COSM process on the host + */ +static int cosm_reboot_event(struct notifier_block *this, unsigned long event, + void *ptr) +{ + struct cosm_msg msg = { .id = COSM_MSG_SHUTDOWN_STATUS }; + int rc; + + event = (event == SYS_RESTART) ? SYSTEM_RESTART : event; + dev_info(&client_spdev->dev, "%s %d received event %ld\n", + __func__, __LINE__, event); + + msg.shutdown_status = event; + rc = scif_send(client_epd, &msg, sizeof(msg), SCIF_SEND_BLOCK); + if (rc < 0) + dev_err(&client_spdev->dev, "%s %d scif_send rc %d\n", + __func__, __LINE__, rc); + + return NOTIFY_DONE; +} + +static struct notifier_block cosm_reboot = { + .notifier_call = cosm_reboot_event, +}; + +/* Set system time from timespec value received from the host */ +static void cosm_set_time(struct cosm_msg *msg) +{ + int rc = do_settimeofday64(&msg->timespec); + + if (rc) + dev_err(&client_spdev->dev, "%s: %d settimeofday rc %d\n", + __func__, __LINE__, rc); +} + +/* COSM client receive message processing */ +static void cosm_client_recv(void) +{ + struct cosm_msg msg; + int rc; + + while (1) { + rc = scif_recv(client_epd, &msg, sizeof(msg), 0); + if (!rc) { + return; + } else if (rc < 0) { + dev_err(&client_spdev->dev, "%s: %d rc %d\n", + __func__, __LINE__, rc); + return; + } + + dev_dbg(&client_spdev->dev, "%s: %d rc %d id 0x%llx\n", + __func__, __LINE__, rc, msg.id); + + switch (msg.id) { + case COSM_MSG_SYNC_TIME: + cosm_set_time(&msg); + break; + case COSM_MSG_SHUTDOWN: + orderly_poweroff(true); + break; + default: + dev_err(&client_spdev->dev, "%s: %d unknown id %lld\n", + __func__, __LINE__, msg.id); + break; + } + } +} + +/* Initiate connection to the COSM server on the host */ +static int cosm_scif_connect(void) +{ + struct scif_port_id port_id; + int i, rc; + + client_epd = scif_open(); + if (!client_epd) { + dev_err(&client_spdev->dev, "%s %d scif_open failed\n", + __func__, __LINE__); + return -ENOMEM; + } + + port_id.node = 0; + port_id.port = SCIF_COSM_LISTEN_PORT; + + for (i = 0; i < COSM_SCIF_MAX_RETRIES; i++) { + rc = scif_connect(client_epd, &port_id); + if (rc < 0) + msleep(1000); + else + break; + } + + if (rc < 0) { + dev_err(&client_spdev->dev, "%s %d scif_connect rc %d\n", + __func__, __LINE__, rc); + scif_close(client_epd); + client_epd = NULL; + } + return rc < 0 ? rc : 0; +} + +/* Close host SCIF connection */ +static void cosm_scif_connect_exit(void) +{ + if (client_epd) { + scif_close(client_epd); + client_epd = NULL; + } +} + +/* + * COSM SCIF client thread function: waits for messages from the host and sends + * a heartbeat to the host + */ +static int cosm_scif_client(void *unused) +{ + struct cosm_msg msg = { .id = COSM_MSG_HEARTBEAT }; + struct scif_pollepd pollepd; + int rc; + + allow_signal(SIGKILL); + + while (!kthread_should_stop()) { + pollepd.epd = client_epd; + pollepd.events = POLLIN; + + rc = scif_poll(&pollepd, 1, COSM_HEARTBEAT_SEND_MSEC); + if (rc < 0) { + if (-EINTR != rc) + dev_err(&client_spdev->dev, + "%s %d scif_poll rc %d\n", + __func__, __LINE__, rc); + continue; + } + + if (pollepd.revents & POLLIN) + cosm_client_recv(); + + msg.id = COSM_MSG_HEARTBEAT; + rc = scif_send(client_epd, &msg, sizeof(msg), SCIF_SEND_BLOCK); + if (rc < 0) + dev_err(&client_spdev->dev, "%s %d scif_send rc %d\n", + __func__, __LINE__, rc); + } + + dev_dbg(&client_spdev->dev, "%s %d Client thread stopped\n", + __func__, __LINE__); + return 0; +} + +static void cosm_scif_probe(struct scif_peer_dev *spdev) +{ + int rc; + + dev_dbg(&spdev->dev, "%s %d: dnode %d\n", + __func__, __LINE__, spdev->dnode); + + /* We are only interested in the host with spdev->dnode == 0 */ + if (spdev->dnode) + return; + + client_spdev = spdev; + rc = cosm_scif_connect(); + if (rc) + goto exit; + + rc = register_reboot_notifier(&cosm_reboot); + if (rc) { + dev_err(&spdev->dev, + "reboot notifier registration failed rc %d\n", rc); + goto connect_exit; + } + + client_thread = kthread_run(cosm_scif_client, NULL, "cosm_client"); + if (IS_ERR(client_thread)) { + rc = PTR_ERR(client_thread); + dev_err(&spdev->dev, "%s %d kthread_run rc %d\n", + __func__, __LINE__, rc); + goto unreg_reboot; + } + return; +unreg_reboot: + unregister_reboot_notifier(&cosm_reboot); +connect_exit: + cosm_scif_connect_exit(); +exit: + client_spdev = NULL; +} + +static void cosm_scif_remove(struct scif_peer_dev *spdev) +{ + int rc; + + dev_dbg(&spdev->dev, "%s %d: dnode %d\n", + __func__, __LINE__, spdev->dnode); + + if (spdev->dnode) + return; + + if (!IS_ERR_OR_NULL(client_thread)) { + rc = send_sig(SIGKILL, client_thread, 0); + if (rc) { + pr_err("%s %d send_sig rc %d\n", + __func__, __LINE__, rc); + return; + } + kthread_stop(client_thread); + } + unregister_reboot_notifier(&cosm_reboot); + cosm_scif_connect_exit(); + client_spdev = NULL; +} + +static struct scif_client scif_client_cosm = { + .name = KBUILD_MODNAME, + .probe = cosm_scif_probe, + .remove = cosm_scif_remove, +}; + +static int __init cosm_client_init(void) +{ + int rc = scif_client_register(&scif_client_cosm); + + if (rc) + pr_err("scif_client_register failed rc %d\n", rc); + return rc; +} + +static void __exit cosm_client_exit(void) +{ + scif_client_unregister(&scif_client_cosm); +} + +module_init(cosm_client_init); +module_exit(cosm_client_exit); + +MODULE_AUTHOR("Intel Corporation"); +MODULE_DESCRIPTION("Intel(R) MIC card OS state management client driver"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/misc/mic/host/Makefile b/drivers/misc/mic/host/Makefile index c2197f999394..004d3db0f990 100644 --- a/drivers/misc/mic/host/Makefile +++ b/drivers/misc/mic/host/Makefile @@ -5,7 +5,6 @@ obj-$(CONFIG_INTEL_MIC_HOST) += mic_host.o mic_host-objs := mic_main.o mic_host-objs += mic_x100.o -mic_host-objs += mic_sysfs.o mic_host-objs += mic_smpt.o mic_host-objs += mic_intr.o mic_host-objs += mic_boot.o diff --git a/drivers/misc/mic/host/mic_boot.c b/drivers/misc/mic/host/mic_boot.c index e5f6a5e7bca1..7845564dff64 100644 --- a/drivers/misc/mic/host/mic_boot.c +++ b/drivers/misc/mic/host/mic_boot.c @@ -22,9 +22,9 @@ #include <linux/firmware.h> #include <linux/pci.h> #include <linux/kmod.h> - #include <linux/mic_common.h> #include <linux/mic_bus.h> +#include "../bus/scif_bus.h" #include "../common/mic_dev.h" #include "mic_device.h" #include "mic_smpt.h" @@ -99,7 +99,7 @@ static int __mic_dma_map_sg(struct device *dev, struct scatterlist *sg, int i, j, ret; dma_addr_t da; - ret = dma_map_sg(mdev->sdev->parent, sg, nents, dir); + ret = dma_map_sg(&mdev->pdev->dev, sg, nents, dir); if (ret <= 0) return 0; @@ -115,7 +115,7 @@ err: mic_unmap(mdev, sg_dma_address(s), s->length); sg_dma_address(s) = mic_to_dma_addr(mdev, sg_dma_address(s)); } - dma_unmap_sg(mdev->sdev->parent, sg, nents, dir); + dma_unmap_sg(&mdev->pdev->dev, sg, nents, dir); return 0; } @@ -135,7 +135,7 @@ static void __mic_dma_unmap_sg(struct device *dev, mic_unmap(mdev, sg_dma_address(s), s->length); sg_dma_address(s) = da; } - dma_unmap_sg(mdev->sdev->parent, sg, nents, dir); + dma_unmap_sg(&mdev->pdev->dev, sg, nents, dir); } static struct dma_map_ops __mic_dma_ops = { @@ -270,48 +270,13 @@ static struct mbus_hw_ops mbus_hw_ops = { .ack_interrupt = _mic_ack_interrupt, }; -/** - * mic_reset - Reset the MIC device. - * @mdev: pointer to mic_device instance - */ -static void mic_reset(struct mic_device *mdev) -{ - int i; - -#define MIC_RESET_TO (45) - - reinit_completion(&mdev->reset_wait); - mdev->ops->reset_fw_ready(mdev); - mdev->ops->reset(mdev); - - for (i = 0; i < MIC_RESET_TO; i++) { - if (mdev->ops->is_fw_ready(mdev)) - goto done; - /* - * Resets typically take 10s of seconds to complete. - * Since an MMIO read is required to check if the - * firmware is ready or not, a 1 second delay works nicely. - */ - msleep(1000); - } - mic_set_state(mdev, MIC_RESET_FAILED); -done: - complete_all(&mdev->reset_wait); -} - /* Initialize the MIC bootparams */ void mic_bootparam_init(struct mic_device *mdev) { struct mic_bootparam *bootparam = mdev->dp; bootparam->magic = cpu_to_le32(MIC_MAGIC); - bootparam->c2h_shutdown_db = mdev->shutdown_db; - bootparam->h2c_shutdown_db = -1; bootparam->h2c_config_db = -1; - bootparam->shutdown_status = 0; - bootparam->shutdown_card = 0; - /* Total nodes = number of MICs + 1 for self node */ - bootparam->tot_nodes = atomic_read(&g_num_mics) + 1; bootparam->node_id = mdev->id + 1; bootparam->scif_host_dma_addr = 0x0; bootparam->scif_card_dma_addr = 0x0; @@ -319,6 +284,26 @@ void mic_bootparam_init(struct mic_device *mdev) bootparam->h2c_scif_db = -1; } +static inline struct mic_device *cosmdev_to_mdev(struct cosm_device *cdev) +{ + return dev_get_drvdata(cdev->dev.parent); +} + +static void _mic_reset(struct cosm_device *cdev) +{ + struct mic_device *mdev = cosmdev_to_mdev(cdev); + + mdev->ops->reset_fw_ready(mdev); + mdev->ops->reset(mdev); +} + +static bool _mic_ready(struct cosm_device *cdev) +{ + struct mic_device *mdev = cosmdev_to_mdev(cdev); + + return mdev->ops->is_fw_ready(mdev); +} + /** * mic_request_dma_chans - Request DMA channels * @mdev: pointer to mic_device instance @@ -336,14 +321,14 @@ static int mic_request_dma_chans(struct mic_device *mdev) do { chan = dma_request_channel(mask, mdev->ops->dma_filter, - mdev->sdev->parent); + &mdev->pdev->dev); if (chan) { mdev->dma_ch[mdev->num_dma_ch++] = chan; if (mdev->num_dma_ch >= MIC_MAX_DMA_CHAN) break; } } while (chan); - dev_info(mdev->sdev->parent, "DMA channels # %d\n", mdev->num_dma_ch); + dev_info(&mdev->pdev->dev, "DMA channels # %d\n", mdev->num_dma_ch); return mdev->num_dma_ch; } @@ -365,34 +350,24 @@ static void mic_free_dma_chans(struct mic_device *mdev) } /** - * mic_start - Start the MIC. - * @mdev: pointer to mic_device instance - * @buf: buffer containing boot string including firmware/ramdisk path. + * _mic_start - Start the MIC. + * @cdev: pointer to cosm_device instance + * @id: MIC device id/index provided by COSM used in other drivers like SCIF * * This function prepares an MIC for boot and initiates boot. * RETURNS: An appropriate -ERRNO error value on error, or zero for success. + * + * For all cosm_hw_ops the caller holds a mutex to ensure serialization. */ -int mic_start(struct mic_device *mdev, const char *buf) +static int _mic_start(struct cosm_device *cdev, int id) { + struct mic_device *mdev = cosmdev_to_mdev(cdev); int rc; - mutex_lock(&mdev->mic_mutex); + mic_bootparam_init(mdev); -retry: - if (MIC_OFFLINE != mdev->state) { - rc = -EINVAL; - goto unlock_ret; - } - if (!mdev->ops->is_fw_ready(mdev)) { - mic_reset(mdev); - /* - * The state will either be MIC_OFFLINE if the reset succeeded - * or MIC_RESET_FAILED if the firmware reset failed. - */ - goto retry; - } - mdev->dma_mbdev = mbus_register_device(mdev->sdev->parent, + mdev->dma_mbdev = mbus_register_device(&mdev->pdev->dev, MBUS_DEV_DMA_HOST, &mic_dma_ops, - &mbus_hw_ops, mdev->mmio.va); + &mbus_hw_ops, id, mdev->mmio.va); if (IS_ERR(mdev->dma_mbdev)) { rc = PTR_ERR(mdev->dma_mbdev); goto unlock_ret; @@ -401,16 +376,18 @@ retry: rc = -ENODEV; goto dma_remove; } - mdev->scdev = scif_register_device(mdev->sdev->parent, MIC_SCIF_DEV, + mdev->scdev = scif_register_device(&mdev->pdev->dev, MIC_SCIF_DEV, &__mic_dma_ops, &scif_hw_ops, - mdev->id + 1, 0, &mdev->mmio, + id + 1, 0, &mdev->mmio, &mdev->aper, mdev->dp, NULL, - mdev->dma_ch, mdev->num_dma_ch); + mdev->dma_ch, mdev->num_dma_ch, + true); if (IS_ERR(mdev->scdev)) { rc = PTR_ERR(mdev->scdev); goto dma_free; } - rc = mdev->ops->load_mic_fw(mdev, buf); + + rc = mdev->ops->load_mic_fw(mdev, NULL); if (rc) goto scif_remove; mic_smpt_restore(mdev); @@ -419,7 +396,6 @@ retry: mdev->ops->write_spad(mdev, MIC_DPLO_SPAD, mdev->dp_dma_addr); mdev->ops->write_spad(mdev, MIC_DPHI_SPAD, mdev->dp_dma_addr >> 32); mdev->ops->send_firmware_intr(mdev); - mic_set_state(mdev, MIC_ONLINE); goto unlock_ret; scif_remove: scif_unregister_device(mdev->scdev); @@ -428,198 +404,79 @@ dma_free: dma_remove: mbus_unregister_device(mdev->dma_mbdev); unlock_ret: - mutex_unlock(&mdev->mic_mutex); return rc; } /** - * mic_stop - Prepare the MIC for reset and trigger reset. - * @mdev: pointer to mic_device instance + * _mic_stop - Prepare the MIC for reset and trigger reset. + * @cdev: pointer to cosm_device instance * @force: force a MIC to reset even if it is already offline. * * RETURNS: None. */ -void mic_stop(struct mic_device *mdev, bool force) -{ - mutex_lock(&mdev->mic_mutex); - if (MIC_OFFLINE != mdev->state || force) { - scif_unregister_device(mdev->scdev); - mic_virtio_reset_devices(mdev); - mic_free_dma_chans(mdev); - mbus_unregister_device(mdev->dma_mbdev); - mic_bootparam_init(mdev); - mic_reset(mdev); - if (MIC_RESET_FAILED == mdev->state) - goto unlock; - mic_set_shutdown_status(mdev, MIC_NOP); - if (MIC_SUSPENDED != mdev->state) - mic_set_state(mdev, MIC_OFFLINE); - } -unlock: - mutex_unlock(&mdev->mic_mutex); -} - -/** - * mic_shutdown - Initiate MIC shutdown. - * @mdev: pointer to mic_device instance - * - * RETURNS: None. - */ -void mic_shutdown(struct mic_device *mdev) +static void _mic_stop(struct cosm_device *cdev, bool force) { - struct mic_bootparam *bootparam = mdev->dp; - s8 db = bootparam->h2c_shutdown_db; - - mutex_lock(&mdev->mic_mutex); - if (MIC_ONLINE == mdev->state && db != -1) { - bootparam->shutdown_card = 1; - mdev->ops->send_intr(mdev, db); - mic_set_state(mdev, MIC_SHUTTING_DOWN); - } - mutex_unlock(&mdev->mic_mutex); -} - -/** - * mic_shutdown_work - Handle shutdown interrupt from MIC. - * @work: The work structure. - * - * This work is scheduled whenever the host has received a shutdown - * interrupt from the MIC. - */ -void mic_shutdown_work(struct work_struct *work) -{ - struct mic_device *mdev = container_of(work, struct mic_device, - shutdown_work); - struct mic_bootparam *bootparam = mdev->dp; - - mutex_lock(&mdev->mic_mutex); - mic_set_shutdown_status(mdev, bootparam->shutdown_status); - bootparam->shutdown_status = 0; + struct mic_device *mdev = cosmdev_to_mdev(cdev); /* - * if state is MIC_SUSPENDED, OSPM suspend is in progress. We do not - * change the state here so as to prevent users from booting the card - * during and after the suspend operation. + * Since SCIF handles card shutdown and reset (using COSM), it will + * will be the first to be registered and the last to be + * unregistered. */ - if (MIC_SHUTTING_DOWN != mdev->state && - MIC_SUSPENDED != mdev->state) - mic_set_state(mdev, MIC_SHUTTING_DOWN); - mutex_unlock(&mdev->mic_mutex); + mic_virtio_reset_devices(mdev); + scif_unregister_device(mdev->scdev); + mic_free_dma_chans(mdev); + mbus_unregister_device(mdev->dma_mbdev); + mic_bootparam_init(mdev); } -/** - * mic_reset_trigger_work - Trigger MIC reset. - * @work: The work structure. - * - * This work is scheduled whenever the host wants to reset the MIC. - */ -void mic_reset_trigger_work(struct work_struct *work) +static ssize_t _mic_family(struct cosm_device *cdev, char *buf) { - struct mic_device *mdev = container_of(work, struct mic_device, - reset_trigger_work); + struct mic_device *mdev = cosmdev_to_mdev(cdev); + static const char *family[MIC_FAMILY_LAST] = { "x100", "Unknown" }; - mic_stop(mdev, false); + return scnprintf(buf, PAGE_SIZE, "%s\n", family[mdev->family]); } -/** - * mic_complete_resume - Complete MIC Resume after an OSPM suspend/hibernate - * event. - * @mdev: pointer to mic_device instance - * - * RETURNS: None. - */ -void mic_complete_resume(struct mic_device *mdev) +static ssize_t _mic_stepping(struct cosm_device *cdev, char *buf) { - if (mdev->state != MIC_SUSPENDED) { - dev_warn(mdev->sdev->parent, "state %d should be %d\n", - mdev->state, MIC_SUSPENDED); - return; - } - - /* Make sure firmware is ready */ - if (!mdev->ops->is_fw_ready(mdev)) - mic_stop(mdev, true); + struct mic_device *mdev = cosmdev_to_mdev(cdev); + const char *string = "??"; - mutex_lock(&mdev->mic_mutex); - mic_set_state(mdev, MIC_OFFLINE); - mutex_unlock(&mdev->mic_mutex); -} - -/** - * mic_prepare_suspend - Handle suspend notification for the MIC device. - * @mdev: pointer to mic_device instance - * - * RETURNS: None. - */ -void mic_prepare_suspend(struct mic_device *mdev) -{ - unsigned long timeout; - -#define MIC_SUSPEND_TIMEOUT (60 * HZ) - - mutex_lock(&mdev->mic_mutex); - switch (mdev->state) { - case MIC_OFFLINE: - /* - * Card is already offline. Set state to MIC_SUSPENDED - * to prevent users from booting the card. - */ - mic_set_state(mdev, MIC_SUSPENDED); - mutex_unlock(&mdev->mic_mutex); + switch (mdev->stepping) { + case MIC_A0_STEP: + string = "A0"; break; - case MIC_ONLINE: - /* - * Card is online. Set state to MIC_SUSPENDING and notify - * MIC user space daemon which will issue card - * shutdown and reset. - */ - mic_set_state(mdev, MIC_SUSPENDING); - mutex_unlock(&mdev->mic_mutex); - timeout = wait_for_completion_timeout(&mdev->reset_wait, - MIC_SUSPEND_TIMEOUT); - /* Force reset the card if the shutdown completion timed out */ - if (!timeout) { - mutex_lock(&mdev->mic_mutex); - mic_set_state(mdev, MIC_SUSPENDED); - mutex_unlock(&mdev->mic_mutex); - mic_stop(mdev, true); - } + case MIC_B0_STEP: + string = "B0"; + break; + case MIC_B1_STEP: + string = "B1"; break; - case MIC_SHUTTING_DOWN: - /* - * Card is shutting down. Set state to MIC_SUSPENDED - * to prevent further boot of the card. - */ - mic_set_state(mdev, MIC_SUSPENDED); - mutex_unlock(&mdev->mic_mutex); - timeout = wait_for_completion_timeout(&mdev->reset_wait, - MIC_SUSPEND_TIMEOUT); - /* Force reset the card if the shutdown completion timed out */ - if (!timeout) - mic_stop(mdev, true); + case MIC_C0_STEP: + string = "C0"; break; default: - mutex_unlock(&mdev->mic_mutex); break; } + return scnprintf(buf, PAGE_SIZE, "%s\n", string); } -/** - * mic_suspend - Initiate MIC suspend. Suspend merely issues card shutdown. - * @mdev: pointer to mic_device instance - * - * RETURNS: None. - */ -void mic_suspend(struct mic_device *mdev) +static struct mic_mw *_mic_aper(struct cosm_device *cdev) { - struct mic_bootparam *bootparam = mdev->dp; - s8 db = bootparam->h2c_shutdown_db; + struct mic_device *mdev = cosmdev_to_mdev(cdev); - mutex_lock(&mdev->mic_mutex); - if (MIC_SUSPENDING == mdev->state && db != -1) { - bootparam->shutdown_card = 1; - mdev->ops->send_intr(mdev, db); - mic_set_state(mdev, MIC_SUSPENDED); - } - mutex_unlock(&mdev->mic_mutex); + return &mdev->aper; } + +struct cosm_hw_ops cosm_hw_ops = { + .reset = _mic_reset, + .force_reset = _mic_reset, + .post_reset = NULL, + .ready = _mic_ready, + .start = _mic_start, + .stop = _mic_stop, + .family = _mic_family, + .stepping = _mic_stepping, + .aper = _mic_aper, +}; diff --git a/drivers/misc/mic/host/mic_debugfs.c b/drivers/misc/mic/host/mic_debugfs.c index 3c9ea4896f3c..10581600777a 100644 --- a/drivers/misc/mic/host/mic_debugfs.c +++ b/drivers/misc/mic/host/mic_debugfs.c @@ -31,71 +31,6 @@ /* Debugfs parent dir */ static struct dentry *mic_dbg; -/** - * mic_log_buf_show - Display MIC kernel log buffer. - * - * log_buf addr/len is read from System.map by user space - * and populated in sysfs entries. - */ -static int mic_log_buf_show(struct seq_file *s, void *unused) -{ - void __iomem *log_buf_va; - int __iomem *log_buf_len_va; - struct mic_device *mdev = s->private; - void *kva; - int size; - unsigned long aper_offset; - - if (!mdev || !mdev->log_buf_addr || !mdev->log_buf_len) - goto done; - /* - * Card kernel will never be relocated and any kernel text/data mapping - * can be translated to phys address by subtracting __START_KERNEL_map. - */ - aper_offset = (unsigned long)mdev->log_buf_len - __START_KERNEL_map; - log_buf_len_va = mdev->aper.va + aper_offset; - aper_offset = (unsigned long)mdev->log_buf_addr - __START_KERNEL_map; - log_buf_va = mdev->aper.va + aper_offset; - size = ioread32(log_buf_len_va); - - kva = kmalloc(size, GFP_KERNEL); - if (!kva) - goto done; - mutex_lock(&mdev->mic_mutex); - memcpy_fromio(kva, log_buf_va, size); - switch (mdev->state) { - case MIC_ONLINE: - /* Fall through */ - case MIC_SHUTTING_DOWN: - seq_write(s, kva, size); - break; - default: - break; - } - mutex_unlock(&mdev->mic_mutex); - kfree(kva); -done: - return 0; -} - -static int mic_log_buf_open(struct inode *inode, struct file *file) -{ - return single_open(file, mic_log_buf_show, inode->i_private); -} - -static int mic_log_buf_release(struct inode *inode, struct file *file) -{ - return single_release(inode, file); -} - -static const struct file_operations log_buf_ops = { - .owner = THIS_MODULE, - .open = mic_log_buf_open, - .read = seq_read, - .llseek = seq_lseek, - .release = mic_log_buf_release -}; - static int mic_smpt_show(struct seq_file *s, void *pos) { int i; @@ -138,32 +73,6 @@ static const struct file_operations smpt_file_ops = { .release = mic_smpt_debug_release }; -static int mic_soft_reset_show(struct seq_file *s, void *pos) -{ - struct mic_device *mdev = s->private; - - mic_stop(mdev, true); - return 0; -} - -static int mic_soft_reset_debug_open(struct inode *inode, struct file *file) -{ - return single_open(file, mic_soft_reset_show, inode->i_private); -} - -static int mic_soft_reset_debug_release(struct inode *inode, struct file *file) -{ - return single_release(inode, file); -} - -static const struct file_operations soft_reset_ops = { - .owner = THIS_MODULE, - .open = mic_soft_reset_debug_open, - .read = seq_read, - .llseek = seq_lseek, - .release = mic_soft_reset_debug_release -}; - static int mic_post_code_show(struct seq_file *s, void *pos) { struct mic_device *mdev = s->private; @@ -204,18 +113,8 @@ static int mic_dp_show(struct seq_file *s, void *pos) seq_printf(s, "Bootparam: magic 0x%x\n", bootparam->magic); - seq_printf(s, "Bootparam: h2c_shutdown_db %d\n", - bootparam->h2c_shutdown_db); seq_printf(s, "Bootparam: h2c_config_db %d\n", bootparam->h2c_config_db); - seq_printf(s, "Bootparam: c2h_shutdown_db %d\n", - bootparam->c2h_shutdown_db); - seq_printf(s, "Bootparam: shutdown_status %d\n", - bootparam->shutdown_status); - seq_printf(s, "Bootparam: shutdown_card %d\n", - bootparam->shutdown_card); - seq_printf(s, "Bootparam: tot_nodes %d\n", - bootparam->tot_nodes); seq_printf(s, "Bootparam: node_id %d\n", bootparam->node_id); seq_printf(s, "Bootparam: c2h_scif_db %d\n", @@ -392,8 +291,7 @@ static int mic_msi_irq_info_show(struct seq_file *s, void *pos) int i, j; u16 entry; u16 vector; - struct pci_dev *pdev = container_of(mdev->sdev->parent, - struct pci_dev, dev); + struct pci_dev *pdev = mdev->pdev; if (pci_dev_msi_enabled(pdev)) { for (i = 0; i < mdev->irq_info.num_vectors; i++) { @@ -454,20 +352,18 @@ static const struct file_operations msi_irq_info_ops = { */ void mic_create_debug_dir(struct mic_device *mdev) { + char name[16]; + if (!mic_dbg) return; - mdev->dbg_dir = debugfs_create_dir(dev_name(mdev->sdev), mic_dbg); + scnprintf(name, sizeof(name), "mic%d", mdev->id); + mdev->dbg_dir = debugfs_create_dir(name, mic_dbg); if (!mdev->dbg_dir) return; - debugfs_create_file("log_buf", 0444, mdev->dbg_dir, mdev, &log_buf_ops); - debugfs_create_file("smpt", 0444, mdev->dbg_dir, mdev, &smpt_file_ops); - debugfs_create_file("soft_reset", 0444, mdev->dbg_dir, mdev, - &soft_reset_ops); - debugfs_create_file("post_code", 0444, mdev->dbg_dir, mdev, &post_code_ops); diff --git a/drivers/misc/mic/host/mic_device.h b/drivers/misc/mic/host/mic_device.h index 01a7555aa648..461184a12fbb 100644 --- a/drivers/misc/mic/host/mic_device.h +++ b/drivers/misc/mic/host/mic_device.h @@ -26,21 +26,12 @@ #include <linux/notifier.h> #include <linux/irqreturn.h> #include <linux/dmaengine.h> +#include <linux/miscdevice.h> #include <linux/mic_bus.h> #include "../bus/scif_bus.h" +#include "../bus/cosm_bus.h" #include "mic_intr.h" -/* The maximum number of MIC devices supported in a single host system. */ -#define MIC_MAX_NUM_DEVS 256 - -/** - * enum mic_hw_family - The hardware family to which a device belongs. - */ -enum mic_hw_family { - MIC_FAMILY_X100 = 0, - MIC_FAMILY_UNKNOWN -}; - /** * enum mic_stepping - MIC stepping ids. */ @@ -51,6 +42,8 @@ enum mic_stepping { MIC_C0_STEP = 0x20, }; +extern struct cosm_hw_ops cosm_hw_ops; + /** * struct mic_device - MIC device information for each card. * @@ -60,8 +53,7 @@ enum mic_stepping { * @ops: MIC HW specific operations. * @id: The unique device id for this MIC device. * @stepping: Stepping ID. - * @attr_group: Pointer to list of sysfs attribute groups. - * @sdev: Device for sysfs entries. + * @pdev: Underlying PCI device. * @mic_mutex: Mutex for synchronizing access to mic_device. * @intr_ops: HW specific interrupt operations. * @smpt_ops: Hardware specific SMPT operations. @@ -69,30 +61,17 @@ enum mic_stepping { * @intr_info: H/W specific interrupt information. * @irq_info: The OS specific irq information * @dbg_dir: debugfs directory of this MIC device. - * @cmdline: Kernel command line. - * @firmware: Firmware file name. - * @ramdisk: Ramdisk file name. - * @bootmode: Boot mode i.e. "linux" or "elf" for flash updates. * @bootaddr: MIC boot address. - * @reset_trigger_work: Work for triggering reset requests. - * @shutdown_work: Work for handling shutdown interrupts. - * @state: MIC state. - * @shutdown_status: MIC status reported by card for shutdown/crashes. - * @state_sysfs: Sysfs dirent for notifying ring 3 about MIC state changes. - * @reset_wait: Waitqueue for sleeping while reset completes. - * @log_buf_addr: Log buffer address for MIC. - * @log_buf_len: Log buffer length address for MIC. * @dp: virtio device page * @dp_dma_addr: virtio device page DMA address. - * @shutdown_db: shutdown doorbell. - * @shutdown_cookie: shutdown cookie. - * @cdev: Character device for MIC. + * @name: name for the misc char device + * @miscdev: registered misc char device * @vdev_list: list of virtio devices. - * @pm_notifier: Handles PM notifications from the OS. * @dma_mbdev: MIC BUS DMA device. * @dma_ch - Array of DMA channels * @num_dma_ch - Number of DMA channels available * @scdev: SCIF device on the SCIF virtual bus. + * @cosm_dev: COSM device */ struct mic_device { struct mic_mw mmio; @@ -101,8 +80,7 @@ struct mic_device { struct mic_hw_ops *ops; int id; enum mic_stepping stepping; - const struct attribute_group **attr_group; - struct device *sdev; + struct pci_dev *pdev; struct mutex mic_mutex; struct mic_hw_intr_ops *intr_ops; struct mic_smpt_ops *smpt_ops; @@ -110,30 +88,17 @@ struct mic_device { struct mic_intr_info *intr_info; struct mic_irq_info irq_info; struct dentry *dbg_dir; - char *cmdline; - char *firmware; - char *ramdisk; - char *bootmode; u32 bootaddr; - struct work_struct reset_trigger_work; - struct work_struct shutdown_work; - u8 state; - u8 shutdown_status; - struct kernfs_node *state_sysfs; - struct completion reset_wait; - void *log_buf_addr; - int *log_buf_len; void *dp; dma_addr_t dp_dma_addr; - int shutdown_db; - struct mic_irq *shutdown_cookie; - struct cdev cdev; + char name[16]; + struct miscdevice miscdev; struct list_head vdev_list; - struct notifier_block pm_notifier; struct mbus_device *dma_mbdev; struct dma_chan *dma_ch[MIC_MAX_DMA_CHAN]; int num_dma_ch; struct scif_hw_dev *scdev; + struct cosm_device *cosm_dev; }; /** @@ -199,38 +164,9 @@ mic_mmio_write(struct mic_mw *mw, u32 val, u32 offset) iowrite32(val, mw->va + offset); } -static inline struct dma_chan *mic_request_dma_chan(struct mic_device *mdev) -{ - dma_cap_mask_t mask; - struct dma_chan *chan; - - dma_cap_zero(mask); - dma_cap_set(DMA_MEMCPY, mask); - chan = dma_request_channel(mask, mdev->ops->dma_filter, - mdev->sdev->parent); - if (chan) - return chan; - dev_err(mdev->sdev->parent, "%s %d unable to acquire channel\n", - __func__, __LINE__); - return NULL; -} - -void mic_sysfs_init(struct mic_device *mdev); -int mic_start(struct mic_device *mdev, const char *buf); -void mic_stop(struct mic_device *mdev, bool force); -void mic_shutdown(struct mic_device *mdev); -void mic_reset_delayed_work(struct work_struct *work); -void mic_reset_trigger_work(struct work_struct *work); -void mic_shutdown_work(struct work_struct *work); void mic_bootparam_init(struct mic_device *mdev); -void mic_set_state(struct mic_device *mdev, u8 state); -void mic_set_shutdown_status(struct mic_device *mdev, u8 status); void mic_create_debug_dir(struct mic_device *dev); void mic_delete_debug_dir(struct mic_device *dev); void __init mic_init_debugfs(void); void mic_exit_debugfs(void); -void mic_prepare_suspend(struct mic_device *mdev); -void mic_complete_resume(struct mic_device *mdev); -void mic_suspend(struct mic_device *mdev); -extern atomic_t g_num_mics; #endif diff --git a/drivers/misc/mic/host/mic_fops.c b/drivers/misc/mic/host/mic_fops.c index 85776d7327f3..8cc1d90cd949 100644 --- a/drivers/misc/mic/host/mic_fops.c +++ b/drivers/misc/mic/host/mic_fops.c @@ -30,8 +30,8 @@ int mic_open(struct inode *inode, struct file *f) { struct mic_vdev *mvdev; - struct mic_device *mdev = container_of(inode->i_cdev, - struct mic_device, cdev); + struct mic_device *mdev = container_of(f->private_data, + struct mic_device, miscdev); mvdev = kzalloc(sizeof(*mvdev), GFP_KERNEL); if (!mvdev) diff --git a/drivers/misc/mic/host/mic_intr.c b/drivers/misc/mic/host/mic_intr.c index b4ca6c884d19..08ca3e372fa4 100644 --- a/drivers/misc/mic/host/mic_intr.c +++ b/drivers/misc/mic/host/mic_intr.c @@ -30,8 +30,7 @@ static irqreturn_t mic_thread_fn(int irq, void *dev) struct mic_intr_info *intr_info = mdev->intr_info; struct mic_irq_info *irq_info = &mdev->irq_info; struct mic_intr_cb *intr_cb; - struct pci_dev *pdev = container_of(mdev->sdev->parent, - struct pci_dev, dev); + struct pci_dev *pdev = mdev->pdev; int i; spin_lock(&irq_info->mic_thread_lock); @@ -57,8 +56,7 @@ static irqreturn_t mic_interrupt(int irq, void *dev) struct mic_intr_info *intr_info = mdev->intr_info; struct mic_irq_info *irq_info = &mdev->irq_info; struct mic_intr_cb *intr_cb; - struct pci_dev *pdev = container_of(mdev->sdev->parent, - struct pci_dev, dev); + struct pci_dev *pdev = mdev->pdev; u32 mask; int i; @@ -83,7 +81,7 @@ static irqreturn_t mic_interrupt(int irq, void *dev) /* Return the interrupt offset from the index. Index is 0 based. */ static u16 mic_map_src_to_offset(struct mic_device *mdev, - int intr_src, enum mic_intr_type type) + int intr_src, enum mic_intr_type type) { if (type >= MIC_NUM_INTR_TYPES) return MIC_NUM_OFFSETS; @@ -214,7 +212,7 @@ static int mic_setup_msix(struct mic_device *mdev, struct pci_dev *pdev) mdev->irq_info.msix_entries[i].entry = i; rc = pci_enable_msix_exact(pdev, mdev->irq_info.msix_entries, - MIC_MIN_MSIX); + MIC_MIN_MSIX); if (rc) { dev_dbg(&pdev->dev, "Error enabling MSIx. rc = %d\n", rc); goto err_enable_msix; @@ -229,7 +227,7 @@ static int mic_setup_msix(struct mic_device *mdev, struct pci_dev *pdev) goto err_nomem2; } - dev_dbg(mdev->sdev->parent, + dev_dbg(&mdev->pdev->dev, "%d MSIx irqs setup\n", mdev->irq_info.num_vectors); return 0; err_nomem2: @@ -281,7 +279,6 @@ static void mic_release_callbacks(struct mic_device *mdev) spin_lock(&mdev->irq_info.mic_thread_lock); spin_lock_irqsave(&mdev->irq_info.mic_intr_lock, flags); for (i = 0; i < MIC_NUM_OFFSETS; i++) { - if (list_empty(&mdev->irq_info.cb_list[i])) break; @@ -443,12 +440,11 @@ mic_request_threaded_irq(struct mic_device *mdev, unsigned long cookie = 0; u16 entry; struct mic_intr_cb *intr_cb; - struct pci_dev *pdev = container_of(mdev->sdev->parent, - struct pci_dev, dev); + struct pci_dev *pdev = mdev->pdev; offset = mic_map_src_to_offset(mdev, intr_src, type); if (offset >= MIC_NUM_OFFSETS) { - dev_err(mdev->sdev->parent, + dev_err(&mdev->pdev->dev, "Error mapping index %d to a valid source id.\n", intr_src); rc = -EINVAL; @@ -458,7 +454,7 @@ mic_request_threaded_irq(struct mic_device *mdev, if (mdev->irq_info.num_vectors > 1) { msix = mic_get_available_vector(mdev); if (!msix) { - dev_err(mdev->sdev->parent, + dev_err(&mdev->pdev->dev, "No MSIx vectors available for use.\n"); rc = -ENOSPC; goto err; @@ -467,7 +463,7 @@ mic_request_threaded_irq(struct mic_device *mdev, rc = request_threaded_irq(msix->vector, handler, thread_fn, 0, name, data); if (rc) { - dev_dbg(mdev->sdev->parent, + dev_dbg(&mdev->pdev->dev, "request irq failed rc = %d\n", rc); goto err; } @@ -476,13 +472,13 @@ mic_request_threaded_irq(struct mic_device *mdev, mdev->intr_ops->program_msi_to_src_map(mdev, entry, offset, true); cookie = MK_COOKIE(entry, offset); - dev_dbg(mdev->sdev->parent, "irq: %d assigned for src: %d\n", + dev_dbg(&mdev->pdev->dev, "irq: %d assigned for src: %d\n", msix->vector, intr_src); } else { intr_cb = mic_register_intr_callback(mdev, offset, handler, thread_fn, data); if (IS_ERR(intr_cb)) { - dev_err(mdev->sdev->parent, + dev_err(&mdev->pdev->dev, "No available callback entries for use\n"); rc = PTR_ERR(intr_cb); goto err; @@ -495,7 +491,7 @@ mic_request_threaded_irq(struct mic_device *mdev, entry, offset, true); } cookie = MK_COOKIE(entry, intr_cb->cb_id); - dev_dbg(mdev->sdev->parent, "callback %d registered for src: %d\n", + dev_dbg(&mdev->pdev->dev, "callback %d registered for src: %d\n", intr_cb->cb_id, intr_src); } return (struct mic_irq *)cookie; @@ -515,20 +511,19 @@ err: * returns: none. */ void mic_free_irq(struct mic_device *mdev, - struct mic_irq *cookie, void *data) + struct mic_irq *cookie, void *data) { u32 offset; u32 entry; u8 src_id; unsigned int irq; - struct pci_dev *pdev = container_of(mdev->sdev->parent, - struct pci_dev, dev); + struct pci_dev *pdev = mdev->pdev; entry = GET_ENTRY((unsigned long)cookie); offset = GET_OFFSET((unsigned long)cookie); if (mdev->irq_info.num_vectors > 1) { if (entry >= mdev->irq_info.num_vectors) { - dev_warn(mdev->sdev->parent, + dev_warn(&mdev->pdev->dev, "entry %d should be < num_irq %d\n", entry, mdev->irq_info.num_vectors); return; @@ -539,12 +534,12 @@ void mic_free_irq(struct mic_device *mdev, mdev->intr_ops->program_msi_to_src_map(mdev, entry, offset, false); - dev_dbg(mdev->sdev->parent, "irq: %d freed\n", irq); + dev_dbg(&mdev->pdev->dev, "irq: %d freed\n", irq); } else { irq = pdev->irq; src_id = mic_unregister_intr_callback(mdev, offset); if (src_id >= MIC_NUM_OFFSETS) { - dev_warn(mdev->sdev->parent, "Error unregistering callback\n"); + dev_warn(&mdev->pdev->dev, "Error unregistering callback\n"); return; } if (pci_dev_msi_enabled(pdev)) { @@ -552,7 +547,7 @@ void mic_free_irq(struct mic_device *mdev, mdev->intr_ops->program_msi_to_src_map(mdev, entry, src_id, false); } - dev_dbg(mdev->sdev->parent, "callback %d unregistered for src: %d\n", + dev_dbg(&mdev->pdev->dev, "callback %d unregistered for src: %d\n", offset, src_id); } } @@ -579,7 +574,7 @@ int mic_setup_interrupts(struct mic_device *mdev, struct pci_dev *pdev) rc = mic_setup_intx(mdev, pdev); if (rc) { - dev_err(mdev->sdev->parent, "no usable interrupts\n"); + dev_err(&mdev->pdev->dev, "no usable interrupts\n"); return rc; } done: @@ -635,8 +630,7 @@ void mic_free_interrupts(struct mic_device *mdev, struct pci_dev *pdev) void mic_intr_restore(struct mic_device *mdev) { int entry, offset; - struct pci_dev *pdev = container_of(mdev->sdev->parent, - struct pci_dev, dev); + struct pci_dev *pdev = mdev->pdev; if (!pci_dev_msi_enabled(pdev)) return; diff --git a/drivers/misc/mic/host/mic_main.c b/drivers/misc/mic/host/mic_main.c index 456462932151..153894e7ed5b 100644 --- a/drivers/misc/mic/host/mic_main.c +++ b/drivers/misc/mic/host/mic_main.c @@ -16,17 +16,11 @@ * the file called "COPYING". * * Intel MIC Host driver. - * - * Global TODO's across the driver to be added after initial base - * patches are accepted upstream: - * 1) Enable DMA support. - * 2) Enable per vring interrupt support. */ #include <linux/fs.h> #include <linux/module.h> #include <linux/pci.h> #include <linux/poll.h> -#include <linux/suspend.h> #include <linux/mic_common.h> #include "../common/mic_dev.h" @@ -63,12 +57,8 @@ MODULE_DEVICE_TABLE(pci, mic_pci_tbl); /* ID allocator for MIC devices */ static struct ida g_mic_ida; -/* Class of MIC devices for sysfs accessibility. */ -static struct class *g_mic_class; /* Base device node number for MIC devices */ static dev_t g_mic_devno; -/* Track the total number of MIC devices */ -atomic_t g_num_mics; static const struct file_operations mic_fops = { .open = mic_open, @@ -83,17 +73,14 @@ static const struct file_operations mic_fops = { static int mic_dp_init(struct mic_device *mdev) { mdev->dp = kzalloc(MIC_DP_SIZE, GFP_KERNEL); - if (!mdev->dp) { - dev_err(mdev->sdev->parent, "%s %d err %d\n", - __func__, __LINE__, -ENOMEM); + if (!mdev->dp) return -ENOMEM; - } mdev->dp_dma_addr = mic_map_single(mdev, mdev->dp, MIC_DP_SIZE); if (mic_map_error(mdev->dp_dma_addr)) { kfree(mdev->dp); - dev_err(mdev->sdev->parent, "%s %d err %d\n", + dev_err(&mdev->pdev->dev, "%s %d err %d\n", __func__, __LINE__, -ENOMEM); return -ENOMEM; } @@ -110,30 +97,6 @@ static void mic_dp_uninit(struct mic_device *mdev) } /** - * mic_shutdown_db - Shutdown doorbell interrupt handler. - */ -static irqreturn_t mic_shutdown_db(int irq, void *data) -{ - struct mic_device *mdev = data; - struct mic_bootparam *bootparam = mdev->dp; - - mdev->ops->intr_workarounds(mdev); - - switch (bootparam->shutdown_status) { - case MIC_HALTED: - case MIC_POWER_OFF: - case MIC_RESTART: - /* Fall through */ - case MIC_CRASHED: - schedule_work(&mdev->shutdown_work); - break; - default: - break; - }; - return IRQ_HANDLED; -} - -/** * mic_ops_init: Initialize HW specific operation tables. * * @mdev: pointer to mic_device instance @@ -190,43 +153,6 @@ static enum mic_hw_family mic_get_family(struct pci_dev *pdev) } /** -* mic_pm_notifier: Notifier callback function that handles -* PM notifications. -* -* @notifier_block: The notifier structure. -* @pm_event: The event for which the driver was notified. -* @unused: Meaningless. Always NULL. -* -* returns NOTIFY_DONE -*/ -static int mic_pm_notifier(struct notifier_block *notifier, - unsigned long pm_event, void *unused) -{ - struct mic_device *mdev = container_of(notifier, - struct mic_device, pm_notifier); - - switch (pm_event) { - case PM_HIBERNATION_PREPARE: - /* Fall through */ - case PM_SUSPEND_PREPARE: - mic_prepare_suspend(mdev); - break; - case PM_POST_HIBERNATION: - /* Fall through */ - case PM_POST_SUSPEND: - /* Fall through */ - case PM_POST_RESTORE: - mic_complete_resume(mdev); - break; - case PM_RESTORE_PREPARE: - break; - default: - break; - } - return NOTIFY_DONE; -} - -/** * mic_device_init - Allocates and initializes the MIC device structure * * @mdev: pointer to mic_device instance @@ -234,52 +160,16 @@ static int mic_pm_notifier(struct notifier_block *notifier, * * returns none. */ -static int +static void mic_device_init(struct mic_device *mdev, struct pci_dev *pdev) { - int rc; - + mdev->pdev = pdev; mdev->family = mic_get_family(pdev); mdev->stepping = pdev->revision; mic_ops_init(mdev); - mic_sysfs_init(mdev); mutex_init(&mdev->mic_mutex); mdev->irq_info.next_avail_src = 0; - INIT_WORK(&mdev->reset_trigger_work, mic_reset_trigger_work); - INIT_WORK(&mdev->shutdown_work, mic_shutdown_work); - init_completion(&mdev->reset_wait); INIT_LIST_HEAD(&mdev->vdev_list); - mdev->pm_notifier.notifier_call = mic_pm_notifier; - rc = register_pm_notifier(&mdev->pm_notifier); - if (rc) { - dev_err(&pdev->dev, "register_pm_notifier failed rc %d\n", - rc); - goto register_pm_notifier_fail; - } - return 0; -register_pm_notifier_fail: - flush_work(&mdev->shutdown_work); - flush_work(&mdev->reset_trigger_work); - return rc; -} - -/** - * mic_device_uninit - Frees resources allocated during mic_device_init(..) - * - * @mdev: pointer to mic_device instance - * - * returns none - */ -static void mic_device_uninit(struct mic_device *mdev) -{ - /* The cmdline sysfs entry might have allocated cmdline */ - kfree(mdev->cmdline); - kfree(mdev->firmware); - kfree(mdev->ramdisk); - kfree(mdev->bootmode); - flush_work(&mdev->reset_trigger_work); - flush_work(&mdev->shutdown_work); - unregister_pm_notifier(&mdev->pm_notifier); } /** @@ -291,7 +181,7 @@ static void mic_device_uninit(struct mic_device *mdev) * returns 0 on success, < 0 on failure. */ static int mic_probe(struct pci_dev *pdev, - const struct pci_device_id *ent) + const struct pci_device_id *ent) { int rc; struct mic_device *mdev; @@ -309,16 +199,12 @@ static int mic_probe(struct pci_dev *pdev, goto ida_fail; } - rc = mic_device_init(mdev, pdev); - if (rc) { - dev_err(&pdev->dev, "mic_device_init failed rc %d\n", rc); - goto device_init_fail; - } + mic_device_init(mdev, pdev); rc = pci_enable_device(pdev); if (rc) { dev_err(&pdev->dev, "failed to enable pci device.\n"); - goto uninit_device; + goto ida_remove; } pci_set_master(pdev); @@ -367,62 +253,39 @@ static int mic_probe(struct pci_dev *pdev, pci_set_drvdata(pdev, mdev); - mdev->sdev = device_create_with_groups(g_mic_class, &pdev->dev, - MKDEV(MAJOR(g_mic_devno), mdev->id), NULL, - mdev->attr_group, "mic%d", mdev->id); - if (IS_ERR(mdev->sdev)) { - rc = PTR_ERR(mdev->sdev); - dev_err(&pdev->dev, - "device_create_with_groups failed rc %d\n", rc); - goto smpt_uninit; - } - mdev->state_sysfs = sysfs_get_dirent(mdev->sdev->kobj.sd, "state"); - if (!mdev->state_sysfs) { - rc = -ENODEV; - dev_err(&pdev->dev, "sysfs_get_dirent failed rc %d\n", rc); - goto destroy_device; - } - rc = mic_dp_init(mdev); if (rc) { dev_err(&pdev->dev, "mic_dp_init failed rc %d\n", rc); - goto sysfs_put; - } - mutex_lock(&mdev->mic_mutex); - - mdev->shutdown_db = mic_next_db(mdev); - mdev->shutdown_cookie = mic_request_threaded_irq(mdev, mic_shutdown_db, - NULL, "shutdown-interrupt", mdev, - mdev->shutdown_db, MIC_INTR_DB); - if (IS_ERR(mdev->shutdown_cookie)) { - rc = PTR_ERR(mdev->shutdown_cookie); - mutex_unlock(&mdev->mic_mutex); - goto dp_uninit; + goto smpt_uninit; } - mutex_unlock(&mdev->mic_mutex); mic_bootparam_init(mdev); mic_create_debug_dir(mdev); - cdev_init(&mdev->cdev, &mic_fops); - mdev->cdev.owner = THIS_MODULE; - rc = cdev_add(&mdev->cdev, MKDEV(MAJOR(g_mic_devno), mdev->id), 1); + + mdev->miscdev.minor = MISC_DYNAMIC_MINOR; + snprintf(mdev->name, sizeof(mdev->name), "mic%d", mdev->id); + mdev->miscdev.name = mdev->name; + mdev->miscdev.fops = &mic_fops; + mdev->miscdev.parent = &mdev->pdev->dev; + rc = misc_register(&mdev->miscdev); if (rc) { - dev_err(&pdev->dev, "cdev_add err id %d rc %d\n", mdev->id, rc); + dev_err(&pdev->dev, "misc_register err id %d rc %d\n", + mdev->id, rc); goto cleanup_debug_dir; } - atomic_inc(&g_num_mics); + + mdev->cosm_dev = cosm_register_device(&mdev->pdev->dev, &cosm_hw_ops); + if (IS_ERR(mdev->cosm_dev)) { + rc = PTR_ERR(mdev->cosm_dev); + dev_err(&pdev->dev, "cosm_add_device failed rc %d\n", rc); + goto misc_dereg; + } return 0; +misc_dereg: + misc_deregister(&mdev->miscdev); cleanup_debug_dir: mic_delete_debug_dir(mdev); - mutex_lock(&mdev->mic_mutex); - mic_free_irq(mdev, mdev->shutdown_cookie, mdev); - mutex_unlock(&mdev->mic_mutex); -dp_uninit: mic_dp_uninit(mdev); -sysfs_put: - sysfs_put(mdev->state_sysfs); -destroy_device: - device_destroy(g_mic_class, MKDEV(MAJOR(g_mic_devno), mdev->id)); smpt_uninit: mic_smpt_uninit(mdev); free_interrupts: @@ -435,9 +298,7 @@ release_regions: pci_release_regions(pdev); disable_device: pci_disable_device(pdev); -uninit_device: - mic_device_uninit(mdev); -device_init_fail: +ida_remove: ida_simple_remove(&g_mic_ida, mdev->id); ida_fail: kfree(mdev); @@ -461,22 +322,14 @@ static void mic_remove(struct pci_dev *pdev) if (!mdev) return; - mic_stop(mdev, false); - atomic_dec(&g_num_mics); - cdev_del(&mdev->cdev); + cosm_unregister_device(mdev->cosm_dev); + misc_deregister(&mdev->miscdev); mic_delete_debug_dir(mdev); - mutex_lock(&mdev->mic_mutex); - mic_free_irq(mdev, mdev->shutdown_cookie, mdev); - mutex_unlock(&mdev->mic_mutex); - flush_work(&mdev->shutdown_work); mic_dp_uninit(mdev); - sysfs_put(mdev->state_sysfs); - device_destroy(g_mic_class, MKDEV(MAJOR(g_mic_devno), mdev->id)); mic_smpt_uninit(mdev); mic_free_interrupts(mdev, pdev); - iounmap(mdev->mmio.va); iounmap(mdev->aper.va); - mic_device_uninit(mdev); + iounmap(mdev->mmio.va); pci_release_regions(pdev); pci_disable_device(pdev); ida_simple_remove(&g_mic_ida, mdev->id); @@ -495,32 +348,23 @@ static int __init mic_init(void) int ret; ret = alloc_chrdev_region(&g_mic_devno, 0, - MIC_MAX_NUM_DEVS, mic_driver_name); + MIC_MAX_NUM_DEVS, mic_driver_name); if (ret) { pr_err("alloc_chrdev_region failed ret %d\n", ret); goto error; } - g_mic_class = class_create(THIS_MODULE, mic_driver_name); - if (IS_ERR(g_mic_class)) { - ret = PTR_ERR(g_mic_class); - pr_err("class_create failed ret %d\n", ret); - goto cleanup_chrdev; - } - mic_init_debugfs(); ida_init(&g_mic_ida); ret = pci_register_driver(&mic_driver); if (ret) { pr_err("pci_register_driver failed ret %d\n", ret); - goto cleanup_debugfs; + goto cleanup_chrdev; } return ret; -cleanup_debugfs: +cleanup_chrdev: ida_destroy(&g_mic_ida); mic_exit_debugfs(); - class_destroy(g_mic_class); -cleanup_chrdev: unregister_chrdev_region(g_mic_devno, MIC_MAX_NUM_DEVS); error: return ret; @@ -531,7 +375,6 @@ static void __exit mic_exit(void) pci_unregister_driver(&mic_driver); ida_destroy(&g_mic_ida); mic_exit_debugfs(); - class_destroy(g_mic_class); unregister_chrdev_region(g_mic_devno, MIC_MAX_NUM_DEVS); } diff --git a/drivers/misc/mic/host/mic_smpt.c b/drivers/misc/mic/host/mic_smpt.c index cec82034875f..c3f958580fb0 100644 --- a/drivers/misc/mic/host/mic_smpt.c +++ b/drivers/misc/mic/host/mic_smpt.c @@ -76,7 +76,7 @@ mic_is_system_addr(struct mic_device *mdev, dma_addr_t pa) /* Populate an SMPT entry and update the reference counts. */ static void mic_add_smpt_entry(int spt, s64 *ref, u64 addr, - int entries, struct mic_device *mdev) + int entries, struct mic_device *mdev) { struct mic_smpt_info *smpt_info = mdev->smpt; int i; @@ -97,7 +97,7 @@ static void mic_add_smpt_entry(int spt, s64 *ref, u64 addr, * for a given DMA address and size. */ static dma_addr_t mic_smpt_op(struct mic_device *mdev, u64 dma_addr, - int entries, s64 *ref, size_t size) + int entries, s64 *ref, size_t size) { int spt; int ae = 0; @@ -148,7 +148,7 @@ found: * and the starting smpt address */ static int mic_get_smpt_ref_count(struct mic_device *mdev, dma_addr_t dma_addr, - size_t size, s64 *ref, u64 *smpt_start) + size_t size, s64 *ref, u64 *smpt_start) { u64 start = dma_addr; u64 end = dma_addr + size; @@ -181,7 +181,7 @@ dma_addr_t mic_to_dma_addr(struct mic_device *mdev, dma_addr_t mic_addr) dma_addr_t dma_addr; if (!mic_is_system_addr(mdev, mic_addr)) { - dev_err(mdev->sdev->parent, + dev_err(&mdev->pdev->dev, "mic_addr is invalid. mic_addr = 0x%llx\n", mic_addr); return -EINVAL; } @@ -218,7 +218,7 @@ dma_addr_t mic_map(struct mic_device *mdev, dma_addr_t dma_addr, size_t size) return mic_addr; num_entries = mic_get_smpt_ref_count(mdev, dma_addr, size, - ref, &smpt_start); + ref, &smpt_start); /* Set the smpt table appropriately and get 16G aligned mic address */ mic_addr = mic_smpt_op(mdev, smpt_start, num_entries, ref, size); @@ -231,7 +231,7 @@ dma_addr_t mic_map(struct mic_device *mdev, dma_addr_t dma_addr, size_t size) * else generate mic_addr by adding the 16G offset in dma_addr */ if (!mic_addr && MIC_FAMILY_X100 == mdev->family) { - dev_err(mdev->sdev->parent, + dev_err(&mdev->pdev->dev, "mic_map failed dma_addr 0x%llx size 0x%lx\n", dma_addr, size); return mic_addr; @@ -264,7 +264,7 @@ void mic_unmap(struct mic_device *mdev, dma_addr_t mic_addr, size_t size) return; if (!mic_is_system_addr(mdev, mic_addr)) { - dev_err(mdev->sdev->parent, + dev_err(&mdev->pdev->dev, "invalid address: 0x%llx\n", mic_addr); return; } @@ -284,7 +284,7 @@ void mic_unmap(struct mic_device *mdev, dma_addr_t mic_addr, size_t size) for (i = spt; i < spt + num_smpt; i++) { smpt_info->entry[i].ref_count -= ref[i - spt]; if (smpt_info->entry[i].ref_count < 0) - dev_warn(mdev->sdev->parent, + dev_warn(&mdev->pdev->dev, "ref count for entry %d is negative\n", i); } spin_unlock_irqrestore(&smpt_info->smpt_lock, flags); @@ -307,15 +307,14 @@ void mic_unmap(struct mic_device *mdev, dma_addr_t mic_addr, size_t size) dma_addr_t mic_map_single(struct mic_device *mdev, void *va, size_t size) { dma_addr_t mic_addr = 0; - struct pci_dev *pdev = container_of(mdev->sdev->parent, - struct pci_dev, dev); + struct pci_dev *pdev = mdev->pdev; dma_addr_t dma_addr = pci_map_single(pdev, va, size, PCI_DMA_BIDIRECTIONAL); if (!pci_dma_mapping_error(pdev, dma_addr)) { mic_addr = mic_map(mdev, dma_addr, size); if (!mic_addr) { - dev_err(mdev->sdev->parent, + dev_err(&mdev->pdev->dev, "mic_map failed dma_addr 0x%llx size 0x%lx\n", dma_addr, size); pci_unmap_single(pdev, dma_addr, @@ -339,8 +338,7 @@ dma_addr_t mic_map_single(struct mic_device *mdev, void *va, size_t size) void mic_unmap_single(struct mic_device *mdev, dma_addr_t mic_addr, size_t size) { - struct pci_dev *pdev = container_of(mdev->sdev->parent, - struct pci_dev, dev); + struct pci_dev *pdev = mdev->pdev; dma_addr_t dma_addr = mic_to_dma_addr(mdev, mic_addr); mic_unmap(mdev, mic_addr, size); pci_unmap_single(pdev, dma_addr, size, PCI_DMA_BIDIRECTIONAL); @@ -399,18 +397,18 @@ void mic_smpt_uninit(struct mic_device *mdev) struct mic_smpt_info *smpt_info = mdev->smpt; int i; - dev_dbg(mdev->sdev->parent, + dev_dbg(&mdev->pdev->dev, "nodeid %d SMPT ref count %lld map %lld unmap %lld\n", mdev->id, smpt_info->ref_count, smpt_info->map_count, smpt_info->unmap_count); for (i = 0; i < smpt_info->info.num_reg; i++) { - dev_dbg(mdev->sdev->parent, + dev_dbg(&mdev->pdev->dev, "SMPT entry[%d] dma_addr = 0x%llx ref_count = %lld\n", i, smpt_info->entry[i].dma_addr, smpt_info->entry[i].ref_count); if (smpt_info->entry[i].ref_count) - dev_warn(mdev->sdev->parent, + dev_warn(&mdev->pdev->dev, "ref count for entry %d is not zero\n", i); } kfree(smpt_info->entry); diff --git a/drivers/misc/mic/host/mic_sysfs.c b/drivers/misc/mic/host/mic_sysfs.c deleted file mode 100644 index 6dd864e4a617..000000000000 --- a/drivers/misc/mic/host/mic_sysfs.c +++ /dev/null @@ -1,459 +0,0 @@ -/* - * Intel MIC Platform Software Stack (MPSS) - * - * Copyright(c) 2013 Intel Corporation. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License, version 2, as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * The full GNU General Public License is included in this distribution in - * the file called "COPYING". - * - * Intel MIC Host driver. - * - */ -#include <linux/pci.h> - -#include <linux/mic_common.h> -#include "../common/mic_dev.h" -#include "mic_device.h" - -/* - * A state-to-string lookup table, for exposing a human readable state - * via sysfs. Always keep in sync with enum mic_states - */ -static const char * const mic_state_string[] = { - [MIC_OFFLINE] = "offline", - [MIC_ONLINE] = "online", - [MIC_SHUTTING_DOWN] = "shutting_down", - [MIC_RESET_FAILED] = "reset_failed", - [MIC_SUSPENDING] = "suspending", - [MIC_SUSPENDED] = "suspended", -}; - -/* - * A shutdown-status-to-string lookup table, for exposing a human - * readable state via sysfs. Always keep in sync with enum mic_shutdown_status - */ -static const char * const mic_shutdown_status_string[] = { - [MIC_NOP] = "nop", - [MIC_CRASHED] = "crashed", - [MIC_HALTED] = "halted", - [MIC_POWER_OFF] = "poweroff", - [MIC_RESTART] = "restart", -}; - -void mic_set_shutdown_status(struct mic_device *mdev, u8 shutdown_status) -{ - dev_dbg(mdev->sdev->parent, "Shutdown Status %s -> %s\n", - mic_shutdown_status_string[mdev->shutdown_status], - mic_shutdown_status_string[shutdown_status]); - mdev->shutdown_status = shutdown_status; -} - -void mic_set_state(struct mic_device *mdev, u8 state) -{ - dev_dbg(mdev->sdev->parent, "State %s -> %s\n", - mic_state_string[mdev->state], - mic_state_string[state]); - mdev->state = state; - sysfs_notify_dirent(mdev->state_sysfs); -} - -static ssize_t -family_show(struct device *dev, struct device_attribute *attr, char *buf) -{ - static const char x100[] = "x100"; - static const char unknown[] = "Unknown"; - const char *card = NULL; - struct mic_device *mdev = dev_get_drvdata(dev->parent); - - if (!mdev) - return -EINVAL; - - switch (mdev->family) { - case MIC_FAMILY_X100: - card = x100; - break; - default: - card = unknown; - break; - } - return scnprintf(buf, PAGE_SIZE, "%s\n", card); -} -static DEVICE_ATTR_RO(family); - -static ssize_t -stepping_show(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct mic_device *mdev = dev_get_drvdata(dev->parent); - char *string = "??"; - - if (!mdev) - return -EINVAL; - - switch (mdev->stepping) { - case MIC_A0_STEP: - string = "A0"; - break; - case MIC_B0_STEP: - string = "B0"; - break; - case MIC_B1_STEP: - string = "B1"; - break; - case MIC_C0_STEP: - string = "C0"; - break; - default: - break; - } - return scnprintf(buf, PAGE_SIZE, "%s\n", string); -} -static DEVICE_ATTR_RO(stepping); - -static ssize_t -state_show(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct mic_device *mdev = dev_get_drvdata(dev->parent); - - if (!mdev || mdev->state >= MIC_LAST) - return -EINVAL; - - return scnprintf(buf, PAGE_SIZE, "%s\n", - mic_state_string[mdev->state]); -} - -static ssize_t -state_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - int rc = 0; - struct mic_device *mdev = dev_get_drvdata(dev->parent); - if (!mdev) - return -EINVAL; - if (sysfs_streq(buf, "boot")) { - rc = mic_start(mdev, buf); - if (rc) { - dev_err(mdev->sdev->parent, - "mic_boot failed rc %d\n", rc); - count = rc; - } - goto done; - } - - if (sysfs_streq(buf, "reset")) { - schedule_work(&mdev->reset_trigger_work); - goto done; - } - - if (sysfs_streq(buf, "shutdown")) { - mic_shutdown(mdev); - goto done; - } - - if (sysfs_streq(buf, "suspend")) { - mic_suspend(mdev); - goto done; - } - - count = -EINVAL; -done: - return count; -} -static DEVICE_ATTR_RW(state); - -static ssize_t shutdown_status_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct mic_device *mdev = dev_get_drvdata(dev->parent); - - if (!mdev || mdev->shutdown_status >= MIC_STATUS_LAST) - return -EINVAL; - - return scnprintf(buf, PAGE_SIZE, "%s\n", - mic_shutdown_status_string[mdev->shutdown_status]); -} -static DEVICE_ATTR_RO(shutdown_status); - -static ssize_t -cmdline_show(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct mic_device *mdev = dev_get_drvdata(dev->parent); - char *cmdline; - - if (!mdev) - return -EINVAL; - - cmdline = mdev->cmdline; - - if (cmdline) - return scnprintf(buf, PAGE_SIZE, "%s\n", cmdline); - return 0; -} - -static ssize_t -cmdline_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct mic_device *mdev = dev_get_drvdata(dev->parent); - - if (!mdev) - return -EINVAL; - - mutex_lock(&mdev->mic_mutex); - kfree(mdev->cmdline); - - mdev->cmdline = kmalloc(count + 1, GFP_KERNEL); - if (!mdev->cmdline) { - count = -ENOMEM; - goto unlock; - } - - strncpy(mdev->cmdline, buf, count); - - if (mdev->cmdline[count - 1] == '\n') - mdev->cmdline[count - 1] = '\0'; - else - mdev->cmdline[count] = '\0'; -unlock: - mutex_unlock(&mdev->mic_mutex); - return count; -} -static DEVICE_ATTR_RW(cmdline); - -static ssize_t -firmware_show(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct mic_device *mdev = dev_get_drvdata(dev->parent); - char *firmware; - - if (!mdev) - return -EINVAL; - - firmware = mdev->firmware; - - if (firmware) - return scnprintf(buf, PAGE_SIZE, "%s\n", firmware); - return 0; -} - -static ssize_t -firmware_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct mic_device *mdev = dev_get_drvdata(dev->parent); - - if (!mdev) - return -EINVAL; - - mutex_lock(&mdev->mic_mutex); - kfree(mdev->firmware); - - mdev->firmware = kmalloc(count + 1, GFP_KERNEL); - if (!mdev->firmware) { - count = -ENOMEM; - goto unlock; - } - strncpy(mdev->firmware, buf, count); - - if (mdev->firmware[count - 1] == '\n') - mdev->firmware[count - 1] = '\0'; - else - mdev->firmware[count] = '\0'; -unlock: - mutex_unlock(&mdev->mic_mutex); - return count; -} -static DEVICE_ATTR_RW(firmware); - -static ssize_t -ramdisk_show(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct mic_device *mdev = dev_get_drvdata(dev->parent); - char *ramdisk; - - if (!mdev) - return -EINVAL; - - ramdisk = mdev->ramdisk; - - if (ramdisk) - return scnprintf(buf, PAGE_SIZE, "%s\n", ramdisk); - return 0; -} - -static ssize_t -ramdisk_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct mic_device *mdev = dev_get_drvdata(dev->parent); - - if (!mdev) - return -EINVAL; - - mutex_lock(&mdev->mic_mutex); - kfree(mdev->ramdisk); - - mdev->ramdisk = kmalloc(count + 1, GFP_KERNEL); - if (!mdev->ramdisk) { - count = -ENOMEM; - goto unlock; - } - - strncpy(mdev->ramdisk, buf, count); - - if (mdev->ramdisk[count - 1] == '\n') - mdev->ramdisk[count - 1] = '\0'; - else - mdev->ramdisk[count] = '\0'; -unlock: - mutex_unlock(&mdev->mic_mutex); - return count; -} -static DEVICE_ATTR_RW(ramdisk); - -static ssize_t -bootmode_show(struct device *dev, struct device_attribute *attr, char *buf) -{ - struct mic_device *mdev = dev_get_drvdata(dev->parent); - char *bootmode; - - if (!mdev) - return -EINVAL; - - bootmode = mdev->bootmode; - - if (bootmode) - return scnprintf(buf, PAGE_SIZE, "%s\n", bootmode); - return 0; -} - -static ssize_t -bootmode_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct mic_device *mdev = dev_get_drvdata(dev->parent); - - if (!mdev) - return -EINVAL; - - if (!sysfs_streq(buf, "linux") && !sysfs_streq(buf, "elf")) - return -EINVAL; - - mutex_lock(&mdev->mic_mutex); - kfree(mdev->bootmode); - - mdev->bootmode = kmalloc(count + 1, GFP_KERNEL); - if (!mdev->bootmode) { - count = -ENOMEM; - goto unlock; - } - - strncpy(mdev->bootmode, buf, count); - - if (mdev->bootmode[count - 1] == '\n') - mdev->bootmode[count - 1] = '\0'; - else - mdev->bootmode[count] = '\0'; -unlock: - mutex_unlock(&mdev->mic_mutex); - return count; -} -static DEVICE_ATTR_RW(bootmode); - -static ssize_t -log_buf_addr_show(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct mic_device *mdev = dev_get_drvdata(dev->parent); - - if (!mdev) - return -EINVAL; - - return scnprintf(buf, PAGE_SIZE, "%p\n", mdev->log_buf_addr); -} - -static ssize_t -log_buf_addr_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct mic_device *mdev = dev_get_drvdata(dev->parent); - int ret; - unsigned long addr; - - if (!mdev) - return -EINVAL; - - ret = kstrtoul(buf, 16, &addr); - if (ret) - goto exit; - - mdev->log_buf_addr = (void *)addr; - ret = count; -exit: - return ret; -} -static DEVICE_ATTR_RW(log_buf_addr); - -static ssize_t -log_buf_len_show(struct device *dev, struct device_attribute *attr, - char *buf) -{ - struct mic_device *mdev = dev_get_drvdata(dev->parent); - - if (!mdev) - return -EINVAL; - - return scnprintf(buf, PAGE_SIZE, "%p\n", mdev->log_buf_len); -} - -static ssize_t -log_buf_len_store(struct device *dev, struct device_attribute *attr, - const char *buf, size_t count) -{ - struct mic_device *mdev = dev_get_drvdata(dev->parent); - int ret; - unsigned long addr; - - if (!mdev) - return -EINVAL; - - ret = kstrtoul(buf, 16, &addr); - if (ret) - goto exit; - - mdev->log_buf_len = (int *)addr; - ret = count; -exit: - return ret; -} -static DEVICE_ATTR_RW(log_buf_len); - -static struct attribute *mic_default_attrs[] = { - &dev_attr_family.attr, - &dev_attr_stepping.attr, - &dev_attr_state.attr, - &dev_attr_shutdown_status.attr, - &dev_attr_cmdline.attr, - &dev_attr_firmware.attr, - &dev_attr_ramdisk.attr, - &dev_attr_bootmode.attr, - &dev_attr_log_buf_addr.attr, - &dev_attr_log_buf_len.attr, - - NULL -}; - -ATTRIBUTE_GROUPS(mic_default); - -void mic_sysfs_init(struct mic_device *mdev) -{ - mdev->attr_group = mic_default_groups; -} diff --git a/drivers/misc/mic/host/mic_virtio.c b/drivers/misc/mic/host/mic_virtio.c index cc08e9f733c9..58b107a24a8b 100644 --- a/drivers/misc/mic/host/mic_virtio.c +++ b/drivers/misc/mic/host/mic_virtio.c @@ -23,7 +23,6 @@ #include <linux/uaccess.h> #include <linux/dmaengine.h> #include <linux/mic_common.h> - #include "../common/mic_dev.h" #include "mic_device.h" #include "mic_smpt.h" @@ -62,7 +61,7 @@ static int mic_sync_dma(struct mic_device *mdev, dma_addr_t dst, } error: if (err) - dev_err(mdev->sdev->parent, "%s %d err %d\n", + dev_err(&mdev->pdev->dev, "%s %d err %d\n", __func__, __LINE__, err); return err; } @@ -440,7 +439,7 @@ void mic_virtio_reset_devices(struct mic_device *mdev) struct list_head *pos, *tmp; struct mic_vdev *mvdev; - dev_dbg(mdev->sdev->parent, "%s\n", __func__); + dev_dbg(&mdev->pdev->dev, "%s\n", __func__); list_for_each_safe(pos, tmp, &mdev->vdev_list) { mvdev = list_entry(pos, struct mic_vdev, list); @@ -686,7 +685,7 @@ int mic_virtio_add_device(struct mic_vdev *mvdev, mvr->head = USHRT_MAX; mvr->mvdev = mvdev; mvr->vrh.notify = mic_notify; - dev_dbg(mdev->sdev->parent, + dev_dbg(&mdev->pdev->dev, "%s %d index %d va %p info %p vr_size 0x%x\n", __func__, __LINE__, i, vr->va, vr->info, vr_size); mvr->buf = (void *)__get_free_pages(GFP_KERNEL, @@ -704,7 +703,7 @@ int mic_virtio_add_device(struct mic_vdev *mvdev, mvdev->virtio_db, MIC_INTR_DB); if (IS_ERR(mvdev->virtio_cookie)) { ret = PTR_ERR(mvdev->virtio_cookie); - dev_dbg(mdev->sdev->parent, "request irq failed\n"); + dev_dbg(&mdev->pdev->dev, "request irq failed\n"); goto err; } @@ -720,7 +719,7 @@ int mic_virtio_add_device(struct mic_vdev *mvdev, smp_wmb(); dd->type = type; - dev_dbg(mdev->sdev->parent, "Added virtio device id %d\n", dd->type); + dev_dbg(&mdev->pdev->dev, "Added virtio device id %d\n", dd->type); db = bootparam->h2c_config_db; if (db != -1) @@ -755,7 +754,7 @@ void mic_virtio_del_device(struct mic_vdev *mvdev) db = bootparam->h2c_config_db; if (db == -1) goto skip_hot_remove; - dev_dbg(mdev->sdev->parent, + dev_dbg(&mdev->pdev->dev, "Requesting hot remove id %d\n", mvdev->virtio_id); mvdev->dc->config_change = MIC_VIRTIO_PARAM_DEV_REMOVE; mdev->ops->send_intr(mdev, db); @@ -765,7 +764,7 @@ void mic_virtio_del_device(struct mic_vdev *mvdev) if (ret) break; } - dev_dbg(mdev->sdev->parent, + dev_dbg(&mdev->pdev->dev, "Device id %d config_change %d guest_ack %d retry %d\n", mvdev->virtio_id, mvdev->dc->config_change, mvdev->dc->guest_ack, retry); @@ -794,7 +793,7 @@ skip_hot_remove: tmp_mvdev = list_entry(pos, struct mic_vdev, list); if (tmp_mvdev == mvdev) { list_del(pos); - dev_dbg(mdev->sdev->parent, + dev_dbg(&mdev->pdev->dev, "Removing virtio device id %d\n", mvdev->virtio_id); break; diff --git a/drivers/misc/mic/host/mic_virtio.h b/drivers/misc/mic/host/mic_virtio.h index d574efb853d9..a80631f2790d 100644 --- a/drivers/misc/mic/host/mic_virtio.h +++ b/drivers/misc/mic/host/mic_virtio.h @@ -124,7 +124,7 @@ void mic_bh_handler(struct work_struct *work); /* Helper API to obtain the MIC PCIe device */ static inline struct device *mic_dev(struct mic_vdev *mvdev) { - return mvdev->mdev->sdev->parent; + return &mvdev->mdev->pdev->dev; } /* Helper API to check if a virtio device is initialized */ diff --git a/drivers/misc/mic/host/mic_x100.c b/drivers/misc/mic/host/mic_x100.c index 3341e90dede4..8118ac48c764 100644 --- a/drivers/misc/mic/host/mic_x100.c +++ b/drivers/misc/mic/host/mic_x100.c @@ -43,7 +43,7 @@ static void mic_x100_write_spad(struct mic_device *mdev, unsigned int idx, u32 val) { - dev_dbg(mdev->sdev->parent, "Writing 0x%x to scratch pad index %d\n", + dev_dbg(&mdev->pdev->dev, "Writing 0x%x to scratch pad index %d\n", val, idx); mic_mmio_write(&mdev->mmio, val, MIC_X100_SBOX_BASE_ADDRESS + @@ -66,7 +66,7 @@ mic_x100_read_spad(struct mic_device *mdev, unsigned int idx) MIC_X100_SBOX_BASE_ADDRESS + MIC_X100_SBOX_SPAD0 + idx * 4); - dev_dbg(mdev->sdev->parent, + dev_dbg(&mdev->pdev->dev, "Reading 0x%x from scratch pad index %d\n", val, idx); return val; } @@ -126,7 +126,7 @@ static void mic_x100_disable_interrupts(struct mic_device *mdev) * @mdev: pointer to mic_device instance */ static void mic_x100_send_sbox_intr(struct mic_device *mdev, - int doorbell) + int doorbell) { struct mic_mw *mw = &mdev->mmio; u64 apic_icr_offset = MIC_X100_SBOX_APICICR0 + doorbell * 8; @@ -147,7 +147,7 @@ static void mic_x100_send_sbox_intr(struct mic_device *mdev, * @mdev: pointer to mic_device instance */ static void mic_x100_send_rdmasr_intr(struct mic_device *mdev, - int doorbell) + int doorbell) { int rdmasr_offset = MIC_X100_SBOX_RDMASR0 + (doorbell << 2); /* Ensure that the interrupt is ordered w.r.t. previous stores. */ @@ -359,15 +359,14 @@ mic_x100_load_command_line(struct mic_device *mdev, const struct firmware *fw) boot_mem = mdev->aper.len >> 20; buf = kzalloc(CMDLINE_SIZE, GFP_KERNEL); - if (!buf) { - dev_err(mdev->sdev->parent, - "%s %d allocation failed\n", __func__, __LINE__); + if (!buf) return -ENOMEM; - } + len += snprintf(buf, CMDLINE_SIZE - len, " mem=%dM", boot_mem); - if (mdev->cmdline) - snprintf(buf + len, CMDLINE_SIZE - len, " %s", mdev->cmdline); + if (mdev->cosm_dev->cmdline) + snprintf(buf + len, CMDLINE_SIZE - len, " %s", + mdev->cosm_dev->cmdline); memcpy_toio(cmd_line_va, buf, strlen(buf) + 1); kfree(buf); return 0; @@ -386,12 +385,11 @@ mic_x100_load_ramdisk(struct mic_device *mdev) int rc; struct boot_params __iomem *bp = mdev->aper.va + mdev->bootaddr; - rc = request_firmware(&fw, - mdev->ramdisk, mdev->sdev->parent); + rc = request_firmware(&fw, mdev->cosm_dev->ramdisk, &mdev->pdev->dev); if (rc < 0) { - dev_err(mdev->sdev->parent, + dev_err(&mdev->pdev->dev, "ramdisk request_firmware failed: %d %s\n", - rc, mdev->ramdisk); + rc, mdev->cosm_dev->ramdisk); goto error; } /* @@ -423,10 +421,10 @@ mic_x100_get_boot_addr(struct mic_device *mdev) scratch2 = mdev->ops->read_spad(mdev, MIC_X100_DOWNLOAD_INFO); boot_addr = MIC_X100_SPAD2_DOWNLOAD_ADDR(scratch2); - dev_dbg(mdev->sdev->parent, "%s %d boot_addr 0x%x\n", + dev_dbg(&mdev->pdev->dev, "%s %d boot_addr 0x%x\n", __func__, __LINE__, boot_addr); if (boot_addr > (1 << 31)) { - dev_err(mdev->sdev->parent, + dev_err(&mdev->pdev->dev, "incorrect bootaddr 0x%x\n", boot_addr); rc = -EINVAL; @@ -454,37 +452,37 @@ mic_x100_load_firmware(struct mic_device *mdev, const char *buf) if (rc) goto error; /* load OS */ - rc = request_firmware(&fw, mdev->firmware, mdev->sdev->parent); + rc = request_firmware(&fw, mdev->cosm_dev->firmware, &mdev->pdev->dev); if (rc < 0) { - dev_err(mdev->sdev->parent, + dev_err(&mdev->pdev->dev, "ramdisk request_firmware failed: %d %s\n", - rc, mdev->firmware); + rc, mdev->cosm_dev->firmware); goto error; } if (mdev->bootaddr > mdev->aper.len - fw->size) { rc = -EINVAL; - dev_err(mdev->sdev->parent, "%s %d rc %d bootaddr 0x%x\n", + dev_err(&mdev->pdev->dev, "%s %d rc %d bootaddr 0x%x\n", __func__, __LINE__, rc, mdev->bootaddr); release_firmware(fw); goto error; } memcpy_toio(mdev->aper.va + mdev->bootaddr, fw->data, fw->size); mdev->ops->write_spad(mdev, MIC_X100_FW_SIZE, fw->size); - if (!strcmp(mdev->bootmode, "elf")) + if (!strcmp(mdev->cosm_dev->bootmode, "flash")) goto done; /* load command line */ rc = mic_x100_load_command_line(mdev, fw); if (rc) { - dev_err(mdev->sdev->parent, "%s %d rc %d\n", + dev_err(&mdev->pdev->dev, "%s %d rc %d\n", __func__, __LINE__, rc); goto error; } release_firmware(fw); /* load ramdisk */ - if (mdev->ramdisk) + if (mdev->cosm_dev->ramdisk) rc = mic_x100_load_ramdisk(mdev); error: - dev_dbg(mdev->sdev->parent, "%s %d rc %d\n", __func__, __LINE__, rc); + dev_dbg(&mdev->pdev->dev, "%s %d rc %d\n", __func__, __LINE__, rc); done: return rc; } diff --git a/drivers/misc/mic/scif/Makefile b/drivers/misc/mic/scif/Makefile index bf10bb7e2b91..29cfc3e51ac9 100644 --- a/drivers/misc/mic/scif/Makefile +++ b/drivers/misc/mic/scif/Makefile @@ -13,3 +13,8 @@ scif-objs += scif_epd.o scif-objs += scif_rb.o scif-objs += scif_nodeqp.o scif-objs += scif_nm.o +scif-objs += scif_dma.o +scif-objs += scif_fence.o +scif-objs += scif_mmap.o +scif-objs += scif_rma.o +scif-objs += scif_rma_list.o diff --git a/drivers/misc/mic/scif/scif_api.c b/drivers/misc/mic/scif/scif_api.c index f39d3135a9ef..ddc9e4b08b5c 100644 --- a/drivers/misc/mic/scif/scif_api.c +++ b/drivers/misc/mic/scif/scif_api.c @@ -37,9 +37,21 @@ enum conn_async_state { ASYNC_CONN_FLUSH_WORK /* async work flush in progress */ }; +/* + * File operations for anonymous inode file associated with a SCIF endpoint, + * used in kernel mode SCIF poll. Kernel mode SCIF poll calls portions of the + * poll API in the kernel and these take in a struct file *. Since a struct + * file is not available to kernel mode SCIF, it uses an anonymous file for + * this purpose. + */ +const struct file_operations scif_anon_fops = { + .owner = THIS_MODULE, +}; + scif_epd_t scif_open(void) { struct scif_endpt *ep; + int err; might_sleep(); ep = kzalloc(sizeof(*ep), GFP_KERNEL); @@ -50,15 +62,22 @@ scif_epd_t scif_open(void) if (!ep->qp_info.qp) goto err_qp_alloc; + err = scif_anon_inode_getfile(ep); + if (err) + goto err_anon_inode; + spin_lock_init(&ep->lock); mutex_init(&ep->sendlock); mutex_init(&ep->recvlock); + scif_rma_ep_init(ep); ep->state = SCIFEP_UNBOUND; dev_dbg(scif_info.mdev.this_device, "SCIFAPI open: ep %p success\n", ep); return ep; +err_anon_inode: + kfree(ep->qp_info.qp); err_qp_alloc: kfree(ep); err_ep_alloc: @@ -166,8 +185,11 @@ int scif_close(scif_epd_t epd) switch (oldstate) { case SCIFEP_ZOMBIE: + dev_err(scif_info.mdev.this_device, + "SCIFAPI close: zombie state unexpected\n"); case SCIFEP_DISCONNECTED: spin_unlock(&ep->lock); + scif_unregister_all_windows(epd); /* Remove from the disconnected list */ mutex_lock(&scif_info.connlock); list_for_each_safe(pos, tmpq, &scif_info.disconnected) { @@ -189,6 +211,7 @@ int scif_close(scif_epd_t epd) case SCIFEP_CLOSING: { spin_unlock(&ep->lock); + scif_unregister_all_windows(epd); scif_disconnect_ep(ep); break; } @@ -200,7 +223,7 @@ int scif_close(scif_epd_t epd) struct scif_endpt *aep; spin_unlock(&ep->lock); - spin_lock(&scif_info.eplock); + mutex_lock(&scif_info.eplock); /* remove from listen list */ list_for_each_safe(pos, tmpq, &scif_info.listen) { @@ -222,7 +245,7 @@ int scif_close(scif_epd_t epd) break; } } - spin_unlock(&scif_info.eplock); + mutex_unlock(&scif_info.eplock); mutex_lock(&scif_info.connlock); list_for_each_safe(pos, tmpq, &scif_info.connected) { tmpep = list_entry(pos, @@ -242,13 +265,13 @@ int scif_close(scif_epd_t epd) } mutex_unlock(&scif_info.connlock); scif_teardown_ep(aep); - spin_lock(&scif_info.eplock); + mutex_lock(&scif_info.eplock); scif_add_epd_to_zombie_list(aep, SCIF_EPLOCK_HELD); ep->acceptcnt--; } spin_lock(&ep->lock); - spin_unlock(&scif_info.eplock); + mutex_unlock(&scif_info.eplock); /* Remove and reject any pending connection requests. */ while (ep->conreqcnt) { @@ -279,6 +302,7 @@ int scif_close(scif_epd_t epd) } } scif_put_port(ep->port.port); + scif_anon_inode_fput(ep); scif_teardown_ep(ep); scif_add_epd_to_zombie_list(ep, !SCIF_EPLOCK_HELD); return 0; @@ -409,9 +433,9 @@ int scif_listen(scif_epd_t epd, int backlog) scif_teardown_ep(ep); ep->qp_info.qp = NULL; - spin_lock(&scif_info.eplock); + mutex_lock(&scif_info.eplock); list_add_tail(&ep->list, &scif_info.listen); - spin_unlock(&scif_info.eplock); + mutex_unlock(&scif_info.eplock); return 0; } EXPORT_SYMBOL_GPL(scif_listen); @@ -450,6 +474,13 @@ static int scif_conn_func(struct scif_endpt *ep) struct scifmsg msg; struct device *spdev; + err = scif_reserve_dma_chan(ep); + if (err) { + dev_err(&ep->remote_dev->sdev->dev, + "%s %d err %d\n", __func__, __LINE__, err); + ep->state = SCIFEP_BOUND; + goto connect_error_simple; + } /* Initiate the first part of the endpoint QP setup */ err = scif_setup_qp_connect(ep->qp_info.qp, &ep->qp_info.qp_offset, SCIF_ENDPT_QP_SIZE, ep->remote_dev); @@ -558,8 +589,10 @@ void scif_conn_handler(struct work_struct *work) list_del(&ep->conn_list); } spin_unlock(&scif_info.nb_connect_lock); - if (ep) + if (ep) { ep->conn_err = scif_conn_func(ep); + wake_up_interruptible(&ep->conn_pend_wq); + } } while (ep); } @@ -660,6 +693,7 @@ int __scif_connect(scif_epd_t epd, struct scif_port_id *dst, bool non_block) ep->remote_dev = &scif_dev[dst->node]; ep->qp_info.qp->magic = SCIFEP_MAGIC; if (ep->conn_async_state == ASYNC_CONN_INPROGRESS) { + init_waitqueue_head(&ep->conn_pend_wq); spin_lock(&scif_info.nb_connect_lock); list_add_tail(&ep->conn_list, &scif_info.nb_connect_list); spin_unlock(&scif_info.nb_connect_lock); @@ -782,12 +816,25 @@ retry_connection: cep->remote_dev = &scif_dev[peer->node]; cep->remote_ep = conreq->msg.payload[0]; + scif_rma_ep_init(cep); + + err = scif_reserve_dma_chan(cep); + if (err) { + dev_err(scif_info.mdev.this_device, + "%s %d err %d\n", __func__, __LINE__, err); + goto scif_accept_error_qpalloc; + } + cep->qp_info.qp = kzalloc(sizeof(*cep->qp_info.qp), GFP_KERNEL); if (!cep->qp_info.qp) { err = -ENOMEM; goto scif_accept_error_qpalloc; } + err = scif_anon_inode_getfile(cep); + if (err) + goto scif_accept_error_anon_inode; + cep->qp_info.qp->magic = SCIFEP_MAGIC; spdev = scif_get_peer_dev(cep->remote_dev); if (IS_ERR(spdev)) { @@ -858,6 +905,8 @@ retry: spin_unlock(&cep->lock); return 0; scif_accept_error_map: + scif_anon_inode_fput(cep); +scif_accept_error_anon_inode: scif_teardown_ep(cep); scif_accept_error_qpalloc: kfree(cep); @@ -1247,6 +1296,134 @@ int scif_recv(scif_epd_t epd, void *msg, int len, int flags) } EXPORT_SYMBOL_GPL(scif_recv); +static inline void _scif_poll_wait(struct file *f, wait_queue_head_t *wq, + poll_table *p, struct scif_endpt *ep) +{ + /* + * Because poll_wait makes a GFP_KERNEL allocation, give up the lock + * and regrab it afterwards. Because the endpoint state might have + * changed while the lock was given up, the state must be checked + * again after re-acquiring the lock. The code in __scif_pollfd(..) + * does this. + */ + spin_unlock(&ep->lock); + poll_wait(f, wq, p); + spin_lock(&ep->lock); +} + +unsigned int +__scif_pollfd(struct file *f, poll_table *wait, struct scif_endpt *ep) +{ + unsigned int mask = 0; + + dev_dbg(scif_info.mdev.this_device, + "SCIFAPI pollfd: ep %p %s\n", ep, scif_ep_states[ep->state]); + + spin_lock(&ep->lock); + + /* Endpoint is waiting for a non-blocking connect to complete */ + if (ep->conn_async_state == ASYNC_CONN_INPROGRESS) { + _scif_poll_wait(f, &ep->conn_pend_wq, wait, ep); + if (ep->conn_async_state == ASYNC_CONN_INPROGRESS) { + if (ep->state == SCIFEP_CONNECTED || + ep->state == SCIFEP_DISCONNECTED || + ep->conn_err) + mask |= POLLOUT; + goto exit; + } + } + + /* Endpoint is listening for incoming connection requests */ + if (ep->state == SCIFEP_LISTENING) { + _scif_poll_wait(f, &ep->conwq, wait, ep); + if (ep->state == SCIFEP_LISTENING) { + if (ep->conreqcnt) + mask |= POLLIN; + goto exit; + } + } + + /* Endpoint is connected or disconnected */ + if (ep->state == SCIFEP_CONNECTED || ep->state == SCIFEP_DISCONNECTED) { + if (poll_requested_events(wait) & POLLIN) + _scif_poll_wait(f, &ep->recvwq, wait, ep); + if (poll_requested_events(wait) & POLLOUT) + _scif_poll_wait(f, &ep->sendwq, wait, ep); + if (ep->state == SCIFEP_CONNECTED || + ep->state == SCIFEP_DISCONNECTED) { + /* Data can be read without blocking */ + if (scif_rb_count(&ep->qp_info.qp->inbound_q, 1)) + mask |= POLLIN; + /* Data can be written without blocking */ + if (scif_rb_space(&ep->qp_info.qp->outbound_q)) + mask |= POLLOUT; + /* Return POLLHUP if endpoint is disconnected */ + if (ep->state == SCIFEP_DISCONNECTED) + mask |= POLLHUP; + goto exit; + } + } + + /* Return POLLERR if the endpoint is in none of the above states */ + mask |= POLLERR; +exit: + spin_unlock(&ep->lock); + return mask; +} + +/** + * scif_poll() - Kernel mode SCIF poll + * @ufds: Array of scif_pollepd structures containing the end points + * and events to poll on + * @nfds: Size of the ufds array + * @timeout_msecs: Timeout in msecs, -ve implies infinite timeout + * + * The code flow in this function is based on do_poll(..) in select.c + * + * Returns the number of endpoints which have pending events or 0 in + * the event of a timeout. If a signal is used for wake up, -EINTR is + * returned. + */ +int +scif_poll(struct scif_pollepd *ufds, unsigned int nfds, long timeout_msecs) +{ + struct poll_wqueues table; + poll_table *pt; + int i, mask, count = 0, timed_out = timeout_msecs == 0; + u64 timeout = timeout_msecs < 0 ? MAX_SCHEDULE_TIMEOUT + : msecs_to_jiffies(timeout_msecs); + + poll_initwait(&table); + pt = &table.pt; + while (1) { + for (i = 0; i < nfds; i++) { + pt->_key = ufds[i].events | POLLERR | POLLHUP; + mask = __scif_pollfd(ufds[i].epd->anon, + pt, ufds[i].epd); + mask &= ufds[i].events | POLLERR | POLLHUP; + if (mask) { + count++; + pt->_qproc = NULL; + } + ufds[i].revents = mask; + } + pt->_qproc = NULL; + if (!count) { + count = table.error; + if (signal_pending(current)) + count = -EINTR; + } + if (count || timed_out) + break; + + if (!schedule_timeout_interruptible(timeout)) + timed_out = 1; + } + poll_freewait(&table); + return count; +} +EXPORT_SYMBOL_GPL(scif_poll); + int scif_get_node_ids(u16 *nodes, int len, u16 *self) { int online = 0; @@ -1274,3 +1451,46 @@ int scif_get_node_ids(u16 *nodes, int len, u16 *self) return online; } EXPORT_SYMBOL_GPL(scif_get_node_ids); + +static int scif_add_client_dev(struct device *dev, struct subsys_interface *si) +{ + struct scif_client *client = + container_of(si, struct scif_client, si); + struct scif_peer_dev *spdev = + container_of(dev, struct scif_peer_dev, dev); + + if (client->probe) + client->probe(spdev); + return 0; +} + +static void scif_remove_client_dev(struct device *dev, + struct subsys_interface *si) +{ + struct scif_client *client = + container_of(si, struct scif_client, si); + struct scif_peer_dev *spdev = + container_of(dev, struct scif_peer_dev, dev); + + if (client->remove) + client->remove(spdev); +} + +void scif_client_unregister(struct scif_client *client) +{ + subsys_interface_unregister(&client->si); +} +EXPORT_SYMBOL_GPL(scif_client_unregister); + +int scif_client_register(struct scif_client *client) +{ + struct subsys_interface *si = &client->si; + + si->name = client->name; + si->subsys = &scif_peer_bus; + si->add_dev = scif_add_client_dev; + si->remove_dev = scif_remove_client_dev; + + return subsys_interface_register(&client->si); +} +EXPORT_SYMBOL_GPL(scif_client_register); diff --git a/drivers/misc/mic/scif/scif_debugfs.c b/drivers/misc/mic/scif/scif_debugfs.c index 51f14e2a1196..6884dad97e17 100644 --- a/drivers/misc/mic/scif/scif_debugfs.c +++ b/drivers/misc/mic/scif/scif_debugfs.c @@ -62,10 +62,87 @@ static const struct file_operations scif_dev_ops = { .release = scif_dev_test_release }; -void __init scif_init_debugfs(void) +static void scif_display_window(struct scif_window *window, struct seq_file *s) +{ + int j; + struct scatterlist *sg; + scif_pinned_pages_t pin = window->pinned_pages; + + seq_printf(s, "window %p type %d temp %d offset 0x%llx ", + window, window->type, window->temp, window->offset); + seq_printf(s, "nr_pages 0x%llx nr_contig_chunks 0x%x prot %d ", + window->nr_pages, window->nr_contig_chunks, window->prot); + seq_printf(s, "ref_count %d magic 0x%llx peer_window 0x%llx ", + window->ref_count, window->magic, window->peer_window); + seq_printf(s, "unreg_state 0x%x va_for_temp 0x%lx\n", + window->unreg_state, window->va_for_temp); + + for (j = 0; j < window->nr_contig_chunks; j++) + seq_printf(s, "page[%d] dma_addr 0x%llx num_pages 0x%llx\n", j, + window->dma_addr[j], window->num_pages[j]); + + if (window->type == SCIF_WINDOW_SELF && pin) + for (j = 0; j < window->nr_pages; j++) + seq_printf(s, "page[%d] = pinned_pages %p address %p\n", + j, pin->pages[j], + page_address(pin->pages[j])); + + if (window->st) + for_each_sg(window->st->sgl, sg, window->st->nents, j) + seq_printf(s, "sg[%d] dma addr 0x%llx length 0x%x\n", + j, sg_dma_address(sg), sg_dma_len(sg)); +} + +static void scif_display_all_windows(struct list_head *head, struct seq_file *s) { - struct dentry *d; + struct list_head *item; + struct scif_window *window; + list_for_each(item, head) { + window = list_entry(item, struct scif_window, list); + scif_display_window(window, s); + } +} + +static int scif_rma_test(struct seq_file *s, void *unused) +{ + struct scif_endpt *ep; + struct list_head *pos; + + mutex_lock(&scif_info.connlock); + list_for_each(pos, &scif_info.connected) { + ep = list_entry(pos, struct scif_endpt, list); + seq_printf(s, "ep %p self windows\n", ep); + mutex_lock(&ep->rma_info.rma_lock); + scif_display_all_windows(&ep->rma_info.reg_list, s); + seq_printf(s, "ep %p remote windows\n", ep); + scif_display_all_windows(&ep->rma_info.remote_reg_list, s); + mutex_unlock(&ep->rma_info.rma_lock); + } + mutex_unlock(&scif_info.connlock); + return 0; +} + +static int scif_rma_test_open(struct inode *inode, struct file *file) +{ + return single_open(file, scif_rma_test, inode->i_private); +} + +static int scif_rma_test_release(struct inode *inode, struct file *file) +{ + return single_release(inode, file); +} + +static const struct file_operations scif_rma_ops = { + .owner = THIS_MODULE, + .open = scif_rma_test_open, + .read = seq_read, + .llseek = seq_lseek, + .release = scif_rma_test_release +}; + +void __init scif_init_debugfs(void) +{ scif_dbg = debugfs_create_dir(KBUILD_MODNAME, NULL); if (!scif_dbg) { dev_err(scif_info.mdev.this_device, @@ -73,8 +150,8 @@ void __init scif_init_debugfs(void) return; } - d = debugfs_create_file("scif_dev", 0444, scif_dbg, - NULL, &scif_dev_ops); + debugfs_create_file("scif_dev", 0444, scif_dbg, NULL, &scif_dev_ops); + debugfs_create_file("scif_rma", 0444, scif_dbg, NULL, &scif_rma_ops); debugfs_create_u8("en_msg_log", 0666, scif_dbg, &scif_info.en_msg_log); debugfs_create_u8("p2p_enable", 0666, scif_dbg, &scif_info.p2p_enable); } diff --git a/drivers/misc/mic/scif/scif_dma.c b/drivers/misc/mic/scif/scif_dma.c new file mode 100644 index 000000000000..95a13c629a8e --- /dev/null +++ b/drivers/misc/mic/scif/scif_dma.c @@ -0,0 +1,1979 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * Intel SCIF driver. + * + */ +#include "scif_main.h" +#include "scif_map.h" + +/* + * struct scif_dma_comp_cb - SCIF DMA completion callback + * + * @dma_completion_func: DMA completion callback + * @cb_cookie: DMA completion callback cookie + * @temp_buf: Temporary buffer + * @temp_buf_to_free: Temporary buffer to be freed + * @is_cache: Is a kmem_cache allocated buffer + * @dst_offset: Destination registration offset + * @dst_window: Destination registration window + * @len: Length of the temp buffer + * @temp_phys: DMA address of the temp buffer + * @sdev: The SCIF device + * @header_padding: padding for cache line alignment + */ +struct scif_dma_comp_cb { + void (*dma_completion_func)(void *cookie); + void *cb_cookie; + u8 *temp_buf; + u8 *temp_buf_to_free; + bool is_cache; + s64 dst_offset; + struct scif_window *dst_window; + size_t len; + dma_addr_t temp_phys; + struct scif_dev *sdev; + int header_padding; +}; + +/** + * struct scif_copy_work - Work for DMA copy + * + * @src_offset: Starting source offset + * @dst_offset: Starting destination offset + * @src_window: Starting src registered window + * @dst_window: Starting dst registered window + * @loopback: true if this is a loopback DMA transfer + * @len: Length of the transfer + * @comp_cb: DMA copy completion callback + * @remote_dev: The remote SCIF peer device + * @fence_type: polling or interrupt based + * @ordered: is this a tail byte ordered DMA transfer + */ +struct scif_copy_work { + s64 src_offset; + s64 dst_offset; + struct scif_window *src_window; + struct scif_window *dst_window; + int loopback; + size_t len; + struct scif_dma_comp_cb *comp_cb; + struct scif_dev *remote_dev; + int fence_type; + bool ordered; +}; + +#ifndef list_entry_next +#define list_entry_next(pos, member) \ + list_entry(pos->member.next, typeof(*pos), member) +#endif + +/** + * scif_reserve_dma_chan: + * @ep: Endpoint Descriptor. + * + * This routine reserves a DMA channel for a particular + * endpoint. All DMA transfers for an endpoint are always + * programmed on the same DMA channel. + */ +int scif_reserve_dma_chan(struct scif_endpt *ep) +{ + int err = 0; + struct scif_dev *scifdev; + struct scif_hw_dev *sdev; + struct dma_chan *chan; + + /* Loopback DMAs are not supported on the management node */ + if (!scif_info.nodeid && scifdev_self(ep->remote_dev)) + return 0; + if (scif_info.nodeid) + scifdev = &scif_dev[0]; + else + scifdev = ep->remote_dev; + sdev = scifdev->sdev; + if (!sdev->num_dma_ch) + return -ENODEV; + chan = sdev->dma_ch[scifdev->dma_ch_idx]; + scifdev->dma_ch_idx = (scifdev->dma_ch_idx + 1) % sdev->num_dma_ch; + mutex_lock(&ep->rma_info.rma_lock); + ep->rma_info.dma_chan = chan; + mutex_unlock(&ep->rma_info.rma_lock); + return err; +} + +#ifdef CONFIG_MMU_NOTIFIER +/** + * scif_rma_destroy_tcw: + * + * This routine destroys temporary cached windows + */ +static +void __scif_rma_destroy_tcw(struct scif_mmu_notif *mmn, + struct scif_endpt *ep, + u64 start, u64 len) +{ + struct list_head *item, *tmp; + struct scif_window *window; + u64 start_va, end_va; + u64 end = start + len; + + if (end <= start) + return; + + list_for_each_safe(item, tmp, &mmn->tc_reg_list) { + window = list_entry(item, struct scif_window, list); + ep = (struct scif_endpt *)window->ep; + if (!len) + break; + start_va = window->va_for_temp; + end_va = start_va + (window->nr_pages << PAGE_SHIFT); + if (start < start_va && end <= start_va) + break; + if (start >= end_va) + continue; + __scif_rma_destroy_tcw_helper(window); + } +} + +static void scif_rma_destroy_tcw(struct scif_mmu_notif *mmn, u64 start, u64 len) +{ + struct scif_endpt *ep = mmn->ep; + + spin_lock(&ep->rma_info.tc_lock); + __scif_rma_destroy_tcw(mmn, ep, start, len); + spin_unlock(&ep->rma_info.tc_lock); +} + +static void scif_rma_destroy_tcw_ep(struct scif_endpt *ep) +{ + struct list_head *item, *tmp; + struct scif_mmu_notif *mmn; + + list_for_each_safe(item, tmp, &ep->rma_info.mmn_list) { + mmn = list_entry(item, struct scif_mmu_notif, list); + scif_rma_destroy_tcw(mmn, 0, ULONG_MAX); + } +} + +static void __scif_rma_destroy_tcw_ep(struct scif_endpt *ep) +{ + struct list_head *item, *tmp; + struct scif_mmu_notif *mmn; + + spin_lock(&ep->rma_info.tc_lock); + list_for_each_safe(item, tmp, &ep->rma_info.mmn_list) { + mmn = list_entry(item, struct scif_mmu_notif, list); + __scif_rma_destroy_tcw(mmn, ep, 0, ULONG_MAX); + } + spin_unlock(&ep->rma_info.tc_lock); +} + +static bool scif_rma_tc_can_cache(struct scif_endpt *ep, size_t cur_bytes) +{ + if ((cur_bytes >> PAGE_SHIFT) > scif_info.rma_tc_limit) + return false; + if ((atomic_read(&ep->rma_info.tcw_total_pages) + + (cur_bytes >> PAGE_SHIFT)) > + scif_info.rma_tc_limit) { + dev_info(scif_info.mdev.this_device, + "%s %d total=%d, current=%zu reached max\n", + __func__, __LINE__, + atomic_read(&ep->rma_info.tcw_total_pages), + (1 + (cur_bytes >> PAGE_SHIFT))); + scif_rma_destroy_tcw_invalid(); + __scif_rma_destroy_tcw_ep(ep); + } + return true; +} + +static void scif_mmu_notifier_release(struct mmu_notifier *mn, + struct mm_struct *mm) +{ + struct scif_mmu_notif *mmn; + + mmn = container_of(mn, struct scif_mmu_notif, ep_mmu_notifier); + scif_rma_destroy_tcw(mmn, 0, ULONG_MAX); + schedule_work(&scif_info.misc_work); +} + +static void scif_mmu_notifier_invalidate_page(struct mmu_notifier *mn, + struct mm_struct *mm, + unsigned long address) +{ + struct scif_mmu_notif *mmn; + + mmn = container_of(mn, struct scif_mmu_notif, ep_mmu_notifier); + scif_rma_destroy_tcw(mmn, address, PAGE_SIZE); +} + +static void scif_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn, + struct mm_struct *mm, + unsigned long start, + unsigned long end) +{ + struct scif_mmu_notif *mmn; + + mmn = container_of(mn, struct scif_mmu_notif, ep_mmu_notifier); + scif_rma_destroy_tcw(mmn, start, end - start); +} + +static void scif_mmu_notifier_invalidate_range_end(struct mmu_notifier *mn, + struct mm_struct *mm, + unsigned long start, + unsigned long end) +{ + /* + * Nothing to do here, everything needed was done in + * invalidate_range_start. + */ +} + +static const struct mmu_notifier_ops scif_mmu_notifier_ops = { + .release = scif_mmu_notifier_release, + .clear_flush_young = NULL, + .invalidate_page = scif_mmu_notifier_invalidate_page, + .invalidate_range_start = scif_mmu_notifier_invalidate_range_start, + .invalidate_range_end = scif_mmu_notifier_invalidate_range_end}; + +static void scif_ep_unregister_mmu_notifier(struct scif_endpt *ep) +{ + struct scif_endpt_rma_info *rma = &ep->rma_info; + struct scif_mmu_notif *mmn = NULL; + struct list_head *item, *tmp; + + mutex_lock(&ep->rma_info.mmn_lock); + list_for_each_safe(item, tmp, &rma->mmn_list) { + mmn = list_entry(item, struct scif_mmu_notif, list); + mmu_notifier_unregister(&mmn->ep_mmu_notifier, mmn->mm); + list_del(item); + kfree(mmn); + } + mutex_unlock(&ep->rma_info.mmn_lock); +} + +static void scif_init_mmu_notifier(struct scif_mmu_notif *mmn, + struct mm_struct *mm, struct scif_endpt *ep) +{ + mmn->ep = ep; + mmn->mm = mm; + mmn->ep_mmu_notifier.ops = &scif_mmu_notifier_ops; + INIT_LIST_HEAD(&mmn->list); + INIT_LIST_HEAD(&mmn->tc_reg_list); +} + +static struct scif_mmu_notif * +scif_find_mmu_notifier(struct mm_struct *mm, struct scif_endpt_rma_info *rma) +{ + struct scif_mmu_notif *mmn; + struct list_head *item; + + list_for_each(item, &rma->mmn_list) { + mmn = list_entry(item, struct scif_mmu_notif, list); + if (mmn->mm == mm) + return mmn; + } + return NULL; +} + +static struct scif_mmu_notif * +scif_add_mmu_notifier(struct mm_struct *mm, struct scif_endpt *ep) +{ + struct scif_mmu_notif *mmn + = kzalloc(sizeof(*mmn), GFP_KERNEL); + + if (!mmn) + return ERR_PTR(ENOMEM); + + scif_init_mmu_notifier(mmn, current->mm, ep); + if (mmu_notifier_register(&mmn->ep_mmu_notifier, + current->mm)) { + kfree(mmn); + return ERR_PTR(EBUSY); + } + list_add(&mmn->list, &ep->rma_info.mmn_list); + return mmn; +} + +/* + * Called from the misc thread to destroy temporary cached windows and + * unregister the MMU notifier for the SCIF endpoint. + */ +void scif_mmu_notif_handler(struct work_struct *work) +{ + struct list_head *pos, *tmpq; + struct scif_endpt *ep; +restart: + scif_rma_destroy_tcw_invalid(); + spin_lock(&scif_info.rmalock); + list_for_each_safe(pos, tmpq, &scif_info.mmu_notif_cleanup) { + ep = list_entry(pos, struct scif_endpt, mmu_list); + list_del(&ep->mmu_list); + spin_unlock(&scif_info.rmalock); + scif_rma_destroy_tcw_ep(ep); + scif_ep_unregister_mmu_notifier(ep); + goto restart; + } + spin_unlock(&scif_info.rmalock); +} + +static bool scif_is_set_reg_cache(int flags) +{ + return !!(flags & SCIF_RMA_USECACHE); +} +#else +static struct scif_mmu_notif * +scif_find_mmu_notifier(struct mm_struct *mm, + struct scif_endpt_rma_info *rma) +{ + return NULL; +} + +static struct scif_mmu_notif * +scif_add_mmu_notifier(struct mm_struct *mm, struct scif_endpt *ep) +{ + return NULL; +} + +void scif_mmu_notif_handler(struct work_struct *work) +{ +} + +static bool scif_is_set_reg_cache(int flags) +{ + return false; +} + +static bool scif_rma_tc_can_cache(struct scif_endpt *ep, size_t cur_bytes) +{ + return false; +} +#endif + +/** + * scif_register_temp: + * @epd: End Point Descriptor. + * @addr: virtual address to/from which to copy + * @len: length of range to copy + * @out_offset: computed offset returned by reference. + * @out_window: allocated registered window returned by reference. + * + * Create a temporary registered window. The peer will not know about this + * window. This API is used for scif_vreadfrom()/scif_vwriteto() API's. + */ +static int +scif_register_temp(scif_epd_t epd, unsigned long addr, size_t len, int prot, + off_t *out_offset, struct scif_window **out_window) +{ + struct scif_endpt *ep = (struct scif_endpt *)epd; + int err; + scif_pinned_pages_t pinned_pages; + size_t aligned_len; + + aligned_len = ALIGN(len, PAGE_SIZE); + + err = __scif_pin_pages((void *)(addr & PAGE_MASK), + aligned_len, &prot, 0, &pinned_pages); + if (err) + return err; + + pinned_pages->prot = prot; + + /* Compute the offset for this registration */ + err = scif_get_window_offset(ep, 0, 0, + aligned_len >> PAGE_SHIFT, + (s64 *)out_offset); + if (err) + goto error_unpin; + + /* Allocate and prepare self registration window */ + *out_window = scif_create_window(ep, aligned_len >> PAGE_SHIFT, + *out_offset, true); + if (!*out_window) { + scif_free_window_offset(ep, NULL, *out_offset); + err = -ENOMEM; + goto error_unpin; + } + + (*out_window)->pinned_pages = pinned_pages; + (*out_window)->nr_pages = pinned_pages->nr_pages; + (*out_window)->prot = pinned_pages->prot; + + (*out_window)->va_for_temp = addr & PAGE_MASK; + err = scif_map_window(ep->remote_dev, *out_window); + if (err) { + /* Something went wrong! Rollback */ + scif_destroy_window(ep, *out_window); + *out_window = NULL; + } else { + *out_offset |= (addr - (*out_window)->va_for_temp); + } + return err; +error_unpin: + if (err) + dev_err(&ep->remote_dev->sdev->dev, + "%s %d err %d\n", __func__, __LINE__, err); + scif_unpin_pages(pinned_pages); + return err; +} + +#define SCIF_DMA_TO (3 * HZ) + +/* + * scif_sync_dma - Program a DMA without an interrupt descriptor + * + * @dev - The address of the pointer to the device instance used + * for DMA registration. + * @chan - DMA channel to be used. + * @sync_wait: Wait for DMA to complete? + * + * Return 0 on success and -errno on error. + */ +static int scif_sync_dma(struct scif_hw_dev *sdev, struct dma_chan *chan, + bool sync_wait) +{ + int err = 0; + struct dma_async_tx_descriptor *tx = NULL; + enum dma_ctrl_flags flags = DMA_PREP_FENCE; + dma_cookie_t cookie; + struct dma_device *ddev; + + if (!chan) { + err = -EIO; + dev_err(&sdev->dev, "%s %d err %d\n", + __func__, __LINE__, err); + return err; + } + ddev = chan->device; + + tx = ddev->device_prep_dma_memcpy(chan, 0, 0, 0, flags); + if (!tx) { + err = -ENOMEM; + dev_err(&sdev->dev, "%s %d err %d\n", + __func__, __LINE__, err); + goto release; + } + cookie = tx->tx_submit(tx); + + if (dma_submit_error(cookie)) { + err = -ENOMEM; + dev_err(&sdev->dev, "%s %d err %d\n", + __func__, __LINE__, err); + goto release; + } + if (!sync_wait) { + dma_async_issue_pending(chan); + } else { + if (dma_sync_wait(chan, cookie) == DMA_COMPLETE) { + err = 0; + } else { + err = -EIO; + dev_err(&sdev->dev, "%s %d err %d\n", + __func__, __LINE__, err); + } + } +release: + return err; +} + +static void scif_dma_callback(void *arg) +{ + struct completion *done = (struct completion *)arg; + + complete(done); +} + +#define SCIF_DMA_SYNC_WAIT true +#define SCIF_DMA_POLL BIT(0) +#define SCIF_DMA_INTR BIT(1) + +/* + * scif_async_dma - Program a DMA with an interrupt descriptor + * + * @dev - The address of the pointer to the device instance used + * for DMA registration. + * @chan - DMA channel to be used. + * Return 0 on success and -errno on error. + */ +static int scif_async_dma(struct scif_hw_dev *sdev, struct dma_chan *chan) +{ + int err = 0; + struct dma_device *ddev; + struct dma_async_tx_descriptor *tx = NULL; + enum dma_ctrl_flags flags = DMA_PREP_INTERRUPT | DMA_PREP_FENCE; + DECLARE_COMPLETION_ONSTACK(done_wait); + dma_cookie_t cookie; + enum dma_status status; + + if (!chan) { + err = -EIO; + dev_err(&sdev->dev, "%s %d err %d\n", + __func__, __LINE__, err); + return err; + } + ddev = chan->device; + + tx = ddev->device_prep_dma_memcpy(chan, 0, 0, 0, flags); + if (!tx) { + err = -ENOMEM; + dev_err(&sdev->dev, "%s %d err %d\n", + __func__, __LINE__, err); + goto release; + } + reinit_completion(&done_wait); + tx->callback = scif_dma_callback; + tx->callback_param = &done_wait; + cookie = tx->tx_submit(tx); + + if (dma_submit_error(cookie)) { + err = -ENOMEM; + dev_err(&sdev->dev, "%s %d err %d\n", + __func__, __LINE__, err); + goto release; + } + dma_async_issue_pending(chan); + + err = wait_for_completion_timeout(&done_wait, SCIF_DMA_TO); + if (!err) { + err = -EIO; + dev_err(&sdev->dev, "%s %d err %d\n", + __func__, __LINE__, err); + goto release; + } + err = 0; + status = dma_async_is_tx_complete(chan, cookie, NULL, NULL); + if (status != DMA_COMPLETE) { + err = -EIO; + dev_err(&sdev->dev, "%s %d err %d\n", + __func__, __LINE__, err); + goto release; + } +release: + return err; +} + +/* + * scif_drain_dma_poll - Drain all outstanding DMA operations for a particular + * DMA channel via polling. + * + * @sdev - The SCIF device + * @chan - DMA channel + * Return 0 on success and -errno on error. + */ +static int scif_drain_dma_poll(struct scif_hw_dev *sdev, struct dma_chan *chan) +{ + if (!chan) + return -EINVAL; + return scif_sync_dma(sdev, chan, SCIF_DMA_SYNC_WAIT); +} + +/* + * scif_drain_dma_intr - Drain all outstanding DMA operations for a particular + * DMA channel via interrupt based blocking wait. + * + * @sdev - The SCIF device + * @chan - DMA channel + * Return 0 on success and -errno on error. + */ +int scif_drain_dma_intr(struct scif_hw_dev *sdev, struct dma_chan *chan) +{ + if (!chan) + return -EINVAL; + return scif_async_dma(sdev, chan); +} + +/** + * scif_rma_destroy_windows: + * + * This routine destroys all windows queued for cleanup + */ +void scif_rma_destroy_windows(void) +{ + struct list_head *item, *tmp; + struct scif_window *window; + struct scif_endpt *ep; + struct dma_chan *chan; + + might_sleep(); +restart: + spin_lock(&scif_info.rmalock); + list_for_each_safe(item, tmp, &scif_info.rma) { + window = list_entry(item, struct scif_window, + list); + ep = (struct scif_endpt *)window->ep; + chan = ep->rma_info.dma_chan; + + list_del_init(&window->list); + spin_unlock(&scif_info.rmalock); + if (!chan || !scifdev_alive(ep) || + !scif_drain_dma_intr(ep->remote_dev->sdev, + ep->rma_info.dma_chan)) + /* Remove window from global list */ + window->unreg_state = OP_COMPLETED; + else + dev_warn(&ep->remote_dev->sdev->dev, + "DMA engine hung?\n"); + if (window->unreg_state == OP_COMPLETED) { + if (window->type == SCIF_WINDOW_SELF) + scif_destroy_window(ep, window); + else + scif_destroy_remote_window(window); + atomic_dec(&ep->rma_info.tw_refcount); + } + goto restart; + } + spin_unlock(&scif_info.rmalock); +} + +/** + * scif_rma_destroy_tcw: + * + * This routine destroys temporary cached registered windows + * which have been queued for cleanup. + */ +void scif_rma_destroy_tcw_invalid(void) +{ + struct list_head *item, *tmp; + struct scif_window *window; + struct scif_endpt *ep; + struct dma_chan *chan; + + might_sleep(); +restart: + spin_lock(&scif_info.rmalock); + list_for_each_safe(item, tmp, &scif_info.rma_tc) { + window = list_entry(item, struct scif_window, list); + ep = (struct scif_endpt *)window->ep; + chan = ep->rma_info.dma_chan; + list_del_init(&window->list); + spin_unlock(&scif_info.rmalock); + mutex_lock(&ep->rma_info.rma_lock); + if (!chan || !scifdev_alive(ep) || + !scif_drain_dma_intr(ep->remote_dev->sdev, + ep->rma_info.dma_chan)) { + atomic_sub(window->nr_pages, + &ep->rma_info.tcw_total_pages); + scif_destroy_window(ep, window); + atomic_dec(&ep->rma_info.tcw_refcount); + } else { + dev_warn(&ep->remote_dev->sdev->dev, + "DMA engine hung?\n"); + } + mutex_unlock(&ep->rma_info.rma_lock); + goto restart; + } + spin_unlock(&scif_info.rmalock); +} + +static inline +void *_get_local_va(off_t off, struct scif_window *window, size_t len) +{ + int page_nr = (off - window->offset) >> PAGE_SHIFT; + off_t page_off = off & ~PAGE_MASK; + void *va = NULL; + + if (window->type == SCIF_WINDOW_SELF) { + struct page **pages = window->pinned_pages->pages; + + va = page_address(pages[page_nr]) + page_off; + } + return va; +} + +static inline +void *ioremap_remote(off_t off, struct scif_window *window, + size_t len, struct scif_dev *dev, + struct scif_window_iter *iter) +{ + dma_addr_t phys = scif_off_to_dma_addr(window, off, NULL, iter); + + /* + * If the DMA address is not card relative then we need the DMA + * addresses to be an offset into the bar. The aperture base was already + * added so subtract it here since scif_ioremap is going to add it again + */ + if (!scifdev_self(dev) && window->type == SCIF_WINDOW_PEER && + dev->sdev->aper && !dev->sdev->card_rel_da) + phys = phys - dev->sdev->aper->pa; + return scif_ioremap(phys, len, dev); +} + +static inline void +iounmap_remote(void *virt, size_t size, struct scif_copy_work *work) +{ + scif_iounmap(virt, size, work->remote_dev); +} + +/* + * Takes care of ordering issue caused by + * 1. Hardware: Only in the case of cpu copy from mgmt node to card + * because of WC memory. + * 2. Software: If memcpy reorders copy instructions for optimization. + * This could happen at both mgmt node and card. + */ +static inline void +scif_ordered_memcpy_toio(char *dst, const char *src, size_t count) +{ + if (!count) + return; + + memcpy_toio((void __iomem __force *)dst, src, --count); + /* Order the last byte with the previous stores */ + wmb(); + *(dst + count) = *(src + count); +} + +static inline void scif_unaligned_cpy_toio(char *dst, const char *src, + size_t count, bool ordered) +{ + if (ordered) + scif_ordered_memcpy_toio(dst, src, count); + else + memcpy_toio((void __iomem __force *)dst, src, count); +} + +static inline +void scif_ordered_memcpy_fromio(char *dst, const char *src, size_t count) +{ + if (!count) + return; + + memcpy_fromio(dst, (void __iomem __force *)src, --count); + /* Order the last byte with the previous loads */ + rmb(); + *(dst + count) = *(src + count); +} + +static inline void scif_unaligned_cpy_fromio(char *dst, const char *src, + size_t count, bool ordered) +{ + if (ordered) + scif_ordered_memcpy_fromio(dst, src, count); + else + memcpy_fromio(dst, (void __iomem __force *)src, count); +} + +#define SCIF_RMA_ERROR_CODE (~(dma_addr_t)0x0) + +/* + * scif_off_to_dma_addr: + * Obtain the dma_addr given the window and the offset. + * @window: Registered window. + * @off: Window offset. + * @nr_bytes: Return the number of contiguous bytes till next DMA addr index. + * @index: Return the index of the dma_addr array found. + * @start_off: start offset of index of the dma addr array found. + * The nr_bytes provides the callee an estimate of the maximum possible + * DMA xfer possible while the index/start_off provide faster lookups + * for the next iteration. + */ +dma_addr_t scif_off_to_dma_addr(struct scif_window *window, s64 off, + size_t *nr_bytes, struct scif_window_iter *iter) +{ + int i, page_nr; + s64 start, end; + off_t page_off; + + if (window->nr_pages == window->nr_contig_chunks) { + page_nr = (off - window->offset) >> PAGE_SHIFT; + page_off = off & ~PAGE_MASK; + + if (nr_bytes) + *nr_bytes = PAGE_SIZE - page_off; + return window->dma_addr[page_nr] | page_off; + } + if (iter) { + i = iter->index; + start = iter->offset; + } else { + i = 0; + start = window->offset; + } + for (; i < window->nr_contig_chunks; i++) { + end = start + (window->num_pages[i] << PAGE_SHIFT); + if (off >= start && off < end) { + if (iter) { + iter->index = i; + iter->offset = start; + } + if (nr_bytes) + *nr_bytes = end - off; + return (window->dma_addr[i] + (off - start)); + } + start += (window->num_pages[i] << PAGE_SHIFT); + } + dev_err(scif_info.mdev.this_device, + "%s %d BUG. Addr not found? window %p off 0x%llx\n", + __func__, __LINE__, window, off); + return SCIF_RMA_ERROR_CODE; +} + +/* + * Copy between rma window and temporary buffer + */ +static void scif_rma_local_cpu_copy(s64 offset, struct scif_window *window, + u8 *temp, size_t rem_len, bool to_temp) +{ + void *window_virt; + size_t loop_len; + int offset_in_page; + s64 end_offset; + + offset_in_page = offset & ~PAGE_MASK; + loop_len = PAGE_SIZE - offset_in_page; + + if (rem_len < loop_len) + loop_len = rem_len; + + window_virt = _get_local_va(offset, window, loop_len); + if (!window_virt) + return; + if (to_temp) + memcpy(temp, window_virt, loop_len); + else + memcpy(window_virt, temp, loop_len); + + offset += loop_len; + temp += loop_len; + rem_len -= loop_len; + + end_offset = window->offset + + (window->nr_pages << PAGE_SHIFT); + while (rem_len) { + if (offset == end_offset) { + window = list_entry_next(window, list); + end_offset = window->offset + + (window->nr_pages << PAGE_SHIFT); + } + loop_len = min(PAGE_SIZE, rem_len); + window_virt = _get_local_va(offset, window, loop_len); + if (!window_virt) + return; + if (to_temp) + memcpy(temp, window_virt, loop_len); + else + memcpy(window_virt, temp, loop_len); + offset += loop_len; + temp += loop_len; + rem_len -= loop_len; + } +} + +/** + * scif_rma_completion_cb: + * @data: RMA cookie + * + * RMA interrupt completion callback. + */ +static void scif_rma_completion_cb(void *data) +{ + struct scif_dma_comp_cb *comp_cb = data; + + /* Free DMA Completion CB. */ + if (comp_cb->dst_window) + scif_rma_local_cpu_copy(comp_cb->dst_offset, + comp_cb->dst_window, + comp_cb->temp_buf + + comp_cb->header_padding, + comp_cb->len, false); + scif_unmap_single(comp_cb->temp_phys, comp_cb->sdev, + SCIF_KMEM_UNALIGNED_BUF_SIZE); + if (comp_cb->is_cache) + kmem_cache_free(unaligned_cache, + comp_cb->temp_buf_to_free); + else + kfree(comp_cb->temp_buf_to_free); +} + +/* Copies between temporary buffer and offsets provided in work */ +static int +scif_rma_list_dma_copy_unaligned(struct scif_copy_work *work, + u8 *temp, struct dma_chan *chan, + bool src_local) +{ + struct scif_dma_comp_cb *comp_cb = work->comp_cb; + dma_addr_t window_dma_addr, temp_dma_addr; + dma_addr_t temp_phys = comp_cb->temp_phys; + size_t loop_len, nr_contig_bytes = 0, remaining_len = work->len; + int offset_in_ca, ret = 0; + s64 end_offset, offset; + struct scif_window *window; + void *window_virt_addr; + size_t tail_len; + struct dma_async_tx_descriptor *tx; + struct dma_device *dev = chan->device; + dma_cookie_t cookie; + + if (src_local) { + offset = work->dst_offset; + window = work->dst_window; + } else { + offset = work->src_offset; + window = work->src_window; + } + + offset_in_ca = offset & (L1_CACHE_BYTES - 1); + if (offset_in_ca) { + loop_len = L1_CACHE_BYTES - offset_in_ca; + loop_len = min(loop_len, remaining_len); + window_virt_addr = ioremap_remote(offset, window, + loop_len, + work->remote_dev, + NULL); + if (!window_virt_addr) + return -ENOMEM; + if (src_local) + scif_unaligned_cpy_toio(window_virt_addr, temp, + loop_len, + work->ordered && + !(remaining_len - loop_len)); + else + scif_unaligned_cpy_fromio(temp, window_virt_addr, + loop_len, work->ordered && + !(remaining_len - loop_len)); + iounmap_remote(window_virt_addr, loop_len, work); + + offset += loop_len; + temp += loop_len; + temp_phys += loop_len; + remaining_len -= loop_len; + } + + offset_in_ca = offset & ~PAGE_MASK; + end_offset = window->offset + + (window->nr_pages << PAGE_SHIFT); + + tail_len = remaining_len & (L1_CACHE_BYTES - 1); + remaining_len -= tail_len; + while (remaining_len) { + if (offset == end_offset) { + window = list_entry_next(window, list); + end_offset = window->offset + + (window->nr_pages << PAGE_SHIFT); + } + if (scif_is_mgmt_node()) + temp_dma_addr = temp_phys; + else + /* Fix if we ever enable IOMMU on the card */ + temp_dma_addr = (dma_addr_t)virt_to_phys(temp); + window_dma_addr = scif_off_to_dma_addr(window, offset, + &nr_contig_bytes, + NULL); + loop_len = min(nr_contig_bytes, remaining_len); + if (src_local) { + if (work->ordered && !tail_len && + !(remaining_len - loop_len) && + loop_len != L1_CACHE_BYTES) { + /* + * Break up the last chunk of the transfer into + * two steps. if there is no tail to guarantee + * DMA ordering. SCIF_DMA_POLLING inserts + * a status update descriptor in step 1 which + * acts as a double sided synchronization fence + * for the DMA engine to ensure that the last + * cache line in step 2 is updated last. + */ + /* Step 1) DMA: Body Length - L1_CACHE_BYTES. */ + tx = + dev->device_prep_dma_memcpy(chan, + window_dma_addr, + temp_dma_addr, + loop_len - + L1_CACHE_BYTES, + DMA_PREP_FENCE); + if (!tx) { + ret = -ENOMEM; + goto err; + } + cookie = tx->tx_submit(tx); + if (dma_submit_error(cookie)) { + ret = -ENOMEM; + goto err; + } + dma_async_issue_pending(chan); + offset += (loop_len - L1_CACHE_BYTES); + temp_dma_addr += (loop_len - L1_CACHE_BYTES); + window_dma_addr += (loop_len - L1_CACHE_BYTES); + remaining_len -= (loop_len - L1_CACHE_BYTES); + loop_len = remaining_len; + + /* Step 2) DMA: L1_CACHE_BYTES */ + tx = + dev->device_prep_dma_memcpy(chan, + window_dma_addr, + temp_dma_addr, + loop_len, 0); + if (!tx) { + ret = -ENOMEM; + goto err; + } + cookie = tx->tx_submit(tx); + if (dma_submit_error(cookie)) { + ret = -ENOMEM; + goto err; + } + dma_async_issue_pending(chan); + } else { + tx = + dev->device_prep_dma_memcpy(chan, + window_dma_addr, + temp_dma_addr, + loop_len, 0); + if (!tx) { + ret = -ENOMEM; + goto err; + } + cookie = tx->tx_submit(tx); + if (dma_submit_error(cookie)) { + ret = -ENOMEM; + goto err; + } + dma_async_issue_pending(chan); + } + } else { + tx = dev->device_prep_dma_memcpy(chan, temp_dma_addr, + window_dma_addr, loop_len, 0); + if (!tx) { + ret = -ENOMEM; + goto err; + } + cookie = tx->tx_submit(tx); + if (dma_submit_error(cookie)) { + ret = -ENOMEM; + goto err; + } + dma_async_issue_pending(chan); + } + if (ret < 0) + goto err; + offset += loop_len; + temp += loop_len; + temp_phys += loop_len; + remaining_len -= loop_len; + offset_in_ca = 0; + } + if (tail_len) { + if (offset == end_offset) { + window = list_entry_next(window, list); + end_offset = window->offset + + (window->nr_pages << PAGE_SHIFT); + } + window_virt_addr = ioremap_remote(offset, window, tail_len, + work->remote_dev, + NULL); + if (!window_virt_addr) + return -ENOMEM; + /* + * The CPU copy for the tail bytes must be initiated only once + * previous DMA transfers for this endpoint have completed + * to guarantee ordering. + */ + if (work->ordered) { + struct scif_dev *rdev = work->remote_dev; + + ret = scif_drain_dma_intr(rdev->sdev, chan); + if (ret) + return ret; + } + if (src_local) + scif_unaligned_cpy_toio(window_virt_addr, temp, + tail_len, work->ordered); + else + scif_unaligned_cpy_fromio(temp, window_virt_addr, + tail_len, work->ordered); + iounmap_remote(window_virt_addr, tail_len, work); + } + tx = dev->device_prep_dma_memcpy(chan, 0, 0, 0, DMA_PREP_INTERRUPT); + if (!tx) { + ret = -ENOMEM; + return ret; + } + tx->callback = &scif_rma_completion_cb; + tx->callback_param = comp_cb; + cookie = tx->tx_submit(tx); + + if (dma_submit_error(cookie)) { + ret = -ENOMEM; + return ret; + } + dma_async_issue_pending(chan); + return 0; +err: + dev_err(scif_info.mdev.this_device, + "%s %d Desc Prog Failed ret %d\n", + __func__, __LINE__, ret); + return ret; +} + +/* + * _scif_rma_list_dma_copy_aligned: + * + * Traverse all the windows and perform DMA copy. + */ +static int _scif_rma_list_dma_copy_aligned(struct scif_copy_work *work, + struct dma_chan *chan) +{ + dma_addr_t src_dma_addr, dst_dma_addr; + size_t loop_len, remaining_len, src_contig_bytes = 0; + size_t dst_contig_bytes = 0; + struct scif_window_iter src_win_iter; + struct scif_window_iter dst_win_iter; + s64 end_src_offset, end_dst_offset; + struct scif_window *src_window = work->src_window; + struct scif_window *dst_window = work->dst_window; + s64 src_offset = work->src_offset, dst_offset = work->dst_offset; + int ret = 0; + struct dma_async_tx_descriptor *tx; + struct dma_device *dev = chan->device; + dma_cookie_t cookie; + + remaining_len = work->len; + + scif_init_window_iter(src_window, &src_win_iter); + scif_init_window_iter(dst_window, &dst_win_iter); + end_src_offset = src_window->offset + + (src_window->nr_pages << PAGE_SHIFT); + end_dst_offset = dst_window->offset + + (dst_window->nr_pages << PAGE_SHIFT); + while (remaining_len) { + if (src_offset == end_src_offset) { + src_window = list_entry_next(src_window, list); + end_src_offset = src_window->offset + + (src_window->nr_pages << PAGE_SHIFT); + scif_init_window_iter(src_window, &src_win_iter); + } + if (dst_offset == end_dst_offset) { + dst_window = list_entry_next(dst_window, list); + end_dst_offset = dst_window->offset + + (dst_window->nr_pages << PAGE_SHIFT); + scif_init_window_iter(dst_window, &dst_win_iter); + } + + /* compute dma addresses for transfer */ + src_dma_addr = scif_off_to_dma_addr(src_window, src_offset, + &src_contig_bytes, + &src_win_iter); + dst_dma_addr = scif_off_to_dma_addr(dst_window, dst_offset, + &dst_contig_bytes, + &dst_win_iter); + loop_len = min(src_contig_bytes, dst_contig_bytes); + loop_len = min(loop_len, remaining_len); + if (work->ordered && !(remaining_len - loop_len)) { + /* + * Break up the last chunk of the transfer into two + * steps to ensure that the last byte in step 2 is + * updated last. + */ + /* Step 1) DMA: Body Length - 1 */ + tx = dev->device_prep_dma_memcpy(chan, dst_dma_addr, + src_dma_addr, + loop_len - 1, + DMA_PREP_FENCE); + if (!tx) { + ret = -ENOMEM; + goto err; + } + cookie = tx->tx_submit(tx); + if (dma_submit_error(cookie)) { + ret = -ENOMEM; + goto err; + } + src_offset += (loop_len - 1); + dst_offset += (loop_len - 1); + src_dma_addr += (loop_len - 1); + dst_dma_addr += (loop_len - 1); + remaining_len -= (loop_len - 1); + loop_len = remaining_len; + + /* Step 2) DMA: 1 BYTES */ + tx = dev->device_prep_dma_memcpy(chan, dst_dma_addr, + src_dma_addr, loop_len, 0); + if (!tx) { + ret = -ENOMEM; + goto err; + } + cookie = tx->tx_submit(tx); + if (dma_submit_error(cookie)) { + ret = -ENOMEM; + goto err; + } + dma_async_issue_pending(chan); + } else { + tx = dev->device_prep_dma_memcpy(chan, dst_dma_addr, + src_dma_addr, loop_len, 0); + if (!tx) { + ret = -ENOMEM; + goto err; + } + cookie = tx->tx_submit(tx); + if (dma_submit_error(cookie)) { + ret = -ENOMEM; + goto err; + } + } + src_offset += loop_len; + dst_offset += loop_len; + remaining_len -= loop_len; + } + return ret; +err: + dev_err(scif_info.mdev.this_device, + "%s %d Desc Prog Failed ret %d\n", + __func__, __LINE__, ret); + return ret; +} + +/* + * scif_rma_list_dma_copy_aligned: + * + * Traverse all the windows and perform DMA copy. + */ +static int scif_rma_list_dma_copy_aligned(struct scif_copy_work *work, + struct dma_chan *chan) +{ + dma_addr_t src_dma_addr, dst_dma_addr; + size_t loop_len, remaining_len, tail_len, src_contig_bytes = 0; + size_t dst_contig_bytes = 0; + int src_cache_off; + s64 end_src_offset, end_dst_offset; + struct scif_window_iter src_win_iter; + struct scif_window_iter dst_win_iter; + void *src_virt, *dst_virt; + struct scif_window *src_window = work->src_window; + struct scif_window *dst_window = work->dst_window; + s64 src_offset = work->src_offset, dst_offset = work->dst_offset; + int ret = 0; + struct dma_async_tx_descriptor *tx; + struct dma_device *dev = chan->device; + dma_cookie_t cookie; + + remaining_len = work->len; + scif_init_window_iter(src_window, &src_win_iter); + scif_init_window_iter(dst_window, &dst_win_iter); + + src_cache_off = src_offset & (L1_CACHE_BYTES - 1); + if (src_cache_off != 0) { + /* Head */ + loop_len = L1_CACHE_BYTES - src_cache_off; + loop_len = min(loop_len, remaining_len); + src_dma_addr = __scif_off_to_dma_addr(src_window, src_offset); + dst_dma_addr = __scif_off_to_dma_addr(dst_window, dst_offset); + if (src_window->type == SCIF_WINDOW_SELF) + src_virt = _get_local_va(src_offset, src_window, + loop_len); + else + src_virt = ioremap_remote(src_offset, src_window, + loop_len, + work->remote_dev, NULL); + if (!src_virt) + return -ENOMEM; + if (dst_window->type == SCIF_WINDOW_SELF) + dst_virt = _get_local_va(dst_offset, dst_window, + loop_len); + else + dst_virt = ioremap_remote(dst_offset, dst_window, + loop_len, + work->remote_dev, NULL); + if (!dst_virt) { + if (src_window->type != SCIF_WINDOW_SELF) + iounmap_remote(src_virt, loop_len, work); + return -ENOMEM; + } + if (src_window->type == SCIF_WINDOW_SELF) + scif_unaligned_cpy_toio(dst_virt, src_virt, loop_len, + remaining_len == loop_len ? + work->ordered : false); + else + scif_unaligned_cpy_fromio(dst_virt, src_virt, loop_len, + remaining_len == loop_len ? + work->ordered : false); + if (src_window->type != SCIF_WINDOW_SELF) + iounmap_remote(src_virt, loop_len, work); + if (dst_window->type != SCIF_WINDOW_SELF) + iounmap_remote(dst_virt, loop_len, work); + src_offset += loop_len; + dst_offset += loop_len; + remaining_len -= loop_len; + } + + end_src_offset = src_window->offset + + (src_window->nr_pages << PAGE_SHIFT); + end_dst_offset = dst_window->offset + + (dst_window->nr_pages << PAGE_SHIFT); + tail_len = remaining_len & (L1_CACHE_BYTES - 1); + remaining_len -= tail_len; + while (remaining_len) { + if (src_offset == end_src_offset) { + src_window = list_entry_next(src_window, list); + end_src_offset = src_window->offset + + (src_window->nr_pages << PAGE_SHIFT); + scif_init_window_iter(src_window, &src_win_iter); + } + if (dst_offset == end_dst_offset) { + dst_window = list_entry_next(dst_window, list); + end_dst_offset = dst_window->offset + + (dst_window->nr_pages << PAGE_SHIFT); + scif_init_window_iter(dst_window, &dst_win_iter); + } + + /* compute dma addresses for transfer */ + src_dma_addr = scif_off_to_dma_addr(src_window, src_offset, + &src_contig_bytes, + &src_win_iter); + dst_dma_addr = scif_off_to_dma_addr(dst_window, dst_offset, + &dst_contig_bytes, + &dst_win_iter); + loop_len = min(src_contig_bytes, dst_contig_bytes); + loop_len = min(loop_len, remaining_len); + if (work->ordered && !tail_len && + !(remaining_len - loop_len)) { + /* + * Break up the last chunk of the transfer into two + * steps. if there is no tail to gurantee DMA ordering. + * Passing SCIF_DMA_POLLING inserts a status update + * descriptor in step 1 which acts as a double sided + * synchronization fence for the DMA engine to ensure + * that the last cache line in step 2 is updated last. + */ + /* Step 1) DMA: Body Length - L1_CACHE_BYTES. */ + tx = dev->device_prep_dma_memcpy(chan, dst_dma_addr, + src_dma_addr, + loop_len - + L1_CACHE_BYTES, + DMA_PREP_FENCE); + if (!tx) { + ret = -ENOMEM; + goto err; + } + cookie = tx->tx_submit(tx); + if (dma_submit_error(cookie)) { + ret = -ENOMEM; + goto err; + } + dma_async_issue_pending(chan); + src_offset += (loop_len - L1_CACHE_BYTES); + dst_offset += (loop_len - L1_CACHE_BYTES); + src_dma_addr += (loop_len - L1_CACHE_BYTES); + dst_dma_addr += (loop_len - L1_CACHE_BYTES); + remaining_len -= (loop_len - L1_CACHE_BYTES); + loop_len = remaining_len; + + /* Step 2) DMA: L1_CACHE_BYTES */ + tx = dev->device_prep_dma_memcpy(chan, dst_dma_addr, + src_dma_addr, + loop_len, 0); + if (!tx) { + ret = -ENOMEM; + goto err; + } + cookie = tx->tx_submit(tx); + if (dma_submit_error(cookie)) { + ret = -ENOMEM; + goto err; + } + dma_async_issue_pending(chan); + } else { + tx = dev->device_prep_dma_memcpy(chan, dst_dma_addr, + src_dma_addr, + loop_len, 0); + if (!tx) { + ret = -ENOMEM; + goto err; + } + cookie = tx->tx_submit(tx); + if (dma_submit_error(cookie)) { + ret = -ENOMEM; + goto err; + } + dma_async_issue_pending(chan); + } + src_offset += loop_len; + dst_offset += loop_len; + remaining_len -= loop_len; + } + remaining_len = tail_len; + if (remaining_len) { + loop_len = remaining_len; + if (src_offset == end_src_offset) + src_window = list_entry_next(src_window, list); + if (dst_offset == end_dst_offset) + dst_window = list_entry_next(dst_window, list); + + src_dma_addr = __scif_off_to_dma_addr(src_window, src_offset); + dst_dma_addr = __scif_off_to_dma_addr(dst_window, dst_offset); + /* + * The CPU copy for the tail bytes must be initiated only once + * previous DMA transfers for this endpoint have completed to + * guarantee ordering. + */ + if (work->ordered) { + struct scif_dev *rdev = work->remote_dev; + + ret = scif_drain_dma_poll(rdev->sdev, chan); + if (ret) + return ret; + } + if (src_window->type == SCIF_WINDOW_SELF) + src_virt = _get_local_va(src_offset, src_window, + loop_len); + else + src_virt = ioremap_remote(src_offset, src_window, + loop_len, + work->remote_dev, NULL); + if (!src_virt) + return -ENOMEM; + + if (dst_window->type == SCIF_WINDOW_SELF) + dst_virt = _get_local_va(dst_offset, dst_window, + loop_len); + else + dst_virt = ioremap_remote(dst_offset, dst_window, + loop_len, + work->remote_dev, NULL); + if (!dst_virt) { + if (src_window->type != SCIF_WINDOW_SELF) + iounmap_remote(src_virt, loop_len, work); + return -ENOMEM; + } + + if (src_window->type == SCIF_WINDOW_SELF) + scif_unaligned_cpy_toio(dst_virt, src_virt, loop_len, + work->ordered); + else + scif_unaligned_cpy_fromio(dst_virt, src_virt, + loop_len, work->ordered); + if (src_window->type != SCIF_WINDOW_SELF) + iounmap_remote(src_virt, loop_len, work); + + if (dst_window->type != SCIF_WINDOW_SELF) + iounmap_remote(dst_virt, loop_len, work); + remaining_len -= loop_len; + } + return ret; +err: + dev_err(scif_info.mdev.this_device, + "%s %d Desc Prog Failed ret %d\n", + __func__, __LINE__, ret); + return ret; +} + +/* + * scif_rma_list_cpu_copy: + * + * Traverse all the windows and perform CPU copy. + */ +static int scif_rma_list_cpu_copy(struct scif_copy_work *work) +{ + void *src_virt, *dst_virt; + size_t loop_len, remaining_len; + int src_page_off, dst_page_off; + s64 src_offset = work->src_offset, dst_offset = work->dst_offset; + struct scif_window *src_window = work->src_window; + struct scif_window *dst_window = work->dst_window; + s64 end_src_offset, end_dst_offset; + int ret = 0; + struct scif_window_iter src_win_iter; + struct scif_window_iter dst_win_iter; + + remaining_len = work->len; + + scif_init_window_iter(src_window, &src_win_iter); + scif_init_window_iter(dst_window, &dst_win_iter); + while (remaining_len) { + src_page_off = src_offset & ~PAGE_MASK; + dst_page_off = dst_offset & ~PAGE_MASK; + loop_len = min(PAGE_SIZE - + max(src_page_off, dst_page_off), + remaining_len); + + if (src_window->type == SCIF_WINDOW_SELF) + src_virt = _get_local_va(src_offset, src_window, + loop_len); + else + src_virt = ioremap_remote(src_offset, src_window, + loop_len, + work->remote_dev, + &src_win_iter); + if (!src_virt) { + ret = -ENOMEM; + goto error; + } + + if (dst_window->type == SCIF_WINDOW_SELF) + dst_virt = _get_local_va(dst_offset, dst_window, + loop_len); + else + dst_virt = ioremap_remote(dst_offset, dst_window, + loop_len, + work->remote_dev, + &dst_win_iter); + if (!dst_virt) { + if (src_window->type == SCIF_WINDOW_PEER) + iounmap_remote(src_virt, loop_len, work); + ret = -ENOMEM; + goto error; + } + + if (work->loopback) { + memcpy(dst_virt, src_virt, loop_len); + } else { + if (src_window->type == SCIF_WINDOW_SELF) + memcpy_toio((void __iomem __force *)dst_virt, + src_virt, loop_len); + else + memcpy_fromio(dst_virt, + (void __iomem __force *)src_virt, + loop_len); + } + if (src_window->type == SCIF_WINDOW_PEER) + iounmap_remote(src_virt, loop_len, work); + + if (dst_window->type == SCIF_WINDOW_PEER) + iounmap_remote(dst_virt, loop_len, work); + + src_offset += loop_len; + dst_offset += loop_len; + remaining_len -= loop_len; + if (remaining_len) { + end_src_offset = src_window->offset + + (src_window->nr_pages << PAGE_SHIFT); + end_dst_offset = dst_window->offset + + (dst_window->nr_pages << PAGE_SHIFT); + if (src_offset == end_src_offset) { + src_window = list_entry_next(src_window, list); + scif_init_window_iter(src_window, + &src_win_iter); + } + if (dst_offset == end_dst_offset) { + dst_window = list_entry_next(dst_window, list); + scif_init_window_iter(dst_window, + &dst_win_iter); + } + } + } +error: + return ret; +} + +static int scif_rma_list_dma_copy_wrapper(struct scif_endpt *epd, + struct scif_copy_work *work, + struct dma_chan *chan, off_t loffset) +{ + int src_cache_off, dst_cache_off; + s64 src_offset = work->src_offset, dst_offset = work->dst_offset; + u8 *temp = NULL; + bool src_local = true, dst_local = false; + struct scif_dma_comp_cb *comp_cb; + dma_addr_t src_dma_addr, dst_dma_addr; + int err; + + if (is_dma_copy_aligned(chan->device, 1, 1, 1)) + return _scif_rma_list_dma_copy_aligned(work, chan); + + src_cache_off = src_offset & (L1_CACHE_BYTES - 1); + dst_cache_off = dst_offset & (L1_CACHE_BYTES - 1); + + if (dst_cache_off == src_cache_off) + return scif_rma_list_dma_copy_aligned(work, chan); + + if (work->loopback) + return scif_rma_list_cpu_copy(work); + src_dma_addr = __scif_off_to_dma_addr(work->src_window, src_offset); + dst_dma_addr = __scif_off_to_dma_addr(work->dst_window, dst_offset); + src_local = work->src_window->type == SCIF_WINDOW_SELF; + dst_local = work->dst_window->type == SCIF_WINDOW_SELF; + + dst_local = dst_local; + /* Allocate dma_completion cb */ + comp_cb = kzalloc(sizeof(*comp_cb), GFP_KERNEL); + if (!comp_cb) + goto error; + + work->comp_cb = comp_cb; + comp_cb->cb_cookie = comp_cb; + comp_cb->dma_completion_func = &scif_rma_completion_cb; + + if (work->len + (L1_CACHE_BYTES << 1) < SCIF_KMEM_UNALIGNED_BUF_SIZE) { + comp_cb->is_cache = false; + /* Allocate padding bytes to align to a cache line */ + temp = kmalloc(work->len + (L1_CACHE_BYTES << 1), + GFP_KERNEL); + if (!temp) + goto free_comp_cb; + comp_cb->temp_buf_to_free = temp; + /* kmalloc(..) does not guarantee cache line alignment */ + if (!IS_ALIGNED((u64)temp, L1_CACHE_BYTES)) + temp = PTR_ALIGN(temp, L1_CACHE_BYTES); + } else { + comp_cb->is_cache = true; + temp = kmem_cache_alloc(unaligned_cache, GFP_KERNEL); + if (!temp) + goto free_comp_cb; + comp_cb->temp_buf_to_free = temp; + } + + if (src_local) { + temp += dst_cache_off; + scif_rma_local_cpu_copy(work->src_offset, work->src_window, + temp, work->len, true); + } else { + comp_cb->dst_window = work->dst_window; + comp_cb->dst_offset = work->dst_offset; + work->src_offset = work->src_offset - src_cache_off; + comp_cb->len = work->len; + work->len = ALIGN(work->len + src_cache_off, L1_CACHE_BYTES); + comp_cb->header_padding = src_cache_off; + } + comp_cb->temp_buf = temp; + + err = scif_map_single(&comp_cb->temp_phys, temp, + work->remote_dev, SCIF_KMEM_UNALIGNED_BUF_SIZE); + if (err) + goto free_temp_buf; + comp_cb->sdev = work->remote_dev; + if (scif_rma_list_dma_copy_unaligned(work, temp, chan, src_local) < 0) + goto free_temp_buf; + if (!src_local) + work->fence_type = SCIF_DMA_INTR; + return 0; +free_temp_buf: + if (comp_cb->is_cache) + kmem_cache_free(unaligned_cache, comp_cb->temp_buf_to_free); + else + kfree(comp_cb->temp_buf_to_free); +free_comp_cb: + kfree(comp_cb); +error: + return -ENOMEM; +} + +/** + * scif_rma_copy: + * @epd: end point descriptor. + * @loffset: offset in local registered address space to/from which to copy + * @addr: user virtual address to/from which to copy + * @len: length of range to copy + * @roffset: offset in remote registered address space to/from which to copy + * @flags: flags + * @dir: LOCAL->REMOTE or vice versa. + * @last_chunk: true if this is the last chunk of a larger transfer + * + * Validate parameters, check if src/dst registered ranges requested for copy + * are valid and initiate either CPU or DMA copy. + */ +static int scif_rma_copy(scif_epd_t epd, off_t loffset, unsigned long addr, + size_t len, off_t roffset, int flags, + enum scif_rma_dir dir, bool last_chunk) +{ + struct scif_endpt *ep = (struct scif_endpt *)epd; + struct scif_rma_req remote_req; + struct scif_rma_req req; + struct scif_window *local_window = NULL; + struct scif_window *remote_window = NULL; + struct scif_copy_work copy_work; + bool loopback; + int err = 0; + struct dma_chan *chan; + struct scif_mmu_notif *mmn = NULL; + bool cache = false; + struct device *spdev; + + err = scif_verify_epd(ep); + if (err) + return err; + + if (flags && !(flags & (SCIF_RMA_USECPU | SCIF_RMA_USECACHE | + SCIF_RMA_SYNC | SCIF_RMA_ORDERED))) + return -EINVAL; + + loopback = scifdev_self(ep->remote_dev) ? true : false; + copy_work.fence_type = ((flags & SCIF_RMA_SYNC) && last_chunk) ? + SCIF_DMA_POLL : 0; + copy_work.ordered = !!((flags & SCIF_RMA_ORDERED) && last_chunk); + + /* Use CPU for Mgmt node <-> Mgmt node copies */ + if (loopback && scif_is_mgmt_node()) { + flags |= SCIF_RMA_USECPU; + copy_work.fence_type = 0x0; + } + + cache = scif_is_set_reg_cache(flags); + + remote_req.out_window = &remote_window; + remote_req.offset = roffset; + remote_req.nr_bytes = len; + /* + * If transfer is from local to remote then the remote window + * must be writeable and vice versa. + */ + remote_req.prot = dir == SCIF_LOCAL_TO_REMOTE ? VM_WRITE : VM_READ; + remote_req.type = SCIF_WINDOW_PARTIAL; + remote_req.head = &ep->rma_info.remote_reg_list; + + spdev = scif_get_peer_dev(ep->remote_dev); + if (IS_ERR(spdev)) { + err = PTR_ERR(spdev); + return err; + } + + if (addr && cache) { + mutex_lock(&ep->rma_info.mmn_lock); + mmn = scif_find_mmu_notifier(current->mm, &ep->rma_info); + if (!mmn) + scif_add_mmu_notifier(current->mm, ep); + mutex_unlock(&ep->rma_info.mmn_lock); + if (IS_ERR(mmn)) { + scif_put_peer_dev(spdev); + return PTR_ERR(mmn); + } + cache = cache && !scif_rma_tc_can_cache(ep, len); + } + mutex_lock(&ep->rma_info.rma_lock); + if (addr) { + req.out_window = &local_window; + req.nr_bytes = ALIGN(len + (addr & ~PAGE_MASK), + PAGE_SIZE); + req.va_for_temp = addr & PAGE_MASK; + req.prot = (dir == SCIF_LOCAL_TO_REMOTE ? + VM_READ : VM_WRITE | VM_READ); + /* Does a valid local window exist? */ + if (mmn) { + spin_lock(&ep->rma_info.tc_lock); + req.head = &mmn->tc_reg_list; + err = scif_query_tcw(ep, &req); + spin_unlock(&ep->rma_info.tc_lock); + } + if (!mmn || err) { + err = scif_register_temp(epd, req.va_for_temp, + req.nr_bytes, req.prot, + &loffset, &local_window); + if (err) { + mutex_unlock(&ep->rma_info.rma_lock); + goto error; + } + if (!cache) + goto skip_cache; + atomic_inc(&ep->rma_info.tcw_refcount); + atomic_add_return(local_window->nr_pages, + &ep->rma_info.tcw_total_pages); + if (mmn) { + spin_lock(&ep->rma_info.tc_lock); + scif_insert_tcw(local_window, + &mmn->tc_reg_list); + spin_unlock(&ep->rma_info.tc_lock); + } + } +skip_cache: + loffset = local_window->offset + + (addr - local_window->va_for_temp); + } else { + req.out_window = &local_window; + req.offset = loffset; + /* + * If transfer is from local to remote then the self window + * must be readable and vice versa. + */ + req.prot = dir == SCIF_LOCAL_TO_REMOTE ? VM_READ : VM_WRITE; + req.nr_bytes = len; + req.type = SCIF_WINDOW_PARTIAL; + req.head = &ep->rma_info.reg_list; + /* Does a valid local window exist? */ + err = scif_query_window(&req); + if (err) { + mutex_unlock(&ep->rma_info.rma_lock); + goto error; + } + } + + /* Does a valid remote window exist? */ + err = scif_query_window(&remote_req); + if (err) { + mutex_unlock(&ep->rma_info.rma_lock); + goto error; + } + + /* + * Prepare copy_work for submitting work to the DMA kernel thread + * or CPU copy routine. + */ + copy_work.len = len; + copy_work.loopback = loopback; + copy_work.remote_dev = ep->remote_dev; + if (dir == SCIF_LOCAL_TO_REMOTE) { + copy_work.src_offset = loffset; + copy_work.src_window = local_window; + copy_work.dst_offset = roffset; + copy_work.dst_window = remote_window; + } else { + copy_work.src_offset = roffset; + copy_work.src_window = remote_window; + copy_work.dst_offset = loffset; + copy_work.dst_window = local_window; + } + + if (flags & SCIF_RMA_USECPU) { + scif_rma_list_cpu_copy(©_work); + } else { + chan = ep->rma_info.dma_chan; + err = scif_rma_list_dma_copy_wrapper(epd, ©_work, + chan, loffset); + } + if (addr && !cache) + atomic_inc(&ep->rma_info.tw_refcount); + + mutex_unlock(&ep->rma_info.rma_lock); + + if (last_chunk) { + struct scif_dev *rdev = ep->remote_dev; + + if (copy_work.fence_type == SCIF_DMA_POLL) + err = scif_drain_dma_poll(rdev->sdev, + ep->rma_info.dma_chan); + else if (copy_work.fence_type == SCIF_DMA_INTR) + err = scif_drain_dma_intr(rdev->sdev, + ep->rma_info.dma_chan); + } + + if (addr && !cache) + scif_queue_for_cleanup(local_window, &scif_info.rma); + scif_put_peer_dev(spdev); + return err; +error: + if (err) { + if (addr && local_window && !cache) + scif_destroy_window(ep, local_window); + dev_err(scif_info.mdev.this_device, + "%s %d err %d len 0x%lx\n", + __func__, __LINE__, err, len); + } + scif_put_peer_dev(spdev); + return err; +} + +int scif_readfrom(scif_epd_t epd, off_t loffset, size_t len, + off_t roffset, int flags) +{ + int err; + + dev_dbg(scif_info.mdev.this_device, + "SCIFAPI readfrom: ep %p loffset 0x%lx len 0x%lx offset 0x%lx flags 0x%x\n", + epd, loffset, len, roffset, flags); + if (scif_unaligned(loffset, roffset)) { + while (len > SCIF_MAX_UNALIGNED_BUF_SIZE) { + err = scif_rma_copy(epd, loffset, 0x0, + SCIF_MAX_UNALIGNED_BUF_SIZE, + roffset, flags, + SCIF_REMOTE_TO_LOCAL, false); + if (err) + goto readfrom_err; + loffset += SCIF_MAX_UNALIGNED_BUF_SIZE; + roffset += SCIF_MAX_UNALIGNED_BUF_SIZE; + len -= SCIF_MAX_UNALIGNED_BUF_SIZE; + } + } + err = scif_rma_copy(epd, loffset, 0x0, len, + roffset, flags, SCIF_REMOTE_TO_LOCAL, true); +readfrom_err: + return err; +} +EXPORT_SYMBOL_GPL(scif_readfrom); + +int scif_writeto(scif_epd_t epd, off_t loffset, size_t len, + off_t roffset, int flags) +{ + int err; + + dev_dbg(scif_info.mdev.this_device, + "SCIFAPI writeto: ep %p loffset 0x%lx len 0x%lx roffset 0x%lx flags 0x%x\n", + epd, loffset, len, roffset, flags); + if (scif_unaligned(loffset, roffset)) { + while (len > SCIF_MAX_UNALIGNED_BUF_SIZE) { + err = scif_rma_copy(epd, loffset, 0x0, + SCIF_MAX_UNALIGNED_BUF_SIZE, + roffset, flags, + SCIF_LOCAL_TO_REMOTE, false); + if (err) + goto writeto_err; + loffset += SCIF_MAX_UNALIGNED_BUF_SIZE; + roffset += SCIF_MAX_UNALIGNED_BUF_SIZE; + len -= SCIF_MAX_UNALIGNED_BUF_SIZE; + } + } + err = scif_rma_copy(epd, loffset, 0x0, len, + roffset, flags, SCIF_LOCAL_TO_REMOTE, true); +writeto_err: + return err; +} +EXPORT_SYMBOL_GPL(scif_writeto); + +int scif_vreadfrom(scif_epd_t epd, void *addr, size_t len, + off_t roffset, int flags) +{ + int err; + + dev_dbg(scif_info.mdev.this_device, + "SCIFAPI vreadfrom: ep %p addr %p len 0x%lx roffset 0x%lx flags 0x%x\n", + epd, addr, len, roffset, flags); + if (scif_unaligned((off_t __force)addr, roffset)) { + if (len > SCIF_MAX_UNALIGNED_BUF_SIZE) + flags &= ~SCIF_RMA_USECACHE; + + while (len > SCIF_MAX_UNALIGNED_BUF_SIZE) { + err = scif_rma_copy(epd, 0, (u64)addr, + SCIF_MAX_UNALIGNED_BUF_SIZE, + roffset, flags, + SCIF_REMOTE_TO_LOCAL, false); + if (err) + goto vreadfrom_err; + addr += SCIF_MAX_UNALIGNED_BUF_SIZE; + roffset += SCIF_MAX_UNALIGNED_BUF_SIZE; + len -= SCIF_MAX_UNALIGNED_BUF_SIZE; + } + } + err = scif_rma_copy(epd, 0, (u64)addr, len, + roffset, flags, SCIF_REMOTE_TO_LOCAL, true); +vreadfrom_err: + return err; +} +EXPORT_SYMBOL_GPL(scif_vreadfrom); + +int scif_vwriteto(scif_epd_t epd, void *addr, size_t len, + off_t roffset, int flags) +{ + int err; + + dev_dbg(scif_info.mdev.this_device, + "SCIFAPI vwriteto: ep %p addr %p len 0x%lx roffset 0x%lx flags 0x%x\n", + epd, addr, len, roffset, flags); + if (scif_unaligned((off_t __force)addr, roffset)) { + if (len > SCIF_MAX_UNALIGNED_BUF_SIZE) + flags &= ~SCIF_RMA_USECACHE; + + while (len > SCIF_MAX_UNALIGNED_BUF_SIZE) { + err = scif_rma_copy(epd, 0, (u64)addr, + SCIF_MAX_UNALIGNED_BUF_SIZE, + roffset, flags, + SCIF_LOCAL_TO_REMOTE, false); + if (err) + goto vwriteto_err; + addr += SCIF_MAX_UNALIGNED_BUF_SIZE; + roffset += SCIF_MAX_UNALIGNED_BUF_SIZE; + len -= SCIF_MAX_UNALIGNED_BUF_SIZE; + } + } + err = scif_rma_copy(epd, 0, (u64)addr, len, + roffset, flags, SCIF_LOCAL_TO_REMOTE, true); +vwriteto_err: + return err; +} +EXPORT_SYMBOL_GPL(scif_vwriteto); diff --git a/drivers/misc/mic/scif/scif_epd.c b/drivers/misc/mic/scif/scif_epd.c index b4bfbb08a8e3..00e5d6d66e7b 100644 --- a/drivers/misc/mic/scif/scif_epd.c +++ b/drivers/misc/mic/scif/scif_epd.c @@ -65,14 +65,14 @@ void scif_teardown_ep(void *endpt) void scif_add_epd_to_zombie_list(struct scif_endpt *ep, bool eplock_held) { if (!eplock_held) - spin_lock(&scif_info.eplock); + mutex_lock(&scif_info.eplock); spin_lock(&ep->lock); ep->state = SCIFEP_ZOMBIE; spin_unlock(&ep->lock); list_add_tail(&ep->list, &scif_info.zombie); scif_info.nr_zombies++; if (!eplock_held) - spin_unlock(&scif_info.eplock); + mutex_unlock(&scif_info.eplock); schedule_work(&scif_info.misc_work); } @@ -81,16 +81,15 @@ static struct scif_endpt *scif_find_listen_ep(u16 port) struct scif_endpt *ep = NULL; struct list_head *pos, *tmpq; - spin_lock(&scif_info.eplock); + mutex_lock(&scif_info.eplock); list_for_each_safe(pos, tmpq, &scif_info.listen) { ep = list_entry(pos, struct scif_endpt, list); if (ep->port.port == port) { - spin_lock(&ep->lock); - spin_unlock(&scif_info.eplock); + mutex_unlock(&scif_info.eplock); return ep; } } - spin_unlock(&scif_info.eplock); + mutex_unlock(&scif_info.eplock); return NULL; } @@ -99,14 +98,17 @@ void scif_cleanup_zombie_epd(void) struct list_head *pos, *tmpq; struct scif_endpt *ep; - spin_lock(&scif_info.eplock); + mutex_lock(&scif_info.eplock); list_for_each_safe(pos, tmpq, &scif_info.zombie) { ep = list_entry(pos, struct scif_endpt, list); - list_del(pos); - scif_info.nr_zombies--; - kfree(ep); + if (scif_rma_ep_can_uninit(ep)) { + list_del(pos); + scif_info.nr_zombies--; + put_iova_domain(&ep->rma_info.iovad); + kfree(ep); + } } - spin_unlock(&scif_info.eplock); + mutex_unlock(&scif_info.eplock); } /** @@ -137,6 +139,8 @@ void scif_cnctreq(struct scif_dev *scifdev, struct scifmsg *msg) if (!ep) /* Send reject due to no listening ports */ goto conreq_sendrej_free; + else + spin_lock(&ep->lock); if (ep->backlog <= ep->conreqcnt) { /* Send reject due to too many pending requests */ diff --git a/drivers/misc/mic/scif/scif_epd.h b/drivers/misc/mic/scif/scif_epd.h index 331322a25213..1771d7a9b8d0 100644 --- a/drivers/misc/mic/scif/scif_epd.h +++ b/drivers/misc/mic/scif/scif_epd.h @@ -96,7 +96,11 @@ struct scif_endpt_qp_info { * @conn_port: Connection port * @conn_err: Errors during connection * @conn_async_state: Async connection + * @conn_pend_wq: Used by poll while waiting for incoming connections * @conn_list: List of async connection requests + * @rma_info: Information for triggering SCIF RMA and DMA operations + * @mmu_list: link to list of MMU notifier cleanup work + * @anon: anonymous file for use in kernel mode scif poll */ struct scif_endpt { enum scif_epd_state state; @@ -125,7 +129,11 @@ struct scif_endpt { struct scif_port_id conn_port; int conn_err; int conn_async_state; + wait_queue_head_t conn_pend_wq; struct list_head conn_list; + struct scif_endpt_rma_info rma_info; + struct list_head mmu_list; + struct file *anon; }; static inline int scifdev_alive(struct scif_endpt *ep) @@ -133,6 +141,43 @@ static inline int scifdev_alive(struct scif_endpt *ep) return _scifdev_alive(ep->remote_dev); } +/* + * scif_verify_epd: + * ep: SCIF endpoint + * + * Checks several generic error conditions and returns the + * appropriate error. + */ +static inline int scif_verify_epd(struct scif_endpt *ep) +{ + if (ep->state == SCIFEP_DISCONNECTED) + return -ECONNRESET; + + if (ep->state != SCIFEP_CONNECTED) + return -ENOTCONN; + + if (!scifdev_alive(ep)) + return -ENODEV; + + return 0; +} + +static inline int scif_anon_inode_getfile(scif_epd_t epd) +{ + epd->anon = anon_inode_getfile("scif", &scif_anon_fops, NULL, 0); + if (IS_ERR(epd->anon)) + return PTR_ERR(epd->anon); + return 0; +} + +static inline void scif_anon_inode_fput(scif_epd_t epd) +{ + if (epd->anon) { + fput(epd->anon); + epd->anon = NULL; + } +} + void scif_cleanup_zombie_epd(void); void scif_teardown_ep(void *endpt); void scif_cleanup_ep_qp(struct scif_endpt *ep); @@ -157,4 +202,9 @@ void scif_clientsend(struct scif_dev *scifdev, struct scifmsg *msg); void scif_clientrcvd(struct scif_dev *scifdev, struct scifmsg *msg); int __scif_connect(scif_epd_t epd, struct scif_port_id *dst, bool non_block); int __scif_flush(scif_epd_t epd); +int scif_mmap(struct vm_area_struct *vma, scif_epd_t epd); +unsigned int __scif_pollfd(struct file *f, poll_table *wait, + struct scif_endpt *ep); +int __scif_pin_pages(void *addr, size_t len, int *out_prot, + int map_flags, scif_pinned_pages_t *pages); #endif /* SCIF_EPD_H */ diff --git a/drivers/misc/mic/scif/scif_fd.c b/drivers/misc/mic/scif/scif_fd.c index eccf7e7135f9..f7e826142a72 100644 --- a/drivers/misc/mic/scif/scif_fd.c +++ b/drivers/misc/mic/scif/scif_fd.c @@ -34,6 +34,20 @@ static int scif_fdclose(struct inode *inode, struct file *f) return scif_close(priv); } +static int scif_fdmmap(struct file *f, struct vm_area_struct *vma) +{ + struct scif_endpt *priv = f->private_data; + + return scif_mmap(vma, priv); +} + +static unsigned int scif_fdpoll(struct file *f, poll_table *wait) +{ + struct scif_endpt *priv = f->private_data; + + return __scif_pollfd(f, wait, priv); +} + static int scif_fdflush(struct file *f, fl_owner_t id) { struct scif_endpt *ep = f->private_data; @@ -140,12 +154,12 @@ static long scif_fdioctl(struct file *f, unsigned int cmd, unsigned long arg) * Add to the list of user mode eps where the second half * of the accept is not yet completed. */ - spin_lock(&scif_info.eplock); + mutex_lock(&scif_info.eplock); list_add_tail(&((*ep)->miacceptlist), &scif_info.uaccept); list_add_tail(&((*ep)->liacceptlist), &priv->li_accept); (*ep)->listenep = priv; priv->acceptcnt++; - spin_unlock(&scif_info.eplock); + mutex_unlock(&scif_info.eplock); return 0; } @@ -163,7 +177,7 @@ static long scif_fdioctl(struct file *f, unsigned int cmd, unsigned long arg) return -EFAULT; /* Remove form the user accept queue */ - spin_lock(&scif_info.eplock); + mutex_lock(&scif_info.eplock); list_for_each_safe(pos, tmpq, &scif_info.uaccept) { tmpep = list_entry(pos, struct scif_endpt, miacceptlist); @@ -175,7 +189,7 @@ static long scif_fdioctl(struct file *f, unsigned int cmd, unsigned long arg) } if (!fep) { - spin_unlock(&scif_info.eplock); + mutex_unlock(&scif_info.eplock); return -ENOENT; } @@ -190,9 +204,10 @@ static long scif_fdioctl(struct file *f, unsigned int cmd, unsigned long arg) } } - spin_unlock(&scif_info.eplock); + mutex_unlock(&scif_info.eplock); /* Free the resources automatically created from the open. */ + scif_anon_inode_fput(priv); scif_teardown_ep(priv); scif_add_epd_to_zombie_list(priv, !SCIF_EPLOCK_HELD); f->private_data = newep; @@ -290,6 +305,157 @@ getnodes_err1: getnodes_err2: return err; } + case SCIF_REG: + { + struct scif_endpt *priv = f->private_data; + struct scifioctl_reg reg; + off_t ret; + + if (copy_from_user(®, argp, sizeof(reg))) { + err = -EFAULT; + goto reg_err; + } + if (reg.flags & SCIF_MAP_KERNEL) { + err = -EINVAL; + goto reg_err; + } + ret = scif_register(priv, (void *)reg.addr, reg.len, + reg.offset, reg.prot, reg.flags); + if (ret < 0) { + err = (int)ret; + goto reg_err; + } + + if (copy_to_user(&((struct scifioctl_reg __user *)argp) + ->out_offset, &ret, sizeof(reg.out_offset))) { + err = -EFAULT; + goto reg_err; + } + err = 0; +reg_err: + scif_err_debug(err, "scif_register"); + return err; + } + case SCIF_UNREG: + { + struct scif_endpt *priv = f->private_data; + struct scifioctl_unreg unreg; + + if (copy_from_user(&unreg, argp, sizeof(unreg))) { + err = -EFAULT; + goto unreg_err; + } + err = scif_unregister(priv, unreg.offset, unreg.len); +unreg_err: + scif_err_debug(err, "scif_unregister"); + return err; + } + case SCIF_READFROM: + { + struct scif_endpt *priv = f->private_data; + struct scifioctl_copy copy; + + if (copy_from_user(©, argp, sizeof(copy))) { + err = -EFAULT; + goto readfrom_err; + } + err = scif_readfrom(priv, copy.loffset, copy.len, copy.roffset, + copy.flags); +readfrom_err: + scif_err_debug(err, "scif_readfrom"); + return err; + } + case SCIF_WRITETO: + { + struct scif_endpt *priv = f->private_data; + struct scifioctl_copy copy; + + if (copy_from_user(©, argp, sizeof(copy))) { + err = -EFAULT; + goto writeto_err; + } + err = scif_writeto(priv, copy.loffset, copy.len, copy.roffset, + copy.flags); +writeto_err: + scif_err_debug(err, "scif_writeto"); + return err; + } + case SCIF_VREADFROM: + { + struct scif_endpt *priv = f->private_data; + struct scifioctl_copy copy; + + if (copy_from_user(©, argp, sizeof(copy))) { + err = -EFAULT; + goto vreadfrom_err; + } + err = scif_vreadfrom(priv, (void __force *)copy.addr, copy.len, + copy.roffset, copy.flags); +vreadfrom_err: + scif_err_debug(err, "scif_vreadfrom"); + return err; + } + case SCIF_VWRITETO: + { + struct scif_endpt *priv = f->private_data; + struct scifioctl_copy copy; + + if (copy_from_user(©, argp, sizeof(copy))) { + err = -EFAULT; + goto vwriteto_err; + } + err = scif_vwriteto(priv, (void __force *)copy.addr, copy.len, + copy.roffset, copy.flags); +vwriteto_err: + scif_err_debug(err, "scif_vwriteto"); + return err; + } + case SCIF_FENCE_MARK: + { + struct scif_endpt *priv = f->private_data; + struct scifioctl_fence_mark mark; + int tmp_mark = 0; + + if (copy_from_user(&mark, argp, sizeof(mark))) { + err = -EFAULT; + goto fence_mark_err; + } + err = scif_fence_mark(priv, mark.flags, &tmp_mark); + if (err) + goto fence_mark_err; + if (copy_to_user((void __user *)mark.mark, &tmp_mark, + sizeof(tmp_mark))) { + err = -EFAULT; + goto fence_mark_err; + } +fence_mark_err: + scif_err_debug(err, "scif_fence_mark"); + return err; + } + case SCIF_FENCE_WAIT: + { + struct scif_endpt *priv = f->private_data; + + err = scif_fence_wait(priv, arg); + scif_err_debug(err, "scif_fence_wait"); + return err; + } + case SCIF_FENCE_SIGNAL: + { + struct scif_endpt *priv = f->private_data; + struct scifioctl_fence_signal signal; + + if (copy_from_user(&signal, argp, sizeof(signal))) { + err = -EFAULT; + goto fence_signal_err; + } + + err = scif_fence_signal(priv, signal.loff, signal.lval, + signal.roff, signal.rval, signal.flags); +fence_signal_err: + scif_err_debug(err, "scif_fence_signal"); + return err; + } } return -EINVAL; } @@ -298,6 +464,8 @@ const struct file_operations scif_fops = { .open = scif_fdopen, .release = scif_fdclose, .unlocked_ioctl = scif_fdioctl, + .mmap = scif_fdmmap, + .poll = scif_fdpoll, .flush = scif_fdflush, .owner = THIS_MODULE, }; diff --git a/drivers/misc/mic/scif/scif_fence.c b/drivers/misc/mic/scif/scif_fence.c new file mode 100644 index 000000000000..7f2c96f57066 --- /dev/null +++ b/drivers/misc/mic/scif/scif_fence.c @@ -0,0 +1,771 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * Intel SCIF driver. + * + */ + +#include "scif_main.h" + +/** + * scif_recv_mark: Handle SCIF_MARK request + * @msg: Interrupt message + * + * The peer has requested a mark. + */ +void scif_recv_mark(struct scif_dev *scifdev, struct scifmsg *msg) +{ + struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; + int mark, err; + + err = _scif_fence_mark(ep, &mark); + if (err) + msg->uop = SCIF_MARK_NACK; + else + msg->uop = SCIF_MARK_ACK; + msg->payload[0] = ep->remote_ep; + msg->payload[2] = mark; + scif_nodeqp_send(ep->remote_dev, msg); +} + +/** + * scif_recv_mark_resp: Handle SCIF_MARK_(N)ACK messages. + * @msg: Interrupt message + * + * The peer has responded to a SCIF_MARK message. + */ +void scif_recv_mark_resp(struct scif_dev *scifdev, struct scifmsg *msg) +{ + struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; + struct scif_fence_info *fence_req = + (struct scif_fence_info *)msg->payload[1]; + + mutex_lock(&ep->rma_info.rma_lock); + if (msg->uop == SCIF_MARK_ACK) { + fence_req->state = OP_COMPLETED; + fence_req->dma_mark = (int)msg->payload[2]; + } else { + fence_req->state = OP_FAILED; + } + mutex_unlock(&ep->rma_info.rma_lock); + complete(&fence_req->comp); +} + +/** + * scif_recv_wait: Handle SCIF_WAIT request + * @msg: Interrupt message + * + * The peer has requested waiting on a fence. + */ +void scif_recv_wait(struct scif_dev *scifdev, struct scifmsg *msg) +{ + struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; + struct scif_remote_fence_info *fence; + + /* + * Allocate structure for remote fence information and + * send a NACK if the allocation failed. The peer will + * return ENOMEM upon receiving a NACK. + */ + fence = kmalloc(sizeof(*fence), GFP_KERNEL); + if (!fence) { + msg->payload[0] = ep->remote_ep; + msg->uop = SCIF_WAIT_NACK; + scif_nodeqp_send(ep->remote_dev, msg); + return; + } + + /* Prepare the fence request */ + memcpy(&fence->msg, msg, sizeof(struct scifmsg)); + INIT_LIST_HEAD(&fence->list); + + /* Insert to the global remote fence request list */ + mutex_lock(&scif_info.fencelock); + atomic_inc(&ep->rma_info.fence_refcount); + list_add_tail(&fence->list, &scif_info.fence); + mutex_unlock(&scif_info.fencelock); + + schedule_work(&scif_info.misc_work); +} + +/** + * scif_recv_wait_resp: Handle SCIF_WAIT_(N)ACK messages. + * @msg: Interrupt message + * + * The peer has responded to a SCIF_WAIT message. + */ +void scif_recv_wait_resp(struct scif_dev *scifdev, struct scifmsg *msg) +{ + struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; + struct scif_fence_info *fence_req = + (struct scif_fence_info *)msg->payload[1]; + + mutex_lock(&ep->rma_info.rma_lock); + if (msg->uop == SCIF_WAIT_ACK) + fence_req->state = OP_COMPLETED; + else + fence_req->state = OP_FAILED; + mutex_unlock(&ep->rma_info.rma_lock); + complete(&fence_req->comp); +} + +/** + * scif_recv_sig_local: Handle SCIF_SIG_LOCAL request + * @msg: Interrupt message + * + * The peer has requested a signal on a local offset. + */ +void scif_recv_sig_local(struct scif_dev *scifdev, struct scifmsg *msg) +{ + struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; + int err; + + err = scif_prog_signal(ep, msg->payload[1], msg->payload[2], + SCIF_WINDOW_SELF); + if (err) + msg->uop = SCIF_SIG_NACK; + else + msg->uop = SCIF_SIG_ACK; + msg->payload[0] = ep->remote_ep; + scif_nodeqp_send(ep->remote_dev, msg); +} + +/** + * scif_recv_sig_remote: Handle SCIF_SIGNAL_REMOTE request + * @msg: Interrupt message + * + * The peer has requested a signal on a remote offset. + */ +void scif_recv_sig_remote(struct scif_dev *scifdev, struct scifmsg *msg) +{ + struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; + int err; + + err = scif_prog_signal(ep, msg->payload[1], msg->payload[2], + SCIF_WINDOW_PEER); + if (err) + msg->uop = SCIF_SIG_NACK; + else + msg->uop = SCIF_SIG_ACK; + msg->payload[0] = ep->remote_ep; + scif_nodeqp_send(ep->remote_dev, msg); +} + +/** + * scif_recv_sig_resp: Handle SCIF_SIG_(N)ACK messages. + * @msg: Interrupt message + * + * The peer has responded to a signal request. + */ +void scif_recv_sig_resp(struct scif_dev *scifdev, struct scifmsg *msg) +{ + struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; + struct scif_fence_info *fence_req = + (struct scif_fence_info *)msg->payload[3]; + + mutex_lock(&ep->rma_info.rma_lock); + if (msg->uop == SCIF_SIG_ACK) + fence_req->state = OP_COMPLETED; + else + fence_req->state = OP_FAILED; + mutex_unlock(&ep->rma_info.rma_lock); + complete(&fence_req->comp); +} + +static inline void *scif_get_local_va(off_t off, struct scif_window *window) +{ + struct page **pages = window->pinned_pages->pages; + int page_nr = (off - window->offset) >> PAGE_SHIFT; + off_t page_off = off & ~PAGE_MASK; + + return page_address(pages[page_nr]) + page_off; +} + +static void scif_prog_signal_cb(void *arg) +{ + struct scif_status *status = arg; + + dma_pool_free(status->ep->remote_dev->signal_pool, status, + status->src_dma_addr); +} + +static int _scif_prog_signal(scif_epd_t epd, dma_addr_t dst, u64 val) +{ + struct scif_endpt *ep = (struct scif_endpt *)epd; + struct dma_chan *chan = ep->rma_info.dma_chan; + struct dma_device *ddev = chan->device; + bool x100 = !is_dma_copy_aligned(chan->device, 1, 1, 1); + struct dma_async_tx_descriptor *tx; + struct scif_status *status = NULL; + dma_addr_t src; + dma_cookie_t cookie; + int err; + + tx = ddev->device_prep_dma_memcpy(chan, 0, 0, 0, DMA_PREP_FENCE); + if (!tx) { + err = -ENOMEM; + dev_err(&ep->remote_dev->sdev->dev, "%s %d err %d\n", + __func__, __LINE__, err); + goto alloc_fail; + } + cookie = tx->tx_submit(tx); + if (dma_submit_error(cookie)) { + err = (int)cookie; + dev_err(&ep->remote_dev->sdev->dev, "%s %d err %d\n", + __func__, __LINE__, err); + goto alloc_fail; + } + dma_async_issue_pending(chan); + if (x100) { + /* + * For X100 use the status descriptor to write the value to + * the destination. + */ + tx = ddev->device_prep_dma_imm_data(chan, dst, val, 0); + } else { + status = dma_pool_alloc(ep->remote_dev->signal_pool, GFP_KERNEL, + &src); + if (!status) { + err = -ENOMEM; + dev_err(&ep->remote_dev->sdev->dev, "%s %d err %d\n", + __func__, __LINE__, err); + goto alloc_fail; + } + status->val = val; + status->src_dma_addr = src; + status->ep = ep; + src += offsetof(struct scif_status, val); + tx = ddev->device_prep_dma_memcpy(chan, dst, src, sizeof(val), + DMA_PREP_INTERRUPT); + } + if (!tx) { + err = -ENOMEM; + dev_err(&ep->remote_dev->sdev->dev, "%s %d err %d\n", + __func__, __LINE__, err); + goto dma_fail; + } + if (!x100) { + tx->callback = scif_prog_signal_cb; + tx->callback_param = status; + } + cookie = tx->tx_submit(tx); + if (dma_submit_error(cookie)) { + err = -EIO; + dev_err(&ep->remote_dev->sdev->dev, "%s %d err %d\n", + __func__, __LINE__, err); + goto dma_fail; + } + dma_async_issue_pending(chan); + return 0; +dma_fail: + if (!x100) + dma_pool_free(ep->remote_dev->signal_pool, status, + status->src_dma_addr); +alloc_fail: + return err; +} + +/* + * scif_prog_signal: + * @epd - Endpoint Descriptor + * @offset - registered address to write @val to + * @val - Value to be written at @offset + * @type - Type of the window. + * + * Arrange to write a value to the registered offset after ensuring that the + * offset provided is indeed valid. + */ +int scif_prog_signal(scif_epd_t epd, off_t offset, u64 val, + enum scif_window_type type) +{ + struct scif_endpt *ep = (struct scif_endpt *)epd; + struct scif_window *window = NULL; + struct scif_rma_req req; + dma_addr_t dst_dma_addr; + int err; + + mutex_lock(&ep->rma_info.rma_lock); + req.out_window = &window; + req.offset = offset; + req.nr_bytes = sizeof(u64); + req.prot = SCIF_PROT_WRITE; + req.type = SCIF_WINDOW_SINGLE; + if (type == SCIF_WINDOW_SELF) + req.head = &ep->rma_info.reg_list; + else + req.head = &ep->rma_info.remote_reg_list; + /* Does a valid window exist? */ + err = scif_query_window(&req); + if (err) { + dev_err(scif_info.mdev.this_device, + "%s %d err %d\n", __func__, __LINE__, err); + goto unlock_ret; + } + + if (scif_is_mgmt_node() && scifdev_self(ep->remote_dev)) { + u64 *dst_virt; + + if (type == SCIF_WINDOW_SELF) + dst_virt = scif_get_local_va(offset, window); + else + dst_virt = + scif_get_local_va(offset, (struct scif_window *) + window->peer_window); + *dst_virt = val; + } else { + dst_dma_addr = __scif_off_to_dma_addr(window, offset); + err = _scif_prog_signal(epd, dst_dma_addr, val); + } +unlock_ret: + mutex_unlock(&ep->rma_info.rma_lock); + return err; +} + +static int _scif_fence_wait(scif_epd_t epd, int mark) +{ + struct scif_endpt *ep = (struct scif_endpt *)epd; + dma_cookie_t cookie = mark & ~SCIF_REMOTE_FENCE; + int err; + + /* Wait for DMA callback in scif_fence_mark_cb(..) */ + err = wait_event_interruptible_timeout(ep->rma_info.markwq, + dma_async_is_tx_complete( + ep->rma_info.dma_chan, + cookie, NULL, NULL) == + DMA_COMPLETE, + SCIF_NODE_ALIVE_TIMEOUT); + if (!err) + err = -ETIMEDOUT; + else if (err > 0) + err = 0; + return err; +} + +/** + * scif_rma_handle_remote_fences: + * + * This routine services remote fence requests. + */ +void scif_rma_handle_remote_fences(void) +{ + struct list_head *item, *tmp; + struct scif_remote_fence_info *fence; + struct scif_endpt *ep; + int mark, err; + + might_sleep(); + mutex_lock(&scif_info.fencelock); + list_for_each_safe(item, tmp, &scif_info.fence) { + fence = list_entry(item, struct scif_remote_fence_info, + list); + /* Remove fence from global list */ + list_del(&fence->list); + + /* Initiate the fence operation */ + ep = (struct scif_endpt *)fence->msg.payload[0]; + mark = fence->msg.payload[2]; + err = _scif_fence_wait(ep, mark); + if (err) + fence->msg.uop = SCIF_WAIT_NACK; + else + fence->msg.uop = SCIF_WAIT_ACK; + fence->msg.payload[0] = ep->remote_ep; + scif_nodeqp_send(ep->remote_dev, &fence->msg); + kfree(fence); + if (!atomic_sub_return(1, &ep->rma_info.fence_refcount)) + schedule_work(&scif_info.misc_work); + } + mutex_unlock(&scif_info.fencelock); +} + +static int _scif_send_fence(scif_epd_t epd, int uop, int mark, int *out_mark) +{ + int err; + struct scifmsg msg; + struct scif_fence_info *fence_req; + struct scif_endpt *ep = (struct scif_endpt *)epd; + + fence_req = kmalloc(sizeof(*fence_req), GFP_KERNEL); + if (!fence_req) { + err = -ENOMEM; + goto error; + } + + fence_req->state = OP_IN_PROGRESS; + init_completion(&fence_req->comp); + + msg.src = ep->port; + msg.uop = uop; + msg.payload[0] = ep->remote_ep; + msg.payload[1] = (u64)fence_req; + if (uop == SCIF_WAIT) + msg.payload[2] = mark; + spin_lock(&ep->lock); + if (ep->state == SCIFEP_CONNECTED) + err = scif_nodeqp_send(ep->remote_dev, &msg); + else + err = -ENOTCONN; + spin_unlock(&ep->lock); + if (err) + goto error_free; +retry: + /* Wait for a SCIF_WAIT_(N)ACK message */ + err = wait_for_completion_timeout(&fence_req->comp, + SCIF_NODE_ALIVE_TIMEOUT); + if (!err && scifdev_alive(ep)) + goto retry; + if (!err) + err = -ENODEV; + if (err > 0) + err = 0; + mutex_lock(&ep->rma_info.rma_lock); + if (err < 0) { + if (fence_req->state == OP_IN_PROGRESS) + fence_req->state = OP_FAILED; + } + if (fence_req->state == OP_FAILED && !err) + err = -ENOMEM; + if (uop == SCIF_MARK && fence_req->state == OP_COMPLETED) + *out_mark = SCIF_REMOTE_FENCE | fence_req->dma_mark; + mutex_unlock(&ep->rma_info.rma_lock); +error_free: + kfree(fence_req); +error: + return err; +} + +/** + * scif_send_fence_mark: + * @epd: end point descriptor. + * @out_mark: Output DMA mark reported by peer. + * + * Send a remote fence mark request. + */ +static int scif_send_fence_mark(scif_epd_t epd, int *out_mark) +{ + return _scif_send_fence(epd, SCIF_MARK, 0, out_mark); +} + +/** + * scif_send_fence_wait: + * @epd: end point descriptor. + * @mark: DMA mark to wait for. + * + * Send a remote fence wait request. + */ +static int scif_send_fence_wait(scif_epd_t epd, int mark) +{ + return _scif_send_fence(epd, SCIF_WAIT, mark, NULL); +} + +static int _scif_send_fence_signal_wait(struct scif_endpt *ep, + struct scif_fence_info *fence_req) +{ + int err; + +retry: + /* Wait for a SCIF_SIG_(N)ACK message */ + err = wait_for_completion_timeout(&fence_req->comp, + SCIF_NODE_ALIVE_TIMEOUT); + if (!err && scifdev_alive(ep)) + goto retry; + if (!err) + err = -ENODEV; + if (err > 0) + err = 0; + if (err < 0) { + mutex_lock(&ep->rma_info.rma_lock); + if (fence_req->state == OP_IN_PROGRESS) + fence_req->state = OP_FAILED; + mutex_unlock(&ep->rma_info.rma_lock); + } + if (fence_req->state == OP_FAILED && !err) + err = -ENXIO; + return err; +} + +/** + * scif_send_fence_signal: + * @epd - endpoint descriptor + * @loff - local offset + * @lval - local value to write to loffset + * @roff - remote offset + * @rval - remote value to write to roffset + * @flags - flags + * + * Sends a remote fence signal request + */ +static int scif_send_fence_signal(scif_epd_t epd, off_t roff, u64 rval, + off_t loff, u64 lval, int flags) +{ + int err = 0; + struct scifmsg msg; + struct scif_fence_info *fence_req; + struct scif_endpt *ep = (struct scif_endpt *)epd; + + fence_req = kmalloc(sizeof(*fence_req), GFP_KERNEL); + if (!fence_req) { + err = -ENOMEM; + goto error; + } + + fence_req->state = OP_IN_PROGRESS; + init_completion(&fence_req->comp); + msg.src = ep->port; + if (flags & SCIF_SIGNAL_LOCAL) { + msg.uop = SCIF_SIG_LOCAL; + msg.payload[0] = ep->remote_ep; + msg.payload[1] = roff; + msg.payload[2] = rval; + msg.payload[3] = (u64)fence_req; + spin_lock(&ep->lock); + if (ep->state == SCIFEP_CONNECTED) + err = scif_nodeqp_send(ep->remote_dev, &msg); + else + err = -ENOTCONN; + spin_unlock(&ep->lock); + if (err) + goto error_free; + err = _scif_send_fence_signal_wait(ep, fence_req); + if (err) + goto error_free; + } + fence_req->state = OP_IN_PROGRESS; + + if (flags & SCIF_SIGNAL_REMOTE) { + msg.uop = SCIF_SIG_REMOTE; + msg.payload[0] = ep->remote_ep; + msg.payload[1] = loff; + msg.payload[2] = lval; + msg.payload[3] = (u64)fence_req; + spin_lock(&ep->lock); + if (ep->state == SCIFEP_CONNECTED) + err = scif_nodeqp_send(ep->remote_dev, &msg); + else + err = -ENOTCONN; + spin_unlock(&ep->lock); + if (err) + goto error_free; + err = _scif_send_fence_signal_wait(ep, fence_req); + } +error_free: + kfree(fence_req); +error: + return err; +} + +static void scif_fence_mark_cb(void *arg) +{ + struct scif_endpt *ep = (struct scif_endpt *)arg; + + wake_up_interruptible(&ep->rma_info.markwq); + atomic_dec(&ep->rma_info.fence_refcount); +} + +/* + * _scif_fence_mark: + * + * @epd - endpoint descriptor + * Set up a mark for this endpoint and return the value of the mark. + */ +int _scif_fence_mark(scif_epd_t epd, int *mark) +{ + struct scif_endpt *ep = (struct scif_endpt *)epd; + struct dma_chan *chan = ep->rma_info.dma_chan; + struct dma_device *ddev = chan->device; + struct dma_async_tx_descriptor *tx; + dma_cookie_t cookie; + int err; + + tx = ddev->device_prep_dma_memcpy(chan, 0, 0, 0, DMA_PREP_FENCE); + if (!tx) { + err = -ENOMEM; + dev_err(&ep->remote_dev->sdev->dev, "%s %d err %d\n", + __func__, __LINE__, err); + return err; + } + cookie = tx->tx_submit(tx); + if (dma_submit_error(cookie)) { + err = (int)cookie; + dev_err(&ep->remote_dev->sdev->dev, "%s %d err %d\n", + __func__, __LINE__, err); + return err; + } + dma_async_issue_pending(chan); + tx = ddev->device_prep_dma_interrupt(chan, DMA_PREP_INTERRUPT); + if (!tx) { + err = -ENOMEM; + dev_err(&ep->remote_dev->sdev->dev, "%s %d err %d\n", + __func__, __LINE__, err); + return err; + } + tx->callback = scif_fence_mark_cb; + tx->callback_param = ep; + *mark = cookie = tx->tx_submit(tx); + if (dma_submit_error(cookie)) { + err = (int)cookie; + dev_err(&ep->remote_dev->sdev->dev, "%s %d err %d\n", + __func__, __LINE__, err); + return err; + } + atomic_inc(&ep->rma_info.fence_refcount); + dma_async_issue_pending(chan); + return 0; +} + +#define SCIF_LOOPB_MAGIC_MARK 0xdead + +int scif_fence_mark(scif_epd_t epd, int flags, int *mark) +{ + struct scif_endpt *ep = (struct scif_endpt *)epd; + int err = 0; + + dev_dbg(scif_info.mdev.this_device, + "SCIFAPI fence_mark: ep %p flags 0x%x mark 0x%x\n", + ep, flags, *mark); + err = scif_verify_epd(ep); + if (err) + return err; + + /* Invalid flags? */ + if (flags & ~(SCIF_FENCE_INIT_SELF | SCIF_FENCE_INIT_PEER)) + return -EINVAL; + + /* At least one of init self or peer RMA should be set */ + if (!(flags & (SCIF_FENCE_INIT_SELF | SCIF_FENCE_INIT_PEER))) + return -EINVAL; + + /* Exactly one of init self or peer RMA should be set but not both */ + if ((flags & SCIF_FENCE_INIT_SELF) && (flags & SCIF_FENCE_INIT_PEER)) + return -EINVAL; + + /* + * Management node loopback does not need to use DMA. + * Return a valid mark to be symmetric. + */ + if (scifdev_self(ep->remote_dev) && scif_is_mgmt_node()) { + *mark = SCIF_LOOPB_MAGIC_MARK; + return 0; + } + + if (flags & SCIF_FENCE_INIT_SELF) + err = _scif_fence_mark(epd, mark); + else + err = scif_send_fence_mark(ep, mark); + + if (err) + dev_err(scif_info.mdev.this_device, + "%s %d err %d\n", __func__, __LINE__, err); + dev_dbg(scif_info.mdev.this_device, + "SCIFAPI fence_mark: ep %p flags 0x%x mark 0x%x err %d\n", + ep, flags, *mark, err); + return err; +} +EXPORT_SYMBOL_GPL(scif_fence_mark); + +int scif_fence_wait(scif_epd_t epd, int mark) +{ + struct scif_endpt *ep = (struct scif_endpt *)epd; + int err = 0; + + dev_dbg(scif_info.mdev.this_device, + "SCIFAPI fence_wait: ep %p mark 0x%x\n", + ep, mark); + err = scif_verify_epd(ep); + if (err) + return err; + /* + * Management node loopback does not need to use DMA. + * The only valid mark provided is 0 so simply + * return success if the mark is valid. + */ + if (scifdev_self(ep->remote_dev) && scif_is_mgmt_node()) { + if (mark == SCIF_LOOPB_MAGIC_MARK) + return 0; + else + return -EINVAL; + } + if (mark & SCIF_REMOTE_FENCE) + err = scif_send_fence_wait(epd, mark); + else + err = _scif_fence_wait(epd, mark); + if (err < 0) + dev_err(scif_info.mdev.this_device, + "%s %d err %d\n", __func__, __LINE__, err); + return err; +} +EXPORT_SYMBOL_GPL(scif_fence_wait); + +int scif_fence_signal(scif_epd_t epd, off_t loff, u64 lval, + off_t roff, u64 rval, int flags) +{ + struct scif_endpt *ep = (struct scif_endpt *)epd; + int err = 0; + + dev_dbg(scif_info.mdev.this_device, + "SCIFAPI fence_signal: ep %p loff 0x%lx lval 0x%llx roff 0x%lx rval 0x%llx flags 0x%x\n", + ep, loff, lval, roff, rval, flags); + err = scif_verify_epd(ep); + if (err) + return err; + + /* Invalid flags? */ + if (flags & ~(SCIF_FENCE_INIT_SELF | SCIF_FENCE_INIT_PEER | + SCIF_SIGNAL_LOCAL | SCIF_SIGNAL_REMOTE)) + return -EINVAL; + + /* At least one of init self or peer RMA should be set */ + if (!(flags & (SCIF_FENCE_INIT_SELF | SCIF_FENCE_INIT_PEER))) + return -EINVAL; + + /* Exactly one of init self or peer RMA should be set but not both */ + if ((flags & SCIF_FENCE_INIT_SELF) && (flags & SCIF_FENCE_INIT_PEER)) + return -EINVAL; + + /* At least one of SCIF_SIGNAL_LOCAL or SCIF_SIGNAL_REMOTE required */ + if (!(flags & (SCIF_SIGNAL_LOCAL | SCIF_SIGNAL_REMOTE))) + return -EINVAL; + + /* Only Dword offsets allowed */ + if ((flags & SCIF_SIGNAL_LOCAL) && (loff & (sizeof(u32) - 1))) + return -EINVAL; + + /* Only Dword aligned offsets allowed */ + if ((flags & SCIF_SIGNAL_REMOTE) && (roff & (sizeof(u32) - 1))) + return -EINVAL; + + if (flags & SCIF_FENCE_INIT_PEER) { + err = scif_send_fence_signal(epd, roff, rval, loff, + lval, flags); + } else { + /* Local Signal in Local RAS */ + if (flags & SCIF_SIGNAL_LOCAL) { + err = scif_prog_signal(epd, loff, lval, + SCIF_WINDOW_SELF); + if (err) + goto error_ret; + } + + /* Signal in Remote RAS */ + if (flags & SCIF_SIGNAL_REMOTE) + err = scif_prog_signal(epd, roff, + rval, SCIF_WINDOW_PEER); + } +error_ret: + if (err) + dev_err(scif_info.mdev.this_device, + "%s %d err %d\n", __func__, __LINE__, err); + return err; +} +EXPORT_SYMBOL_GPL(scif_fence_signal); diff --git a/drivers/misc/mic/scif/scif_main.c b/drivers/misc/mic/scif/scif_main.c index 6ce851f5c7e6..36d847af1209 100644 --- a/drivers/misc/mic/scif/scif_main.c +++ b/drivers/misc/mic/scif/scif_main.c @@ -34,6 +34,7 @@ struct scif_info scif_info = { }; struct scif_dev *scif_dev; +struct kmem_cache *unaligned_cache; static atomic_t g_loopb_cnt; /* Runs in the context of intr_wq */ @@ -80,35 +81,6 @@ irqreturn_t scif_intr_handler(int irq, void *data) return IRQ_HANDLED; } -static int scif_peer_probe(struct scif_peer_dev *spdev) -{ - struct scif_dev *scifdev = &scif_dev[spdev->dnode]; - - mutex_lock(&scif_info.conflock); - scif_info.total++; - scif_info.maxid = max_t(u32, spdev->dnode, scif_info.maxid); - mutex_unlock(&scif_info.conflock); - rcu_assign_pointer(scifdev->spdev, spdev); - - /* In the future SCIF kernel client devices will be added here */ - return 0; -} - -static void scif_peer_remove(struct scif_peer_dev *spdev) -{ - struct scif_dev *scifdev = &scif_dev[spdev->dnode]; - - /* In the future SCIF kernel client devices will be removed here */ - spdev = rcu_dereference(scifdev->spdev); - if (spdev) - RCU_INIT_POINTER(scifdev->spdev, NULL); - synchronize_rcu(); - - mutex_lock(&scif_info.conflock); - scif_info.total--; - mutex_unlock(&scif_info.conflock); -} - static void scif_qp_setup_handler(struct work_struct *work) { struct scif_dev *scifdev = container_of(work, struct scif_dev, @@ -139,20 +111,13 @@ static void scif_qp_setup_handler(struct work_struct *work) } } -static int scif_setup_scifdev(struct scif_hw_dev *sdev) +static int scif_setup_scifdev(void) { + /* We support a maximum of 129 SCIF nodes including the mgmt node */ +#define MAX_SCIF_NODES 129 int i; - u8 num_nodes; - - if (sdev->snode) { - struct mic_bootparam __iomem *bp = sdev->rdp; - - num_nodes = ioread8(&bp->tot_nodes); - } else { - struct mic_bootparam *bp = sdev->dp; + u8 num_nodes = MAX_SCIF_NODES; - num_nodes = bp->tot_nodes; - } scif_dev = kcalloc(num_nodes, sizeof(*scif_dev), GFP_KERNEL); if (!scif_dev) return -ENOMEM; @@ -163,7 +128,7 @@ static int scif_setup_scifdev(struct scif_hw_dev *sdev) scifdev->exit = OP_IDLE; init_waitqueue_head(&scifdev->disconn_wq); mutex_init(&scifdev->lock); - INIT_WORK(&scifdev->init_msg_work, scif_qp_response_ack); + INIT_WORK(&scifdev->peer_add_work, scif_add_peer_device); INIT_DELAYED_WORK(&scifdev->p2p_dwork, scif_poll_qp_state); INIT_DELAYED_WORK(&scifdev->qp_dwork, @@ -181,27 +146,21 @@ static void scif_destroy_scifdev(void) static int scif_probe(struct scif_hw_dev *sdev) { - struct scif_dev *scifdev; + struct scif_dev *scifdev = &scif_dev[sdev->dnode]; int rc; dev_set_drvdata(&sdev->dev, sdev); + scifdev->sdev = sdev; + if (1 == atomic_add_return(1, &g_loopb_cnt)) { - struct scif_dev *loopb_dev; + struct scif_dev *loopb_dev = &scif_dev[sdev->snode]; - rc = scif_setup_scifdev(sdev); - if (rc) - goto exit; - scifdev = &scif_dev[sdev->dnode]; - scifdev->sdev = sdev; - loopb_dev = &scif_dev[sdev->snode]; loopb_dev->sdev = sdev; rc = scif_setup_loopback_qp(loopb_dev); if (rc) - goto free_sdev; - } else { - scifdev = &scif_dev[sdev->dnode]; - scifdev->sdev = sdev; + goto exit; } + rc = scif_setup_intr_wq(scifdev); if (rc) goto destroy_loopb; @@ -237,8 +196,6 @@ destroy_intr: destroy_loopb: if (atomic_dec_and_test(&g_loopb_cnt)) scif_destroy_loopback_qp(&scif_dev[sdev->snode]); -free_sdev: - scif_destroy_scifdev(); exit: return rc; } @@ -290,13 +247,6 @@ static void scif_remove(struct scif_hw_dev *sdev) scifdev->sdev = NULL; } -static struct scif_peer_driver scif_peer_driver = { - .driver.name = KBUILD_MODNAME, - .driver.owner = THIS_MODULE, - .probe = scif_peer_probe, - .remove = scif_peer_remove, -}; - static struct scif_hw_dev_id id_table[] = { { MIC_SCIF_DEV, SCIF_DEV_ANY_ID }, { 0 }, @@ -312,29 +262,54 @@ static struct scif_driver scif_driver = { static int _scif_init(void) { - spin_lock_init(&scif_info.eplock); + int rc; + + mutex_init(&scif_info.eplock); + spin_lock_init(&scif_info.rmalock); spin_lock_init(&scif_info.nb_connect_lock); spin_lock_init(&scif_info.port_lock); mutex_init(&scif_info.conflock); mutex_init(&scif_info.connlock); + mutex_init(&scif_info.fencelock); INIT_LIST_HEAD(&scif_info.uaccept); INIT_LIST_HEAD(&scif_info.listen); INIT_LIST_HEAD(&scif_info.zombie); INIT_LIST_HEAD(&scif_info.connected); INIT_LIST_HEAD(&scif_info.disconnected); + INIT_LIST_HEAD(&scif_info.rma); + INIT_LIST_HEAD(&scif_info.rma_tc); + INIT_LIST_HEAD(&scif_info.mmu_notif_cleanup); + INIT_LIST_HEAD(&scif_info.fence); INIT_LIST_HEAD(&scif_info.nb_connect_list); init_waitqueue_head(&scif_info.exitwq); + scif_info.rma_tc_limit = SCIF_RMA_TEMP_CACHE_LIMIT; scif_info.en_msg_log = 0; scif_info.p2p_enable = 1; + rc = scif_setup_scifdev(); + if (rc) + goto error; + unaligned_cache = kmem_cache_create("Unaligned_DMA", + SCIF_KMEM_UNALIGNED_BUF_SIZE, + 0, SLAB_HWCACHE_ALIGN, NULL); + if (!unaligned_cache) { + rc = -ENOMEM; + goto free_sdev; + } INIT_WORK(&scif_info.misc_work, scif_misc_handler); + INIT_WORK(&scif_info.mmu_notif_work, scif_mmu_notif_handler); INIT_WORK(&scif_info.conn_work, scif_conn_handler); idr_init(&scif_ports); return 0; +free_sdev: + scif_destroy_scifdev(); +error: + return rc; } static void _scif_exit(void) { idr_destroy(&scif_ports); + kmem_cache_destroy(unaligned_cache); scif_destroy_scifdev(); } @@ -344,15 +319,13 @@ static int __init scif_init(void) int rc; _scif_init(); + iova_cache_get(); rc = scif_peer_bus_init(); if (rc) goto exit; - rc = scif_peer_register_driver(&scif_peer_driver); - if (rc) - goto peer_bus_exit; rc = scif_register_driver(&scif_driver); if (rc) - goto unreg_scif_peer; + goto peer_bus_exit; rc = misc_register(mdev); if (rc) goto unreg_scif; @@ -360,8 +333,6 @@ static int __init scif_init(void) return 0; unreg_scif: scif_unregister_driver(&scif_driver); -unreg_scif_peer: - scif_peer_unregister_driver(&scif_peer_driver); peer_bus_exit: scif_peer_bus_exit(); exit: @@ -374,8 +345,8 @@ static void __exit scif_exit(void) scif_exit_debugfs(); misc_deregister(&scif_info.mdev); scif_unregister_driver(&scif_driver); - scif_peer_unregister_driver(&scif_peer_driver); scif_peer_bus_exit(); + iova_cache_put(); _scif_exit(); } diff --git a/drivers/misc/mic/scif/scif_main.h b/drivers/misc/mic/scif/scif_main.h index 580bc63e1b23..a08f0b600a9e 100644 --- a/drivers/misc/mic/scif/scif_main.h +++ b/drivers/misc/mic/scif/scif_main.h @@ -22,15 +22,18 @@ #include <linux/pci.h> #include <linux/miscdevice.h> #include <linux/dmaengine.h> +#include <linux/iova.h> +#include <linux/anon_inodes.h> #include <linux/file.h> +#include <linux/vmalloc.h> #include <linux/scif.h> - #include "../common/mic_dev.h" #define SCIF_MGMT_NODE 0 #define SCIF_DEFAULT_WATCHDOG_TO 30 #define SCIF_NODE_ACCEPT_TIMEOUT (3 * HZ) #define SCIF_NODE_ALIVE_TIMEOUT (SCIF_DEFAULT_WATCHDOG_TO * HZ) +#define SCIF_RMA_TEMP_CACHE_LIMIT 0x20000 /* * Generic state used for certain node QP message exchanges @@ -73,13 +76,21 @@ enum scif_msg_state { * @loopb_work: Used for submitting work to loopb_wq * @loopb_recv_q: List of messages received on the loopb_wq * @card_initiated_exit: set when the card has initiated the exit + * @rmalock: Synchronize access to RMA operations + * @fencelock: Synchronize access to list of remote fences requested. + * @rma: List of temporary registered windows to be destroyed. + * @rma_tc: List of temporary registered & cached Windows to be destroyed + * @fence: List of remote fence requests + * @mmu_notif_work: Work for registration caching MMU notifier workqueue + * @mmu_notif_cleanup: List of temporary cached windows for reg cache + * @rma_tc_limit: RMA temporary cache limit */ struct scif_info { u8 nodeid; u8 maxid; u8 total; u32 nr_zombies; - spinlock_t eplock; + struct mutex eplock; struct mutex connlock; spinlock_t nb_connect_lock; spinlock_t port_lock; @@ -102,6 +113,14 @@ struct scif_info { struct work_struct loopb_work; struct list_head loopb_recv_q; bool card_initiated_exit; + spinlock_t rmalock; + struct mutex fencelock; + struct list_head rma; + struct list_head rma_tc; + struct list_head fence; + struct work_struct mmu_notif_work; + struct list_head mmu_notif_cleanup; + unsigned long rma_tc_limit; }; /* @@ -139,7 +158,7 @@ struct scif_p2p_info { * @db: doorbell the peer will trigger to generate an interrupt on self * @rdb: Doorbell to trigger on the peer to generate an interrupt on the peer * @cookie: Cookie received while registering the interrupt handler - * init_msg_work: work scheduled for SCIF_INIT message processing + * @peer_add_work: Work for handling device_add for peer devices * @p2p_dwork: Delayed work to enable polling for P2P state * @qp_dwork: Delayed work for enabling polling for remote QP information * @p2p_retry: Number of times to retry polling of P2P state @@ -152,6 +171,8 @@ struct scif_p2p_info { * @disconn_rescnt: Keeps track of number of node remove requests sent * @exit: Status of exit message * @qp_dma_addr: Queue pair DMA address passed to the peer + * @dma_ch_idx: Round robin index for DMA channels + * @signal_pool: DMA pool used for scheduling scif_fence_signal DMA's */ struct scif_dev { u8 node; @@ -165,7 +186,7 @@ struct scif_dev { int db; int rdb; struct mic_irq *cookie; - struct work_struct init_msg_work; + struct work_struct peer_add_work; struct delayed_work p2p_dwork; struct delayed_work qp_dwork; int p2p_retry; @@ -178,17 +199,25 @@ struct scif_dev { atomic_t disconn_rescnt; enum scif_msg_state exit; dma_addr_t qp_dma_addr; + int dma_ch_idx; + struct dma_pool *signal_pool; }; +extern bool scif_reg_cache_enable; +extern bool scif_ulimit_check; extern struct scif_info scif_info; extern struct idr scif_ports; +extern struct bus_type scif_peer_bus; extern struct scif_dev *scif_dev; extern const struct file_operations scif_fops; +extern const struct file_operations scif_anon_fops; /* Size of the RB for the Node QP */ #define SCIF_NODE_QP_SIZE 0x10000 #include "scif_nodeqp.h" +#include "scif_rma.h" +#include "scif_rma_list.h" /* * scifdev_self: diff --git a/drivers/misc/mic/scif/scif_map.h b/drivers/misc/mic/scif/scif_map.h index 20e50b4e19b2..3e86360ba5a6 100644 --- a/drivers/misc/mic/scif/scif_map.h +++ b/drivers/misc/mic/scif/scif_map.h @@ -80,7 +80,7 @@ scif_unmap_single(dma_addr_t local, struct scif_dev *scifdev, size_t size) { if (!scifdev_self(scifdev)) { - if (scifdev_is_p2p(scifdev) && local > scifdev->base_addr) + if (scifdev_is_p2p(scifdev)) local = local - scifdev->base_addr; dma_unmap_single(&scifdev->sdev->dev, local, size, DMA_BIDIRECTIONAL); @@ -110,4 +110,27 @@ scif_iounmap(void *virt, size_t len, struct scif_dev *scifdev) sdev->hw_ops->iounmap(sdev, (void __force __iomem *)virt); } } + +static __always_inline int +scif_map_page(dma_addr_t *dma_handle, struct page *page, + struct scif_dev *scifdev) +{ + int err = 0; + + if (scifdev_self(scifdev)) { + *dma_handle = page_to_phys(page); + } else { + struct scif_hw_dev *sdev = scifdev->sdev; + *dma_handle = dma_map_page(&sdev->dev, + page, 0x0, PAGE_SIZE, + DMA_BIDIRECTIONAL); + if (dma_mapping_error(&sdev->dev, *dma_handle)) + err = -ENOMEM; + else if (scifdev_is_p2p(scifdev)) + *dma_handle = *dma_handle + scifdev->base_addr; + } + if (err) + *dma_handle = 0; + return err; +} #endif /* SCIF_MAP_H */ diff --git a/drivers/misc/mic/scif/scif_mmap.c b/drivers/misc/mic/scif/scif_mmap.c new file mode 100644 index 000000000000..49cb8f7b4672 --- /dev/null +++ b/drivers/misc/mic/scif/scif_mmap.c @@ -0,0 +1,699 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * Intel SCIF driver. + * + */ +#include "scif_main.h" + +/* + * struct scif_vma_info - Information about a remote memory mapping + * created via scif_mmap(..) + * @vma: VM area struct + * @list: link to list of active vmas + */ +struct scif_vma_info { + struct vm_area_struct *vma; + struct list_head list; +}; + +void scif_recv_munmap(struct scif_dev *scifdev, struct scifmsg *msg) +{ + struct scif_rma_req req; + struct scif_window *window = NULL; + struct scif_window *recv_window = + (struct scif_window *)msg->payload[0]; + struct scif_endpt *ep; + + ep = (struct scif_endpt *)recv_window->ep; + req.out_window = &window; + req.offset = recv_window->offset; + req.prot = recv_window->prot; + req.nr_bytes = recv_window->nr_pages << PAGE_SHIFT; + req.type = SCIF_WINDOW_FULL; + req.head = &ep->rma_info.reg_list; + msg->payload[0] = ep->remote_ep; + + mutex_lock(&ep->rma_info.rma_lock); + /* Does a valid window exist? */ + if (scif_query_window(&req)) { + dev_err(&scifdev->sdev->dev, + "%s %d -ENXIO\n", __func__, __LINE__); + msg->uop = SCIF_UNREGISTER_ACK; + goto error; + } + + scif_put_window(window, window->nr_pages); + + if (!window->ref_count) { + atomic_inc(&ep->rma_info.tw_refcount); + ep->rma_info.async_list_del = 1; + list_del_init(&window->list); + scif_free_window_offset(ep, window, window->offset); + } +error: + mutex_unlock(&ep->rma_info.rma_lock); + if (window && !window->ref_count) + scif_queue_for_cleanup(window, &scif_info.rma); +} + +/* + * Remove valid remote memory mappings created via scif_mmap(..) from the + * process address space since the remote node is lost + */ +static void __scif_zap_mmaps(struct scif_endpt *ep) +{ + struct list_head *item; + struct scif_vma_info *info; + struct vm_area_struct *vma; + unsigned long size; + + spin_lock(&ep->lock); + list_for_each(item, &ep->rma_info.vma_list) { + info = list_entry(item, struct scif_vma_info, list); + vma = info->vma; + size = vma->vm_end - vma->vm_start; + zap_vma_ptes(vma, vma->vm_start, size); + dev_dbg(scif_info.mdev.this_device, + "%s ep %p zap vma %p size 0x%lx\n", + __func__, ep, info->vma, size); + } + spin_unlock(&ep->lock); +} + +/* + * Traverse the list of endpoints for a particular remote node and + * zap valid remote memory mappings since the remote node is lost + */ +static void _scif_zap_mmaps(int node, struct list_head *head) +{ + struct scif_endpt *ep; + struct list_head *item; + + mutex_lock(&scif_info.connlock); + list_for_each(item, head) { + ep = list_entry(item, struct scif_endpt, list); + if (ep->remote_dev->node == node) + __scif_zap_mmaps(ep); + } + mutex_unlock(&scif_info.connlock); +} + +/* + * Wrapper for removing remote memory mappings for a particular node. This API + * is called by peer nodes as part of handling a lost node. + */ +void scif_zap_mmaps(int node) +{ + _scif_zap_mmaps(node, &scif_info.connected); + _scif_zap_mmaps(node, &scif_info.disconnected); +} + +/* + * This API is only called while handling a lost node: + * a) Remote node is dead. + * b) Remote memory mappings have been zapped + * So we can traverse the remote_reg_list without any locks. Since + * the window has not yet been unregistered we can drop the ref count + * and queue it to the cleanup thread. + */ +static void __scif_cleanup_rma_for_zombies(struct scif_endpt *ep) +{ + struct list_head *pos, *tmp; + struct scif_window *window; + + list_for_each_safe(pos, tmp, &ep->rma_info.remote_reg_list) { + window = list_entry(pos, struct scif_window, list); + if (window->ref_count) + scif_put_window(window, window->nr_pages); + else + dev_err(scif_info.mdev.this_device, + "%s %d unexpected\n", + __func__, __LINE__); + if (!window->ref_count) { + atomic_inc(&ep->rma_info.tw_refcount); + list_del_init(&window->list); + scif_queue_for_cleanup(window, &scif_info.rma); + } + } +} + +/* Cleanup remote registration lists for zombie endpoints */ +void scif_cleanup_rma_for_zombies(int node) +{ + struct scif_endpt *ep; + struct list_head *item; + + mutex_lock(&scif_info.eplock); + list_for_each(item, &scif_info.zombie) { + ep = list_entry(item, struct scif_endpt, list); + if (ep->remote_dev && ep->remote_dev->node == node) + __scif_cleanup_rma_for_zombies(ep); + } + mutex_unlock(&scif_info.eplock); + flush_work(&scif_info.misc_work); +} + +/* Insert the VMA into the per endpoint VMA list */ +static int scif_insert_vma(struct scif_endpt *ep, struct vm_area_struct *vma) +{ + struct scif_vma_info *info; + int err = 0; + + info = kzalloc(sizeof(*info), GFP_KERNEL); + if (!info) { + err = -ENOMEM; + goto done; + } + info->vma = vma; + spin_lock(&ep->lock); + list_add_tail(&info->list, &ep->rma_info.vma_list); + spin_unlock(&ep->lock); +done: + return err; +} + +/* Delete the VMA from the per endpoint VMA list */ +static void scif_delete_vma(struct scif_endpt *ep, struct vm_area_struct *vma) +{ + struct list_head *item; + struct scif_vma_info *info; + + spin_lock(&ep->lock); + list_for_each(item, &ep->rma_info.vma_list) { + info = list_entry(item, struct scif_vma_info, list); + if (info->vma == vma) { + list_del(&info->list); + kfree(info); + break; + } + } + spin_unlock(&ep->lock); +} + +static phys_addr_t scif_get_phys(phys_addr_t phys, struct scif_endpt *ep) +{ + struct scif_dev *scifdev = (struct scif_dev *)ep->remote_dev; + struct scif_hw_dev *sdev = scifdev->sdev; + phys_addr_t out_phys, apt_base = 0; + + /* + * If the DMA address is card relative then we need to add the + * aperture base for mmap to work correctly + */ + if (!scifdev_self(scifdev) && sdev->aper && sdev->card_rel_da) + apt_base = sdev->aper->pa; + out_phys = apt_base + phys; + return out_phys; +} + +int scif_get_pages(scif_epd_t epd, off_t offset, size_t len, + struct scif_range **pages) +{ + struct scif_endpt *ep = (struct scif_endpt *)epd; + struct scif_rma_req req; + struct scif_window *window = NULL; + int nr_pages, err, i; + + dev_dbg(scif_info.mdev.this_device, + "SCIFAPI get_pinned_pages: ep %p offset 0x%lx len 0x%lx\n", + ep, offset, len); + err = scif_verify_epd(ep); + if (err) + return err; + + if (!len || (offset < 0) || + (offset + len < offset) || + (ALIGN(offset, PAGE_SIZE) != offset) || + (ALIGN(len, PAGE_SIZE) != len)) + return -EINVAL; + + nr_pages = len >> PAGE_SHIFT; + + req.out_window = &window; + req.offset = offset; + req.prot = 0; + req.nr_bytes = len; + req.type = SCIF_WINDOW_SINGLE; + req.head = &ep->rma_info.remote_reg_list; + + mutex_lock(&ep->rma_info.rma_lock); + /* Does a valid window exist? */ + err = scif_query_window(&req); + if (err) { + dev_err(&ep->remote_dev->sdev->dev, + "%s %d err %d\n", __func__, __LINE__, err); + goto error; + } + + /* Allocate scif_range */ + *pages = kzalloc(sizeof(**pages), GFP_KERNEL); + if (!*pages) { + err = -ENOMEM; + goto error; + } + + /* Allocate phys addr array */ + (*pages)->phys_addr = scif_zalloc(nr_pages * sizeof(dma_addr_t)); + if (!((*pages)->phys_addr)) { + err = -ENOMEM; + goto error; + } + + if (scif_is_mgmt_node() && !scifdev_self(ep->remote_dev)) { + /* Allocate virtual address array */ + ((*pages)->va = scif_zalloc(nr_pages * sizeof(void *))); + if (!(*pages)->va) { + err = -ENOMEM; + goto error; + } + } + /* Populate the values */ + (*pages)->cookie = window; + (*pages)->nr_pages = nr_pages; + (*pages)->prot_flags = window->prot; + + for (i = 0; i < nr_pages; i++) { + (*pages)->phys_addr[i] = + __scif_off_to_dma_addr(window, offset + + (i * PAGE_SIZE)); + (*pages)->phys_addr[i] = scif_get_phys((*pages)->phys_addr[i], + ep); + if (scif_is_mgmt_node() && !scifdev_self(ep->remote_dev)) + (*pages)->va[i] = + ep->remote_dev->sdev->aper->va + + (*pages)->phys_addr[i] - + ep->remote_dev->sdev->aper->pa; + } + + scif_get_window(window, nr_pages); +error: + mutex_unlock(&ep->rma_info.rma_lock); + if (err) { + if (*pages) { + scif_free((*pages)->phys_addr, + nr_pages * sizeof(dma_addr_t)); + scif_free((*pages)->va, + nr_pages * sizeof(void *)); + kfree(*pages); + *pages = NULL; + } + dev_err(&ep->remote_dev->sdev->dev, + "%s %d err %d\n", __func__, __LINE__, err); + } + return err; +} +EXPORT_SYMBOL_GPL(scif_get_pages); + +int scif_put_pages(struct scif_range *pages) +{ + struct scif_endpt *ep; + struct scif_window *window; + struct scifmsg msg; + + if (!pages || !pages->cookie) + return -EINVAL; + + window = pages->cookie; + + if (!window || window->magic != SCIFEP_MAGIC) + return -EINVAL; + + ep = (struct scif_endpt *)window->ep; + /* + * If the state is SCIFEP_CONNECTED or SCIFEP_DISCONNECTED then the + * callee should be allowed to release references to the pages, + * else the endpoint was not connected in the first place, + * hence the ENOTCONN. + */ + if (ep->state != SCIFEP_CONNECTED && ep->state != SCIFEP_DISCONNECTED) + return -ENOTCONN; + + mutex_lock(&ep->rma_info.rma_lock); + + scif_put_window(window, pages->nr_pages); + + /* Initiate window destruction if ref count is zero */ + if (!window->ref_count) { + list_del(&window->list); + mutex_unlock(&ep->rma_info.rma_lock); + scif_drain_dma_intr(ep->remote_dev->sdev, + ep->rma_info.dma_chan); + /* Inform the peer about this window being destroyed. */ + msg.uop = SCIF_MUNMAP; + msg.src = ep->port; + msg.payload[0] = window->peer_window; + /* No error handling for notification messages */ + scif_nodeqp_send(ep->remote_dev, &msg); + /* Destroy this window from the peer's registered AS */ + scif_destroy_remote_window(window); + } else { + mutex_unlock(&ep->rma_info.rma_lock); + } + + scif_free(pages->phys_addr, pages->nr_pages * sizeof(dma_addr_t)); + scif_free(pages->va, pages->nr_pages * sizeof(void *)); + kfree(pages); + return 0; +} +EXPORT_SYMBOL_GPL(scif_put_pages); + +/* + * scif_rma_list_mmap: + * + * Traverse the remote registration list starting from start_window: + * 1) Create VtoP mappings via remap_pfn_range(..) + * 2) Once step 1) and 2) complete successfully then traverse the range of + * windows again and bump the reference count. + * RMA lock must be held. + */ +static int scif_rma_list_mmap(struct scif_window *start_window, s64 offset, + int nr_pages, struct vm_area_struct *vma) +{ + s64 end_offset, loop_offset = offset; + struct scif_window *window = start_window; + int loop_nr_pages, nr_pages_left = nr_pages; + struct scif_endpt *ep = (struct scif_endpt *)start_window->ep; + struct list_head *head = &ep->rma_info.remote_reg_list; + int i, err = 0; + dma_addr_t phys_addr; + struct scif_window_iter src_win_iter; + size_t contig_bytes = 0; + + might_sleep(); + list_for_each_entry_from(window, head, list) { + end_offset = window->offset + + (window->nr_pages << PAGE_SHIFT); + loop_nr_pages = min_t(int, + (end_offset - loop_offset) >> PAGE_SHIFT, + nr_pages_left); + scif_init_window_iter(window, &src_win_iter); + for (i = 0; i < loop_nr_pages; i++) { + phys_addr = scif_off_to_dma_addr(window, loop_offset, + &contig_bytes, + &src_win_iter); + phys_addr = scif_get_phys(phys_addr, ep); + err = remap_pfn_range(vma, + vma->vm_start + + loop_offset - offset, + phys_addr >> PAGE_SHIFT, + PAGE_SIZE, + vma->vm_page_prot); + if (err) + goto error; + loop_offset += PAGE_SIZE; + } + nr_pages_left -= loop_nr_pages; + if (!nr_pages_left) + break; + } + /* + * No more failures expected. Bump up the ref count for all + * the windows. Another traversal from start_window required + * for handling errors encountered across windows during + * remap_pfn_range(..). + */ + loop_offset = offset; + nr_pages_left = nr_pages; + window = start_window; + head = &ep->rma_info.remote_reg_list; + list_for_each_entry_from(window, head, list) { + end_offset = window->offset + + (window->nr_pages << PAGE_SHIFT); + loop_nr_pages = min_t(int, + (end_offset - loop_offset) >> PAGE_SHIFT, + nr_pages_left); + scif_get_window(window, loop_nr_pages); + nr_pages_left -= loop_nr_pages; + loop_offset += (loop_nr_pages << PAGE_SHIFT); + if (!nr_pages_left) + break; + } +error: + if (err) + dev_err(scif_info.mdev.this_device, + "%s %d err %d\n", __func__, __LINE__, err); + return err; +} + +/* + * scif_rma_list_munmap: + * + * Traverse the remote registration list starting from window: + * 1) Decrement ref count. + * 2) If the ref count drops to zero then send a SCIF_MUNMAP message to peer. + * RMA lock must be held. + */ +static void scif_rma_list_munmap(struct scif_window *start_window, + s64 offset, int nr_pages) +{ + struct scifmsg msg; + s64 loop_offset = offset, end_offset; + int loop_nr_pages, nr_pages_left = nr_pages; + struct scif_endpt *ep = (struct scif_endpt *)start_window->ep; + struct list_head *head = &ep->rma_info.remote_reg_list; + struct scif_window *window = start_window, *_window; + + msg.uop = SCIF_MUNMAP; + msg.src = ep->port; + loop_offset = offset; + nr_pages_left = nr_pages; + list_for_each_entry_safe_from(window, _window, head, list) { + end_offset = window->offset + + (window->nr_pages << PAGE_SHIFT); + loop_nr_pages = min_t(int, + (end_offset - loop_offset) >> PAGE_SHIFT, + nr_pages_left); + scif_put_window(window, loop_nr_pages); + if (!window->ref_count) { + struct scif_dev *rdev = ep->remote_dev; + + scif_drain_dma_intr(rdev->sdev, + ep->rma_info.dma_chan); + /* Inform the peer about this munmap */ + msg.payload[0] = window->peer_window; + /* No error handling for Notification messages. */ + scif_nodeqp_send(ep->remote_dev, &msg); + list_del(&window->list); + /* Destroy this window from the peer's registered AS */ + scif_destroy_remote_window(window); + } + nr_pages_left -= loop_nr_pages; + loop_offset += (loop_nr_pages << PAGE_SHIFT); + if (!nr_pages_left) + break; + } +} + +/* + * The private data field of each VMA used to mmap a remote window + * points to an instance of struct vma_pvt + */ +struct vma_pvt { + struct scif_endpt *ep; /* End point for remote window */ + s64 offset; /* offset within remote window */ + bool valid_offset; /* offset is valid only if the original + * mmap request was for a single page + * else the offset within the vma is + * the correct offset + */ + struct kref ref; +}; + +static void vma_pvt_release(struct kref *ref) +{ + struct vma_pvt *vmapvt = container_of(ref, struct vma_pvt, ref); + + kfree(vmapvt); +} + +/** + * scif_vma_open - VMA open driver callback + * @vma: VMM memory area. + * The open method is called by the kernel to allow the subsystem implementing + * the VMA to initialize the area. This method is invoked any time a new + * reference to the VMA is made (when a process forks, for example). + * The one exception happens when the VMA is first created by mmap; + * in this case, the driver's mmap method is called instead. + * This function is also invoked when an existing VMA is split by the kernel + * due to a call to munmap on a subset of the VMA resulting in two VMAs. + * The kernel invokes this function only on one of the two VMAs. + */ +static void scif_vma_open(struct vm_area_struct *vma) +{ + struct vma_pvt *vmapvt = vma->vm_private_data; + + dev_dbg(scif_info.mdev.this_device, + "SCIFAPI vma open: vma_start 0x%lx vma_end 0x%lx\n", + vma->vm_start, vma->vm_end); + scif_insert_vma(vmapvt->ep, vma); + kref_get(&vmapvt->ref); +} + +/** + * scif_munmap - VMA close driver callback. + * @vma: VMM memory area. + * When an area is destroyed, the kernel calls its close operation. + * Note that there's no usage count associated with VMA's; the area + * is opened and closed exactly once by each process that uses it. + */ +static void scif_munmap(struct vm_area_struct *vma) +{ + struct scif_endpt *ep; + struct vma_pvt *vmapvt = vma->vm_private_data; + int nr_pages = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT; + s64 offset; + struct scif_rma_req req; + struct scif_window *window = NULL; + int err; + + might_sleep(); + dev_dbg(scif_info.mdev.this_device, + "SCIFAPI munmap: vma_start 0x%lx vma_end 0x%lx\n", + vma->vm_start, vma->vm_end); + ep = vmapvt->ep; + offset = vmapvt->valid_offset ? vmapvt->offset : + (vma->vm_pgoff) << PAGE_SHIFT; + dev_dbg(scif_info.mdev.this_device, + "SCIFAPI munmap: ep %p nr_pages 0x%x offset 0x%llx\n", + ep, nr_pages, offset); + req.out_window = &window; + req.offset = offset; + req.nr_bytes = vma->vm_end - vma->vm_start; + req.prot = vma->vm_flags & (VM_READ | VM_WRITE); + req.type = SCIF_WINDOW_PARTIAL; + req.head = &ep->rma_info.remote_reg_list; + + mutex_lock(&ep->rma_info.rma_lock); + + err = scif_query_window(&req); + if (err) + dev_err(scif_info.mdev.this_device, + "%s %d err %d\n", __func__, __LINE__, err); + else + scif_rma_list_munmap(window, offset, nr_pages); + + mutex_unlock(&ep->rma_info.rma_lock); + /* + * The kernel probably zeroes these out but we still want + * to clean up our own mess just in case. + */ + vma->vm_ops = NULL; + vma->vm_private_data = NULL; + kref_put(&vmapvt->ref, vma_pvt_release); + scif_delete_vma(ep, vma); +} + +static const struct vm_operations_struct scif_vm_ops = { + .open = scif_vma_open, + .close = scif_munmap, +}; + +/** + * scif_mmap - Map pages in virtual address space to a remote window. + * @vma: VMM memory area. + * @epd: endpoint descriptor + * + * Return: Upon successful completion, scif_mmap() returns zero + * else an apt error is returned as documented in scif.h + */ +int scif_mmap(struct vm_area_struct *vma, scif_epd_t epd) +{ + struct scif_rma_req req; + struct scif_window *window = NULL; + struct scif_endpt *ep = (struct scif_endpt *)epd; + s64 start_offset = vma->vm_pgoff << PAGE_SHIFT; + int nr_pages = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT; + int err; + struct vma_pvt *vmapvt; + + dev_dbg(scif_info.mdev.this_device, + "SCIFAPI mmap: ep %p start_offset 0x%llx nr_pages 0x%x\n", + ep, start_offset, nr_pages); + err = scif_verify_epd(ep); + if (err) + return err; + + might_sleep(); + + err = scif_insert_vma(ep, vma); + if (err) + return err; + + vmapvt = kzalloc(sizeof(*vmapvt), GFP_KERNEL); + if (!vmapvt) { + scif_delete_vma(ep, vma); + return -ENOMEM; + } + + vmapvt->ep = ep; + kref_init(&vmapvt->ref); + + req.out_window = &window; + req.offset = start_offset; + req.nr_bytes = vma->vm_end - vma->vm_start; + req.prot = vma->vm_flags & (VM_READ | VM_WRITE); + req.type = SCIF_WINDOW_PARTIAL; + req.head = &ep->rma_info.remote_reg_list; + + mutex_lock(&ep->rma_info.rma_lock); + /* Does a valid window exist? */ + err = scif_query_window(&req); + if (err) { + dev_err(&ep->remote_dev->sdev->dev, + "%s %d err %d\n", __func__, __LINE__, err); + goto error_unlock; + } + + /* Default prot for loopback */ + if (!scifdev_self(ep->remote_dev)) + vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot); + + /* + * VM_DONTCOPY - Do not copy this vma on fork + * VM_DONTEXPAND - Cannot expand with mremap() + * VM_RESERVED - Count as reserved_vm like IO + * VM_PFNMAP - Page-ranges managed without "struct page" + * VM_IO - Memory mapped I/O or similar + * + * We do not want to copy this VMA automatically on a fork(), + * expand this VMA due to mremap() or swap out these pages since + * the VMA is actually backed by physical pages in the remote + * node's physical memory and not via a struct page. + */ + vma->vm_flags |= VM_DONTCOPY | VM_DONTEXPAND | VM_DONTDUMP; + + if (!scifdev_self(ep->remote_dev)) + vma->vm_flags |= VM_IO | VM_PFNMAP; + + /* Map this range of windows */ + err = scif_rma_list_mmap(window, start_offset, nr_pages, vma); + if (err) { + dev_err(&ep->remote_dev->sdev->dev, + "%s %d err %d\n", __func__, __LINE__, err); + goto error_unlock; + } + /* Set up the driver call back */ + vma->vm_ops = &scif_vm_ops; + vma->vm_private_data = vmapvt; +error_unlock: + mutex_unlock(&ep->rma_info.rma_lock); + if (err) { + kfree(vmapvt); + dev_err(&ep->remote_dev->sdev->dev, + "%s %d err %d\n", __func__, __LINE__, err); + scif_delete_vma(ep, vma); + } + return err; +} diff --git a/drivers/misc/mic/scif/scif_nm.c b/drivers/misc/mic/scif/scif_nm.c index 9b4c5382d6a7..79f26a02a1cb 100644 --- a/drivers/misc/mic/scif/scif_nm.c +++ b/drivers/misc/mic/scif/scif_nm.c @@ -34,6 +34,7 @@ static void scif_invalidate_ep(int node) list_for_each_safe(pos, tmpq, &scif_info.disconnected) { ep = list_entry(pos, struct scif_endpt, list); if (ep->remote_dev->node == node) { + scif_unmap_all_windows(ep); spin_lock(&ep->lock); scif_cleanup_ep_qp(ep); spin_unlock(&ep->lock); @@ -50,6 +51,7 @@ static void scif_invalidate_ep(int node) wake_up_interruptible(&ep->sendwq); wake_up_interruptible(&ep->recvwq); spin_unlock(&ep->lock); + scif_unmap_all_windows(ep); } } mutex_unlock(&scif_info.connlock); @@ -61,8 +63,8 @@ void scif_free_qp(struct scif_dev *scifdev) if (!qp) return; - scif_free_coherent((void *)qp->inbound_q.rb_base, - qp->local_buf, scifdev, qp->inbound_q.size); + scif_unmap_single(qp->local_buf, scifdev, qp->inbound_q.size); + kfree(qp->inbound_q.rb_base); scif_unmap_single(qp->local_qp, scifdev, sizeof(struct scif_qp)); kfree(scifdev->qpairs); scifdev->qpairs = NULL; @@ -125,8 +127,12 @@ void scif_cleanup_scifdev(struct scif_dev *dev) } scif_destroy_intr_wq(dev); } + flush_work(&scif_info.misc_work); scif_destroy_p2p(dev); scif_invalidate_ep(dev->node); + scif_zap_mmaps(dev->node); + scif_cleanup_rma_for_zombies(dev->node); + flush_work(&scif_info.misc_work); scif_send_acks(dev); if (!dev->node && scif_info.card_initiated_exit) { /* @@ -147,14 +153,8 @@ void scif_cleanup_scifdev(struct scif_dev *dev) void scif_handle_remove_node(int node) { struct scif_dev *scifdev = &scif_dev[node]; - struct scif_peer_dev *spdev; - - rcu_read_lock(); - spdev = rcu_dereference(scifdev->spdev); - rcu_read_unlock(); - if (spdev) - scif_peer_unregister_device(spdev); - else + + if (scif_peer_unregister_device(scifdev)) scif_send_acks(scifdev); } diff --git a/drivers/misc/mic/scif/scif_nodeqp.c b/drivers/misc/mic/scif/scif_nodeqp.c index 6dfdae3452d6..7180d566c74b 100644 --- a/drivers/misc/mic/scif/scif_nodeqp.c +++ b/drivers/misc/mic/scif/scif_nodeqp.c @@ -105,18 +105,22 @@ int scif_setup_qp_connect(struct scif_qp *qp, dma_addr_t *qp_offset, int local_size, struct scif_dev *scifdev) { - void *local_q = NULL; + void *local_q = qp->inbound_q.rb_base; int err = 0; u32 tmp_rd = 0; spin_lock_init(&qp->send_lock); spin_lock_init(&qp->recv_lock); - local_q = kzalloc(local_size, GFP_KERNEL); + /* Allocate rb only if not already allocated */ if (!local_q) { - err = -ENOMEM; - return err; + local_q = kzalloc(local_size, GFP_KERNEL); + if (!local_q) { + err = -ENOMEM; + return err; + } } + err = scif_map_single(&qp->local_buf, local_q, scifdev, local_size); if (err) goto kfree; @@ -260,6 +264,11 @@ int scif_setup_qp_connect_response(struct scif_dev *scifdev, r_buf, get_count_order(remote_size)); /* + * Because the node QP may already be processing an INIT message, set + * the read pointer so the cached read offset isn't lost + */ + qp->remote_qp->local_read = qp->inbound_q.current_read_offset; + /* * resetup the inbound_q now that we know where the * inbound_read really is. */ @@ -529,27 +538,6 @@ static void scif_p2p_setup(void) } } -void scif_qp_response_ack(struct work_struct *work) -{ - struct scif_dev *scifdev = container_of(work, struct scif_dev, - init_msg_work); - struct scif_peer_dev *spdev; - - /* Drop the INIT message if it has already been received */ - if (_scifdev_alive(scifdev)) - return; - - spdev = scif_peer_register_device(scifdev); - if (IS_ERR(spdev)) - return; - - if (scif_is_mgmt_node()) { - mutex_lock(&scif_info.conflock); - scif_p2p_setup(); - mutex_unlock(&scif_info.conflock); - } -} - static char *message_types[] = {"BAD", "INIT", "EXIT", @@ -568,7 +556,29 @@ static char *message_types[] = {"BAD", "DISCNT_ACK", "CLIENT_SENT", "CLIENT_RCVD", - "SCIF_GET_NODE_INFO"}; + "SCIF_GET_NODE_INFO", + "REGISTER", + "REGISTER_ACK", + "REGISTER_NACK", + "UNREGISTER", + "UNREGISTER_ACK", + "UNREGISTER_NACK", + "ALLOC_REQ", + "ALLOC_GNT", + "ALLOC_REJ", + "FREE_PHYS", + "FREE_VIRT", + "MUNMAP", + "MARK", + "MARK_ACK", + "MARK_NACK", + "WAIT", + "WAIT_ACK", + "WAIT_NACK", + "SIGNAL_LOCAL", + "SIGNAL_REMOTE", + "SIG_ACK", + "SIG_NACK"}; static void scif_display_message(struct scif_dev *scifdev, struct scifmsg *msg, @@ -662,10 +672,16 @@ int scif_nodeqp_send(struct scif_dev *scifdev, struct scifmsg *msg) * * Work queue handler for servicing miscellaneous SCIF tasks. * Examples include: - * 1) Cleanup of zombie endpoints. + * 1) Remote fence requests. + * 2) Destruction of temporary registered windows + * created during scif_vreadfrom()/scif_vwriteto(). + * 3) Cleanup of zombie endpoints. */ void scif_misc_handler(struct work_struct *work) { + scif_rma_handle_remote_fences(); + scif_rma_destroy_windows(); + scif_rma_destroy_tcw_invalid(); scif_cleanup_zombie_epd(); } @@ -682,13 +698,14 @@ scif_init(struct scif_dev *scifdev, struct scifmsg *msg) * address to complete initializing the inbound_q. */ flush_delayed_work(&scifdev->qp_dwork); - /* - * Delegate the peer device registration to a workqueue, otherwise if - * SCIF client probe (called during peer device registration) calls - * scif_connect(..), it will block the message processing thread causing - * a deadlock. - */ - schedule_work(&scifdev->init_msg_work); + + scif_peer_register_device(scifdev); + + if (scif_is_mgmt_node()) { + mutex_lock(&scif_info.conflock); + scif_p2p_setup(); + mutex_unlock(&scif_info.conflock); + } } /** @@ -838,13 +855,13 @@ void scif_poll_qp_state(struct work_struct *work) msecs_to_jiffies(SCIF_NODE_QP_TIMEOUT)); return; } - scif_peer_register_device(peerdev); return; timeout: dev_err(&peerdev->sdev->dev, "%s %d remote node %d offline, state = 0x%x\n", __func__, __LINE__, peerdev->node, qp->qp_state); qp->remote_qp->qp_state = SCIF_QP_OFFLINE; + scif_peer_unregister_device(peerdev); scif_cleanup_scifdev(peerdev); } @@ -894,6 +911,9 @@ scif_node_add_ack(struct scif_dev *scifdev, struct scifmsg *msg) goto local_error; peerdev->rdb = msg->payload[2]; qp->remote_qp->qp_state = SCIF_QP_ONLINE; + + scif_peer_register_device(peerdev); + schedule_delayed_work(&peerdev->p2p_dwork, 0); return; local_error: @@ -1007,6 +1027,27 @@ static void (*scif_intr_func[SCIF_MAX_MSG + 1]) scif_clientsend, /* SCIF_CLIENT_SENT */ scif_clientrcvd, /* SCIF_CLIENT_RCVD */ scif_get_node_info_resp,/* SCIF_GET_NODE_INFO */ + scif_recv_reg, /* SCIF_REGISTER */ + scif_recv_reg_ack, /* SCIF_REGISTER_ACK */ + scif_recv_reg_nack, /* SCIF_REGISTER_NACK */ + scif_recv_unreg, /* SCIF_UNREGISTER */ + scif_recv_unreg_ack, /* SCIF_UNREGISTER_ACK */ + scif_recv_unreg_nack, /* SCIF_UNREGISTER_NACK */ + scif_alloc_req, /* SCIF_ALLOC_REQ */ + scif_alloc_gnt_rej, /* SCIF_ALLOC_GNT */ + scif_alloc_gnt_rej, /* SCIF_ALLOC_REJ */ + scif_free_virt, /* SCIF_FREE_VIRT */ + scif_recv_munmap, /* SCIF_MUNMAP */ + scif_recv_mark, /* SCIF_MARK */ + scif_recv_mark_resp, /* SCIF_MARK_ACK */ + scif_recv_mark_resp, /* SCIF_MARK_NACK */ + scif_recv_wait, /* SCIF_WAIT */ + scif_recv_wait_resp, /* SCIF_WAIT_ACK */ + scif_recv_wait_resp, /* SCIF_WAIT_NACK */ + scif_recv_sig_local, /* SCIF_SIG_LOCAL */ + scif_recv_sig_remote, /* SCIF_SIG_REMOTE */ + scif_recv_sig_resp, /* SCIF_SIG_ACK */ + scif_recv_sig_resp, /* SCIF_SIG_NACK */ }; /** @@ -1169,7 +1210,6 @@ int scif_setup_loopback_qp(struct scif_dev *scifdev) int err = 0; void *local_q; struct scif_qp *qp; - struct scif_peer_dev *spdev; err = scif_setup_intr_wq(scifdev); if (err) @@ -1216,15 +1256,11 @@ int scif_setup_loopback_qp(struct scif_dev *scifdev) &qp->local_write, local_q, get_count_order(SCIF_NODE_QP_SIZE)); scif_info.nodeid = scifdev->node; - spdev = scif_peer_register_device(scifdev); - if (IS_ERR(spdev)) { - err = PTR_ERR(spdev); - goto free_local_q; - } + + scif_peer_register_device(scifdev); + scif_info.loopb_dev = scifdev; return err; -free_local_q: - kfree(local_q); free_qpairs: kfree(scifdev->qpairs); destroy_loopb_wq: @@ -1243,13 +1279,7 @@ exit: */ int scif_destroy_loopback_qp(struct scif_dev *scifdev) { - struct scif_peer_dev *spdev; - - rcu_read_lock(); - spdev = rcu_dereference(scifdev->spdev); - rcu_read_unlock(); - if (spdev) - scif_peer_unregister_device(spdev); + scif_peer_unregister_device(scifdev); destroy_workqueue(scif_info.loopb_wq); scif_destroy_intr_wq(scifdev); kfree(scifdev->qpairs->outbound_q.rb_base); diff --git a/drivers/misc/mic/scif/scif_nodeqp.h b/drivers/misc/mic/scif/scif_nodeqp.h index 6c0ed6783479..95896273138e 100644 --- a/drivers/misc/mic/scif/scif_nodeqp.h +++ b/drivers/misc/mic/scif/scif_nodeqp.h @@ -74,7 +74,28 @@ #define SCIF_CLIENT_SENT 16 /* Notify the peer that data has been written */ #define SCIF_CLIENT_RCVD 17 /* Notify the peer that data has been read */ #define SCIF_GET_NODE_INFO 18 /* Get current node mask from the mgmt node*/ -#define SCIF_MAX_MSG SCIF_GET_NODE_INFO +#define SCIF_REGISTER 19 /* Tell peer about a new registered window */ +#define SCIF_REGISTER_ACK 20 /* Notify peer about unregistration success */ +#define SCIF_REGISTER_NACK 21 /* Notify peer about registration success */ +#define SCIF_UNREGISTER 22 /* Tell peer about unregistering a window */ +#define SCIF_UNREGISTER_ACK 23 /* Notify peer about registration failure */ +#define SCIF_UNREGISTER_NACK 24 /* Notify peer about unregistration failure */ +#define SCIF_ALLOC_REQ 25 /* Request a mapped buffer */ +#define SCIF_ALLOC_GNT 26 /* Notify peer about allocation success */ +#define SCIF_ALLOC_REJ 27 /* Notify peer about allocation failure */ +#define SCIF_FREE_VIRT 28 /* Free previously allocated virtual memory */ +#define SCIF_MUNMAP 29 /* Acknowledgment for a SCIF_MMAP request */ +#define SCIF_MARK 30 /* SCIF Remote Fence Mark Request */ +#define SCIF_MARK_ACK 31 /* SCIF Remote Fence Mark Success */ +#define SCIF_MARK_NACK 32 /* SCIF Remote Fence Mark Failure */ +#define SCIF_WAIT 33 /* SCIF Remote Fence Wait Request */ +#define SCIF_WAIT_ACK 34 /* SCIF Remote Fence Wait Success */ +#define SCIF_WAIT_NACK 35 /* SCIF Remote Fence Wait Failure */ +#define SCIF_SIG_LOCAL 36 /* SCIF Remote Fence Local Signal Request */ +#define SCIF_SIG_REMOTE 37 /* SCIF Remote Fence Remote Signal Request */ +#define SCIF_SIG_ACK 38 /* SCIF Remote Fence Remote Signal Success */ +#define SCIF_SIG_NACK 39 /* SCIF Remote Fence Remote Signal Failure */ +#define SCIF_MAX_MSG SCIF_SIG_NACK /* * struct scifmsg - Node QP message format @@ -92,6 +113,24 @@ struct scifmsg { } __packed; /* + * struct scif_allocmsg - Used with SCIF_ALLOC_REQ to request + * the remote note to allocate memory + * + * phys_addr: Physical address of the buffer + * vaddr: Virtual address of the buffer + * size: Size of the buffer + * state: Current state + * allocwq: wait queue for status + */ +struct scif_allocmsg { + dma_addr_t phys_addr; + unsigned long vaddr; + size_t size; + enum scif_msg_state state; + wait_queue_head_t allocwq; +}; + +/* * struct scif_qp - Node Queue Pair * * Interesting structure -- a little difficult because we can only @@ -158,7 +197,6 @@ int scif_setup_qp_connect_response(struct scif_dev *scifdev, int scif_setup_loopback_qp(struct scif_dev *scifdev); int scif_destroy_loopback_qp(struct scif_dev *scifdev); void scif_poll_qp_state(struct work_struct *work); -void scif_qp_response_ack(struct work_struct *work); void scif_destroy_p2p(struct scif_dev *scifdev); void scif_send_exit(struct scif_dev *scifdev); static inline struct device *scif_get_peer_dev(struct scif_dev *scifdev) diff --git a/drivers/misc/mic/scif/scif_peer_bus.c b/drivers/misc/mic/scif/scif_peer_bus.c index 589ae9ad2501..6ffa3bdbd45b 100644 --- a/drivers/misc/mic/scif/scif_peer_bus.c +++ b/drivers/misc/mic/scif/scif_peer_bus.c @@ -24,93 +24,152 @@ dev_to_scif_peer(struct device *dev) return container_of(dev, struct scif_peer_dev, dev); } -static inline struct scif_peer_driver * -drv_to_scif_peer(struct device_driver *drv) -{ - return container_of(drv, struct scif_peer_driver, driver); -} +struct bus_type scif_peer_bus = { + .name = "scif_peer_bus", +}; -static int scif_peer_dev_match(struct device *dv, struct device_driver *dr) +static void scif_peer_release_dev(struct device *d) { - return !strncmp(dev_name(dv), dr->name, 4); + struct scif_peer_dev *sdev = dev_to_scif_peer(d); + struct scif_dev *scifdev = &scif_dev[sdev->dnode]; + + scif_cleanup_scifdev(scifdev); + kfree(sdev); } -static int scif_peer_dev_probe(struct device *d) +static int scif_peer_initialize_device(struct scif_dev *scifdev) { - struct scif_peer_dev *dev = dev_to_scif_peer(d); - struct scif_peer_driver *drv = drv_to_scif_peer(dev->dev.driver); + struct scif_peer_dev *spdev; + int ret; - return drv->probe(dev); -} + spdev = kzalloc(sizeof(*spdev), GFP_KERNEL); + if (!spdev) { + ret = -ENOMEM; + goto err; + } -static int scif_peer_dev_remove(struct device *d) -{ - struct scif_peer_dev *dev = dev_to_scif_peer(d); - struct scif_peer_driver *drv = drv_to_scif_peer(dev->dev.driver); + spdev->dev.parent = scifdev->sdev->dev.parent; + spdev->dev.release = scif_peer_release_dev; + spdev->dnode = scifdev->node; + spdev->dev.bus = &scif_peer_bus; + dev_set_name(&spdev->dev, "scif_peer-dev%u", spdev->dnode); + + device_initialize(&spdev->dev); + get_device(&spdev->dev); + rcu_assign_pointer(scifdev->spdev, spdev); - drv->remove(dev); + mutex_lock(&scif_info.conflock); + scif_info.total++; + scif_info.maxid = max_t(u32, spdev->dnode, scif_info.maxid); + mutex_unlock(&scif_info.conflock); return 0; +err: + dev_err(&scifdev->sdev->dev, + "dnode %d: initialize_device rc %d\n", scifdev->node, ret); + return ret; } -static struct bus_type scif_peer_bus = { - .name = "scif_peer_bus", - .match = scif_peer_dev_match, - .probe = scif_peer_dev_probe, - .remove = scif_peer_dev_remove, -}; - -int scif_peer_register_driver(struct scif_peer_driver *driver) +static int scif_peer_add_device(struct scif_dev *scifdev) { - driver->driver.bus = &scif_peer_bus; - return driver_register(&driver->driver); + struct scif_peer_dev *spdev = rcu_dereference(scifdev->spdev); + char pool_name[16]; + int ret; + + ret = device_add(&spdev->dev); + put_device(&spdev->dev); + if (ret) { + dev_err(&scifdev->sdev->dev, + "dnode %d: peer device_add failed\n", scifdev->node); + goto put_spdev; + } + + scnprintf(pool_name, sizeof(pool_name), "scif-%d", spdev->dnode); + scifdev->signal_pool = dmam_pool_create(pool_name, &scifdev->sdev->dev, + sizeof(struct scif_status), 1, + 0); + if (!scifdev->signal_pool) { + dev_err(&scifdev->sdev->dev, + "dnode %d: dmam_pool_create failed\n", scifdev->node); + ret = -ENOMEM; + goto del_spdev; + } + dev_dbg(&spdev->dev, "Added peer dnode %d\n", spdev->dnode); + return 0; +del_spdev: + device_del(&spdev->dev); +put_spdev: + RCU_INIT_POINTER(scifdev->spdev, NULL); + synchronize_rcu(); + put_device(&spdev->dev); + + mutex_lock(&scif_info.conflock); + scif_info.total--; + mutex_unlock(&scif_info.conflock); + return ret; } -void scif_peer_unregister_driver(struct scif_peer_driver *driver) +void scif_add_peer_device(struct work_struct *work) { - driver_unregister(&driver->driver); + struct scif_dev *scifdev = container_of(work, struct scif_dev, + peer_add_work); + + scif_peer_add_device(scifdev); } -static void scif_peer_release_dev(struct device *d) +/* + * Peer device registration is split into a device_initialize and a device_add. + * The reason for doing this is as follows: First, peer device registration + * itself cannot be done in the message processing thread and must be delegated + * to another workqueue, otherwise if SCIF client probe, called during peer + * device registration, calls scif_connect(..), it will block the message + * processing thread causing a deadlock. Next, device_initialize is done in the + * "top-half" message processing thread and device_add in the "bottom-half" + * workqueue. If this is not done, SCIF_CNCT_REQ message processing executing + * concurrently with SCIF_INIT message processing is unable to get a reference + * on the peer device, thereby failing the connect request. + */ +void scif_peer_register_device(struct scif_dev *scifdev) { - struct scif_peer_dev *sdev = dev_to_scif_peer(d); - struct scif_dev *scifdev = &scif_dev[sdev->dnode]; + int ret; - scif_cleanup_scifdev(scifdev); - kfree(sdev); + mutex_lock(&scifdev->lock); + ret = scif_peer_initialize_device(scifdev); + if (ret) + goto exit; + schedule_work(&scifdev->peer_add_work); +exit: + mutex_unlock(&scifdev->lock); } -struct scif_peer_dev * -scif_peer_register_device(struct scif_dev *scifdev) +int scif_peer_unregister_device(struct scif_dev *scifdev) { - int ret; struct scif_peer_dev *spdev; - spdev = kzalloc(sizeof(*spdev), GFP_KERNEL); - if (!spdev) - return ERR_PTR(-ENOMEM); - - spdev->dev.parent = scifdev->sdev->dev.parent; - spdev->dev.release = scif_peer_release_dev; - spdev->dnode = scifdev->node; - spdev->dev.bus = &scif_peer_bus; + mutex_lock(&scifdev->lock); + /* Flush work to ensure device register is complete */ + flush_work(&scifdev->peer_add_work); - dev_set_name(&spdev->dev, "scif_peer-dev%u", spdev->dnode); /* - * device_register() causes the bus infrastructure to look for a - * matching driver. + * Continue holding scifdev->lock since theoretically unregister_device + * can be called simultaneously from multiple threads */ - ret = device_register(&spdev->dev); - if (ret) - goto free_spdev; - return spdev; -free_spdev: - kfree(spdev); - return ERR_PTR(ret); -} - -void scif_peer_unregister_device(struct scif_peer_dev *sdev) -{ - device_unregister(&sdev->dev); + spdev = rcu_dereference(scifdev->spdev); + if (!spdev) { + mutex_unlock(&scifdev->lock); + return -ENODEV; + } + + RCU_INIT_POINTER(scifdev->spdev, NULL); + synchronize_rcu(); + mutex_unlock(&scifdev->lock); + + dev_dbg(&spdev->dev, "Removing peer dnode %d\n", spdev->dnode); + device_unregister(&spdev->dev); + + mutex_lock(&scif_info.conflock); + scif_info.total--; + mutex_unlock(&scif_info.conflock); + return 0; } int scif_peer_bus_init(void) diff --git a/drivers/misc/mic/scif/scif_peer_bus.h b/drivers/misc/mic/scif/scif_peer_bus.h index 33f0dbb30152..a3b8dd2edaa5 100644 --- a/drivers/misc/mic/scif/scif_peer_bus.h +++ b/drivers/misc/mic/scif/scif_peer_bus.h @@ -19,47 +19,13 @@ #include <linux/device.h> #include <linux/mic_common.h> - -/* - * Peer devices show up as PCIe devices for the mgmt node but not the cards. - * The mgmt node discovers all the cards on the PCIe bus and informs the other - * cards about their peers. Upon notification of a peer a node adds a peer - * device to the peer bus to maintain symmetry in the way devices are - * discovered across all nodes in the SCIF network. - */ -/** - * scif_peer_dev - representation of a peer SCIF device - * @dev: underlying device - * @dnode - The destination node which this device will communicate with. - */ -struct scif_peer_dev { - struct device dev; - u8 dnode; -}; - -/** - * scif_peer_driver - operations for a scif_peer I/O driver - * @driver: underlying device driver (populate name and owner). - * @id_table: the ids serviced by this driver. - * @probe: the function to call when a device is found. Returns 0 or -errno. - * @remove: the function to call when a device is removed. - */ -struct scif_peer_driver { - struct device_driver driver; - const struct scif_peer_dev_id *id_table; - - int (*probe)(struct scif_peer_dev *dev); - void (*remove)(struct scif_peer_dev *dev); -}; +#include <linux/scif.h> struct scif_dev; -int scif_peer_register_driver(struct scif_peer_driver *driver); -void scif_peer_unregister_driver(struct scif_peer_driver *driver); - -struct scif_peer_dev *scif_peer_register_device(struct scif_dev *sdev); -void scif_peer_unregister_device(struct scif_peer_dev *sdev); - +void scif_add_peer_device(struct work_struct *work); +void scif_peer_register_device(struct scif_dev *sdev); +int scif_peer_unregister_device(struct scif_dev *scifdev); int scif_peer_bus_init(void); void scif_peer_bus_exit(void); #endif /* _SCIF_PEER_BUS_H */ diff --git a/drivers/misc/mic/scif/scif_rma.c b/drivers/misc/mic/scif/scif_rma.c new file mode 100644 index 000000000000..e2889967036a --- /dev/null +++ b/drivers/misc/mic/scif/scif_rma.c @@ -0,0 +1,1770 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * Intel SCIF driver. + * + */ +#include <linux/dma_remapping.h> +#include <linux/pagemap.h> +#include "scif_main.h" +#include "scif_map.h" + +/* Used to skip ulimit checks for registrations with SCIF_MAP_KERNEL flag */ +#define SCIF_MAP_ULIMIT 0x40 + +bool scif_ulimit_check = 1; + +/** + * scif_rma_ep_init: + * @ep: end point + * + * Initialize RMA per EP data structures. + */ +void scif_rma_ep_init(struct scif_endpt *ep) +{ + struct scif_endpt_rma_info *rma = &ep->rma_info; + + mutex_init(&rma->rma_lock); + init_iova_domain(&rma->iovad, PAGE_SIZE, SCIF_IOVA_START_PFN, + SCIF_DMA_64BIT_PFN); + spin_lock_init(&rma->tc_lock); + mutex_init(&rma->mmn_lock); + INIT_LIST_HEAD(&rma->reg_list); + INIT_LIST_HEAD(&rma->remote_reg_list); + atomic_set(&rma->tw_refcount, 0); + atomic_set(&rma->tcw_refcount, 0); + atomic_set(&rma->tcw_total_pages, 0); + atomic_set(&rma->fence_refcount, 0); + + rma->async_list_del = 0; + rma->dma_chan = NULL; + INIT_LIST_HEAD(&rma->mmn_list); + INIT_LIST_HEAD(&rma->vma_list); + init_waitqueue_head(&rma->markwq); +} + +/** + * scif_rma_ep_can_uninit: + * @ep: end point + * + * Returns 1 if an endpoint can be uninitialized and 0 otherwise. + */ +int scif_rma_ep_can_uninit(struct scif_endpt *ep) +{ + int ret = 0; + + mutex_lock(&ep->rma_info.rma_lock); + /* Destroy RMA Info only if both lists are empty */ + if (list_empty(&ep->rma_info.reg_list) && + list_empty(&ep->rma_info.remote_reg_list) && + list_empty(&ep->rma_info.mmn_list) && + !atomic_read(&ep->rma_info.tw_refcount) && + !atomic_read(&ep->rma_info.tcw_refcount) && + !atomic_read(&ep->rma_info.fence_refcount)) + ret = 1; + mutex_unlock(&ep->rma_info.rma_lock); + return ret; +} + +/** + * scif_create_pinned_pages: + * @nr_pages: number of pages in window + * @prot: read/write protection + * + * Allocate and prepare a set of pinned pages. + */ +static struct scif_pinned_pages * +scif_create_pinned_pages(int nr_pages, int prot) +{ + struct scif_pinned_pages *pin; + + might_sleep(); + pin = scif_zalloc(sizeof(*pin)); + if (!pin) + goto error; + + pin->pages = scif_zalloc(nr_pages * sizeof(*pin->pages)); + if (!pin->pages) + goto error_free_pinned_pages; + + pin->prot = prot; + pin->magic = SCIFEP_MAGIC; + return pin; + +error_free_pinned_pages: + scif_free(pin, sizeof(*pin)); +error: + return NULL; +} + +/** + * scif_destroy_pinned_pages: + * @pin: A set of pinned pages. + * + * Deallocate resources for pinned pages. + */ +static int scif_destroy_pinned_pages(struct scif_pinned_pages *pin) +{ + int j; + int writeable = pin->prot & SCIF_PROT_WRITE; + int kernel = SCIF_MAP_KERNEL & pin->map_flags; + + for (j = 0; j < pin->nr_pages; j++) { + if (pin->pages[j] && !kernel) { + if (writeable) + SetPageDirty(pin->pages[j]); + put_page(pin->pages[j]); + } + } + + scif_free(pin->pages, + pin->nr_pages * sizeof(*pin->pages)); + scif_free(pin, sizeof(*pin)); + return 0; +} + +/* + * scif_create_window: + * @ep: end point + * @nr_pages: number of pages + * @offset: registration offset + * @temp: true if a temporary window is being created + * + * Allocate and prepare a self registration window. + */ +struct scif_window *scif_create_window(struct scif_endpt *ep, int nr_pages, + s64 offset, bool temp) +{ + struct scif_window *window; + + might_sleep(); + window = scif_zalloc(sizeof(*window)); + if (!window) + goto error; + + window->dma_addr = scif_zalloc(nr_pages * sizeof(*window->dma_addr)); + if (!window->dma_addr) + goto error_free_window; + + window->num_pages = scif_zalloc(nr_pages * sizeof(*window->num_pages)); + if (!window->num_pages) + goto error_free_window; + + window->offset = offset; + window->ep = (u64)ep; + window->magic = SCIFEP_MAGIC; + window->reg_state = OP_IDLE; + init_waitqueue_head(&window->regwq); + window->unreg_state = OP_IDLE; + init_waitqueue_head(&window->unregwq); + INIT_LIST_HEAD(&window->list); + window->type = SCIF_WINDOW_SELF; + window->temp = temp; + return window; + +error_free_window: + scif_free(window->dma_addr, + nr_pages * sizeof(*window->dma_addr)); + scif_free(window, sizeof(*window)); +error: + return NULL; +} + +/** + * scif_destroy_incomplete_window: + * @ep: end point + * @window: registration window + * + * Deallocate resources for self window. + */ +static void scif_destroy_incomplete_window(struct scif_endpt *ep, + struct scif_window *window) +{ + int err; + int nr_pages = window->nr_pages; + struct scif_allocmsg *alloc = &window->alloc_handle; + struct scifmsg msg; + +retry: + /* Wait for a SCIF_ALLOC_GNT/REJ message */ + err = wait_event_timeout(alloc->allocwq, + alloc->state != OP_IN_PROGRESS, + SCIF_NODE_ALIVE_TIMEOUT); + if (!err && scifdev_alive(ep)) + goto retry; + + mutex_lock(&ep->rma_info.rma_lock); + if (alloc->state == OP_COMPLETED) { + msg.uop = SCIF_FREE_VIRT; + msg.src = ep->port; + msg.payload[0] = ep->remote_ep; + msg.payload[1] = window->alloc_handle.vaddr; + msg.payload[2] = (u64)window; + msg.payload[3] = SCIF_REGISTER; + _scif_nodeqp_send(ep->remote_dev, &msg); + } + mutex_unlock(&ep->rma_info.rma_lock); + + scif_free_window_offset(ep, window, window->offset); + scif_free(window->dma_addr, nr_pages * sizeof(*window->dma_addr)); + scif_free(window->num_pages, nr_pages * sizeof(*window->num_pages)); + scif_free(window, sizeof(*window)); +} + +/** + * scif_unmap_window: + * @remote_dev: SCIF remote device + * @window: registration window + * + * Delete any DMA mappings created for a registered self window + */ +void scif_unmap_window(struct scif_dev *remote_dev, struct scif_window *window) +{ + int j; + + if (scif_is_iommu_enabled() && !scifdev_self(remote_dev)) { + if (window->st) { + dma_unmap_sg(&remote_dev->sdev->dev, + window->st->sgl, window->st->nents, + DMA_BIDIRECTIONAL); + sg_free_table(window->st); + kfree(window->st); + window->st = NULL; + } + } else { + for (j = 0; j < window->nr_contig_chunks; j++) { + if (window->dma_addr[j]) { + scif_unmap_single(window->dma_addr[j], + remote_dev, + window->num_pages[j] << + PAGE_SHIFT); + window->dma_addr[j] = 0x0; + } + } + } +} + +static inline struct mm_struct *__scif_acquire_mm(void) +{ + if (scif_ulimit_check) + return get_task_mm(current); + return NULL; +} + +static inline void __scif_release_mm(struct mm_struct *mm) +{ + if (mm) + mmput(mm); +} + +static inline int +__scif_dec_pinned_vm_lock(struct mm_struct *mm, + int nr_pages, bool try_lock) +{ + if (!mm || !nr_pages || !scif_ulimit_check) + return 0; + if (try_lock) { + if (!down_write_trylock(&mm->mmap_sem)) { + dev_err(scif_info.mdev.this_device, + "%s %d err\n", __func__, __LINE__); + return -1; + } + } else { + down_write(&mm->mmap_sem); + } + mm->pinned_vm -= nr_pages; + up_write(&mm->mmap_sem); + return 0; +} + +static inline int __scif_check_inc_pinned_vm(struct mm_struct *mm, + int nr_pages) +{ + unsigned long locked, lock_limit; + + if (!mm || !nr_pages || !scif_ulimit_check) + return 0; + + locked = nr_pages; + locked += mm->pinned_vm; + lock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT; + if ((locked > lock_limit) && !capable(CAP_IPC_LOCK)) { + dev_err(scif_info.mdev.this_device, + "locked(%lu) > lock_limit(%lu)\n", + locked, lock_limit); + return -ENOMEM; + } + mm->pinned_vm = locked; + return 0; +} + +/** + * scif_destroy_window: + * @ep: end point + * @window: registration window + * + * Deallocate resources for self window. + */ +int scif_destroy_window(struct scif_endpt *ep, struct scif_window *window) +{ + int j; + struct scif_pinned_pages *pinned_pages = window->pinned_pages; + int nr_pages = window->nr_pages; + + might_sleep(); + if (!window->temp && window->mm) { + __scif_dec_pinned_vm_lock(window->mm, window->nr_pages, 0); + __scif_release_mm(window->mm); + window->mm = NULL; + } + + scif_free_window_offset(ep, window, window->offset); + scif_unmap_window(ep->remote_dev, window); + /* + * Decrement references for this set of pinned pages from + * this window. + */ + j = atomic_sub_return(1, &pinned_pages->ref_count); + if (j < 0) + dev_err(scif_info.mdev.this_device, + "%s %d incorrect ref count %d\n", + __func__, __LINE__, j); + /* + * If the ref count for pinned_pages is zero then someone + * has already called scif_unpin_pages() for it and we should + * destroy the page cache. + */ + if (!j) + scif_destroy_pinned_pages(window->pinned_pages); + scif_free(window->dma_addr, nr_pages * sizeof(*window->dma_addr)); + scif_free(window->num_pages, nr_pages * sizeof(*window->num_pages)); + window->magic = 0; + scif_free(window, sizeof(*window)); + return 0; +} + +/** + * scif_create_remote_lookup: + * @remote_dev: SCIF remote device + * @window: remote window + * + * Allocate and prepare lookup entries for the remote + * end to copy over the physical addresses. + * Returns 0 on success and appropriate errno on failure. + */ +static int scif_create_remote_lookup(struct scif_dev *remote_dev, + struct scif_window *window) +{ + int i, j, err = 0; + int nr_pages = window->nr_pages; + bool vmalloc_dma_phys, vmalloc_num_pages; + + might_sleep(); + /* Map window */ + err = scif_map_single(&window->mapped_offset, + window, remote_dev, sizeof(*window)); + if (err) + goto error_window; + + /* Compute the number of lookup entries. 21 == 2MB Shift */ + window->nr_lookup = ALIGN(nr_pages * PAGE_SIZE, + ((2) * 1024 * 1024)) >> 21; + + window->dma_addr_lookup.lookup = + scif_alloc_coherent(&window->dma_addr_lookup.offset, + remote_dev, window->nr_lookup * + sizeof(*window->dma_addr_lookup.lookup), + GFP_KERNEL | __GFP_ZERO); + if (!window->dma_addr_lookup.lookup) + goto error_window; + + window->num_pages_lookup.lookup = + scif_alloc_coherent(&window->num_pages_lookup.offset, + remote_dev, window->nr_lookup * + sizeof(*window->num_pages_lookup.lookup), + GFP_KERNEL | __GFP_ZERO); + if (!window->num_pages_lookup.lookup) + goto error_window; + + vmalloc_dma_phys = is_vmalloc_addr(&window->dma_addr[0]); + vmalloc_num_pages = is_vmalloc_addr(&window->num_pages[0]); + + /* Now map each of the pages containing physical addresses */ + for (i = 0, j = 0; i < nr_pages; i += SCIF_NR_ADDR_IN_PAGE, j++) { + err = scif_map_page(&window->dma_addr_lookup.lookup[j], + vmalloc_dma_phys ? + vmalloc_to_page(&window->dma_addr[i]) : + virt_to_page(&window->dma_addr[i]), + remote_dev); + if (err) + goto error_window; + err = scif_map_page(&window->num_pages_lookup.lookup[j], + vmalloc_dma_phys ? + vmalloc_to_page(&window->num_pages[i]) : + virt_to_page(&window->num_pages[i]), + remote_dev); + if (err) + goto error_window; + } + return 0; +error_window: + return err; +} + +/** + * scif_destroy_remote_lookup: + * @remote_dev: SCIF remote device + * @window: remote window + * + * Destroy lookup entries used for the remote + * end to copy over the physical addresses. + */ +static void scif_destroy_remote_lookup(struct scif_dev *remote_dev, + struct scif_window *window) +{ + int i, j; + + if (window->nr_lookup) { + struct scif_rma_lookup *lup = &window->dma_addr_lookup; + struct scif_rma_lookup *npup = &window->num_pages_lookup; + + for (i = 0, j = 0; i < window->nr_pages; + i += SCIF_NR_ADDR_IN_PAGE, j++) { + if (lup->lookup && lup->lookup[j]) + scif_unmap_single(lup->lookup[j], + remote_dev, + PAGE_SIZE); + if (npup->lookup && npup->lookup[j]) + scif_unmap_single(npup->lookup[j], + remote_dev, + PAGE_SIZE); + } + if (lup->lookup) + scif_free_coherent(lup->lookup, lup->offset, + remote_dev, window->nr_lookup * + sizeof(*lup->lookup)); + if (npup->lookup) + scif_free_coherent(npup->lookup, npup->offset, + remote_dev, window->nr_lookup * + sizeof(*npup->lookup)); + if (window->mapped_offset) + scif_unmap_single(window->mapped_offset, + remote_dev, sizeof(*window)); + window->nr_lookup = 0; + } +} + +/** + * scif_create_remote_window: + * @ep: end point + * @nr_pages: number of pages in window + * + * Allocate and prepare a remote registration window. + */ +static struct scif_window * +scif_create_remote_window(struct scif_dev *scifdev, int nr_pages) +{ + struct scif_window *window; + + might_sleep(); + window = scif_zalloc(sizeof(*window)); + if (!window) + goto error_ret; + + window->magic = SCIFEP_MAGIC; + window->nr_pages = nr_pages; + + window->dma_addr = scif_zalloc(nr_pages * sizeof(*window->dma_addr)); + if (!window->dma_addr) + goto error_window; + + window->num_pages = scif_zalloc(nr_pages * + sizeof(*window->num_pages)); + if (!window->num_pages) + goto error_window; + + if (scif_create_remote_lookup(scifdev, window)) + goto error_window; + + window->type = SCIF_WINDOW_PEER; + window->unreg_state = OP_IDLE; + INIT_LIST_HEAD(&window->list); + return window; +error_window: + scif_destroy_remote_window(window); +error_ret: + return NULL; +} + +/** + * scif_destroy_remote_window: + * @ep: end point + * @window: remote registration window + * + * Deallocate resources for remote window. + */ +void +scif_destroy_remote_window(struct scif_window *window) +{ + scif_free(window->dma_addr, window->nr_pages * + sizeof(*window->dma_addr)); + scif_free(window->num_pages, window->nr_pages * + sizeof(*window->num_pages)); + window->magic = 0; + scif_free(window, sizeof(*window)); +} + +/** + * scif_iommu_map: create DMA mappings if the IOMMU is enabled + * @remote_dev: SCIF remote device + * @window: remote registration window + * + * Map the physical pages using dma_map_sg(..) and then detect the number + * of contiguous DMA mappings allocated + */ +static int scif_iommu_map(struct scif_dev *remote_dev, + struct scif_window *window) +{ + struct scatterlist *sg; + int i, err; + scif_pinned_pages_t pin = window->pinned_pages; + + window->st = kzalloc(sizeof(*window->st), GFP_KERNEL); + if (!window->st) + return -ENOMEM; + + err = sg_alloc_table(window->st, window->nr_pages, GFP_KERNEL); + if (err) + return err; + + for_each_sg(window->st->sgl, sg, window->st->nents, i) + sg_set_page(sg, pin->pages[i], PAGE_SIZE, 0x0); + + err = dma_map_sg(&remote_dev->sdev->dev, window->st->sgl, + window->st->nents, DMA_BIDIRECTIONAL); + if (!err) + return -ENOMEM; + /* Detect contiguous ranges of DMA mappings */ + sg = window->st->sgl; + for (i = 0; sg; i++) { + dma_addr_t last_da; + + window->dma_addr[i] = sg_dma_address(sg); + window->num_pages[i] = sg_dma_len(sg) >> PAGE_SHIFT; + last_da = sg_dma_address(sg) + sg_dma_len(sg); + while ((sg = sg_next(sg)) && sg_dma_address(sg) == last_da) { + window->num_pages[i] += + (sg_dma_len(sg) >> PAGE_SHIFT); + last_da = window->dma_addr[i] + + sg_dma_len(sg); + } + window->nr_contig_chunks++; + } + return 0; +} + +/** + * scif_map_window: + * @remote_dev: SCIF remote device + * @window: self registration window + * + * Map pages of a window into the aperture/PCI. + * Also determine addresses required for DMA. + */ +int +scif_map_window(struct scif_dev *remote_dev, struct scif_window *window) +{ + int i, j, k, err = 0, nr_contig_pages; + scif_pinned_pages_t pin; + phys_addr_t phys_prev, phys_curr; + + might_sleep(); + + pin = window->pinned_pages; + + if (intel_iommu_enabled && !scifdev_self(remote_dev)) + return scif_iommu_map(remote_dev, window); + + for (i = 0, j = 0; i < window->nr_pages; i += nr_contig_pages, j++) { + phys_prev = page_to_phys(pin->pages[i]); + nr_contig_pages = 1; + + /* Detect physically contiguous chunks */ + for (k = i + 1; k < window->nr_pages; k++) { + phys_curr = page_to_phys(pin->pages[k]); + if (phys_curr != (phys_prev + PAGE_SIZE)) + break; + phys_prev = phys_curr; + nr_contig_pages++; + } + window->num_pages[j] = nr_contig_pages; + window->nr_contig_chunks++; + if (scif_is_mgmt_node()) { + /* + * Management node has to deal with SMPT on X100 and + * hence the DMA mapping is required + */ + err = scif_map_single(&window->dma_addr[j], + phys_to_virt(page_to_phys( + pin->pages[i])), + remote_dev, + nr_contig_pages << PAGE_SHIFT); + if (err) + return err; + } else { + window->dma_addr[j] = page_to_phys(pin->pages[i]); + } + } + return err; +} + +/** + * scif_send_scif_unregister: + * @ep: end point + * @window: self registration window + * + * Send a SCIF_UNREGISTER message. + */ +static int scif_send_scif_unregister(struct scif_endpt *ep, + struct scif_window *window) +{ + struct scifmsg msg; + + msg.uop = SCIF_UNREGISTER; + msg.src = ep->port; + msg.payload[0] = window->alloc_handle.vaddr; + msg.payload[1] = (u64)window; + return scif_nodeqp_send(ep->remote_dev, &msg); +} + +/** + * scif_unregister_window: + * @window: self registration window + * + * Send an unregistration request and wait for a response. + */ +int scif_unregister_window(struct scif_window *window) +{ + int err = 0; + struct scif_endpt *ep = (struct scif_endpt *)window->ep; + bool send_msg = false; + + might_sleep(); + switch (window->unreg_state) { + case OP_IDLE: + { + window->unreg_state = OP_IN_PROGRESS; + send_msg = true; + /* fall through */ + } + case OP_IN_PROGRESS: + { + scif_get_window(window, 1); + mutex_unlock(&ep->rma_info.rma_lock); + if (send_msg) { + err = scif_send_scif_unregister(ep, window); + if (err) { + window->unreg_state = OP_COMPLETED; + goto done; + } + } else { + /* Return ENXIO since unregistration is in progress */ + return -ENXIO; + } +retry: + /* Wait for a SCIF_UNREGISTER_(N)ACK message */ + err = wait_event_timeout(window->unregwq, + window->unreg_state != OP_IN_PROGRESS, + SCIF_NODE_ALIVE_TIMEOUT); + if (!err && scifdev_alive(ep)) + goto retry; + if (!err) { + err = -ENODEV; + window->unreg_state = OP_COMPLETED; + dev_err(scif_info.mdev.this_device, + "%s %d err %d\n", __func__, __LINE__, err); + } + if (err > 0) + err = 0; +done: + mutex_lock(&ep->rma_info.rma_lock); + scif_put_window(window, 1); + break; + } + case OP_FAILED: + { + if (!scifdev_alive(ep)) { + err = -ENODEV; + window->unreg_state = OP_COMPLETED; + } + break; + } + case OP_COMPLETED: + break; + default: + err = -ENODEV; + } + + if (window->unreg_state == OP_COMPLETED && window->ref_count) + scif_put_window(window, window->nr_pages); + + if (!window->ref_count) { + atomic_inc(&ep->rma_info.tw_refcount); + list_del_init(&window->list); + scif_free_window_offset(ep, window, window->offset); + mutex_unlock(&ep->rma_info.rma_lock); + if ((!!(window->pinned_pages->map_flags & SCIF_MAP_KERNEL)) && + scifdev_alive(ep)) { + scif_drain_dma_intr(ep->remote_dev->sdev, + ep->rma_info.dma_chan); + } else { + if (!__scif_dec_pinned_vm_lock(window->mm, + window->nr_pages, 1)) { + __scif_release_mm(window->mm); + window->mm = NULL; + } + } + scif_queue_for_cleanup(window, &scif_info.rma); + mutex_lock(&ep->rma_info.rma_lock); + } + return err; +} + +/** + * scif_send_alloc_request: + * @ep: end point + * @window: self registration window + * + * Send a remote window allocation request + */ +static int scif_send_alloc_request(struct scif_endpt *ep, + struct scif_window *window) +{ + struct scifmsg msg; + struct scif_allocmsg *alloc = &window->alloc_handle; + + /* Set up the Alloc Handle */ + alloc->state = OP_IN_PROGRESS; + init_waitqueue_head(&alloc->allocwq); + + /* Send out an allocation request */ + msg.uop = SCIF_ALLOC_REQ; + msg.payload[1] = window->nr_pages; + msg.payload[2] = (u64)&window->alloc_handle; + return _scif_nodeqp_send(ep->remote_dev, &msg); +} + +/** + * scif_prep_remote_window: + * @ep: end point + * @window: self registration window + * + * Send a remote window allocation request, wait for an allocation response, + * and prepares the remote window by copying over the page lists + */ +static int scif_prep_remote_window(struct scif_endpt *ep, + struct scif_window *window) +{ + struct scifmsg msg; + struct scif_window *remote_window; + struct scif_allocmsg *alloc = &window->alloc_handle; + dma_addr_t *dma_phys_lookup, *tmp, *num_pages_lookup, *tmp1; + int i = 0, j = 0; + int nr_contig_chunks, loop_nr_contig_chunks; + int remaining_nr_contig_chunks, nr_lookup; + int err, map_err; + + map_err = scif_map_window(ep->remote_dev, window); + if (map_err) + dev_err(&ep->remote_dev->sdev->dev, + "%s %d map_err %d\n", __func__, __LINE__, map_err); + remaining_nr_contig_chunks = window->nr_contig_chunks; + nr_contig_chunks = window->nr_contig_chunks; +retry: + /* Wait for a SCIF_ALLOC_GNT/REJ message */ + err = wait_event_timeout(alloc->allocwq, + alloc->state != OP_IN_PROGRESS, + SCIF_NODE_ALIVE_TIMEOUT); + mutex_lock(&ep->rma_info.rma_lock); + /* Synchronize with the thread waking up allocwq */ + mutex_unlock(&ep->rma_info.rma_lock); + if (!err && scifdev_alive(ep)) + goto retry; + + if (!err) + err = -ENODEV; + + if (err > 0) + err = 0; + else + return err; + + /* Bail out. The remote end rejected this request */ + if (alloc->state == OP_FAILED) + return -ENOMEM; + + if (map_err) { + dev_err(&ep->remote_dev->sdev->dev, + "%s %d err %d\n", __func__, __LINE__, map_err); + msg.uop = SCIF_FREE_VIRT; + msg.src = ep->port; + msg.payload[0] = ep->remote_ep; + msg.payload[1] = window->alloc_handle.vaddr; + msg.payload[2] = (u64)window; + msg.payload[3] = SCIF_REGISTER; + spin_lock(&ep->lock); + if (ep->state == SCIFEP_CONNECTED) + err = _scif_nodeqp_send(ep->remote_dev, &msg); + else + err = -ENOTCONN; + spin_unlock(&ep->lock); + return err; + } + + remote_window = scif_ioremap(alloc->phys_addr, sizeof(*window), + ep->remote_dev); + + /* Compute the number of lookup entries. 21 == 2MB Shift */ + nr_lookup = ALIGN(nr_contig_chunks, SCIF_NR_ADDR_IN_PAGE) + >> ilog2(SCIF_NR_ADDR_IN_PAGE); + + dma_phys_lookup = + scif_ioremap(remote_window->dma_addr_lookup.offset, + nr_lookup * + sizeof(*remote_window->dma_addr_lookup.lookup), + ep->remote_dev); + num_pages_lookup = + scif_ioremap(remote_window->num_pages_lookup.offset, + nr_lookup * + sizeof(*remote_window->num_pages_lookup.lookup), + ep->remote_dev); + + while (remaining_nr_contig_chunks) { + loop_nr_contig_chunks = min_t(int, remaining_nr_contig_chunks, + (int)SCIF_NR_ADDR_IN_PAGE); + /* #1/2 - Copy physical addresses over to the remote side */ + + /* #2/2 - Copy DMA addresses (addresses that are fed into the + * DMA engine) We transfer bus addresses which are then + * converted into a MIC physical address on the remote + * side if it is a MIC, if the remote node is a mgmt node we + * transfer the MIC physical address + */ + tmp = scif_ioremap(dma_phys_lookup[j], + loop_nr_contig_chunks * + sizeof(*window->dma_addr), + ep->remote_dev); + tmp1 = scif_ioremap(num_pages_lookup[j], + loop_nr_contig_chunks * + sizeof(*window->num_pages), + ep->remote_dev); + if (scif_is_mgmt_node()) { + memcpy_toio((void __force __iomem *)tmp, + &window->dma_addr[i], loop_nr_contig_chunks + * sizeof(*window->dma_addr)); + memcpy_toio((void __force __iomem *)tmp1, + &window->num_pages[i], loop_nr_contig_chunks + * sizeof(*window->num_pages)); + } else { + if (scifdev_is_p2p(ep->remote_dev)) { + /* + * add remote node's base address for this node + * to convert it into a MIC address + */ + int m; + dma_addr_t dma_addr; + + for (m = 0; m < loop_nr_contig_chunks; m++) { + dma_addr = window->dma_addr[i + m] + + ep->remote_dev->base_addr; + writeq(dma_addr, + (void __force __iomem *)&tmp[m]); + } + memcpy_toio((void __force __iomem *)tmp1, + &window->num_pages[i], + loop_nr_contig_chunks + * sizeof(*window->num_pages)); + } else { + /* Mgmt node or loopback - transfer DMA + * addresses as is, this is the same as a + * MIC physical address (we use the dma_addr + * and not the phys_addr array since the + * phys_addr is only setup if there is a mmap() + * request from the mgmt node) + */ + memcpy_toio((void __force __iomem *)tmp, + &window->dma_addr[i], + loop_nr_contig_chunks * + sizeof(*window->dma_addr)); + memcpy_toio((void __force __iomem *)tmp1, + &window->num_pages[i], + loop_nr_contig_chunks * + sizeof(*window->num_pages)); + } + } + remaining_nr_contig_chunks -= loop_nr_contig_chunks; + i += loop_nr_contig_chunks; + j++; + scif_iounmap(tmp, loop_nr_contig_chunks * + sizeof(*window->dma_addr), ep->remote_dev); + scif_iounmap(tmp1, loop_nr_contig_chunks * + sizeof(*window->num_pages), ep->remote_dev); + } + + /* Prepare the remote window for the peer */ + remote_window->peer_window = (u64)window; + remote_window->offset = window->offset; + remote_window->prot = window->prot; + remote_window->nr_contig_chunks = nr_contig_chunks; + remote_window->ep = ep->remote_ep; + scif_iounmap(num_pages_lookup, + nr_lookup * + sizeof(*remote_window->num_pages_lookup.lookup), + ep->remote_dev); + scif_iounmap(dma_phys_lookup, + nr_lookup * + sizeof(*remote_window->dma_addr_lookup.lookup), + ep->remote_dev); + scif_iounmap(remote_window, sizeof(*remote_window), ep->remote_dev); + window->peer_window = alloc->vaddr; + return err; +} + +/** + * scif_send_scif_register: + * @ep: end point + * @window: self registration window + * + * Send a SCIF_REGISTER message if EP is connected and wait for a + * SCIF_REGISTER_(N)ACK message else send a SCIF_FREE_VIRT + * message so that the peer can free its remote window allocated earlier. + */ +static int scif_send_scif_register(struct scif_endpt *ep, + struct scif_window *window) +{ + int err = 0; + struct scifmsg msg; + + msg.src = ep->port; + msg.payload[0] = ep->remote_ep; + msg.payload[1] = window->alloc_handle.vaddr; + msg.payload[2] = (u64)window; + spin_lock(&ep->lock); + if (ep->state == SCIFEP_CONNECTED) { + msg.uop = SCIF_REGISTER; + window->reg_state = OP_IN_PROGRESS; + err = _scif_nodeqp_send(ep->remote_dev, &msg); + spin_unlock(&ep->lock); + if (!err) { +retry: + /* Wait for a SCIF_REGISTER_(N)ACK message */ + err = wait_event_timeout(window->regwq, + window->reg_state != + OP_IN_PROGRESS, + SCIF_NODE_ALIVE_TIMEOUT); + if (!err && scifdev_alive(ep)) + goto retry; + err = !err ? -ENODEV : 0; + if (window->reg_state == OP_FAILED) + err = -ENOTCONN; + } + } else { + msg.uop = SCIF_FREE_VIRT; + msg.payload[3] = SCIF_REGISTER; + err = _scif_nodeqp_send(ep->remote_dev, &msg); + spin_unlock(&ep->lock); + if (!err) + err = -ENOTCONN; + } + return err; +} + +/** + * scif_get_window_offset: + * @ep: end point descriptor + * @flags: flags + * @offset: offset hint + * @num_pages: number of pages + * @out_offset: computed offset returned by reference. + * + * Compute/Claim a new offset for this EP. + */ +int scif_get_window_offset(struct scif_endpt *ep, int flags, s64 offset, + int num_pages, s64 *out_offset) +{ + s64 page_index; + struct iova *iova_ptr; + int err = 0; + + if (flags & SCIF_MAP_FIXED) { + page_index = SCIF_IOVA_PFN(offset); + iova_ptr = reserve_iova(&ep->rma_info.iovad, page_index, + page_index + num_pages - 1); + if (!iova_ptr) + err = -EADDRINUSE; + } else { + iova_ptr = alloc_iova(&ep->rma_info.iovad, num_pages, + SCIF_DMA_63BIT_PFN - 1, 0); + if (!iova_ptr) + err = -ENOMEM; + } + if (!err) + *out_offset = (iova_ptr->pfn_lo) << PAGE_SHIFT; + return err; +} + +/** + * scif_free_window_offset: + * @ep: end point descriptor + * @window: registration window + * @offset: Offset to be freed + * + * Free offset for this EP. The callee is supposed to grab + * the RMA mutex before calling this API. + */ +void scif_free_window_offset(struct scif_endpt *ep, + struct scif_window *window, s64 offset) +{ + if ((window && !window->offset_freed) || !window) { + free_iova(&ep->rma_info.iovad, offset >> PAGE_SHIFT); + if (window) + window->offset_freed = true; + } +} + +/** + * scif_alloc_req: Respond to SCIF_ALLOC_REQ interrupt message + * @msg: Interrupt message + * + * Remote side is requesting a memory allocation. + */ +void scif_alloc_req(struct scif_dev *scifdev, struct scifmsg *msg) +{ + int err; + struct scif_window *window = NULL; + int nr_pages = msg->payload[1]; + + window = scif_create_remote_window(scifdev, nr_pages); + if (!window) { + err = -ENOMEM; + goto error; + } + + /* The peer's allocation request is granted */ + msg->uop = SCIF_ALLOC_GNT; + msg->payload[0] = (u64)window; + msg->payload[1] = window->mapped_offset; + err = scif_nodeqp_send(scifdev, msg); + if (err) + scif_destroy_remote_window(window); + return; +error: + /* The peer's allocation request is rejected */ + dev_err(&scifdev->sdev->dev, + "%s %d error %d alloc_ptr %p nr_pages 0x%x\n", + __func__, __LINE__, err, window, nr_pages); + msg->uop = SCIF_ALLOC_REJ; + scif_nodeqp_send(scifdev, msg); +} + +/** + * scif_alloc_gnt_rej: Respond to SCIF_ALLOC_GNT/REJ interrupt message + * @msg: Interrupt message + * + * Remote side responded to a memory allocation. + */ +void scif_alloc_gnt_rej(struct scif_dev *scifdev, struct scifmsg *msg) +{ + struct scif_allocmsg *handle = (struct scif_allocmsg *)msg->payload[2]; + struct scif_window *window = container_of(handle, struct scif_window, + alloc_handle); + struct scif_endpt *ep = (struct scif_endpt *)window->ep; + + mutex_lock(&ep->rma_info.rma_lock); + handle->vaddr = msg->payload[0]; + handle->phys_addr = msg->payload[1]; + if (msg->uop == SCIF_ALLOC_GNT) + handle->state = OP_COMPLETED; + else + handle->state = OP_FAILED; + wake_up(&handle->allocwq); + mutex_unlock(&ep->rma_info.rma_lock); +} + +/** + * scif_free_virt: Respond to SCIF_FREE_VIRT interrupt message + * @msg: Interrupt message + * + * Free up memory kmalloc'd earlier. + */ +void scif_free_virt(struct scif_dev *scifdev, struct scifmsg *msg) +{ + struct scif_window *window = (struct scif_window *)msg->payload[1]; + + scif_destroy_remote_window(window); +} + +static void +scif_fixup_aper_base(struct scif_dev *dev, struct scif_window *window) +{ + int j; + struct scif_hw_dev *sdev = dev->sdev; + phys_addr_t apt_base = 0; + + /* + * Add the aperture base if the DMA address is not card relative + * since the DMA addresses need to be an offset into the bar + */ + if (!scifdev_self(dev) && window->type == SCIF_WINDOW_PEER && + sdev->aper && !sdev->card_rel_da) + apt_base = sdev->aper->pa; + else + return; + + for (j = 0; j < window->nr_contig_chunks; j++) { + if (window->num_pages[j]) + window->dma_addr[j] += apt_base; + else + break; + } +} + +/** + * scif_recv_reg: Respond to SCIF_REGISTER interrupt message + * @msg: Interrupt message + * + * Update remote window list with a new registered window. + */ +void scif_recv_reg(struct scif_dev *scifdev, struct scifmsg *msg) +{ + struct scif_endpt *ep = (struct scif_endpt *)msg->payload[0]; + struct scif_window *window = + (struct scif_window *)msg->payload[1]; + + mutex_lock(&ep->rma_info.rma_lock); + spin_lock(&ep->lock); + if (ep->state == SCIFEP_CONNECTED) { + msg->uop = SCIF_REGISTER_ACK; + scif_nodeqp_send(ep->remote_dev, msg); + scif_fixup_aper_base(ep->remote_dev, window); + /* No further failures expected. Insert new window */ + scif_insert_window(window, &ep->rma_info.remote_reg_list); + } else { + msg->uop = SCIF_REGISTER_NACK; + scif_nodeqp_send(ep->remote_dev, msg); + } + spin_unlock(&ep->lock); + mutex_unlock(&ep->rma_info.rma_lock); + /* free up any lookup resources now that page lists are transferred */ + scif_destroy_remote_lookup(ep->remote_dev, window); + /* + * We could not insert the window but we need to + * destroy the window. + */ + if (msg->uop == SCIF_REGISTER_NACK) + scif_destroy_remote_window(window); +} + +/** + * scif_recv_unreg: Respond to SCIF_UNREGISTER interrupt message + * @msg: Interrupt message + * + * Remove window from remote registration list; + */ +void scif_recv_unreg(struct scif_dev *scifdev, struct scifmsg *msg) +{ + struct scif_rma_req req; + struct scif_window *window = NULL; + struct scif_window *recv_window = + (struct scif_window *)msg->payload[0]; + struct scif_endpt *ep; + int del_window = 0; + + ep = (struct scif_endpt *)recv_window->ep; + req.out_window = &window; + req.offset = recv_window->offset; + req.prot = 0; + req.nr_bytes = recv_window->nr_pages << PAGE_SHIFT; + req.type = SCIF_WINDOW_FULL; + req.head = &ep->rma_info.remote_reg_list; + msg->payload[0] = ep->remote_ep; + + mutex_lock(&ep->rma_info.rma_lock); + /* Does a valid window exist? */ + if (scif_query_window(&req)) { + dev_err(&scifdev->sdev->dev, + "%s %d -ENXIO\n", __func__, __LINE__); + msg->uop = SCIF_UNREGISTER_ACK; + goto error; + } + if (window) { + if (window->ref_count) + scif_put_window(window, window->nr_pages); + else + dev_err(&scifdev->sdev->dev, + "%s %d ref count should be +ve\n", + __func__, __LINE__); + window->unreg_state = OP_COMPLETED; + if (!window->ref_count) { + msg->uop = SCIF_UNREGISTER_ACK; + atomic_inc(&ep->rma_info.tw_refcount); + ep->rma_info.async_list_del = 1; + list_del_init(&window->list); + del_window = 1; + } else { + /* NACK! There are valid references to this window */ + msg->uop = SCIF_UNREGISTER_NACK; + } + } else { + /* The window did not make its way to the list at all. ACK */ + msg->uop = SCIF_UNREGISTER_ACK; + scif_destroy_remote_window(recv_window); + } +error: + mutex_unlock(&ep->rma_info.rma_lock); + if (del_window) + scif_drain_dma_intr(ep->remote_dev->sdev, + ep->rma_info.dma_chan); + scif_nodeqp_send(ep->remote_dev, msg); + if (del_window) + scif_queue_for_cleanup(window, &scif_info.rma); +} + +/** + * scif_recv_reg_ack: Respond to SCIF_REGISTER_ACK interrupt message + * @msg: Interrupt message + * + * Wake up the window waiting to complete registration. + */ +void scif_recv_reg_ack(struct scif_dev *scifdev, struct scifmsg *msg) +{ + struct scif_window *window = + (struct scif_window *)msg->payload[2]; + struct scif_endpt *ep = (struct scif_endpt *)window->ep; + + mutex_lock(&ep->rma_info.rma_lock); + window->reg_state = OP_COMPLETED; + wake_up(&window->regwq); + mutex_unlock(&ep->rma_info.rma_lock); +} + +/** + * scif_recv_reg_nack: Respond to SCIF_REGISTER_NACK interrupt message + * @msg: Interrupt message + * + * Wake up the window waiting to inform it that registration + * cannot be completed. + */ +void scif_recv_reg_nack(struct scif_dev *scifdev, struct scifmsg *msg) +{ + struct scif_window *window = + (struct scif_window *)msg->payload[2]; + struct scif_endpt *ep = (struct scif_endpt *)window->ep; + + mutex_lock(&ep->rma_info.rma_lock); + window->reg_state = OP_FAILED; + wake_up(&window->regwq); + mutex_unlock(&ep->rma_info.rma_lock); +} + +/** + * scif_recv_unreg_ack: Respond to SCIF_UNREGISTER_ACK interrupt message + * @msg: Interrupt message + * + * Wake up the window waiting to complete unregistration. + */ +void scif_recv_unreg_ack(struct scif_dev *scifdev, struct scifmsg *msg) +{ + struct scif_window *window = + (struct scif_window *)msg->payload[1]; + struct scif_endpt *ep = (struct scif_endpt *)window->ep; + + mutex_lock(&ep->rma_info.rma_lock); + window->unreg_state = OP_COMPLETED; + wake_up(&window->unregwq); + mutex_unlock(&ep->rma_info.rma_lock); +} + +/** + * scif_recv_unreg_nack: Respond to SCIF_UNREGISTER_NACK interrupt message + * @msg: Interrupt message + * + * Wake up the window waiting to inform it that unregistration + * cannot be completed immediately. + */ +void scif_recv_unreg_nack(struct scif_dev *scifdev, struct scifmsg *msg) +{ + struct scif_window *window = + (struct scif_window *)msg->payload[1]; + struct scif_endpt *ep = (struct scif_endpt *)window->ep; + + mutex_lock(&ep->rma_info.rma_lock); + window->unreg_state = OP_FAILED; + wake_up(&window->unregwq); + mutex_unlock(&ep->rma_info.rma_lock); +} + +int __scif_pin_pages(void *addr, size_t len, int *out_prot, + int map_flags, scif_pinned_pages_t *pages) +{ + struct scif_pinned_pages *pinned_pages; + int nr_pages, err = 0, i; + bool vmalloc_addr = false; + bool try_upgrade = false; + int prot = *out_prot; + int ulimit = 0; + struct mm_struct *mm = NULL; + + /* Unsupported flags */ + if (map_flags & ~(SCIF_MAP_KERNEL | SCIF_MAP_ULIMIT)) + return -EINVAL; + ulimit = !!(map_flags & SCIF_MAP_ULIMIT); + + /* Unsupported protection requested */ + if (prot & ~(SCIF_PROT_READ | SCIF_PROT_WRITE)) + return -EINVAL; + + /* addr/len must be page aligned. len should be non zero */ + if (!len || + (ALIGN((u64)addr, PAGE_SIZE) != (u64)addr) || + (ALIGN((u64)len, PAGE_SIZE) != (u64)len)) + return -EINVAL; + + might_sleep(); + + nr_pages = len >> PAGE_SHIFT; + + /* Allocate a set of pinned pages */ + pinned_pages = scif_create_pinned_pages(nr_pages, prot); + if (!pinned_pages) + return -ENOMEM; + + if (map_flags & SCIF_MAP_KERNEL) { + if (is_vmalloc_addr(addr)) + vmalloc_addr = true; + + for (i = 0; i < nr_pages; i++) { + if (vmalloc_addr) + pinned_pages->pages[i] = + vmalloc_to_page(addr + (i * PAGE_SIZE)); + else + pinned_pages->pages[i] = + virt_to_page(addr + (i * PAGE_SIZE)); + } + pinned_pages->nr_pages = nr_pages; + pinned_pages->map_flags = SCIF_MAP_KERNEL; + } else { + /* + * SCIF supports registration caching. If a registration has + * been requested with read only permissions, then we try + * to pin the pages with RW permissions so that a subsequent + * transfer with RW permission can hit the cache instead of + * invalidating it. If the upgrade fails with RW then we + * revert back to R permission and retry + */ + if (prot == SCIF_PROT_READ) + try_upgrade = true; + prot |= SCIF_PROT_WRITE; +retry: + mm = current->mm; + down_write(&mm->mmap_sem); + if (ulimit) { + err = __scif_check_inc_pinned_vm(mm, nr_pages); + if (err) { + up_write(&mm->mmap_sem); + pinned_pages->nr_pages = 0; + goto error_unmap; + } + } + + pinned_pages->nr_pages = get_user_pages( + current, + mm, + (u64)addr, + nr_pages, + !!(prot & SCIF_PROT_WRITE), + 0, + pinned_pages->pages, + NULL); + up_write(&mm->mmap_sem); + if (nr_pages != pinned_pages->nr_pages) { + if (try_upgrade) { + if (ulimit) + __scif_dec_pinned_vm_lock(mm, + nr_pages, 0); + /* Roll back any pinned pages */ + for (i = 0; i < pinned_pages->nr_pages; i++) { + if (pinned_pages->pages[i]) + put_page( + pinned_pages->pages[i]); + } + prot &= ~SCIF_PROT_WRITE; + try_upgrade = false; + goto retry; + } + } + pinned_pages->map_flags = 0; + } + + if (pinned_pages->nr_pages < nr_pages) { + err = -EFAULT; + pinned_pages->nr_pages = nr_pages; + goto dec_pinned; + } + + *out_prot = prot; + atomic_set(&pinned_pages->ref_count, 1); + *pages = pinned_pages; + return err; +dec_pinned: + if (ulimit) + __scif_dec_pinned_vm_lock(mm, nr_pages, 0); + /* Something went wrong! Rollback */ +error_unmap: + pinned_pages->nr_pages = nr_pages; + scif_destroy_pinned_pages(pinned_pages); + *pages = NULL; + dev_dbg(scif_info.mdev.this_device, + "%s %d err %d len 0x%lx\n", __func__, __LINE__, err, len); + return err; +} + +int scif_pin_pages(void *addr, size_t len, int prot, + int map_flags, scif_pinned_pages_t *pages) +{ + return __scif_pin_pages(addr, len, &prot, map_flags, pages); +} +EXPORT_SYMBOL_GPL(scif_pin_pages); + +int scif_unpin_pages(scif_pinned_pages_t pinned_pages) +{ + int err = 0, ret; + + if (!pinned_pages || SCIFEP_MAGIC != pinned_pages->magic) + return -EINVAL; + + ret = atomic_sub_return(1, &pinned_pages->ref_count); + if (ret < 0) { + dev_err(scif_info.mdev.this_device, + "%s %d scif_unpin_pages called without pinning? rc %d\n", + __func__, __LINE__, ret); + return -EINVAL; + } + /* + * Destroy the window if the ref count for this set of pinned + * pages has dropped to zero. If it is positive then there is + * a valid registered window which is backed by these pages and + * it will be destroyed once all such windows are unregistered. + */ + if (!ret) + err = scif_destroy_pinned_pages(pinned_pages); + + return err; +} +EXPORT_SYMBOL_GPL(scif_unpin_pages); + +static inline void +scif_insert_local_window(struct scif_window *window, struct scif_endpt *ep) +{ + mutex_lock(&ep->rma_info.rma_lock); + scif_insert_window(window, &ep->rma_info.reg_list); + mutex_unlock(&ep->rma_info.rma_lock); +} + +off_t scif_register_pinned_pages(scif_epd_t epd, + scif_pinned_pages_t pinned_pages, + off_t offset, int map_flags) +{ + struct scif_endpt *ep = (struct scif_endpt *)epd; + s64 computed_offset; + struct scif_window *window; + int err; + size_t len; + struct device *spdev; + + /* Unsupported flags */ + if (map_flags & ~SCIF_MAP_FIXED) + return -EINVAL; + + len = pinned_pages->nr_pages << PAGE_SHIFT; + + /* + * Offset is not page aligned/negative or offset+len + * wraps around with SCIF_MAP_FIXED. + */ + if ((map_flags & SCIF_MAP_FIXED) && + ((ALIGN(offset, PAGE_SIZE) != offset) || + (offset < 0) || + (offset + (off_t)len < offset))) + return -EINVAL; + + might_sleep(); + + err = scif_verify_epd(ep); + if (err) + return err; + /* + * It is an error to pass pinned_pages to scif_register_pinned_pages() + * after calling scif_unpin_pages(). + */ + if (!atomic_add_unless(&pinned_pages->ref_count, 1, 0)) + return -EINVAL; + + /* Compute the offset for this registration */ + err = scif_get_window_offset(ep, map_flags, offset, + len, &computed_offset); + if (err) { + atomic_sub(1, &pinned_pages->ref_count); + return err; + } + + /* Allocate and prepare self registration window */ + window = scif_create_window(ep, pinned_pages->nr_pages, + computed_offset, false); + if (!window) { + atomic_sub(1, &pinned_pages->ref_count); + scif_free_window_offset(ep, NULL, computed_offset); + return -ENOMEM; + } + + window->pinned_pages = pinned_pages; + window->nr_pages = pinned_pages->nr_pages; + window->prot = pinned_pages->prot; + + spdev = scif_get_peer_dev(ep->remote_dev); + if (IS_ERR(spdev)) { + err = PTR_ERR(spdev); + scif_destroy_window(ep, window); + return err; + } + err = scif_send_alloc_request(ep, window); + if (err) { + dev_err(&ep->remote_dev->sdev->dev, + "%s %d err %d\n", __func__, __LINE__, err); + goto error_unmap; + } + + /* Prepare the remote registration window */ + err = scif_prep_remote_window(ep, window); + if (err) { + dev_err(&ep->remote_dev->sdev->dev, + "%s %d err %d\n", __func__, __LINE__, err); + goto error_unmap; + } + + /* Tell the peer about the new window */ + err = scif_send_scif_register(ep, window); + if (err) { + dev_err(&ep->remote_dev->sdev->dev, + "%s %d err %d\n", __func__, __LINE__, err); + goto error_unmap; + } + + scif_put_peer_dev(spdev); + /* No further failures expected. Insert new window */ + scif_insert_local_window(window, ep); + return computed_offset; +error_unmap: + scif_destroy_window(ep, window); + scif_put_peer_dev(spdev); + dev_err(&ep->remote_dev->sdev->dev, + "%s %d err %d\n", __func__, __LINE__, err); + return err; +} +EXPORT_SYMBOL_GPL(scif_register_pinned_pages); + +off_t scif_register(scif_epd_t epd, void *addr, size_t len, off_t offset, + int prot, int map_flags) +{ + scif_pinned_pages_t pinned_pages; + off_t err; + struct scif_endpt *ep = (struct scif_endpt *)epd; + s64 computed_offset; + struct scif_window *window; + struct mm_struct *mm = NULL; + struct device *spdev; + + dev_dbg(scif_info.mdev.this_device, + "SCIFAPI register: ep %p addr %p len 0x%lx offset 0x%lx prot 0x%x map_flags 0x%x\n", + epd, addr, len, offset, prot, map_flags); + /* Unsupported flags */ + if (map_flags & ~(SCIF_MAP_FIXED | SCIF_MAP_KERNEL)) + return -EINVAL; + + /* + * Offset is not page aligned/negative or offset+len + * wraps around with SCIF_MAP_FIXED. + */ + if ((map_flags & SCIF_MAP_FIXED) && + ((ALIGN(offset, PAGE_SIZE) != offset) || + (offset < 0) || + (offset + (off_t)len < offset))) + return -EINVAL; + + /* Unsupported protection requested */ + if (prot & ~(SCIF_PROT_READ | SCIF_PROT_WRITE)) + return -EINVAL; + + /* addr/len must be page aligned. len should be non zero */ + if (!len || (ALIGN((u64)addr, PAGE_SIZE) != (u64)addr) || + (ALIGN(len, PAGE_SIZE) != len)) + return -EINVAL; + + might_sleep(); + + err = scif_verify_epd(ep); + if (err) + return err; + + /* Compute the offset for this registration */ + err = scif_get_window_offset(ep, map_flags, offset, + len >> PAGE_SHIFT, &computed_offset); + if (err) + return err; + + spdev = scif_get_peer_dev(ep->remote_dev); + if (IS_ERR(spdev)) { + err = PTR_ERR(spdev); + scif_free_window_offset(ep, NULL, computed_offset); + return err; + } + /* Allocate and prepare self registration window */ + window = scif_create_window(ep, len >> PAGE_SHIFT, + computed_offset, false); + if (!window) { + scif_free_window_offset(ep, NULL, computed_offset); + scif_put_peer_dev(spdev); + return -ENOMEM; + } + + window->nr_pages = len >> PAGE_SHIFT; + + err = scif_send_alloc_request(ep, window); + if (err) { + scif_destroy_incomplete_window(ep, window); + scif_put_peer_dev(spdev); + return err; + } + + if (!(map_flags & SCIF_MAP_KERNEL)) { + mm = __scif_acquire_mm(); + map_flags |= SCIF_MAP_ULIMIT; + } + /* Pin down the pages */ + err = __scif_pin_pages(addr, len, &prot, + map_flags & (SCIF_MAP_KERNEL | SCIF_MAP_ULIMIT), + &pinned_pages); + if (err) { + scif_destroy_incomplete_window(ep, window); + __scif_release_mm(mm); + goto error; + } + + window->pinned_pages = pinned_pages; + window->prot = pinned_pages->prot; + window->mm = mm; + + /* Prepare the remote registration window */ + err = scif_prep_remote_window(ep, window); + if (err) { + dev_err(&ep->remote_dev->sdev->dev, + "%s %d err %ld\n", __func__, __LINE__, err); + goto error_unmap; + } + + /* Tell the peer about the new window */ + err = scif_send_scif_register(ep, window); + if (err) { + dev_err(&ep->remote_dev->sdev->dev, + "%s %d err %ld\n", __func__, __LINE__, err); + goto error_unmap; + } + + scif_put_peer_dev(spdev); + /* No further failures expected. Insert new window */ + scif_insert_local_window(window, ep); + dev_dbg(&ep->remote_dev->sdev->dev, + "SCIFAPI register: ep %p addr %p len 0x%lx computed_offset 0x%llx\n", + epd, addr, len, computed_offset); + return computed_offset; +error_unmap: + scif_destroy_window(ep, window); +error: + scif_put_peer_dev(spdev); + dev_err(&ep->remote_dev->sdev->dev, + "%s %d err %ld\n", __func__, __LINE__, err); + return err; +} +EXPORT_SYMBOL_GPL(scif_register); + +int +scif_unregister(scif_epd_t epd, off_t offset, size_t len) +{ + struct scif_endpt *ep = (struct scif_endpt *)epd; + struct scif_window *window = NULL; + struct scif_rma_req req; + int nr_pages, err; + struct device *spdev; + + dev_dbg(scif_info.mdev.this_device, + "SCIFAPI unregister: ep %p offset 0x%lx len 0x%lx\n", + ep, offset, len); + /* len must be page aligned. len should be non zero */ + if (!len || + (ALIGN((u64)len, PAGE_SIZE) != (u64)len)) + return -EINVAL; + + /* Offset is not page aligned or offset+len wraps around */ + if ((ALIGN(offset, PAGE_SIZE) != offset) || + (offset + (off_t)len < offset)) + return -EINVAL; + + err = scif_verify_epd(ep); + if (err) + return err; + + might_sleep(); + nr_pages = len >> PAGE_SHIFT; + + req.out_window = &window; + req.offset = offset; + req.prot = 0; + req.nr_bytes = len; + req.type = SCIF_WINDOW_FULL; + req.head = &ep->rma_info.reg_list; + + spdev = scif_get_peer_dev(ep->remote_dev); + if (IS_ERR(spdev)) { + err = PTR_ERR(spdev); + return err; + } + mutex_lock(&ep->rma_info.rma_lock); + /* Does a valid window exist? */ + err = scif_query_window(&req); + if (err) { + dev_err(&ep->remote_dev->sdev->dev, + "%s %d err %d\n", __func__, __LINE__, err); + goto error; + } + /* Unregister all the windows in this range */ + err = scif_rma_list_unregister(window, offset, nr_pages); + if (err) + dev_err(&ep->remote_dev->sdev->dev, + "%s %d err %d\n", __func__, __LINE__, err); +error: + mutex_unlock(&ep->rma_info.rma_lock); + scif_put_peer_dev(spdev); + return err; +} +EXPORT_SYMBOL_GPL(scif_unregister); diff --git a/drivers/misc/mic/scif/scif_rma.h b/drivers/misc/mic/scif/scif_rma.h new file mode 100644 index 000000000000..fa6722279196 --- /dev/null +++ b/drivers/misc/mic/scif/scif_rma.h @@ -0,0 +1,464 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * This file is provided under a dual BSD/GPLv2 license. When using or + * redistributing this file, you may do so under either license. + * + * GPL LICENSE SUMMARY + * + * Copyright(c) 2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * BSD LICENSE + * + * Copyright(c) 2015 Intel Corporation. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in + * the documentation and/or other materials provided with the + * distribution. + * * Neither the name of Intel Corporation nor the names of its + * contributors may be used to endorse or promote products derived + * from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS + * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT + * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR + * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT + * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, + * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY + * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE + * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * Intel SCIF driver. + * + */ +#ifndef SCIF_RMA_H +#define SCIF_RMA_H + +#include <linux/dma_remapping.h> +#include <linux/mmu_notifier.h> + +#include "../bus/scif_bus.h" + +/* If this bit is set then the mark is a remote fence mark */ +#define SCIF_REMOTE_FENCE_BIT 31 +/* Magic value used to indicate a remote fence request */ +#define SCIF_REMOTE_FENCE BIT_ULL(SCIF_REMOTE_FENCE_BIT) + +#define SCIF_MAX_UNALIGNED_BUF_SIZE (1024 * 1024ULL) +#define SCIF_KMEM_UNALIGNED_BUF_SIZE (SCIF_MAX_UNALIGNED_BUF_SIZE + \ + (L1_CACHE_BYTES << 1)) + +#define SCIF_IOVA_START_PFN (1) +#define SCIF_IOVA_PFN(addr) ((addr) >> PAGE_SHIFT) +#define SCIF_DMA_64BIT_PFN SCIF_IOVA_PFN(DMA_BIT_MASK(64)) +#define SCIF_DMA_63BIT_PFN SCIF_IOVA_PFN(DMA_BIT_MASK(63)) + +/* + * struct scif_endpt_rma_info - Per Endpoint Remote Memory Access Information + * + * @reg_list: List of registration windows for self + * @remote_reg_list: List of registration windows for peer + * @iovad: Offset generator + * @rma_lock: Synchronizes access to self/remote list and also protects the + * window from being destroyed while RMAs are in progress. + * @tc_lock: Synchronizes access to temporary cached windows list + * for SCIF Registration Caching. + * @mmn_lock: Synchronizes access to the list of MMU notifiers registered + * @tw_refcount: Keeps track of number of outstanding temporary registered + * windows created by scif_vreadfrom/scif_vwriteto which have + * not been destroyed. + * @tcw_refcount: Same as tw_refcount but for temporary cached windows + * @tcw_total_pages: Same as tcw_refcount but in terms of pages pinned + * @mmn_list: MMU notifier so that we can destroy the windows when required + * @fence_refcount: Keeps track of number of outstanding remote fence + * requests which have been received by the peer. + * @dma_chan: DMA channel used for all DMA transfers for this endpoint. + * @async_list_del: Detect asynchronous list entry deletion + * @vma_list: List of vmas with remote memory mappings + * @markwq: Wait queue used for scif_fence_mark/scif_fence_wait +*/ +struct scif_endpt_rma_info { + struct list_head reg_list; + struct list_head remote_reg_list; + struct iova_domain iovad; + struct mutex rma_lock; + spinlock_t tc_lock; + struct mutex mmn_lock; + atomic_t tw_refcount; + atomic_t tcw_refcount; + atomic_t tcw_total_pages; + struct list_head mmn_list; + atomic_t fence_refcount; + struct dma_chan *dma_chan; + int async_list_del; + struct list_head vma_list; + wait_queue_head_t markwq; +}; + +/* + * struct scif_fence_info - used for tracking fence requests + * + * @state: State of this transfer + * @wq: Fences wait on this queue + * @dma_mark: Used for storing the DMA mark + */ +struct scif_fence_info { + enum scif_msg_state state; + struct completion comp; + int dma_mark; +}; + +/* + * struct scif_remote_fence_info - used for tracking remote fence requests + * + * @msg: List of SCIF node QP fence messages + * @list: Link to list of remote fence requests + */ +struct scif_remote_fence_info { + struct scifmsg msg; + struct list_head list; +}; + +/* + * Specifies whether an RMA operation can span across partial windows, a single + * window or multiple contiguous windows. Mmaps can span across partial windows. + * Unregistration can span across complete windows. scif_get_pages() can span a + * single window. A window can also be of type self or peer. + */ +enum scif_window_type { + SCIF_WINDOW_PARTIAL, + SCIF_WINDOW_SINGLE, + SCIF_WINDOW_FULL, + SCIF_WINDOW_SELF, + SCIF_WINDOW_PEER +}; + +/* The number of physical addresses that can be stored in a PAGE. */ +#define SCIF_NR_ADDR_IN_PAGE (0x1000 >> 3) + +/* + * struct scif_rma_lookup - RMA lookup data structure for page list transfers + * + * Store an array of lookup offsets. Each offset in this array maps + * one 4K page containing 512 physical addresses i.e. 2MB. 512 such + * offsets in a 4K page will correspond to 1GB of registered address space. + + * @lookup: Array of offsets + * @offset: DMA offset of lookup array + */ +struct scif_rma_lookup { + dma_addr_t *lookup; + dma_addr_t offset; +}; + +/* + * struct scif_pinned_pages - A set of pinned pages obtained with + * scif_pin_pages() which could be part of multiple registered + * windows across different end points. + * + * @nr_pages: Number of pages which is defined as a s64 instead of an int + * to avoid sign extension with buffers >= 2GB + * @prot: read/write protections + * @map_flags: Flags specified during the pin operation + * @ref_count: Reference count bumped in terms of number of pages + * @magic: A magic value + * @pages: Array of pointers to struct pages populated with get_user_pages(..) + */ +struct scif_pinned_pages { + s64 nr_pages; + int prot; + int map_flags; + atomic_t ref_count; + u64 magic; + struct page **pages; +}; + +/* + * struct scif_status - Stores DMA status update information + * + * @src_dma_addr: Source buffer DMA address + * @val: src location for value to be written to the destination + * @ep: SCIF endpoint + */ +struct scif_status { + dma_addr_t src_dma_addr; + u64 val; + struct scif_endpt *ep; +}; + +/* + * struct scif_window - Registration Window for Self and Remote + * + * @nr_pages: Number of pages which is defined as a s64 instead of an int + * to avoid sign extension with buffers >= 2GB + * @nr_contig_chunks: Number of contiguous physical chunks + * @prot: read/write protections + * @ref_count: reference count in terms of number of pages + * @magic: Cookie to detect corruption + * @offset: registered offset + * @va_for_temp: va address that this window represents + * @dma_mark: Used to determine if all DMAs against the window are done + * @ep: Pointer to EP. Useful for passing EP around with messages to + avoid expensive list traversals. + * @list: link to list of windows for the endpoint + * @type: self or peer window + * @peer_window: Pointer to peer window. Useful for sending messages to peer + * without requiring an extra list traversal + * @unreg_state: unregistration state + * @offset_freed: True if the offset has been freed + * @temp: True for temporary windows created via scif_vreadfrom/scif_vwriteto + * @mm: memory descriptor for the task_struct which initiated the RMA + * @st: scatter gather table for DMA mappings with IOMMU enabled + * @pinned_pages: The set of pinned_pages backing this window + * @alloc_handle: Handle for sending ALLOC_REQ + * @regwq: Wait Queue for an registration (N)ACK + * @reg_state: Registration state + * @unregwq: Wait Queue for an unregistration (N)ACK + * @dma_addr_lookup: Lookup for physical addresses used for DMA + * @nr_lookup: Number of entries in lookup + * @mapped_offset: Offset used to map the window by the peer + * @dma_addr: Array of physical addresses used for Mgmt node & MIC initiated DMA + * @num_pages: Array specifying number of pages for each physical address + */ +struct scif_window { + s64 nr_pages; + int nr_contig_chunks; + int prot; + int ref_count; + u64 magic; + s64 offset; + unsigned long va_for_temp; + int dma_mark; + u64 ep; + struct list_head list; + enum scif_window_type type; + u64 peer_window; + enum scif_msg_state unreg_state; + bool offset_freed; + bool temp; + struct mm_struct *mm; + struct sg_table *st; + union { + struct { + struct scif_pinned_pages *pinned_pages; + struct scif_allocmsg alloc_handle; + wait_queue_head_t regwq; + enum scif_msg_state reg_state; + wait_queue_head_t unregwq; + }; + struct { + struct scif_rma_lookup dma_addr_lookup; + struct scif_rma_lookup num_pages_lookup; + int nr_lookup; + dma_addr_t mapped_offset; + }; + }; + dma_addr_t *dma_addr; + u64 *num_pages; +} __packed; + +/* + * scif_mmu_notif - SCIF mmu notifier information + * + * @mmu_notifier ep_mmu_notifier: MMU notifier operations + * @tc_reg_list: List of temp registration windows for self + * @mm: memory descriptor for the task_struct which initiated the RMA + * @ep: SCIF endpoint + * @list: link to list of MMU notifier information + */ +struct scif_mmu_notif { +#ifdef CONFIG_MMU_NOTIFIER + struct mmu_notifier ep_mmu_notifier; +#endif + struct list_head tc_reg_list; + struct mm_struct *mm; + struct scif_endpt *ep; + struct list_head list; +}; + +enum scif_rma_dir { + SCIF_LOCAL_TO_REMOTE, + SCIF_REMOTE_TO_LOCAL +}; + +extern struct kmem_cache *unaligned_cache; +/* Initialize RMA for this EP */ +void scif_rma_ep_init(struct scif_endpt *ep); +/* Check if epd can be uninitialized */ +int scif_rma_ep_can_uninit(struct scif_endpt *ep); +/* Obtain a new offset. Callee must grab RMA lock */ +int scif_get_window_offset(struct scif_endpt *ep, int flags, + s64 offset, int nr_pages, s64 *out_offset); +/* Free offset. Callee must grab RMA lock */ +void scif_free_window_offset(struct scif_endpt *ep, + struct scif_window *window, s64 offset); +/* Create self registration window */ +struct scif_window *scif_create_window(struct scif_endpt *ep, int nr_pages, + s64 offset, bool temp); +/* Destroy self registration window.*/ +int scif_destroy_window(struct scif_endpt *ep, struct scif_window *window); +void scif_unmap_window(struct scif_dev *remote_dev, struct scif_window *window); +/* Map pages of self window to Aperture/PCI */ +int scif_map_window(struct scif_dev *remote_dev, + struct scif_window *window); +/* Unregister a self window */ +int scif_unregister_window(struct scif_window *window); +/* Destroy remote registration window */ +void +scif_destroy_remote_window(struct scif_window *window); +/* remove valid remote memory mappings from process address space */ +void scif_zap_mmaps(int node); +/* Query if any applications have remote memory mappings */ +bool scif_rma_do_apps_have_mmaps(int node); +/* Cleanup remote registration lists for zombie endpoints */ +void scif_cleanup_rma_for_zombies(int node); +/* Reserve a DMA channel for a particular endpoint */ +int scif_reserve_dma_chan(struct scif_endpt *ep); +/* Setup a DMA mark for an endpoint */ +int _scif_fence_mark(scif_epd_t epd, int *mark); +int scif_prog_signal(scif_epd_t epd, off_t offset, u64 val, + enum scif_window_type type); +void scif_alloc_req(struct scif_dev *scifdev, struct scifmsg *msg); +void scif_alloc_gnt_rej(struct scif_dev *scifdev, struct scifmsg *msg); +void scif_free_virt(struct scif_dev *scifdev, struct scifmsg *msg); +void scif_recv_reg(struct scif_dev *scifdev, struct scifmsg *msg); +void scif_recv_unreg(struct scif_dev *scifdev, struct scifmsg *msg); +void scif_recv_reg_ack(struct scif_dev *scifdev, struct scifmsg *msg); +void scif_recv_reg_nack(struct scif_dev *scifdev, struct scifmsg *msg); +void scif_recv_unreg_ack(struct scif_dev *scifdev, struct scifmsg *msg); +void scif_recv_unreg_nack(struct scif_dev *scifdev, struct scifmsg *msg); +void scif_recv_munmap(struct scif_dev *scifdev, struct scifmsg *msg); +void scif_recv_mark(struct scif_dev *scifdev, struct scifmsg *msg); +void scif_recv_mark_resp(struct scif_dev *scifdev, struct scifmsg *msg); +void scif_recv_wait(struct scif_dev *scifdev, struct scifmsg *msg); +void scif_recv_wait_resp(struct scif_dev *scifdev, struct scifmsg *msg); +void scif_recv_sig_local(struct scif_dev *scifdev, struct scifmsg *msg); +void scif_recv_sig_remote(struct scif_dev *scifdev, struct scifmsg *msg); +void scif_recv_sig_resp(struct scif_dev *scifdev, struct scifmsg *msg); +void scif_mmu_notif_handler(struct work_struct *work); +void scif_rma_handle_remote_fences(void); +void scif_rma_destroy_windows(void); +void scif_rma_destroy_tcw_invalid(void); +int scif_drain_dma_intr(struct scif_hw_dev *sdev, struct dma_chan *chan); + +struct scif_window_iter { + s64 offset; + int index; +}; + +static inline void +scif_init_window_iter(struct scif_window *window, struct scif_window_iter *iter) +{ + iter->offset = window->offset; + iter->index = 0; +} + +dma_addr_t scif_off_to_dma_addr(struct scif_window *window, s64 off, + size_t *nr_bytes, + struct scif_window_iter *iter); +static inline +dma_addr_t __scif_off_to_dma_addr(struct scif_window *window, s64 off) +{ + return scif_off_to_dma_addr(window, off, NULL, NULL); +} + +static inline bool scif_unaligned(off_t src_offset, off_t dst_offset) +{ + src_offset = src_offset & (L1_CACHE_BYTES - 1); + dst_offset = dst_offset & (L1_CACHE_BYTES - 1); + return !(src_offset == dst_offset); +} + +/* + * scif_zalloc: + * @size: Size of the allocation request. + * + * Helper API which attempts to allocate zeroed pages via + * __get_free_pages(..) first and then falls back on + * vzalloc(..) if that fails. + */ +static inline void *scif_zalloc(size_t size) +{ + void *ret = NULL; + size_t align = ALIGN(size, PAGE_SIZE); + + if (align && get_order(align) < MAX_ORDER) + ret = (void *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, + get_order(align)); + return ret ? ret : vzalloc(align); +} + +/* + * scif_free: + * @addr: Address to be freed. + * @size: Size of the allocation. + * Helper API which frees memory allocated via scif_zalloc(). + */ +static inline void scif_free(void *addr, size_t size) +{ + size_t align = ALIGN(size, PAGE_SIZE); + + if (is_vmalloc_addr(addr)) + vfree(addr); + else + free_pages((unsigned long)addr, get_order(align)); +} + +static inline void scif_get_window(struct scif_window *window, int nr_pages) +{ + window->ref_count += nr_pages; +} + +static inline void scif_put_window(struct scif_window *window, int nr_pages) +{ + window->ref_count -= nr_pages; +} + +static inline void scif_set_window_ref(struct scif_window *window, int nr_pages) +{ + window->ref_count = nr_pages; +} + +static inline void +scif_queue_for_cleanup(struct scif_window *window, struct list_head *list) +{ + spin_lock(&scif_info.rmalock); + list_add_tail(&window->list, list); + spin_unlock(&scif_info.rmalock); + schedule_work(&scif_info.misc_work); +} + +static inline void __scif_rma_destroy_tcw_helper(struct scif_window *window) +{ + list_del_init(&window->list); + scif_queue_for_cleanup(window, &scif_info.rma_tc); +} + +static inline bool scif_is_iommu_enabled(void) +{ +#ifdef CONFIG_INTEL_IOMMU + return intel_iommu_enabled; +#else + return false; +#endif +} +#endif /* SCIF_RMA_H */ diff --git a/drivers/misc/mic/scif/scif_rma_list.c b/drivers/misc/mic/scif/scif_rma_list.c new file mode 100644 index 000000000000..e1ef8daedd5a --- /dev/null +++ b/drivers/misc/mic/scif/scif_rma_list.c @@ -0,0 +1,291 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * Intel SCIF driver. + * + */ +#include "scif_main.h" +#include <linux/mmu_notifier.h> +#include <linux/highmem.h> + +/* + * scif_insert_tcw: + * + * Insert a temp window to the temp registration list sorted by va_for_temp. + * RMA lock must be held. + */ +void scif_insert_tcw(struct scif_window *window, struct list_head *head) +{ + struct scif_window *curr = NULL; + struct scif_window *prev = list_entry(head, struct scif_window, list); + struct list_head *item; + + INIT_LIST_HEAD(&window->list); + /* Compare with tail and if the entry is new tail add it to the end */ + if (!list_empty(head)) { + curr = list_entry(head->prev, struct scif_window, list); + if (curr->va_for_temp < window->va_for_temp) { + list_add_tail(&window->list, head); + return; + } + } + list_for_each(item, head) { + curr = list_entry(item, struct scif_window, list); + if (curr->va_for_temp > window->va_for_temp) + break; + prev = curr; + } + list_add(&window->list, &prev->list); +} + +/* + * scif_insert_window: + * + * Insert a window to the self registration list sorted by offset. + * RMA lock must be held. + */ +void scif_insert_window(struct scif_window *window, struct list_head *head) +{ + struct scif_window *curr = NULL, *prev = NULL; + struct list_head *item; + + INIT_LIST_HEAD(&window->list); + list_for_each(item, head) { + curr = list_entry(item, struct scif_window, list); + if (curr->offset > window->offset) + break; + prev = curr; + } + if (!prev) + list_add(&window->list, head); + else + list_add(&window->list, &prev->list); + scif_set_window_ref(window, window->nr_pages); +} + +/* + * scif_query_tcw: + * + * Query the temp cached registration list of ep for an overlapping window + * in case of permission mismatch, destroy the previous window. if permissions + * match and overlap is partial, destroy the window but return the new range + * RMA lock must be held. + */ +int scif_query_tcw(struct scif_endpt *ep, struct scif_rma_req *req) +{ + struct list_head *item, *temp, *head = req->head; + struct scif_window *window; + u64 start_va_window, start_va_req = req->va_for_temp; + u64 end_va_window, end_va_req = start_va_req + req->nr_bytes; + + if (!req->nr_bytes) + return -EINVAL; + /* + * Avoid traversing the entire list to find out that there + * is no entry that matches + */ + if (!list_empty(head)) { + window = list_last_entry(head, struct scif_window, list); + end_va_window = window->va_for_temp + + (window->nr_pages << PAGE_SHIFT); + if (start_va_req > end_va_window) + return -ENXIO; + } + list_for_each_safe(item, temp, head) { + window = list_entry(item, struct scif_window, list); + start_va_window = window->va_for_temp; + end_va_window = window->va_for_temp + + (window->nr_pages << PAGE_SHIFT); + if (start_va_req < start_va_window && + end_va_req < start_va_window) + break; + if (start_va_req >= end_va_window) + continue; + if ((window->prot & req->prot) == req->prot) { + if (start_va_req >= start_va_window && + end_va_req <= end_va_window) { + *req->out_window = window; + return 0; + } + /* expand window */ + if (start_va_req < start_va_window) { + req->nr_bytes += + start_va_window - start_va_req; + req->va_for_temp = start_va_window; + } + if (end_va_req >= end_va_window) + req->nr_bytes += end_va_window - end_va_req; + } + /* Destroy the old window to create a new one */ + __scif_rma_destroy_tcw_helper(window); + break; + } + return -ENXIO; +} + +/* + * scif_query_window: + * + * Query the registration list and check if a valid contiguous + * range of windows exist. + * RMA lock must be held. + */ +int scif_query_window(struct scif_rma_req *req) +{ + struct list_head *item; + struct scif_window *window; + s64 end_offset, offset = req->offset; + u64 tmp_min, nr_bytes_left = req->nr_bytes; + + if (!req->nr_bytes) + return -EINVAL; + + list_for_each(item, req->head) { + window = list_entry(item, struct scif_window, list); + end_offset = window->offset + + (window->nr_pages << PAGE_SHIFT); + if (offset < window->offset) + /* Offset not found! */ + return -ENXIO; + if (offset >= end_offset) + continue; + /* Check read/write protections. */ + if ((window->prot & req->prot) != req->prot) + return -EPERM; + if (nr_bytes_left == req->nr_bytes) + /* Store the first window */ + *req->out_window = window; + tmp_min = min((u64)end_offset - offset, nr_bytes_left); + nr_bytes_left -= tmp_min; + offset += tmp_min; + /* + * Range requested encompasses + * multiple windows contiguously. + */ + if (!nr_bytes_left) { + /* Done for partial window */ + if (req->type == SCIF_WINDOW_PARTIAL || + req->type == SCIF_WINDOW_SINGLE) + return 0; + /* Extra logic for full windows */ + if (offset == end_offset) + /* Spanning multiple whole windows */ + return 0; + /* Not spanning multiple whole windows */ + return -ENXIO; + } + if (req->type == SCIF_WINDOW_SINGLE) + break; + } + dev_err(scif_info.mdev.this_device, + "%s %d ENXIO\n", __func__, __LINE__); + return -ENXIO; +} + +/* + * scif_rma_list_unregister: + * + * Traverse the self registration list starting from window: + * 1) Call scif_unregister_window(..) + * RMA lock must be held. + */ +int scif_rma_list_unregister(struct scif_window *window, + s64 offset, int nr_pages) +{ + struct scif_endpt *ep = (struct scif_endpt *)window->ep; + struct list_head *head = &ep->rma_info.reg_list; + s64 end_offset; + int err = 0; + int loop_nr_pages; + struct scif_window *_window; + + list_for_each_entry_safe_from(window, _window, head, list) { + end_offset = window->offset + (window->nr_pages << PAGE_SHIFT); + loop_nr_pages = min((int)((end_offset - offset) >> PAGE_SHIFT), + nr_pages); + err = scif_unregister_window(window); + if (err) + return err; + nr_pages -= loop_nr_pages; + offset += (loop_nr_pages << PAGE_SHIFT); + if (!nr_pages) + break; + } + return 0; +} + +/* + * scif_unmap_all_window: + * + * Traverse all the windows in the self registration list and: + * 1) Delete any DMA mappings created + */ +void scif_unmap_all_windows(scif_epd_t epd) +{ + struct list_head *item, *tmp; + struct scif_window *window; + struct scif_endpt *ep = (struct scif_endpt *)epd; + struct list_head *head = &ep->rma_info.reg_list; + + mutex_lock(&ep->rma_info.rma_lock); + list_for_each_safe(item, tmp, head) { + window = list_entry(item, struct scif_window, list); + scif_unmap_window(ep->remote_dev, window); + } + mutex_unlock(&ep->rma_info.rma_lock); +} + +/* + * scif_unregister_all_window: + * + * Traverse all the windows in the self registration list and: + * 1) Call scif_unregister_window(..) + * RMA lock must be held. + */ +int scif_unregister_all_windows(scif_epd_t epd) +{ + struct list_head *item, *tmp; + struct scif_window *window; + struct scif_endpt *ep = (struct scif_endpt *)epd; + struct list_head *head = &ep->rma_info.reg_list; + int err = 0; + + mutex_lock(&ep->rma_info.rma_lock); +retry: + item = NULL; + tmp = NULL; + list_for_each_safe(item, tmp, head) { + window = list_entry(item, struct scif_window, list); + ep->rma_info.async_list_del = 0; + err = scif_unregister_window(window); + if (err) + dev_err(scif_info.mdev.this_device, + "%s %d err %d\n", + __func__, __LINE__, err); + /* + * Need to restart list traversal if there has been + * an asynchronous list entry deletion. + */ + if (ACCESS_ONCE(ep->rma_info.async_list_del)) + goto retry; + } + mutex_unlock(&ep->rma_info.rma_lock); + if (!list_empty(&ep->rma_info.mmn_list)) { + spin_lock(&scif_info.rmalock); + list_add_tail(&ep->mmu_list, &scif_info.mmu_notif_cleanup); + spin_unlock(&scif_info.rmalock); + schedule_work(&scif_info.mmu_notif_work); + } + return err; +} diff --git a/drivers/misc/mic/scif/scif_rma_list.h b/drivers/misc/mic/scif/scif_rma_list.h new file mode 100644 index 000000000000..7d58d1d551b0 --- /dev/null +++ b/drivers/misc/mic/scif/scif_rma_list.h @@ -0,0 +1,57 @@ +/* + * Intel MIC Platform Software Stack (MPSS) + * + * Copyright(c) 2015 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but + * WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * General Public License for more details. + * + * Intel SCIF driver. + * + */ +#ifndef SCIF_RMA_LIST_H +#define SCIF_RMA_LIST_H + +/* + * struct scif_rma_req - Self Registration list RMA Request query + * + * @out_window - Returns the window if found + * @offset: Starting offset + * @nr_bytes: number of bytes + * @prot: protection requested i.e. read or write or both + * @type: Specify single, partial or multiple windows + * @head: Head of list on which to search + * @va_for_temp: VA for searching temporary cached windows + */ +struct scif_rma_req { + struct scif_window **out_window; + union { + s64 offset; + unsigned long va_for_temp; + }; + size_t nr_bytes; + int prot; + enum scif_window_type type; + struct list_head *head; +}; + +/* Insert */ +void scif_insert_window(struct scif_window *window, struct list_head *head); +void scif_insert_tcw(struct scif_window *window, + struct list_head *head); +/* Query */ +int scif_query_window(struct scif_rma_req *request); +int scif_query_tcw(struct scif_endpt *ep, struct scif_rma_req *request); +/* Called from close to unregister all self windows */ +int scif_unregister_all_windows(scif_epd_t epd); +void scif_unmap_all_windows(scif_epd_t epd); +/* Traverse list and unregister */ +int scif_rma_list_unregister(struct scif_window *window, s64 offset, + int nr_pages); +#endif /* SCIF_RMA_LIST_H */ diff --git a/drivers/misc/sgi-gru/gruhandles.c b/drivers/misc/sgi-gru/gruhandles.c index 2f30badc6ffd..1ee8e82ba710 100644 --- a/drivers/misc/sgi-gru/gruhandles.c +++ b/drivers/misc/sgi-gru/gruhandles.c @@ -196,12 +196,6 @@ void tfh_write_restart(struct gru_tlb_fault_handle *tfh, start_instruction(tfh); } -void tfh_restart(struct gru_tlb_fault_handle *tfh) -{ - tfh->opc = TFHOP_RESTART; - start_instruction(tfh); -} - void tfh_user_polling_mode(struct gru_tlb_fault_handle *tfh) { tfh->opc = TFHOP_USER_POLLING_MODE; diff --git a/drivers/misc/sgi-gru/gruhandles.h b/drivers/misc/sgi-gru/gruhandles.h index 3f998b924d8f..3d7bd36a1c89 100644 --- a/drivers/misc/sgi-gru/gruhandles.h +++ b/drivers/misc/sgi-gru/gruhandles.h @@ -524,7 +524,6 @@ int tfh_write_only(struct gru_tlb_fault_handle *tfh, unsigned long paddr, int gaa, unsigned long vaddr, int asid, int dirty, int pagesize); void tfh_write_restart(struct gru_tlb_fault_handle *tfh, unsigned long paddr, int gaa, unsigned long vaddr, int asid, int dirty, int pagesize); -void tfh_restart(struct gru_tlb_fault_handle *tfh); void tfh_user_polling_mode(struct gru_tlb_fault_handle *tfh); void tfh_exception(struct gru_tlb_fault_handle *tfh); diff --git a/drivers/misc/sgi-gru/grukdump.c b/drivers/misc/sgi-gru/grukdump.c index a3700a56b8ff..313da3150262 100644 --- a/drivers/misc/sgi-gru/grukdump.c +++ b/drivers/misc/sgi-gru/grukdump.c @@ -78,11 +78,10 @@ static int gru_dump_tfm(struct gru_state *gru, void __user *ubuf, void __user *ubufend) { struct gru_tlb_fault_map *tfm; - int i, ret, bytes; + int i; - bytes = GRU_NUM_TFM * GRU_CACHE_LINE_BYTES; - if (bytes > ubufend - ubuf) - ret = -EFBIG; + if (GRU_NUM_TFM * GRU_CACHE_LINE_BYTES > ubufend - ubuf) + return -EFBIG; for (i = 0; i < GRU_NUM_TFM; i++) { tfm = get_tfm(gru->gs_gru_base_vaddr, i); @@ -99,11 +98,10 @@ static int gru_dump_tgh(struct gru_state *gru, void __user *ubuf, void __user *ubufend) { struct gru_tlb_global_handle *tgh; - int i, ret, bytes; + int i; - bytes = GRU_NUM_TGH * GRU_CACHE_LINE_BYTES; - if (bytes > ubufend - ubuf) - ret = -EFBIG; + if (GRU_NUM_TGH * GRU_CACHE_LINE_BYTES > ubufend - ubuf) + return -EFBIG; for (i = 0; i < GRU_NUM_TGH; i++) { tgh = get_tgh(gru->gs_gru_base_vaddr, i); @@ -196,7 +194,7 @@ int gru_dump_chiplet_request(unsigned long arg) return -EFAULT; /* Currently, only dump by gid is implemented */ - if (req.gid >= gru_max_gids || req.gid < 0) + if (req.gid >= gru_max_gids) return -EINVAL; gru = GID_TO_GRU(req.gid); diff --git a/drivers/misc/sgi-gru/grukservices.c b/drivers/misc/sgi-gru/grukservices.c index 913de07e577c..967b9dd24fe9 100644 --- a/drivers/misc/sgi-gru/grukservices.c +++ b/drivers/misc/sgi-gru/grukservices.c @@ -160,7 +160,12 @@ static void gru_load_kernel_context(struct gru_blade_state *bs, int blade_id) down_write(&bs->bs_kgts_sema); if (!bs->bs_kgts) { - bs->bs_kgts = gru_alloc_gts(NULL, 0, 0, 0, 0, 0); + do { + bs->bs_kgts = gru_alloc_gts(NULL, 0, 0, 0, 0, 0); + if (!IS_ERR(bs->bs_kgts)) + break; + msleep(1); + } while (true); bs->bs_kgts->ts_user_blade_id = blade_id; } kgts = bs->bs_kgts; @@ -429,8 +434,8 @@ int gru_get_cb_exception_detail(void *cb, return 0; } -char *gru_get_cb_exception_detail_str(int ret, void *cb, - char *buf, int size) +static char *gru_get_cb_exception_detail_str(int ret, void *cb, + char *buf, int size) { struct gru_control_block_status *gen = (void *)cb; struct control_block_extended_exc_detail excdet; @@ -505,7 +510,7 @@ int gru_wait_proc(void *cb) return ret; } -void gru_abort(int ret, void *cb, char *str) +static void gru_abort(int ret, void *cb, char *str) { char buf[GRU_EXC_STR_SIZE]; @@ -997,7 +1002,6 @@ static int quicktest1(unsigned long arg) { struct gru_message_queue_desc mqd; void *p, *mq; - unsigned long *dw; int i, ret = -EIO; char mes[GRU_CACHE_LINE_BYTES], *m; @@ -1007,7 +1011,6 @@ static int quicktest1(unsigned long arg) return -ENOMEM; mq = ALIGNUP(p, 1024); memset(mes, 0xee, sizeof(mes)); - dw = mq; gru_create_message_queue(&mqd, mq, 8 * GRU_CACHE_LINE_BYTES, 0, 0, 0); for (i = 0; i < 6; i++) { diff --git a/drivers/misc/sgi-gru/grumain.c b/drivers/misc/sgi-gru/grumain.c index ae16c8cb4f3e..1525870f460a 100644 --- a/drivers/misc/sgi-gru/grumain.c +++ b/drivers/misc/sgi-gru/grumain.c @@ -930,6 +930,7 @@ int gru_fault(struct vm_area_struct *vma, struct vm_fault *vmf) { struct gru_thread_state *gts; unsigned long paddr, vaddr; + unsigned long expires; vaddr = (unsigned long)vmf->virtual_address; gru_dbg(grudev, "vma %p, vaddr 0x%lx (0x%lx)\n", @@ -954,7 +955,8 @@ again: mutex_unlock(>s->ts_ctxlock); set_current_state(TASK_INTERRUPTIBLE); schedule_timeout(GRU_ASSIGN_DELAY); /* true hack ZZZ */ - if (gts->ts_steal_jiffies + GRU_STEAL_DELAY < jiffies) + expires = gts->ts_steal_jiffies + GRU_STEAL_DELAY; + if (time_before(expires, jiffies)) gru_steal_context(gts); goto again; } diff --git a/drivers/misc/sgi-gru/grutlbpurge.c b/drivers/misc/sgi-gru/grutlbpurge.c index 2129274ef7ab..e936d43895d2 100644 --- a/drivers/misc/sgi-gru/grutlbpurge.c +++ b/drivers/misc/sgi-gru/grutlbpurge.c @@ -306,19 +306,20 @@ struct gru_mm_struct *gru_register_mmu_notifier(void) atomic_inc(&gms->ms_refcnt); } else { gms = kzalloc(sizeof(*gms), GFP_KERNEL); - if (gms) { - STAT(gms_alloc); - spin_lock_init(&gms->ms_asid_lock); - gms->ms_notifier.ops = &gru_mmuops; - atomic_set(&gms->ms_refcnt, 1); - init_waitqueue_head(&gms->ms_wait_queue); - err = __mmu_notifier_register(&gms->ms_notifier, current->mm); - if (err) - goto error; - } + if (!gms) + return ERR_PTR(-ENOMEM); + STAT(gms_alloc); + spin_lock_init(&gms->ms_asid_lock); + gms->ms_notifier.ops = &gru_mmuops; + atomic_set(&gms->ms_refcnt, 1); + init_waitqueue_head(&gms->ms_wait_queue); + err = __mmu_notifier_register(&gms->ms_notifier, current->mm); + if (err) + goto error; } - gru_dbg(grudev, "gms %p, refcnt %d\n", gms, - atomic_read(&gms->ms_refcnt)); + if (gms) + gru_dbg(grudev, "gms %p, refcnt %d\n", gms, + atomic_read(&gms->ms_refcnt)); return gms; error: kfree(gms); diff --git a/drivers/misc/ti-st/st_core.c b/drivers/misc/ti-st/st_core.c index c8c6a363069c..6e3af8b42cdd 100644 --- a/drivers/misc/ti-st/st_core.c +++ b/drivers/misc/ti-st/st_core.c @@ -460,6 +460,13 @@ static void st_int_enqueue(struct st_data_s *st_gdata, struct sk_buff *skb) * - TTY layer when write's finished * - st_write (in context of the protocol stack) */ +static void work_fn_write_wakeup(struct work_struct *work) +{ + struct st_data_s *st_gdata = container_of(work, struct st_data_s, + work_write_wakeup); + + st_tx_wakeup((void *)st_gdata); +} void st_tx_wakeup(struct st_data_s *st_data) { struct sk_buff *skb; @@ -812,8 +819,12 @@ static void st_tty_wakeup(struct tty_struct *tty) /* don't do an wakeup for now */ clear_bit(TTY_DO_WRITE_WAKEUP, &tty->flags); - /* call our internal wakeup */ - st_tx_wakeup((void *)st_gdata); + /* + * schedule the internal wakeup instead of calling directly to + * avoid lockup (port->lock needed in tty->ops->write is + * already taken here + */ + schedule_work(&st_gdata->work_write_wakeup); } static void st_tty_flush_buffer(struct tty_struct *tty) @@ -881,6 +892,9 @@ int st_core_init(struct st_data_s **core_data) pr_err("unable to un-register ldisc"); return err; } + + INIT_WORK(&st_gdata->work_write_wakeup, work_fn_write_wakeup); + *core_data = st_gdata; return 0; } diff --git a/drivers/misc/vmw_balloon.c b/drivers/misc/vmw_balloon.c index ffb56340d0c7..89300870fefb 100644 --- a/drivers/misc/vmw_balloon.c +++ b/drivers/misc/vmw_balloon.c @@ -1,7 +1,7 @@ /* * VMware Balloon driver. * - * Copyright (C) 2000-2010, VMware, Inc. All Rights Reserved. + * Copyright (C) 2000-2014, VMware, Inc. All Rights Reserved. * * This program is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License as published by the @@ -37,16 +37,19 @@ #include <linux/types.h> #include <linux/kernel.h> #include <linux/mm.h> +#include <linux/vmalloc.h> #include <linux/sched.h> #include <linux/module.h> #include <linux/workqueue.h> #include <linux/debugfs.h> #include <linux/seq_file.h> +#include <linux/vmw_vmci_defs.h> +#include <linux/vmw_vmci_api.h> #include <asm/hypervisor.h> MODULE_AUTHOR("VMware, Inc."); MODULE_DESCRIPTION("VMware Memory Control (Balloon) Driver"); -MODULE_VERSION("1.3.0.0-k"); +MODULE_VERSION("1.5.0.0-k"); MODULE_ALIAS("dmi:*:svnVMware*:*"); MODULE_ALIAS("vmware_vmmemctl"); MODULE_LICENSE("GPL"); @@ -57,12 +60,6 @@ MODULE_LICENSE("GPL"); */ /* - * Rate of allocating memory when there is no memory pressure - * (driver performs non-sleeping allocations). - */ -#define VMW_BALLOON_NOSLEEP_ALLOC_MAX 16384U - -/* * Rates of memory allocaton when guest experiences memory pressure * (driver performs sleeping allocations). */ @@ -71,13 +68,6 @@ MODULE_LICENSE("GPL"); #define VMW_BALLOON_RATE_ALLOC_INC 16U /* - * Rates for releasing pages while deflating balloon. - */ -#define VMW_BALLOON_RATE_FREE_MIN 512U -#define VMW_BALLOON_RATE_FREE_MAX 16384U -#define VMW_BALLOON_RATE_FREE_INC 16U - -/* * When guest is under memory pressure, use a reduced page allocation * rate for next several cycles. */ @@ -99,9 +89,6 @@ MODULE_LICENSE("GPL"); */ #define VMW_PAGE_ALLOC_CANSLEEP (GFP_HIGHUSER) -/* Maximum number of page allocations without yielding processor */ -#define VMW_BALLOON_YIELD_THRESHOLD 1024 - /* Maximum number of refused pages we accumulate during inflation cycle */ #define VMW_BALLOON_MAX_REFUSED 16 @@ -116,17 +103,45 @@ enum vmwballoon_capabilities { /* * Bit 0 is reserved and not associated to any capability. */ - VMW_BALLOON_BASIC_CMDS = (1 << 1), - VMW_BALLOON_BATCHED_CMDS = (1 << 2) + VMW_BALLOON_BASIC_CMDS = (1 << 1), + VMW_BALLOON_BATCHED_CMDS = (1 << 2), + VMW_BALLOON_BATCHED_2M_CMDS = (1 << 3), + VMW_BALLOON_SIGNALLED_WAKEUP_CMD = (1 << 4), }; -#define VMW_BALLOON_CAPABILITIES (VMW_BALLOON_BASIC_CMDS) +#define VMW_BALLOON_CAPABILITIES (VMW_BALLOON_BASIC_CMDS \ + | VMW_BALLOON_BATCHED_CMDS \ + | VMW_BALLOON_BATCHED_2M_CMDS \ + | VMW_BALLOON_SIGNALLED_WAKEUP_CMD) + +#define VMW_BALLOON_2M_SHIFT (9) +#define VMW_BALLOON_NUM_PAGE_SIZES (2) + +/* + * Backdoor commands availability: + * + * START, GET_TARGET and GUEST_ID are always available, + * + * VMW_BALLOON_BASIC_CMDS: + * LOCK and UNLOCK commands, + * VMW_BALLOON_BATCHED_CMDS: + * BATCHED_LOCK and BATCHED_UNLOCK commands. + * VMW BALLOON_BATCHED_2M_CMDS: + * BATCHED_2M_LOCK and BATCHED_2M_UNLOCK commands, + * VMW VMW_BALLOON_SIGNALLED_WAKEUP_CMD: + * VMW_BALLOON_CMD_VMCI_DOORBELL_SET command. + */ +#define VMW_BALLOON_CMD_START 0 +#define VMW_BALLOON_CMD_GET_TARGET 1 +#define VMW_BALLOON_CMD_LOCK 2 +#define VMW_BALLOON_CMD_UNLOCK 3 +#define VMW_BALLOON_CMD_GUEST_ID 4 +#define VMW_BALLOON_CMD_BATCHED_LOCK 6 +#define VMW_BALLOON_CMD_BATCHED_UNLOCK 7 +#define VMW_BALLOON_CMD_BATCHED_2M_LOCK 8 +#define VMW_BALLOON_CMD_BATCHED_2M_UNLOCK 9 +#define VMW_BALLOON_CMD_VMCI_DOORBELL_SET 10 -#define VMW_BALLOON_CMD_START 0 -#define VMW_BALLOON_CMD_GET_TARGET 1 -#define VMW_BALLOON_CMD_LOCK 2 -#define VMW_BALLOON_CMD_UNLOCK 3 -#define VMW_BALLOON_CMD_GUEST_ID 4 /* error codes */ #define VMW_BALLOON_SUCCESS 0 @@ -142,18 +157,60 @@ enum vmwballoon_capabilities { #define VMW_BALLOON_SUCCESS_WITH_CAPABILITIES (0x03000000) -#define VMWARE_BALLOON_CMD(cmd, data, result) \ +/* Batch page description */ + +/* + * Layout of a page in the batch page: + * + * +-------------+----------+--------+ + * | | | | + * | Page number | Reserved | Status | + * | | | | + * +-------------+----------+--------+ + * 64 PAGE_SHIFT 6 0 + * + * The reserved field should be set to 0. + */ +#define VMW_BALLOON_BATCH_MAX_PAGES (PAGE_SIZE / sizeof(u64)) +#define VMW_BALLOON_BATCH_STATUS_MASK ((1UL << 5) - 1) +#define VMW_BALLOON_BATCH_PAGE_MASK (~((1UL << PAGE_SHIFT) - 1)) + +struct vmballoon_batch_page { + u64 pages[VMW_BALLOON_BATCH_MAX_PAGES]; +}; + +static u64 vmballoon_batch_get_pa(struct vmballoon_batch_page *batch, int idx) +{ + return batch->pages[idx] & VMW_BALLOON_BATCH_PAGE_MASK; +} + +static int vmballoon_batch_get_status(struct vmballoon_batch_page *batch, + int idx) +{ + return (int)(batch->pages[idx] & VMW_BALLOON_BATCH_STATUS_MASK); +} + +static void vmballoon_batch_set_pa(struct vmballoon_batch_page *batch, int idx, + u64 pa) +{ + batch->pages[idx] = pa; +} + + +#define VMWARE_BALLOON_CMD(cmd, arg1, arg2, result) \ ({ \ - unsigned long __status, __dummy1, __dummy2; \ + unsigned long __status, __dummy1, __dummy2, __dummy3; \ __asm__ __volatile__ ("inl %%dx" : \ "=a"(__status), \ "=c"(__dummy1), \ "=d"(__dummy2), \ - "=b"(result) : \ + "=b"(result), \ + "=S" (__dummy3) : \ "0"(VMW_BALLOON_HV_MAGIC), \ "1"(VMW_BALLOON_CMD_##cmd), \ "2"(VMW_BALLOON_HV_PORT), \ - "3"(data) : \ + "3"(arg1), \ + "4" (arg2) : \ "memory"); \ if (VMW_BALLOON_CMD_##cmd == VMW_BALLOON_CMD_START) \ result = __dummy1; \ @@ -164,27 +221,30 @@ enum vmwballoon_capabilities { #ifdef CONFIG_DEBUG_FS struct vmballoon_stats { unsigned int timer; + unsigned int doorbell; /* allocation statistics */ - unsigned int alloc; - unsigned int alloc_fail; + unsigned int alloc[VMW_BALLOON_NUM_PAGE_SIZES]; + unsigned int alloc_fail[VMW_BALLOON_NUM_PAGE_SIZES]; unsigned int sleep_alloc; unsigned int sleep_alloc_fail; - unsigned int refused_alloc; - unsigned int refused_free; - unsigned int free; + unsigned int refused_alloc[VMW_BALLOON_NUM_PAGE_SIZES]; + unsigned int refused_free[VMW_BALLOON_NUM_PAGE_SIZES]; + unsigned int free[VMW_BALLOON_NUM_PAGE_SIZES]; /* monitor operations */ - unsigned int lock; - unsigned int lock_fail; - unsigned int unlock; - unsigned int unlock_fail; + unsigned int lock[VMW_BALLOON_NUM_PAGE_SIZES]; + unsigned int lock_fail[VMW_BALLOON_NUM_PAGE_SIZES]; + unsigned int unlock[VMW_BALLOON_NUM_PAGE_SIZES]; + unsigned int unlock_fail[VMW_BALLOON_NUM_PAGE_SIZES]; unsigned int target; unsigned int target_fail; unsigned int start; unsigned int start_fail; unsigned int guest_type; unsigned int guest_type_fail; + unsigned int doorbell_set; + unsigned int doorbell_unset; }; #define STATS_INC(stat) (stat)++ @@ -192,14 +252,30 @@ struct vmballoon_stats { #define STATS_INC(stat) #endif -struct vmballoon { +struct vmballoon; +struct vmballoon_ops { + void (*add_page)(struct vmballoon *b, int idx, struct page *p); + int (*lock)(struct vmballoon *b, unsigned int num_pages, + bool is_2m_pages, unsigned int *target); + int (*unlock)(struct vmballoon *b, unsigned int num_pages, + bool is_2m_pages, unsigned int *target); +}; + +struct vmballoon_page_size { /* list of reserved physical pages */ struct list_head pages; /* transient list of non-balloonable pages */ struct list_head refused_pages; unsigned int n_refused_pages; +}; + +struct vmballoon { + struct vmballoon_page_size page_sizes[VMW_BALLOON_NUM_PAGE_SIZES]; + + /* supported page sizes. 1 == 4k pages only, 2 == 4k and 2m pages */ + unsigned supported_page_sizes; /* balloon size in pages */ unsigned int size; @@ -210,11 +286,18 @@ struct vmballoon { /* adjustment rates (pages per second) */ unsigned int rate_alloc; - unsigned int rate_free; /* slowdown page allocations for next few cycles */ unsigned int slow_allocation_cycles; + unsigned long capabilities; + + struct vmballoon_batch_page *batch_page; + unsigned int batch_max_pages; + struct page *page; + + const struct vmballoon_ops *ops; + #ifdef CONFIG_DEBUG_FS /* statistics */ struct vmballoon_stats stats; @@ -226,6 +309,8 @@ struct vmballoon { struct sysinfo sysinfo; struct delayed_work dwork; + + struct vmci_handle vmci_doorbell; }; static struct vmballoon balloon; @@ -234,20 +319,38 @@ static struct vmballoon balloon; * Send "start" command to the host, communicating supported version * of the protocol. */ -static bool vmballoon_send_start(struct vmballoon *b) +static bool vmballoon_send_start(struct vmballoon *b, unsigned long req_caps) { - unsigned long status, capabilities; + unsigned long status, capabilities, dummy = 0; + bool success; STATS_INC(b->stats.start); - status = VMWARE_BALLOON_CMD(START, VMW_BALLOON_CAPABILITIES, - capabilities); - if (status == VMW_BALLOON_SUCCESS) - return true; + status = VMWARE_BALLOON_CMD(START, req_caps, dummy, capabilities); - pr_debug("%s - failed, hv returns %ld\n", __func__, status); - STATS_INC(b->stats.start_fail); - return false; + switch (status) { + case VMW_BALLOON_SUCCESS_WITH_CAPABILITIES: + b->capabilities = capabilities; + success = true; + break; + case VMW_BALLOON_SUCCESS: + b->capabilities = VMW_BALLOON_BASIC_CMDS; + success = true; + break; + default: + success = false; + } + + if (b->capabilities & VMW_BALLOON_BATCHED_2M_CMDS) + b->supported_page_sizes = 2; + else + b->supported_page_sizes = 1; + + if (!success) { + pr_debug("%s - failed, hv returns %ld\n", __func__, status); + STATS_INC(b->stats.start_fail); + } + return success; } static bool vmballoon_check_status(struct vmballoon *b, unsigned long status) @@ -273,9 +376,10 @@ static bool vmballoon_check_status(struct vmballoon *b, unsigned long status) */ static bool vmballoon_send_guest_id(struct vmballoon *b) { - unsigned long status, dummy; + unsigned long status, dummy = 0; - status = VMWARE_BALLOON_CMD(GUEST_ID, VMW_BALLOON_GUEST_ID, dummy); + status = VMWARE_BALLOON_CMD(GUEST_ID, VMW_BALLOON_GUEST_ID, dummy, + dummy); STATS_INC(b->stats.guest_type); @@ -287,6 +391,14 @@ static bool vmballoon_send_guest_id(struct vmballoon *b) return false; } +static u16 vmballoon_page_size(bool is_2m_page) +{ + if (is_2m_page) + return 1 << VMW_BALLOON_2M_SHIFT; + + return 1; +} + /* * Retrieve desired balloon size from the host. */ @@ -295,6 +407,7 @@ static bool vmballoon_send_get_target(struct vmballoon *b, u32 *new_target) unsigned long status; unsigned long target; unsigned long limit; + unsigned long dummy = 0; u32 limit32; /* @@ -313,7 +426,7 @@ static bool vmballoon_send_get_target(struct vmballoon *b, u32 *new_target) /* update stats */ STATS_INC(b->stats.target); - status = VMWARE_BALLOON_CMD(GET_TARGET, limit, target); + status = VMWARE_BALLOON_CMD(GET_TARGET, limit, dummy, target); if (vmballoon_check_status(b, status)) { *new_target = target; return true; @@ -330,23 +443,46 @@ static bool vmballoon_send_get_target(struct vmballoon *b, u32 *new_target) * check the return value and maybe submit a different page. */ static int vmballoon_send_lock_page(struct vmballoon *b, unsigned long pfn, - unsigned int *hv_status) + unsigned int *hv_status, unsigned int *target) { - unsigned long status, dummy; + unsigned long status, dummy = 0; u32 pfn32; pfn32 = (u32)pfn; if (pfn32 != pfn) return -1; - STATS_INC(b->stats.lock); + STATS_INC(b->stats.lock[false]); - *hv_status = status = VMWARE_BALLOON_CMD(LOCK, pfn, dummy); + *hv_status = status = VMWARE_BALLOON_CMD(LOCK, pfn, dummy, *target); if (vmballoon_check_status(b, status)) return 0; pr_debug("%s - ppn %lx, hv returns %ld\n", __func__, pfn, status); - STATS_INC(b->stats.lock_fail); + STATS_INC(b->stats.lock_fail[false]); + return 1; +} + +static int vmballoon_send_batched_lock(struct vmballoon *b, + unsigned int num_pages, bool is_2m_pages, unsigned int *target) +{ + unsigned long status; + unsigned long pfn = page_to_pfn(b->page); + + STATS_INC(b->stats.lock[is_2m_pages]); + + if (is_2m_pages) + status = VMWARE_BALLOON_CMD(BATCHED_2M_LOCK, pfn, num_pages, + *target); + else + status = VMWARE_BALLOON_CMD(BATCHED_LOCK, pfn, num_pages, + *target); + + if (vmballoon_check_status(b, status)) + return 0; + + pr_debug("%s - batch ppn %lx, hv returns %ld\n", __func__, pfn, status); + STATS_INC(b->stats.lock_fail[is_2m_pages]); return 1; } @@ -354,26 +490,66 @@ static int vmballoon_send_lock_page(struct vmballoon *b, unsigned long pfn, * Notify the host that guest intends to release given page back into * the pool of available (to the guest) pages. */ -static bool vmballoon_send_unlock_page(struct vmballoon *b, unsigned long pfn) +static bool vmballoon_send_unlock_page(struct vmballoon *b, unsigned long pfn, + unsigned int *target) { - unsigned long status, dummy; + unsigned long status, dummy = 0; u32 pfn32; pfn32 = (u32)pfn; if (pfn32 != pfn) return false; - STATS_INC(b->stats.unlock); + STATS_INC(b->stats.unlock[false]); - status = VMWARE_BALLOON_CMD(UNLOCK, pfn, dummy); + status = VMWARE_BALLOON_CMD(UNLOCK, pfn, dummy, *target); if (vmballoon_check_status(b, status)) return true; pr_debug("%s - ppn %lx, hv returns %ld\n", __func__, pfn, status); - STATS_INC(b->stats.unlock_fail); + STATS_INC(b->stats.unlock_fail[false]); + return false; +} + +static bool vmballoon_send_batched_unlock(struct vmballoon *b, + unsigned int num_pages, bool is_2m_pages, unsigned int *target) +{ + unsigned long status; + unsigned long pfn = page_to_pfn(b->page); + + STATS_INC(b->stats.unlock[is_2m_pages]); + + if (is_2m_pages) + status = VMWARE_BALLOON_CMD(BATCHED_2M_UNLOCK, pfn, num_pages, + *target); + else + status = VMWARE_BALLOON_CMD(BATCHED_UNLOCK, pfn, num_pages, + *target); + + if (vmballoon_check_status(b, status)) + return true; + + pr_debug("%s - batch ppn %lx, hv returns %ld\n", __func__, pfn, status); + STATS_INC(b->stats.unlock_fail[is_2m_pages]); return false; } +static struct page *vmballoon_alloc_page(gfp_t flags, bool is_2m_page) +{ + if (is_2m_page) + return alloc_pages(flags, VMW_BALLOON_2M_SHIFT); + + return alloc_page(flags); +} + +static void vmballoon_free_page(struct page *page, bool is_2m_page) +{ + if (is_2m_page) + __free_pages(page, VMW_BALLOON_2M_SHIFT); + else + __free_page(page); +} + /* * Quickly release all pages allocated for the balloon. This function is * called when host decides to "reset" balloon for one reason or another. @@ -383,35 +559,31 @@ static bool vmballoon_send_unlock_page(struct vmballoon *b, unsigned long pfn) static void vmballoon_pop(struct vmballoon *b) { struct page *page, *next; - unsigned int count = 0; - - list_for_each_entry_safe(page, next, &b->pages, lru) { - list_del(&page->lru); - __free_page(page); - STATS_INC(b->stats.free); - b->size--; - - if (++count >= b->rate_free) { - count = 0; + unsigned is_2m_pages; + + for (is_2m_pages = 0; is_2m_pages < VMW_BALLOON_NUM_PAGE_SIZES; + is_2m_pages++) { + struct vmballoon_page_size *page_size = + &b->page_sizes[is_2m_pages]; + u16 size_per_page = vmballoon_page_size(is_2m_pages); + + list_for_each_entry_safe(page, next, &page_size->pages, lru) { + list_del(&page->lru); + vmballoon_free_page(page, is_2m_pages); + STATS_INC(b->stats.free[is_2m_pages]); + b->size -= size_per_page; cond_resched(); } } -} -/* - * Perform standard reset sequence by popping the balloon (in case it - * is not empty) and then restarting protocol. This operation normally - * happens when host responds with VMW_BALLOON_ERROR_RESET to a command. - */ -static void vmballoon_reset(struct vmballoon *b) -{ - /* free all pages, skipping monitor unlock */ - vmballoon_pop(b); + if (b->batch_page) { + vunmap(b->batch_page); + b->batch_page = NULL; + } - if (vmballoon_send_start(b)) { - b->reset_required = false; - if (!vmballoon_send_guest_id(b)) - pr_err("failed to send guest ID to the host\n"); + if (b->page) { + __free_page(b->page); + b->page = NULL; } } @@ -420,17 +592,23 @@ static void vmballoon_reset(struct vmballoon *b) * refuse list, those refused page are then released at the end of the * inflation cycle. */ -static int vmballoon_lock_page(struct vmballoon *b, struct page *page) +static int vmballoon_lock_page(struct vmballoon *b, unsigned int num_pages, + bool is_2m_pages, unsigned int *target) { int locked, hv_status; + struct page *page = b->page; + struct vmballoon_page_size *page_size = &b->page_sizes[false]; + + /* is_2m_pages can never happen as 2m pages support implies batching */ - locked = vmballoon_send_lock_page(b, page_to_pfn(page), &hv_status); + locked = vmballoon_send_lock_page(b, page_to_pfn(page), &hv_status, + target); if (locked > 0) { - STATS_INC(b->stats.refused_alloc); + STATS_INC(b->stats.refused_alloc[false]); if (hv_status == VMW_BALLOON_ERROR_RESET || hv_status == VMW_BALLOON_ERROR_PPN_NOTNEEDED) { - __free_page(page); + vmballoon_free_page(page, false); return -EIO; } @@ -439,17 +617,17 @@ static int vmballoon_lock_page(struct vmballoon *b, struct page *page) * and retry allocation, unless we already accumulated * too many of them, in which case take a breather. */ - if (b->n_refused_pages < VMW_BALLOON_MAX_REFUSED) { - b->n_refused_pages++; - list_add(&page->lru, &b->refused_pages); + if (page_size->n_refused_pages < VMW_BALLOON_MAX_REFUSED) { + page_size->n_refused_pages++; + list_add(&page->lru, &page_size->refused_pages); } else { - __free_page(page); + vmballoon_free_page(page, false); } return -EIO; } /* track allocated page */ - list_add(&page->lru, &b->pages); + list_add(&page->lru, &page_size->pages); /* update balloon size */ b->size++; @@ -457,21 +635,81 @@ static int vmballoon_lock_page(struct vmballoon *b, struct page *page) return 0; } +static int vmballoon_lock_batched_page(struct vmballoon *b, + unsigned int num_pages, bool is_2m_pages, unsigned int *target) +{ + int locked, i; + u16 size_per_page = vmballoon_page_size(is_2m_pages); + + locked = vmballoon_send_batched_lock(b, num_pages, is_2m_pages, + target); + if (locked > 0) { + for (i = 0; i < num_pages; i++) { + u64 pa = vmballoon_batch_get_pa(b->batch_page, i); + struct page *p = pfn_to_page(pa >> PAGE_SHIFT); + + vmballoon_free_page(p, is_2m_pages); + } + + return -EIO; + } + + for (i = 0; i < num_pages; i++) { + u64 pa = vmballoon_batch_get_pa(b->batch_page, i); + struct page *p = pfn_to_page(pa >> PAGE_SHIFT); + struct vmballoon_page_size *page_size = + &b->page_sizes[is_2m_pages]; + + locked = vmballoon_batch_get_status(b->batch_page, i); + + switch (locked) { + case VMW_BALLOON_SUCCESS: + list_add(&p->lru, &page_size->pages); + b->size += size_per_page; + break; + case VMW_BALLOON_ERROR_PPN_PINNED: + case VMW_BALLOON_ERROR_PPN_INVALID: + if (page_size->n_refused_pages + < VMW_BALLOON_MAX_REFUSED) { + list_add(&p->lru, &page_size->refused_pages); + page_size->n_refused_pages++; + break; + } + /* Fallthrough */ + case VMW_BALLOON_ERROR_RESET: + case VMW_BALLOON_ERROR_PPN_NOTNEEDED: + vmballoon_free_page(p, is_2m_pages); + break; + default: + /* This should never happen */ + WARN_ON_ONCE(true); + } + } + + return 0; +} + /* * Release the page allocated for the balloon. Note that we first notify * the host so it can make sure the page will be available for the guest * to use, if needed. */ -static int vmballoon_release_page(struct vmballoon *b, struct page *page) +static int vmballoon_unlock_page(struct vmballoon *b, unsigned int num_pages, + bool is_2m_pages, unsigned int *target) { - if (!vmballoon_send_unlock_page(b, page_to_pfn(page))) - return -EIO; + struct page *page = b->page; + struct vmballoon_page_size *page_size = &b->page_sizes[false]; + + /* is_2m_pages can never happen as 2m pages support implies batching */ - list_del(&page->lru); + if (!vmballoon_send_unlock_page(b, page_to_pfn(page), target)) { + list_add(&page->lru, &page_size->pages); + return -EIO; + } /* deallocate page */ - __free_page(page); - STATS_INC(b->stats.free); + vmballoon_free_page(page, false); + STATS_INC(b->stats.free[false]); /* update balloon size */ b->size--; @@ -479,21 +717,76 @@ static int vmballoon_release_page(struct vmballoon *b, struct page *page) return 0; } +static int vmballoon_unlock_batched_page(struct vmballoon *b, + unsigned int num_pages, bool is_2m_pages, + unsigned int *target) +{ + int locked, i, ret = 0; + bool hv_success; + u16 size_per_page = vmballoon_page_size(is_2m_pages); + + hv_success = vmballoon_send_batched_unlock(b, num_pages, is_2m_pages, + target); + if (!hv_success) + ret = -EIO; + + for (i = 0; i < num_pages; i++) { + u64 pa = vmballoon_batch_get_pa(b->batch_page, i); + struct page *p = pfn_to_page(pa >> PAGE_SHIFT); + struct vmballoon_page_size *page_size = + &b->page_sizes[is_2m_pages]; + + locked = vmballoon_batch_get_status(b->batch_page, i); + if (!hv_success || locked != VMW_BALLOON_SUCCESS) { + /* + * That page wasn't successfully unlocked by the + * hypervisor, re-add it to the list of pages owned by + * the balloon driver. + */ + list_add(&p->lru, &page_size->pages); + } else { + /* deallocate page */ + vmballoon_free_page(p, is_2m_pages); + STATS_INC(b->stats.free[is_2m_pages]); + + /* update balloon size */ + b->size -= size_per_page; + } + } + + return ret; +} + /* * Release pages that were allocated while attempting to inflate the * balloon but were refused by the host for one reason or another. */ -static void vmballoon_release_refused_pages(struct vmballoon *b) +static void vmballoon_release_refused_pages(struct vmballoon *b, + bool is_2m_pages) { struct page *page, *next; + struct vmballoon_page_size *page_size = + &b->page_sizes[is_2m_pages]; - list_for_each_entry_safe(page, next, &b->refused_pages, lru) { + list_for_each_entry_safe(page, next, &page_size->refused_pages, lru) { list_del(&page->lru); - __free_page(page); - STATS_INC(b->stats.refused_free); + vmballoon_free_page(page, is_2m_pages); + STATS_INC(b->stats.refused_free[is_2m_pages]); } - b->n_refused_pages = 0; + page_size->n_refused_pages = 0; +} + +static void vmballoon_add_page(struct vmballoon *b, int idx, struct page *p) +{ + b->page = p; +} + +static void vmballoon_add_batched_page(struct vmballoon *b, int idx, + struct page *p) +{ + vmballoon_batch_set_pa(b->batch_page, idx, + (u64)page_to_pfn(p) << PAGE_SHIFT); } /* @@ -503,12 +796,12 @@ static void vmballoon_release_refused_pages(struct vmballoon *b) */ static void vmballoon_inflate(struct vmballoon *b) { - unsigned int goal; - unsigned int rate; - unsigned int i; + unsigned rate; unsigned int allocations = 0; + unsigned int num_pages = 0; int error = 0; gfp_t flags = VMW_PAGE_ALLOC_NOSLEEP; + bool is_2m_pages; pr_debug("%s - size: %d, target %d\n", __func__, b->size, b->target); @@ -527,27 +820,50 @@ static void vmballoon_inflate(struct vmballoon *b) * slowdown page allocations considerably. */ - goal = b->target - b->size; /* * Start with no sleep allocation rate which may be higher * than sleeping allocation rate. */ - rate = b->slow_allocation_cycles ? - b->rate_alloc : VMW_BALLOON_NOSLEEP_ALLOC_MAX; + if (b->slow_allocation_cycles) { + rate = b->rate_alloc; + is_2m_pages = false; + } else { + rate = UINT_MAX; + is_2m_pages = + b->supported_page_sizes == VMW_BALLOON_NUM_PAGE_SIZES; + } - pr_debug("%s - goal: %d, no-sleep rate: %d, sleep rate: %d\n", - __func__, goal, rate, b->rate_alloc); + pr_debug("%s - goal: %d, no-sleep rate: %u, sleep rate: %d\n", + __func__, b->target - b->size, rate, b->rate_alloc); - for (i = 0; i < goal; i++) { + while (!b->reset_required && + b->size + num_pages * vmballoon_page_size(is_2m_pages) + < b->target) { struct page *page; if (flags == VMW_PAGE_ALLOC_NOSLEEP) - STATS_INC(b->stats.alloc); + STATS_INC(b->stats.alloc[is_2m_pages]); else STATS_INC(b->stats.sleep_alloc); - page = alloc_page(flags); + page = vmballoon_alloc_page(flags, is_2m_pages); if (!page) { + STATS_INC(b->stats.alloc_fail[is_2m_pages]); + + if (is_2m_pages) { + b->ops->lock(b, num_pages, true, &b->target); + + /* + * ignore errors from locking as we now switch + * to 4k pages and we might get different + * errors. + */ + + num_pages = 0; + is_2m_pages = false; + continue; + } + if (flags == VMW_PAGE_ALLOC_CANSLEEP) { /* * CANSLEEP page allocation failed, so guest @@ -559,7 +875,6 @@ static void vmballoon_inflate(struct vmballoon *b) STATS_INC(b->stats.sleep_alloc_fail); break; } - STATS_INC(b->stats.alloc_fail); /* * NOSLEEP page allocation failed, so the guest is @@ -571,7 +886,7 @@ static void vmballoon_inflate(struct vmballoon *b) */ b->slow_allocation_cycles = VMW_BALLOON_SLOW_CYCLES; - if (i >= b->rate_alloc) + if (allocations >= b->rate_alloc) break; flags = VMW_PAGE_ALLOC_CANSLEEP; @@ -580,34 +895,40 @@ static void vmballoon_inflate(struct vmballoon *b) continue; } - error = vmballoon_lock_page(b, page); - if (error) - break; - - if (++allocations > VMW_BALLOON_YIELD_THRESHOLD) { - cond_resched(); - allocations = 0; + b->ops->add_page(b, num_pages++, page); + if (num_pages == b->batch_max_pages) { + error = b->ops->lock(b, num_pages, is_2m_pages, + &b->target); + num_pages = 0; + if (error) + break; } - if (i >= rate) { + cond_resched(); + + if (allocations >= rate) { /* We allocated enough pages, let's take a break. */ break; } } + if (num_pages > 0) + b->ops->lock(b, num_pages, is_2m_pages, &b->target); + /* * We reached our goal without failures so try increasing * allocation rate. */ - if (error == 0 && i >= b->rate_alloc) { - unsigned int mult = i / b->rate_alloc; + if (error == 0 && allocations >= b->rate_alloc) { + unsigned int mult = allocations / b->rate_alloc; b->rate_alloc = min(b->rate_alloc + mult * VMW_BALLOON_RATE_ALLOC_INC, VMW_BALLOON_RATE_ALLOC_MAX); } - vmballoon_release_refused_pages(b); + vmballoon_release_refused_pages(b, true); + vmballoon_release_refused_pages(b, false); } /* @@ -615,35 +936,176 @@ static void vmballoon_inflate(struct vmballoon *b) */ static void vmballoon_deflate(struct vmballoon *b) { - struct page *page, *next; - unsigned int i = 0; - unsigned int goal; - int error; + unsigned is_2m_pages; pr_debug("%s - size: %d, target %d\n", __func__, b->size, b->target); - /* limit deallocation rate */ - goal = min(b->size - b->target, b->rate_free); + /* free pages to reach target */ + for (is_2m_pages = 0; is_2m_pages < b->supported_page_sizes; + is_2m_pages++) { + struct page *page, *next; + unsigned int num_pages = 0; + struct vmballoon_page_size *page_size = + &b->page_sizes[is_2m_pages]; + + list_for_each_entry_safe(page, next, &page_size->pages, lru) { + if (b->reset_required || + (b->target > 0 && + b->size - num_pages + * vmballoon_page_size(is_2m_pages) + < b->target + vmballoon_page_size(true))) + break; + + list_del(&page->lru); + b->ops->add_page(b, num_pages++, page); - pr_debug("%s - goal: %d, rate: %d\n", __func__, goal, b->rate_free); + if (num_pages == b->batch_max_pages) { + int error; - /* free pages to reach target */ - list_for_each_entry_safe(page, next, &b->pages, lru) { - error = vmballoon_release_page(b, page); - if (error) { - /* quickly decrease rate in case of error */ - b->rate_free = max(b->rate_free / 2, - VMW_BALLOON_RATE_FREE_MIN); - return; + error = b->ops->unlock(b, num_pages, + is_2m_pages, &b->target); + num_pages = 0; + if (error) + return; + } + + cond_resched(); } - if (++i >= goal) - break; + if (num_pages > 0) + b->ops->unlock(b, num_pages, is_2m_pages, &b->target); + } +} + +static const struct vmballoon_ops vmballoon_basic_ops = { + .add_page = vmballoon_add_page, + .lock = vmballoon_lock_page, + .unlock = vmballoon_unlock_page +}; + +static const struct vmballoon_ops vmballoon_batched_ops = { + .add_page = vmballoon_add_batched_page, + .lock = vmballoon_lock_batched_page, + .unlock = vmballoon_unlock_batched_page +}; + +static bool vmballoon_init_batching(struct vmballoon *b) +{ + b->page = alloc_page(VMW_PAGE_ALLOC_NOSLEEP); + if (!b->page) + return false; + + b->batch_page = vmap(&b->page, 1, VM_MAP, PAGE_KERNEL); + if (!b->batch_page) { + __free_page(b->page); + return false; + } + + return true; +} + +/* + * Receive notification and resize balloon + */ +static void vmballoon_doorbell(void *client_data) +{ + struct vmballoon *b = client_data; + + STATS_INC(b->stats.doorbell); + + mod_delayed_work(system_freezable_wq, &b->dwork, 0); +} + +/* + * Clean up vmci doorbell + */ +static void vmballoon_vmci_cleanup(struct vmballoon *b) +{ + int error; + + VMWARE_BALLOON_CMD(VMCI_DOORBELL_SET, VMCI_INVALID_ID, + VMCI_INVALID_ID, error); + STATS_INC(b->stats.doorbell_unset); + + if (!vmci_handle_is_invalid(b->vmci_doorbell)) { + vmci_doorbell_destroy(b->vmci_doorbell); + b->vmci_doorbell = VMCI_INVALID_HANDLE; + } +} + +/* + * Initialize vmci doorbell, to get notified as soon as balloon changes + */ +static int vmballoon_vmci_init(struct vmballoon *b) +{ + int error = 0; + + if ((b->capabilities & VMW_BALLOON_SIGNALLED_WAKEUP_CMD) != 0) { + error = vmci_doorbell_create(&b->vmci_doorbell, + VMCI_FLAG_DELAYED_CB, + VMCI_PRIVILEGE_FLAG_RESTRICTED, + vmballoon_doorbell, b); + + if (error == VMCI_SUCCESS) { + VMWARE_BALLOON_CMD(VMCI_DOORBELL_SET, + b->vmci_doorbell.context, + b->vmci_doorbell.resource, error); + STATS_INC(b->stats.doorbell_set); + } + } + + if (error != 0) { + vmballoon_vmci_cleanup(b); + + return -EIO; } - /* slowly increase rate if there were no errors */ - b->rate_free = min(b->rate_free + VMW_BALLOON_RATE_FREE_INC, - VMW_BALLOON_RATE_FREE_MAX); + return 0; +} + +/* + * Perform standard reset sequence by popping the balloon (in case it + * is not empty) and then restarting protocol. This operation normally + * happens when host responds with VMW_BALLOON_ERROR_RESET to a command. + */ +static void vmballoon_reset(struct vmballoon *b) +{ + int error; + + vmballoon_vmci_cleanup(b); + + /* free all pages, skipping monitor unlock */ + vmballoon_pop(b); + + if (!vmballoon_send_start(b, VMW_BALLOON_CAPABILITIES)) + return; + + if ((b->capabilities & VMW_BALLOON_BATCHED_CMDS) != 0) { + b->ops = &vmballoon_batched_ops; + b->batch_max_pages = VMW_BALLOON_BATCH_MAX_PAGES; + if (!vmballoon_init_batching(b)) { + /* + * We failed to initialize batching, inform the monitor + * about it by sending a null capability. + * + * The guest will retry in one second. + */ + vmballoon_send_start(b, 0); + return; + } + } else if ((b->capabilities & VMW_BALLOON_BASIC_CMDS) != 0) { + b->ops = &vmballoon_basic_ops; + b->batch_max_pages = 1; + } + + b->reset_required = false; + + error = vmballoon_vmci_init(b); + if (error) + pr_err("failed to initialize vmci doorbell\n"); + + if (!vmballoon_send_guest_id(b)) + pr_err("failed to send guest ID to the host\n"); } /* @@ -664,13 +1126,14 @@ static void vmballoon_work(struct work_struct *work) if (b->slow_allocation_cycles > 0) b->slow_allocation_cycles--; - if (vmballoon_send_get_target(b, &target)) { + if (!b->reset_required && vmballoon_send_get_target(b, &target)) { /* update target, adjust size */ b->target = target; if (b->size < target) vmballoon_inflate(b); - else if (b->size > target) + else if (target == 0 || + b->size > target + vmballoon_page_size(true)) vmballoon_deflate(b); } @@ -692,6 +1155,14 @@ static int vmballoon_debug_show(struct seq_file *f, void *offset) struct vmballoon *b = f->private; struct vmballoon_stats *stats = &b->stats; + /* format capabilities info */ + seq_printf(f, + "balloon capabilities: %#4x\n" + "used capabilities: %#4lx\n" + "is resetting: %c\n", + VMW_BALLOON_CAPABILITIES, b->capabilities, + b->reset_required ? 'y' : 'n'); + /* format size info */ seq_printf(f, "target: %8d pages\n" @@ -700,35 +1171,48 @@ static int vmballoon_debug_show(struct seq_file *f, void *offset) /* format rate info */ seq_printf(f, - "rateNoSleepAlloc: %8d pages/sec\n" - "rateSleepAlloc: %8d pages/sec\n" - "rateFree: %8d pages/sec\n", - VMW_BALLOON_NOSLEEP_ALLOC_MAX, - b->rate_alloc, b->rate_free); + "rateSleepAlloc: %8d pages/sec\n", + b->rate_alloc); seq_printf(f, "\n" "timer: %8u\n" + "doorbell: %8u\n" "start: %8u (%4u failed)\n" "guestType: %8u (%4u failed)\n" + "2m-lock: %8u (%4u failed)\n" "lock: %8u (%4u failed)\n" + "2m-unlock: %8u (%4u failed)\n" "unlock: %8u (%4u failed)\n" "target: %8u (%4u failed)\n" + "prim2mAlloc: %8u (%4u failed)\n" "primNoSleepAlloc: %8u (%4u failed)\n" "primCanSleepAlloc: %8u (%4u failed)\n" + "prim2mFree: %8u\n" "primFree: %8u\n" + "err2mAlloc: %8u\n" "errAlloc: %8u\n" - "errFree: %8u\n", + "err2mFree: %8u\n" + "errFree: %8u\n" + "doorbellSet: %8u\n" + "doorbellUnset: %8u\n", stats->timer, + stats->doorbell, stats->start, stats->start_fail, stats->guest_type, stats->guest_type_fail, - stats->lock, stats->lock_fail, - stats->unlock, stats->unlock_fail, + stats->lock[true], stats->lock_fail[true], + stats->lock[false], stats->lock_fail[false], + stats->unlock[true], stats->unlock_fail[true], + stats->unlock[false], stats->unlock_fail[false], stats->target, stats->target_fail, - stats->alloc, stats->alloc_fail, + stats->alloc[true], stats->alloc_fail[true], + stats->alloc[false], stats->alloc_fail[false], stats->sleep_alloc, stats->sleep_alloc_fail, - stats->free, - stats->refused_alloc, stats->refused_free); + stats->free[true], + stats->free[false], + stats->refused_alloc[true], stats->refused_alloc[false], + stats->refused_free[true], stats->refused_free[false], + stats->doorbell_set, stats->doorbell_unset); return 0; } @@ -782,7 +1266,7 @@ static inline void vmballoon_debugfs_exit(struct vmballoon *b) static int __init vmballoon_init(void) { int error; - + unsigned is_2m_pages; /* * Check if we are running on VMware's hypervisor and bail out * if we are not. @@ -790,32 +1274,26 @@ static int __init vmballoon_init(void) if (x86_hyper != &x86_hyper_vmware) return -ENODEV; - INIT_LIST_HEAD(&balloon.pages); - INIT_LIST_HEAD(&balloon.refused_pages); + for (is_2m_pages = 0; is_2m_pages < VMW_BALLOON_NUM_PAGE_SIZES; + is_2m_pages++) { + INIT_LIST_HEAD(&balloon.page_sizes[is_2m_pages].pages); + INIT_LIST_HEAD(&balloon.page_sizes[is_2m_pages].refused_pages); + } /* initialize rates */ balloon.rate_alloc = VMW_BALLOON_RATE_ALLOC_MAX; - balloon.rate_free = VMW_BALLOON_RATE_FREE_MAX; INIT_DELAYED_WORK(&balloon.dwork, vmballoon_work); - /* - * Start balloon. - */ - if (!vmballoon_send_start(&balloon)) { - pr_err("failed to send start command to the host\n"); - return -EIO; - } - - if (!vmballoon_send_guest_id(&balloon)) { - pr_err("failed to send guest ID to the host\n"); - return -EIO; - } - error = vmballoon_debugfs_init(&balloon); if (error) return error; + balloon.vmci_doorbell = VMCI_INVALID_HANDLE; + balloon.batch_page = NULL; + balloon.page = NULL; + balloon.reset_required = true; + queue_delayed_work(system_freezable_wq, &balloon.dwork, 0); return 0; @@ -824,6 +1302,7 @@ module_init(vmballoon_init); static void __exit vmballoon_exit(void) { + vmballoon_vmci_cleanup(&balloon); cancel_delayed_work_sync(&balloon.dwork); vmballoon_debugfs_exit(&balloon); @@ -833,7 +1312,7 @@ static void __exit vmballoon_exit(void) * Reset connection before deallocating memory to avoid potential for * additional spurious resets from guest touching deallocated pages. */ - vmballoon_send_start(&balloon); + vmballoon_send_start(&balloon, 0); vmballoon_pop(&balloon); } module_exit(vmballoon_exit); diff --git a/drivers/misc/vmw_vmci/vmci_datagram.c b/drivers/misc/vmw_vmci/vmci_datagram.c index 822665245588..8a4b6bbe1bee 100644 --- a/drivers/misc/vmw_vmci/vmci_datagram.c +++ b/drivers/misc/vmw_vmci/vmci_datagram.c @@ -276,11 +276,10 @@ static int dg_dispatch_as_host(u32 context_id, struct vmci_datagram *dg) } /* We make a copy to enqueue. */ - new_dg = kmalloc(dg_size, GFP_KERNEL); + new_dg = kmemdup(dg, dg_size, GFP_KERNEL); if (new_dg == NULL) return VMCI_ERROR_NO_MEM; - memcpy(new_dg, dg, dg_size); retval = vmci_ctx_enqueue_datagram(dg->dst.context, new_dg); if (retval < VMCI_SUCCESS) { kfree(new_dg); diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c index 0520064dc33b..a3eb20bdcd97 100644 --- a/drivers/mmc/core/core.c +++ b/drivers/mmc/core/core.c @@ -134,9 +134,11 @@ void mmc_request_done(struct mmc_host *host, struct mmc_request *mrq) int err = cmd->error; /* Flag re-tuning needed on CRC errors */ - if (err == -EILSEQ || (mrq->sbc && mrq->sbc->error == -EILSEQ) || + if ((cmd->opcode != MMC_SEND_TUNING_BLOCK && + cmd->opcode != MMC_SEND_TUNING_BLOCK_HS200) && + (err == -EILSEQ || (mrq->sbc && mrq->sbc->error == -EILSEQ) || (mrq->data && mrq->data->error == -EILSEQ) || - (mrq->stop && mrq->stop->error == -EILSEQ)) + (mrq->stop && mrq->stop->error == -EILSEQ))) mmc_retune_needed(host); if (err && cmd->retries && mmc_host_is_spi(host)) { diff --git a/drivers/mmc/core/host.c b/drivers/mmc/core/host.c index abd933b7029b..5466f25f0281 100644 --- a/drivers/mmc/core/host.c +++ b/drivers/mmc/core/host.c @@ -457,7 +457,7 @@ int mmc_of_parse(struct mmc_host *host) 0, &cd_gpio_invert); if (!ret) dev_info(host->parent, "Got CD GPIO\n"); - else if (ret != -ENOENT) + else if (ret != -ENOENT && ret != -ENOSYS) return ret; /* @@ -481,7 +481,7 @@ int mmc_of_parse(struct mmc_host *host) ret = mmc_gpiod_request_ro(host, "wp", 0, false, 0, &ro_gpio_invert); if (!ret) dev_info(host->parent, "Got WP GPIO\n"); - else if (ret != -ENOENT) + else if (ret != -ENOENT && ret != -ENOSYS) return ret; if (of_property_read_bool(np, "disable-wp")) diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c index 781e4db31767..7fb0753abe30 100644 --- a/drivers/mmc/host/omap_hsmmc.c +++ b/drivers/mmc/host/omap_hsmmc.c @@ -182,6 +182,7 @@ struct omap_hsmmc_host { struct clk *fclk; struct clk *dbclk; struct regulator *pbias; + bool pbias_enabled; void __iomem *base; int vqmmc_enabled; resource_size_t mapbase; @@ -328,20 +329,22 @@ static int omap_hsmmc_set_pbias(struct omap_hsmmc_host *host, bool power_on, return ret; } - if (!regulator_is_enabled(host->pbias)) { + if (host->pbias_enabled == 0) { ret = regulator_enable(host->pbias); if (ret) { dev_err(host->dev, "pbias reg enable fail\n"); return ret; } + host->pbias_enabled = 1; } } else { - if (regulator_is_enabled(host->pbias)) { + if (host->pbias_enabled == 1) { ret = regulator_disable(host->pbias); if (ret) { dev_err(host->dev, "pbias reg disable fail\n"); return ret; } + host->pbias_enabled = 0; } } @@ -475,7 +478,7 @@ static int omap_hsmmc_reg_get(struct omap_hsmmc_host *host) mmc->supply.vmmc = devm_regulator_get_optional(host->dev, "vmmc"); if (IS_ERR(mmc->supply.vmmc)) { ret = PTR_ERR(mmc->supply.vmmc); - if (ret != -ENODEV) + if ((ret != -ENODEV) && host->dev->of_node) return ret; dev_dbg(host->dev, "unable to get vmmc regulator %ld\n", PTR_ERR(mmc->supply.vmmc)); @@ -490,7 +493,7 @@ static int omap_hsmmc_reg_get(struct omap_hsmmc_host *host) mmc->supply.vqmmc = devm_regulator_get_optional(host->dev, "vmmc_aux"); if (IS_ERR(mmc->supply.vqmmc)) { ret = PTR_ERR(mmc->supply.vqmmc); - if (ret != -ENODEV) + if ((ret != -ENODEV) && host->dev->of_node) return ret; dev_dbg(host->dev, "unable to get vmmc_aux regulator %ld\n", PTR_ERR(mmc->supply.vqmmc)); @@ -500,7 +503,7 @@ static int omap_hsmmc_reg_get(struct omap_hsmmc_host *host) host->pbias = devm_regulator_get_optional(host->dev, "pbias"); if (IS_ERR(host->pbias)) { ret = PTR_ERR(host->pbias); - if (ret != -ENODEV) + if ((ret != -ENODEV) && host->dev->of_node) return ret; dev_dbg(host->dev, "unable to get pbias regulator %ld\n", PTR_ERR(host->pbias)); @@ -2053,6 +2056,7 @@ static int omap_hsmmc_probe(struct platform_device *pdev) host->base = base + pdata->reg_offset; host->power_mode = MMC_POWER_OFF; host->next_data.cookie = 1; + host->pbias_enabled = 0; host->vqmmc_enabled = 0; ret = omap_hsmmc_gpio_init(mmc, host, pdata); diff --git a/drivers/mmc/host/pxamci.c b/drivers/mmc/host/pxamci.c index 1420f29628c7..8cadd74e8407 100644 --- a/drivers/mmc/host/pxamci.c +++ b/drivers/mmc/host/pxamci.c @@ -28,6 +28,7 @@ #include <linux/clk.h> #include <linux/err.h> #include <linux/mmc/host.h> +#include <linux/mmc/slot-gpio.h> #include <linux/io.h> #include <linux/regulator/consumer.h> #include <linux/gpio.h> @@ -454,12 +455,8 @@ static int pxamci_get_ro(struct mmc_host *mmc) { struct pxamci_host *host = mmc_priv(mmc); - if (host->pdata && gpio_is_valid(host->pdata->gpio_card_ro)) { - if (host->pdata->gpio_card_ro_invert) - return !gpio_get_value(host->pdata->gpio_card_ro); - else - return gpio_get_value(host->pdata->gpio_card_ro); - } + if (host->pdata && gpio_is_valid(host->pdata->gpio_card_ro)) + return mmc_gpio_get_ro(mmc); if (host->pdata && host->pdata->get_ro) return !!host->pdata->get_ro(mmc_dev(mmc)); /* @@ -551,6 +548,7 @@ static void pxamci_enable_sdio_irq(struct mmc_host *host, int enable) static const struct mmc_host_ops pxamci_ops = { .request = pxamci_request, + .get_cd = mmc_gpio_get_cd, .get_ro = pxamci_get_ro, .set_ios = pxamci_set_ios, .enable_sdio_irq = pxamci_enable_sdio_irq, @@ -790,37 +788,31 @@ static int pxamci_probe(struct platform_device *pdev) gpio_power = host->pdata->gpio_power; } if (gpio_is_valid(gpio_power)) { - ret = gpio_request(gpio_power, "mmc card power"); + ret = devm_gpio_request(&pdev->dev, gpio_power, + "mmc card power"); if (ret) { - dev_err(&pdev->dev, "Failed requesting gpio_power %d\n", gpio_power); + dev_err(&pdev->dev, "Failed requesting gpio_power %d\n", + gpio_power); goto out; } gpio_direction_output(gpio_power, host->pdata->gpio_power_invert); } - if (gpio_is_valid(gpio_ro)) { - ret = gpio_request(gpio_ro, "mmc card read only"); - if (ret) { - dev_err(&pdev->dev, "Failed requesting gpio_ro %d\n", gpio_ro); - goto err_gpio_ro; - } - gpio_direction_input(gpio_ro); + if (gpio_is_valid(gpio_ro)) + ret = mmc_gpio_request_ro(mmc, gpio_ro); + if (ret) { + dev_err(&pdev->dev, "Failed requesting gpio_ro %d\n", gpio_ro); + goto out; + } else { + mmc->caps |= host->pdata->gpio_card_ro_invert ? + MMC_CAP2_RO_ACTIVE_HIGH : 0; } - if (gpio_is_valid(gpio_cd)) { - ret = gpio_request(gpio_cd, "mmc card detect"); - if (ret) { - dev_err(&pdev->dev, "Failed requesting gpio_cd %d\n", gpio_cd); - goto err_gpio_cd; - } - gpio_direction_input(gpio_cd); - ret = request_irq(gpio_to_irq(gpio_cd), pxamci_detect_irq, - IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING, - "mmc card detect", mmc); - if (ret) { - dev_err(&pdev->dev, "failed to request card detect IRQ\n"); - goto err_request_irq; - } + if (gpio_is_valid(gpio_cd)) + ret = mmc_gpio_request_cd(mmc, gpio_cd, 0); + if (ret) { + dev_err(&pdev->dev, "Failed requesting gpio_cd %d\n", gpio_cd); + goto out; } if (host->pdata && host->pdata->init) @@ -835,13 +827,7 @@ static int pxamci_probe(struct platform_device *pdev) return 0; -err_request_irq: - gpio_free(gpio_cd); -err_gpio_cd: - gpio_free(gpio_ro); -err_gpio_ro: - gpio_free(gpio_power); - out: +out: if (host) { if (host->dma_chan_rx) dma_release_channel(host->dma_chan_rx); @@ -873,14 +859,6 @@ static int pxamci_remove(struct platform_device *pdev) gpio_ro = host->pdata->gpio_card_ro; gpio_power = host->pdata->gpio_power; } - if (gpio_is_valid(gpio_cd)) { - free_irq(gpio_to_irq(gpio_cd), mmc); - gpio_free(gpio_cd); - } - if (gpio_is_valid(gpio_ro)) - gpio_free(gpio_ro); - if (gpio_is_valid(gpio_power)) - gpio_free(gpio_power); if (host->vcc) regulator_put(host->vcc); diff --git a/drivers/mmc/host/sdhci-of-at91.c b/drivers/mmc/host/sdhci-of-at91.c index d1556643a41d..a0f05de5409f 100644 --- a/drivers/mmc/host/sdhci-of-at91.c +++ b/drivers/mmc/host/sdhci-of-at91.c @@ -43,6 +43,7 @@ static const struct sdhci_ops sdhci_at91_sama5d2_ops = { static const struct sdhci_pltfm_data soc_data_sama5d2 = { .ops = &sdhci_at91_sama5d2_ops, + .quirks2 = SDHCI_QUIRK2_NEED_DELAY_AFTER_INT_CLK_RST, }; static const struct of_device_id sdhci_at91_dt_match[] = { diff --git a/drivers/mmc/host/sdhci-pxav3.c b/drivers/mmc/host/sdhci-pxav3.c index 946d37f94a31..f5edf9d3a18a 100644 --- a/drivers/mmc/host/sdhci-pxav3.c +++ b/drivers/mmc/host/sdhci-pxav3.c @@ -135,6 +135,7 @@ static int armada_38x_quirks(struct platform_device *pdev, struct sdhci_pxa *pxa = pltfm_host->priv; struct resource *res; + host->quirks &= ~SDHCI_QUIRK_CAP_CLOCK_BASE_BROKEN; host->quirks |= SDHCI_QUIRK_MISSING_CAPS; res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "conf-sdio3"); @@ -290,6 +291,9 @@ static void pxav3_set_uhs_signaling(struct sdhci_host *host, unsigned int uhs) uhs == MMC_TIMING_UHS_DDR50) { reg_val &= ~SDIO3_CONF_CLK_INV; reg_val |= SDIO3_CONF_SD_FB_CLK; + } else if (uhs == MMC_TIMING_MMC_HS) { + reg_val &= ~SDIO3_CONF_CLK_INV; + reg_val &= ~SDIO3_CONF_SD_FB_CLK; } else { reg_val |= SDIO3_CONF_CLK_INV; reg_val &= ~SDIO3_CONF_SD_FB_CLK; @@ -398,7 +402,7 @@ static int sdhci_pxav3_probe(struct platform_device *pdev) if (of_device_is_compatible(np, "marvell,armada-380-sdhci")) { ret = armada_38x_quirks(pdev, host); if (ret < 0) - goto err_clk_get; + goto err_mbus_win; ret = mv_conf_mbus_windows(pdev, mv_mbus_dram_info()); if (ret < 0) goto err_mbus_win; diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c index 64b7fdbd1a9c..fbc7efdddcb5 100644 --- a/drivers/mmc/host/sdhci.c +++ b/drivers/mmc/host/sdhci.c @@ -1160,6 +1160,8 @@ void sdhci_set_clock(struct sdhci_host *host, unsigned int clock) host->mmc->actual_clock = 0; sdhci_writew(host, 0, SDHCI_CLOCK_CONTROL); + if (host->quirks2 & SDHCI_QUIRK2_NEED_DELAY_AFTER_INT_CLK_RST) + mdelay(1); if (clock == 0) return; diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h index 7c02ff46c8ac..9d4aa31b683a 100644 --- a/drivers/mmc/host/sdhci.h +++ b/drivers/mmc/host/sdhci.h @@ -412,6 +412,11 @@ struct sdhci_host { #define SDHCI_QUIRK2_ACMD23_BROKEN (1<<14) /* Broken Clock divider zero in controller */ #define SDHCI_QUIRK2_CLOCK_DIV_ZERO_BROKEN (1<<15) +/* + * When internal clock is disabled, a delay is needed before modifying the + * SD clock frequency or enabling back the internal clock. + */ +#define SDHCI_QUIRK2_NEED_DELAY_AFTER_INT_CLK_RST (1<<16) int irq; /* Device IRQ */ void __iomem *ioaddr; /* Mapped address */ diff --git a/drivers/mmc/host/sunxi-mmc.c b/drivers/mmc/host/sunxi-mmc.c index a7b7a6771598..b981b8552e43 100644 --- a/drivers/mmc/host/sunxi-mmc.c +++ b/drivers/mmc/host/sunxi-mmc.c @@ -210,6 +210,16 @@ #define SDXC_IDMAC_DES0_CES BIT(30) /* card error summary */ #define SDXC_IDMAC_DES0_OWN BIT(31) /* 1-idma owns it, 0-host owns it */ +#define SDXC_CLK_400K 0 +#define SDXC_CLK_25M 1 +#define SDXC_CLK_50M 2 +#define SDXC_CLK_50M_DDR 3 + +struct sunxi_mmc_clk_delay { + u32 output; + u32 sample; +}; + struct sunxi_idma_des { u32 config; u32 buf_size; @@ -229,6 +239,7 @@ struct sunxi_mmc_host { struct clk *clk_mmc; struct clk *clk_sample; struct clk *clk_output; + const struct sunxi_mmc_clk_delay *clk_delays; /* irq */ spinlock_t lock; @@ -654,25 +665,19 @@ static int sunxi_mmc_clk_set_rate(struct sunxi_mmc_host *host, /* determine delays */ if (rate <= 400000) { - oclk_dly = 180; - sclk_dly = 42; + oclk_dly = host->clk_delays[SDXC_CLK_400K].output; + sclk_dly = host->clk_delays[SDXC_CLK_400K].sample; } else if (rate <= 25000000) { - oclk_dly = 180; - sclk_dly = 75; + oclk_dly = host->clk_delays[SDXC_CLK_25M].output; + sclk_dly = host->clk_delays[SDXC_CLK_25M].sample; } else if (rate <= 50000000) { if (ios->timing == MMC_TIMING_UHS_DDR50) { - oclk_dly = 60; - sclk_dly = 120; + oclk_dly = host->clk_delays[SDXC_CLK_50M_DDR].output; + sclk_dly = host->clk_delays[SDXC_CLK_50M_DDR].sample; } else { - oclk_dly = 90; - sclk_dly = 150; + oclk_dly = host->clk_delays[SDXC_CLK_50M].output; + sclk_dly = host->clk_delays[SDXC_CLK_50M].sample; } - } else if (rate <= 100000000) { - oclk_dly = 6; - sclk_dly = 24; - } else if (rate <= 200000000) { - oclk_dly = 3; - sclk_dly = 12; } else { return -EINVAL; } @@ -871,6 +876,7 @@ static void sunxi_mmc_request(struct mmc_host *mmc, struct mmc_request *mrq) static const struct of_device_id sunxi_mmc_of_match[] = { { .compatible = "allwinner,sun4i-a10-mmc", }, { .compatible = "allwinner,sun5i-a13-mmc", }, + { .compatible = "allwinner,sun9i-a80-mmc", }, { /* sentinel */ } }; MODULE_DEVICE_TABLE(of, sunxi_mmc_of_match); @@ -884,6 +890,20 @@ static struct mmc_host_ops sunxi_mmc_ops = { .hw_reset = sunxi_mmc_hw_reset, }; +static const struct sunxi_mmc_clk_delay sunxi_mmc_clk_delays[] = { + [SDXC_CLK_400K] = { .output = 180, .sample = 180 }, + [SDXC_CLK_25M] = { .output = 180, .sample = 75 }, + [SDXC_CLK_50M] = { .output = 90, .sample = 120 }, + [SDXC_CLK_50M_DDR] = { .output = 60, .sample = 120 }, +}; + +static const struct sunxi_mmc_clk_delay sun9i_mmc_clk_delays[] = { + [SDXC_CLK_400K] = { .output = 180, .sample = 180 }, + [SDXC_CLK_25M] = { .output = 180, .sample = 75 }, + [SDXC_CLK_50M] = { .output = 150, .sample = 120 }, + [SDXC_CLK_50M_DDR] = { .output = 90, .sample = 120 }, +}; + static int sunxi_mmc_resource_request(struct sunxi_mmc_host *host, struct platform_device *pdev) { @@ -895,6 +915,11 @@ static int sunxi_mmc_resource_request(struct sunxi_mmc_host *host, else host->idma_des_size_bits = 16; + if (of_device_is_compatible(np, "allwinner,sun9i-a80-mmc")) + host->clk_delays = sun9i_mmc_clk_delays; + else + host->clk_delays = sunxi_mmc_clk_delays; + ret = mmc_regulator_get_supply(host->mmc); if (ret) { if (ret != -EPROBE_DEFER) diff --git a/drivers/mtd/nand/mxc_nand.c b/drivers/mtd/nand/mxc_nand.c index 2426db88db36..f04445b992f5 100644 --- a/drivers/mtd/nand/mxc_nand.c +++ b/drivers/mtd/nand/mxc_nand.c @@ -879,7 +879,7 @@ static void copy_spare(struct mtd_info *mtd, bool bfrom) oob_chunk_size); /* the last chunk */ - memcpy16_toio(&s[oob_chunk_size * sparebuf_size], + memcpy16_toio(&s[i * sparebuf_size], &d[i * oob_chunk_size], host->used_oobsize - i * oob_chunk_size); } diff --git a/drivers/mtd/nand/sunxi_nand.c b/drivers/mtd/nand/sunxi_nand.c index f97a58d6aae1..e7d333c162be 100644 --- a/drivers/mtd/nand/sunxi_nand.c +++ b/drivers/mtd/nand/sunxi_nand.c @@ -147,6 +147,10 @@ #define NFC_ECC_MODE GENMASK(15, 12) #define NFC_RANDOM_SEED GENMASK(30, 16) +/* NFC_USER_DATA helper macros */ +#define NFC_BUF_TO_USER_DATA(buf) ((buf)[0] | ((buf)[1] << 8) | \ + ((buf)[2] << 16) | ((buf)[3] << 24)) + #define NFC_DEFAULT_TIMEOUT_MS 1000 #define NFC_SRAM_SIZE 1024 @@ -646,15 +650,9 @@ static int sunxi_nfc_hw_ecc_write_page(struct mtd_info *mtd, offset = layout->eccpos[i * ecc->bytes] - 4 + mtd->writesize; /* Fill OOB data in */ - if (oob_required) { - tmp = 0xffffffff; - memcpy_toio(nfc->regs + NFC_REG_USER_DATA_BASE, &tmp, - 4); - } else { - memcpy_toio(nfc->regs + NFC_REG_USER_DATA_BASE, - chip->oob_poi + offset - mtd->writesize, - 4); - } + writel(NFC_BUF_TO_USER_DATA(chip->oob_poi + + layout->oobfree[i].offset), + nfc->regs + NFC_REG_USER_DATA_BASE); chip->cmdfunc(mtd, NAND_CMD_RNDIN, offset, -1); @@ -784,14 +782,8 @@ static int sunxi_nfc_hw_syndrome_ecc_write_page(struct mtd_info *mtd, offset += ecc->size; /* Fill OOB data in */ - if (oob_required) { - tmp = 0xffffffff; - memcpy_toio(nfc->regs + NFC_REG_USER_DATA_BASE, &tmp, - 4); - } else { - memcpy_toio(nfc->regs + NFC_REG_USER_DATA_BASE, oob, - 4); - } + writel(NFC_BUF_TO_USER_DATA(oob), + nfc->regs + NFC_REG_USER_DATA_BASE); tmp = NFC_DATA_TRANS | NFC_DATA_SWAP_METHOD | NFC_ACCESS_DIR | (1 << 30); @@ -1389,6 +1381,7 @@ static void sunxi_nand_chips_cleanup(struct sunxi_nfc *nfc) node); nand_release(&chip->mtd); sunxi_nand_ecc_cleanup(&chip->nand.ecc); + list_del(&chip->node); } } diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c index 5bbd1f094f4e..1fc23e48fe8e 100644 --- a/drivers/mtd/ubi/io.c +++ b/drivers/mtd/ubi/io.c @@ -926,6 +926,11 @@ static int validate_vid_hdr(const struct ubi_device *ubi, goto bad; } + if (data_size > ubi->leb_size) { + ubi_err(ubi, "bad data_size"); + goto bad; + } + if (vol_type == UBI_VID_STATIC) { /* * Although from high-level point of view static volumes may diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c index 80bdd5b88bac..d85c19762160 100644 --- a/drivers/mtd/ubi/vtbl.c +++ b/drivers/mtd/ubi/vtbl.c @@ -649,6 +649,7 @@ static int init_volumes(struct ubi_device *ubi, if (ubi->corr_peb_count) ubi_err(ubi, "%d PEBs are corrupted and not used", ubi->corr_peb_count); + return -ENOSPC; } ubi->rsvd_pebs += reserved_pebs; ubi->avail_pebs -= reserved_pebs; diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c index 275d9fb6fe5c..eb4489f9082f 100644 --- a/drivers/mtd/ubi/wl.c +++ b/drivers/mtd/ubi/wl.c @@ -1601,6 +1601,7 @@ int ubi_wl_init(struct ubi_device *ubi, struct ubi_attach_info *ai) if (ubi->corr_peb_count) ubi_err(ubi, "%d PEBs are corrupted and not used", ubi->corr_peb_count); + err = -ENOSPC; goto out_free; } ubi->avail_pebs -= reserved_pebs; diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c index f8baa897d1a0..1f7dd927cc5e 100644 --- a/drivers/net/dsa/mv88e6xxx.c +++ b/drivers/net/dsa/mv88e6xxx.c @@ -2051,6 +2051,8 @@ static int mv88e6xxx_setup_port(struct dsa_switch *ds, int port) reg |= PORT_CONTROL_FRAME_ETHER_TYPE_DSA; else reg |= PORT_CONTROL_FRAME_MODE_DSA; + reg |= PORT_CONTROL_FORWARD_UNKNOWN | + PORT_CONTROL_FORWARD_UNKNOWN_MC; } if (mv88e6xxx_6352_family(ds) || mv88e6xxx_6351_family(ds) || diff --git a/drivers/net/ethernet/brocade/bna/bfa_ioc.c b/drivers/net/ethernet/brocade/bna/bfa_ioc.c index b7a0f7879de2..9e59663a6ead 100644 --- a/drivers/net/ethernet/brocade/bna/bfa_ioc.c +++ b/drivers/net/ethernet/brocade/bna/bfa_ioc.c @@ -1543,7 +1543,7 @@ bfa_flash_cmd_act_check(void __iomem *pci_bar) } /* Flush FLI data fifo. */ -static u32 +static int bfa_flash_fifo_flush(void __iomem *pci_bar) { u32 i; @@ -1573,11 +1573,11 @@ bfa_flash_fifo_flush(void __iomem *pci_bar) } /* Read flash status. */ -static u32 +static int bfa_flash_status_read(void __iomem *pci_bar) { union bfa_flash_dev_status_reg dev_status; - u32 status; + int status; u32 ret_status; int i; @@ -1611,11 +1611,11 @@ bfa_flash_status_read(void __iomem *pci_bar) } /* Start flash read operation. */ -static u32 +static int bfa_flash_read_start(void __iomem *pci_bar, u32 offset, u32 len, char *buf) { - u32 status; + int status; /* len must be mutiple of 4 and not exceeding fifo size */ if (len == 0 || len > BFA_FLASH_FIFO_SIZE || (len & 0x03) != 0) @@ -1703,7 +1703,8 @@ static enum bfa_status bfa_flash_raw_read(void __iomem *pci_bar, u32 offset, char *buf, u32 len) { - u32 n, status; + u32 n; + int status; u32 off, l, s, residue, fifo_sz; residue = len; diff --git a/drivers/net/ethernet/hisilicon/hip04_eth.c b/drivers/net/ethernet/hisilicon/hip04_eth.c index cc2d8b4b18e3..253f8ed0537a 100644 --- a/drivers/net/ethernet/hisilicon/hip04_eth.c +++ b/drivers/net/ethernet/hisilicon/hip04_eth.c @@ -816,7 +816,7 @@ static int hip04_mac_probe(struct platform_device *pdev) struct net_device *ndev; struct hip04_priv *priv; struct resource *res; - unsigned int irq; + int irq; int ret; ndev = alloc_etherdev(sizeof(struct hip04_priv)); diff --git a/drivers/net/ethernet/ibm/emac/core.h b/drivers/net/ethernet/ibm/emac/core.h index 28df37420da9..ac02c675c59c 100644 --- a/drivers/net/ethernet/ibm/emac/core.h +++ b/drivers/net/ethernet/ibm/emac/core.h @@ -460,8 +460,8 @@ struct emac_ethtool_regs_subhdr { u32 index; }; -#define EMAC_ETHTOOL_REGS_VER 0 -#define EMAC4_ETHTOOL_REGS_VER 1 -#define EMAC4SYNC_ETHTOOL_REGS_VER 2 +#define EMAC_ETHTOOL_REGS_VER 3 +#define EMAC4_ETHTOOL_REGS_VER 4 +#define EMAC4SYNC_ETHTOOL_REGS_VER 5 #endif /* __IBM_NEWEMAC_CORE_H */ diff --git a/drivers/net/ethernet/intel/i40e/i40e_adminq.c b/drivers/net/ethernet/intel/i40e/i40e_adminq.c index 3e0d20037675..62488a67149d 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_adminq.c +++ b/drivers/net/ethernet/intel/i40e/i40e_adminq.c @@ -946,6 +946,13 @@ i40e_status i40e_clean_arq_element(struct i40e_hw *hw, /* take the lock before we start messing with the ring */ mutex_lock(&hw->aq.arq_mutex); + if (hw->aq.arq.count == 0) { + i40e_debug(hw, I40E_DEBUG_AQ_MESSAGE, + "AQRX: Admin queue not initialized.\n"); + ret_code = I40E_ERR_QUEUE_EMPTY; + goto clean_arq_element_err; + } + /* set next_to_use to head */ ntu = (rd32(hw, hw->aq.arq.head) & I40E_PF_ARQH_ARQH_MASK); if (ntu == ntc) { @@ -1007,6 +1014,8 @@ clean_arq_element_out: /* Set pending if needed, unlock and return */ if (pending != NULL) *pending = (ntc > ntu ? hw->aq.arq.count : 0) + (ntu - ntc); + +clean_arq_element_err: mutex_unlock(&hw->aq.arq_mutex); if (i40e_is_nvm_update_op(&e->desc)) { diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index 851c1a159be8..2fdf978ae6a5 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -2672,7 +2672,8 @@ static int i40e_configure_rx_ring(struct i40e_ring *ring) rx_ctx.lrxqthresh = 2; rx_ctx.crcstrip = 1; rx_ctx.l2tsel = 1; - rx_ctx.showiv = 1; + /* this controls whether VLAN is stripped from inner headers */ + rx_ctx.showiv = 0; #ifdef I40E_FCOE rx_ctx.fc_ena = (vsi->type == I40E_VSI_FCOE); #endif diff --git a/drivers/net/ethernet/intel/i40evf/i40e_adminq.c b/drivers/net/ethernet/intel/i40evf/i40e_adminq.c index f08450b90774..929d47152bf2 100644 --- a/drivers/net/ethernet/intel/i40evf/i40e_adminq.c +++ b/drivers/net/ethernet/intel/i40evf/i40e_adminq.c @@ -887,6 +887,13 @@ i40e_status i40evf_clean_arq_element(struct i40e_hw *hw, /* take the lock before we start messing with the ring */ mutex_lock(&hw->aq.arq_mutex); + if (hw->aq.arq.count == 0) { + i40e_debug(hw, I40E_DEBUG_AQ_MESSAGE, + "AQRX: Admin queue not initialized.\n"); + ret_code = I40E_ERR_QUEUE_EMPTY; + goto clean_arq_element_err; + } + /* set next_to_use to head */ ntu = (rd32(hw, hw->aq.arq.head) & I40E_VF_ARQH1_ARQH_MASK); if (ntu == ntc) { @@ -948,6 +955,8 @@ clean_arq_element_out: /* Set pending if needed, unlock and return */ if (pending != NULL) *pending = (ntc > ntu ? hw->aq.arq.count : 0) + (ntu - ntc); + +clean_arq_element_err: mutex_unlock(&hw->aq.arq_mutex); return ret_code; diff --git a/drivers/net/ethernet/mellanox/mlx4/mcg.c b/drivers/net/ethernet/mellanox/mlx4/mcg.c index bd9ea0d01aae..1d4e2e054647 100644 --- a/drivers/net/ethernet/mellanox/mlx4/mcg.c +++ b/drivers/net/ethernet/mellanox/mlx4/mcg.c @@ -1184,10 +1184,11 @@ out: if (prot == MLX4_PROT_ETH) { /* manage the steering entry for promisc mode */ if (new_entry) - new_steering_entry(dev, port, steer, index, qp->qpn); + err = new_steering_entry(dev, port, steer, + index, qp->qpn); else - existing_steering_entry(dev, port, steer, - index, qp->qpn); + err = existing_steering_entry(dev, port, steer, + index, qp->qpn); } if (err && link && index != -1) { if (index < dev->caps.num_mgms) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fw.c b/drivers/net/ethernet/mellanox/mlx5/core/fw.c index aa0d5ffe92d8..9335e5ae18cc 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fw.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fw.c @@ -200,25 +200,3 @@ int mlx5_cmd_teardown_hca(struct mlx5_core_dev *dev) return err; } - -int mlx5_core_query_special_context(struct mlx5_core_dev *dev, u32 *rsvd_lkey) -{ - struct mlx5_cmd_query_special_contexts_mbox_in in; - struct mlx5_cmd_query_special_contexts_mbox_out out; - int err; - - memset(&in, 0, sizeof(in)); - memset(&out, 0, sizeof(out)); - in.hdr.opcode = cpu_to_be16(MLX5_CMD_OP_QUERY_SPECIAL_CONTEXTS); - err = mlx5_cmd_exec(dev, &in, sizeof(in), &out, sizeof(out)); - if (err) - return err; - - if (out.hdr.status) - err = mlx5_cmd_status_to_err(&out.hdr); - - *rsvd_lkey = be32_to_cpu(out.resd_lkey); - - return err; -} -EXPORT_SYMBOL(mlx5_core_query_special_context); diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c index 2b32e0c5a0b4..b4f21232019a 100644 --- a/drivers/net/ethernet/realtek/r8169.c +++ b/drivers/net/ethernet/realtek/r8169.c @@ -6081,7 +6081,7 @@ static void rtl_hw_start_8168h_1(struct rtl8169_private *tp) { void __iomem *ioaddr = tp->mmio_addr; struct pci_dev *pdev = tp->pci_dev; - u16 rg_saw_cnt; + int rg_saw_cnt; u32 data; static const struct ephy_info e_info_8168h_1[] = { { 0x1e, 0x0800, 0x0001 }, diff --git a/drivers/nfc/mei_phy.c b/drivers/nfc/mei_phy.c index 754a9bb0f58d..83deda4bb4d6 100644 --- a/drivers/nfc/mei_phy.c +++ b/drivers/nfc/mei_phy.c @@ -118,7 +118,7 @@ static int mei_nfc_if_version(struct nfc_mei_phy *phy) cmd.sub_command = MEI_NFC_SUBCMD_IF_VERSION; MEI_DUMP_NFC_HDR("version", &cmd.hdr); - r = mei_cl_send(phy->device, (u8 *)&cmd, sizeof(struct mei_nfc_cmd)); + r = mei_cldev_send(phy->cldev, (u8 *)&cmd, sizeof(struct mei_nfc_cmd)); if (r < 0) { pr_err("Could not send IF version cmd\n"); return r; @@ -132,7 +132,7 @@ static int mei_nfc_if_version(struct nfc_mei_phy *phy) if (!reply) return -ENOMEM; - bytes_recv = mei_cl_recv(phy->device, (u8 *)reply, if_version_length); + bytes_recv = mei_cldev_recv(phy->cldev, (u8 *)reply, if_version_length); if (bytes_recv < 0 || bytes_recv < sizeof(struct mei_nfc_reply)) { pr_err("Could not read IF version\n"); r = -EIO; @@ -186,13 +186,14 @@ static int mei_nfc_connect(struct nfc_mei_phy *phy) connect->vendor_id = phy->vendor_id; MEI_DUMP_NFC_HDR("connect request", &cmd->hdr); - r = mei_cl_send(phy->device, (u8 *)cmd, connect_length); + r = mei_cldev_send(phy->cldev, (u8 *)cmd, connect_length); if (r < 0) { pr_err("Could not send connect cmd %d\n", r); goto err; } - bytes_recv = mei_cl_recv(phy->device, (u8 *)reply, connect_resp_length); + bytes_recv = mei_cldev_recv(phy->cldev, (u8 *)reply, + connect_resp_length); if (bytes_recv < 0) { r = bytes_recv; pr_err("Could not read connect response %d\n", r); @@ -238,7 +239,7 @@ static int mei_nfc_send(struct nfc_mei_phy *phy, u8 *buf, size_t length) MEI_DUMP_NFC_HDR("send", hdr); memcpy(mei_buf + MEI_NFC_HEADER_SIZE, buf, length); - err = mei_cl_send(phy->device, mei_buf, length + MEI_NFC_HEADER_SIZE); + err = mei_cldev_send(phy->cldev, mei_buf, length + MEI_NFC_HEADER_SIZE); if (err < 0) goto out; @@ -278,7 +279,7 @@ static int mei_nfc_recv(struct nfc_mei_phy *phy, u8 *buf, size_t length) struct mei_nfc_hdr *hdr; int received_length; - received_length = mei_cl_recv(phy->device, buf, length); + received_length = mei_cldev_recv(phy->cldev, buf, length); if (received_length < 0) return received_length; @@ -296,7 +297,7 @@ static int mei_nfc_recv(struct nfc_mei_phy *phy, u8 *buf, size_t length) } -static void nfc_mei_event_cb(struct mei_cl_device *device, u32 events, +static void nfc_mei_event_cb(struct mei_cl_device *cldev, u32 events, void *context) { struct nfc_mei_phy *phy = context; @@ -337,7 +338,7 @@ static int nfc_mei_phy_enable(void *phy_id) if (phy->powered == 1) return 0; - r = mei_cl_enable_device(phy->device); + r = mei_cldev_enable(phy->cldev); if (r < 0) { pr_err("Could not enable device %d\n", r); return r; @@ -355,7 +356,7 @@ static int nfc_mei_phy_enable(void *phy_id) goto err; } - r = mei_cl_register_event_cb(phy->device, BIT(MEI_CL_EVENT_RX), + r = mei_cldev_register_event_cb(phy->cldev, BIT(MEI_CL_EVENT_RX), nfc_mei_event_cb, phy); if (r) { pr_err("Event cb registration failed %d\n", r); @@ -368,7 +369,7 @@ static int nfc_mei_phy_enable(void *phy_id) err: phy->powered = 0; - mei_cl_disable_device(phy->device); + mei_cldev_disable(phy->cldev); return r; } @@ -378,7 +379,7 @@ static void nfc_mei_phy_disable(void *phy_id) pr_info("%s\n", __func__); - mei_cl_disable_device(phy->device); + mei_cldev_disable(phy->cldev); phy->powered = 0; } @@ -390,7 +391,7 @@ struct nfc_phy_ops mei_phy_ops = { }; EXPORT_SYMBOL_GPL(mei_phy_ops); -struct nfc_mei_phy *nfc_mei_phy_alloc(struct mei_cl_device *device) +struct nfc_mei_phy *nfc_mei_phy_alloc(struct mei_cl_device *cldev) { struct nfc_mei_phy *phy; @@ -398,9 +399,9 @@ struct nfc_mei_phy *nfc_mei_phy_alloc(struct mei_cl_device *device) if (!phy) return NULL; - phy->device = device; + phy->cldev = cldev; init_waitqueue_head(&phy->send_wq); - mei_cl_set_drvdata(device, phy); + mei_cldev_set_drvdata(cldev, phy); return phy; } @@ -408,7 +409,7 @@ EXPORT_SYMBOL_GPL(nfc_mei_phy_alloc); void nfc_mei_phy_free(struct nfc_mei_phy *phy) { - mei_cl_disable_device(phy->device); + mei_cldev_disable(phy->cldev); kfree(phy); } EXPORT_SYMBOL_GPL(nfc_mei_phy_free); diff --git a/drivers/nfc/mei_phy.h b/drivers/nfc/mei_phy.h index fbfa3e61738f..acd3a1fc69e6 100644 --- a/drivers/nfc/mei_phy.h +++ b/drivers/nfc/mei_phy.h @@ -13,7 +13,7 @@ /** * struct nfc_mei_phy * - * @device: mei device + * @cldev: mei client device * @hdev: nfc hci device * @send_wq: send completion wait queue @@ -28,7 +28,7 @@ * and prevents normal operation. */ struct nfc_mei_phy { - struct mei_cl_device *device; + struct mei_cl_device *cldev; struct nfc_hci_dev *hdev; wait_queue_head_t send_wq; diff --git a/drivers/nfc/microread/mei.c b/drivers/nfc/microread/mei.c index f9f5fc97cdd7..3092501f26c4 100644 --- a/drivers/nfc/microread/mei.c +++ b/drivers/nfc/microread/mei.c @@ -29,7 +29,7 @@ #define MICROREAD_DRIVER_NAME "microread" -static int microread_mei_probe(struct mei_cl_device *device, +static int microread_mei_probe(struct mei_cl_device *cldev, const struct mei_cl_device_id *id) { struct nfc_mei_phy *phy; @@ -37,7 +37,7 @@ static int microread_mei_probe(struct mei_cl_device *device, pr_info("Probing NFC microread\n"); - phy = nfc_mei_phy_alloc(device); + phy = nfc_mei_phy_alloc(cldev); if (!phy) { pr_err("Cannot allocate memory for microread mei phy.\n"); return -ENOMEM; @@ -55,9 +55,9 @@ static int microread_mei_probe(struct mei_cl_device *device, return 0; } -static int microread_mei_remove(struct mei_cl_device *device) +static int microread_mei_remove(struct mei_cl_device *cldev) { - struct nfc_mei_phy *phy = mei_cl_get_drvdata(device); + struct nfc_mei_phy *phy = mei_cldev_get_drvdata(cldev); microread_remove(phy->hdev); @@ -67,7 +67,7 @@ static int microread_mei_remove(struct mei_cl_device *device) } static struct mei_cl_device_id microread_mei_tbl[] = { - { MICROREAD_DRIVER_NAME, MEI_NFC_UUID}, + { MICROREAD_DRIVER_NAME, MEI_NFC_UUID, MEI_CL_VERSION_ANY}, /* required last entry */ { } @@ -88,7 +88,7 @@ static int microread_mei_init(void) pr_debug(DRIVER_DESC ": %s\n", __func__); - r = mei_cl_driver_register(µread_driver); + r = mei_cldev_driver_register(µread_driver); if (r) { pr_err(MICROREAD_DRIVER_NAME ": driver registration failed\n"); return r; @@ -99,7 +99,7 @@ static int microread_mei_init(void) static void microread_mei_exit(void) { - mei_cl_driver_unregister(µread_driver); + mei_cldev_driver_unregister(µread_driver); } module_init(microread_mei_init); diff --git a/drivers/nfc/pn544/mei.c b/drivers/nfc/pn544/mei.c index 101a37e12efa..46d0eb24eef9 100644 --- a/drivers/nfc/pn544/mei.c +++ b/drivers/nfc/pn544/mei.c @@ -27,7 +27,7 @@ #define PN544_DRIVER_NAME "pn544" -static int pn544_mei_probe(struct mei_cl_device *device, +static int pn544_mei_probe(struct mei_cl_device *cldev, const struct mei_cl_device_id *id) { struct nfc_mei_phy *phy; @@ -35,7 +35,7 @@ static int pn544_mei_probe(struct mei_cl_device *device, pr_info("Probing NFC pn544\n"); - phy = nfc_mei_phy_alloc(device); + phy = nfc_mei_phy_alloc(cldev); if (!phy) { pr_err("Cannot allocate memory for pn544 mei phy.\n"); return -ENOMEM; @@ -53,9 +53,9 @@ static int pn544_mei_probe(struct mei_cl_device *device, return 0; } -static int pn544_mei_remove(struct mei_cl_device *device) +static int pn544_mei_remove(struct mei_cl_device *cldev) { - struct nfc_mei_phy *phy = mei_cl_get_drvdata(device); + struct nfc_mei_phy *phy = mei_cldev_get_drvdata(cldev); pr_info("Removing pn544\n"); @@ -67,7 +67,7 @@ static int pn544_mei_remove(struct mei_cl_device *device) } static struct mei_cl_device_id pn544_mei_tbl[] = { - { PN544_DRIVER_NAME, MEI_NFC_UUID}, + { PN544_DRIVER_NAME, MEI_NFC_UUID, MEI_CL_VERSION_ANY}, /* required last entry */ { } @@ -88,7 +88,7 @@ static int pn544_mei_init(void) pr_debug(DRIVER_DESC ": %s\n", __func__); - r = mei_cl_driver_register(&pn544_driver); + r = mei_cldev_driver_register(&pn544_driver); if (r) { pr_err(PN544_DRIVER_NAME ": driver registration failed\n"); return r; @@ -99,7 +99,7 @@ static int pn544_mei_init(void) static void pn544_mei_exit(void) { - mei_cl_driver_unregister(&pn544_driver); + mei_cldev_driver_unregister(&pn544_driver); } module_init(pn544_mei_init); diff --git a/drivers/nvmem/Kconfig b/drivers/nvmem/Kconfig index 8db297821f78..bc4ea585b42e 100644 --- a/drivers/nvmem/Kconfig +++ b/drivers/nvmem/Kconfig @@ -14,6 +14,28 @@ menuconfig NVMEM if NVMEM +config NVMEM_IMX_OCOTP + tristate "i.MX6 On-Chip OTP Controller support" + depends on SOC_IMX6 + help + This is a driver for the On-Chip OTP Controller (OCOTP) available on + i.MX6 SoCs, providing access to 4 Kbits of one-time programmable + eFuses. + + This driver can also be built as a module. If so, the module + will be called nvmem-imx-ocotp. + +config NVMEM_MXS_OCOTP + tristate "Freescale MXS On-Chip OTP Memory Support" + depends on ARCH_MXS || COMPILE_TEST + help + If you say Y here, you will get readonly access to the + One Time Programmable memory pages that are stored + on the Freescale i.MX23/i.MX28 processor. + + This driver can also be built as a module. If so, the module + will be called nvmem-mxs-ocotp. + config QCOM_QFPROM tristate "QCOM QFPROM Support" depends on ARCH_QCOM || COMPILE_TEST @@ -25,6 +47,16 @@ config QCOM_QFPROM This driver can also be built as a module. If so, the module will be called nvmem_qfprom. +config ROCKCHIP_EFUSE + tristate "Rockchip eFuse Support" + depends on ARCH_ROCKCHIP || COMPILE_TEST + help + This is a simple drive to dump specified values of Rockchip SoC + from eFuse, such as cpu-leakage. + + This driver can also be built as a module. If so, the module + will be called nvmem_rockchip_efuse. + config NVMEM_SUNXI_SID tristate "Allwinner SoCs SID support" depends on ARCH_SUNXI @@ -36,4 +68,14 @@ config NVMEM_SUNXI_SID This driver can also be built as a module. If so, the module will be called nvmem_sunxi_sid. +config NVMEM_VF610_OCOTP + tristate "VF610 SoC OCOTP support" + depends on SOC_VF610 || COMPILE_TEST + help + This is a driver for the 'OCOTP' peripheral available on Vybrid + devices like VF5xx and VF6xx. + + This driver can also be build as a module. If so, the module will + be called nvmem-vf610-ocotp. + endif diff --git a/drivers/nvmem/Makefile b/drivers/nvmem/Makefile index 4328b930ad9a..95dde3f8f085 100644 --- a/drivers/nvmem/Makefile +++ b/drivers/nvmem/Makefile @@ -6,7 +6,15 @@ obj-$(CONFIG_NVMEM) += nvmem_core.o nvmem_core-y := core.o # Devices +obj-$(CONFIG_NVMEM_IMX_OCOTP) += nvmem-imx-ocotp.o +nvmem-imx-ocotp-y := imx-ocotp.o +obj-$(CONFIG_NVMEM_MXS_OCOTP) += nvmem-mxs-ocotp.o +nvmem-mxs-ocotp-y := mxs-ocotp.o obj-$(CONFIG_QCOM_QFPROM) += nvmem_qfprom.o nvmem_qfprom-y := qfprom.o +obj-$(CONFIG_ROCKCHIP_EFUSE) += nvmem_rockchip_efuse.o +nvmem_rockchip_efuse-y := rockchip-efuse.o obj-$(CONFIG_NVMEM_SUNXI_SID) += nvmem_sunxi_sid.o nvmem_sunxi_sid-y := sunxi_sid.o +obj-$(CONFIG_NVMEM_VF610_OCOTP) += nvmem-vf610-ocotp.o +nvmem-vf610-ocotp-y := vf610-ocotp.o diff --git a/drivers/nvmem/core.c b/drivers/nvmem/core.c index d3c6676b3c0c..6fd4e5a5ef4a 100644 --- a/drivers/nvmem/core.c +++ b/drivers/nvmem/core.c @@ -67,7 +67,7 @@ static ssize_t bin_attr_nvmem_read(struct file *filp, struct kobject *kobj, int rc; /* Stop the user from reading */ - if (pos > nvmem->size) + if (pos >= nvmem->size) return 0; if (pos + count > nvmem->size) @@ -92,7 +92,7 @@ static ssize_t bin_attr_nvmem_write(struct file *filp, struct kobject *kobj, int rc; /* Stop the user from writing */ - if (pos > nvmem->size) + if (pos >= nvmem->size) return 0; if (pos + count > nvmem->size) @@ -825,7 +825,7 @@ static int __nvmem_cell_read(struct nvmem_device *nvmem, return rc; /* shift bits in-place */ - if (cell->bit_offset || cell->bit_offset) + if (cell->bit_offset || cell->nbits) nvmem_shift_read_buffer_in_place(cell, buf); *len = cell->bytes; @@ -938,7 +938,7 @@ int nvmem_cell_write(struct nvmem_cell *cell, void *buf, size_t len) rc = regmap_raw_write(nvmem->regmap, cell->offset, buf, cell->bytes); /* free the tmp buffer */ - if (cell->bit_offset) + if (cell->bit_offset || cell->nbits) kfree(buf); if (IS_ERR_VALUE(rc)) diff --git a/drivers/nvmem/imx-ocotp.c b/drivers/nvmem/imx-ocotp.c new file mode 100644 index 000000000000..b7971d410b60 --- /dev/null +++ b/drivers/nvmem/imx-ocotp.c @@ -0,0 +1,154 @@ +/* + * i.MX6 OCOTP fusebox driver + * + * Copyright (c) 2015 Pengutronix, Philipp Zabel <p.zabel@pengutronix.de> + * + * Based on the barebox ocotp driver, + * Copyright (c) 2010 Baruch Siach <baruch@tkos.co.il>, + * Orex Computed Radiography + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation. + * + * http://www.opensource.org/licenses/gpl-license.html + * http://www.gnu.org/copyleft/gpl.html + */ + +#include <linux/device.h> +#include <linux/io.h> +#include <linux/module.h> +#include <linux/nvmem-provider.h> +#include <linux/of.h> +#include <linux/of_device.h> +#include <linux/platform_device.h> +#include <linux/regmap.h> +#include <linux/slab.h> + +struct ocotp_priv { + struct device *dev; + void __iomem *base; + unsigned int nregs; +}; + +static int imx_ocotp_read(void *context, const void *reg, size_t reg_size, + void *val, size_t val_size) +{ + struct ocotp_priv *priv = context; + unsigned int offset = *(u32 *)reg; + unsigned int count; + int i; + u32 index; + + index = offset >> 2; + count = val_size >> 2; + + if (count > (priv->nregs - index)) + count = priv->nregs - index; + + for (i = index; i < (index + count); i++) { + *(u32 *)val = readl(priv->base + 0x400 + i * 0x10); + val += 4; + } + + return (i - index) * 4; +} + +static int imx_ocotp_write(void *context, const void *data, size_t count) +{ + /* Not implemented */ + return 0; +} + +static struct regmap_bus imx_ocotp_bus = { + .read = imx_ocotp_read, + .write = imx_ocotp_write, + .reg_format_endian_default = REGMAP_ENDIAN_NATIVE, + .val_format_endian_default = REGMAP_ENDIAN_NATIVE, +}; + +static bool imx_ocotp_writeable_reg(struct device *dev, unsigned int reg) +{ + return false; +} + +static struct regmap_config imx_ocotp_regmap_config = { + .reg_bits = 32, + .val_bits = 32, + .reg_stride = 4, + .writeable_reg = imx_ocotp_writeable_reg, + .name = "imx-ocotp", +}; + +static struct nvmem_config imx_ocotp_nvmem_config = { + .name = "imx-ocotp", + .read_only = true, + .owner = THIS_MODULE, +}; + +static const struct of_device_id imx_ocotp_dt_ids[] = { + { .compatible = "fsl,imx6q-ocotp", (void *)128 }, + { .compatible = "fsl,imx6sl-ocotp", (void *)32 }, + { .compatible = "fsl,imx6sx-ocotp", (void *)128 }, + { }, +}; +MODULE_DEVICE_TABLE(of, imx_ocotp_dt_ids); + +static int imx_ocotp_probe(struct platform_device *pdev) +{ + const struct of_device_id *of_id; + struct device *dev = &pdev->dev; + struct resource *res; + struct regmap *regmap; + struct ocotp_priv *priv; + struct nvmem_device *nvmem; + + priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL); + if (!priv) + return -ENOMEM; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + priv->base = devm_ioremap_resource(dev, res); + if (IS_ERR(priv->base)) + return PTR_ERR(priv->base); + + of_id = of_match_device(imx_ocotp_dt_ids, dev); + priv->nregs = (unsigned int)of_id->data; + imx_ocotp_regmap_config.max_register = 4 * priv->nregs - 4; + + regmap = devm_regmap_init(dev, &imx_ocotp_bus, priv, + &imx_ocotp_regmap_config); + if (IS_ERR(regmap)) { + dev_err(dev, "regmap init failed\n"); + return PTR_ERR(regmap); + } + imx_ocotp_nvmem_config.dev = dev; + nvmem = nvmem_register(&imx_ocotp_nvmem_config); + if (IS_ERR(nvmem)) + return PTR_ERR(nvmem); + + platform_set_drvdata(pdev, nvmem); + + return 0; +} + +static int imx_ocotp_remove(struct platform_device *pdev) +{ + struct nvmem_device *nvmem = platform_get_drvdata(pdev); + + return nvmem_unregister(nvmem); +} + +static struct platform_driver imx_ocotp_driver = { + .probe = imx_ocotp_probe, + .remove = imx_ocotp_remove, + .driver = { + .name = "imx_ocotp", + .of_match_table = imx_ocotp_dt_ids, + }, +}; +module_platform_driver(imx_ocotp_driver); + +MODULE_AUTHOR("Philipp Zabel <p.zabel@pengutronix.de>"); +MODULE_DESCRIPTION("i.MX6 OCOTP fuse box driver"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/nvmem/mxs-ocotp.c b/drivers/nvmem/mxs-ocotp.c new file mode 100644 index 000000000000..8ba19bba3156 --- /dev/null +++ b/drivers/nvmem/mxs-ocotp.c @@ -0,0 +1,257 @@ +/* + * Freescale MXS On-Chip OTP driver + * + * Copyright (C) 2015 Stefan Wahren <stefan.wahren@i2se.com> + * + * Based on the driver from Huang Shijie and Christoph G. Baumann + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + */ +#include <linux/clk.h> +#include <linux/delay.h> +#include <linux/device.h> +#include <linux/err.h> +#include <linux/io.h> +#include <linux/module.h> +#include <linux/nvmem-provider.h> +#include <linux/of_device.h> +#include <linux/platform_device.h> +#include <linux/regmap.h> +#include <linux/slab.h> +#include <linux/stmp_device.h> + +/* OCOTP registers and bits */ + +#define BM_OCOTP_CTRL_RD_BANK_OPEN BIT(12) +#define BM_OCOTP_CTRL_ERROR BIT(9) +#define BM_OCOTP_CTRL_BUSY BIT(8) + +#define OCOTP_TIMEOUT 10000 +#define OCOTP_DATA_OFFSET 0x20 + +struct mxs_ocotp { + struct clk *clk; + void __iomem *base; + struct nvmem_device *nvmem; +}; + +static int mxs_ocotp_wait(struct mxs_ocotp *otp) +{ + int timeout = OCOTP_TIMEOUT; + unsigned int status = 0; + + while (timeout--) { + status = readl(otp->base); + + if (!(status & (BM_OCOTP_CTRL_BUSY | BM_OCOTP_CTRL_ERROR))) + break; + + cpu_relax(); + } + + if (status & BM_OCOTP_CTRL_BUSY) + return -EBUSY; + else if (status & BM_OCOTP_CTRL_ERROR) + return -EIO; + + return 0; +} + +static int mxs_ocotp_read(void *context, const void *reg, size_t reg_size, + void *val, size_t val_size) +{ + struct mxs_ocotp *otp = context; + unsigned int offset = *(u32 *)reg; + u32 *buf = val; + int ret; + + ret = clk_enable(otp->clk); + if (ret) + return ret; + + writel(BM_OCOTP_CTRL_ERROR, otp->base + STMP_OFFSET_REG_CLR); + + ret = mxs_ocotp_wait(otp); + if (ret) + goto disable_clk; + + /* open OCOTP banks for read */ + writel(BM_OCOTP_CTRL_RD_BANK_OPEN, otp->base + STMP_OFFSET_REG_SET); + + /* approximately wait 33 hclk cycles */ + udelay(1); + + ret = mxs_ocotp_wait(otp); + if (ret) + goto close_banks; + + while (val_size) { + if ((offset < OCOTP_DATA_OFFSET) || (offset % 16)) { + /* fill up non-data register */ + *buf = 0; + } else { + *buf = readl(otp->base + offset); + } + + buf++; + val_size--; + offset += reg_size; + } + +close_banks: + /* close banks for power saving */ + writel(BM_OCOTP_CTRL_RD_BANK_OPEN, otp->base + STMP_OFFSET_REG_CLR); + +disable_clk: + clk_disable(otp->clk); + + return ret; +} + +static int mxs_ocotp_write(void *context, const void *data, size_t count) +{ + /* We don't want to support writing */ + return 0; +} + +static bool mxs_ocotp_writeable_reg(struct device *dev, unsigned int reg) +{ + return false; +} + +static struct nvmem_config ocotp_config = { + .name = "mxs-ocotp", + .owner = THIS_MODULE, +}; + +static const struct regmap_range imx23_ranges[] = { + regmap_reg_range(OCOTP_DATA_OFFSET, 0x210), +}; + +static const struct regmap_access_table imx23_access = { + .yes_ranges = imx23_ranges, + .n_yes_ranges = ARRAY_SIZE(imx23_ranges), +}; + +static const struct regmap_range imx28_ranges[] = { + regmap_reg_range(OCOTP_DATA_OFFSET, 0x290), +}; + +static const struct regmap_access_table imx28_access = { + .yes_ranges = imx28_ranges, + .n_yes_ranges = ARRAY_SIZE(imx28_ranges), +}; + +static struct regmap_bus mxs_ocotp_bus = { + .read = mxs_ocotp_read, + .write = mxs_ocotp_write, /* make regmap_init() happy */ + .reg_format_endian_default = REGMAP_ENDIAN_NATIVE, + .val_format_endian_default = REGMAP_ENDIAN_NATIVE, +}; + +static struct regmap_config mxs_ocotp_config = { + .reg_bits = 32, + .val_bits = 32, + .reg_stride = 16, + .writeable_reg = mxs_ocotp_writeable_reg, +}; + +static const struct of_device_id mxs_ocotp_match[] = { + { .compatible = "fsl,imx23-ocotp", .data = &imx23_access }, + { .compatible = "fsl,imx28-ocotp", .data = &imx28_access }, + { /* sentinel */}, +}; +MODULE_DEVICE_TABLE(of, mxs_ocotp_match); + +static int mxs_ocotp_probe(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct mxs_ocotp *otp; + struct resource *res; + const struct of_device_id *match; + struct regmap *regmap; + const struct regmap_access_table *access; + int ret; + + match = of_match_device(dev->driver->of_match_table, dev); + if (!match || !match->data) + return -EINVAL; + + otp = devm_kzalloc(dev, sizeof(*otp), GFP_KERNEL); + if (!otp) + return -ENOMEM; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + otp->base = devm_ioremap_resource(dev, res); + if (IS_ERR(otp->base)) + return PTR_ERR(otp->base); + + otp->clk = devm_clk_get(&pdev->dev, NULL); + if (IS_ERR(otp->clk)) + return PTR_ERR(otp->clk); + + ret = clk_prepare(otp->clk); + if (ret < 0) { + dev_err(dev, "failed to prepare clk: %d\n", ret); + return ret; + } + + access = match->data; + mxs_ocotp_config.rd_table = access; + mxs_ocotp_config.max_register = access->yes_ranges[0].range_max; + + regmap = devm_regmap_init(dev, &mxs_ocotp_bus, otp, &mxs_ocotp_config); + if (IS_ERR(regmap)) { + dev_err(dev, "regmap init failed\n"); + ret = PTR_ERR(regmap); + goto err_clk; + } + + ocotp_config.dev = dev; + otp->nvmem = nvmem_register(&ocotp_config); + if (IS_ERR(otp->nvmem)) { + ret = PTR_ERR(otp->nvmem); + goto err_clk; + } + + platform_set_drvdata(pdev, otp); + + return 0; + +err_clk: + clk_unprepare(otp->clk); + + return ret; +} + +static int mxs_ocotp_remove(struct platform_device *pdev) +{ + struct mxs_ocotp *otp = platform_get_drvdata(pdev); + + clk_unprepare(otp->clk); + + return nvmem_unregister(otp->nvmem); +} + +static struct platform_driver mxs_ocotp_driver = { + .probe = mxs_ocotp_probe, + .remove = mxs_ocotp_remove, + .driver = { + .name = "mxs-ocotp", + .of_match_table = mxs_ocotp_match, + }, +}; + +module_platform_driver(mxs_ocotp_driver); +MODULE_AUTHOR("Stefan Wahren <stefan.wahren@i2se.com>"); +MODULE_DESCRIPTION("driver for OCOTP in i.MX23/i.MX28"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/nvmem/rockchip-efuse.c b/drivers/nvmem/rockchip-efuse.c new file mode 100644 index 000000000000..f55213424222 --- /dev/null +++ b/drivers/nvmem/rockchip-efuse.c @@ -0,0 +1,186 @@ +/* + * Rockchip eFuse Driver + * + * Copyright (c) 2015 Rockchip Electronics Co. Ltd. + * Author: Caesar Wang <wxt@rock-chips.com> + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of version 2 of the GNU General Public License as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ + +#include <linux/platform_device.h> +#include <linux/nvmem-provider.h> +#include <linux/slab.h> +#include <linux/regmap.h> +#include <linux/device.h> +#include <linux/io.h> +#include <linux/module.h> +#include <linux/delay.h> +#include <linux/of.h> +#include <linux/clk.h> + +#define EFUSE_A_SHIFT 6 +#define EFUSE_A_MASK 0x3ff +#define EFUSE_PGENB BIT(3) +#define EFUSE_LOAD BIT(2) +#define EFUSE_STROBE BIT(1) +#define EFUSE_CSB BIT(0) + +#define REG_EFUSE_CTRL 0x0000 +#define REG_EFUSE_DOUT 0x0004 + +struct rockchip_efuse_context { + struct device *dev; + void __iomem *base; + struct clk *efuse_clk; +}; + +static int rockchip_efuse_write(void *context, const void *data, size_t count) +{ + /* Nothing TBD, Read-Only */ + return 0; +} + +static int rockchip_efuse_read(void *context, + const void *reg, size_t reg_size, + void *val, size_t val_size) +{ + unsigned int offset = *(u32 *)reg; + struct rockchip_efuse_context *_context = context; + void __iomem *base = _context->base; + struct clk *clk = _context->efuse_clk; + u8 *buf = val; + int ret; + + ret = clk_prepare_enable(clk); + if (ret < 0) { + dev_err(_context->dev, "failed to prepare/enable efuse clk\n"); + return ret; + } + + writel(EFUSE_LOAD | EFUSE_PGENB, base + REG_EFUSE_CTRL); + udelay(1); + while (val_size) { + writel(readl(base + REG_EFUSE_CTRL) & + (~(EFUSE_A_MASK << EFUSE_A_SHIFT)), + base + REG_EFUSE_CTRL); + writel(readl(base + REG_EFUSE_CTRL) | + ((offset & EFUSE_A_MASK) << EFUSE_A_SHIFT), + base + REG_EFUSE_CTRL); + udelay(1); + writel(readl(base + REG_EFUSE_CTRL) | + EFUSE_STROBE, base + REG_EFUSE_CTRL); + udelay(1); + *buf++ = readb(base + REG_EFUSE_DOUT); + writel(readl(base + REG_EFUSE_CTRL) & + (~EFUSE_STROBE), base + REG_EFUSE_CTRL); + udelay(1); + + val_size -= 1; + offset += 1; + } + + /* Switch to standby mode */ + writel(EFUSE_PGENB | EFUSE_CSB, base + REG_EFUSE_CTRL); + + clk_disable_unprepare(clk); + + return 0; +} + +static struct regmap_bus rockchip_efuse_bus = { + .read = rockchip_efuse_read, + .write = rockchip_efuse_write, + .reg_format_endian_default = REGMAP_ENDIAN_NATIVE, + .val_format_endian_default = REGMAP_ENDIAN_NATIVE, +}; + +static struct regmap_config rockchip_efuse_regmap_config = { + .reg_bits = 32, + .reg_stride = 1, + .val_bits = 8, +}; + +static struct nvmem_config econfig = { + .name = "rockchip-efuse", + .owner = THIS_MODULE, + .read_only = true, +}; + +static const struct of_device_id rockchip_efuse_match[] = { + { .compatible = "rockchip,rockchip-efuse",}, + { /* sentinel */}, +}; +MODULE_DEVICE_TABLE(of, rockchip_efuse_match); + +static int rockchip_efuse_probe(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct resource *res; + struct nvmem_device *nvmem; + struct regmap *regmap; + void __iomem *base; + struct clk *clk; + struct rockchip_efuse_context *context; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + base = devm_ioremap_resource(dev, res); + if (IS_ERR(base)) + return PTR_ERR(base); + + context = devm_kzalloc(dev, sizeof(struct rockchip_efuse_context), + GFP_KERNEL); + if (IS_ERR(context)) + return PTR_ERR(context); + + clk = devm_clk_get(dev, "pclk_efuse"); + if (IS_ERR(clk)) + return PTR_ERR(clk); + + context->dev = dev; + context->base = base; + context->efuse_clk = clk; + + rockchip_efuse_regmap_config.max_register = resource_size(res) - 1; + + regmap = devm_regmap_init(dev, &rockchip_efuse_bus, + context, &rockchip_efuse_regmap_config); + if (IS_ERR(regmap)) { + dev_err(dev, "regmap init failed\n"); + return PTR_ERR(regmap); + } + econfig.dev = dev; + nvmem = nvmem_register(&econfig); + if (IS_ERR(nvmem)) + return PTR_ERR(nvmem); + + platform_set_drvdata(pdev, nvmem); + + return 0; +} + +static int rockchip_efuse_remove(struct platform_device *pdev) +{ + struct nvmem_device *nvmem = platform_get_drvdata(pdev); + + return nvmem_unregister(nvmem); +} + +static struct platform_driver rockchip_efuse_driver = { + .probe = rockchip_efuse_probe, + .remove = rockchip_efuse_remove, + .driver = { + .name = "rockchip-efuse", + .of_match_table = rockchip_efuse_match, + }, +}; + +module_platform_driver(rockchip_efuse_driver); +MODULE_DESCRIPTION("rockchip_efuse driver"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/nvmem/sunxi_sid.c b/drivers/nvmem/sunxi_sid.c index 14777dd5212d..cfa3b85064dd 100644 --- a/drivers/nvmem/sunxi_sid.c +++ b/drivers/nvmem/sunxi_sid.c @@ -103,7 +103,7 @@ static int sunxi_sid_probe(struct platform_device *pdev) struct nvmem_device *nvmem; struct regmap *regmap; struct sunxi_sid *sid; - int i, size; + int ret, i, size; char *randomness; sid = devm_kzalloc(dev, sizeof(*sid), GFP_KERNEL); @@ -131,6 +131,11 @@ static int sunxi_sid_probe(struct platform_device *pdev) return PTR_ERR(nvmem); randomness = kzalloc(sizeof(u8) * size, GFP_KERNEL); + if (!randomness) { + ret = -EINVAL; + goto err_unreg_nvmem; + } + for (i = 0; i < size; i++) randomness[i] = sunxi_sid_read_byte(sid, i); @@ -140,6 +145,10 @@ static int sunxi_sid_probe(struct platform_device *pdev) platform_set_drvdata(pdev, nvmem); return 0; + +err_unreg_nvmem: + nvmem_unregister(nvmem); + return ret; } static int sunxi_sid_remove(struct platform_device *pdev) diff --git a/drivers/nvmem/vf610-ocotp.c b/drivers/nvmem/vf610-ocotp.c new file mode 100644 index 000000000000..8641319efeda --- /dev/null +++ b/drivers/nvmem/vf610-ocotp.c @@ -0,0 +1,302 @@ +/* + * Copyright (C) 2015 Toradex AG. + * + * Author: Sanchayan Maity <sanchayan.maity@toradex.com> + * + * Based on the barebox ocotp driver, + * Copyright (c) 2010 Baruch Siach <baruch@tkos.co.il> + * Orex Computed Radiography + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 and + * only version 2 as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <linux/clk.h> +#include <linux/delay.h> +#include <linux/device.h> +#include <linux/io.h> +#include <linux/module.h> +#include <linux/nvmem-provider.h> +#include <linux/of.h> +#include <linux/platform_device.h> +#include <linux/regmap.h> +#include <linux/slab.h> + +/* OCOTP Register Offsets */ +#define OCOTP_CTRL_REG 0x00 +#define OCOTP_CTRL_SET 0x04 +#define OCOTP_CTRL_CLR 0x08 +#define OCOTP_TIMING 0x10 +#define OCOTP_DATA 0x20 +#define OCOTP_READ_CTRL_REG 0x30 +#define OCOTP_READ_FUSE_DATA 0x40 + +/* OCOTP Register bits and masks */ +#define OCOTP_CTRL_WR_UNLOCK 16 +#define OCOTP_CTRL_WR_UNLOCK_KEY 0x3E77 +#define OCOTP_CTRL_WR_UNLOCK_MASK GENMASK(31, 16) +#define OCOTP_CTRL_ADDR 0 +#define OCOTP_CTRL_ADDR_MASK GENMASK(6, 0) +#define OCOTP_CTRL_RELOAD_SHADOWS BIT(10) +#define OCOTP_CTRL_ERR BIT(9) +#define OCOTP_CTRL_BUSY BIT(8) + +#define OCOTP_TIMING_STROBE_READ 16 +#define OCOTP_TIMING_STROBE_READ_MASK GENMASK(21, 16) +#define OCOTP_TIMING_RELAX 12 +#define OCOTP_TIMING_RELAX_MASK GENMASK(15, 12) +#define OCOTP_TIMING_STROBE_PROG 0 +#define OCOTP_TIMING_STROBE_PROG_MASK GENMASK(11, 0) + +#define OCOTP_READ_CTRL_READ_FUSE 0x1 + +#define VF610_OCOTP_TIMEOUT 100000 + +#define BF(value, field) (((value) << field) & field##_MASK) + +#define DEF_RELAX 20 + +static const int base_to_fuse_addr_mappings[][2] = { + {0x400, 0x00}, + {0x410, 0x01}, + {0x420, 0x02}, + {0x450, 0x05}, + {0x4F0, 0x0F}, + {0x600, 0x20}, + {0x610, 0x21}, + {0x620, 0x22}, + {0x630, 0x23}, + {0x640, 0x24}, + {0x650, 0x25}, + {0x660, 0x26}, + {0x670, 0x27}, + {0x6F0, 0x2F}, + {0x880, 0x38}, + {0x890, 0x39}, + {0x8A0, 0x3A}, + {0x8B0, 0x3B}, + {0x8C0, 0x3C}, + {0x8D0, 0x3D}, + {0x8E0, 0x3E}, + {0x8F0, 0x3F}, + {0xC80, 0x78}, + {0xC90, 0x79}, + {0xCA0, 0x7A}, + {0xCB0, 0x7B}, + {0xCC0, 0x7C}, + {0xCD0, 0x7D}, + {0xCE0, 0x7E}, + {0xCF0, 0x7F}, +}; + +struct vf610_ocotp { + void __iomem *base; + struct clk *clk; + struct device *dev; + struct nvmem_device *nvmem; + int timing; +}; + +static int vf610_ocotp_wait_busy(void __iomem *base) +{ + int timeout = VF610_OCOTP_TIMEOUT; + + while ((readl(base) & OCOTP_CTRL_BUSY) && --timeout) + udelay(10); + + if (!timeout) { + writel(OCOTP_CTRL_ERR, base + OCOTP_CTRL_CLR); + return -ETIMEDOUT; + } + + udelay(10); + + return 0; +} + +static int vf610_ocotp_calculate_timing(struct vf610_ocotp *ocotp_dev) +{ + u32 clk_rate; + u32 relax, strobe_read, strobe_prog; + u32 timing; + + clk_rate = clk_get_rate(ocotp_dev->clk); + + /* Refer section OTP read/write timing parameters in TRM */ + relax = clk_rate / (1000000000 / DEF_RELAX) - 1; + strobe_prog = clk_rate / (1000000000 / 10000) + 2 * (DEF_RELAX + 1) - 1; + strobe_read = clk_rate / (1000000000 / 40) + 2 * (DEF_RELAX + 1) - 1; + + timing = BF(relax, OCOTP_TIMING_RELAX); + timing |= BF(strobe_read, OCOTP_TIMING_STROBE_READ); + timing |= BF(strobe_prog, OCOTP_TIMING_STROBE_PROG); + + return timing; +} + +static int vf610_get_fuse_address(int base_addr_offset) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(base_to_fuse_addr_mappings); i++) { + if (base_to_fuse_addr_mappings[i][0] == base_addr_offset) + return base_to_fuse_addr_mappings[i][1]; + } + + return -EINVAL; +} + +static int vf610_ocotp_write(void *context, const void *data, size_t count) +{ + return 0; +} + +static int vf610_ocotp_read(void *context, + const void *off, size_t reg_size, + void *val, size_t val_size) +{ + struct vf610_ocotp *ocotp = context; + void __iomem *base = ocotp->base; + unsigned int offset = *(u32 *)off; + u32 reg, *buf = val; + int fuse_addr; + int ret; + + while (val_size > 0) { + fuse_addr = vf610_get_fuse_address(offset); + if (fuse_addr > 0) { + writel(ocotp->timing, base + OCOTP_TIMING); + ret = vf610_ocotp_wait_busy(base + OCOTP_CTRL_REG); + if (ret) + return ret; + + reg = readl(base + OCOTP_CTRL_REG); + reg &= ~OCOTP_CTRL_ADDR_MASK; + reg &= ~OCOTP_CTRL_WR_UNLOCK_MASK; + reg |= BF(fuse_addr, OCOTP_CTRL_ADDR); + writel(reg, base + OCOTP_CTRL_REG); + + writel(OCOTP_READ_CTRL_READ_FUSE, + base + OCOTP_READ_CTRL_REG); + ret = vf610_ocotp_wait_busy(base + OCOTP_CTRL_REG); + if (ret) + return ret; + + if (readl(base) & OCOTP_CTRL_ERR) { + dev_dbg(ocotp->dev, "Error reading from fuse address %x\n", + fuse_addr); + writel(OCOTP_CTRL_ERR, base + OCOTP_CTRL_CLR); + } + + /* + * In case of error, we do not abort and expect to read + * 0xBADABADA as mentioned by the TRM. We just read this + * value and return. + */ + *buf = readl(base + OCOTP_READ_FUSE_DATA); + } else { + *buf = 0; + } + + buf++; + val_size--; + offset += reg_size; + } + + return 0; +} + +static struct regmap_bus vf610_ocotp_bus = { + .read = vf610_ocotp_read, + .write = vf610_ocotp_write, + .reg_format_endian_default = REGMAP_ENDIAN_NATIVE, + .val_format_endian_default = REGMAP_ENDIAN_NATIVE, +}; + +static struct regmap_config ocotp_regmap_config = { + .reg_bits = 32, + .val_bits = 32, + .reg_stride = 4, +}; + +static struct nvmem_config ocotp_config = { + .name = "ocotp", + .owner = THIS_MODULE, +}; + +static const struct of_device_id ocotp_of_match[] = { + { .compatible = "fsl,vf610-ocotp", }, + {/* sentinel */}, +}; +MODULE_DEVICE_TABLE(of, ocotp_of_match); + +static int vf610_ocotp_remove(struct platform_device *pdev) +{ + struct vf610_ocotp *ocotp_dev = platform_get_drvdata(pdev); + + return nvmem_unregister(ocotp_dev->nvmem); +} + +static int vf610_ocotp_probe(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct resource *res; + struct regmap *regmap; + struct vf610_ocotp *ocotp_dev; + + ocotp_dev = devm_kzalloc(&pdev->dev, + sizeof(struct vf610_ocotp), GFP_KERNEL); + if (!ocotp_dev) + return -ENOMEM; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + ocotp_dev->base = devm_ioremap_resource(dev, res); + if (IS_ERR(ocotp_dev->base)) + return PTR_ERR(ocotp_dev->base); + + ocotp_dev->clk = devm_clk_get(dev, NULL); + if (IS_ERR(ocotp_dev->clk)) { + dev_err(dev, "failed getting clock, err = %ld\n", + PTR_ERR(ocotp_dev->clk)); + return PTR_ERR(ocotp_dev->clk); + } + + ocotp_regmap_config.max_register = resource_size(res); + regmap = devm_regmap_init(dev, + &vf610_ocotp_bus, ocotp_dev, &ocotp_regmap_config); + if (IS_ERR(regmap)) { + dev_err(dev, "regmap init failed\n"); + return PTR_ERR(regmap); + } + ocotp_config.dev = dev; + + ocotp_dev->nvmem = nvmem_register(&ocotp_config); + if (IS_ERR(ocotp_dev->nvmem)) + return PTR_ERR(ocotp_dev->nvmem); + + ocotp_dev->dev = dev; + platform_set_drvdata(pdev, ocotp_dev); + + ocotp_dev->timing = vf610_ocotp_calculate_timing(ocotp_dev); + + return 0; +} + +static struct platform_driver vf610_ocotp_driver = { + .probe = vf610_ocotp_probe, + .remove = vf610_ocotp_remove, + .driver = { + .name = "vf610-ocotp", + .of_match_table = ocotp_of_match, + }, +}; +module_platform_driver(vf610_ocotp_driver); +MODULE_AUTHOR("Sanchayan Maity <sanchayan.maity@toradex.com>"); +MODULE_DESCRIPTION("Vybrid OCOTP driver"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index dd652f2ae03d..108a3118ace7 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -299,9 +299,10 @@ static long local_pci_probe(void *_ddi) * Unbound PCI devices are always put in D0, regardless of * runtime PM status. During probe, the device is set to * active and the usage count is incremented. If the driver - * supports runtime PM, it should call pm_runtime_put_noidle() - * in its probe routine and pm_runtime_get_noresume() in its - * remove routine. + * supports runtime PM, it should call pm_runtime_put_noidle(), + * or any other runtime PM helper function decrementing the usage + * count, in its probe routine and pm_runtime_get_noresume() in + * its remove routine. */ pm_runtime_get_sync(dev); pci_dev->driver = pci_drv; diff --git a/drivers/pcmcia/ds.c b/drivers/pcmcia/ds.c index 0decee6c556e..489ea1098c96 100644 --- a/drivers/pcmcia/ds.c +++ b/drivers/pcmcia/ds.c @@ -468,12 +468,10 @@ static int pcmcia_device_query(struct pcmcia_device *p_dev) if ((length < 2) || (length > 255)) continue; - new = kmalloc(sizeof(char) * length, GFP_KERNEL); + new = kstrdup(tmp, GFP_KERNEL); if (!new) continue; - new = strncpy(new, tmp, length); - tmp = p_dev->prod_id[i]; p_dev->prod_id[i] = new; kfree(tmp); diff --git a/drivers/phy/phy-berlin-sata.c b/drivers/phy/phy-berlin-sata.c index 0062027afb1e..77a2e054fdea 100644 --- a/drivers/phy/phy-berlin-sata.c +++ b/drivers/phy/phy-berlin-sata.c @@ -276,6 +276,7 @@ static const struct of_device_id phy_berlin_sata_of_match[] = { { .compatible = "marvell,berlin2q-sata-phy" }, { }, }; +MODULE_DEVICE_TABLE(of, phy_berlin_sata_of_match); static struct platform_driver phy_berlin_sata_driver = { .probe = phy_berlin_sata_probe, diff --git a/drivers/phy/phy-qcom-ufs.c b/drivers/phy/phy-qcom-ufs.c index 49a1ed0cef56..107cb57c3513 100644 --- a/drivers/phy/phy-qcom-ufs.c +++ b/drivers/phy/phy-qcom-ufs.c @@ -432,6 +432,7 @@ out_disable_src: out: return ret; } +EXPORT_SYMBOL_GPL(ufs_qcom_phy_enable_ref_clk); static int ufs_qcom_phy_disable_vreg(struct phy *phy, @@ -474,6 +475,7 @@ void ufs_qcom_phy_disable_ref_clk(struct phy *generic_phy) phy->is_ref_clk_enabled = false; } } +EXPORT_SYMBOL_GPL(ufs_qcom_phy_disable_ref_clk); #define UFS_REF_CLK_EN (1 << 5) @@ -517,11 +519,13 @@ void ufs_qcom_phy_enable_dev_ref_clk(struct phy *generic_phy) { ufs_qcom_phy_dev_ref_clk_ctrl(generic_phy, true); } +EXPORT_SYMBOL_GPL(ufs_qcom_phy_enable_dev_ref_clk); void ufs_qcom_phy_disable_dev_ref_clk(struct phy *generic_phy) { ufs_qcom_phy_dev_ref_clk_ctrl(generic_phy, false); } +EXPORT_SYMBOL_GPL(ufs_qcom_phy_disable_dev_ref_clk); /* Turn ON M-PHY RMMI interface clocks */ int ufs_qcom_phy_enable_iface_clk(struct phy *generic_phy) @@ -550,6 +554,7 @@ int ufs_qcom_phy_enable_iface_clk(struct phy *generic_phy) out: return ret; } +EXPORT_SYMBOL_GPL(ufs_qcom_phy_enable_iface_clk); /* Turn OFF M-PHY RMMI interface clocks */ void ufs_qcom_phy_disable_iface_clk(struct phy *generic_phy) @@ -562,6 +567,7 @@ void ufs_qcom_phy_disable_iface_clk(struct phy *generic_phy) phy->is_iface_clk_enabled = false; } } +EXPORT_SYMBOL_GPL(ufs_qcom_phy_disable_iface_clk); int ufs_qcom_phy_start_serdes(struct phy *generic_phy) { @@ -578,6 +584,7 @@ int ufs_qcom_phy_start_serdes(struct phy *generic_phy) return ret; } +EXPORT_SYMBOL_GPL(ufs_qcom_phy_start_serdes); int ufs_qcom_phy_set_tx_lane_enable(struct phy *generic_phy, u32 tx_lanes) { @@ -595,6 +602,7 @@ int ufs_qcom_phy_set_tx_lane_enable(struct phy *generic_phy, u32 tx_lanes) return ret; } +EXPORT_SYMBOL_GPL(ufs_qcom_phy_set_tx_lane_enable); void ufs_qcom_phy_save_controller_version(struct phy *generic_phy, u8 major, u16 minor, u16 step) @@ -605,6 +613,7 @@ void ufs_qcom_phy_save_controller_version(struct phy *generic_phy, ufs_qcom_phy->host_ctrl_rev_minor = minor; ufs_qcom_phy->host_ctrl_rev_step = step; } +EXPORT_SYMBOL_GPL(ufs_qcom_phy_save_controller_version); int ufs_qcom_phy_calibrate_phy(struct phy *generic_phy, bool is_rate_B) { @@ -625,6 +634,7 @@ int ufs_qcom_phy_calibrate_phy(struct phy *generic_phy, bool is_rate_B) return ret; } +EXPORT_SYMBOL_GPL(ufs_qcom_phy_calibrate_phy); int ufs_qcom_phy_remove(struct phy *generic_phy, struct ufs_qcom_phy *ufs_qcom_phy) @@ -662,6 +672,7 @@ int ufs_qcom_phy_is_pcs_ready(struct phy *generic_phy) return ufs_qcom_phy->phy_spec_ops-> is_physical_coding_sublayer_ready(ufs_qcom_phy); } +EXPORT_SYMBOL_GPL(ufs_qcom_phy_is_pcs_ready); int ufs_qcom_phy_power_on(struct phy *generic_phy) { diff --git a/drivers/phy/phy-rockchip-usb.c b/drivers/phy/phy-rockchip-usb.c index 5a5c073e72fe..91d6f342c565 100644 --- a/drivers/phy/phy-rockchip-usb.c +++ b/drivers/phy/phy-rockchip-usb.c @@ -98,6 +98,7 @@ static int rockchip_usb_phy_probe(struct platform_device *pdev) struct device_node *child; struct regmap *grf; unsigned int reg_offset; + int err; grf = syscon_regmap_lookup_by_phandle(dev->of_node, "rockchip,grf"); if (IS_ERR(grf)) { @@ -129,6 +130,11 @@ static int rockchip_usb_phy_probe(struct platform_device *pdev) return PTR_ERR(rk_phy->phy); } phy_set_drvdata(rk_phy->phy, rk_phy); + + /* only power up usb phy when it use, so disable it when init*/ + err = rockchip_usb_phy_power(rk_phy, 1); + if (err) + return err; } phy_provider = devm_of_phy_provider_register(dev, of_phy_simple_xlate); diff --git a/drivers/regulator/axp20x-regulator.c b/drivers/regulator/axp20x-regulator.c index 01bf3476a791..a9567af7cec0 100644 --- a/drivers/regulator/axp20x-regulator.c +++ b/drivers/regulator/axp20x-regulator.c @@ -192,9 +192,9 @@ static const struct regulator_desc axp22x_regulators[] = { AXP_DESC(AXP22X, DCDC3, "dcdc3", "vin3", 600, 1860, 20, AXP22X_DCDC3_V_OUT, 0x3f, AXP22X_PWR_OUT_CTRL1, BIT(3)), AXP_DESC(AXP22X, DCDC4, "dcdc4", "vin4", 600, 1540, 20, - AXP22X_DCDC4_V_OUT, 0x3f, AXP22X_PWR_OUT_CTRL1, BIT(3)), + AXP22X_DCDC4_V_OUT, 0x3f, AXP22X_PWR_OUT_CTRL1, BIT(4)), AXP_DESC(AXP22X, DCDC5, "dcdc5", "vin5", 1000, 2550, 50, - AXP22X_DCDC5_V_OUT, 0x1f, AXP22X_PWR_OUT_CTRL1, BIT(4)), + AXP22X_DCDC5_V_OUT, 0x1f, AXP22X_PWR_OUT_CTRL1, BIT(5)), /* secondary switchable output of DCDC1 */ AXP_DESC_SW(AXP22X, DC1SW, "dc1sw", "dcdc1", 1600, 3400, 100, AXP22X_DCDC1_V_OUT, 0x1f, AXP22X_PWR_OUT_CTRL2, BIT(7)), diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c index 7849187d91ae..8a34f6acc801 100644 --- a/drivers/regulator/core.c +++ b/drivers/regulator/core.c @@ -1403,6 +1403,10 @@ static int regulator_resolve_supply(struct regulator_dev *rdev) return 0; } + /* Did the lookup explicitly defer for us? */ + if (ret == -EPROBE_DEFER) + return ret; + if (have_full_constraints()) { r = dummy_regulator_rdev; } else { diff --git a/drivers/scsi/3w-9xxx.c b/drivers/scsi/3w-9xxx.c index add419d6ff34..a56a7b243e91 100644 --- a/drivers/scsi/3w-9xxx.c +++ b/drivers/scsi/3w-9xxx.c @@ -212,6 +212,17 @@ static const struct file_operations twa_fops = { .llseek = noop_llseek, }; +/* + * The controllers use an inline buffer instead of a mapped SGL for small, + * single entry buffers. Note that we treat a zero-length transfer like + * a mapped SGL. + */ +static bool twa_command_mapped(struct scsi_cmnd *cmd) +{ + return scsi_sg_count(cmd) != 1 || + scsi_bufflen(cmd) >= TW_MIN_SGL_LENGTH; +} + /* This function will complete an aen request from the isr */ static int twa_aen_complete(TW_Device_Extension *tw_dev, int request_id) { @@ -1339,7 +1350,8 @@ static irqreturn_t twa_interrupt(int irq, void *dev_instance) } /* Now complete the io */ - scsi_dma_unmap(cmd); + if (twa_command_mapped(cmd)) + scsi_dma_unmap(cmd); cmd->scsi_done(cmd); tw_dev->state[request_id] = TW_S_COMPLETED; twa_free_request_id(tw_dev, request_id); @@ -1582,7 +1594,8 @@ static int twa_reset_device_extension(TW_Device_Extension *tw_dev) struct scsi_cmnd *cmd = tw_dev->srb[i]; cmd->result = (DID_RESET << 16); - scsi_dma_unmap(cmd); + if (twa_command_mapped(cmd)) + scsi_dma_unmap(cmd); cmd->scsi_done(cmd); } } @@ -1765,12 +1778,14 @@ static int twa_scsi_queue_lck(struct scsi_cmnd *SCpnt, void (*done)(struct scsi_ retval = twa_scsiop_execute_scsi(tw_dev, request_id, NULL, 0, NULL); switch (retval) { case SCSI_MLQUEUE_HOST_BUSY: - scsi_dma_unmap(SCpnt); + if (twa_command_mapped(SCpnt)) + scsi_dma_unmap(SCpnt); twa_free_request_id(tw_dev, request_id); break; case 1: SCpnt->result = (DID_ERROR << 16); - scsi_dma_unmap(SCpnt); + if (twa_command_mapped(SCpnt)) + scsi_dma_unmap(SCpnt); done(SCpnt); tw_dev->state[request_id] = TW_S_COMPLETED; twa_free_request_id(tw_dev, request_id); @@ -1831,8 +1846,7 @@ static int twa_scsiop_execute_scsi(TW_Device_Extension *tw_dev, int request_id, /* Map sglist from scsi layer to cmd packet */ if (scsi_sg_count(srb)) { - if ((scsi_sg_count(srb) == 1) && - (scsi_bufflen(srb) < TW_MIN_SGL_LENGTH)) { + if (!twa_command_mapped(srb)) { if (srb->sc_data_direction == DMA_TO_DEVICE || srb->sc_data_direction == DMA_BIDIRECTIONAL) scsi_sg_copy_to_buffer(srb, @@ -1905,7 +1919,7 @@ static void twa_scsiop_execute_scsi_complete(TW_Device_Extension *tw_dev, int re { struct scsi_cmnd *cmd = tw_dev->srb[request_id]; - if (scsi_bufflen(cmd) < TW_MIN_SGL_LENGTH && + if (!twa_command_mapped(cmd) && (cmd->sc_data_direction == DMA_FROM_DEVICE || cmd->sc_data_direction == DMA_BIDIRECTIONAL)) { if (scsi_sg_count(cmd) == 1) { diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c index 33c74d3436c9..6bffd91b973a 100644 --- a/drivers/scsi/libiscsi.c +++ b/drivers/scsi/libiscsi.c @@ -976,13 +976,13 @@ static void iscsi_tmf_rsp(struct iscsi_conn *conn, struct iscsi_hdr *hdr) wake_up(&conn->ehwait); } -static void iscsi_send_nopout(struct iscsi_conn *conn, struct iscsi_nopin *rhdr) +static int iscsi_send_nopout(struct iscsi_conn *conn, struct iscsi_nopin *rhdr) { struct iscsi_nopout hdr; struct iscsi_task *task; if (!rhdr && conn->ping_task) - return; + return -EINVAL; memset(&hdr, 0, sizeof(struct iscsi_nopout)); hdr.opcode = ISCSI_OP_NOOP_OUT | ISCSI_OP_IMMEDIATE; @@ -996,13 +996,16 @@ static void iscsi_send_nopout(struct iscsi_conn *conn, struct iscsi_nopin *rhdr) hdr.ttt = RESERVED_ITT; task = __iscsi_conn_send_pdu(conn, (struct iscsi_hdr *)&hdr, NULL, 0); - if (!task) + if (!task) { iscsi_conn_printk(KERN_ERR, conn, "Could not send nopout\n"); - else if (!rhdr) { + return -EIO; + } else if (!rhdr) { /* only track our nops */ conn->ping_task = task; conn->last_ping = jiffies; } + + return 0; } static int iscsi_nop_out_rsp(struct iscsi_task *task, @@ -2092,8 +2095,10 @@ static void iscsi_check_transport_timeouts(unsigned long data) if (time_before_eq(last_recv + recv_timeout, jiffies)) { /* send a ping to try to provoke some traffic */ ISCSI_DBG_CONN(conn, "Sending nopout as ping\n"); - iscsi_send_nopout(conn, NULL); - next_timeout = conn->last_ping + (conn->ping_timeout * HZ); + if (iscsi_send_nopout(conn, NULL)) + next_timeout = jiffies + (1 * HZ); + else + next_timeout = conn->last_ping + (conn->ping_timeout * HZ); } else next_timeout = last_recv + recv_timeout; diff --git a/drivers/scsi/scsi_dh.c b/drivers/scsi/scsi_dh.c index edb044a7b56d..0a2168e69bbc 100644 --- a/drivers/scsi/scsi_dh.c +++ b/drivers/scsi/scsi_dh.c @@ -111,7 +111,7 @@ static struct scsi_device_handler *scsi_dh_lookup(const char *name) dh = __scsi_dh_lookup(name); if (!dh) { - request_module(name); + request_module("scsi_dh_%s", name); dh = __scsi_dh_lookup(name); } diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index cbfc5990052b..126a48c6431e 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -1957,7 +1957,7 @@ static int scsi_mq_prep_fn(struct request *req) static void scsi_mq_done(struct scsi_cmnd *cmd) { trace_scsi_dispatch_cmd_done(cmd); - blk_mq_complete_request(cmd->request); + blk_mq_complete_request(cmd->request, cmd->request->errors); } static int scsi_queue_rq(struct blk_mq_hw_ctx *hctx, diff --git a/drivers/spi/spi-davinci.c b/drivers/spi/spi-davinci.c index 3cf9faa6cc3f..a85d863d4a44 100644 --- a/drivers/spi/spi-davinci.c +++ b/drivers/spi/spi-davinci.c @@ -992,11 +992,12 @@ static int davinci_spi_probe(struct platform_device *pdev) goto free_master; } - dspi->irq = platform_get_irq(pdev, 0); - if (dspi->irq <= 0) { + ret = platform_get_irq(pdev, 0); + if (ret == 0) ret = -EINVAL; + if (ret < 0) goto free_master; - } + dspi->irq = ret; ret = devm_request_threaded_irq(&pdev->dev, dspi->irq, davinci_spi_irq, dummy_thread_fn, 0, dev_name(&pdev->dev), dspi); diff --git a/drivers/spmi/spmi-pmic-arb.c b/drivers/spmi/spmi-pmic-arb.c index 4a3cf9ba152f..96615d807694 100644 --- a/drivers/spmi/spmi-pmic-arb.c +++ b/drivers/spmi/spmi-pmic-arb.c @@ -168,11 +168,6 @@ struct pmic_arb_ver_ops { u32 (*irq_clear)(u8 n); }; -static inline u32 pmic_arb_base_read(struct spmi_pmic_arb_dev *dev, u32 offset) -{ - return readl_relaxed(dev->rd_base + offset); -} - static inline void pmic_arb_base_write(struct spmi_pmic_arb_dev *dev, u32 offset, u32 val) { @@ -193,7 +188,7 @@ static inline void pmic_arb_set_rd_cmd(struct spmi_pmic_arb_dev *dev, */ static void pa_read_data(struct spmi_pmic_arb_dev *dev, u8 *buf, u32 reg, u8 bc) { - u32 data = pmic_arb_base_read(dev, reg); + u32 data = __raw_readl(dev->rd_base + reg); memcpy(buf, &data, (bc & 3) + 1); } @@ -208,7 +203,7 @@ pa_write_data(struct spmi_pmic_arb_dev *dev, const u8 *buf, u32 reg, u8 bc) { u32 data = 0; memcpy(&data, buf, (bc & 3) + 1); - pmic_arb_base_write(dev, reg, data); + __raw_writel(data, dev->wr_base + reg); } static int pmic_arb_wait_for_done(struct spmi_controller *ctrl, @@ -365,7 +360,7 @@ static int pmic_arb_write_cmd(struct spmi_controller *ctrl, u8 opc, u8 sid, opc = PMIC_ARB_OP_EXT_WRITE; else if (opc >= 0x30 && opc <= 0x37) opc = PMIC_ARB_OP_EXT_WRITEL; - else if (opc >= 0x80 && opc <= 0xFF) + else if (opc >= 0x80) opc = PMIC_ARB_OP_ZERO_WRITE; else return -EINVAL; diff --git a/drivers/spmi/spmi.c b/drivers/spmi/spmi.c index 11467e17bdd8..6b3da1bb0d63 100644 --- a/drivers/spmi/spmi.c +++ b/drivers/spmi/spmi.c @@ -560,12 +560,13 @@ EXPORT_SYMBOL_GPL(spmi_controller_remove); * This API will register the client driver with the SPMI framework. * It is typically called from the driver's module-init function. */ -int spmi_driver_register(struct spmi_driver *sdrv) +int __spmi_driver_register(struct spmi_driver *sdrv, struct module *owner) { sdrv->driver.bus = &spmi_bus_type; + sdrv->driver.owner = owner; return driver_register(&sdrv->driver); } -EXPORT_SYMBOL_GPL(spmi_driver_register); +EXPORT_SYMBOL_GPL(__spmi_driver_register); static void __exit spmi_exit(void) { diff --git a/drivers/staging/speakup/fakekey.c b/drivers/staging/speakup/fakekey.c index 4299cf45f947..5e1f16c36b49 100644 --- a/drivers/staging/speakup/fakekey.c +++ b/drivers/staging/speakup/fakekey.c @@ -81,6 +81,7 @@ void speakup_fake_down_arrow(void) __this_cpu_write(reporting_keystroke, true); input_report_key(virt_keyboard, KEY_DOWN, PRESSED); input_report_key(virt_keyboard, KEY_DOWN, RELEASED); + input_sync(virt_keyboard); __this_cpu_write(reporting_keystroke, false); /* reenable preemption */ diff --git a/drivers/thermal/power_allocator.c b/drivers/thermal/power_allocator.c index 7ff96270c933..e570ff084add 100644 --- a/drivers/thermal/power_allocator.c +++ b/drivers/thermal/power_allocator.c @@ -144,6 +144,16 @@ static void estimate_pid_constants(struct thermal_zone_device *tz, switch_on_temp = 0; temperature_threshold = control_temp - switch_on_temp; + /* + * estimate_pid_constants() tries to find appropriate default + * values for thermal zones that don't provide them. If a + * system integrator has configured a thermal zone with two + * passive trip points at the same temperature, that person + * hasn't put any effort to set up the thermal zone properly + * so just give up. + */ + if (!temperature_threshold) + return; if (!tz->tzp->k_po || force) tz->tzp->k_po = int_to_frac(sustainable_power) / diff --git a/drivers/tty/n_tty.c b/drivers/tty/n_tty.c index 20932cc9c8f7..b09023b07169 100644 --- a/drivers/tty/n_tty.c +++ b/drivers/tty/n_tty.c @@ -343,8 +343,7 @@ static void n_tty_packet_mode_flush(struct tty_struct *tty) spin_lock_irqsave(&tty->ctrl_lock, flags); tty->ctrl_status |= TIOCPKT_FLUSHREAD; spin_unlock_irqrestore(&tty->ctrl_lock, flags); - if (waitqueue_active(&tty->link->read_wait)) - wake_up_interruptible(&tty->link->read_wait); + wake_up_interruptible(&tty->link->read_wait); } } @@ -1382,8 +1381,7 @@ handle_newline: put_tty_queue(c, ldata); smp_store_release(&ldata->canon_head, ldata->read_head); kill_fasync(&tty->fasync, SIGIO, POLL_IN); - if (waitqueue_active(&tty->read_wait)) - wake_up_interruptible_poll(&tty->read_wait, POLLIN); + wake_up_interruptible_poll(&tty->read_wait, POLLIN); return 0; } } @@ -1667,8 +1665,7 @@ static void __receive_buf(struct tty_struct *tty, const unsigned char *cp, if ((read_cnt(ldata) >= ldata->minimum_to_wake) || L_EXTPROC(tty)) { kill_fasync(&tty->fasync, SIGIO, POLL_IN); - if (waitqueue_active(&tty->read_wait)) - wake_up_interruptible_poll(&tty->read_wait, POLLIN); + wake_up_interruptible_poll(&tty->read_wait, POLLIN); } } @@ -1887,10 +1884,8 @@ static void n_tty_set_termios(struct tty_struct *tty, struct ktermios *old) } /* The termios change make the tty ready for I/O */ - if (waitqueue_active(&tty->write_wait)) - wake_up_interruptible(&tty->write_wait); - if (waitqueue_active(&tty->read_wait)) - wake_up_interruptible(&tty->read_wait); + wake_up_interruptible(&tty->write_wait); + wake_up_interruptible(&tty->read_wait); } /** diff --git a/drivers/tty/serial/8250/8250_port.c b/drivers/tty/serial/8250/8250_port.c index b1e0ba3e525b..0bbf34035d6a 100644 --- a/drivers/tty/serial/8250/8250_port.c +++ b/drivers/tty/serial/8250/8250_port.c @@ -261,6 +261,14 @@ configured less than Maximum supported fifo bytes */ UART_FCR7_64BYTE, .flags = UART_CAP_FIFO, }, + [PORT_RT2880] = { + .name = "Palmchip BK-3103", + .fifo_size = 16, + .tx_loadsz = 16, + .fcr = UART_FCR_ENABLE_FIFO | UART_FCR_R_TRIG_10, + .rxtrig_bytes = {1, 4, 8, 14}, + .flags = UART_CAP_FIFO, + }, }; /* Uart divisor latch read */ diff --git a/drivers/tty/serial/atmel_serial.c b/drivers/tty/serial/atmel_serial.c index 5ca5cf3e9359..538ea03bc101 100644 --- a/drivers/tty/serial/atmel_serial.c +++ b/drivers/tty/serial/atmel_serial.c @@ -2786,7 +2786,7 @@ static int atmel_serial_probe(struct platform_device *pdev) ret = atmel_init_gpios(port, &pdev->dev); if (ret < 0) { dev_err(&pdev->dev, "Failed to initialize GPIOs."); - goto err; + goto err_clear_bit; } ret = atmel_init_port(port, pdev); diff --git a/drivers/tty/serial/imx.c b/drivers/tty/serial/imx.c index fe3d41cc8416..d0388a071ba1 100644 --- a/drivers/tty/serial/imx.c +++ b/drivers/tty/serial/imx.c @@ -1631,12 +1631,12 @@ imx_console_write(struct console *co, const char *s, unsigned int count) int locked = 1; int retval; - retval = clk_prepare_enable(sport->clk_per); + retval = clk_enable(sport->clk_per); if (retval) return; - retval = clk_prepare_enable(sport->clk_ipg); + retval = clk_enable(sport->clk_ipg); if (retval) { - clk_disable_unprepare(sport->clk_per); + clk_disable(sport->clk_per); return; } @@ -1675,8 +1675,8 @@ imx_console_write(struct console *co, const char *s, unsigned int count) if (locked) spin_unlock_irqrestore(&sport->port.lock, flags); - clk_disable_unprepare(sport->clk_ipg); - clk_disable_unprepare(sport->clk_per); + clk_disable(sport->clk_ipg); + clk_disable(sport->clk_per); } /* @@ -1777,7 +1777,15 @@ imx_console_setup(struct console *co, char *options) retval = uart_set_options(&sport->port, co, baud, parity, bits, flow); - clk_disable_unprepare(sport->clk_ipg); + clk_disable(sport->clk_ipg); + if (retval) { + clk_unprepare(sport->clk_ipg); + goto error_console; + } + + retval = clk_prepare(sport->clk_per); + if (retval) + clk_disable_unprepare(sport->clk_ipg); error_console: return retval; diff --git a/drivers/tty/tty_buffer.c b/drivers/tty/tty_buffer.c index 5a3fa8913880..a660ab181cca 100644 --- a/drivers/tty/tty_buffer.c +++ b/drivers/tty/tty_buffer.c @@ -242,7 +242,10 @@ void tty_buffer_flush(struct tty_struct *tty, struct tty_ldisc *ld) atomic_inc(&buf->priority); mutex_lock(&buf->lock); - while ((next = buf->head->next) != NULL) { + /* paired w/ release in __tty_buffer_request_room; ensures there are + * no pending memory accesses to the freed buffer + */ + while ((next = smp_load_acquire(&buf->head->next)) != NULL) { tty_buffer_free(port, buf->head); buf->head = next; } @@ -290,7 +293,10 @@ static int __tty_buffer_request_room(struct tty_port *port, size_t size, if (n != NULL) { n->flags = flags; buf->tail = n; - b->commit = b->used; + /* paired w/ acquire in flush_to_ldisc(); ensures + * flush_to_ldisc() sees buffer data. + */ + smp_store_release(&b->commit, b->used); /* paired w/ acquire in flush_to_ldisc(); ensures the * latest commit value can be read before the head is * advanced to the next buffer @@ -393,7 +399,10 @@ void tty_schedule_flip(struct tty_port *port) { struct tty_bufhead *buf = &port->buf; - buf->tail->commit = buf->tail->used; + /* paired w/ acquire in flush_to_ldisc(); ensures + * flush_to_ldisc() sees buffer data. + */ + smp_store_release(&buf->tail->commit, buf->tail->used); schedule_work(&buf->work); } EXPORT_SYMBOL(tty_schedule_flip); @@ -467,7 +476,7 @@ static void flush_to_ldisc(struct work_struct *work) struct tty_struct *tty; struct tty_ldisc *disc; - tty = port->itty; + tty = READ_ONCE(port->itty); if (tty == NULL) return; @@ -491,7 +500,10 @@ static void flush_to_ldisc(struct work_struct *work) * is advancing to the next buffer */ next = smp_load_acquire(&head->next); - count = head->commit - head->read; + /* paired w/ release in __tty_buffer_request_room() or in + * tty_buffer_flush(); ensures we see the committed buffer data + */ + count = smp_load_acquire(&head->commit) - head->read; if (!count) { if (next == NULL) { check_other_closed(tty); diff --git a/drivers/tty/tty_io.c b/drivers/tty/tty_io.c index 02785d844354..2eefaa6e3e3a 100644 --- a/drivers/tty/tty_io.c +++ b/drivers/tty/tty_io.c @@ -2128,8 +2128,24 @@ retry_open: if (!noctty && current->signal->leader && !current->signal->tty && - tty->session == NULL) - __proc_set_tty(tty); + tty->session == NULL) { + /* + * Don't let a process that only has write access to the tty + * obtain the privileges associated with having a tty as + * controlling terminal (being able to reopen it with full + * access through /dev/tty, being able to perform pushback). + * Many distributions set the group of all ttys to "tty" and + * grant write-only access to all terminals for setgid tty + * binaries, which should not imply full privileges on all ttys. + * + * This could theoretically break old code that performs open() + * on a write-only file descriptor. In that case, it might be + * necessary to also permit this if + * inode_permission(inode, MAY_READ) == 0. + */ + if (filp->f_mode & FMODE_READ) + __proc_set_tty(tty); + } spin_unlock_irq(¤t->sighand->siglock); read_unlock(&tasklist_lock); tty_unlock(tty); @@ -2418,7 +2434,7 @@ static int fionbio(struct file *file, int __user *p) * Takes ->siglock() when updating signal->tty */ -static int tiocsctty(struct tty_struct *tty, int arg) +static int tiocsctty(struct tty_struct *tty, struct file *file, int arg) { int ret = 0; @@ -2452,6 +2468,13 @@ static int tiocsctty(struct tty_struct *tty, int arg) goto unlock; } } + + /* See the comment in tty_open(). */ + if ((file->f_mode & FMODE_READ) == 0 && !capable(CAP_SYS_ADMIN)) { + ret = -EPERM; + goto unlock; + } + proc_set_tty(tty); unlock: read_unlock(&tasklist_lock); @@ -2844,7 +2867,7 @@ long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg) no_tty(); return 0; case TIOCSCTTY: - return tiocsctty(tty, arg); + return tiocsctty(tty, file, arg); case TIOCGPGRP: return tiocgpgrp(tty, real_tty, p); case TIOCSPGRP: @@ -3151,13 +3174,18 @@ struct class *tty_class; static int tty_cdev_add(struct tty_driver *driver, dev_t dev, unsigned int index, unsigned int count) { + int err; + /* init here, since reused cdevs cause crashes */ driver->cdevs[index] = cdev_alloc(); if (!driver->cdevs[index]) return -ENOMEM; - cdev_init(driver->cdevs[index], &tty_fops); + driver->cdevs[index]->ops = &tty_fops; driver->cdevs[index]->owner = driver->owner; - return cdev_add(driver->cdevs[index], dev, count); + err = cdev_add(driver->cdevs[index], dev, count); + if (err) + kobject_put(&driver->cdevs[index]->kobj); + return err; } /** diff --git a/drivers/uio/uio.c b/drivers/uio/uio.c index 8196581f54c2..bcc1fc027311 100644 --- a/drivers/uio/uio.c +++ b/drivers/uio/uio.c @@ -524,6 +524,7 @@ static ssize_t uio_read(struct file *filep, char __user *buf, event_count = atomic_read(&idev->event); if (event_count != listener->event_count) { + __set_current_state(TASK_RUNNING); if (copy_to_user(buf, &event_count, count)) retval = -EFAULT; else { diff --git a/drivers/uio/uio_fsl_elbc_gpcm.c b/drivers/uio/uio_fsl_elbc_gpcm.c index 2bcf80c159c1..b46323d9dc18 100644 --- a/drivers/uio/uio_fsl_elbc_gpcm.c +++ b/drivers/uio/uio_fsl_elbc_gpcm.c @@ -470,6 +470,7 @@ static const struct of_device_id uio_fsl_elbc_gpcm_match[] = { { .compatible = "fsl,elbc-gpcm-uio", }, {} }; +MODULE_DEVICE_TABLE(of, uio_fsl_elbc_gpcm_match); static struct platform_driver uio_fsl_elbc_gpcm_driver = { .driver = { diff --git a/drivers/usb/core/quirks.c b/drivers/usb/core/quirks.c index d85abfed84cc..f5a381945db2 100644 --- a/drivers/usb/core/quirks.c +++ b/drivers/usb/core/quirks.c @@ -54,6 +54,13 @@ static const struct usb_device_id usb_quirk_list[] = { { USB_DEVICE(0x046d, 0x082d), .driver_info = USB_QUIRK_DELAY_INIT }, { USB_DEVICE(0x046d, 0x0843), .driver_info = USB_QUIRK_DELAY_INIT }, + /* Logitech ConferenceCam CC3000e */ + { USB_DEVICE(0x046d, 0x0847), .driver_info = USB_QUIRK_DELAY_INIT }, + { USB_DEVICE(0x046d, 0x0848), .driver_info = USB_QUIRK_DELAY_INIT }, + + /* Logitech PTZ Pro Camera */ + { USB_DEVICE(0x046d, 0x0853), .driver_info = USB_QUIRK_DELAY_INIT }, + /* Logitech Quickcam Fusion */ { USB_DEVICE(0x046d, 0x08c1), .driver_info = USB_QUIRK_RESET_RESUME }, @@ -78,6 +85,12 @@ static const struct usb_device_id usb_quirk_list[] = { /* Philips PSC805 audio device */ { USB_DEVICE(0x0471, 0x0155), .driver_info = USB_QUIRK_RESET_RESUME }, + /* Plantronic Audio 655 DSP */ + { USB_DEVICE(0x047f, 0xc008), .driver_info = USB_QUIRK_RESET_RESUME }, + + /* Plantronic Audio 648 USB */ + { USB_DEVICE(0x047f, 0xc013), .driver_info = USB_QUIRK_RESET_RESUME }, + /* Artisman Watchdog Dongle */ { USB_DEVICE(0x04b4, 0x0526), .driver_info = USB_QUIRK_CONFIG_INTF_STRINGS }, diff --git a/drivers/usb/gadget/udc/bdc/bdc_ep.c b/drivers/usb/gadget/udc/bdc/bdc_ep.c index d1b81539d632..d6199507f861 100644 --- a/drivers/usb/gadget/udc/bdc/bdc_ep.c +++ b/drivers/usb/gadget/udc/bdc/bdc_ep.c @@ -159,8 +159,10 @@ static int ep_bd_list_alloc(struct bdc_ep *ep) bd_table->start_bd = dma_pool_alloc(bdc->bd_table_pool, GFP_ATOMIC, &dma); - if (!bd_table->start_bd) + if (!bd_table->start_bd) { + kfree(bd_table); goto fail; + } bd_table->dma = dma; diff --git a/drivers/usb/misc/chaoskey.c b/drivers/usb/misc/chaoskey.c index 3ad5d19e4d04..23c794813e6a 100644 --- a/drivers/usb/misc/chaoskey.c +++ b/drivers/usb/misc/chaoskey.c @@ -472,7 +472,7 @@ static int chaoskey_rng_read(struct hwrng *rng, void *data, if (this_time > max) this_time = max; - memcpy(data, dev->buf, this_time); + memcpy(data, dev->buf + dev->used, this_time); dev->used += this_time; diff --git a/drivers/usb/renesas_usbhs/common.c b/drivers/usb/renesas_usbhs/common.c index 7b98e1d9194c..d82fa36c3465 100644 --- a/drivers/usb/renesas_usbhs/common.c +++ b/drivers/usb/renesas_usbhs/common.c @@ -476,6 +476,11 @@ static const struct of_device_id usbhs_of_match[] = { .compatible = "renesas,usbhs-r8a7794", .data = (void *)USBHS_TYPE_RCAR_GEN2, }, + { + /* Gen3 is compatible with Gen2 */ + .compatible = "renesas,usbhs-r8a7795", + .data = (void *)USBHS_TYPE_RCAR_GEN2, + }, { }, }; MODULE_DEVICE_TABLE(of, usbhs_of_match); @@ -493,7 +498,7 @@ static struct renesas_usbhs_platform_info *usbhs_parse_dt(struct device *dev) return NULL; dparam = &info->driver_param; - dparam->type = of_id ? (u32)of_id->data : 0; + dparam->type = of_id ? (uintptr_t)of_id->data : 0; if (!of_property_read_u32(dev->of_node, "renesas,buswait", &tmp)) dparam->buswait_bwait = tmp; gpio = of_get_named_gpio_flags(dev->of_node, "renesas,enable-gpio", 0, diff --git a/drivers/video/fbdev/broadsheetfb.c b/drivers/video/fbdev/broadsheetfb.c index 0e5fde1d3ffb..9f9a7bef1ff6 100644 --- a/drivers/video/fbdev/broadsheetfb.c +++ b/drivers/video/fbdev/broadsheetfb.c @@ -752,7 +752,7 @@ static ssize_t broadsheet_loadstore_waveform(struct device *dev, if ((fw_entry->size < 8*1024) || (fw_entry->size > 64*1024)) { dev_err(dev, "Invalid waveform\n"); err = -EINVAL; - goto err_failed; + goto err_fw; } mutex_lock(&(par->io_lock)); @@ -762,13 +762,15 @@ static ssize_t broadsheet_loadstore_waveform(struct device *dev, mutex_unlock(&(par->io_lock)); if (err < 0) { dev_err(dev, "Failed to store broadsheet waveform\n"); - goto err_failed; + goto err_fw; } dev_info(dev, "Stored broadsheet waveform, size %zd\n", fw_entry->size); - return len; + err = len; +err_fw: + release_firmware(fw_entry); err_failed: return err; } diff --git a/drivers/video/fbdev/fsl-diu-fb.c b/drivers/video/fbdev/fsl-diu-fb.c index 7fa2e6f9e322..b335c1ae8625 100644 --- a/drivers/video/fbdev/fsl-diu-fb.c +++ b/drivers/video/fbdev/fsl-diu-fb.c @@ -1628,9 +1628,16 @@ static int fsl_diu_suspend(struct platform_device *ofdev, pm_message_t state) static int fsl_diu_resume(struct platform_device *ofdev) { struct fsl_diu_data *data; + unsigned int i; data = dev_get_drvdata(&ofdev->dev); - enable_lcdc(data->fsl_diu_info); + + fsl_diu_enable_interrupts(data); + update_lcdc(data->fsl_diu_info); + for (i = 0; i < NUM_AOIS; i++) { + if (data->mfb[i].count) + fsl_diu_enable_panel(&data->fsl_diu_info[i]); + } return 0; } diff --git a/drivers/video/fbdev/mb862xx/mb862xxfbdrv.c b/drivers/video/fbdev/mb862xx/mb862xxfbdrv.c index 9b8bebdf8f86..f9ec5c0484fa 100644 --- a/drivers/video/fbdev/mb862xx/mb862xxfbdrv.c +++ b/drivers/video/fbdev/mb862xx/mb862xxfbdrv.c @@ -831,6 +831,7 @@ static struct of_device_id of_platform_mb862xx_tbl[] = { { .compatible = "fujitsu,coral", }, { /* end */ } }; +MODULE_DEVICE_TABLE(of, of_platform_mb862xx_tbl); static struct platform_driver of_platform_mb862xxfb_driver = { .driver = { diff --git a/drivers/video/fbdev/omap2/displays-new/connector-dvi.c b/drivers/video/fbdev/omap2/displays-new/connector-dvi.c index a8ce920fa797..d811e6dcaef7 100644 --- a/drivers/video/fbdev/omap2/displays-new/connector-dvi.c +++ b/drivers/video/fbdev/omap2/displays-new/connector-dvi.c @@ -294,7 +294,7 @@ static int dvic_probe_of(struct platform_device *pdev) adapter_node = of_parse_phandle(node, "ddc-i2c-bus", 0); if (adapter_node) { - adapter = of_find_i2c_adapter_by_node(adapter_node); + adapter = of_get_i2c_adapter_by_node(adapter_node); if (adapter == NULL) { dev_err(&pdev->dev, "failed to parse ddc-i2c-bus\n"); omap_dss_put_device(ddata->in); diff --git a/drivers/video/fbdev/omap2/displays-new/panel-sony-acx565akm.c b/drivers/video/fbdev/omap2/displays-new/panel-sony-acx565akm.c index 90cbc4c3406c..c581231c74a5 100644 --- a/drivers/video/fbdev/omap2/displays-new/panel-sony-acx565akm.c +++ b/drivers/video/fbdev/omap2/displays-new/panel-sony-acx565akm.c @@ -898,6 +898,7 @@ static const struct of_device_id acx565akm_of_match[] = { { .compatible = "omapdss,sony,acx565akm", }, {}, }; +MODULE_DEVICE_TABLE(of, acx565akm_of_match); static struct spi_driver acx565akm_driver = { .driver = { diff --git a/drivers/video/fbdev/tridentfb.c b/drivers/video/fbdev/tridentfb.c index 7ed9a227f5ea..01b43e9ce941 100644 --- a/drivers/video/fbdev/tridentfb.c +++ b/drivers/video/fbdev/tridentfb.c @@ -226,7 +226,7 @@ static void blade_image_blit(struct tridentfb_par *par, const char *data, writemmr(par, DST1, point(x, y)); writemmr(par, DST2, point(x + w - 1, y + h - 1)); - memcpy(par->io_virt + 0x10000, data, 4 * size); + iowrite32_rep(par->io_virt + 0x10000, data, size); } static void blade_copy_rect(struct tridentfb_par *par, @@ -673,8 +673,14 @@ static int get_nativex(struct tridentfb_par *par) static inline void set_lwidth(struct tridentfb_par *par, int width) { write3X4(par, VGA_CRTC_OFFSET, width & 0xFF); - write3X4(par, AddColReg, - (read3X4(par, AddColReg) & 0xCF) | ((width & 0x300) >> 4)); + /* chips older than TGUI9660 have only 1 width bit in AddColReg */ + /* touching the other one breaks I2C/DDC */ + if (par->chip_id == TGUI9440 || par->chip_id == CYBER9320) + write3X4(par, AddColReg, + (read3X4(par, AddColReg) & 0xEF) | ((width & 0x100) >> 4)); + else + write3X4(par, AddColReg, + (read3X4(par, AddColReg) & 0xCF) | ((width & 0x300) >> 4)); } /* For resolutions smaller than FP resolution stretch */ diff --git a/drivers/video/of_display_timing.c b/drivers/video/of_display_timing.c index 32d8275e4c88..8a1076beecd3 100644 --- a/drivers/video/of_display_timing.c +++ b/drivers/video/of_display_timing.c @@ -210,6 +210,7 @@ struct display_timings *of_get_display_timings(struct device_node *np) */ pr_err("%s: error in timing %d\n", of_node_full_name(np), disp->num_timings + 1); + kfree(dt); goto timingfail; } diff --git a/drivers/w1/masters/omap_hdq.c b/drivers/w1/masters/omap_hdq.c index e7d448963a24..0e2f43bccf1f 100644 --- a/drivers/w1/masters/omap_hdq.c +++ b/drivers/w1/masters/omap_hdq.c @@ -17,6 +17,7 @@ #include <linux/io.h> #include <linux/sched.h> #include <linux/pm_runtime.h> +#include <linux/of.h> #include "../w1.h" #include "../w1_int.h" @@ -27,21 +28,23 @@ #define OMAP_HDQ_TX_DATA 0x04 #define OMAP_HDQ_RX_DATA 0x08 #define OMAP_HDQ_CTRL_STATUS 0x0c -#define OMAP_HDQ_CTRL_STATUS_INTERRUPTMASK (1<<6) -#define OMAP_HDQ_CTRL_STATUS_CLOCKENABLE (1<<5) -#define OMAP_HDQ_CTRL_STATUS_GO (1<<4) -#define OMAP_HDQ_CTRL_STATUS_INITIALIZATION (1<<2) -#define OMAP_HDQ_CTRL_STATUS_DIR (1<<1) -#define OMAP_HDQ_CTRL_STATUS_MODE (1<<0) +#define OMAP_HDQ_CTRL_STATUS_SINGLE BIT(7) +#define OMAP_HDQ_CTRL_STATUS_INTERRUPTMASK BIT(6) +#define OMAP_HDQ_CTRL_STATUS_CLOCKENABLE BIT(5) +#define OMAP_HDQ_CTRL_STATUS_GO BIT(4) +#define OMAP_HDQ_CTRL_STATUS_PRESENCE BIT(3) +#define OMAP_HDQ_CTRL_STATUS_INITIALIZATION BIT(2) +#define OMAP_HDQ_CTRL_STATUS_DIR BIT(1) #define OMAP_HDQ_INT_STATUS 0x10 -#define OMAP_HDQ_INT_STATUS_TXCOMPLETE (1<<2) -#define OMAP_HDQ_INT_STATUS_RXCOMPLETE (1<<1) -#define OMAP_HDQ_INT_STATUS_TIMEOUT (1<<0) +#define OMAP_HDQ_INT_STATUS_TXCOMPLETE BIT(2) +#define OMAP_HDQ_INT_STATUS_RXCOMPLETE BIT(1) +#define OMAP_HDQ_INT_STATUS_TIMEOUT BIT(0) #define OMAP_HDQ_SYSCONFIG 0x14 -#define OMAP_HDQ_SYSCONFIG_SOFTRESET (1<<1) -#define OMAP_HDQ_SYSCONFIG_AUTOIDLE (1<<0) +#define OMAP_HDQ_SYSCONFIG_SOFTRESET BIT(1) +#define OMAP_HDQ_SYSCONFIG_AUTOIDLE BIT(0) +#define OMAP_HDQ_SYSCONFIG_NOIDLE 0x0 #define OMAP_HDQ_SYSSTATUS 0x18 -#define OMAP_HDQ_SYSSTATUS_RESETDONE (1<<0) +#define OMAP_HDQ_SYSSTATUS_RESETDONE BIT(0) #define OMAP_HDQ_FLAG_CLEAR 0 #define OMAP_HDQ_FLAG_SET 1 @@ -67,6 +70,10 @@ struct hdq_data { * the data wrire or read. */ int init_trans; + int rrw; + /* mode: 0-HDQ 1-W1 */ + int mode; + }; static int omap_hdq_probe(struct platform_device *pdev); @@ -74,6 +81,7 @@ static int omap_hdq_remove(struct platform_device *pdev); static const struct of_device_id omap_hdq_dt_ids[] = { { .compatible = "ti,omap3-1w" }, + { .compatible = "ti,am4372-hdq" }, {} }; MODULE_DEVICE_TABLE(of, omap_hdq_dt_ids); @@ -90,15 +98,12 @@ static struct platform_driver omap_hdq_driver = { static u8 omap_w1_read_byte(void *_hdq); static void omap_w1_write_byte(void *_hdq, u8 byte); static u8 omap_w1_reset_bus(void *_hdq); -static void omap_w1_search_bus(void *_hdq, struct w1_master *master_dev, - u8 search_type, w1_slave_found_callback slave_found); static struct w1_bus_master omap_w1_master = { .read_byte = omap_w1_read_byte, .write_byte = omap_w1_write_byte, .reset_bus = omap_w1_reset_bus, - .search = omap_w1_search_bus, }; /* HDQ register I/O routines */ @@ -122,6 +127,15 @@ static inline u8 hdq_reg_merge(struct hdq_data *hdq_data, u32 offset, return new_val; } +static void hdq_disable_interrupt(struct hdq_data *hdq_data, u32 offset, + u32 mask) +{ + u32 ie; + + ie = readl(hdq_data->hdq_base + offset); + writel(ie & mask, hdq_data->hdq_base + offset); +} + /* * Wait for one or more bits in flag change. * HDQ_FLAG_SET: wait until any bit in the flag is set. @@ -229,13 +243,7 @@ static irqreturn_t hdq_isr(int irq, void *_hdq) return IRQ_HANDLED; } -/* HDQ Mode: always return success */ -static u8 omap_w1_reset_bus(void *_hdq) -{ - return 0; -} - -/* W1 search callback function */ +/* W1 search callback function in HDQ mode */ static void omap_w1_search_bus(void *_hdq, struct w1_master *master_dev, u8 search_type, w1_slave_found_callback slave_found) { @@ -262,9 +270,10 @@ static int _omap_hdq_reset(struct hdq_data *hdq_data) int ret; u8 tmp_status; - hdq_reg_out(hdq_data, OMAP_HDQ_SYSCONFIG, OMAP_HDQ_SYSCONFIG_SOFTRESET); + hdq_reg_out(hdq_data, OMAP_HDQ_SYSCONFIG, + OMAP_HDQ_SYSCONFIG_SOFTRESET); /* - * Select HDQ mode & enable clocks. + * Select HDQ/1W mode & enable clocks. * It is observed that INT flags can't be cleared via a read and GO/INIT * won't return to zero if interrupt is disabled. So we always enable * interrupt. @@ -282,7 +291,8 @@ static int _omap_hdq_reset(struct hdq_data *hdq_data) else { hdq_reg_out(hdq_data, OMAP_HDQ_CTRL_STATUS, OMAP_HDQ_CTRL_STATUS_CLOCKENABLE | - OMAP_HDQ_CTRL_STATUS_INTERRUPTMASK); + OMAP_HDQ_CTRL_STATUS_INTERRUPTMASK | + hdq_data->mode); hdq_reg_out(hdq_data, OMAP_HDQ_SYSCONFIG, OMAP_HDQ_SYSCONFIG_AUTOIDLE); } @@ -334,6 +344,18 @@ static int omap_hdq_break(struct hdq_data *hdq_data) ret = -ETIMEDOUT; goto out; } + + /* + * check for the presence detect bit to get + * set to show that the slave is responding + */ + if (!(hdq_reg_in(hdq_data, OMAP_HDQ_CTRL_STATUS) & + OMAP_HDQ_CTRL_STATUS_PRESENCE)) { + dev_dbg(hdq_data->dev, "Presence bit not set\n"); + ret = -ETIMEDOUT; + goto out; + } + /* * wait for both INIT and GO bits rerurn to zero. * zero wait time expected for interrupt mode. @@ -368,6 +390,8 @@ static int hdq_read_byte(struct hdq_data *hdq_data, u8 *val) goto out; } + hdq_data->hdq_irqstatus = 0; + if (!(hdq_data->hdq_irqstatus & OMAP_HDQ_INT_STATUS_RXCOMPLETE)) { hdq_reg_merge(hdq_data, OMAP_HDQ_CTRL_STATUS, OMAP_HDQ_CTRL_STATUS_DIR | OMAP_HDQ_CTRL_STATUS_GO, @@ -400,7 +424,7 @@ rtn: } -/* Enable clocks and set the controller to HDQ mode */ +/* Enable clocks and set the controller to HDQ/1W mode */ static int omap_hdq_get(struct hdq_data *hdq_data) { int ret = 0; @@ -422,7 +446,7 @@ static int omap_hdq_get(struct hdq_data *hdq_data) pm_runtime_get_sync(hdq_data->dev); - /* make sure HDQ is out of reset */ + /* make sure HDQ/1W is out of reset */ if (!(hdq_reg_in(hdq_data, OMAP_HDQ_SYSSTATUS) & OMAP_HDQ_SYSSTATUS_RESETDONE)) { ret = _omap_hdq_reset(hdq_data); @@ -430,12 +454,13 @@ static int omap_hdq_get(struct hdq_data *hdq_data) /* back up the count */ hdq_data->hdq_usecount--; } else { - /* select HDQ mode & enable clocks */ + /* select HDQ/1W mode & enable clocks */ hdq_reg_out(hdq_data, OMAP_HDQ_CTRL_STATUS, OMAP_HDQ_CTRL_STATUS_CLOCKENABLE | - OMAP_HDQ_CTRL_STATUS_INTERRUPTMASK); + OMAP_HDQ_CTRL_STATUS_INTERRUPTMASK | + hdq_data->mode); hdq_reg_out(hdq_data, OMAP_HDQ_SYSCONFIG, - OMAP_HDQ_SYSCONFIG_AUTOIDLE); + OMAP_HDQ_SYSCONFIG_NOIDLE); hdq_reg_in(hdq_data, OMAP_HDQ_INT_STATUS); } } @@ -456,6 +481,8 @@ static int omap_hdq_put(struct hdq_data *hdq_data) if (ret < 0) return -EINTR; + hdq_reg_out(hdq_data, OMAP_HDQ_SYSCONFIG, + OMAP_HDQ_SYSCONFIG_AUTOIDLE); if (0 == hdq_data->hdq_usecount) { dev_dbg(hdq_data->dev, "attempt to decrement use count" " when it is zero"); @@ -471,6 +498,100 @@ static int omap_hdq_put(struct hdq_data *hdq_data) return ret; } +/* + * W1 triplet callback function - used for searching ROM addresses. + * Registered only when controller is in 1-wire mode. + */ +static u8 omap_w1_triplet(void *_hdq, u8 bdir) +{ + u8 id_bit, comp_bit; + int err; + u8 ret = 0x3; /* no slaves responded */ + struct hdq_data *hdq_data = _hdq; + u8 ctrl = OMAP_HDQ_CTRL_STATUS_SINGLE | OMAP_HDQ_CTRL_STATUS_GO | + OMAP_HDQ_CTRL_STATUS_INTERRUPTMASK; + u8 mask = ctrl | OMAP_HDQ_CTRL_STATUS_DIR; + + omap_hdq_get(_hdq); + + err = mutex_lock_interruptible(&hdq_data->hdq_mutex); + if (err < 0) { + dev_dbg(hdq_data->dev, "Could not acquire mutex\n"); + goto rtn; + } + + hdq_data->hdq_irqstatus = 0; + /* read id_bit */ + hdq_reg_merge(_hdq, OMAP_HDQ_CTRL_STATUS, + ctrl | OMAP_HDQ_CTRL_STATUS_DIR, mask); + err = wait_event_timeout(hdq_wait_queue, + (hdq_data->hdq_irqstatus + & OMAP_HDQ_INT_STATUS_RXCOMPLETE), + OMAP_HDQ_TIMEOUT); + if (err == 0) { + dev_dbg(hdq_data->dev, "RX wait elapsed\n"); + goto out; + } + id_bit = (hdq_reg_in(_hdq, OMAP_HDQ_RX_DATA) & 0x01); + + hdq_data->hdq_irqstatus = 0; + /* read comp_bit */ + hdq_reg_merge(_hdq, OMAP_HDQ_CTRL_STATUS, + ctrl | OMAP_HDQ_CTRL_STATUS_DIR, mask); + err = wait_event_timeout(hdq_wait_queue, + (hdq_data->hdq_irqstatus + & OMAP_HDQ_INT_STATUS_RXCOMPLETE), + OMAP_HDQ_TIMEOUT); + if (err == 0) { + dev_dbg(hdq_data->dev, "RX wait elapsed\n"); + goto out; + } + comp_bit = (hdq_reg_in(_hdq, OMAP_HDQ_RX_DATA) & 0x01); + + if (id_bit && comp_bit) { + ret = 0x03; /* no slaves responded */ + goto out; + } + if (!id_bit && !comp_bit) { + /* Both bits are valid, take the direction given */ + ret = bdir ? 0x04 : 0; + } else { + /* Only one bit is valid, take that direction */ + bdir = id_bit; + ret = id_bit ? 0x05 : 0x02; + } + + /* write bdir bit */ + hdq_reg_out(_hdq, OMAP_HDQ_TX_DATA, bdir); + hdq_reg_merge(_hdq, OMAP_HDQ_CTRL_STATUS, ctrl, mask); + err = wait_event_timeout(hdq_wait_queue, + (hdq_data->hdq_irqstatus + & OMAP_HDQ_INT_STATUS_TXCOMPLETE), + OMAP_HDQ_TIMEOUT); + if (err == 0) { + dev_dbg(hdq_data->dev, "TX wait elapsed\n"); + goto out; + } + + hdq_reg_merge(_hdq, OMAP_HDQ_CTRL_STATUS, 0, + OMAP_HDQ_CTRL_STATUS_SINGLE); + +out: + mutex_unlock(&hdq_data->hdq_mutex); +rtn: + omap_hdq_put(_hdq); + return ret; +} + +/* reset callback */ +static u8 omap_w1_reset_bus(void *_hdq) +{ + omap_hdq_get(_hdq); + omap_hdq_break(_hdq); + omap_hdq_put(_hdq); + return 0; +} + /* Read a byte of data from the device */ static u8 omap_w1_read_byte(void *_hdq) { @@ -478,6 +599,10 @@ static u8 omap_w1_read_byte(void *_hdq) u8 val = 0; int ret; + /* First write to initialize the transfer */ + if (hdq_data->init_trans == 0) + omap_hdq_get(hdq_data); + ret = hdq_read_byte(hdq_data, &val); if (ret) { ret = mutex_lock_interruptible(&hdq_data->hdq_mutex); @@ -491,6 +616,10 @@ static u8 omap_w1_read_byte(void *_hdq) return -1; } + hdq_disable_interrupt(hdq_data, OMAP_HDQ_CTRL_STATUS, + ~OMAP_HDQ_CTRL_STATUS_INTERRUPTMASK); + hdq_data->hdq_usecount = 0; + /* Write followed by a read, release the module */ if (hdq_data->init_trans) { ret = mutex_lock_interruptible(&hdq_data->hdq_mutex); @@ -517,6 +646,14 @@ static void omap_w1_write_byte(void *_hdq, u8 byte) if (hdq_data->init_trans == 0) omap_hdq_get(hdq_data); + /* + * We need to reset the slave before + * issuing the SKIP ROM command, else + * the slave will not work. + */ + if (byte == W1_SKIP_ROM) + omap_hdq_break(hdq_data); + ret = mutex_lock_interruptible(&hdq_data->hdq_mutex); if (ret < 0) { dev_dbg(hdq_data->dev, "Could not acquire mutex\n"); @@ -551,6 +688,7 @@ static int omap_hdq_probe(struct platform_device *pdev) struct resource *res; int ret, irq; u8 rev; + const char *mode; hdq_data = devm_kzalloc(dev, sizeof(*hdq_data), GFP_KERNEL); if (!hdq_data) { @@ -567,10 +705,21 @@ static int omap_hdq_probe(struct platform_device *pdev) return PTR_ERR(hdq_data->hdq_base); hdq_data->hdq_usecount = 0; + hdq_data->rrw = 0; mutex_init(&hdq_data->hdq_mutex); pm_runtime_enable(&pdev->dev); - pm_runtime_get_sync(&pdev->dev); + ret = pm_runtime_get_sync(&pdev->dev); + if (ret < 0) { + dev_dbg(&pdev->dev, "pm_runtime_get_sync failed\n"); + goto err_w1; + } + + ret = _omap_hdq_reset(hdq_data); + if (ret) { + dev_dbg(&pdev->dev, "reset failed\n"); + return -EINVAL; + } rev = hdq_reg_in(hdq_data, OMAP_HDQ_REVISION); dev_info(&pdev->dev, "OMAP HDQ Hardware Rev %c.%c. Driver in %s mode\n", @@ -594,6 +743,15 @@ static int omap_hdq_probe(struct platform_device *pdev) pm_runtime_put_sync(&pdev->dev); + ret = of_property_read_string(pdev->dev.of_node, "ti,mode", &mode); + if (ret < 0 || !strcmp(mode, "hdq")) { + hdq_data->mode = 0; + omap_w1_master.search = omap_w1_search_bus; + } else { + hdq_data->mode = 1; + omap_w1_master.triplet = omap_w1_triplet; + } + omap_w1_master.data = hdq_data; ret = w1_add_master_device(&omap_w1_master); @@ -635,8 +793,8 @@ static int omap_hdq_remove(struct platform_device *pdev) module_platform_driver(omap_hdq_driver); module_param(w1_id, int, S_IRUSR); -MODULE_PARM_DESC(w1_id, "1-wire id for the slave detection"); +MODULE_PARM_DESC(w1_id, "1-wire id for the slave detection in HDQ mode"); MODULE_AUTHOR("Texas Instruments"); -MODULE_DESCRIPTION("HDQ driver Library"); +MODULE_DESCRIPTION("HDQ-1W driver Library"); MODULE_LICENSE("GPL"); diff --git a/drivers/w1/w1_int.c b/drivers/w1/w1_int.c index 47249a30eae3..20f766afa4c7 100644 --- a/drivers/w1/w1_int.c +++ b/drivers/w1/w1_int.c @@ -91,8 +91,7 @@ static struct w1_master *w1_alloc_dev(u32 id, int slave_count, int slave_ttl, err = device_register(&dev->dev); if (err) { pr_err("Failed to register master device. err=%d\n", err); - memset(dev, 0, sizeof(struct w1_master)); - kfree(dev); + put_device(&dev->dev); dev = NULL; } diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig index c68edc16aa54..79e1aa1b0959 100644 --- a/drivers/watchdog/Kconfig +++ b/drivers/watchdog/Kconfig @@ -817,8 +817,9 @@ config ITCO_WDT tristate "Intel TCO Timer/Watchdog" depends on (X86 || IA64) && PCI select WATCHDOG_CORE + depends on I2C || I2C=n select LPC_ICH if !EXPERT - select I2C_I801 if !EXPERT + select I2C_I801 if !EXPERT && I2C ---help--- Hardware driver for the intel TCO timer based watchdog devices. These drivers are included in the Intel 82801 I/O Controller diff --git a/drivers/watchdog/bcm2835_wdt.c b/drivers/watchdog/bcm2835_wdt.c index 66c3e656a616..8a5ce5b5a0b6 100644 --- a/drivers/watchdog/bcm2835_wdt.c +++ b/drivers/watchdog/bcm2835_wdt.c @@ -36,6 +36,13 @@ #define PM_RSTC_WRCFG_FULL_RESET 0x00000020 #define PM_RSTC_RESET 0x00000102 +/* + * The Raspberry Pi firmware uses the RSTS register to know which partiton + * to boot from. The partiton value is spread into bits 0, 2, 4, 6, 8, 10. + * Partiton 63 is a special partition used by the firmware to indicate halt. + */ +#define PM_RSTS_RASPBERRYPI_HALT 0x555 + #define SECS_TO_WDOG_TICKS(x) ((x) << 16) #define WDOG_TICKS_TO_SECS(x) ((x) >> 16) @@ -151,8 +158,7 @@ static void bcm2835_power_off(void) * hard reset. */ val = readl_relaxed(wdt->base + PM_RSTS); - val &= PM_RSTC_WRCFG_CLR; - val |= PM_PASSWORD | PM_RSTS_HADWRH_SET; + val |= PM_PASSWORD | PM_RSTS_RASPBERRYPI_HALT; writel_relaxed(val, wdt->base + PM_RSTS); /* Continue with normal reset mechanism */ diff --git a/drivers/watchdog/gef_wdt.c b/drivers/watchdog/gef_wdt.c index cc1bdfc2ff71..006e2348022c 100644 --- a/drivers/watchdog/gef_wdt.c +++ b/drivers/watchdog/gef_wdt.c @@ -303,6 +303,7 @@ static const struct of_device_id gef_wdt_ids[] = { }, {}, }; +MODULE_DEVICE_TABLE(of, gef_wdt_ids); static struct platform_driver gef_wdt_driver = { .driver = { diff --git a/drivers/watchdog/mena21_wdt.c b/drivers/watchdog/mena21_wdt.c index 69013007dc47..098fa9c34d6d 100644 --- a/drivers/watchdog/mena21_wdt.c +++ b/drivers/watchdog/mena21_wdt.c @@ -253,6 +253,7 @@ static const struct of_device_id a21_wdt_ids[] = { { .compatible = "men,a021-wdt" }, { }, }; +MODULE_DEVICE_TABLE(of, a21_wdt_ids); static struct platform_driver a21_wdt_driver = { .probe = a21_wdt_probe, diff --git a/drivers/watchdog/moxart_wdt.c b/drivers/watchdog/moxart_wdt.c index 2789da2c0515..60b0605bd7e6 100644 --- a/drivers/watchdog/moxart_wdt.c +++ b/drivers/watchdog/moxart_wdt.c @@ -168,6 +168,7 @@ static const struct of_device_id moxart_watchdog_match[] = { { .compatible = "moxa,moxart-watchdog" }, { }, }; +MODULE_DEVICE_TABLE(of, moxart_watchdog_match); static struct platform_driver moxart_wdt_driver = { .probe = moxart_wdt_probe, diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 295795aebe0b..1e60d00d4ea7 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -2847,6 +2847,8 @@ int open_ctree(struct super_block *sb, !extent_buffer_uptodate(chunk_root->node)) { printk(KERN_ERR "BTRFS: failed to read chunk root on %s\n", sb->s_id); + if (!IS_ERR(chunk_root->node)) + free_extent_buffer(chunk_root->node); chunk_root->node = NULL; goto fail_tree_roots; } @@ -2885,6 +2887,8 @@ retry_root_backup: !extent_buffer_uptodate(tree_root->node)) { printk(KERN_WARNING "BTRFS: failed to read tree root on %s\n", sb->s_id); + if (!IS_ERR(tree_root->node)) + free_extent_buffer(tree_root->node); tree_root->node = NULL; goto recovery_tree_root; } diff --git a/fs/btrfs/export.c b/fs/btrfs/export.c index 8d052209f473..2513a7f53334 100644 --- a/fs/btrfs/export.c +++ b/fs/btrfs/export.c @@ -112,11 +112,11 @@ static struct dentry *btrfs_fh_to_parent(struct super_block *sb, struct fid *fh, u32 generation; if (fh_type == FILEID_BTRFS_WITH_PARENT) { - if (fh_len != BTRFS_FID_SIZE_CONNECTABLE) + if (fh_len < BTRFS_FID_SIZE_CONNECTABLE) return NULL; root_objectid = fid->root_objectid; } else if (fh_type == FILEID_BTRFS_WITH_PARENT_ROOT) { - if (fh_len != BTRFS_FID_SIZE_CONNECTABLE_ROOT) + if (fh_len < BTRFS_FID_SIZE_CONNECTABLE_ROOT) return NULL; root_objectid = fid->parent_root_objectid; } else @@ -136,11 +136,11 @@ static struct dentry *btrfs_fh_to_dentry(struct super_block *sb, struct fid *fh, u32 generation; if ((fh_type != FILEID_BTRFS_WITH_PARENT || - fh_len != BTRFS_FID_SIZE_CONNECTABLE) && + fh_len < BTRFS_FID_SIZE_CONNECTABLE) && (fh_type != FILEID_BTRFS_WITH_PARENT_ROOT || - fh_len != BTRFS_FID_SIZE_CONNECTABLE_ROOT) && + fh_len < BTRFS_FID_SIZE_CONNECTABLE_ROOT) && (fh_type != FILEID_BTRFS_WITHOUT_PARENT || - fh_len != BTRFS_FID_SIZE_NON_CONNECTABLE)) + fh_len < BTRFS_FID_SIZE_NON_CONNECTABLE)) return NULL; objectid = fid->objectid; diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 9f9604201333..601d7d45d164 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -2828,6 +2828,7 @@ int btrfs_run_delayed_refs(struct btrfs_trans_handle *trans, struct btrfs_delayed_ref_head *head; int ret; int run_all = count == (unsigned long)-1; + bool can_flush_pending_bgs = trans->can_flush_pending_bgs; /* We'll clean this up in btrfs_cleanup_transaction */ if (trans->aborted) @@ -2844,6 +2845,7 @@ again: #ifdef SCRAMBLE_DELAYED_REFS delayed_refs->run_delayed_start = find_middle(&delayed_refs->root); #endif + trans->can_flush_pending_bgs = false; ret = __btrfs_run_delayed_refs(trans, root, count); if (ret < 0) { btrfs_abort_transaction(trans, root, ret); @@ -2893,6 +2895,7 @@ again: } out: assert_qgroups_uptodate(trans); + trans->can_flush_pending_bgs = can_flush_pending_bgs; return 0; } @@ -4306,7 +4309,8 @@ out: * the block groups that were made dirty during the lifetime of the * transaction. */ - if (trans->chunk_bytes_reserved >= (2 * 1024 * 1024ull)) { + if (trans->can_flush_pending_bgs && + trans->chunk_bytes_reserved >= (2 * 1024 * 1024ull)) { btrfs_create_pending_block_groups(trans, trans->root); btrfs_trans_release_chunk_metadata(trans); } @@ -9560,7 +9564,9 @@ void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans, struct btrfs_block_group_item item; struct btrfs_key key; int ret = 0; + bool can_flush_pending_bgs = trans->can_flush_pending_bgs; + trans->can_flush_pending_bgs = false; list_for_each_entry_safe(block_group, tmp, &trans->new_bgs, bg_list) { if (ret) goto next; @@ -9581,6 +9587,7 @@ void btrfs_create_pending_block_groups(struct btrfs_trans_handle *trans, next: list_del_init(&block_group->bg_list); } + trans->can_flush_pending_bgs = can_flush_pending_bgs; } int btrfs_make_block_group(struct btrfs_trans_handle *trans, diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index e2357e31609a..3915c9473e94 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -3132,12 +3132,12 @@ static inline void __do_contiguous_readpages(struct extent_io_tree *tree, get_extent_t *get_extent, struct extent_map **em_cached, struct bio **bio, int mirror_num, - unsigned long *bio_flags, int rw) + unsigned long *bio_flags, int rw, + u64 *prev_em_start) { struct inode *inode; struct btrfs_ordered_extent *ordered; int index; - u64 prev_em_start = (u64)-1; inode = pages[0]->mapping->host; while (1) { @@ -3153,7 +3153,7 @@ static inline void __do_contiguous_readpages(struct extent_io_tree *tree, for (index = 0; index < nr_pages; index++) { __do_readpage(tree, pages[index], get_extent, em_cached, bio, - mirror_num, bio_flags, rw, &prev_em_start); + mirror_num, bio_flags, rw, prev_em_start); page_cache_release(pages[index]); } } @@ -3163,7 +3163,8 @@ static void __extent_readpages(struct extent_io_tree *tree, int nr_pages, get_extent_t *get_extent, struct extent_map **em_cached, struct bio **bio, int mirror_num, - unsigned long *bio_flags, int rw) + unsigned long *bio_flags, int rw, + u64 *prev_em_start) { u64 start = 0; u64 end = 0; @@ -3184,7 +3185,7 @@ static void __extent_readpages(struct extent_io_tree *tree, index - first_index, start, end, get_extent, em_cached, bio, mirror_num, bio_flags, - rw); + rw, prev_em_start); start = page_start; end = start + PAGE_CACHE_SIZE - 1; first_index = index; @@ -3195,7 +3196,8 @@ static void __extent_readpages(struct extent_io_tree *tree, __do_contiguous_readpages(tree, &pages[first_index], index - first_index, start, end, get_extent, em_cached, bio, - mirror_num, bio_flags, rw); + mirror_num, bio_flags, rw, + prev_em_start); } static int __extent_read_full_page(struct extent_io_tree *tree, @@ -4207,6 +4209,7 @@ int extent_readpages(struct extent_io_tree *tree, struct page *page; struct extent_map *em_cached = NULL; int nr = 0; + u64 prev_em_start = (u64)-1; for (page_idx = 0; page_idx < nr_pages; page_idx++) { page = list_entry(pages->prev, struct page, lru); @@ -4223,12 +4226,12 @@ int extent_readpages(struct extent_io_tree *tree, if (nr < ARRAY_SIZE(pagepool)) continue; __extent_readpages(tree, pagepool, nr, get_extent, &em_cached, - &bio, 0, &bio_flags, READ); + &bio, 0, &bio_flags, READ, &prev_em_start); nr = 0; } if (nr) __extent_readpages(tree, pagepool, nr, get_extent, &em_cached, - &bio, 0, &bio_flags, READ); + &bio, 0, &bio_flags, READ, &prev_em_start); if (em_cached) free_extent_map(em_cached); diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index aa72bfd28f7d..a739b825bdd3 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -1920,10 +1920,12 @@ static int did_overwrite_ref(struct send_ctx *sctx, /* * We know that it is or will be overwritten. Check this now. * The current inode being processed might have been the one that caused - * inode 'ino' to be orphanized, therefore ow_inode can actually be the - * same as sctx->send_progress. + * inode 'ino' to be orphanized, therefore check if ow_inode matches + * the current inode being processed. */ - if (ow_inode <= sctx->send_progress) + if ((ow_inode < sctx->send_progress) || + (ino != sctx->cur_ino && ow_inode == sctx->cur_ino && + gen == sctx->cur_inode_gen)) ret = 1; else ret = 0; diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 74bc3338418b..a5b06442f0bf 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -557,6 +557,7 @@ again: h->delayed_ref_elem.seq = 0; h->type = type; h->allocating_chunk = false; + h->can_flush_pending_bgs = true; h->reloc_reserved = false; h->sync = false; INIT_LIST_HEAD(&h->qgroup_ref_list); diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h index 87964bf8892d..a994bb097ee5 100644 --- a/fs/btrfs/transaction.h +++ b/fs/btrfs/transaction.h @@ -118,6 +118,7 @@ struct btrfs_trans_handle { short aborted; short adding_csums; bool allocating_chunk; + bool can_flush_pending_bgs; bool reloc_reserved; bool sync; unsigned int type; diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 27aea110e923..c3cc1609025f 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -136,5 +136,5 @@ extern long cifs_ioctl(struct file *filep, unsigned int cmd, unsigned long arg); extern const struct export_operations cifs_export_ops; #endif /* CONFIG_CIFS_NFSD_EXPORT */ -#define CIFS_VERSION "2.07" +#define CIFS_VERSION "2.08" #endif /* _CIFSFS_H */ diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index f621b44cb800..6b66dd5d1540 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -2034,7 +2034,6 @@ cifs_set_file_size(struct inode *inode, struct iattr *attrs, struct tcon_link *tlink = NULL; struct cifs_tcon *tcon = NULL; struct TCP_Server_Info *server; - struct cifs_io_parms io_parms; /* * To avoid spurious oplock breaks from server, in the case of @@ -2056,18 +2055,6 @@ cifs_set_file_size(struct inode *inode, struct iattr *attrs, rc = -ENOSYS; cifsFileInfo_put(open_file); cifs_dbg(FYI, "SetFSize for attrs rc = %d\n", rc); - if ((rc == -EINVAL) || (rc == -EOPNOTSUPP)) { - unsigned int bytes_written; - - io_parms.netfid = open_file->fid.netfid; - io_parms.pid = open_file->pid; - io_parms.tcon = tcon; - io_parms.offset = 0; - io_parms.length = attrs->ia_size; - rc = CIFSSMBWrite(xid, &io_parms, &bytes_written, - NULL, NULL, 1); - cifs_dbg(FYI, "Wrt seteof rc %d\n", rc); - } } else rc = -EINVAL; @@ -2093,28 +2080,7 @@ cifs_set_file_size(struct inode *inode, struct iattr *attrs, else rc = -ENOSYS; cifs_dbg(FYI, "SetEOF by path (setattrs) rc = %d\n", rc); - if ((rc == -EINVAL) || (rc == -EOPNOTSUPP)) { - __u16 netfid; - int oplock = 0; - rc = SMBLegacyOpen(xid, tcon, full_path, FILE_OPEN, - GENERIC_WRITE, CREATE_NOT_DIR, &netfid, - &oplock, NULL, cifs_sb->local_nls, - cifs_remap(cifs_sb)); - if (rc == 0) { - unsigned int bytes_written; - - io_parms.netfid = netfid; - io_parms.pid = current->tgid; - io_parms.tcon = tcon; - io_parms.offset = 0; - io_parms.length = attrs->ia_size; - rc = CIFSSMBWrite(xid, &io_parms, &bytes_written, NULL, - NULL, 1); - cifs_dbg(FYI, "wrt seteof rc %d\n", rc); - CIFSSMBClose(xid, tcon, netfid); - } - } if (tlink) cifs_put_tlink(tlink); diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c index ce83e2edbe0a..597a417ba94d 100644 --- a/fs/cifs/smb2pdu.c +++ b/fs/cifs/smb2pdu.c @@ -922,7 +922,7 @@ SMB2_tcon(const unsigned int xid, struct cifs_ses *ses, const char *tree, if (tcon && tcon->bad_network_name) return -ENOENT; - if ((tcon->seal) && + if ((tcon && tcon->seal) && ((ses->server->capabilities & SMB2_GLOBAL_CAP_ENCRYPTION) == 0)) { cifs_dbg(VFS, "encryption requested but no server support"); return -EOPNOTSUPP; @@ -569,8 +569,20 @@ int __dax_pmd_fault(struct vm_area_struct *vma, unsigned long address, if (!buffer_size_valid(&bh) || bh.b_size < PMD_SIZE) goto fallback; + sector = bh.b_blocknr << (blkbits - 9); + if (buffer_unwritten(&bh) || buffer_new(&bh)) { int i; + + length = bdev_direct_access(bh.b_bdev, sector, &kaddr, &pfn, + bh.b_size); + if (length < 0) { + result = VM_FAULT_SIGBUS; + goto out; + } + if ((length < PMD_SIZE) || (pfn & PG_PMD_COLOUR)) + goto fallback; + for (i = 0; i < PTRS_PER_PMD; i++) clear_pmem(kaddr + i * PAGE_SIZE, PAGE_SIZE); wmb_pmem(); @@ -623,7 +635,6 @@ int __dax_pmd_fault(struct vm_area_struct *vma, unsigned long address, result = VM_FAULT_NOPAGE; spin_unlock(ptl); } else { - sector = bh.b_blocknr << (blkbits - 9); length = bdev_direct_access(bh.b_bdev, sector, &kaddr, &pfn, bh.b_size); if (length < 0) { diff --git a/fs/namei.c b/fs/namei.c index 726d211db484..33e9495a3129 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1558,8 +1558,6 @@ static int lookup_fast(struct nameidata *nd, negative = d_is_negative(dentry); if (read_seqcount_retry(&dentry->d_seq, seq)) return -ECHILD; - if (negative) - return -ENOENT; /* * This sequence count validates that the parent had no @@ -1580,6 +1578,12 @@ static int lookup_fast(struct nameidata *nd, goto unlazy; } } + /* + * Note: do negative dentry check after revalidation in + * case that drops it. + */ + if (negative) + return -ENOENT; path->mnt = mnt; path->dentry = dentry; if (likely(__follow_mount_rcu(nd, path, inode, seqp))) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index f93b9cdb4934..5133bb18830e 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -1458,12 +1458,18 @@ nfs4_opendata_check_deleg(struct nfs4_opendata *data, struct nfs4_state *state) if (delegation) delegation_flags = delegation->flags; rcu_read_unlock(); - if (data->o_arg.claim == NFS4_OPEN_CLAIM_DELEGATE_CUR) { + switch (data->o_arg.claim) { + default: + break; + case NFS4_OPEN_CLAIM_DELEGATE_CUR: + case NFS4_OPEN_CLAIM_DELEG_CUR_FH: pr_err_ratelimited("NFS: Broken NFSv4 server %s is " "returning a delegation for " "OPEN(CLAIM_DELEGATE_CUR)\n", clp->cl_hostname); - } else if ((delegation_flags & 1UL<<NFS_DELEGATION_NEED_RECLAIM) == 0) + return; + } + if ((delegation_flags & 1UL<<NFS_DELEGATION_NEED_RECLAIM) == 0) nfs_inode_set_delegation(state->inode, data->owner->so_cred, &data->o_res); @@ -1771,6 +1777,9 @@ int nfs4_open_delegation_recall(struct nfs_open_context *ctx, if (IS_ERR(opendata)) return PTR_ERR(opendata); nfs4_stateid_copy(&opendata->o_arg.u.delegation, stateid); + write_seqlock(&state->seqlock); + nfs4_stateid_copy(&state->stateid, &state->open_stateid); + write_sequnlock(&state->seqlock); clear_bit(NFS_DELEGATED_STATE, &state->flags); switch (type & (FMODE_READ|FMODE_WRITE)) { case FMODE_READ|FMODE_WRITE: @@ -1863,6 +1872,8 @@ static int _nfs4_proc_open_confirm(struct nfs4_opendata *data) data->rpc_done = 0; data->rpc_status = 0; data->timestamp = jiffies; + if (data->is_recover) + nfs4_set_sequence_privileged(&data->c_arg.seq_args); task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) return PTR_ERR(task); diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 5db324635e92..d854693a15b0 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -1725,7 +1725,8 @@ restart: if (!test_and_clear_bit(ops->owner_flag_bit, &sp->so_flags)) continue; - atomic_inc(&sp->so_count); + if (!atomic_inc_not_zero(&sp->so_count)) + continue; spin_unlock(&clp->cl_lock); rcu_read_unlock(); diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index 28df12e525ba..671cf68fe56b 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -409,7 +409,7 @@ DECLARE_EVENT_CLASS(nfs4_open_event, __entry->flags = flags; __entry->fmode = (__force unsigned int)ctx->mode; __entry->dev = ctx->dentry->d_sb->s_dev; - if (!IS_ERR(state)) + if (!IS_ERR_OR_NULL(state)) inode = state->inode; if (inode != NULL) { __entry->fileid = NFS_FILEID(inode); diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 72624dc4a623..75ab7622e0cc 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -569,19 +569,17 @@ static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio, if (!nfs_pageio_add_request(pgio, req)) { nfs_redirty_request(req); ret = pgio->pg_error; - } + } else + nfs_add_stats(page_file_mapping(page)->host, + NFSIOS_WRITEPAGES, 1); out: return ret; } static int nfs_do_writepage(struct page *page, struct writeback_control *wbc, struct nfs_pageio_descriptor *pgio) { - struct inode *inode = page_file_mapping(page)->host; int ret; - nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGE); - nfs_add_stats(inode, NFSIOS_WRITEPAGES, 1); - nfs_pageio_cond_complete(pgio, page_file_index(page)); ret = nfs_page_async_flush(pgio, page, wbc->sync_mode == WB_SYNC_NONE); if (ret == -EAGAIN) { @@ -597,9 +595,11 @@ static int nfs_do_writepage(struct page *page, struct writeback_control *wbc, st static int nfs_writepage_locked(struct page *page, struct writeback_control *wbc) { struct nfs_pageio_descriptor pgio; + struct inode *inode = page_file_mapping(page)->host; int err; - nfs_pageio_init_write(&pgio, page->mapping->host, wb_priority(wbc), + nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGE); + nfs_pageio_init_write(&pgio, inode, wb_priority(wbc), false, &nfs_async_write_completion_ops); err = nfs_do_writepage(page, wbc, &pgio); nfs_pageio_complete(&pgio); @@ -1223,7 +1223,7 @@ static int nfs_can_extend_write(struct file *file, struct page *page, struct ino return 1; if (!flctx || (list_empty_careful(&flctx->flc_flock) && list_empty_careful(&flctx->flc_posix))) - return 0; + return 1; /* Check to see if there are whole file write locks */ ret = 0; diff --git a/fs/ubifs/xattr.c b/fs/ubifs/xattr.c index 96f3448b6eb4..fd65b3f1923c 100644 --- a/fs/ubifs/xattr.c +++ b/fs/ubifs/xattr.c @@ -652,11 +652,8 @@ int ubifs_init_security(struct inode *dentry, struct inode *inode, { int err; - mutex_lock(&inode->i_mutex); err = security_inode_init_security(inode, dentry, qstr, &init_xattrs, 0); - mutex_unlock(&inode->i_mutex); - if (err) { struct ubifs_info *c = dentry->i_sb->s_fs_info; ubifs_err(c, "cannot initialize security for inode %lu, error %d", diff --git a/include/asm-generic/word-at-a-time.h b/include/asm-generic/word-at-a-time.h index 94f9ea8abcae..011dde083f23 100644 --- a/include/asm-generic/word-at-a-time.h +++ b/include/asm-generic/word-at-a-time.h @@ -1,15 +1,10 @@ #ifndef _ASM_WORD_AT_A_TIME_H #define _ASM_WORD_AT_A_TIME_H -/* - * This says "generic", but it's actually big-endian only. - * Little-endian can use more efficient versions of these - * interfaces, see for example - * arch/x86/include/asm/word-at-a-time.h - * for those. - */ - #include <linux/kernel.h> +#include <asm/byteorder.h> + +#ifdef __BIG_ENDIAN struct word_at_a_time { const unsigned long high_bits, low_bits; @@ -53,4 +48,73 @@ static inline bool has_zero(unsigned long val, unsigned long *data, const struct #define zero_bytemask(mask) (~1ul << __fls(mask)) #endif +#else + +/* + * The optimal byte mask counting is probably going to be something + * that is architecture-specific. If you have a reliably fast + * bit count instruction, that might be better than the multiply + * and shift, for example. + */ +struct word_at_a_time { + const unsigned long one_bits, high_bits; +}; + +#define WORD_AT_A_TIME_CONSTANTS { REPEAT_BYTE(0x01), REPEAT_BYTE(0x80) } + +#ifdef CONFIG_64BIT + +/* + * Jan Achrenius on G+: microoptimized version of + * the simpler "(mask & ONEBYTES) * ONEBYTES >> 56" + * that works for the bytemasks without having to + * mask them first. + */ +static inline long count_masked_bytes(unsigned long mask) +{ + return mask*0x0001020304050608ul >> 56; +} + +#else /* 32-bit case */ + +/* Carl Chatfield / Jan Achrenius G+ version for 32-bit */ +static inline long count_masked_bytes(long mask) +{ + /* (000000 0000ff 00ffff ffffff) -> ( 1 1 2 3 ) */ + long a = (0x0ff0001+mask) >> 23; + /* Fix the 1 for 00 case */ + return a & mask; +} + +#endif + +/* Return nonzero if it has a zero */ +static inline unsigned long has_zero(unsigned long a, unsigned long *bits, const struct word_at_a_time *c) +{ + unsigned long mask = ((a - c->one_bits) & ~a) & c->high_bits; + *bits = mask; + return mask; +} + +static inline unsigned long prep_zero_mask(unsigned long a, unsigned long bits, const struct word_at_a_time *c) +{ + return bits; +} + +static inline unsigned long create_zero_mask(unsigned long bits) +{ + bits = (bits - 1) & ~bits; + return bits >> 7; +} + +/* The mask we created is directly usable as a bytemask */ +#define zero_bytemask(mask) (mask) + +static inline unsigned long find_zero(unsigned long mask) +{ + return count_masked_bytes(mask); +} + +#endif /* __BIG_ENDIAN */ + #endif /* _ASM_WORD_AT_A_TIME_H */ diff --git a/include/drm/drm_crtc_helper.h b/include/drm/drm_crtc_helper.h index 2a747a91fded..3febb4b9fce9 100644 --- a/include/drm/drm_crtc_helper.h +++ b/include/drm/drm_crtc_helper.h @@ -240,5 +240,6 @@ extern void drm_kms_helper_hotplug_event(struct drm_device *dev); extern void drm_kms_helper_poll_disable(struct drm_device *dev); extern void drm_kms_helper_poll_enable(struct drm_device *dev); +extern void drm_kms_helper_poll_enable_locked(struct drm_device *dev); #endif diff --git a/include/drm/drm_dp_helper.h b/include/drm/drm_dp_helper.h index 499e9f625aef..0212d139a480 100644 --- a/include/drm/drm_dp_helper.h +++ b/include/drm/drm_dp_helper.h @@ -568,6 +568,10 @@ #define MODE_I2C_READ 4 #define MODE_I2C_STOP 8 +/* DP 1.2 MST PORTs - Section 2.5.1 v1.2a spec */ +#define DP_MST_PHYSICAL_PORT_0 0 +#define DP_MST_LOGICAL_PORT_0 8 + #define DP_LINK_STATUS_SIZE 6 bool drm_dp_channel_eq_ok(const u8 link_status[DP_LINK_STATUS_SIZE], int lane_count); diff --git a/include/drm/drm_dp_mst_helper.h b/include/drm/drm_dp_mst_helper.h index 86d0b25ed054..0f408b002d98 100644 --- a/include/drm/drm_dp_mst_helper.h +++ b/include/drm/drm_dp_mst_helper.h @@ -374,6 +374,7 @@ struct drm_dp_mst_topology_mgr; struct drm_dp_mst_topology_cbs { /* create a connector for a port */ struct drm_connector *(*add_connector)(struct drm_dp_mst_topology_mgr *mgr, struct drm_dp_mst_port *port, const char *path); + void (*register_connector)(struct drm_connector *connector); void (*destroy_connector)(struct drm_dp_mst_topology_mgr *mgr, struct drm_connector *connector); void (*hotplug)(struct drm_dp_mst_topology_mgr *mgr); diff --git a/include/linux/acpi.h b/include/linux/acpi.h index 7235c4851460..43856d19cf4d 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -217,6 +217,7 @@ struct pci_dev; int acpi_pci_irq_enable (struct pci_dev *dev); void acpi_penalize_isa_irq(int irq, int active); +bool acpi_isa_irq_available(int irq); void acpi_penalize_sci_irq(int irq, int trigger, int polarity); void acpi_pci_irq_disable (struct pci_dev *dev); diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h index 37d1602c4f7a..5e7d43ab61c0 100644 --- a/include/linux/blk-mq.h +++ b/include/linux/blk-mq.h @@ -145,7 +145,6 @@ enum { BLK_MQ_F_SHOULD_MERGE = 1 << 0, BLK_MQ_F_TAG_SHARED = 1 << 1, BLK_MQ_F_SG_MERGE = 1 << 2, - BLK_MQ_F_SYSFS_UP = 1 << 3, BLK_MQ_F_DEFER_ISSUE = 1 << 4, BLK_MQ_F_ALLOC_POLICY_START_BIT = 8, BLK_MQ_F_ALLOC_POLICY_BITS = 1, @@ -215,7 +214,7 @@ void blk_mq_add_to_requeue_list(struct request *rq, bool at_head); void blk_mq_cancel_requeue_work(struct request_queue *q); void blk_mq_kick_requeue_list(struct request_queue *q); void blk_mq_abort_requeue_list(struct request_queue *q); -void blk_mq_complete_request(struct request *rq); +void blk_mq_complete_request(struct request *rq, int error); void blk_mq_stop_hw_queue(struct blk_mq_hw_ctx *hctx); void blk_mq_start_hw_queue(struct blk_mq_hw_ctx *hctx); @@ -224,8 +223,6 @@ void blk_mq_start_hw_queues(struct request_queue *q); void blk_mq_start_stopped_hw_queues(struct request_queue *q, bool async); void blk_mq_run_hw_queues(struct request_queue *q, bool async); void blk_mq_delay_queue(struct blk_mq_hw_ctx *hctx, unsigned long msecs); -void blk_mq_tag_busy_iter(struct blk_mq_hw_ctx *hctx, busy_iter_fn *fn, - void *priv); void blk_mq_all_tag_busy_iter(struct blk_mq_tags *tags, busy_tag_iter_fn *fn, void *priv); void blk_mq_freeze_queue(struct request_queue *q); diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 99da9ebc7377..19c2e947d4d1 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -456,6 +456,8 @@ struct request_queue { struct blk_mq_tag_set *tag_set; struct list_head tag_set_list; struct bio_set *bio_split; + + bool mq_sysfs_init_done; }; #define QUEUE_FLAG_QUEUED 1 /* uses generic tag queueing */ diff --git a/include/linux/coresight.h b/include/linux/coresight.h index c69e1b932809..a7cabfa23b55 100644 --- a/include/linux/coresight.h +++ b/include/linux/coresight.h @@ -207,7 +207,7 @@ struct coresight_ops_link { * Operations available for sources. * @trace_id: returns the value of the component's trace ID as known to the HW. - * @enable: enables tracing from a source. + * @enable: enables tracing for a source. * @disable: disables tracing for a source. */ struct coresight_ops_source { diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h index 7ea9184eaa13..c47c68e535e8 100644 --- a/include/linux/dmaengine.h +++ b/include/linux/dmaengine.h @@ -645,6 +645,7 @@ enum dmaengine_alignment { * The function takes a buffer of size buf_len. The callback function will * be called after period_len bytes have been transferred. * @device_prep_interleaved_dma: Transfer expression in a generic way. + * @device_prep_dma_imm_data: DMA's 8 byte immediate data to the dst address * @device_config: Pushes a new configuration to a channel, return 0 or an error * code * @device_pause: Pauses any transfer happening on a channel. Returns @@ -727,6 +728,9 @@ struct dma_device { struct dma_async_tx_descriptor *(*device_prep_interleaved_dma)( struct dma_chan *chan, struct dma_interleaved_template *xt, unsigned long flags); + struct dma_async_tx_descriptor *(*device_prep_dma_imm_data)( + struct dma_chan *chan, dma_addr_t dst, u64 data, + unsigned long flags); int (*device_config)(struct dma_chan *chan, struct dma_slave_config *config); diff --git a/include/linux/fpga/fpga-mgr.h b/include/linux/fpga/fpga-mgr.h new file mode 100644 index 000000000000..0940bf45e2f2 --- /dev/null +++ b/include/linux/fpga/fpga-mgr.h @@ -0,0 +1,127 @@ +/* + * FPGA Framework + * + * Copyright (C) 2013-2015 Altera Corporation + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see <http://www.gnu.org/licenses/>. + */ +#include <linux/mutex.h> +#include <linux/platform_device.h> + +#ifndef _LINUX_FPGA_MGR_H +#define _LINUX_FPGA_MGR_H + +struct fpga_manager; + +/** + * enum fpga_mgr_states - fpga framework states + * @FPGA_MGR_STATE_UNKNOWN: can't determine state + * @FPGA_MGR_STATE_POWER_OFF: FPGA power is off + * @FPGA_MGR_STATE_POWER_UP: FPGA reports power is up + * @FPGA_MGR_STATE_RESET: FPGA in reset state + * @FPGA_MGR_STATE_FIRMWARE_REQ: firmware request in progress + * @FPGA_MGR_STATE_FIRMWARE_REQ_ERR: firmware request failed + * @FPGA_MGR_STATE_WRITE_INIT: preparing FPGA for programming + * @FPGA_MGR_STATE_WRITE_INIT_ERR: Error during WRITE_INIT stage + * @FPGA_MGR_STATE_WRITE: writing image to FPGA + * @FPGA_MGR_STATE_WRITE_ERR: Error while writing FPGA + * @FPGA_MGR_STATE_WRITE_COMPLETE: Doing post programming steps + * @FPGA_MGR_STATE_WRITE_COMPLETE_ERR: Error during WRITE_COMPLETE + * @FPGA_MGR_STATE_OPERATING: FPGA is programmed and operating + */ +enum fpga_mgr_states { + /* default FPGA states */ + FPGA_MGR_STATE_UNKNOWN, + FPGA_MGR_STATE_POWER_OFF, + FPGA_MGR_STATE_POWER_UP, + FPGA_MGR_STATE_RESET, + + /* getting an image for loading */ + FPGA_MGR_STATE_FIRMWARE_REQ, + FPGA_MGR_STATE_FIRMWARE_REQ_ERR, + + /* write sequence: init, write, complete */ + FPGA_MGR_STATE_WRITE_INIT, + FPGA_MGR_STATE_WRITE_INIT_ERR, + FPGA_MGR_STATE_WRITE, + FPGA_MGR_STATE_WRITE_ERR, + FPGA_MGR_STATE_WRITE_COMPLETE, + FPGA_MGR_STATE_WRITE_COMPLETE_ERR, + + /* fpga is programmed and operating */ + FPGA_MGR_STATE_OPERATING, +}; + +/* + * FPGA Manager flags + * FPGA_MGR_PARTIAL_RECONFIG: do partial reconfiguration if supported + */ +#define FPGA_MGR_PARTIAL_RECONFIG BIT(0) + +/** + * struct fpga_manager_ops - ops for low level fpga manager drivers + * @state: returns an enum value of the FPGA's state + * @write_init: prepare the FPGA to receive confuration data + * @write: write count bytes of configuration data to the FPGA + * @write_complete: set FPGA to operating state after writing is done + * @fpga_remove: optional: Set FPGA into a specific state during driver remove + * + * fpga_manager_ops are the low level functions implemented by a specific + * fpga manager driver. The optional ones are tested for NULL before being + * called, so leaving them out is fine. + */ +struct fpga_manager_ops { + enum fpga_mgr_states (*state)(struct fpga_manager *mgr); + int (*write_init)(struct fpga_manager *mgr, u32 flags, + const char *buf, size_t count); + int (*write)(struct fpga_manager *mgr, const char *buf, size_t count); + int (*write_complete)(struct fpga_manager *mgr, u32 flags); + void (*fpga_remove)(struct fpga_manager *mgr); +}; + +/** + * struct fpga_manager - fpga manager structure + * @name: name of low level fpga manager + * @dev: fpga manager device + * @ref_mutex: only allows one reference to fpga manager + * @state: state of fpga manager + * @mops: pointer to struct of fpga manager ops + * @priv: low level driver private date + */ +struct fpga_manager { + const char *name; + struct device dev; + struct mutex ref_mutex; + enum fpga_mgr_states state; + const struct fpga_manager_ops *mops; + void *priv; +}; + +#define to_fpga_manager(d) container_of(d, struct fpga_manager, dev) + +int fpga_mgr_buf_load(struct fpga_manager *mgr, u32 flags, + const char *buf, size_t count); + +int fpga_mgr_firmware_load(struct fpga_manager *mgr, u32 flags, + const char *image_name); + +struct fpga_manager *of_fpga_mgr_get(struct device_node *node); + +void fpga_mgr_put(struct fpga_manager *mgr); + +int fpga_mgr_register(struct device *dev, const char *name, + const struct fpga_manager_ops *mops, void *priv); + +void fpga_mgr_unregister(struct device *dev); + +#endif /*_LINUX_FPGA_MGR_H */ diff --git a/include/linux/iova.h b/include/linux/iova.h index 3920a19d8194..92f7177db2ce 100644 --- a/include/linux/iova.h +++ b/include/linux/iova.h @@ -68,8 +68,8 @@ static inline unsigned long iova_pfn(struct iova_domain *iovad, dma_addr_t iova) return iova >> iova_shift(iovad); } -int iommu_iova_cache_init(void); -void iommu_iova_cache_destroy(void); +int iova_cache_get(void); +void iova_cache_put(void); struct iova *alloc_iova_mem(void); void free_iova_mem(struct iova *iova); diff --git a/include/linux/irqdomain.h b/include/linux/irqdomain.h index d3ca79236fb0..f644fdb06dd6 100644 --- a/include/linux/irqdomain.h +++ b/include/linux/irqdomain.h @@ -161,6 +161,11 @@ enum { IRQ_DOMAIN_FLAG_NONCORE = (1 << 16), }; +static inline struct device_node *irq_domain_get_of_node(struct irq_domain *d) +{ + return d->of_node; +} + #ifdef CONFIG_IRQ_DOMAIN struct irq_domain *__irq_domain_add(struct device_node *of_node, int size, irq_hw_number_t hwirq_max, int direct_max, diff --git a/include/linux/mei_cl_bus.h b/include/linux/mei_cl_bus.h index 0962b2ca628a..e746919530f5 100644 --- a/include/linux/mei_cl_bus.h +++ b/include/linux/mei_cl_bus.h @@ -8,8 +8,8 @@ struct mei_cl_device; struct mei_device; -typedef void (*mei_cl_event_cb_t)(struct mei_cl_device *device, - u32 events, void *context); +typedef void (*mei_cldev_event_cb_t)(struct mei_cl_device *cldev, + u32 events, void *context); /** * struct mei_cl_device - MEI device handle @@ -45,7 +45,7 @@ struct mei_cl_device { char name[MEI_CL_NAME_SIZE]; struct work_struct event_work; - mei_cl_event_cb_t event_cb; + mei_cldev_event_cb_t event_cb; void *event_context; unsigned long events_mask; unsigned long events; @@ -62,33 +62,37 @@ struct mei_cl_driver { const struct mei_cl_device_id *id_table; - int (*probe)(struct mei_cl_device *dev, + int (*probe)(struct mei_cl_device *cldev, const struct mei_cl_device_id *id); - int (*remove)(struct mei_cl_device *dev); + int (*remove)(struct mei_cl_device *cldev); }; -int __mei_cl_driver_register(struct mei_cl_driver *driver, +int __mei_cldev_driver_register(struct mei_cl_driver *cldrv, struct module *owner); -#define mei_cl_driver_register(driver) \ - __mei_cl_driver_register(driver, THIS_MODULE) +#define mei_cldev_driver_register(cldrv) \ + __mei_cldev_driver_register(cldrv, THIS_MODULE) -void mei_cl_driver_unregister(struct mei_cl_driver *driver); +void mei_cldev_driver_unregister(struct mei_cl_driver *cldrv); -ssize_t mei_cl_send(struct mei_cl_device *device, u8 *buf, size_t length); -ssize_t mei_cl_recv(struct mei_cl_device *device, u8 *buf, size_t length); +ssize_t mei_cldev_send(struct mei_cl_device *cldev, u8 *buf, size_t length); +ssize_t mei_cldev_recv(struct mei_cl_device *cldev, u8 *buf, size_t length); -int mei_cl_register_event_cb(struct mei_cl_device *device, - unsigned long event_mask, - mei_cl_event_cb_t read_cb, void *context); +int mei_cldev_register_event_cb(struct mei_cl_device *cldev, + unsigned long event_mask, + mei_cldev_event_cb_t read_cb, void *context); #define MEI_CL_EVENT_RX 0 #define MEI_CL_EVENT_TX 1 #define MEI_CL_EVENT_NOTIF 2 -void *mei_cl_get_drvdata(const struct mei_cl_device *device); -void mei_cl_set_drvdata(struct mei_cl_device *device, void *data); +const uuid_le *mei_cldev_uuid(const struct mei_cl_device *cldev); +u8 mei_cldev_ver(const struct mei_cl_device *cldev); -int mei_cl_enable_device(struct mei_cl_device *device); -int mei_cl_disable_device(struct mei_cl_device *device); +void *mei_cldev_get_drvdata(const struct mei_cl_device *cldev); +void mei_cldev_set_drvdata(struct mei_cl_device *cldev, void *data); + +int mei_cldev_enable(struct mei_cl_device *cldev); +int mei_cldev_disable(struct mei_cl_device *cldev); +bool mei_cldev_enabled(struct mei_cl_device *cldev); #endif /* _LINUX_MEI_CL_BUS_H */ diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index ad800e62cb7a..6452ff4c463f 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -242,7 +242,6 @@ struct mem_cgroup { * percpu counter. */ struct mem_cgroup_stat_cpu __percpu *stat; - spinlock_t pcp_counter_lock; #if defined(CONFIG_MEMCG_KMEM) && defined(CONFIG_INET) struct cg_proto tcp_mem; diff --git a/include/linux/mic_bus.h b/include/linux/mic_bus.h index d5b5f76d57ef..27d7c95fd0da 100644 --- a/include/linux/mic_bus.h +++ b/include/linux/mic_bus.h @@ -91,7 +91,8 @@ struct mbus_hw_ops { struct mbus_device * mbus_register_device(struct device *pdev, int id, struct dma_map_ops *dma_ops, - struct mbus_hw_ops *hw_ops, void __iomem *mmio_va); + struct mbus_hw_ops *hw_ops, int index, + void __iomem *mmio_va); void mbus_unregister_device(struct mbus_device *mbdev); int mbus_register_driver(struct mbus_driver *drv); diff --git a/include/linux/mlx5/device.h b/include/linux/mlx5/device.h index 8eb3b19af2a4..250b1ff8b48d 100644 --- a/include/linux/mlx5/device.h +++ b/include/linux/mlx5/device.h @@ -402,17 +402,6 @@ struct mlx5_cmd_teardown_hca_mbox_out { u8 rsvd[8]; }; -struct mlx5_cmd_query_special_contexts_mbox_in { - struct mlx5_inbox_hdr hdr; - u8 rsvd[8]; -}; - -struct mlx5_cmd_query_special_contexts_mbox_out { - struct mlx5_outbox_hdr hdr; - __be32 dump_fill_mkey; - __be32 resd_lkey; -}; - struct mlx5_cmd_layout { u8 type; u8 rsvd0[3]; diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 27b53f9a24ad..8b6d6f2154a4 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -845,7 +845,6 @@ void *mlx5_get_protocol_dev(struct mlx5_core_dev *mdev, int protocol); int mlx5_register_interface(struct mlx5_interface *intf); void mlx5_unregister_interface(struct mlx5_interface *intf); int mlx5_core_query_vendor_id(struct mlx5_core_dev *mdev, u32 *vendor_id); -int mlx5_core_query_special_context(struct mlx5_core_dev *dev, u32 *rsvd_lkey); struct mlx5_profile { u64 mask; diff --git a/include/linux/mm.h b/include/linux/mm.h index 91c08f6f0dc9..80001de019ba 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -905,6 +905,27 @@ static inline void set_page_links(struct page *page, enum zone_type zone, #endif } +#ifdef CONFIG_MEMCG +static inline struct mem_cgroup *page_memcg(struct page *page) +{ + return page->mem_cgroup; +} + +static inline void set_page_memcg(struct page *page, struct mem_cgroup *memcg) +{ + page->mem_cgroup = memcg; +} +#else +static inline struct mem_cgroup *page_memcg(struct page *page) +{ + return NULL; +} + +static inline void set_page_memcg(struct page *page, struct mem_cgroup *memcg) +{ +} +#endif + /* * Some inline functions in vmstat.h depend on page_zone() */ diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h index 688997a24aad..6975cbf1435b 100644 --- a/include/linux/mod_devicetable.h +++ b/include/linux/mod_devicetable.h @@ -601,15 +601,13 @@ struct ipack_device_id { #define MEI_CL_MODULE_PREFIX "mei:" #define MEI_CL_NAME_SIZE 32 -#define MEI_CL_UUID_FMT "%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x%02x" -#define MEI_CL_UUID_ARGS(_u) \ - _u[0], _u[1], _u[2], _u[3], _u[4], _u[5], _u[6], _u[7], \ - _u[8], _u[9], _u[10], _u[11], _u[12], _u[13], _u[14], _u[15] +#define MEI_CL_VERSION_ANY 0xff /** * struct mei_cl_device_id - MEI client device identifier * @name: helper name * @uuid: client uuid + * @version: client protocol version * @driver_info: information used by the driver. * * identifies mei client device by uuid and name @@ -617,6 +615,7 @@ struct ipack_device_id { struct mei_cl_device_id { char name[MEI_CL_NAME_SIZE]; uuid_le uuid; + __u8 version; kernel_ulong_t driver_info; }; diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index ff476515f716..581abf848566 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -230,12 +230,11 @@ void __wait_rcu_gp(bool checktiny, int n, call_rcu_func_t *crcu_array, struct rcu_synchronize *rs_array); #define _wait_rcu_gp(checktiny, ...) \ -do { \ - call_rcu_func_t __crcu_array[] = { __VA_ARGS__ }; \ - const int __n = ARRAY_SIZE(__crcu_array); \ - struct rcu_synchronize __rs_array[__n]; \ - \ - __wait_rcu_gp(checktiny, __n, __crcu_array, __rs_array); \ +do { \ + call_rcu_func_t __crcu_array[] = { __VA_ARGS__ }; \ + struct rcu_synchronize __rs_array[ARRAY_SIZE(__crcu_array)]; \ + __wait_rcu_gp(checktiny, ARRAY_SIZE(__crcu_array), \ + __crcu_array, __rs_array); \ } while (0) #define wait_rcu_gp(...) _wait_rcu_gp(false, __VA_ARGS__) diff --git a/include/linux/scif.h b/include/linux/scif.h index 44f4f3898bbe..49a35d6edc94 100644 --- a/include/linux/scif.h +++ b/include/linux/scif.h @@ -55,6 +55,7 @@ #include <linux/types.h> #include <linux/poll.h> +#include <linux/device.h> #include <linux/scif_ioctl.h> #define SCIF_ACCEPT_SYNC 1 @@ -92,6 +93,70 @@ enum { #define SCIF_PORT_RSVD 1088 typedef struct scif_endpt *scif_epd_t; +typedef struct scif_pinned_pages *scif_pinned_pages_t; + +/** + * struct scif_range - SCIF registered range used in kernel mode + * @cookie: cookie used internally by SCIF + * @nr_pages: number of pages of PAGE_SIZE + * @prot_flags: R/W protection + * @phys_addr: Array of bus addresses + * @va: Array of kernel virtual addresses backed by the pages in the phys_addr + * array. The va is populated only when called on the host for a remote + * SCIF connection on MIC. This is required to support the use case of DMA + * between MIC and another device which is not a SCIF node e.g., an IB or + * ethernet NIC. + */ +struct scif_range { + void *cookie; + int nr_pages; + int prot_flags; + dma_addr_t *phys_addr; + void __iomem **va; +}; + +/** + * struct scif_pollepd - SCIF endpoint to be monitored via scif_poll + * @epd: SCIF endpoint + * @events: requested events + * @revents: returned events + */ +struct scif_pollepd { + scif_epd_t epd; + short events; + short revents; +}; + +/** + * scif_peer_dev - representation of a peer SCIF device + * + * Peer devices show up as PCIe devices for the mgmt node but not the cards. + * The mgmt node discovers all the cards on the PCIe bus and informs the other + * cards about their peers. Upon notification of a peer a node adds a peer + * device to the peer bus to maintain symmetry in the way devices are + * discovered across all nodes in the SCIF network. + * + * @dev: underlying device + * @dnode - The destination node which this device will communicate with. + */ +struct scif_peer_dev { + struct device dev; + u8 dnode; +}; + +/** + * scif_client - representation of a SCIF client + * @name: client name + * @probe - client method called when a peer device is registered + * @remove - client method called when a peer device is unregistered + * @si - subsys_interface used internally for implementing SCIF clients + */ +struct scif_client { + const char *name; + void (*probe)(struct scif_peer_dev *spdev); + void (*remove)(struct scif_peer_dev *spdev); + struct subsys_interface si; +}; #define SCIF_OPEN_FAILED ((scif_epd_t)-1) #define SCIF_REGISTER_FAILED ((off_t)-1) @@ -345,7 +410,6 @@ int scif_close(scif_epd_t epd); * Errors: * EBADF, ENOTTY - epd is not a valid endpoint descriptor * ECONNRESET - Connection reset by peer - * EFAULT - An invalid address was specified for a parameter * EINVAL - flags is invalid, or len is negative * ENODEV - The remote node is lost or existed, but is not currently in the * network since it may have crashed @@ -398,7 +462,6 @@ int scif_send(scif_epd_t epd, void *msg, int len, int flags); * EAGAIN - The destination node is returning from a low power state * EBADF, ENOTTY - epd is not a valid endpoint descriptor * ECONNRESET - Connection reset by peer - * EFAULT - An invalid address was specified for a parameter * EINVAL - flags is invalid, or len is negative * ENODEV - The remote node is lost or existed, but is not currently in the * network since it may have crashed @@ -461,9 +524,6 @@ int scif_recv(scif_epd_t epd, void *msg, int len, int flags); * SCIF_PROT_READ - allow read operations from the window * SCIF_PROT_WRITE - allow write operations to the window * - * The map_flags argument can be set to SCIF_MAP_FIXED which interprets a - * fixed offset. - * * Return: * Upon successful completion, scif_register() returns the offset at which the * mapping was placed (po); otherwise in user mode SCIF_REGISTER_FAILED (that @@ -476,7 +536,6 @@ int scif_recv(scif_epd_t epd, void *msg, int len, int flags); * EAGAIN - The mapping could not be performed due to lack of resources * EBADF, ENOTTY - epd is not a valid endpoint descriptor * ECONNRESET - Connection reset by peer - * EFAULT - Addresses in the range [addr, addr + len - 1] are invalid * EINVAL - map_flags is invalid, or prot_flags is invalid, or SCIF_MAP_FIXED is * set in flags, and offset is not a multiple of the page size, or addr is not a * multiple of the page size, or len is not a multiple of the page size, or is @@ -759,7 +818,6 @@ int scif_writeto(scif_epd_t epd, off_t loffset, size_t len, off_t * EACCESS - Attempt to write to a read-only range * EBADF, ENOTTY - epd is not a valid endpoint descriptor * ECONNRESET - Connection reset by peer - * EFAULT - Addresses in the range [addr, addr + len - 1] are invalid * EINVAL - rma_flags is invalid * ENODEV - The remote node is lost or existed, but is not currently in the * network since it may have crashed @@ -840,7 +898,6 @@ int scif_vreadfrom(scif_epd_t epd, void *addr, size_t len, off_t roffset, * EACCESS - Attempt to write to a read-only range * EBADF, ENOTTY - epd is not a valid endpoint descriptor * ECONNRESET - Connection reset by peer - * EFAULT - Addresses in the range [addr, addr + len - 1] are invalid * EINVAL - rma_flags is invalid * ENODEV - The remote node is lost or existed, but is not currently in the * network since it may have crashed @@ -984,10 +1041,299 @@ int scif_fence_signal(scif_epd_t epd, off_t loff, u64 lval, off_t roff, * online nodes in the SCIF network including 'self'; otherwise in user mode * -1 is returned and errno is set to indicate the error; in kernel mode no * errors are returned. + */ +int scif_get_node_ids(u16 *nodes, int len, u16 *self); + +/** + * scif_pin_pages() - Pin a set of pages + * @addr: Virtual address of range to pin + * @len: Length of range to pin + * @prot_flags: Page protection flags + * @map_flags: Page classification flags + * @pinned_pages: Handle to pinned pages + * + * scif_pin_pages() pins (locks in physical memory) the physical pages which + * back the range of virtual address pages starting at addr and continuing for + * len bytes. addr and len are constrained to be multiples of the page size. A + * successful scif_pin_pages() call returns a handle to pinned_pages which may + * be used in subsequent calls to scif_register_pinned_pages(). + * + * The pages will remain pinned as long as there is a reference against the + * scif_pinned_pages_t value returned by scif_pin_pages() and until + * scif_unpin_pages() is called, passing the scif_pinned_pages_t value. A + * reference is added to a scif_pinned_pages_t value each time a window is + * created by calling scif_register_pinned_pages() and passing the + * scif_pinned_pages_t value. A reference is removed from a + * scif_pinned_pages_t value each time such a window is deleted. + * + * Subsequent operations which change the memory pages to which virtual + * addresses are mapped (such as mmap(), munmap()) have no effect on the + * scif_pinned_pages_t value or windows created against it. + * + * If the process will fork(), it is recommended that the registered + * virtual address range be marked with MADV_DONTFORK. Doing so will prevent + * problems due to copy-on-write semantics. + * + * The prot_flags argument is formed by OR'ing together one or more of the + * following values. + * SCIF_PROT_READ - allow read operations against the pages + * SCIF_PROT_WRITE - allow write operations against the pages + * The map_flags argument can be set as SCIF_MAP_KERNEL to interpret addr as a + * kernel space address. By default, addr is interpreted as a user space + * address. + * + * Return: + * Upon successful completion, scif_pin_pages() returns 0; otherwise the + * negative of one of the following errors is returned. * * Errors: - * EFAULT - Bad address + * EINVAL - prot_flags is invalid, map_flags is invalid, or offset is negative + * ENOMEM - Not enough space */ -int scif_get_node_ids(u16 *nodes, int len, u16 *self); +int scif_pin_pages(void *addr, size_t len, int prot_flags, int map_flags, + scif_pinned_pages_t *pinned_pages); + +/** + * scif_unpin_pages() - Unpin a set of pages + * @pinned_pages: Handle to pinned pages to be unpinned + * + * scif_unpin_pages() prevents scif_register_pinned_pages() from registering new + * windows against pinned_pages. The physical pages represented by pinned_pages + * will remain pinned until all windows previously registered against + * pinned_pages are deleted (the window is scif_unregister()'d and all + * references to the window are removed (see scif_unregister()). + * + * pinned_pages must have been obtain from a previous call to scif_pin_pages(). + * After calling scif_unpin_pages(), it is an error to pass pinned_pages to + * scif_register_pinned_pages(). + * + * Return: + * Upon successful completion, scif_unpin_pages() returns 0; otherwise the + * negative of one of the following errors is returned. + * + * Errors: + * EINVAL - pinned_pages is not valid + */ +int scif_unpin_pages(scif_pinned_pages_t pinned_pages); + +/** + * scif_register_pinned_pages() - Mark a memory region for remote access. + * @epd: endpoint descriptor + * @pinned_pages: Handle to pinned pages + * @offset: Registered address space offset + * @map_flags: Flags which control where pages are mapped + * + * The scif_register_pinned_pages() function opens a window, a range of whole + * pages of the registered address space of the endpoint epd, starting at + * offset po. The value of po, further described below, is a function of the + * parameters offset and pinned_pages, and the value of map_flags. Each page of + * the window represents a corresponding physical memory page of the range + * represented by pinned_pages; the length of the window is the same as the + * length of range represented by pinned_pages. A successful + * scif_register_pinned_pages() call returns po as the return value. + * + * When SCIF_MAP_FIXED is set in the map_flags argument, po will be offset + * exactly, and offset is constrained to be a multiple of the page size. The + * mapping established by scif_register_pinned_pages() will not replace any + * existing registration; an error is returned if any page of the new window + * would intersect an existing window. + * + * When SCIF_MAP_FIXED is not set, the implementation uses offset in an + * implementation-defined manner to arrive at po. The po so chosen will be an + * area of the registered address space that the implementation deems suitable + * for a mapping of the required size. An offset value of 0 is interpreted as + * granting the implementation complete freedom in selecting po, subject to + * constraints described below. A non-zero value of offset is taken to be a + * suggestion of an offset near which the mapping should be placed. When the + * implementation selects a value for po, it does not replace any extant + * window. In all cases, po will be a multiple of the page size. + * + * The physical pages which are so represented by a window are available for + * access in calls to scif_get_pages(), scif_readfrom(), scif_writeto(), + * scif_vreadfrom(), and scif_vwriteto(). While a window is registered, the + * physical pages represented by the window will not be reused by the memory + * subsystem for any other purpose. Note that the same physical page may be + * represented by multiple windows. + * + * Windows created by scif_register_pinned_pages() are unregistered by + * scif_unregister(). + * + * The map_flags argument can be set to SCIF_MAP_FIXED which interprets a + * fixed offset. + * + * Return: + * Upon successful completion, scif_register_pinned_pages() returns the offset + * at which the mapping was placed (po); otherwise the negative of one of the + * following errors is returned. + * + * Errors: + * EADDRINUSE - SCIF_MAP_FIXED is set in map_flags and pages in the new window + * would intersect an existing window + * EAGAIN - The mapping could not be performed due to lack of resources + * ECONNRESET - Connection reset by peer + * EINVAL - map_flags is invalid, or SCIF_MAP_FIXED is set in map_flags, and + * offset is not a multiple of the page size, or offset is negative + * ENODEV - The remote node is lost or existed, but is not currently in the + * network since it may have crashed + * ENOMEM - Not enough space + * ENOTCONN - The endpoint is not connected + */ +off_t scif_register_pinned_pages(scif_epd_t epd, + scif_pinned_pages_t pinned_pages, + off_t offset, int map_flags); + +/** + * scif_get_pages() - Add references to remote registered pages + * @epd: endpoint descriptor + * @offset: remote registered offset + * @len: length of range of pages + * @pages: returned scif_range structure + * + * scif_get_pages() returns the addresses of the physical pages represented by + * those pages of the registered address space of the peer of epd, starting at + * offset and continuing for len bytes. offset and len are constrained to be + * multiples of the page size. + * + * All of the pages in the specified range [offset, offset + len - 1] must be + * within a single window of the registered address space of the peer of epd. + * + * The addresses are returned as a virtually contiguous array pointed to by the + * phys_addr component of the scif_range structure whose address is returned in + * pages. The nr_pages component of scif_range is the length of the array. The + * prot_flags component of scif_range holds the protection flag value passed + * when the pages were registered. + * + * Each physical page whose address is returned by scif_get_pages() remains + * available and will not be released for reuse until the scif_range structure + * is returned in a call to scif_put_pages(). The scif_range structure returned + * by scif_get_pages() must be unmodified. + * + * It is an error to call scif_close() on an endpoint on which a scif_range + * structure of that endpoint has not been returned to scif_put_pages(). + * + * Return: + * Upon successful completion, scif_get_pages() returns 0; otherwise the + * negative of one of the following errors is returned. + * Errors: + * ECONNRESET - Connection reset by peer. + * EINVAL - offset is not a multiple of the page size, or offset is negative, or + * len is not a multiple of the page size + * ENODEV - The remote node is lost or existed, but is not currently in the + * network since it may have crashed + * ENOTCONN - The endpoint is not connected + * ENXIO - Offsets in the range [offset, offset + len - 1] are invalid + * for the registered address space of the peer epd + */ +int scif_get_pages(scif_epd_t epd, off_t offset, size_t len, + struct scif_range **pages); + +/** + * scif_put_pages() - Remove references from remote registered pages + * @pages: pages to be returned + * + * scif_put_pages() releases a scif_range structure previously obtained by + * calling scif_get_pages(). The physical pages represented by pages may + * be reused when the window which represented those pages is unregistered. + * Therefore, those pages must not be accessed after calling scif_put_pages(). + * + * Return: + * Upon successful completion, scif_put_pages() returns 0; otherwise the + * negative of one of the following errors is returned. + * Errors: + * EINVAL - pages does not point to a valid scif_range structure, or + * the scif_range structure pointed to by pages was already returned + * ENODEV - The remote node is lost or existed, but is not currently in the + * network since it may have crashed + * ENOTCONN - The endpoint is not connected + */ +int scif_put_pages(struct scif_range *pages); + +/** + * scif_poll() - Wait for some event on an endpoint + * @epds: Array of endpoint descriptors + * @nepds: Length of epds + * @timeout: Upper limit on time for which scif_poll() will block + * + * scif_poll() waits for one of a set of endpoints to become ready to perform + * an I/O operation. + * + * The epds argument specifies the endpoint descriptors to be examined and the + * events of interest for each endpoint descriptor. epds is a pointer to an + * array with one member for each open endpoint descriptor of interest. + * + * The number of items in the epds array is specified in nepds. The epd field + * of scif_pollepd is an endpoint descriptor of an open endpoint. The field + * events is a bitmask specifying the events which the application is + * interested in. The field revents is an output parameter, filled by the + * kernel with the events that actually occurred. The bits returned in revents + * can include any of those specified in events, or one of the values POLLERR, + * POLLHUP, or POLLNVAL. (These three bits are meaningless in the events + * field, and will be set in the revents field whenever the corresponding + * condition is true.) + * + * If none of the events requested (and no error) has occurred for any of the + * endpoint descriptors, then scif_poll() blocks until one of the events occurs. + * + * The timeout argument specifies an upper limit on the time for which + * scif_poll() will block, in milliseconds. Specifying a negative value in + * timeout means an infinite timeout. + * + * The following bits may be set in events and returned in revents. + * POLLIN - Data may be received without blocking. For a connected + * endpoint, this means that scif_recv() may be called without blocking. For a + * listening endpoint, this means that scif_accept() may be called without + * blocking. + * POLLOUT - Data may be sent without blocking. For a connected endpoint, this + * means that scif_send() may be called without blocking. POLLOUT may also be + * used to block waiting for a non-blocking connect to complete. This bit value + * has no meaning for a listening endpoint and is ignored if specified. + * + * The following bits are only returned in revents, and are ignored if set in + * events. + * POLLERR - An error occurred on the endpoint + * POLLHUP - The connection to the peer endpoint was disconnected + * POLLNVAL - The specified endpoint descriptor is invalid. + * + * Return: + * Upon successful completion, scif_poll() returns a non-negative value. A + * positive value indicates the total number of endpoint descriptors that have + * been selected (that is, endpoint descriptors for which the revents member is + * non-zero). A value of 0 indicates that the call timed out and no endpoint + * descriptors have been selected. Otherwise in user mode -1 is returned and + * errno is set to indicate the error; in kernel mode the negative of one of + * the following errors is returned. + * + * Errors: + * EINTR - A signal occurred before any requested event + * EINVAL - The nepds argument is greater than {OPEN_MAX} + * ENOMEM - There was no space to allocate file descriptor tables + */ +int scif_poll(struct scif_pollepd *epds, unsigned int nepds, long timeout); + +/** + * scif_client_register() - Register a SCIF client + * @client: client to be registered + * + * scif_client_register() registers a SCIF client. The probe() method + * of the client is called when SCIF peer devices come online and the + * remove() method is called when the peer devices disappear. + * + * Return: + * Upon successful completion, scif_client_register() returns a non-negative + * value. Otherwise the return value is the same as subsys_interface_register() + * in the kernel. + */ +int scif_client_register(struct scif_client *client); + +/** + * scif_client_unregister() - Unregister a SCIF client + * @client: client to be unregistered + * + * scif_client_unregister() unregisters a SCIF client. + * + * Return: + * None + */ +void scif_client_unregister(struct scif_client *client); #endif /* __SCIF_H__ */ diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 2b0a30a6e31c..4398411236f1 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -2708,7 +2708,7 @@ static inline void skb_postpull_rcsum(struct sk_buff *skb, if (skb->ip_summed == CHECKSUM_COMPLETE) skb->csum = csum_sub(skb->csum, csum_partial(start, len, 0)); else if (skb->ip_summed == CHECKSUM_PARTIAL && - skb_checksum_start_offset(skb) <= len) + skb_checksum_start_offset(skb) < 0) skb->ip_summed = CHECKSUM_NONE; } diff --git a/include/linux/spmi.h b/include/linux/spmi.h index f84212cd3b7d..1396a255d2a2 100644 --- a/include/linux/spmi.h +++ b/include/linux/spmi.h @@ -153,7 +153,9 @@ static inline struct spmi_driver *to_spmi_driver(struct device_driver *d) return container_of(d, struct spmi_driver, driver); } -int spmi_driver_register(struct spmi_driver *sdrv); +#define spmi_driver_register(sdrv) \ + __spmi_driver_register(sdrv, THIS_MODULE) +int __spmi_driver_register(struct spmi_driver *sdrv, struct module *owner); /** * spmi_driver_unregister() - unregister an SPMI client driver diff --git a/include/linux/stm.h b/include/linux/stm.h new file mode 100644 index 000000000000..9d0083d364e6 --- /dev/null +++ b/include/linux/stm.h @@ -0,0 +1,126 @@ +/* + * System Trace Module (STM) infrastructure apis + * Copyright (C) 2014 Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ + +#ifndef _STM_H_ +#define _STM_H_ + +#include <linux/device.h> + +/** + * enum stp_packet_type - STP packets that an STM driver sends + */ +enum stp_packet_type { + STP_PACKET_DATA = 0, + STP_PACKET_FLAG, + STP_PACKET_USER, + STP_PACKET_MERR, + STP_PACKET_GERR, + STP_PACKET_TRIG, + STP_PACKET_XSYNC, +}; + +/** + * enum stp_packet_flags - STP packet modifiers + */ +enum stp_packet_flags { + STP_PACKET_MARKED = 0x1, + STP_PACKET_TIMESTAMPED = 0x2, +}; + +struct stp_policy; + +struct stm_device; + +/** + * struct stm_data - STM device description and callbacks + * @name: device name + * @stm: internal structure, only used by stm class code + * @sw_start: first STP master available to software + * @sw_end: last STP master available to software + * @sw_nchannels: number of STP channels per master + * @sw_mmiosz: size of one channel's IO space, for mmap, optional + * @packet: callback that sends an STP packet + * @mmio_addr: mmap callback, optional + * @link: called when a new stm_source gets linked to us, optional + * @unlink: likewise for unlinking, again optional + * @set_options: set device-specific options on a channel + * + * Fill out this structure before calling stm_register_device() to create + * an STM device and stm_unregister_device() to destroy it. It will also be + * passed back to @packet(), @mmio_addr(), @link(), @unlink() and @set_options() + * callbacks. + * + * Normally, an STM device will have a range of masters available to software + * and the rest being statically assigned to various hardware trace sources. + * The former is defined by the the range [@sw_start..@sw_end] of the device + * description. That is, the lowest master that can be allocated to software + * writers is @sw_start and data from this writer will appear is @sw_start + * master in the STP stream. + */ +struct stm_data { + const char *name; + struct stm_device *stm; + unsigned int sw_start; + unsigned int sw_end; + unsigned int sw_nchannels; + unsigned int sw_mmiosz; + ssize_t (*packet)(struct stm_data *, unsigned int, + unsigned int, unsigned int, + unsigned int, unsigned int, + const unsigned char *); + phys_addr_t (*mmio_addr)(struct stm_data *, unsigned int, + unsigned int, unsigned int); + int (*link)(struct stm_data *, unsigned int, + unsigned int); + void (*unlink)(struct stm_data *, unsigned int, + unsigned int); + long (*set_options)(struct stm_data *, unsigned int, + unsigned int, unsigned int, + unsigned long); +}; + +int stm_register_device(struct device *parent, struct stm_data *stm_data, + struct module *owner); +void stm_unregister_device(struct stm_data *stm_data); + +struct stm_source_device; + +/** + * struct stm_source_data - STM source device description and callbacks + * @name: device name, will be used for policy lookup + * @src: internal structure, only used by stm class code + * @nr_chans: number of channels to allocate + * @link: called when this source gets linked to an STM device + * @unlink: called when this source is about to get unlinked from its STM + * + * Fill in this structure before calling stm_source_register_device() to + * register a source device. Also pass it to unregister and write calls. + */ +struct stm_source_data { + const char *name; + struct stm_source_device *src; + unsigned int percpu; + unsigned int nr_chans; + int (*link)(struct stm_source_data *data); + void (*unlink)(struct stm_source_data *data); +}; + +int stm_source_register_device(struct device *parent, + struct stm_source_data *data); +void stm_source_unregister_device(struct stm_source_data *data); + +int stm_source_write(struct stm_source_data *data, unsigned int chan, + const char *buf, size_t count); + +#endif /* _STM_H_ */ diff --git a/include/linux/string.h b/include/linux/string.h index a8d90db9c4b0..9ef7795e65e4 100644 --- a/include/linux/string.h +++ b/include/linux/string.h @@ -25,6 +25,9 @@ extern char * strncpy(char *,const char *, __kernel_size_t); #ifndef __HAVE_ARCH_STRLCPY size_t strlcpy(char *, const char *, size_t); #endif +#ifndef __HAVE_ARCH_STRSCPY +ssize_t __must_check strscpy(char *, const char *, size_t); +#endif #ifndef __HAVE_ARCH_STRCAT extern char * strcat(char *, const char *); #endif diff --git a/include/linux/ti_wilink_st.h b/include/linux/ti_wilink_st.h index d4217eff489f..0a0d56834c8e 100644 --- a/include/linux/ti_wilink_st.h +++ b/include/linux/ti_wilink_st.h @@ -158,6 +158,7 @@ struct st_data_s { unsigned long ll_state; void *kim_data; struct tty_struct *tty; + struct work_struct work_write_wakeup; }; /* diff --git a/include/linux/usb/renesas_usbhs.h b/include/linux/usb/renesas_usbhs.h index 3dd5a781da99..bfb74723f151 100644 --- a/include/linux/usb/renesas_usbhs.h +++ b/include/linux/usb/renesas_usbhs.h @@ -157,7 +157,7 @@ struct renesas_usbhs_driver_param { */ int pio_dma_border; /* default is 64byte */ - u32 type; + uintptr_t type; u32 enable_gpio; /* diff --git a/include/net/af_unix.h b/include/net/af_unix.h index 4a167b30a12f..cb1b9bbda332 100644 --- a/include/net/af_unix.h +++ b/include/net/af_unix.h @@ -63,7 +63,11 @@ struct unix_sock { #define UNIX_GC_MAYBE_CYCLE 1 struct socket_wq peer_wq; }; -#define unix_sk(__sk) ((struct unix_sock *)__sk) + +static inline struct unix_sock *unix_sk(struct sock *sk) +{ + return (struct unix_sock *)sk; +} #define peer_wait peer_wq.wait diff --git a/include/uapi/linux/mic_common.h b/include/uapi/linux/mic_common.h index 302a2ced373c..e9686372029d 100644 --- a/include/uapi/linux/mic_common.h +++ b/include/uapi/linux/mic_common.h @@ -75,12 +75,7 @@ struct mic_device_ctrl { * struct mic_bootparam: Virtio device independent information in device page * * @magic: A magic value used by the card to ensure it can see the host - * @c2h_shutdown_db: Card to Host shutdown doorbell set by host - * @h2c_shutdown_db: Host to Card shutdown doorbell set by card * @h2c_config_db: Host to Card Virtio config doorbell set by card - * @shutdown_status: Card shutdown status set by card - * @shutdown_card: Set to 1 by the host when a card shutdown is initiated - * @tot_nodes: Total number of nodes in the SCIF network * @node_id: Unique id of the node * @h2c_scif_db - Host to card SCIF doorbell set by card * @c2h_scif_db - Card to host SCIF doorbell set by host @@ -89,12 +84,7 @@ struct mic_device_ctrl { */ struct mic_bootparam { __le32 magic; - __s8 c2h_shutdown_db; - __s8 h2c_shutdown_db; __s8 h2c_config_db; - __u8 shutdown_status; - __u8 shutdown_card; - __u8 tot_nodes; __u8 node_id; __u8 h2c_scif_db; __u8 c2h_scif_db; @@ -219,12 +209,12 @@ static inline unsigned mic_total_desc_size(struct mic_device_desc *desc) * enum mic_states - MIC states. */ enum mic_states { - MIC_OFFLINE = 0, + MIC_READY = 0, + MIC_BOOTING, MIC_ONLINE, MIC_SHUTTING_DOWN, + MIC_RESETTING, MIC_RESET_FAILED, - MIC_SUSPENDING, - MIC_SUSPENDED, MIC_LAST }; diff --git a/include/uapi/linux/scif_ioctl.h b/include/uapi/linux/scif_ioctl.h index 4a94d917cf99..d9048918be52 100644 --- a/include/uapi/linux/scif_ioctl.h +++ b/include/uapi/linux/scif_ioctl.h @@ -107,6 +107,82 @@ struct scifioctl_msg { }; /** + * struct scifioctl_reg - used for SCIF_REG IOCTL + * @addr: starting virtual address + * @len: length of range + * @offset: offset of window + * @prot: read/write protection + * @flags: flags + * @out_offset: offset returned + */ +struct scifioctl_reg { + __u64 addr; + __u64 len; + __s64 offset; + __s32 prot; + __s32 flags; + __s64 out_offset; +}; + +/** + * struct scifioctl_unreg - used for SCIF_UNREG IOCTL + * @offset: start of range to unregister + * @len: length of range to unregister + */ +struct scifioctl_unreg { + __s64 offset; + __u64 len; +}; + +/** + * struct scifioctl_copy - used for SCIF DMA copy IOCTLs + * + * @loffset: offset in local registered address space to/from + * which to copy + * @len: length of range to copy + * @roffset: offset in remote registered address space to/from + * which to copy + * @addr: user virtual address to/from which to copy + * @flags: flags + * + * This structure is used for SCIF_READFROM, SCIF_WRITETO, SCIF_VREADFROM + * and SCIF_VREADFROM IOCTL's. + */ +struct scifioctl_copy { + __s64 loffset; + __u64 len; + __s64 roffset; + __u64 addr; + __s32 flags; +}; + +/** + * struct scifioctl_fence_mark - used for SCIF_FENCE_MARK IOCTL + * @flags: flags + * @mark: fence handle which is a pointer to a __s32 + */ +struct scifioctl_fence_mark { + __s32 flags; + __u64 mark; +}; + +/** + * struct scifioctl_fence_signal - used for SCIF_FENCE_SIGNAL IOCTL + * @loff: local offset + * @lval: value to write to loffset + * @roff: remote offset + * @rval: value to write to roffset + * @flags: flags + */ +struct scifioctl_fence_signal { + __s64 loff; + __u64 lval; + __s64 roff; + __u64 rval; + __s32 flags; +}; + +/** * struct scifioctl_node_ids - used for SCIF_GET_NODEIDS IOCTL * @nodes: pointer to an array of node_ids * @self: ID of the current node @@ -125,6 +201,15 @@ struct scifioctl_node_ids { #define SCIF_ACCEPTREG _IOWR('s', 5, __u64) #define SCIF_SEND _IOWR('s', 6, struct scifioctl_msg) #define SCIF_RECV _IOWR('s', 7, struct scifioctl_msg) +#define SCIF_REG _IOWR('s', 8, struct scifioctl_reg) +#define SCIF_UNREG _IOWR('s', 9, struct scifioctl_unreg) +#define SCIF_READFROM _IOWR('s', 10, struct scifioctl_copy) +#define SCIF_WRITETO _IOWR('s', 11, struct scifioctl_copy) +#define SCIF_VREADFROM _IOWR('s', 12, struct scifioctl_copy) +#define SCIF_VWRITETO _IOWR('s', 13, struct scifioctl_copy) #define SCIF_GET_NODEIDS _IOWR('s', 14, struct scifioctl_node_ids) +#define SCIF_FENCE_MARK _IOWR('s', 15, struct scifioctl_fence_mark) +#define SCIF_FENCE_WAIT _IOWR('s', 16, __s32) +#define SCIF_FENCE_SIGNAL _IOWR('s', 17, struct scifioctl_fence_signal) #endif /* SCIF_IOCTL_H */ diff --git a/include/uapi/linux/stm.h b/include/uapi/linux/stm.h new file mode 100644 index 000000000000..626a8d3f63b5 --- /dev/null +++ b/include/uapi/linux/stm.h @@ -0,0 +1,50 @@ +/* + * System Trace Module (STM) userspace interfaces + * Copyright (c) 2014, Intel Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * STM class implements generic infrastructure for System Trace Module devices + * as defined in MIPI STPv2 specification. + */ + +#ifndef _UAPI_LINUX_STM_H +#define _UAPI_LINUX_STM_H + +#include <linux/types.h> + +/** + * struct stp_policy_id - identification for the STP policy + * @size: size of the structure including real id[] length + * @master: assigned master + * @channel: first assigned channel + * @width: number of requested channels + * @id: identification string + * + * User must calculate the total size of the structure and put it into + * @size field, fill out the @id and desired @width. In return, kernel + * fills out @master, @channel and @width. + */ +struct stp_policy_id { + __u32 size; + __u16 master; + __u16 channel; + __u16 width; + /* padding */ + __u16 __reserved_0; + __u32 __reserved_1; + char id[0]; +}; + +#define STP_POLICY_ID_SET _IOWR('%', 0, struct stp_policy_id) +#define STP_POLICY_ID_GET _IOR('%', 1, struct stp_policy_id) +#define STP_SET_OPTIONS _IOW('%', 2, __u64) + +#endif /* _UAPI_LINUX_STM_H */ diff --git a/include/uapi/linux/userfaultfd.h b/include/uapi/linux/userfaultfd.h index df0e09bb7dd5..9057d7af3ae1 100644 --- a/include/uapi/linux/userfaultfd.h +++ b/include/uapi/linux/userfaultfd.h @@ -11,8 +11,6 @@ #include <linux/types.h> -#include <linux/compiler.h> - #define UFFD_API ((__u64)0xAA) /* * After implementing the respective features it will become: diff --git a/include/xen/interface/sched.h b/include/xen/interface/sched.h index 9ce083960a25..f18490985fc8 100644 --- a/include/xen/interface/sched.h +++ b/include/xen/interface/sched.h @@ -107,5 +107,13 @@ struct sched_watchdog { #define SHUTDOWN_suspend 2 /* Clean up, save suspend info, kill. */ #define SHUTDOWN_crash 3 /* Tell controller we've crashed. */ #define SHUTDOWN_watchdog 4 /* Restart because watchdog time expired. */ +/* + * Domain asked to perform 'soft reset' for it. The expected behavior is to + * reset internal Xen state for the domain returning it to the point where it + * was created but leaving the domain's memory contents and vCPU contexts + * intact. This will allow the domain to start over and set up all Xen specific + * interfaces again. + */ +#define SHUTDOWN_soft_reset 5 #endif /* __XEN_PUBLIC_SCHED_H__ */ diff --git a/ipc/msg.c b/ipc/msg.c index 66c4f567eb73..1471db9a7e61 100644 --- a/ipc/msg.c +++ b/ipc/msg.c @@ -137,13 +137,6 @@ static int newque(struct ipc_namespace *ns, struct ipc_params *params) return retval; } - /* ipc_addid() locks msq upon success. */ - id = ipc_addid(&msg_ids(ns), &msq->q_perm, ns->msg_ctlmni); - if (id < 0) { - ipc_rcu_putref(msq, msg_rcu_free); - return id; - } - msq->q_stime = msq->q_rtime = 0; msq->q_ctime = get_seconds(); msq->q_cbytes = msq->q_qnum = 0; @@ -153,6 +146,13 @@ static int newque(struct ipc_namespace *ns, struct ipc_params *params) INIT_LIST_HEAD(&msq->q_receivers); INIT_LIST_HEAD(&msq->q_senders); + /* ipc_addid() locks msq upon success. */ + id = ipc_addid(&msg_ids(ns), &msq->q_perm, ns->msg_ctlmni); + if (id < 0) { + ipc_rcu_putref(msq, msg_rcu_free); + return id; + } + ipc_unlock_object(&msq->q_perm); rcu_read_unlock(); diff --git a/ipc/shm.c b/ipc/shm.c index 222131e8e38f..41787276e141 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -551,12 +551,6 @@ static int newseg(struct ipc_namespace *ns, struct ipc_params *params) if (IS_ERR(file)) goto no_file; - id = ipc_addid(&shm_ids(ns), &shp->shm_perm, ns->shm_ctlmni); - if (id < 0) { - error = id; - goto no_id; - } - shp->shm_cprid = task_tgid_vnr(current); shp->shm_lprid = 0; shp->shm_atim = shp->shm_dtim = 0; @@ -565,6 +559,13 @@ static int newseg(struct ipc_namespace *ns, struct ipc_params *params) shp->shm_nattch = 0; shp->shm_file = file; shp->shm_creator = current; + + id = ipc_addid(&shm_ids(ns), &shp->shm_perm, ns->shm_ctlmni); + if (id < 0) { + error = id; + goto no_id; + } + list_add(&shp->shm_clist, ¤t->sysvshm.shm_clist); /* diff --git a/ipc/util.c b/ipc/util.c index be4230020a1f..0f401d94b7c6 100644 --- a/ipc/util.c +++ b/ipc/util.c @@ -237,6 +237,10 @@ int ipc_addid(struct ipc_ids *ids, struct kern_ipc_perm *new, int size) rcu_read_lock(); spin_lock(&new->lock); + current_euid_egid(&euid, &egid); + new->cuid = new->uid = euid; + new->gid = new->cgid = egid; + id = idr_alloc(&ids->ipcs_idr, new, (next_id < 0) ? 0 : ipcid_to_idx(next_id), 0, GFP_NOWAIT); @@ -249,10 +253,6 @@ int ipc_addid(struct ipc_ids *ids, struct kern_ipc_perm *new, int size) ids->in_use++; - current_euid_egid(&euid, &egid); - new->cuid = new->uid = euid; - new->gid = new->cgid = egid; - if (next_id < 0) { new->seq = ids->seq++; if (ids->seq > IPCID_SEQ_MAX) diff --git a/kernel/events/core.c b/kernel/events/core.c index f548f69c4299..b11756f9b6dc 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -1243,11 +1243,7 @@ static inline void perf_event__state_init(struct perf_event *event) PERF_EVENT_STATE_INACTIVE; } -/* - * Called at perf_event creation and when events are attached/detached from a - * group. - */ -static void perf_event__read_size(struct perf_event *event) +static void __perf_event_read_size(struct perf_event *event, int nr_siblings) { int entry = sizeof(u64); /* value */ int size = 0; @@ -1263,7 +1259,7 @@ static void perf_event__read_size(struct perf_event *event) entry += sizeof(u64); if (event->attr.read_format & PERF_FORMAT_GROUP) { - nr += event->group_leader->nr_siblings; + nr += nr_siblings; size += sizeof(u64); } @@ -1271,14 +1267,11 @@ static void perf_event__read_size(struct perf_event *event) event->read_size = size; } -static void perf_event__header_size(struct perf_event *event) +static void __perf_event_header_size(struct perf_event *event, u64 sample_type) { struct perf_sample_data *data; - u64 sample_type = event->attr.sample_type; u16 size = 0; - perf_event__read_size(event); - if (sample_type & PERF_SAMPLE_IP) size += sizeof(data->ip); @@ -1303,6 +1296,17 @@ static void perf_event__header_size(struct perf_event *event) event->header_size = size; } +/* + * Called at perf_event creation and when events are attached/detached from a + * group. + */ +static void perf_event__header_size(struct perf_event *event) +{ + __perf_event_read_size(event, + event->group_leader->nr_siblings); + __perf_event_header_size(event, event->attr.sample_type); +} + static void perf_event__id_header_size(struct perf_event *event) { struct perf_sample_data *data; @@ -1330,6 +1334,27 @@ static void perf_event__id_header_size(struct perf_event *event) event->id_header_size = size; } +static bool perf_event_validate_size(struct perf_event *event) +{ + /* + * The values computed here will be over-written when we actually + * attach the event. + */ + __perf_event_read_size(event, event->group_leader->nr_siblings + 1); + __perf_event_header_size(event, event->attr.sample_type & ~PERF_SAMPLE_READ); + perf_event__id_header_size(event); + + /* + * Sum the lot; should not exceed the 64k limit we have on records. + * Conservative limit to allow for callchains and other variable fields. + */ + if (event->read_size + event->header_size + + event->id_header_size + sizeof(struct perf_event_header) >= 16*1024) + return false; + + return true; +} + static void perf_group_attach(struct perf_event *event) { struct perf_event *group_leader = event->group_leader, *pos; @@ -8297,13 +8322,35 @@ SYSCALL_DEFINE5(perf_event_open, if (move_group) { gctx = group_leader->ctx; + mutex_lock_double(&gctx->mutex, &ctx->mutex); + } else { + mutex_lock(&ctx->mutex); + } + if (!perf_event_validate_size(event)) { + err = -E2BIG; + goto err_locked; + } + + /* + * Must be under the same ctx::mutex as perf_install_in_context(), + * because we need to serialize with concurrent event creation. + */ + if (!exclusive_event_installable(event, ctx)) { + /* exclusive and group stuff are assumed mutually exclusive */ + WARN_ON_ONCE(move_group); + + err = -EBUSY; + goto err_locked; + } + + WARN_ON_ONCE(ctx->parent_ctx); + + if (move_group) { /* * See perf_event_ctx_lock() for comments on the details * of swizzling perf_event::ctx. */ - mutex_lock_double(&gctx->mutex, &ctx->mutex); - perf_remove_from_context(group_leader, false); list_for_each_entry(sibling, &group_leader->sibling_list, @@ -8311,13 +8358,7 @@ SYSCALL_DEFINE5(perf_event_open, perf_remove_from_context(sibling, false); put_ctx(gctx); } - } else { - mutex_lock(&ctx->mutex); - } - WARN_ON_ONCE(ctx->parent_ctx); - - if (move_group) { /* * Wait for everybody to stop referencing the events through * the old lists, before installing it on new lists. @@ -8349,22 +8390,29 @@ SYSCALL_DEFINE5(perf_event_open, perf_event__state_init(group_leader); perf_install_in_context(ctx, group_leader, group_leader->cpu); get_ctx(ctx); - } - if (!exclusive_event_installable(event, ctx)) { - err = -EBUSY; - mutex_unlock(&ctx->mutex); - fput(event_file); - goto err_context; + /* + * Now that all events are installed in @ctx, nothing + * references @gctx anymore, so drop the last reference we have + * on it. + */ + put_ctx(gctx); } + /* + * Precalculate sample_data sizes; do while holding ctx::mutex such + * that we're serialized against further additions and before + * perf_install_in_context() which is the point the event is active and + * can use these values. + */ + perf_event__header_size(event); + perf_event__id_header_size(event); + perf_install_in_context(ctx, event, event->cpu); perf_unpin_context(ctx); - if (move_group) { + if (move_group) mutex_unlock(&gctx->mutex); - put_ctx(gctx); - } mutex_unlock(&ctx->mutex); put_online_cpus(); @@ -8376,12 +8424,6 @@ SYSCALL_DEFINE5(perf_event_open, mutex_unlock(¤t->perf_event_mutex); /* - * Precalculate sample_data sizes - */ - perf_event__header_size(event); - perf_event__id_header_size(event); - - /* * Drop the reference on the group_event after placing the * new event on the sibling_list. This ensures destruction * of the group leader will find the pointer to itself in @@ -8391,6 +8433,12 @@ SYSCALL_DEFINE5(perf_event_open, fd_install(event_fd, event_file); return event_fd; +err_locked: + if (move_group) + mutex_unlock(&gctx->mutex); + mutex_unlock(&ctx->mutex); +/* err_file: */ + fput(event_file); err_context: perf_unpin_context(ctx); put_ctx(ctx); diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index de41a68fc038..e25a83b67cce 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -22,7 +22,6 @@ /** * handle_bad_irq - handle spurious and unhandled irqs - * @irq: the interrupt number * @desc: description of the interrupt * * Handles spurious and unhandled IRQ's. It also prints a debugmessage. @@ -35,6 +34,7 @@ void handle_bad_irq(struct irq_desc *desc) kstat_incr_irqs_this_cpu(desc); ack_bad_irq(irq); } +EXPORT_SYMBOL_GPL(handle_bad_irq); /* * Special, empty irq handler: diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c index e3a8c9577ba6..a50ddc9417ff 100644 --- a/kernel/irq/proc.c +++ b/kernel/irq/proc.c @@ -12,6 +12,7 @@ #include <linux/seq_file.h> #include <linux/interrupt.h> #include <linux/kernel_stat.h> +#include <linux/mutex.h> #include "internals.h" @@ -323,18 +324,29 @@ void register_handler_proc(unsigned int irq, struct irqaction *action) void register_irq_proc(unsigned int irq, struct irq_desc *desc) { + static DEFINE_MUTEX(register_lock); char name [MAX_NAMELEN]; - if (!root_irq_dir || (desc->irq_data.chip == &no_irq_chip) || desc->dir) + if (!root_irq_dir || (desc->irq_data.chip == &no_irq_chip)) return; + /* + * irq directories are registered only when a handler is + * added, not when the descriptor is created, so multiple + * tasks might try to register at the same time. + */ + mutex_lock(®ister_lock); + + if (desc->dir) + goto out_unlock; + memset(name, 0, MAX_NAMELEN); sprintf(name, "%d", irq); /* create /proc/irq/1234 */ desc->dir = proc_mkdir(name, root_irq_dir); if (!desc->dir) - return; + goto out_unlock; #ifdef CONFIG_SMP /* create /proc/irq/<irq>/smp_affinity */ @@ -355,6 +367,9 @@ void register_irq_proc(unsigned int irq, struct irq_desc *desc) proc_create_data("spurious", 0444, desc->dir, &irq_spurious_proc_fops, (void *)(long)irq); + +out_unlock: + mutex_unlock(®ister_lock); } void unregister_irq_proc(unsigned int irq, struct irq_desc *desc) diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index 8acfbf773e06..4e49cc4c9952 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -3068,7 +3068,7 @@ static int __lock_is_held(struct lockdep_map *lock); static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, int trylock, int read, int check, int hardirqs_off, struct lockdep_map *nest_lock, unsigned long ip, - int references) + int references, int pin_count) { struct task_struct *curr = current; struct lock_class *class = NULL; @@ -3157,7 +3157,7 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, hlock->waittime_stamp = 0; hlock->holdtime_stamp = lockstat_clock(); #endif - hlock->pin_count = 0; + hlock->pin_count = pin_count; if (check && !mark_irqflags(curr, hlock)) return 0; @@ -3343,7 +3343,7 @@ found_it: hlock_class(hlock)->subclass, hlock->trylock, hlock->read, hlock->check, hlock->hardirqs_off, hlock->nest_lock, hlock->acquire_ip, - hlock->references)) + hlock->references, hlock->pin_count)) return 0; } @@ -3433,7 +3433,7 @@ found_it: hlock_class(hlock)->subclass, hlock->trylock, hlock->read, hlock->check, hlock->hardirqs_off, hlock->nest_lock, hlock->acquire_ip, - hlock->references)) + hlock->references, hlock->pin_count)) return 0; } @@ -3583,7 +3583,7 @@ void lock_acquire(struct lockdep_map *lock, unsigned int subclass, current->lockdep_recursion = 1; trace_lock_acquire(lock, subclass, trylock, read, check, nest_lock, ip); __lock_acquire(lock, subclass, trylock, read, check, - irqs_disabled_flags(flags), nest_lock, ip, 0); + irqs_disabled_flags(flags), nest_lock, ip, 0, 0); current->lockdep_recursion = 0; raw_local_irq_restore(flags); } diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 9f75f25cc5d9..775d36cc0050 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -3868,6 +3868,7 @@ static void rcu_init_new_rnp(struct rcu_node *rnp_leaf) static void __init rcu_boot_init_percpu_data(int cpu, struct rcu_state *rsp) { + static struct lock_class_key rcu_exp_sched_rdp_class; unsigned long flags; struct rcu_data *rdp = per_cpu_ptr(rsp->rda, cpu); struct rcu_node *rnp = rcu_get_root(rsp); @@ -3883,6 +3884,10 @@ rcu_boot_init_percpu_data(int cpu, struct rcu_state *rsp) mutex_init(&rdp->exp_funnel_mutex); rcu_boot_init_nocb_percpu_data(rdp); raw_spin_unlock_irqrestore(&rnp->lock, flags); + if (rsp == &rcu_sched_state) + lockdep_set_class_and_name(&rdp->exp_funnel_mutex, + &rcu_exp_sched_rdp_class, + "rcu_data_exp_sched"); } /* diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 2f9c92884817..10a8faa1b0d4 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -2517,11 +2517,11 @@ static struct rq *finish_task_switch(struct task_struct *prev) * If a task dies, then it sets TASK_DEAD in tsk->state and calls * schedule one last time. The schedule call will never return, and * the scheduled task must drop that reference. - * The test for TASK_DEAD must occur while the runqueue locks are - * still held, otherwise prev could be scheduled on another cpu, die - * there before we look at prev->state, and then the reference would - * be dropped twice. - * Manfred Spraul <manfred@colorfullife.com> + * + * We must observe prev->state before clearing prev->on_cpu (in + * finish_lock_switch), otherwise a concurrent wakeup can get prev + * running on another CPU and we could rave with its RUNNING -> DEAD + * transition, resulting in a double drop. */ prev_state = prev->state; vtime_task_switch(prev); @@ -4934,7 +4934,15 @@ void init_idle(struct task_struct *idle, int cpu) idle->state = TASK_RUNNING; idle->se.exec_start = sched_clock(); - do_set_cpus_allowed(idle, cpumask_of(cpu)); +#ifdef CONFIG_SMP + /* + * Its possible that init_idle() gets called multiple times on a task, + * in that case do_set_cpus_allowed() will not do the right thing. + * + * And since this is boot we can forgo the serialization. + */ + set_cpus_allowed_common(idle, cpumask_of(cpu)); +#endif /* * We're having a chicken and egg problem, even though we are * holding rq->lock, the cpu isn't yet set to this cpu so the @@ -4951,7 +4959,7 @@ void init_idle(struct task_struct *idle, int cpu) rq->curr = rq->idle = idle; idle->on_rq = TASK_ON_RQ_QUEUED; -#if defined(CONFIG_SMP) +#ifdef CONFIG_SMP idle->on_cpu = 1; #endif raw_spin_unlock(&rq->lock); @@ -4966,7 +4974,7 @@ void init_idle(struct task_struct *idle, int cpu) idle->sched_class = &idle_sched_class; ftrace_graph_init_idle_task(idle, cpu); vtime_init_idle(idle, cpu); -#if defined(CONFIG_SMP) +#ifdef CONFIG_SMP sprintf(idle->comm, "%s/%d", INIT_TASK_COMM, cpu); #endif } diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 68cda117574c..6d2a119c7ad9 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -1078,9 +1078,10 @@ static inline void finish_lock_switch(struct rq *rq, struct task_struct *prev) * After ->on_cpu is cleared, the task can be moved to a different CPU. * We must ensure this doesn't happen until the switch is completely * finished. + * + * Pairs with the control dependency and rmb in try_to_wake_up(). */ - smp_wmb(); - prev->on_cpu = 0; + smp_store_release(&prev->on_cpu, 0); #endif #ifdef CONFIG_DEBUG_SPINLOCK /* this is a valid case when another task releases the spinlock */ diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index 841b72f720e8..3a38775b50c2 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -217,7 +217,7 @@ static void clocksource_watchdog(unsigned long data) continue; /* Check the deviation from the watchdog clocksource. */ - if ((abs(cs_nsec - wd_nsec) > WATCHDOG_THRESHOLD)) { + if (abs64(cs_nsec - wd_nsec) > WATCHDOG_THRESHOLD) { pr_warn("timekeeping watchdog: Marking clocksource '%s' as unstable because the skew is too large:\n", cs->name); pr_warn(" '%s' wd_now: %llx wd_last: %llx mask: %llx\n", diff --git a/lib/string.c b/lib/string.c index 13d1e84ddb80..84775ba873b9 100644 --- a/lib/string.c +++ b/lib/string.c @@ -27,6 +27,10 @@ #include <linux/bug.h> #include <linux/errno.h> +#include <asm/byteorder.h> +#include <asm/word-at-a-time.h> +#include <asm/page.h> + #ifndef __HAVE_ARCH_STRNCASECMP /** * strncasecmp - Case insensitive, length-limited string comparison @@ -146,6 +150,91 @@ size_t strlcpy(char *dest, const char *src, size_t size) EXPORT_SYMBOL(strlcpy); #endif +#ifndef __HAVE_ARCH_STRSCPY +/** + * strscpy - Copy a C-string into a sized buffer + * @dest: Where to copy the string to + * @src: Where to copy the string from + * @count: Size of destination buffer + * + * Copy the string, or as much of it as fits, into the dest buffer. + * The routine returns the number of characters copied (not including + * the trailing NUL) or -E2BIG if the destination buffer wasn't big enough. + * The behavior is undefined if the string buffers overlap. + * The destination buffer is always NUL terminated, unless it's zero-sized. + * + * Preferred to strlcpy() since the API doesn't require reading memory + * from the src string beyond the specified "count" bytes, and since + * the return value is easier to error-check than strlcpy()'s. + * In addition, the implementation is robust to the string changing out + * from underneath it, unlike the current strlcpy() implementation. + * + * Preferred to strncpy() since it always returns a valid string, and + * doesn't unnecessarily force the tail of the destination buffer to be + * zeroed. If the zeroing is desired, it's likely cleaner to use strscpy() + * with an overflow test, then just memset() the tail of the dest buffer. + */ +ssize_t strscpy(char *dest, const char *src, size_t count) +{ + const struct word_at_a_time constants = WORD_AT_A_TIME_CONSTANTS; + size_t max = count; + long res = 0; + + if (count == 0) + return -E2BIG; + +#ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS + /* + * If src is unaligned, don't cross a page boundary, + * since we don't know if the next page is mapped. + */ + if ((long)src & (sizeof(long) - 1)) { + size_t limit = PAGE_SIZE - ((long)src & (PAGE_SIZE - 1)); + if (limit < max) + max = limit; + } +#else + /* If src or dest is unaligned, don't do word-at-a-time. */ + if (((long) dest | (long) src) & (sizeof(long) - 1)) + max = 0; +#endif + + while (max >= sizeof(unsigned long)) { + unsigned long c, data; + + c = *(unsigned long *)(src+res); + if (has_zero(c, &data, &constants)) { + data = prep_zero_mask(c, data, &constants); + data = create_zero_mask(data); + *(unsigned long *)(dest+res) = c & zero_bytemask(data); + return res + find_zero(data); + } + *(unsigned long *)(dest+res) = c; + res += sizeof(unsigned long); + count -= sizeof(unsigned long); + max -= sizeof(unsigned long); + } + + while (count) { + char c; + + c = src[res]; + dest[res] = c; + if (!c) + return res; + res++; + count--; + } + + /* Hit buffer length without finding a NUL; force NUL-termination. */ + if (res) + dest[res-1] = '\0'; + + return -E2BIG; +} +EXPORT_SYMBOL(strscpy); +#endif + #ifndef __HAVE_ARCH_STRCAT /** * strcat - Append one %NUL-terminated string to another diff --git a/mm/dmapool.c b/mm/dmapool.c index 71a8998cd03a..312a716fa14c 100644 --- a/mm/dmapool.c +++ b/mm/dmapool.c @@ -394,7 +394,7 @@ static struct dma_page *pool_find_page(struct dma_pool *pool, dma_addr_t dma) list_for_each_entry(page, &pool->page_list, page_list) { if (dma < page->dma) continue; - if (dma < (page->dma + pool->allocation)) + if ((dma - page->dma) < pool->allocation) return page; } return NULL; diff --git a/mm/filemap.c b/mm/filemap.c index 72940fb38666..1cc5467cf36c 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -2473,6 +2473,21 @@ ssize_t generic_perform_write(struct file *file, iov_iter_count(i)); again: + /* + * Bring in the user page that we will copy from _first_. + * Otherwise there's a nasty deadlock on copying from the + * same page as we're writing to, without it being marked + * up-to-date. + * + * Not only is this an optimisation, but it is also required + * to check that the address is actually valid, when atomic + * usercopies are used, below. + */ + if (unlikely(iov_iter_fault_in_readable(i, bytes))) { + status = -EFAULT; + break; + } + status = a_ops->write_begin(file, mapping, pos, bytes, flags, &page, &fsdata); if (unlikely(status < 0)) @@ -2480,17 +2495,8 @@ again: if (mapping_writably_mapped(mapping)) flush_dcache_page(page); - /* - * 'page' is now locked. If we are trying to copy from a - * mapping of 'page' in userspace, the copy might fault and - * would need PageUptodate() to complete. But, page can not be - * made Uptodate without acquiring the page lock, which we hold. - * Deadlock. Avoid with pagefault_disable(). Fix up below with - * iov_iter_fault_in_readable(). - */ - pagefault_disable(); + copied = iov_iter_copy_from_user_atomic(page, i, offset, bytes); - pagefault_enable(); flush_dcache_page(page); status = a_ops->write_end(file, mapping, pos, bytes, copied, @@ -2513,14 +2519,6 @@ again: */ bytes = min_t(unsigned long, PAGE_CACHE_SIZE - offset, iov_iter_single_seg_count(i)); - /* - * This is the fallback to recover if the copy from - * userspace above faults. - */ - if (unlikely(iov_iter_fault_in_readable(i, bytes))) { - status = -EFAULT; - break; - } goto again; } pos += copied; diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 999fb0aef8f1..9cc773483624 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -3202,6 +3202,14 @@ static void unmap_ref_private(struct mm_struct *mm, struct vm_area_struct *vma, continue; /* + * Shared VMAs have their own reserves and do not affect + * MAP_PRIVATE accounting but it is possible that a shared + * VMA is using the same page so check and skip such VMAs. + */ + if (iter_vma->vm_flags & VM_MAYSHARE) + continue; + + /* * Unmap the page from other VMAs without their own reserves. * They get marked to be SIGKILLed if they fault in these * areas. This is because a future no-page fault on this VMA diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 6ddaeba34e09..1fedbde68f59 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -644,12 +644,14 @@ mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_zone *mctz) } /* + * Return page count for single (non recursive) @memcg. + * * Implementation Note: reading percpu statistics for memcg. * * Both of vmstat[] and percpu_counter has threshold and do periodic * synchronization to implement "quick" read. There are trade-off between * reading cost and precision of value. Then, we may have a chance to implement - * a periodic synchronizion of counter in memcg's counter. + * a periodic synchronization of counter in memcg's counter. * * But this _read() function is used for user interface now. The user accounts * memory usage by memory cgroup and he _always_ requires exact value because @@ -659,17 +661,24 @@ mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_zone *mctz) * * If there are kernel internal actions which can make use of some not-exact * value, and reading all cpu value can be performance bottleneck in some - * common workload, threashold and synchonization as vmstat[] should be + * common workload, threshold and synchronization as vmstat[] should be * implemented. */ -static long mem_cgroup_read_stat(struct mem_cgroup *memcg, - enum mem_cgroup_stat_index idx) +static unsigned long +mem_cgroup_read_stat(struct mem_cgroup *memcg, enum mem_cgroup_stat_index idx) { long val = 0; int cpu; + /* Per-cpu values can be negative, use a signed accumulator */ for_each_possible_cpu(cpu) val += per_cpu(memcg->stat->count[idx], cpu); + /* + * Summing races with updates, so val may be negative. Avoid exposing + * transient negative values. + */ + if (val < 0) + val = 0; return val; } @@ -1254,7 +1263,7 @@ void mem_cgroup_print_oom_info(struct mem_cgroup *memcg, struct task_struct *p) for (i = 0; i < MEM_CGROUP_STAT_NSTATS; i++) { if (i == MEM_CGROUP_STAT_SWAP && !do_swap_account) continue; - pr_cont(" %s:%ldKB", mem_cgroup_stat_names[i], + pr_cont(" %s:%luKB", mem_cgroup_stat_names[i], K(mem_cgroup_read_stat(iter, i))); } @@ -2819,14 +2828,11 @@ static unsigned long tree_stat(struct mem_cgroup *memcg, enum mem_cgroup_stat_index idx) { struct mem_cgroup *iter; - long val = 0; + unsigned long val = 0; - /* Per-cpu values can be negative, use a signed accumulator */ for_each_mem_cgroup_tree(iter, memcg) val += mem_cgroup_read_stat(iter, idx); - if (val < 0) /* race ? */ - val = 0; return val; } @@ -3169,7 +3175,7 @@ static int memcg_stat_show(struct seq_file *m, void *v) for (i = 0; i < MEM_CGROUP_STAT_NSTATS; i++) { if (i == MEM_CGROUP_STAT_SWAP && !do_swap_account) continue; - seq_printf(m, "%s %ld\n", mem_cgroup_stat_names[i], + seq_printf(m, "%s %lu\n", mem_cgroup_stat_names[i], mem_cgroup_read_stat(memcg, i) * PAGE_SIZE); } @@ -3194,13 +3200,13 @@ static int memcg_stat_show(struct seq_file *m, void *v) (u64)memsw * PAGE_SIZE); for (i = 0; i < MEM_CGROUP_STAT_NSTATS; i++) { - long long val = 0; + unsigned long long val = 0; if (i == MEM_CGROUP_STAT_SWAP && !do_swap_account) continue; for_each_mem_cgroup_tree(mi, memcg) val += mem_cgroup_read_stat(mi, i) * PAGE_SIZE; - seq_printf(m, "total_%s %lld\n", mem_cgroup_stat_names[i], val); + seq_printf(m, "total_%s %llu\n", mem_cgroup_stat_names[i], val); } for (i = 0; i < MEM_CGROUP_EVENTS_NSTATS; i++) { @@ -4179,7 +4185,6 @@ static struct mem_cgroup *mem_cgroup_alloc(void) if (memcg_wb_domain_init(memcg, GFP_KERNEL)) goto out_free_stat; - spin_lock_init(&memcg->pcp_counter_lock); return memcg; out_free_stat: diff --git a/mm/migrate.c b/mm/migrate.c index 7452a00bbb50..842ecd7aaf7f 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -740,6 +740,15 @@ static int move_to_new_page(struct page *newpage, struct page *page, if (PageSwapBacked(page)) SetPageSwapBacked(newpage); + /* + * Indirectly called below, migrate_page_copy() copies PG_dirty and thus + * needs newpage's memcg set to transfer memcg dirty page accounting. + * So perform memcg migration in two steps: + * 1. set newpage->mem_cgroup (here) + * 2. clear page->mem_cgroup (below) + */ + set_page_memcg(newpage, page_memcg(page)); + mapping = page_mapping(page); if (!mapping) rc = migrate_page(mapping, newpage, page, mode); @@ -756,9 +765,10 @@ static int move_to_new_page(struct page *newpage, struct page *page, rc = fallback_migrate_page(mapping, newpage, page, mode); if (rc != MIGRATEPAGE_SUCCESS) { + set_page_memcg(newpage, NULL); newpage->mapping = NULL; } else { - mem_cgroup_migrate(page, newpage, false); + set_page_memcg(page, NULL); if (page_was_mapped) remove_migration_ptes(page, newpage); page->mapping = NULL; diff --git a/mm/slab.c b/mm/slab.c index c77ebe6cc87c..4fcc5dd8d5a6 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -2190,9 +2190,16 @@ __kmem_cache_create (struct kmem_cache *cachep, unsigned long flags) size += BYTES_PER_WORD; } #if FORCED_DEBUG && defined(CONFIG_DEBUG_PAGEALLOC) - if (size >= kmalloc_size(INDEX_NODE + 1) - && cachep->object_size > cache_line_size() - && ALIGN(size, cachep->align) < PAGE_SIZE) { + /* + * To activate debug pagealloc, off-slab management is necessary + * requirement. In early phase of initialization, small sized slab + * doesn't get initialized so it would not be possible. So, we need + * to check size >= 256. It guarantees that all necessary small + * sized slab is initialized in current slab initialization sequence. + */ + if (!slab_early_init && size >= kmalloc_size(INDEX_NODE) && + size >= 256 && cachep->object_size > cache_line_size() && + ALIGN(size, cachep->align) < PAGE_SIZE) { cachep->obj_offset += PAGE_SIZE - ALIGN(size, cachep->align); size = PAGE_SIZE; } diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c index 805a95a48107..830f8a7c1cb1 100644 --- a/net/core/net-sysfs.c +++ b/net/core/net-sysfs.c @@ -31,7 +31,6 @@ static const char fmt_hex[] = "%#x\n"; static const char fmt_long_hex[] = "%#lx\n"; static const char fmt_dec[] = "%d\n"; -static const char fmt_udec[] = "%u\n"; static const char fmt_ulong[] = "%lu\n"; static const char fmt_u64[] = "%llu\n"; @@ -202,7 +201,7 @@ static ssize_t speed_show(struct device *dev, if (netif_running(netdev)) { struct ethtool_cmd cmd; if (!__ethtool_get_settings(netdev, &cmd)) - ret = sprintf(buf, fmt_udec, ethtool_cmd_speed(&cmd)); + ret = sprintf(buf, fmt_dec, ethtool_cmd_speed(&cmd)); } rtnl_unlock(); return ret; diff --git a/net/core/skbuff.c b/net/core/skbuff.c index dad4dd37e2aa..fab4599ba8b2 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -2958,11 +2958,12 @@ EXPORT_SYMBOL_GPL(skb_append_pagefrags); */ unsigned char *skb_pull_rcsum(struct sk_buff *skb, unsigned int len) { + unsigned char *data = skb->data; + BUG_ON(len > skb->len); - skb->len -= len; - BUG_ON(skb->len < skb->data_len); - skb_postpull_rcsum(skb, skb->data, len); - return skb->data += len; + __skb_pull(skb, len); + skb_postpull_rcsum(skb, data, len); + return skb->data; } EXPORT_SYMBOL_GPL(skb_pull_rcsum); diff --git a/net/dsa/slave.c b/net/dsa/slave.c index cce97385f743..7d91f4612ac0 100644 --- a/net/dsa/slave.c +++ b/net/dsa/slave.c @@ -458,12 +458,17 @@ static int dsa_slave_stp_update(struct net_device *dev, u8 state) static int dsa_slave_port_attr_set(struct net_device *dev, struct switchdev_attr *attr) { - int ret = 0; + struct dsa_slave_priv *p = netdev_priv(dev); + struct dsa_switch *ds = p->parent; + int ret; switch (attr->id) { case SWITCHDEV_ATTR_PORT_STP_STATE: - if (attr->trans == SWITCHDEV_TRANS_COMMIT) - ret = dsa_slave_stp_update(dev, attr->u.stp_state); + if (attr->trans == SWITCHDEV_TRANS_PREPARE) + ret = ds->drv->port_stp_update ? 0 : -EOPNOTSUPP; + else + ret = ds->drv->port_stp_update(ds, p->port, + attr->u.stp_state); break; default: ret = -EOPNOTSUPP; diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index 6fcbd215cdbc..690bcbc59f26 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -340,6 +340,7 @@ static int __fib_validate_source(struct sk_buff *skb, __be32 src, __be32 dst, fl4.flowi4_tos = tos; fl4.flowi4_scope = RT_SCOPE_UNIVERSE; fl4.flowi4_tun_key.tun_id = 0; + fl4.flowi4_flags = 0; no_addr = idev->ifa_list == NULL; diff --git a/net/ipv4/route.c b/net/ipv4/route.c index c6ad99ad0ffb..c81deb85acb4 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -1737,6 +1737,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr, fl4.flowi4_mark = skb->mark; fl4.flowi4_tos = tos; fl4.flowi4_scope = RT_SCOPE_UNIVERSE; + fl4.flowi4_flags = 0; fl4.daddr = daddr; fl4.saddr = saddr; err = fib_lookup(net, &fl4, &res, 0); diff --git a/net/ipv6/route.c b/net/ipv6/route.c index f204089e854c..cb32ce250db0 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1193,7 +1193,8 @@ struct dst_entry *ip6_route_output(struct net *net, const struct sock *sk, fl6->flowi6_iif = LOOPBACK_IFINDEX; - if ((sk && sk->sk_bound_dev_if) || rt6_need_strict(&fl6->daddr)) + if ((sk && sk->sk_bound_dev_if) || rt6_need_strict(&fl6->daddr) || + fl6->flowi6_oif) flags |= RT6_LOOKUP_F_IFACE; if (!ipv6_addr_any(&fl6->saddr)) diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index f6b090df3930..afca2eb4dfa7 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -1319,7 +1319,7 @@ static void l2tp_tunnel_del_work(struct work_struct *work) tunnel = container_of(work, struct l2tp_tunnel, del_work); sk = l2tp_tunnel_sock_lookup(tunnel); if (!sk) - return; + goto out; sock = sk->sk_socket; @@ -1341,6 +1341,8 @@ static void l2tp_tunnel_del_work(struct work_struct *work) } l2tp_tunnel_sock_put(sk); +out: + l2tp_tunnel_dec_refcount(tunnel); } /* Create a socket for the tunnel, if one isn't set up by @@ -1636,8 +1638,13 @@ EXPORT_SYMBOL_GPL(l2tp_tunnel_create); */ int l2tp_tunnel_delete(struct l2tp_tunnel *tunnel) { + l2tp_tunnel_inc_refcount(tunnel); l2tp_tunnel_closeall(tunnel); - return (false == queue_work(l2tp_wq, &tunnel->del_work)); + if (false == queue_work(l2tp_wq, &tunnel->del_work)) { + l2tp_tunnel_dec_refcount(tunnel); + return 1; + } + return 0; } EXPORT_SYMBOL_GPL(l2tp_tunnel_delete); diff --git a/net/sctp/associola.c b/net/sctp/associola.c index 197c3f59ecbf..b00f1f9611d6 100644 --- a/net/sctp/associola.c +++ b/net/sctp/associola.c @@ -1208,20 +1208,22 @@ void sctp_assoc_update(struct sctp_association *asoc, * within this document. * * Our basic strategy is to round-robin transports in priorities - * according to sctp_state_prio_map[] e.g., if no such + * according to sctp_trans_score() e.g., if no such * transport with state SCTP_ACTIVE exists, round-robin through * SCTP_UNKNOWN, etc. You get the picture. */ -static const u8 sctp_trans_state_to_prio_map[] = { - [SCTP_ACTIVE] = 3, /* best case */ - [SCTP_UNKNOWN] = 2, - [SCTP_PF] = 1, - [SCTP_INACTIVE] = 0, /* worst case */ -}; - static u8 sctp_trans_score(const struct sctp_transport *trans) { - return sctp_trans_state_to_prio_map[trans->state]; + switch (trans->state) { + case SCTP_ACTIVE: + return 3; /* best case */ + case SCTP_UNKNOWN: + return 2; + case SCTP_PF: + return 1; + default: /* case SCTP_INACTIVE */ + return 0; /* worst case */ + } } static struct sctp_transport *sctp_trans_elect_tie(struct sctp_transport *trans1, diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index 35df1266bf07..6098d4c42fa9 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -244,12 +244,13 @@ void sctp_generate_t3_rtx_event(unsigned long peer) int error; struct sctp_transport *transport = (struct sctp_transport *) peer; struct sctp_association *asoc = transport->asoc; - struct net *net = sock_net(asoc->base.sk); + struct sock *sk = asoc->base.sk; + struct net *net = sock_net(sk); /* Check whether a task is in the sock. */ - bh_lock_sock(asoc->base.sk); - if (sock_owned_by_user(asoc->base.sk)) { + bh_lock_sock(sk); + if (sock_owned_by_user(sk)) { pr_debug("%s: sock is busy\n", __func__); /* Try again later. */ @@ -272,10 +273,10 @@ void sctp_generate_t3_rtx_event(unsigned long peer) transport, GFP_ATOMIC); if (error) - asoc->base.sk->sk_err = -error; + sk->sk_err = -error; out_unlock: - bh_unlock_sock(asoc->base.sk); + bh_unlock_sock(sk); sctp_transport_put(transport); } @@ -285,11 +286,12 @@ out_unlock: static void sctp_generate_timeout_event(struct sctp_association *asoc, sctp_event_timeout_t timeout_type) { - struct net *net = sock_net(asoc->base.sk); + struct sock *sk = asoc->base.sk; + struct net *net = sock_net(sk); int error = 0; - bh_lock_sock(asoc->base.sk); - if (sock_owned_by_user(asoc->base.sk)) { + bh_lock_sock(sk); + if (sock_owned_by_user(sk)) { pr_debug("%s: sock is busy: timer %d\n", __func__, timeout_type); @@ -312,10 +314,10 @@ static void sctp_generate_timeout_event(struct sctp_association *asoc, (void *)timeout_type, GFP_ATOMIC); if (error) - asoc->base.sk->sk_err = -error; + sk->sk_err = -error; out_unlock: - bh_unlock_sock(asoc->base.sk); + bh_unlock_sock(sk); sctp_association_put(asoc); } @@ -365,10 +367,11 @@ void sctp_generate_heartbeat_event(unsigned long data) int error = 0; struct sctp_transport *transport = (struct sctp_transport *) data; struct sctp_association *asoc = transport->asoc; - struct net *net = sock_net(asoc->base.sk); + struct sock *sk = asoc->base.sk; + struct net *net = sock_net(sk); - bh_lock_sock(asoc->base.sk); - if (sock_owned_by_user(asoc->base.sk)) { + bh_lock_sock(sk); + if (sock_owned_by_user(sk)) { pr_debug("%s: sock is busy\n", __func__); /* Try again later. */ @@ -388,11 +391,11 @@ void sctp_generate_heartbeat_event(unsigned long data) asoc->state, asoc->ep, asoc, transport, GFP_ATOMIC); - if (error) - asoc->base.sk->sk_err = -error; + if (error) + sk->sk_err = -error; out_unlock: - bh_unlock_sock(asoc->base.sk); + bh_unlock_sock(sk); sctp_transport_put(transport); } @@ -403,10 +406,11 @@ void sctp_generate_proto_unreach_event(unsigned long data) { struct sctp_transport *transport = (struct sctp_transport *) data; struct sctp_association *asoc = transport->asoc; - struct net *net = sock_net(asoc->base.sk); + struct sock *sk = asoc->base.sk; + struct net *net = sock_net(sk); - bh_lock_sock(asoc->base.sk); - if (sock_owned_by_user(asoc->base.sk)) { + bh_lock_sock(sk); + if (sock_owned_by_user(sk)) { pr_debug("%s: sock is busy\n", __func__); /* Try again later. */ @@ -427,7 +431,7 @@ void sctp_generate_proto_unreach_event(unsigned long data) asoc->state, asoc->ep, asoc, transport, GFP_ATOMIC); out_unlock: - bh_unlock_sock(asoc->base.sk); + bh_unlock_sock(sk); sctp_association_put(asoc); } diff --git a/net/sunrpc/xprtrdma/fmr_ops.c b/net/sunrpc/xprtrdma/fmr_ops.c index cb25c89da623..f1e8dafbd507 100644 --- a/net/sunrpc/xprtrdma/fmr_ops.c +++ b/net/sunrpc/xprtrdma/fmr_ops.c @@ -39,25 +39,6 @@ static int fmr_op_open(struct rpcrdma_ia *ia, struct rpcrdma_ep *ep, struct rpcrdma_create_data_internal *cdata) { - struct ib_device_attr *devattr = &ia->ri_devattr; - struct ib_mr *mr; - - /* Obtain an lkey to use for the regbufs, which are - * protected from remote access. - */ - if (devattr->device_cap_flags & IB_DEVICE_LOCAL_DMA_LKEY) { - ia->ri_dma_lkey = ia->ri_device->local_dma_lkey; - } else { - mr = ib_get_dma_mr(ia->ri_pd, IB_ACCESS_LOCAL_WRITE); - if (IS_ERR(mr)) { - pr_err("%s: ib_get_dma_mr for failed with %lX\n", - __func__, PTR_ERR(mr)); - return -ENOMEM; - } - ia->ri_dma_lkey = ia->ri_dma_mr->lkey; - ia->ri_dma_mr = mr; - } - return 0; } diff --git a/net/sunrpc/xprtrdma/frwr_ops.c b/net/sunrpc/xprtrdma/frwr_ops.c index d6653f5d0830..5318951b3b53 100644 --- a/net/sunrpc/xprtrdma/frwr_ops.c +++ b/net/sunrpc/xprtrdma/frwr_ops.c @@ -189,11 +189,6 @@ frwr_op_open(struct rpcrdma_ia *ia, struct rpcrdma_ep *ep, struct ib_device_attr *devattr = &ia->ri_devattr; int depth, delta; - /* Obtain an lkey to use for the regbufs, which are - * protected from remote access. - */ - ia->ri_dma_lkey = ia->ri_device->local_dma_lkey; - ia->ri_max_frmr_depth = min_t(unsigned int, RPCRDMA_MAX_DATA_SEGS, devattr->max_fast_reg_page_list_len); diff --git a/net/sunrpc/xprtrdma/physical_ops.c b/net/sunrpc/xprtrdma/physical_ops.c index 72cf8b15bbb4..617b76f22154 100644 --- a/net/sunrpc/xprtrdma/physical_ops.c +++ b/net/sunrpc/xprtrdma/physical_ops.c @@ -23,7 +23,6 @@ static int physical_op_open(struct rpcrdma_ia *ia, struct rpcrdma_ep *ep, struct rpcrdma_create_data_internal *cdata) { - struct ib_device_attr *devattr = &ia->ri_devattr; struct ib_mr *mr; /* Obtain an rkey to use for RPC data payloads. @@ -37,15 +36,8 @@ physical_op_open(struct rpcrdma_ia *ia, struct rpcrdma_ep *ep, __func__, PTR_ERR(mr)); return -ENOMEM; } - ia->ri_dma_mr = mr; - - /* Obtain an lkey to use for regbufs. - */ - if (devattr->device_cap_flags & IB_DEVICE_LOCAL_DMA_LKEY) - ia->ri_dma_lkey = ia->ri_device->local_dma_lkey; - else - ia->ri_dma_lkey = ia->ri_dma_mr->lkey; + ia->ri_dma_mr = mr; return 0; } diff --git a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c index cb5174284074..5f6ca47092b0 100644 --- a/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c +++ b/net/sunrpc/xprtrdma/svc_rdma_recvfrom.c @@ -136,7 +136,8 @@ int rdma_read_chunk_lcl(struct svcxprt_rdma *xprt, ctxt->direction = DMA_FROM_DEVICE; ctxt->read_hdr = head; pages_needed = min_t(int, pages_needed, xprt->sc_max_sge_rd); - read = min_t(int, pages_needed << PAGE_SHIFT, rs_length); + read = min_t(int, (pages_needed << PAGE_SHIFT) - *page_offset, + rs_length); for (pno = 0; pno < pages_needed; pno++) { int len = min_t(int, rs_length, PAGE_SIZE - pg_off); @@ -235,7 +236,8 @@ int rdma_read_chunk_frmr(struct svcxprt_rdma *xprt, ctxt->direction = DMA_FROM_DEVICE; ctxt->frmr = frmr; pages_needed = min_t(int, pages_needed, xprt->sc_frmr_pg_list_len); - read = min_t(int, pages_needed << PAGE_SHIFT, rs_length); + read = min_t(int, (pages_needed << PAGE_SHIFT) - *page_offset, + rs_length); frmr->kva = page_address(rqstp->rq_arg.pages[pg_no]); frmr->direction = DMA_FROM_DEVICE; diff --git a/net/sunrpc/xprtrdma/transport.c b/net/sunrpc/xprtrdma/transport.c index 64443eb754ad..41e452bc580c 100644 --- a/net/sunrpc/xprtrdma/transport.c +++ b/net/sunrpc/xprtrdma/transport.c @@ -270,8 +270,8 @@ xprt_rdma_destroy(struct rpc_xprt *xprt) xprt_clear_connected(xprt); - rpcrdma_buffer_destroy(&r_xprt->rx_buf); rpcrdma_ep_destroy(&r_xprt->rx_ep, &r_xprt->rx_ia); + rpcrdma_buffer_destroy(&r_xprt->rx_buf); rpcrdma_ia_close(&r_xprt->rx_ia); xprt_rdma_free_addresses(xprt); diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c index 682996779970..8a477e27bad7 100644 --- a/net/sunrpc/xprtrdma/verbs.c +++ b/net/sunrpc/xprtrdma/verbs.c @@ -755,19 +755,22 @@ rpcrdma_ep_destroy(struct rpcrdma_ep *ep, struct rpcrdma_ia *ia) cancel_delayed_work_sync(&ep->rep_connect_worker); - if (ia->ri_id->qp) { + if (ia->ri_id->qp) rpcrdma_ep_disconnect(ep, ia); + + rpcrdma_clean_cq(ep->rep_attr.recv_cq); + rpcrdma_clean_cq(ep->rep_attr.send_cq); + + if (ia->ri_id->qp) { rdma_destroy_qp(ia->ri_id); ia->ri_id->qp = NULL; } - rpcrdma_clean_cq(ep->rep_attr.recv_cq); rc = ib_destroy_cq(ep->rep_attr.recv_cq); if (rc) dprintk("RPC: %s: ib_destroy_cq returned %i\n", __func__, rc); - rpcrdma_clean_cq(ep->rep_attr.send_cq); rc = ib_destroy_cq(ep->rep_attr.send_cq); if (rc) dprintk("RPC: %s: ib_destroy_cq returned %i\n", @@ -1252,7 +1255,7 @@ rpcrdma_alloc_regbuf(struct rpcrdma_ia *ia, size_t size, gfp_t flags) goto out_free; iov->length = size; - iov->lkey = ia->ri_dma_lkey; + iov->lkey = ia->ri_pd->local_dma_lkey; rb->rg_size = size; rb->rg_owner = NULL; return rb; diff --git a/net/sunrpc/xprtrdma/xprt_rdma.h b/net/sunrpc/xprtrdma/xprt_rdma.h index 02512221b8bc..c09414e6f91b 100644 --- a/net/sunrpc/xprtrdma/xprt_rdma.h +++ b/net/sunrpc/xprtrdma/xprt_rdma.h @@ -65,7 +65,6 @@ struct rpcrdma_ia { struct rdma_cm_id *ri_id; struct ib_pd *ri_pd; struct ib_mr *ri_dma_mr; - u32 ri_dma_lkey; struct completion ri_done; int ri_async_rc; unsigned int ri_max_frmr_depth; diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 03ee4d359f6a..ef31b40ad550 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -2179,8 +2179,21 @@ unlock: if (UNIXCB(skb).fp) scm.fp = scm_fp_dup(UNIXCB(skb).fp); - sk_peek_offset_fwd(sk, chunk); + if (skip) { + sk_peek_offset_fwd(sk, chunk); + skip -= chunk; + } + if (UNIXCB(skb).fp) + break; + + last = skb; + last_len = skb->len; + unix_state_lock(sk); + skb = skb_peek_next(skb, &sk->sk_receive_queue); + if (skb) + goto again; + unix_state_unlock(sk); break; } } while (size); diff --git a/samples/kprobes/jprobe_example.c b/samples/kprobes/jprobe_example.c index 9119ac6a8270..c285a3b8a9f1 100644 --- a/samples/kprobes/jprobe_example.c +++ b/samples/kprobes/jprobe_example.c @@ -1,13 +1,13 @@ /* * Here's a sample kernel module showing the use of jprobes to dump - * the arguments of do_fork(). + * the arguments of _do_fork(). * * For more information on theory of operation of jprobes, see * Documentation/kprobes.txt * * Build and insert the kernel module as done in the kprobe example. * You will see the trace data in /var/log/messages and on the - * console whenever do_fork() is invoked to create a new process. + * console whenever _do_fork() is invoked to create a new process. * (Some messages may be suppressed if syslogd is configured to * eliminate duplicate messages.) */ @@ -17,13 +17,13 @@ #include <linux/kprobes.h> /* - * Jumper probe for do_fork. + * Jumper probe for _do_fork. * Mirror principle enables access to arguments of the probed routine * from the probe handler. */ -/* Proxy routine having the same arguments as actual do_fork() routine */ -static long jdo_fork(unsigned long clone_flags, unsigned long stack_start, +/* Proxy routine having the same arguments as actual _do_fork() routine */ +static long j_do_fork(unsigned long clone_flags, unsigned long stack_start, unsigned long stack_size, int __user *parent_tidptr, int __user *child_tidptr) { @@ -36,9 +36,9 @@ static long jdo_fork(unsigned long clone_flags, unsigned long stack_start, } static struct jprobe my_jprobe = { - .entry = jdo_fork, + .entry = j_do_fork, .kp = { - .symbol_name = "do_fork", + .symbol_name = "_do_fork", }, }; diff --git a/samples/kprobes/kprobe_example.c b/samples/kprobes/kprobe_example.c index 366db1a9fb65..727eb21c9c56 100644 --- a/samples/kprobes/kprobe_example.c +++ b/samples/kprobes/kprobe_example.c @@ -1,13 +1,13 @@ /* * NOTE: This example is works on x86 and powerpc. * Here's a sample kernel module showing the use of kprobes to dump a - * stack trace and selected registers when do_fork() is called. + * stack trace and selected registers when _do_fork() is called. * * For more information on theory of operation of kprobes, see * Documentation/kprobes.txt * * You will see the trace data in /var/log/messages and on the console - * whenever do_fork() is invoked to create a new process. + * whenever _do_fork() is invoked to create a new process. */ #include <linux/kernel.h> @@ -16,7 +16,7 @@ /* For each probe you need to allocate a kprobe structure */ static struct kprobe kp = { - .symbol_name = "do_fork", + .symbol_name = "_do_fork", }; /* kprobe pre_handler: called just before the probed instruction is executed */ diff --git a/samples/kprobes/kretprobe_example.c b/samples/kprobes/kretprobe_example.c index 1041b6731598..ebb1d1aed547 100644 --- a/samples/kprobes/kretprobe_example.c +++ b/samples/kprobes/kretprobe_example.c @@ -7,7 +7,7 @@ * * usage: insmod kretprobe_example.ko func=<func_name> * - * If no func_name is specified, do_fork is instrumented + * If no func_name is specified, _do_fork is instrumented * * For more information on theory of operation of kretprobes, see * Documentation/kprobes.txt @@ -25,7 +25,7 @@ #include <linux/limits.h> #include <linux/sched.h> -static char func_name[NAME_MAX] = "do_fork"; +static char func_name[NAME_MAX] = "_do_fork"; module_param_string(func, func_name, NAME_MAX, S_IRUGO); MODULE_PARM_DESC(func, "Function to kretprobe; this module will report the" " function's execution time"); diff --git a/scripts/extract-cert.c b/scripts/extract-cert.c index 6ce5945a0b89..b071bf476fea 100644 --- a/scripts/extract-cert.c +++ b/scripts/extract-cert.c @@ -17,13 +17,9 @@ #include <stdint.h> #include <stdbool.h> #include <string.h> -#include <getopt.h> #include <err.h> -#include <arpa/inet.h> #include <openssl/bio.h> -#include <openssl/evp.h> #include <openssl/pem.h> -#include <openssl/pkcs7.h> #include <openssl/err.h> #include <openssl/engine.h> diff --git a/scripts/mod/devicetable-offsets.c b/scripts/mod/devicetable-offsets.c index e70fcd12eeeb..5a6edacc85d9 100644 --- a/scripts/mod/devicetable-offsets.c +++ b/scripts/mod/devicetable-offsets.c @@ -185,6 +185,7 @@ int main(void) DEVID(mei_cl_device_id); DEVID_FIELD(mei_cl_device_id, name); DEVID_FIELD(mei_cl_device_id, uuid); + DEVID_FIELD(mei_cl_device_id, version); DEVID(rio_device_id); DEVID_FIELD(rio_device_id, did); diff --git a/scripts/mod/file2alias.c b/scripts/mod/file2alias.c index 5f2088209132..9bc2cfe0ee37 100644 --- a/scripts/mod/file2alias.c +++ b/scripts/mod/file2alias.c @@ -137,10 +137,12 @@ static inline void add_wildcard(char *str) static inline void add_uuid(char *str, uuid_le uuid) { int len = strlen(str); - int i; - for (i = 0; i < 16; i++) - sprintf(str + len + (i << 1), "%02x", uuid.b[i]); + sprintf(str + len, "%02x%02x%02x%02x-%02x%02x-%02x%02x-%02x%02x-%02x%02x%02x%02x%02x%02x", + uuid.b[3], uuid.b[2], uuid.b[1], uuid.b[0], + uuid.b[5], uuid.b[4], uuid.b[7], uuid.b[6], + uuid.b[8], uuid.b[9], uuid.b[10], uuid.b[11], + uuid.b[12], uuid.b[13], uuid.b[14], uuid.b[15]); } /** @@ -1200,16 +1202,18 @@ static int do_cpu_entry(const char *filename, void *symval, char *alias) } ADD_TO_DEVTABLE("cpu", cpu_feature, do_cpu_entry); -/* Looks like: mei:S:uuid */ +/* Looks like: mei:S:uuid:N:* */ static int do_mei_entry(const char *filename, void *symval, char *alias) { DEF_FIELD_ADDR(symval, mei_cl_device_id, name); DEF_FIELD_ADDR(symval, mei_cl_device_id, uuid); + DEF_FIELD(symval, mei_cl_device_id, version); sprintf(alias, MEI_CL_MODULE_PREFIX); sprintf(alias + strlen(alias), "%s:", (*name)[0] ? *name : "*"); add_uuid(alias, *uuid); + ADD(alias, ":", version != MEI_CL_VERSION_ANY, version); strcat(alias, ":*"); diff --git a/scripts/sign-file.c b/scripts/sign-file.c index c3899ca4811c..250a7a645033 100755 --- a/scripts/sign-file.c +++ b/scripts/sign-file.c @@ -20,13 +20,34 @@ #include <getopt.h> #include <err.h> #include <arpa/inet.h> +#include <openssl/opensslv.h> #include <openssl/bio.h> #include <openssl/evp.h> #include <openssl/pem.h> -#include <openssl/cms.h> #include <openssl/err.h> #include <openssl/engine.h> +/* + * Use CMS if we have openssl-1.0.0 or newer available - otherwise we have to + * assume that it's not available and its header file is missing and that we + * should use PKCS#7 instead. Switching to the older PKCS#7 format restricts + * the options we have on specifying the X.509 certificate we want. + * + * Further, older versions of OpenSSL don't support manually adding signers to + * the PKCS#7 message so have to accept that we get a certificate included in + * the signature message. Nor do such older versions of OpenSSL support + * signing with anything other than SHA1 - so we're stuck with that if such is + * the case. + */ +#if OPENSSL_VERSION_NUMBER < 0x10000000L +#define USE_PKCS7 +#endif +#ifndef USE_PKCS7 +#include <openssl/cms.h> +#else +#include <openssl/pkcs7.h> +#endif + struct module_signature { uint8_t algo; /* Public-key crypto algorithm [0] */ uint8_t hash; /* Digest algorithm [0] */ @@ -110,30 +131,42 @@ int main(int argc, char **argv) struct module_signature sig_info = { .id_type = PKEY_ID_PKCS7 }; char *hash_algo = NULL; char *private_key_name, *x509_name, *module_name, *dest_name; - bool save_cms = false, replace_orig; + bool save_sig = false, replace_orig; bool sign_only = false; unsigned char buf[4096]; - unsigned long module_size, cms_size; - unsigned int use_keyid = 0, use_signed_attrs = CMS_NOATTR; + unsigned long module_size, sig_size; + unsigned int use_signed_attrs; const EVP_MD *digest_algo; EVP_PKEY *private_key; +#ifndef USE_PKCS7 CMS_ContentInfo *cms; + unsigned int use_keyid = 0; +#else + PKCS7 *pkcs7; +#endif X509 *x509; BIO *b, *bd = NULL, *bm; int opt, n; - OpenSSL_add_all_algorithms(); ERR_load_crypto_strings(); ERR_clear_error(); key_pass = getenv("KBUILD_SIGN_PIN"); +#ifndef USE_PKCS7 + use_signed_attrs = CMS_NOATTR; +#else + use_signed_attrs = PKCS7_NOATTR; +#endif + do { opt = getopt(argc, argv, "dpk"); switch (opt) { - case 'p': save_cms = true; break; - case 'd': sign_only = true; save_cms = true; break; + case 'p': save_sig = true; break; + case 'd': sign_only = true; save_sig = true; break; +#ifndef USE_PKCS7 case 'k': use_keyid = CMS_USE_KEYID; break; +#endif case -1: break; default: format(); } @@ -157,6 +190,14 @@ int main(int argc, char **argv) replace_orig = true; } +#ifdef USE_PKCS7 + if (strcmp(hash_algo, "sha1") != 0) { + fprintf(stderr, "sign-file: %s only supports SHA1 signing\n", + OPENSSL_VERSION_TEXT); + exit(3); + } +#endif + /* Read the private key and the X.509 cert the PKCS#7 message * will point to. */ @@ -213,7 +254,8 @@ int main(int argc, char **argv) bm = BIO_new_file(module_name, "rb"); ERR(!bm, "%s", module_name); - /* Load the CMS message from the digest buffer. */ +#ifndef USE_PKCS7 + /* Load the signature message from the digest buffer. */ cms = CMS_sign(NULL, NULL, NULL, NULL, CMS_NOCERTS | CMS_PARTIAL | CMS_BINARY | CMS_DETACHED | CMS_STREAM); ERR(!cms, "CMS_sign"); @@ -221,17 +263,31 @@ int main(int argc, char **argv) ERR(!CMS_add1_signer(cms, x509, private_key, digest_algo, CMS_NOCERTS | CMS_BINARY | CMS_NOSMIMECAP | use_keyid | use_signed_attrs), - "CMS_sign_add_signer"); + "CMS_add1_signer"); ERR(CMS_final(cms, bm, NULL, CMS_NOCERTS | CMS_BINARY) < 0, "CMS_final"); - if (save_cms) { - char *cms_name; +#else + pkcs7 = PKCS7_sign(x509, private_key, NULL, bm, + PKCS7_NOCERTS | PKCS7_BINARY | + PKCS7_DETACHED | use_signed_attrs); + ERR(!pkcs7, "PKCS7_sign"); +#endif - ERR(asprintf(&cms_name, "%s.p7s", module_name) < 0, "asprintf"); - b = BIO_new_file(cms_name, "wb"); - ERR(!b, "%s", cms_name); - ERR(i2d_CMS_bio_stream(b, cms, NULL, 0) < 0, "%s", cms_name); + if (save_sig) { + char *sig_file_name; + + ERR(asprintf(&sig_file_name, "%s.p7s", module_name) < 0, + "asprintf"); + b = BIO_new_file(sig_file_name, "wb"); + ERR(!b, "%s", sig_file_name); +#ifndef USE_PKCS7 + ERR(i2d_CMS_bio_stream(b, cms, NULL, 0) < 0, + "%s", sig_file_name); +#else + ERR(i2d_PKCS7_bio(b, pkcs7) < 0, + "%s", sig_file_name); +#endif BIO_free(b); } @@ -247,9 +303,13 @@ int main(int argc, char **argv) ERR(n < 0, "%s", module_name); module_size = BIO_number_written(bd); +#ifndef USE_PKCS7 ERR(i2d_CMS_bio_stream(bd, cms, NULL, 0) < 0, "%s", dest_name); - cms_size = BIO_number_written(bd) - module_size; - sig_info.sig_len = htonl(cms_size); +#else + ERR(i2d_PKCS7_bio(bd, pkcs7) < 0, "%s", dest_name); +#endif + sig_size = BIO_number_written(bd) - module_size; + sig_info.sig_len = htonl(sig_size); ERR(BIO_write(bd, &sig_info, sizeof(sig_info)) < 0, "%s", dest_name); ERR(BIO_write(bd, magic_number, sizeof(magic_number) - 1) < 0, "%s", dest_name); diff --git a/security/keys/gc.c b/security/keys/gc.c index c7952375ac53..39eac1fd5706 100644 --- a/security/keys/gc.c +++ b/security/keys/gc.c @@ -134,6 +134,10 @@ static noinline void key_gc_unused_keys(struct list_head *keys) kdebug("- %u", key->serial); key_check(key); + /* Throw away the key data */ + if (key->type->destroy) + key->type->destroy(key); + security_key_free(key); /* deal with the user's key tracking and quota */ @@ -148,10 +152,6 @@ static noinline void key_gc_unused_keys(struct list_head *keys) if (test_bit(KEY_FLAG_INSTANTIATED, &key->flags)) atomic_dec(&key->user->nikeys); - /* now throw away the key memory */ - if (key->type->destroy) - key->type->destroy(key); - key_user_put(key->user); kfree(key->description); diff --git a/sound/pci/hda/patch_cirrus.c b/sound/pci/hda/patch_cirrus.c index 584a0343ab0c..85813de26da8 100644 --- a/sound/pci/hda/patch_cirrus.c +++ b/sound/pci/hda/patch_cirrus.c @@ -633,6 +633,7 @@ static const struct snd_pci_quirk cs4208_mac_fixup_tbl[] = { SND_PCI_QUIRK(0x106b, 0x5e00, "MacBookPro 11,2", CS4208_MBP11), SND_PCI_QUIRK(0x106b, 0x7100, "MacBookAir 6,1", CS4208_MBA6), SND_PCI_QUIRK(0x106b, 0x7200, "MacBookAir 6,2", CS4208_MBA6), + SND_PCI_QUIRK(0x106b, 0x7b00, "MacBookPro 12,1", CS4208_MBP11), {} /* terminator */ }; diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index afec6dc9f91f..16b8dcba5c12 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -5306,6 +5306,7 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x17aa, 0x2212, "Thinkpad T440", ALC292_FIXUP_TPT440_DOCK), SND_PCI_QUIRK(0x17aa, 0x2214, "Thinkpad X240", ALC292_FIXUP_TPT440_DOCK), SND_PCI_QUIRK(0x17aa, 0x2215, "Thinkpad", ALC269_FIXUP_LIMIT_INT_MIC_BOOST), + SND_PCI_QUIRK(0x17aa, 0x2223, "ThinkPad T550", ALC292_FIXUP_TPT440_DOCK), SND_PCI_QUIRK(0x17aa, 0x2226, "ThinkPad X250", ALC292_FIXUP_TPT440_DOCK), SND_PCI_QUIRK(0x17aa, 0x3977, "IdeaPad S210", ALC283_FIXUP_INT_MIC), SND_PCI_QUIRK(0x17aa, 0x3978, "IdeaPad Y410P", ALC269_FIXUP_NO_SHUTUP), diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c index 9d947aef2c8b..def5cc8dff02 100644 --- a/sound/pci/hda/patch_sigmatel.c +++ b/sound/pci/hda/patch_sigmatel.c @@ -4520,7 +4520,11 @@ static int patch_stac92hd73xx(struct hda_codec *codec) return err; spec = codec->spec; - codec->power_save_node = 1; + /* enable power_save_node only for new 92HD89xx chips, as it causes + * click noises on old 92HD73xx chips. + */ + if ((codec->core.vendor_id & 0xfffffff0) != 0x111d7670) + codec->power_save_node = 1; spec->linear_tone_beep = 0; spec->gen.mixer_nid = 0x1d; spec->have_spdif_mux = 1; diff --git a/sound/soc/au1x/db1200.c b/sound/soc/au1x/db1200.c index 58c3164802b8..8c907ebea189 100644 --- a/sound/soc/au1x/db1200.c +++ b/sound/soc/au1x/db1200.c @@ -129,6 +129,8 @@ static struct snd_soc_dai_link db1300_i2s_dai = { .cpu_dai_name = "au1xpsc_i2s.2", .platform_name = "au1xpsc-pcm.2", .codec_name = "wm8731.0-001b", + .dai_fmt = SND_SOC_DAIFMT_LEFT_J | SND_SOC_DAIFMT_NB_NF | + SND_SOC_DAIFMT_CBM_CFM, .ops = &db1200_i2s_wm8731_ops, }; @@ -146,6 +148,8 @@ static struct snd_soc_dai_link db1550_i2s_dai = { .cpu_dai_name = "au1xpsc_i2s.3", .platform_name = "au1xpsc-pcm.3", .codec_name = "wm8731.0-001b", + .dai_fmt = SND_SOC_DAIFMT_LEFT_J | SND_SOC_DAIFMT_NB_NF | + SND_SOC_DAIFMT_CBM_CFM, .ops = &db1200_i2s_wm8731_ops, }; diff --git a/sound/soc/codecs/rt5645.c b/sound/soc/codecs/rt5645.c index 268a28bd1df4..5c101af0ac63 100644 --- a/sound/soc/codecs/rt5645.c +++ b/sound/soc/codecs/rt5645.c @@ -519,11 +519,11 @@ static const struct snd_kcontrol_new rt5645_snd_controls[] = { RT5645_L_VOL_SFT + 1, RT5645_R_VOL_SFT + 1, 63, 0, adc_vol_tlv), /* ADC Boost Volume Control */ - SOC_DOUBLE_TLV("STO1 ADC Boost Gain", RT5645_ADC_BST_VOL1, + SOC_DOUBLE_TLV("ADC Boost Capture Volume", RT5645_ADC_BST_VOL1, RT5645_STO1_ADC_L_BST_SFT, RT5645_STO1_ADC_R_BST_SFT, 3, 0, adc_bst_tlv), - SOC_DOUBLE_TLV("STO2 ADC Boost Gain", RT5645_ADC_BST_VOL1, - RT5645_STO2_ADC_L_BST_SFT, RT5645_STO2_ADC_R_BST_SFT, 3, 0, + SOC_DOUBLE_TLV("Mono ADC Boost Capture Volume", RT5645_ADC_BST_VOL2, + RT5645_MONO_ADC_L_BST_SFT, RT5645_MONO_ADC_R_BST_SFT, 3, 0, adc_bst_tlv), /* I2S2 function select */ diff --git a/sound/soc/codecs/rt5645.h b/sound/soc/codecs/rt5645.h index 0e4cfc6ac649..8c964cfb120d 100644 --- a/sound/soc/codecs/rt5645.h +++ b/sound/soc/codecs/rt5645.h @@ -39,8 +39,8 @@ #define RT5645_STO1_ADC_DIG_VOL 0x1c #define RT5645_MONO_ADC_DIG_VOL 0x1d #define RT5645_ADC_BST_VOL1 0x1e -/* Mixer - D-D */ #define RT5645_ADC_BST_VOL2 0x20 +/* Mixer - D-D */ #define RT5645_STO1_ADC_MIXER 0x27 #define RT5645_MONO_ADC_MIXER 0x28 #define RT5645_AD_DA_MIXER 0x29 @@ -315,12 +315,14 @@ #define RT5645_STO1_ADC_R_BST_SFT 12 #define RT5645_STO1_ADC_COMP_MASK (0x3 << 10) #define RT5645_STO1_ADC_COMP_SFT 10 -#define RT5645_STO2_ADC_L_BST_MASK (0x3 << 8) -#define RT5645_STO2_ADC_L_BST_SFT 8 -#define RT5645_STO2_ADC_R_BST_MASK (0x3 << 6) -#define RT5645_STO2_ADC_R_BST_SFT 6 -#define RT5645_STO2_ADC_COMP_MASK (0x3 << 4) -#define RT5645_STO2_ADC_COMP_SFT 4 + +/* ADC Boost Volume Control (0x20) */ +#define RT5645_MONO_ADC_L_BST_MASK (0x3 << 14) +#define RT5645_MONO_ADC_L_BST_SFT 14 +#define RT5645_MONO_ADC_R_BST_MASK (0x3 << 12) +#define RT5645_MONO_ADC_R_BST_SFT 12 +#define RT5645_MONO_ADC_COMP_MASK (0x3 << 10) +#define RT5645_MONO_ADC_COMP_SFT 10 /* Stereo2 ADC Mixer Control (0x26) */ #define RT5645_STO2_ADC_SRC_MASK (0x1 << 15) diff --git a/sound/soc/codecs/sgtl5000.c b/sound/soc/codecs/sgtl5000.c index bfda25ef0dd4..f540f82b1f27 100644 --- a/sound/soc/codecs/sgtl5000.c +++ b/sound/soc/codecs/sgtl5000.c @@ -1376,8 +1376,8 @@ static int sgtl5000_probe(struct snd_soc_codec *codec) sgtl5000->micbias_resistor << SGTL5000_BIAS_R_SHIFT); snd_soc_update_bits(codec, SGTL5000_CHIP_MIC_CTRL, - SGTL5000_BIAS_R_MASK, - sgtl5000->micbias_voltage << SGTL5000_BIAS_R_SHIFT); + SGTL5000_BIAS_VOLT_MASK, + sgtl5000->micbias_voltage << SGTL5000_BIAS_VOLT_SHIFT); /* * disable DAP * TODO: @@ -1549,7 +1549,7 @@ static int sgtl5000_i2c_probe(struct i2c_client *client, else { sgtl5000->micbias_voltage = 0; dev_err(&client->dev, - "Unsuitable MicBias resistor\n"); + "Unsuitable MicBias voltage\n"); } } else { sgtl5000->micbias_voltage = 0; diff --git a/sound/soc/codecs/tas2552.c b/sound/soc/codecs/tas2552.c index e3a0bca28bcf..cc1d3981fa4b 100644 --- a/sound/soc/codecs/tas2552.c +++ b/sound/soc/codecs/tas2552.c @@ -549,7 +549,7 @@ static struct snd_soc_dai_driver tas2552_dai[] = { /* * DAC digital volumes. From -7 to 24 dB in 1 dB steps */ -static DECLARE_TLV_DB_SCALE(dac_tlv, -7, 100, 0); +static DECLARE_TLV_DB_SCALE(dac_tlv, -700, 100, 0); static const char * const tas2552_din_source_select[] = { "Muted", diff --git a/sound/soc/codecs/tlv320aic3x.c b/sound/soc/codecs/tlv320aic3x.c index 1a82b19b2644..8739126a1f6f 100644 --- a/sound/soc/codecs/tlv320aic3x.c +++ b/sound/soc/codecs/tlv320aic3x.c @@ -1509,14 +1509,17 @@ static int aic3x_init(struct snd_soc_codec *codec) snd_soc_write(codec, PGAL_2_LLOPM_VOL, DEFAULT_VOL); snd_soc_write(codec, PGAR_2_RLOPM_VOL, DEFAULT_VOL); - /* Line2 to HP Bypass default volume, disconnect from Output Mixer */ - snd_soc_write(codec, LINE2L_2_HPLOUT_VOL, DEFAULT_VOL); - snd_soc_write(codec, LINE2R_2_HPROUT_VOL, DEFAULT_VOL); - snd_soc_write(codec, LINE2L_2_HPLCOM_VOL, DEFAULT_VOL); - snd_soc_write(codec, LINE2R_2_HPRCOM_VOL, DEFAULT_VOL); - /* Line2 Line Out default volume, disconnect from Output Mixer */ - snd_soc_write(codec, LINE2L_2_LLOPM_VOL, DEFAULT_VOL); - snd_soc_write(codec, LINE2R_2_RLOPM_VOL, DEFAULT_VOL); + /* On tlv320aic3104, these registers are reserved and must not be written */ + if (aic3x->model != AIC3X_MODEL_3104) { + /* Line2 to HP Bypass default volume, disconnect from Output Mixer */ + snd_soc_write(codec, LINE2L_2_HPLOUT_VOL, DEFAULT_VOL); + snd_soc_write(codec, LINE2R_2_HPROUT_VOL, DEFAULT_VOL); + snd_soc_write(codec, LINE2L_2_HPLCOM_VOL, DEFAULT_VOL); + snd_soc_write(codec, LINE2R_2_HPRCOM_VOL, DEFAULT_VOL); + /* Line2 Line Out default volume, disconnect from Output Mixer */ + snd_soc_write(codec, LINE2L_2_LLOPM_VOL, DEFAULT_VOL); + snd_soc_write(codec, LINE2R_2_RLOPM_VOL, DEFAULT_VOL); + } switch (aic3x->model) { case AIC3X_MODEL_3X: diff --git a/sound/soc/codecs/wm8962.c b/sound/soc/codecs/wm8962.c index 293e47a6ff59..2fbc6ef8cbdb 100644 --- a/sound/soc/codecs/wm8962.c +++ b/sound/soc/codecs/wm8962.c @@ -3760,7 +3760,7 @@ static int wm8962_i2c_probe(struct i2c_client *i2c, ret = snd_soc_register_codec(&i2c->dev, &soc_codec_dev_wm8962, &wm8962_dai, 1); if (ret < 0) - goto err_enable; + goto err_pm_runtime; regcache_cache_only(wm8962->regmap, true); @@ -3769,6 +3769,8 @@ static int wm8962_i2c_probe(struct i2c_client *i2c, return 0; +err_pm_runtime: + pm_runtime_disable(&i2c->dev); err_enable: regulator_bulk_disable(ARRAY_SIZE(wm8962->supplies), wm8962->supplies); err: @@ -3778,6 +3780,7 @@ err: static int wm8962_i2c_remove(struct i2c_client *client) { snd_soc_unregister_codec(&client->dev); + pm_runtime_disable(&client->dev); return 0; } diff --git a/sound/soc/dwc/designware_i2s.c b/sound/soc/dwc/designware_i2s.c index a3e97b46b64e..ba34252b7bba 100644 --- a/sound/soc/dwc/designware_i2s.c +++ b/sound/soc/dwc/designware_i2s.c @@ -131,23 +131,32 @@ static inline void i2s_clear_irqs(struct dw_i2s_dev *dev, u32 stream) if (stream == SNDRV_PCM_STREAM_PLAYBACK) { for (i = 0; i < 4; i++) - i2s_write_reg(dev->i2s_base, TOR(i), 0); + i2s_read_reg(dev->i2s_base, TOR(i)); } else { for (i = 0; i < 4; i++) - i2s_write_reg(dev->i2s_base, ROR(i), 0); + i2s_read_reg(dev->i2s_base, ROR(i)); } } static void i2s_start(struct dw_i2s_dev *dev, struct snd_pcm_substream *substream) { - + u32 i, irq; i2s_write_reg(dev->i2s_base, IER, 1); - if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) + if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) { + for (i = 0; i < 4; i++) { + irq = i2s_read_reg(dev->i2s_base, IMR(i)); + i2s_write_reg(dev->i2s_base, IMR(i), irq & ~0x30); + } i2s_write_reg(dev->i2s_base, ITER, 1); - else + } else { + for (i = 0; i < 4; i++) { + irq = i2s_read_reg(dev->i2s_base, IMR(i)); + i2s_write_reg(dev->i2s_base, IMR(i), irq & ~0x03); + } i2s_write_reg(dev->i2s_base, IRER, 1); + } i2s_write_reg(dev->i2s_base, CER, 1); } diff --git a/sound/soc/fsl/imx-ssi.c b/sound/soc/fsl/imx-ssi.c index 48b2d24dd1f0..b95132e2f9dc 100644 --- a/sound/soc/fsl/imx-ssi.c +++ b/sound/soc/fsl/imx-ssi.c @@ -95,7 +95,8 @@ static int imx_ssi_set_dai_fmt(struct snd_soc_dai *cpu_dai, unsigned int fmt) switch (fmt & SND_SOC_DAIFMT_FORMAT_MASK) { case SND_SOC_DAIFMT_I2S: /* data on rising edge of bclk, frame low 1clk before data */ - strcr |= SSI_STCR_TFSI | SSI_STCR_TEFS | SSI_STCR_TXBIT0; + strcr |= SSI_STCR_TXBIT0 | SSI_STCR_TSCKP | SSI_STCR_TFSI | + SSI_STCR_TEFS; scr |= SSI_SCR_NET; if (ssi->flags & IMX_SSI_USE_I2S_SLAVE) { scr &= ~SSI_I2S_MODE_MASK; @@ -104,33 +105,31 @@ static int imx_ssi_set_dai_fmt(struct snd_soc_dai *cpu_dai, unsigned int fmt) break; case SND_SOC_DAIFMT_LEFT_J: /* data on rising edge of bclk, frame high with data */ - strcr |= SSI_STCR_TXBIT0; + strcr |= SSI_STCR_TXBIT0 | SSI_STCR_TSCKP; break; case SND_SOC_DAIFMT_DSP_B: /* data on rising edge of bclk, frame high with data */ - strcr |= SSI_STCR_TFSL | SSI_STCR_TXBIT0; + strcr |= SSI_STCR_TXBIT0 | SSI_STCR_TSCKP | SSI_STCR_TFSL; break; case SND_SOC_DAIFMT_DSP_A: /* data on rising edge of bclk, frame high 1clk before data */ - strcr |= SSI_STCR_TFSL | SSI_STCR_TXBIT0 | SSI_STCR_TEFS; + strcr |= SSI_STCR_TXBIT0 | SSI_STCR_TSCKP | SSI_STCR_TFSL | + SSI_STCR_TEFS; break; } /* DAI clock inversion */ switch (fmt & SND_SOC_DAIFMT_INV_MASK) { case SND_SOC_DAIFMT_IB_IF: - strcr |= SSI_STCR_TFSI; - strcr &= ~SSI_STCR_TSCKP; + strcr ^= SSI_STCR_TSCKP | SSI_STCR_TFSI; break; case SND_SOC_DAIFMT_IB_NF: - strcr &= ~(SSI_STCR_TSCKP | SSI_STCR_TFSI); + strcr ^= SSI_STCR_TSCKP; break; case SND_SOC_DAIFMT_NB_IF: - strcr |= SSI_STCR_TFSI | SSI_STCR_TSCKP; + strcr ^= SSI_STCR_TFSI; break; case SND_SOC_DAIFMT_NB_NF: - strcr &= ~SSI_STCR_TFSI; - strcr |= SSI_STCR_TSCKP; break; } diff --git a/sound/synth/emux/emux_oss.c b/sound/synth/emux/emux_oss.c index 82e350e9501c..ac75816ada7c 100644 --- a/sound/synth/emux/emux_oss.c +++ b/sound/synth/emux/emux_oss.c @@ -69,7 +69,8 @@ snd_emux_init_seq_oss(struct snd_emux *emu) struct snd_seq_oss_reg *arg; struct snd_seq_device *dev; - if (snd_seq_device_new(emu->card, 0, SNDRV_SEQ_DEV_ID_OSS, + /* using device#1 here for avoiding conflicts with OPL3 */ + if (snd_seq_device_new(emu->card, 1, SNDRV_SEQ_DEV_ID_OSS, sizeof(struct snd_seq_oss_reg), &dev) < 0) return; diff --git a/tools/build/Makefile.feature b/tools/build/Makefile.feature index 2975632d51e2..c8fe6d177119 100644 --- a/tools/build/Makefile.feature +++ b/tools/build/Makefile.feature @@ -41,6 +41,7 @@ FEATURE_TESTS ?= \ libelf-getphdrnum \ libelf-mmap \ libnuma \ + numa_num_possible_cpus \ libperl \ libpython \ libpython-version \ @@ -51,7 +52,8 @@ FEATURE_TESTS ?= \ timerfd \ libdw-dwarf-unwind \ zlib \ - lzma + lzma \ + get_cpuid FEATURE_DISPLAY ?= \ dwarf \ @@ -61,13 +63,15 @@ FEATURE_DISPLAY ?= \ libbfd \ libelf \ libnuma \ + numa_num_possible_cpus \ libperl \ libpython \ libslang \ libunwind \ libdw-dwarf-unwind \ zlib \ - lzma + lzma \ + get_cpuid # Set FEATURE_CHECK_(C|LD)FLAGS-all for all FEATURE_TESTS features. # If in the future we need per-feature checks/flags for features not diff --git a/tools/build/feature/Makefile b/tools/build/feature/Makefile index 74ca42093d70..e43a2971bf56 100644 --- a/tools/build/feature/Makefile +++ b/tools/build/feature/Makefile @@ -19,6 +19,7 @@ FILES= \ test-libelf-getphdrnum.bin \ test-libelf-mmap.bin \ test-libnuma.bin \ + test-numa_num_possible_cpus.bin \ test-libperl.bin \ test-libpython.bin \ test-libpython-version.bin \ @@ -34,7 +35,8 @@ FILES= \ test-compile-x32.bin \ test-zlib.bin \ test-lzma.bin \ - test-bpf.bin + test-bpf.bin \ + test-get_cpuid.bin CC := $(CROSS_COMPILE)gcc -MD PKG_CONFIG := $(CROSS_COMPILE)pkg-config @@ -87,6 +89,9 @@ test-libelf-getphdrnum.bin: test-libnuma.bin: $(BUILD) -lnuma +test-numa_num_possible_cpus.bin: + $(BUILD) -lnuma + test-libunwind.bin: $(BUILD) -lelf @@ -162,6 +167,9 @@ test-zlib.bin: test-lzma.bin: $(BUILD) -llzma +test-get_cpuid.bin: + $(BUILD) + test-bpf.bin: $(BUILD) diff --git a/tools/build/feature/test-all.c b/tools/build/feature/test-all.c index 84689a67814a..33cf6f20bd4e 100644 --- a/tools/build/feature/test-all.c +++ b/tools/build/feature/test-all.c @@ -77,6 +77,10 @@ # include "test-libnuma.c" #undef main +#define main main_test_numa_num_possible_cpus +# include "test-numa_num_possible_cpus.c" +#undef main + #define main main_test_timerfd # include "test-timerfd.c" #undef main @@ -117,6 +121,10 @@ # include "test-lzma.c" #undef main +#define main main_test_get_cpuid +# include "test-get_cpuid.c" +#undef main + int main(int argc, char *argv[]) { main_test_libpython(); @@ -136,6 +144,7 @@ int main(int argc, char *argv[]) main_test_libbfd(); main_test_backtrace(); main_test_libnuma(); + main_test_numa_num_possible_cpus(); main_test_timerfd(); main_test_stackprotector_all(); main_test_libdw_dwarf_unwind(); @@ -143,6 +152,7 @@ int main(int argc, char *argv[]) main_test_zlib(); main_test_pthread_attr_setaffinity_np(); main_test_lzma(); + main_test_get_cpuid(); return 0; } diff --git a/tools/build/feature/test-get_cpuid.c b/tools/build/feature/test-get_cpuid.c new file mode 100644 index 000000000000..d7a2c407130d --- /dev/null +++ b/tools/build/feature/test-get_cpuid.c @@ -0,0 +1,7 @@ +#include <cpuid.h> + +int main(void) +{ + unsigned int eax = 0, ebx = 0, ecx = 0, edx = 0; + return __get_cpuid(0x15, &eax, &ebx, &ecx, &edx); +} diff --git a/tools/build/feature/test-numa_num_possible_cpus.c b/tools/build/feature/test-numa_num_possible_cpus.c new file mode 100644 index 000000000000..2606e94b0659 --- /dev/null +++ b/tools/build/feature/test-numa_num_possible_cpus.c @@ -0,0 +1,6 @@ +#include <numa.h> + +int main(void) +{ + return numa_num_possible_cpus(); +} diff --git a/tools/lib/traceevent/event-parse.c b/tools/lib/traceevent/event-parse.c index 4d885934b919..cf42b090477b 100644 --- a/tools/lib/traceevent/event-parse.c +++ b/tools/lib/traceevent/event-parse.c @@ -3795,7 +3795,7 @@ static void print_str_arg(struct trace_seq *s, void *data, int size, struct format_field *field; struct printk_map *printk; long long val, fval; - unsigned long addr; + unsigned long long addr; char *str; unsigned char *hex; int print; @@ -3828,13 +3828,30 @@ static void print_str_arg(struct trace_seq *s, void *data, int size, */ if (!(field->flags & FIELD_IS_ARRAY) && field->size == pevent->long_size) { - addr = *(unsigned long *)(data + field->offset); + + /* Handle heterogeneous recording and processing + * architectures + * + * CASE I: + * Traces recorded on 32-bit devices (32-bit + * addressing) and processed on 64-bit devices: + * In this case, only 32 bits should be read. + * + * CASE II: + * Traces recorded on 64 bit devices and processed + * on 32-bit devices: + * In this case, 64 bits must be read. + */ + addr = (pevent->long_size == 8) ? + *(unsigned long long *)(data + field->offset) : + (unsigned long long)*(unsigned int *)(data + field->offset); + /* Check if it matches a print format */ printk = find_printk(pevent, addr); if (printk) trace_seq_puts(s, printk->printk); else - trace_seq_printf(s, "%lx", addr); + trace_seq_printf(s, "%llx", addr); break; } str = malloc(len + 1); diff --git a/tools/perf/Documentation/intel-pt.txt b/tools/perf/Documentation/intel-pt.txt index 4a0501d7a3b4..c94c9de3173e 100644 --- a/tools/perf/Documentation/intel-pt.txt +++ b/tools/perf/Documentation/intel-pt.txt @@ -364,21 +364,6 @@ cyc_thresh Specifies how frequently CYC packets are produced - see cyc CYC packets are not requested by default. -no_force_psb This is a driver option and is not in the IA32_RTIT_CTL MSR. - - It stops the driver resetting the byte count to zero whenever - enabling the trace (for example on context switches) which in - turn results in no PSB being forced. However some processors - will produce a PSB anyway. - - In any case, there is still a PSB when the trace is enabled for - the first time. - - no_force_psb can be used to slightly decrease the trace size but - may make it harder for the decoder to recover from errors. - - no_force_psb is not selected by default. - new snapshot option ------------------- diff --git a/tools/perf/config/Makefile b/tools/perf/config/Makefile index 827557fc7511..38a08539f4bf 100644 --- a/tools/perf/config/Makefile +++ b/tools/perf/config/Makefile @@ -573,9 +573,14 @@ ifndef NO_LIBNUMA msg := $(warning No numa.h found, disables 'perf bench numa mem' benchmark, please install numactl-devel/libnuma-devel/libnuma-dev); NO_LIBNUMA := 1 else - CFLAGS += -DHAVE_LIBNUMA_SUPPORT - EXTLIBS += -lnuma - $(call detected,CONFIG_NUMA) + ifeq ($(feature-numa_num_possible_cpus), 0) + msg := $(warning Old numa library found, disables 'perf bench numa mem' benchmark, please install numactl-devel/libnuma-devel/libnuma-dev >= 2.0.8); + NO_LIBNUMA := 1 + else + CFLAGS += -DHAVE_LIBNUMA_SUPPORT + EXTLIBS += -lnuma + $(call detected,CONFIG_NUMA) + endif endif endif @@ -621,8 +626,13 @@ ifdef LIBBABELTRACE endif ifndef NO_AUXTRACE - $(call detected,CONFIG_AUXTRACE) - CFLAGS += -DHAVE_AUXTRACE_SUPPORT + ifeq ($(feature-get_cpuid), 0) + msg := $(warning Your gcc lacks the __get_cpuid() builtin, disables support for auxtrace/Intel PT, please install a newer gcc); + NO_AUXTRACE := 1 + else + $(call detected,CONFIG_AUXTRACE) + CFLAGS += -DHAVE_AUXTRACE_SUPPORT + endif endif # Among the variables below, these: diff --git a/tools/perf/util/Build b/tools/perf/util/Build index 349bc96ca1fe..e5f18a288b74 100644 --- a/tools/perf/util/Build +++ b/tools/perf/util/Build @@ -17,6 +17,7 @@ libperf-y += levenshtein.o libperf-y += llvm-utils.o libperf-y += parse-options.o libperf-y += parse-events.o +libperf-y += perf_regs.o libperf-y += path.o libperf-y += rbtree.o libperf-y += bitmap.o @@ -103,7 +104,6 @@ libperf-$(CONFIG_LIBBABELTRACE) += data-convert-bt.o libperf-y += scripting-engines/ -libperf-$(CONFIG_PERF_REGS) += perf_regs.o libperf-$(CONFIG_ZLIB) += zlib.o libperf-$(CONFIG_LZMA) += lzma.o diff --git a/tools/perf/util/perf_regs.c b/tools/perf/util/perf_regs.c index 885e8ac83997..6b8eb13e14e4 100644 --- a/tools/perf/util/perf_regs.c +++ b/tools/perf/util/perf_regs.c @@ -6,6 +6,7 @@ const struct sample_reg __weak sample_reg_masks[] = { SMPL_REG_END }; +#ifdef HAVE_PERF_REGS_SUPPORT int perf_reg_value(u64 *valp, struct regs_dump *regs, int id) { int i, idx = 0; @@ -29,3 +30,4 @@ out: *valp = regs->cache_regs[id]; return 0; } +#endif diff --git a/tools/perf/util/perf_regs.h b/tools/perf/util/perf_regs.h index 2984dcc54d67..679d6e493962 100644 --- a/tools/perf/util/perf_regs.h +++ b/tools/perf/util/perf_regs.h @@ -2,6 +2,7 @@ #define __PERF_REGS_H #include <linux/types.h> +#include <linux/compiler.h> struct regs_dump; diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c index eb5f18b75402..c6f9af78f6f5 100644 --- a/tools/perf/util/probe-event.c +++ b/tools/perf/util/probe-event.c @@ -270,12 +270,13 @@ static int kernel_get_module_dso(const char *module, struct dso **pdso) int ret = 0; if (module) { - list_for_each_entry(dso, &host_machine->dsos.head, node) { - if (!dso->kernel) - continue; - if (strncmp(dso->short_name + 1, module, - dso->short_name_len - 2) == 0) - goto found; + char module_name[128]; + + snprintf(module_name, sizeof(module_name), "[%s]", module); + map = map_groups__find_by_name(&host_machine->kmaps, MAP__FUNCTION, module_name); + if (map) { + dso = map->dso; + goto found; } pr_debug("Failed to find module %s.\n", module); return -ENOENT; diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c index 8a4537ee9bc3..fc3f7c922f99 100644 --- a/tools/perf/util/session.c +++ b/tools/perf/util/session.c @@ -1580,7 +1580,10 @@ static int __perf_session__process_events(struct perf_session *session, file_offset = page_offset; head = data_offset - page_offset; - if (data_size && (data_offset + data_size < file_size)) + if (data_size == 0) + goto out; + + if (data_offset + data_size < file_size) file_size = data_offset + data_size; ui_progress__init(&prog, file_size, "Processing events..."); diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c index 415c359de465..2d065d065b67 100644 --- a/tools/perf/util/stat.c +++ b/tools/perf/util/stat.c @@ -196,7 +196,8 @@ static void zero_per_pkg(struct perf_evsel *counter) memset(counter->per_pkg_mask, 0, MAX_NR_CPUS); } -static int check_per_pkg(struct perf_evsel *counter, int cpu, bool *skip) +static int check_per_pkg(struct perf_evsel *counter, + struct perf_counts_values *vals, int cpu, bool *skip) { unsigned long *mask = counter->per_pkg_mask; struct cpu_map *cpus = perf_evsel__cpus(counter); @@ -218,6 +219,17 @@ static int check_per_pkg(struct perf_evsel *counter, int cpu, bool *skip) counter->per_pkg_mask = mask; } + /* + * we do not consider an event that has not run as a good + * instance to mark a package as used (skip=1). Otherwise + * we may run into a situation where the first CPU in a package + * is not running anything, yet the second is, and this function + * would mark the package as used after the first CPU and would + * not read the values from the second CPU. + */ + if (!(vals->run && vals->ena)) + return 0; + s = cpu_map__get_socket(cpus, cpu); if (s < 0) return -1; @@ -235,7 +247,7 @@ process_counter_values(struct perf_stat_config *config, struct perf_evsel *evsel static struct perf_counts_values zero; bool skip = false; - if (check_per_pkg(evsel, cpu, &skip)) { + if (check_per_pkg(evsel, count, cpu, &skip)) { pr_err("failed to read per-pkg counter\n"); return -1; } diff --git a/tools/perf/util/symbol-elf.c b/tools/perf/util/symbol-elf.c index 53bb5f59ec58..475d88d0a1c9 100644 --- a/tools/perf/util/symbol-elf.c +++ b/tools/perf/util/symbol-elf.c @@ -38,7 +38,7 @@ static inline char *bfd_demangle(void __maybe_unused *v, #endif #ifndef HAVE_ELF_GETPHDRNUM_SUPPORT -int elf_getphdrnum(Elf *elf, size_t *dst) +static int elf_getphdrnum(Elf *elf, size_t *dst) { GElf_Ehdr gehdr; GElf_Ehdr *ehdr; @@ -1271,8 +1271,6 @@ out_close: static int kcore__init(struct kcore *kcore, char *filename, int elfclass, bool temp) { - GElf_Ehdr *ehdr; - kcore->elfclass = elfclass; if (temp) @@ -1289,9 +1287,7 @@ static int kcore__init(struct kcore *kcore, char *filename, int elfclass, if (!gelf_newehdr(kcore->elf, elfclass)) goto out_end; - ehdr = gelf_getehdr(kcore->elf, &kcore->ehdr); - if (!ehdr) - goto out_end; + memset(&kcore->ehdr, 0, sizeof(GElf_Ehdr)); return 0; @@ -1348,23 +1344,18 @@ static int kcore__copy_hdr(struct kcore *from, struct kcore *to, size_t count) static int kcore__add_phdr(struct kcore *kcore, int idx, off_t offset, u64 addr, u64 len) { - GElf_Phdr gphdr; - GElf_Phdr *phdr; - - phdr = gelf_getphdr(kcore->elf, idx, &gphdr); - if (!phdr) - return -1; - - phdr->p_type = PT_LOAD; - phdr->p_flags = PF_R | PF_W | PF_X; - phdr->p_offset = offset; - phdr->p_vaddr = addr; - phdr->p_paddr = 0; - phdr->p_filesz = len; - phdr->p_memsz = len; - phdr->p_align = page_size; - - if (!gelf_update_phdr(kcore->elf, idx, phdr)) + GElf_Phdr phdr = { + .p_type = PT_LOAD, + .p_flags = PF_R | PF_W | PF_X, + .p_offset = offset, + .p_vaddr = addr, + .p_paddr = 0, + .p_filesz = len, + .p_memsz = len, + .p_align = page_size, + }; + + if (!gelf_update_phdr(kcore->elf, idx, &phdr)) return -1; return 0; diff --git a/tools/perf/util/util.c b/tools/perf/util/util.c index 7acafb3c5592..c2cd9bf2348b 100644 --- a/tools/perf/util/util.c +++ b/tools/perf/util/util.c @@ -709,7 +709,7 @@ bool find_process(const char *name) dir = opendir(procfs__mountpoint()); if (!dir) - return -1; + return false; /* Walk through the directory. */ while (ret && (d = readdir(dir)) != NULL) { diff --git a/tools/power/x86/turbostat/turbostat.c b/tools/power/x86/turbostat/turbostat.c index 9655cb49c7cb..bde0ef1a63df 100644 --- a/tools/power/x86/turbostat/turbostat.c +++ b/tools/power/x86/turbostat/turbostat.c @@ -71,8 +71,11 @@ unsigned int extra_msr_offset32; unsigned int extra_msr_offset64; unsigned int extra_delta_offset32; unsigned int extra_delta_offset64; +unsigned int aperf_mperf_multiplier = 1; int do_smi; double bclk; +double base_hz; +double tsc_tweak = 1.0; unsigned int show_pkg; unsigned int show_core; unsigned int show_cpu; @@ -502,7 +505,7 @@ int format_counters(struct thread_data *t, struct core_data *c, /* %Busy */ if (has_aperf) { if (!skip_c0) - outp += sprintf(outp, "%8.2f", 100.0 * t->mperf/t->tsc); + outp += sprintf(outp, "%8.2f", 100.0 * t->mperf/t->tsc/tsc_tweak); else outp += sprintf(outp, "********"); } @@ -510,7 +513,7 @@ int format_counters(struct thread_data *t, struct core_data *c, /* Bzy_MHz */ if (has_aperf) outp += sprintf(outp, "%8.0f", - 1.0 * t->tsc / units * t->aperf / t->mperf / interval_float); + 1.0 * t->tsc * tsc_tweak / units * t->aperf / t->mperf / interval_float); /* TSC_MHz */ outp += sprintf(outp, "%8.0f", 1.0 * t->tsc/units/interval_float); @@ -984,6 +987,8 @@ int get_counters(struct thread_data *t, struct core_data *c, struct pkg_data *p) return -3; if (get_msr(cpu, MSR_IA32_MPERF, &t->mperf)) return -4; + t->aperf = t->aperf * aperf_mperf_multiplier; + t->mperf = t->mperf * aperf_mperf_multiplier; } if (do_smi) { @@ -1149,6 +1154,19 @@ int slv_pkg_cstate_limits[16] = {PCL__0, PCL__1, PCLRSV, PCLRSV, PCL__4, PCLRSV, int amt_pkg_cstate_limits[16] = {PCL__0, PCL__1, PCL__2, PCLRSV, PCLRSV, PCLRSV, PCL__6, PCL__7, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV}; int phi_pkg_cstate_limits[16] = {PCL__0, PCL__2, PCL_6N, PCL_6R, PCLRSV, PCLRSV, PCLRSV, PCLUNL, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV, PCLRSV}; + +static void +calculate_tsc_tweak() +{ + unsigned long long msr; + unsigned int base_ratio; + + get_msr(base_cpu, MSR_NHM_PLATFORM_INFO, &msr); + base_ratio = (msr >> 8) & 0xFF; + base_hz = base_ratio * bclk * 1000000; + tsc_tweak = base_hz / tsc_hz; +} + static void dump_nhm_platform_info(void) { @@ -1926,8 +1944,6 @@ int has_config_tdp(unsigned int family, unsigned int model) switch (model) { case 0x3A: /* IVB */ - case 0x3E: /* IVB Xeon */ - case 0x3C: /* HSW */ case 0x3F: /* HSX */ case 0x45: /* HSW */ @@ -2543,6 +2559,13 @@ int is_knl(unsigned int family, unsigned int model) return 0; } +unsigned int get_aperf_mperf_multiplier(unsigned int family, unsigned int model) +{ + if (is_knl(family, model)) + return 1024; + return 1; +} + #define SLM_BCLK_FREQS 5 double slm_freq_table[SLM_BCLK_FREQS] = { 83.3, 100.0, 133.3, 116.7, 80.0}; @@ -2744,6 +2767,9 @@ void process_cpuid() } } + if (has_aperf) + aperf_mperf_multiplier = get_aperf_mperf_multiplier(family, model); + do_nhm_platform_info = do_nhm_cstates = do_smi = probe_nhm_msrs(family, model); do_snb_cstates = has_snb_msrs(family, model); do_pc2 = do_snb_cstates && (pkg_cstate_limit >= PCL__2); @@ -2762,6 +2788,9 @@ void process_cpuid() if (debug) dump_cstate_pstate_config_info(); + if (has_skl_msrs(family, model)) + calculate_tsc_tweak(); + return; } @@ -3090,7 +3119,7 @@ int get_and_dump_counters(void) } void print_version() { - fprintf(stderr, "turbostat version 4.7 17-June, 2015" + fprintf(stderr, "turbostat version 4.8 26-Sep, 2015" " - Len Brown <lenb@kernel.org>\n"); } |