summaryrefslogtreecommitdiff
path: root/drivers/gpu
AgeCommit message (Collapse)Author
2016-12-28drm: Unconditionally do the range check in drm_mm_scan_add_block()Chris Wilson
Doing the check is trivial (low cost in comparison to overall eviction) and helps simplify the code. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-29-chris@chris-wilson.co.uk
2016-12-28drm: Rename prev_node to hole in drm_mm_scan_add_block()Chris Wilson
Acknowledging that we were building up the hole was more useful to me when reading the code, than knowing the relationship between this node and the previous node. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-28-chris@chris-wilson.co.uk
2016-12-28drm: Fix O= out-of-tree builds for selftestsDaniel Vetter
Kbuild really doesn't like non-recursive Makefiles, but they do work as long as you build without O= Reported-by: kbuild test robot <fengguang.wu@intel.com> Fixes: 50f0033d1a0f ("drm: Add some kselftests for the DRM range manager (struct drm_mm)") Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Christian König <christian.koenig@amd.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Daniel Vetter <daniel.vetter@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1482918077-30027-1-git-send-email-daniel.vetter@ffwll.ch
2016-12-27drm: Extract struct drm_mm_scan from struct drm_mmChris Wilson
The scan state occupies a large proportion of the struct drm_mm and is rarely used and only contains temporary state. That makes it suitable to moving to its struct and onto the stack of the callers. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> [danvet: Fix up etnaviv to compile, was missing a BUG_ON.] Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2016-12-27drm: Add asserts to catch overflow in drm_mm_init() and drm_mm_init_scan()Chris Wilson
A simple assert to ensure that we don't overflow start + size when initialising the drm_mm, or its scanner. In future, we may want to switch to tracking the value of ranges (rather than size) so that we can cover the full u64, for example like resource tracking. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-26-chris@chris-wilson.co.uk
2016-12-27drm: Simplify drm_mm_clean()Chris Wilson
Since commit ea7b1dd44867 ("drm: mm: track free areas implicitly"), to test whether there are any nodes allocated within the range manager, we merely have to ask whether the node_list is empty. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-25-chris@chris-wilson.co.uk
2016-12-27drm: Detect overflow in drm_mm_reserve_node()Chris Wilson
Protect ourselves from a caller passing in node.start + node.size that will overflow and trick us into reserving that node. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-24-chris@chris-wilson.co.uk
2016-12-27drm: Fix kerneldoc for drm_mm_scan_remove_block()Chris Wilson
The nodes must be removed in the *reverse* order. This is correct in the overview, but backwards in the function description. Whilst here add Intel's copyright statement and tweak some formatting. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-23-chris@chris-wilson.co.uk
2016-12-27drm: Promote drm_mm alignment to u64Chris Wilson
In places (e.g. i915.ko), the alignment is exported to userspace as u64 and there now exists hardware for which we can indeed utilize a u64 alignment. As such, we need to keep 64bit integers throughout when handling alignment. Testcase: igt/drm_mm/align64 Testcase: igt/gem_exec_alignment Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Christian König <christian.koenig@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-22-chris@chris-wilson.co.uk
2016-12-27drm: kselftest for drm_mm and restricted color evictionChris Wilson
Check that after applying the driver's color adjustment, restricted eviction scanning finds a suitable hole. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-20-chris@chris-wilson.co.uk
2016-12-27drm: kselftest for drm_mm and color evictionChris Wilson
Check that after applying the driver's color adjustment, eviction scanning finds a suitable hole. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-19-chris@chris-wilson.co.uk
2016-12-27drm: kselftest for drm_mm and color adjustmentChris Wilson
Check that after applying the driver's color adjustment, fitting of the node and its alignment are still correct. v2: s/no_color_touching/separate_adjacent_colors/ Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-18-chris@chris-wilson.co.uk
2016-12-27drm: kselftest for drm_mm and top-down allocationChris Wilson
Check that if we request top-down allocation from drm_mm_insert_node() we receive the next available hole from the top. v2: Flip sign on conditional assert. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-17-chris@chris-wilson.co.uk
2016-12-27drm: kselftest for drm_mm and range restricted evictionChris Wilson
Check that we add arbitrary blocks to a restrited eviction scanner in order to find the first minimal hole that matches our request. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-16-chris@chris-wilson.co.uk
2016-12-27drm: kselftest for drm_mm and evictionChris Wilson
Check that we add arbitrary blocks to the eviction scanner in order to find the first minimal hole that matches our request. v2: Refactor out some common eviction code for later Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-15-chris@chris-wilson.co.uk
2016-12-27drm: kselftest for drm_mm and alignmentChris Wilson
Check that we can request alignment to any power-of-two or prime using a plain drm_mm_node_insert(), and also handle a reasonable selection of primes. v2: Exercise all allocation flags Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-14-chris@chris-wilson.co.uk
2016-12-27drm: kselftest for drm_mm_insert_node_in_range()Chris Wilson
Exercise drm_mm_insert_node_in_range(), check that we only allocate from the specified range. v2: Use all allocation flags v3: Don't pass in invalid ranges - these will be asserted later. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-13-chris@chris-wilson.co.uk
2016-12-27drm: kselftest for drm_mm_replace_node()Chris Wilson
Reuse drm_mm_insert_node() with a temporary node to exercise drm_mm_replace_node(). We use the previous test in order to exercise the various lists following replacement. v2: Check that we copy across the important (user) details of the node. The internal details (such as lists and hole tracking) we hope to detect errors by exercise. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-12-chris@chris-wilson.co.uk
2016-12-27drm: kselftest for drm_mm_insert_node()Chris Wilson
Exercise drm_mm_insert_node(), check that we can't overfill a range and that the lists are correct after reserving/removing. v2: Extract helpers for the repeated tests v3: Iterate over all allocation flags Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-11-chris@chris-wilson.co.uk
2016-12-27drm: kselftest for drm_mm_reserve_node()Chris Wilson
Exercise drm_mm_reserve_node(), check that we can't reserve an already occupied range and that the lists are correct after reserving/removing. v2: Check for invalid node reservation. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-10-chris@chris-wilson.co.uk
2016-12-27drm: kselftest for drm_mm_debug()Chris Wilson
Simple test to just exercise calling the debug dumper on the drm_mm. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-9-chris@chris-wilson.co.uk
2016-12-27drm: kselftest for drm_mm_init()Chris Wilson
Simple first test to just exercise initialisation of struct drm_mm. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-8-chris@chris-wilson.co.uk
2016-12-27drm: Add some kselftests for the DRM range manager (struct drm_mm)Chris Wilson
First we introduce a smattering of infrastructure for writing selftests. The idea is that we have a test module that exercises a particular portion of the exported API, and that module provides a set of tests that can either be run as an ensemble via kselftest or individually via an igt harness (in this case igt/drm_mm). To accommodate selecting individual tests, we export a boolean parameter to control selection of each test - that is hidden inside a bunch of reusable boilerplate macros to keep writing the tests simple. v2: Choose a random random_seed unless one is specified by the user. v3: More parameters to control max_iterations and max_prime of the tests. Testcase: igt/drm_mm Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Acked-by: Christian König <christian.koenig@amd.com> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-7-chris@chris-wilson.co.uk
2016-12-27drm: Add a simple generator of random permutationsChris Wilson
When testing, we want a random but yet reproducible order in which to process elements. Here we create an array which is a random (using the Tausworthe PRNG) permutation of the order in which to execute. Note these are simple helpers intended to be merged upstream in lib/ v2: Tidier code by David Herrmann v3: Add reminder that this code is intended to be temporary, with at least the bulk of the prandom changes going to lib/ Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: David Herrmann <dh.herrmann@gmail.com> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-6-chris@chris-wilson.co.uk
2016-12-27drm: Compile time enabling for asserts in drm_mmChris Wilson
Use CONFIG_DRM_DEBUG_MM to conditionally enable the internal and validation checking using BUG_ON. Ideally these paths should all be exercised by CI selftests (with the asserts enabled). Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-4-chris@chris-wilson.co.uk Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-4-chris@chris-wilson.co.uk
2016-12-27drm: Use drm_mm_nodes() as shorthand for the list of nodes under struct drm_mmChris Wilson
Fairly commonly we want to inspect the node list on the struct drm_mm, which is buried within an embedded node. Bring it to the surface with a bit of syntatic sugar. Note this was intended to be split from commit ad579002c8ec ("drm: Add drm_mm_for_each_node_safe()") before being applied, but my timing sucks. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/20161222083641.2691-3-chris@chris-wilson.co.uk
2016-12-27drm: Get atomic property value even if DRIVER_ATOMIC is not setDhinakaran Pandiyan
i915 does not set DRIVER_ATOMIC by default yet but uses atomic_check and atomic_commit. drm_object_property_get_value() does not read the correct value of atomic properties if DRIVER_ATOMIC is not set. Checking whether the driver uses atomic modeset is a better check instead as the property values are tracked in the state structures. v2: Included header Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/1482396643-32456-2-git-send-email-dhinakaran.pandiyan@intel.com
2016-12-27drm: Wrap the check for atomic_commit implementationDhinakaran Pandiyan
This check is useful for drivers that do not have DRIVER_ATOMIC set but have atomic modesetting internally implemented. Wrap the check into a function since this is used in many places and as a bonus, the function name helps to document what the check is for. v2: Change return type to bool (Ville) Move the function drm_atomic.h (Daniel) Fixed comment marker for documentation Suggested-by: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: Ben Skeggs <bskeggs@redhat.com> Signed-off-by: Dhinakaran Pandiyan <dhinakaran.pandiyan@intel.com> [danvet: Move back to drmP.h because include hell.] Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch> Link: http://patchwork.freedesktop.org/patch/msgid/1482396643-32456-1-git-send-email-dhinakaran.pandiyan@intel.com
2016-12-26drm/i915: Update DRIVER_DATE to 20161226Daniel Vetter
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2016-12-26drm/i915/gvt: fix typo in cfg_space range checkPei Zhang
PCI basic config space's size is 256 bytes. When check if access crosses space range, should use "> 256". Signed-off-by: Pei Zhang <pei.zhang@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
2016-12-26drm/i915/gvt: fix an issue in emulating cfg space PCI_COMMANDMin He
There's an issue in current cfg space emulation for PCI_COMMAND (offset 0x4): when guest changes some bits other than PCI_COMMAND_MEMORY, this write operation will not be written to virutal cfg space successfully. This patch is to fix the wrong behavior above. Signed-off-by: Min He <min.he@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
2016-12-26drm/i915/gvt/kvmgt: trival: code cleanupJike Song
Don't introduce local variables unless necessary. Signed-off-by: Jike Song <jike.song@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
2016-12-26drm/i915/gvt/kvmgt: prevent double-release of vgpuJike Song
The release action might be triggered from either user's closing mdev or the detaching event of kvm and vfio_group, so this patch introduces an atomic to prevent double-release. Signed-off-by: Jike Song <jike.song@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
2016-12-26drm/i915/gvt/kvmgt: check returned slot for gfnJike Song
gfn_to_memslot() may return NULL if the gfn is mmio or invalid. A malicious user might input a bad gfn to panic the host if we don't check it. Signed-off-by: Jike Song <jike.song@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
2016-12-26drm/i915/gvt/kvmgt: dereference the pointer within lockJike Song
Though there is no issue exposed yet, it's possible that another thread releases the entry while our trying to deref it out of the lock. Fit it by moving the dereference within lock. Signed-off-by: Jike Song <jike.song@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
2016-12-26drm/i915/gvt: reset the GGTT entry when vGPU createdPing Gao
The GGTT space is partitioned between vGPUs, it could be reused by next vGPU after previous one is release, the stale entries need point to scratch page when vGPU created. v2: Reset logic move to vGPU create. v3: Correct the commit msg. v4: Move the reset function to vGPU init gtt function, as result it's no need explicitly in vGPU reset logic as vGPU init gtt called during reset. Signed-off-by: Ping Gao <ping.a.gao@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
2016-12-26drm/i915/gvt: fix an error in opregion handlingMin He
It should be vgpu_opregion(vgpu)->va, not vgpu_opregion(vgpu). Signed-off-by: Min He <min.he@intel.com> Signed-off-by: Zhenyu Wang <zhenyuw@linux.intel.com>
2016-12-25ktime: Cleanup ktime_set() usageThomas Gleixner
ktime_set(S,N) was required for the timespec storage type and is still useful for situations where a Seconds and Nanoseconds part of a time value needs to be converted. For anything where the Seconds argument is 0, this is pointless and can be replaced with a simple assignment. Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Cc: Peter Zijlstra <peterz@infradead.org>
2016-12-24drm/i915: re-use computed offset bias for context pinDaniele Ceraolo Spurio
The context has to obey the same offset requirements as the ring, so we can re-use the same bias value we computed for the ring instead of unconditionally using GUC_WOPCM_TOP. Suggested-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Link: http://patchwork.freedesktop.org/patch/msgid/1482537382-28584-2-git-send-email-daniele.ceraolospurio@intel.com Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2016-12-24drm/i915: request ring to be pinned above GUC_WOPCM_TOPDaniele Ceraolo Spurio
GuC will validate the ring offset and fail if it is in the [0, GUC_WOPCM_TOP) range. The bias is conditionally applied only if GuC loading is enabled (we can't check for guc submission enabled as in other cases because HuC loading requires this fix). Note that the default context is processed before enable_guc_loading is sanitized, so we might still apply the bias to its ring even if it is not needed. v2: compute the value during ctx init and pass it to intel_ring_pin (Chris), updated commit message Signed-off-by: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com> Cc: Chris Wilson <chris@chris-wilson.co.uk> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Arkadiusz Hiler <arkadiusz.hiler@intel.com> Cc: Anusha Srivatsa <anusha.srivatsa@intel.com> Cc: Michał Winiarski <michal.winiarski@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1482537382-28584-1-git-send-email-daniele.ceraolospurio@intel.com Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
2016-12-23drm/i915: Prevent timeline updates whilst performing resetChris Wilson
As the fence may be signaled concurrently from an interrupt on another device, it is possible for the list of requests on the timeline to be modified as we walk it. Take both (the context's timeline and the global timeline) locks to prevent such modifications. Fixes: 80b204bce8f2 ("drm/i915: Enable multiple timelines") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com> Cc: <drm-intel-fixes@lists.freedesktop.org> Reviewed-by: Mika Kuoppala <mika.kuoppala@linux.intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20161223145804.6605-10-chris@chris-wilson.co.uk
2016-12-23drm/i915: Assert that the partial VMA fits within the objectChris Wilson
When creating a partial VMA assert that it first fits with the parent object, and that if it covers the whole of the parent a normal view was created instead. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20161223145804.6605-5-chris@chris-wilson.co.uk
2016-12-23drm/i915: Silence allocation failure during sg_trim()Chris Wilson
As trimming the sg table is merely an optimisation that gracefully fails if we cannot allocate a new table, we do not need to report the failure either. Fixes: 0c40ce130e38 ("drm/i915: Trim the object sg table") Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20161223145804.6605-4-chris@chris-wilson.co.uk
2016-12-23drm/i915: Don't clflush before release phys objectChris Wilson
When we teardown the backing storage for the phys object, we copy from the coherent contiguous block back to the shmemfs object, clflushing as we go. Trying to clflush the invalid sg beforehand just oops and would be redundant (due to it already being coherent, and clflushed afterwards). Reported-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Cc: <drm-intel-fixes@lists.freedesktop.org> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20161223145804.6605-3-chris@chris-wilson.co.uk
2016-12-23drm/i915: Repeat flush of idle work during suspendChris Wilson
The idle work handler is self-arming - if it detects that it needs to run again it will queue itself from its work handler. Take greater care when trying to drain the idle work, and double check that it is flushed. The free worker has a similar issue where it is armed by an RCU task which may be running concurrently with us. This should hopefully help with the sporadic WARN_ON(dev_priv->gt.awake) from i915_gem_suspend. v2: Reuse drain_freed_objects. v3: Don't try to flush the freed objects from the shrinker, as it may be underneath the struct_mutex already. v4: do while and comment upon the excess rcu_barrier in drain_freed_objects Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20161223145804.6605-2-chris@chris-wilson.co.uk
2016-12-23drm/i915: Break after walking all GGTT vma in bump_inactive_ggttChris Wilson
Since commit db6c2b4151f2 ("drm/i915: Store the vma in an rbtree under the object") the vma are once again sorted into GGTT first, then ppGTT so that the typical case of walking the GGTT vma can stop as soon as we find a non-ppGTT. Apply that optimisation. Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20161223145804.6605-1-chris@chris-wilson.co.uk
2016-12-23drm/i915: Kill the 830 MI_OVERLAY_OFF workaroundVille Syrjälä
Now that we're disabling L2 clock gating MI_OVERLAY_OFF actually works on 830, so let's use it. v2: Nuke the unused dev_priv variable Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20161222195222.9618-1-ville.syrjala@linux.intel.com Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk>
2016-12-23drm/i915/dsi: Move disable pll call outside of clear_device_ready()Hans de Goede
On enable intel_dsi_enable() directly calls intel_enable_dsi_pll(), make intel_dsi_disable() also directly call intel_disable_dsi_pll(), rather then hiding the call in intel_dsi_clear_device_ready(), no functional changes. Signed-off-by: Hans de Goede <hdegoede@redhat.com> Reviewed-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/20161201202925.12220-7-hdegoede@redhat.com
2016-12-23drm/915: Parsing the missed out DTD fields from the VBTVincente Tsou
The upper bits of the vsync width, vsync offset and hsync width were not parsed from the VBT. Parse these fields in this patch. V2: Renamed lvds dvo timing structure members and code identation fix (Jani's review comments) V3: Corrected commit message, used "from the VBT" Signed-off-by: Vincente Tsou <vincente.tsou@intel.com> Signed-off-by: Madhav Chauhan <madhav.chauhan@intel.com> Reviewed-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1482430993-3265-1-git-send-email-madhav.chauhan@intel.com
2016-12-22drm/i915: Disable L2 cache clock gating on 830 when using the overlayVille Syrjälä
BSpec says: "Overlay Clock Gating Must be Disabled: Overlay & L2 Cache clock gating must be disabled in order to prevent device hangs when turning off overlay.SW must turn off Ovrunit clock gating (6200h) and L2 Cache clock gating (C8h)." We only turned off the overlay clock gating (due to lack of docs I presume). After a bit of experimentation it looks like the the magic C8h register lives in the PCI config space of device 0, and the magic bit appears to be bit 2. Or at the very least this eliminates the GPU death after MI_OVERLAY_OFF. L2 clock gating seems to save ~80mW, so let's keep it on unless we need to actually use the overlay. Also let's move the OVRUNIT clock gating to the same place since we can, and 845 supposedly doesn't need it. Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com> Link: http://patchwork.freedesktop.org/patch/msgid/1481131693-27993-11-git-send-email-ville.syrjala@linux.intel.com Acked-by: Chris Wilson <chris@chris-wilson.co.uk>
generated by cgit v1.2.3 (git 2.46.0) at 2025-07-23 14:41:23 +0000