diff options
158 files changed, 2144 insertions, 1634 deletions
diff --git a/Documentation/filesystems/nilfs2.txt b/Documentation/filesystems/nilfs2.txt index d5c0cef38a71..873a2ab2e9f8 100644 --- a/Documentation/filesystems/nilfs2.txt +++ b/Documentation/filesystems/nilfs2.txt @@ -40,7 +40,6 @@ Features which NILFS2 does not support yet: - POSIX ACLs - quotas - fsck - - resize - defragmentation Mount options diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index d3d653a5f9b9..bfe924217f24 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -346,7 +346,7 @@ tcp_orphan_retries - INTEGER when RTO retransmissions remain unacknowledged. See tcp_retries2 for more details. - The default value is 7. + The default value is 8. If your machine is a loaded WEB server, you should think about lowering this value, such sockets may consume significant resources. Cf. tcp_max_orphans. diff --git a/Documentation/x86/boot.txt b/Documentation/x86/boot.txt index 9b7221a86df2..7c3a8801b7ce 100644 --- a/Documentation/x86/boot.txt +++ b/Documentation/x86/boot.txt @@ -674,7 +674,7 @@ Protocol: 2.10+ Field name: init_size Type: read -Offset/size: 0x25c/4 +Offset/size: 0x260/4 This field indicates the amount of linear contiguous memory starting at the kernel runtime start address that the kernel needs before it @@ -1,7 +1,7 @@ VERSION = 3 PATCHLEVEL = 0 SUBLEVEL = 0 -EXTRAVERSION = -rc7 +EXTRAVERSION = NAME = Sneaky Weasel # *DOCUMENTATION* diff --git a/arch/arm/mach-davinci/board-dm365-evm.c b/arch/arm/mach-davinci/board-dm365-evm.c index c67f684ee3e5..09a87e61ffcf 100644 --- a/arch/arm/mach-davinci/board-dm365-evm.c +++ b/arch/arm/mach-davinci/board-dm365-evm.c @@ -520,7 +520,7 @@ fail: */ if (have_imager()) { label = "HD imager"; - mux |= 1; + mux |= 2; /* externally mux MMC1/ENET/AIC33 to imager */ mux |= BIT(6) | BIT(5) | BIT(3); @@ -540,7 +540,7 @@ fail: resets &= ~BIT(1); if (have_tvp7002()) { - mux |= 2; + mux |= 1; resets &= ~BIT(2); label = "tvp7002 HD"; } else { diff --git a/arch/arm/mach-davinci/gpio.c b/arch/arm/mach-davinci/gpio.c index e7221398e5af..cafbe13a82a5 100644 --- a/arch/arm/mach-davinci/gpio.c +++ b/arch/arm/mach-davinci/gpio.c @@ -254,8 +254,10 @@ gpio_irq_handler(unsigned irq, struct irq_desc *desc) { struct davinci_gpio_regs __iomem *g; u32 mask = 0xffff; + struct davinci_gpio_controller *d; - g = (__force struct davinci_gpio_regs __iomem *) irq_desc_get_handler_data(desc); + d = (struct davinci_gpio_controller *)irq_desc_get_handler_data(desc); + g = (struct davinci_gpio_regs __iomem *)d->regs; /* we only care about one bank */ if (irq & 1) @@ -274,11 +276,14 @@ gpio_irq_handler(unsigned irq, struct irq_desc *desc) if (!status) break; __raw_writel(status, &g->intstat); - if (irq & 1) - status >>= 16; /* now demux them to the right lowlevel handler */ - n = (int)irq_get_handler_data(irq); + n = d->irq_base; + if (irq & 1) { + n += 16; + status >>= 16; + } + while (status) { res = ffs(status); n += res; @@ -424,7 +429,13 @@ static int __init davinci_gpio_irq_setup(void) /* set up all irqs in this bank */ irq_set_chained_handler(bank_irq, gpio_irq_handler); - irq_set_handler_data(bank_irq, (__force void *)g); + + /* + * Each chip handles 32 gpios, and each irq bank consists of 16 + * gpio irqs. Pass the irq bank's corresponding controller to + * the chained irq handler. + */ + irq_set_handler_data(bank_irq, &chips[gpio / 32]); for (i = 0; i < 16 && gpio < ngpio; i++, irq++, gpio++) { irq_set_chip(irq, &gpio_irqchip); diff --git a/arch/arm/mach-davinci/irq.c b/arch/arm/mach-davinci/irq.c index d8c1af025931..952dc126c390 100644 --- a/arch/arm/mach-davinci/irq.c +++ b/arch/arm/mach-davinci/irq.c @@ -52,6 +52,12 @@ davinci_alloc_gc(void __iomem *base, unsigned int irq_start, unsigned int num) struct irq_chip_type *ct; gc = irq_alloc_generic_chip("AINTC", 1, irq_start, base, handle_edge_irq); + if (!gc) { + pr_err("%s: irq_alloc_generic_chip for IRQ %u failed\n", + __func__, irq_start); + return; + } + ct = gc->chip_types; ct->chip.irq_ack = irq_gc_ack_set_bit; ct->chip.irq_mask = irq_gc_mask_clr_bit; diff --git a/arch/arm/mach-ixp4xx/common.c b/arch/arm/mach-ixp4xx/common.c index 74ed81a3cb1a..07772575d7ab 100644 --- a/arch/arm/mach-ixp4xx/common.c +++ b/arch/arm/mach-ixp4xx/common.c @@ -419,14 +419,20 @@ static void notrace ixp4xx_update_sched_clock(void) /* * clocksource */ + +static cycle_t ixp4xx_clocksource_read(struct clocksource *c) +{ + return *IXP4XX_OSTS; +} + unsigned long ixp4xx_timer_freq = IXP4XX_TIMER_FREQ; EXPORT_SYMBOL(ixp4xx_timer_freq); static void __init ixp4xx_clocksource_init(void) { init_sched_clock(&cd, ixp4xx_update_sched_clock, 32, ixp4xx_timer_freq); - clocksource_mmio_init(&IXP4XX_OSTS, "OSTS", ixp4xx_timer_freq, 200, 32, - clocksource_mmio_readl_up); + clocksource_mmio_init(NULL, "OSTS", ixp4xx_timer_freq, 200, 32, + ixp4xx_clocksource_read); } /* diff --git a/arch/arm/mach-s3c64xx/dma.c b/arch/arm/mach-s3c64xx/dma.c index b197171e7d03..204bfafe4bfc 100644 --- a/arch/arm/mach-s3c64xx/dma.c +++ b/arch/arm/mach-s3c64xx/dma.c @@ -113,7 +113,7 @@ found: return chan; } -int s3c2410_dma_config(unsigned int channel, int xferunit) +int s3c2410_dma_config(enum dma_ch channel, int xferunit) { struct s3c2410_dma_chan *chan = s3c_dma_lookup_channel(channel); @@ -297,7 +297,7 @@ static int s3c64xx_dma_flush(struct s3c2410_dma_chan *chan) return 0; } -int s3c2410_dma_ctrl(unsigned int channel, enum s3c2410_chan_op op) +int s3c2410_dma_ctrl(enum dma_ch channel, enum s3c2410_chan_op op) { struct s3c2410_dma_chan *chan = s3c_dma_lookup_channel(channel); @@ -331,7 +331,7 @@ EXPORT_SYMBOL(s3c2410_dma_ctrl); * */ -int s3c2410_dma_enqueue(unsigned int channel, void *id, +int s3c2410_dma_enqueue(enum dma_ch channel, void *id, dma_addr_t data, int size) { struct s3c2410_dma_chan *chan = s3c_dma_lookup_channel(channel); @@ -415,7 +415,7 @@ err_buff: EXPORT_SYMBOL(s3c2410_dma_enqueue); -int s3c2410_dma_devconfig(unsigned int channel, +int s3c2410_dma_devconfig(enum dma_ch channel, enum s3c2410_dmasrc source, unsigned long devaddr) { @@ -463,7 +463,7 @@ int s3c2410_dma_devconfig(unsigned int channel, EXPORT_SYMBOL(s3c2410_dma_devconfig); -int s3c2410_dma_getposition(unsigned int channel, +int s3c2410_dma_getposition(enum dma_ch channel, dma_addr_t *src, dma_addr_t *dst) { struct s3c2410_dma_chan *chan = s3c_dma_lookup_channel(channel); @@ -487,7 +487,7 @@ EXPORT_SYMBOL(s3c2410_dma_getposition); * get control of an dma channel */ -int s3c2410_dma_request(unsigned int channel, +int s3c2410_dma_request(enum dma_ch channel, struct s3c2410_dma_client *client, void *dev) { @@ -533,7 +533,7 @@ EXPORT_SYMBOL(s3c2410_dma_request); * allowed to go through. */ -int s3c2410_dma_free(unsigned int channel, struct s3c2410_dma_client *client) +int s3c2410_dma_free(enum dma_ch channel, struct s3c2410_dma_client *client) { struct s3c2410_dma_chan *chan = s3c_dma_lookup_channel(channel); unsigned long flags; diff --git a/arch/arm/plat-s3c24xx/dma.c b/arch/arm/plat-s3c24xx/dma.c index a79a8ccd25f6..539bd0e3defd 100644 --- a/arch/arm/plat-s3c24xx/dma.c +++ b/arch/arm/plat-s3c24xx/dma.c @@ -712,7 +712,7 @@ static struct s3c2410_dma_chan *s3c2410_dma_map_channel(int channel); * get control of an dma channel */ -int s3c2410_dma_request(unsigned int channel, +int s3c2410_dma_request(enum dma_ch channel, struct s3c2410_dma_client *client, void *dev) { @@ -783,7 +783,7 @@ EXPORT_SYMBOL(s3c2410_dma_request); * allowed to go through. */ -int s3c2410_dma_free(unsigned int channel, struct s3c2410_dma_client *client) +int s3c2410_dma_free(enum dma_ch channel, struct s3c2410_dma_client *client) { struct s3c2410_dma_chan *chan = s3c_dma_lookup_channel(channel); unsigned long flags; @@ -974,7 +974,7 @@ static int s3c2410_dma_started(struct s3c2410_dma_chan *chan) } int -s3c2410_dma_ctrl(unsigned int channel, enum s3c2410_chan_op op) +s3c2410_dma_ctrl(enum dma_ch channel, enum s3c2410_chan_op op) { struct s3c2410_dma_chan *chan = s3c_dma_lookup_channel(channel); @@ -1021,7 +1021,7 @@ EXPORT_SYMBOL(s3c2410_dma_ctrl); * xfersize: size of unit in bytes (1,2,4) */ -int s3c2410_dma_config(unsigned int channel, +int s3c2410_dma_config(enum dma_ch channel, int xferunit) { struct s3c2410_dma_chan *chan = s3c_dma_lookup_channel(channel); @@ -1100,7 +1100,7 @@ EXPORT_SYMBOL(s3c2410_dma_config); * devaddr: physical address of the source */ -int s3c2410_dma_devconfig(unsigned int channel, +int s3c2410_dma_devconfig(enum dma_ch channel, enum s3c2410_dmasrc source, unsigned long devaddr) { @@ -1173,7 +1173,7 @@ EXPORT_SYMBOL(s3c2410_dma_devconfig); * returns the current transfer points for the dma source and destination */ -int s3c2410_dma_getposition(unsigned int channel, dma_addr_t *src, dma_addr_t *dst) +int s3c2410_dma_getposition(enum dma_ch channel, dma_addr_t *src, dma_addr_t *dst) { struct s3c2410_dma_chan *chan = s3c_dma_lookup_channel(channel); diff --git a/arch/arm/plat-samsung/dma.c b/arch/arm/plat-samsung/dma.c index cb459dd95459..6143aa147688 100644 --- a/arch/arm/plat-samsung/dma.c +++ b/arch/arm/plat-samsung/dma.c @@ -41,7 +41,7 @@ struct s3c2410_dma_chan *s3c_dma_lookup_channel(unsigned int channel) * irq? */ -int s3c2410_dma_set_opfn(unsigned int channel, s3c2410_dma_opfn_t rtn) +int s3c2410_dma_set_opfn(enum dma_ch channel, s3c2410_dma_opfn_t rtn) { struct s3c2410_dma_chan *chan = s3c_dma_lookup_channel(channel); @@ -56,7 +56,7 @@ int s3c2410_dma_set_opfn(unsigned int channel, s3c2410_dma_opfn_t rtn) } EXPORT_SYMBOL(s3c2410_dma_set_opfn); -int s3c2410_dma_set_buffdone_fn(unsigned int channel, s3c2410_dma_cbfn_t rtn) +int s3c2410_dma_set_buffdone_fn(enum dma_ch channel, s3c2410_dma_cbfn_t rtn) { struct s3c2410_dma_chan *chan = s3c_dma_lookup_channel(channel); @@ -71,7 +71,7 @@ int s3c2410_dma_set_buffdone_fn(unsigned int channel, s3c2410_dma_cbfn_t rtn) } EXPORT_SYMBOL(s3c2410_dma_set_buffdone_fn); -int s3c2410_dma_setflags(unsigned int channel, unsigned int flags) +int s3c2410_dma_setflags(enum dma_ch channel, unsigned int flags) { struct s3c2410_dma_chan *chan = s3c_dma_lookup_channel(channel); diff --git a/arch/arm/plat-samsung/include/plat/dma.h b/arch/arm/plat-samsung/include/plat/dma.h index 2e8f8c6560d7..8c273b7a6f56 100644 --- a/arch/arm/plat-samsung/include/plat/dma.h +++ b/arch/arm/plat-samsung/include/plat/dma.h @@ -42,6 +42,7 @@ struct s3c2410_dma_client { }; struct s3c2410_dma_chan; +enum dma_ch; /* s3c2410_dma_cbfn_t * @@ -62,7 +63,7 @@ typedef int (*s3c2410_dma_opfn_t)(struct s3c2410_dma_chan *, * request a dma channel exclusivley */ -extern int s3c2410_dma_request(unsigned int channel, +extern int s3c2410_dma_request(enum dma_ch channel, struct s3c2410_dma_client *, void *dev); @@ -71,14 +72,14 @@ extern int s3c2410_dma_request(unsigned int channel, * change the state of the dma channel */ -extern int s3c2410_dma_ctrl(unsigned int channel, enum s3c2410_chan_op op); +extern int s3c2410_dma_ctrl(enum dma_ch channel, enum s3c2410_chan_op op); /* s3c2410_dma_setflags * * set the channel's flags to a given state */ -extern int s3c2410_dma_setflags(unsigned int channel, +extern int s3c2410_dma_setflags(enum dma_ch channel, unsigned int flags); /* s3c2410_dma_free @@ -86,7 +87,7 @@ extern int s3c2410_dma_setflags(unsigned int channel, * free the dma channel (will also abort any outstanding operations) */ -extern int s3c2410_dma_free(unsigned int channel, struct s3c2410_dma_client *); +extern int s3c2410_dma_free(enum dma_ch channel, struct s3c2410_dma_client *); /* s3c2410_dma_enqueue * @@ -95,7 +96,7 @@ extern int s3c2410_dma_free(unsigned int channel, struct s3c2410_dma_client *); * drained before the buffer is given to the DMA system. */ -extern int s3c2410_dma_enqueue(unsigned int channel, void *id, +extern int s3c2410_dma_enqueue(enum dma_ch channel, void *id, dma_addr_t data, int size); /* s3c2410_dma_config @@ -103,14 +104,14 @@ extern int s3c2410_dma_enqueue(unsigned int channel, void *id, * configure the dma channel */ -extern int s3c2410_dma_config(unsigned int channel, int xferunit); +extern int s3c2410_dma_config(enum dma_ch channel, int xferunit); /* s3c2410_dma_devconfig * * configure the device we're talking to */ -extern int s3c2410_dma_devconfig(unsigned int channel, +extern int s3c2410_dma_devconfig(enum dma_ch channel, enum s3c2410_dmasrc source, unsigned long devaddr); /* s3c2410_dma_getposition @@ -118,10 +119,10 @@ extern int s3c2410_dma_devconfig(unsigned int channel, * get the position that the dma transfer is currently at */ -extern int s3c2410_dma_getposition(unsigned int channel, +extern int s3c2410_dma_getposition(enum dma_ch channel, dma_addr_t *src, dma_addr_t *dest); -extern int s3c2410_dma_set_opfn(unsigned int, s3c2410_dma_opfn_t rtn); -extern int s3c2410_dma_set_buffdone_fn(unsigned int, s3c2410_dma_cbfn_t rtn); +extern int s3c2410_dma_set_opfn(enum dma_ch, s3c2410_dma_opfn_t rtn); +extern int s3c2410_dma_set_buffdone_fn(enum dma_ch, s3c2410_dma_cbfn_t rtn); diff --git a/arch/arm/plat-samsung/irq-uart.c b/arch/arm/plat-samsung/irq-uart.c index 0e46588d847b..657405c481d0 100644 --- a/arch/arm/plat-samsung/irq-uart.c +++ b/arch/arm/plat-samsung/irq-uart.c @@ -54,6 +54,13 @@ static void __init s3c_init_uart_irq(struct s3c_uart_irq *uirq) gc = irq_alloc_generic_chip("s3c-uart", 1, uirq->base_irq, reg_base, handle_level_irq); + + if (!gc) { + pr_err("%s: irq_alloc_generic_chip for IRQ %u failed\n", + __func__, uirq->base_irq); + return; + } + ct = gc->chip_types; ct->chip.irq_ack = irq_gc_ack_set_bit; ct->chip.irq_mask = irq_gc_mask_set_bit; diff --git a/arch/arm/plat-samsung/irq-vic-timer.c b/arch/arm/plat-samsung/irq-vic-timer.c index a607546ddbd0..f714d060370d 100644 --- a/arch/arm/plat-samsung/irq-vic-timer.c +++ b/arch/arm/plat-samsung/irq-vic-timer.c @@ -54,6 +54,13 @@ void __init s3c_init_vic_timer_irq(unsigned int num, unsigned int timer_irq) s3c_tgc = irq_alloc_generic_chip("s3c-timer", 1, timer_irq, S3C64XX_TINT_CSTAT, handle_level_irq); + + if (!s3c_tgc) { + pr_err("%s: irq_alloc_generic_chip for IRQ %d failed\n", + __func__, timer_irq); + return; + } + ct = s3c_tgc->chip_types; ct->chip.irq_mask = irq_gc_mask_clr_bit; ct->chip.irq_unmask = irq_gc_mask_set_bit; diff --git a/arch/mips/kernel/i8259.c b/arch/mips/kernel/i8259.c index c018696765d4..5c74eb797f08 100644 --- a/arch/mips/kernel/i8259.c +++ b/arch/mips/kernel/i8259.c @@ -14,7 +14,7 @@ #include <linux/interrupt.h> #include <linux/kernel.h> #include <linux/spinlock.h> -#include <linux/sysdev.h> +#include <linux/syscore_ops.h> #include <linux/irq.h> #include <asm/i8259.h> @@ -215,14 +215,13 @@ spurious_8259A_irq: } } -static int i8259A_resume(struct sys_device *dev) +static void i8259A_resume(void) { if (i8259A_auto_eoi >= 0) init_8259A(i8259A_auto_eoi); - return 0; } -static int i8259A_shutdown(struct sys_device *dev) +static void i8259A_shutdown(void) { /* Put the i8259A into a quiescent state that * the kernel initialization code can get it @@ -232,26 +231,17 @@ static int i8259A_shutdown(struct sys_device *dev) outb(0xff, PIC_MASTER_IMR); /* mask all of 8259A-1 */ outb(0xff, PIC_SLAVE_IMR); /* mask all of 8259A-1 */ } - return 0; } -static struct sysdev_class i8259_sysdev_class = { - .name = "i8259", +static struct syscore_ops i8259_syscore_ops = { .resume = i8259A_resume, .shutdown = i8259A_shutdown, }; -static struct sys_device device_i8259A = { - .id = 0, - .cls = &i8259_sysdev_class, -}; - static int __init i8259A_init_sysfs(void) { - int error = sysdev_class_register(&i8259_sysdev_class); - if (!error) - error = sysdev_register(&device_i8259A); - return error; + register_syscore_ops(&i8259_syscore_ops); + return 0; } device_initcall(i8259A_init_sysfs); diff --git a/arch/sparc/include/asm/irqflags_32.h b/arch/sparc/include/asm/irqflags_32.h index d4d0711de0f9..14848909e0de 100644 --- a/arch/sparc/include/asm/irqflags_32.h +++ b/arch/sparc/include/asm/irqflags_32.h @@ -18,7 +18,7 @@ extern void arch_local_irq_restore(unsigned long); extern unsigned long arch_local_irq_save(void); extern void arch_local_irq_enable(void); -static inline unsigned long arch_local_save_flags(void) +static inline notrace unsigned long arch_local_save_flags(void) { unsigned long flags; @@ -26,17 +26,17 @@ static inline unsigned long arch_local_save_flags(void) return flags; } -static inline void arch_local_irq_disable(void) +static inline notrace void arch_local_irq_disable(void) { arch_local_irq_save(); } -static inline bool arch_irqs_disabled_flags(unsigned long flags) +static inline notrace bool arch_irqs_disabled_flags(unsigned long flags) { return (flags & PSR_PIL) != 0; } -static inline bool arch_irqs_disabled(void) +static inline notrace bool arch_irqs_disabled(void) { return arch_irqs_disabled_flags(arch_local_save_flags()); } diff --git a/arch/sparc/include/asm/irqflags_64.h b/arch/sparc/include/asm/irqflags_64.h index aab969c82c2b..23cd27f6beb4 100644 --- a/arch/sparc/include/asm/irqflags_64.h +++ b/arch/sparc/include/asm/irqflags_64.h @@ -14,7 +14,7 @@ #ifndef __ASSEMBLY__ -static inline unsigned long arch_local_save_flags(void) +static inline notrace unsigned long arch_local_save_flags(void) { unsigned long flags; @@ -26,7 +26,7 @@ static inline unsigned long arch_local_save_flags(void) return flags; } -static inline void arch_local_irq_restore(unsigned long flags) +static inline notrace void arch_local_irq_restore(unsigned long flags) { __asm__ __volatile__( "wrpr %0, %%pil" @@ -36,7 +36,7 @@ static inline void arch_local_irq_restore(unsigned long flags) ); } -static inline void arch_local_irq_disable(void) +static inline notrace void arch_local_irq_disable(void) { __asm__ __volatile__( "wrpr %0, %%pil" @@ -46,7 +46,7 @@ static inline void arch_local_irq_disable(void) ); } -static inline void arch_local_irq_enable(void) +static inline notrace void arch_local_irq_enable(void) { __asm__ __volatile__( "wrpr 0, %%pil" @@ -56,17 +56,17 @@ static inline void arch_local_irq_enable(void) ); } -static inline int arch_irqs_disabled_flags(unsigned long flags) +static inline notrace int arch_irqs_disabled_flags(unsigned long flags) { return (flags > 0); } -static inline int arch_irqs_disabled(void) +static inline notrace int arch_irqs_disabled(void) { return arch_irqs_disabled_flags(arch_local_save_flags()); } -static inline unsigned long arch_local_irq_save(void) +static inline notrace unsigned long arch_local_irq_save(void) { unsigned long flags, tmp; diff --git a/arch/sparc/include/asm/ptrace.h b/arch/sparc/include/asm/ptrace.h index c7ad3fe2b252..b928b31424b1 100644 --- a/arch/sparc/include/asm/ptrace.h +++ b/arch/sparc/include/asm/ptrace.h @@ -205,6 +205,7 @@ do { current_thread_info()->syscall_noerror = 1; \ } while (0) #define user_mode(regs) (!((regs)->tstate & TSTATE_PRIV)) #define instruction_pointer(regs) ((regs)->tpc) +#define instruction_pointer_set(regs, val) ((regs)->tpc = (val)) #define user_stack_pointer(regs) ((regs)->u_regs[UREG_FP]) #define regs_return_value(regs) ((regs)->u_regs[UREG_I0]) #ifdef CONFIG_SMP diff --git a/arch/sparc/kernel/entry.S b/arch/sparc/kernel/entry.S index 9fe08a1ea6c6..f445e98463e6 100644 --- a/arch/sparc/kernel/entry.S +++ b/arch/sparc/kernel/entry.S @@ -293,7 +293,7 @@ maybe_smp4m_msg: WRITE_PAUSE wr %l4, PSR_ET, %psr WRITE_PAUSE - sll %o3, 28, %o2 ! shift for simpler checks below + srl %o3, 28, %o2 ! shift for simpler checks below maybe_smp4m_msg_check_single: andcc %o2, 0x1, %g0 beq,a maybe_smp4m_msg_check_mask diff --git a/arch/sparc/mm/leon_mm.c b/arch/sparc/mm/leon_mm.c index c0e01297e64e..e485a6804998 100644 --- a/arch/sparc/mm/leon_mm.c +++ b/arch/sparc/mm/leon_mm.c @@ -226,7 +226,7 @@ void leon3_getCacheRegs(struct leon3_cacheregs *regs) * Leon2 and Leon3 differ in their way of telling cache information * */ -int leon_flush_needed(void) +int __init leon_flush_needed(void) { int flush_needed = -1; unsigned int ssize, sets; diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index da349723d411..37357a599dca 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -1170,7 +1170,7 @@ comment "NUMA (Summit) requires SMP, 64GB highmem support, ACPI" config AMD_NUMA def_bool y prompt "Old style AMD Opteron NUMA detection" - depends on NUMA && PCI + depends on X86_64 && NUMA && PCI ---help--- Enable AMD NUMA node topology detection. You should say Y here if you have a multi processor AMD system. This uses an old method to diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 4f0d46fefa7f..9242436e9937 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -419,6 +419,30 @@ static struct dmi_system_id __initdata pci_reboot_dmi_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "iMac9,1"), }, }, + { /* Handle problems with rebooting on the Latitude E6320. */ + .callback = set_pci_reboot, + .ident = "Dell Latitude E6320", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "Latitude E6320"), + }, + }, + { /* Handle problems with rebooting on the Latitude E5420. */ + .callback = set_pci_reboot, + .ident = "Dell Latitude E5420", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "Latitude E5420"), + }, + }, + { /* Handle problems with rebooting on the Latitude E6420. */ + .callback = set_pci_reboot, + .ident = "Dell Latitude E6420", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "Latitude E6420"), + }, + }, { } }; diff --git a/drivers/acpi/apei/hest.c b/drivers/acpi/apei/hest.c index abda3786a5d7..181bc2f7bb74 100644 --- a/drivers/acpi/apei/hest.c +++ b/drivers/acpi/apei/hest.c @@ -139,13 +139,23 @@ static int __init hest_parse_ghes(struct acpi_hest_header *hest_hdr, void *data) { struct platform_device *ghes_dev; struct ghes_arr *ghes_arr = data; - int rc; + int rc, i; if (hest_hdr->type != ACPI_HEST_TYPE_GENERIC_ERROR) return 0; if (!((struct acpi_hest_generic *)hest_hdr)->enabled) return 0; + for (i = 0; i < ghes_arr->count; i++) { + struct acpi_hest_header *hdr; + ghes_dev = ghes_arr->ghes_devs[i]; + hdr = *(struct acpi_hest_header **)ghes_dev->dev.platform_data; + if (hdr->source_id == hest_hdr->source_id) { + pr_warning(FW_WARN HEST_PFX "Duplicated hardware error source ID: %d.\n", + hdr->source_id); + return -EIO; + } + } ghes_dev = platform_device_alloc("GHES", hest_hdr->source_id); if (!ghes_dev) return -ENOMEM; diff --git a/drivers/acpi/osl.c b/drivers/acpi/osl.c index 52ca9649d769..372f9b70f7f4 100644 --- a/drivers/acpi/osl.c +++ b/drivers/acpi/osl.c @@ -1333,23 +1333,6 @@ int acpi_resources_are_enforced(void) EXPORT_SYMBOL(acpi_resources_are_enforced); /* - * Create and initialize a spinlock. - */ -acpi_status -acpi_os_create_lock(acpi_spinlock *out_handle) -{ - spinlock_t *lock; - - lock = ACPI_ALLOCATE(sizeof(spinlock_t)); - if (!lock) - return AE_NO_MEMORY; - spin_lock_init(lock); - *out_handle = lock; - - return AE_OK; -} - -/* * Deallocate the memory for a spinlock. */ void acpi_os_delete_lock(acpi_spinlock handle) diff --git a/drivers/char/agp/intel-agp.h b/drivers/char/agp/intel-agp.h index 999803ce10dc..5da67f165afa 100644 --- a/drivers/char/agp/intel-agp.h +++ b/drivers/char/agp/intel-agp.h @@ -90,9 +90,10 @@ #define G4x_GMCH_SIZE_MASK (0xf << 8) #define G4x_GMCH_SIZE_1M (0x1 << 8) #define G4x_GMCH_SIZE_2M (0x3 << 8) -#define G4x_GMCH_SIZE_VT_1M (0x9 << 8) -#define G4x_GMCH_SIZE_VT_1_5M (0xa << 8) -#define G4x_GMCH_SIZE_VT_2M (0xc << 8) +#define G4x_GMCH_SIZE_VT_EN (0x8 << 8) +#define G4x_GMCH_SIZE_VT_1M (G4x_GMCH_SIZE_1M | G4x_GMCH_SIZE_VT_EN) +#define G4x_GMCH_SIZE_VT_1_5M ((0x2 << 8) | G4x_GMCH_SIZE_VT_EN) +#define G4x_GMCH_SIZE_VT_2M (G4x_GMCH_SIZE_2M | G4x_GMCH_SIZE_VT_EN) #define GFX_FLSH_CNTL 0x2170 /* 915+ */ diff --git a/drivers/gpio/wm831x-gpio.c b/drivers/gpio/wm831x-gpio.c index 309644cf4d9b..2bcfb0be09ff 100644 --- a/drivers/gpio/wm831x-gpio.c +++ b/drivers/gpio/wm831x-gpio.c @@ -180,6 +180,7 @@ static void wm831x_gpio_dbg_show(struct seq_file *s, struct gpio_chip *chip) break; case WM831X_GPIO_PULL_UP: pull = "pullup"; + break; default: pull = "INVALID PULL"; break; diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index f245c588ae95..ce7914c4c044 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -262,6 +262,7 @@ enum intel_pch { }; #define QUIRK_PIPEA_FORCE (1<<0) +#define QUIRK_LVDS_SSC_DISABLE (1<<1) struct intel_fbdev; @@ -1194,7 +1195,9 @@ void i915_gem_free_all_phys_object(struct drm_device *dev); void i915_gem_release(struct drm_device *dev, struct drm_file *file); uint32_t -i915_gem_get_unfenced_gtt_alignment(struct drm_i915_gem_object *obj); +i915_gem_get_unfenced_gtt_alignment(struct drm_device *dev, + uint32_t size, + int tiling_mode); /* i915_gem_gtt.c */ void i915_gem_restore_gtt_mappings(struct drm_device *dev); diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 5c0d1247f453..a087e1bf0c2f 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -1374,25 +1374,24 @@ i915_gem_free_mmap_offset(struct drm_i915_gem_object *obj) } static uint32_t -i915_gem_get_gtt_size(struct drm_i915_gem_object *obj) +i915_gem_get_gtt_size(struct drm_device *dev, uint32_t size, int tiling_mode) { - struct drm_device *dev = obj->base.dev; - uint32_t size; + uint32_t gtt_size; if (INTEL_INFO(dev)->gen >= 4 || - obj->tiling_mode == I915_TILING_NONE) - return obj->base.size; + tiling_mode == I915_TILING_NONE) + return size; /* Previous chips need a power-of-two fence region when tiling */ if (INTEL_INFO(dev)->gen == 3) - size = 1024*1024; + gtt_size = 1024*1024; else - size = 512*1024; + gtt_size = 512*1024; - while (size < obj->base.size) - size <<= 1; + while (gtt_size < size) + gtt_size <<= 1; - return size; + return gtt_size; } /** @@ -1403,59 +1402,52 @@ i915_gem_get_gtt_size(struct drm_i915_gem_object *obj) * potential fence register mapping. */ static uint32_t -i915_gem_get_gtt_alignment(struct drm_i915_gem_object *obj) +i915_gem_get_gtt_alignment(struct drm_device *dev, + uint32_t size, + int tiling_mode) { - struct drm_device *dev = obj->base.dev; - /* * Minimum alignment is 4k (GTT page size), but might be greater * if a fence register is needed for the object. */ if (INTEL_INFO(dev)->gen >= 4 || - obj->tiling_mode == I915_TILING_NONE) + tiling_mode == I915_TILING_NONE) return 4096; /* * Previous chips need to be aligned to the size of the smallest * fence register that can contain the object. */ - return i915_gem_get_gtt_size(obj); + return i915_gem_get_gtt_size(dev, size, tiling_mode); } /** * i915_gem_get_unfenced_gtt_alignment - return required GTT alignment for an * unfenced object - * @obj: object to check + * @dev: the device + * @size: size of the object + * @tiling_mode: tiling mode of the object * * Return the required GTT alignment for an object, only taking into account * unfenced tiled surface requirements. */ uint32_t -i915_gem_get_unfenced_gtt_alignment(struct drm_i915_gem_object *obj) +i915_gem_get_unfenced_gtt_alignment(struct drm_device *dev, + uint32_t size, + int tiling_mode) { - struct drm_device *dev = obj->base.dev; - int tile_height; - /* * Minimum alignment is 4k (GTT page size) for sane hw. */ if (INTEL_INFO(dev)->gen >= 4 || IS_G33(dev) || - obj->tiling_mode == I915_TILING_NONE) + tiling_mode == I915_TILING_NONE) return 4096; - /* - * Older chips need unfenced tiled buffers to be aligned to the left - * edge of an even tile row (where tile rows are counted as if the bo is - * placed in a fenced gtt region). + /* Previous hardware however needs to be aligned to a power-of-two + * tile height. The simplest method for determining this is to reuse + * the power-of-tile object size. */ - if (IS_GEN2(dev)) - tile_height = 16; - else if (obj->tiling_mode == I915_TILING_Y && HAS_128_BYTE_Y_TILING(dev)) - tile_height = 32; - else - tile_height = 8; - - return tile_height * obj->stride * 2; + return i915_gem_get_gtt_size(dev, size, tiling_mode); } int @@ -2744,9 +2736,16 @@ i915_gem_object_bind_to_gtt(struct drm_i915_gem_object *obj, return -EINVAL; } - fence_size = i915_gem_get_gtt_size(obj); - fence_alignment = i915_gem_get_gtt_alignment(obj); - unfenced_alignment = i915_gem_get_unfenced_gtt_alignment(obj); + fence_size = i915_gem_get_gtt_size(dev, + obj->base.size, + obj->tiling_mode); + fence_alignment = i915_gem_get_gtt_alignment(dev, + obj->base.size, + obj->tiling_mode); + unfenced_alignment = + i915_gem_get_unfenced_gtt_alignment(dev, + obj->base.size, + obj->tiling_mode); if (alignment == 0) alignment = map_and_fenceable ? fence_alignment : diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c index 82d70fd9e933..99c4faa59d8f 100644 --- a/drivers/gpu/drm/i915/i915_gem_tiling.c +++ b/drivers/gpu/drm/i915/i915_gem_tiling.c @@ -348,7 +348,9 @@ i915_gem_set_tiling(struct drm_device *dev, void *data, /* Rebind if we need a change of alignment */ if (!obj->map_and_fenceable) { u32 unfenced_alignment = - i915_gem_get_unfenced_gtt_alignment(obj); + i915_gem_get_unfenced_gtt_alignment(dev, + obj->base.size, + args->tiling_mode); if (obj->gtt_offset & (unfenced_alignment - 1)) ret = i915_gem_object_unbind(obj); } diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 21b6f93fe919..0f1c799afea1 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -4305,7 +4305,8 @@ static void intel_update_watermarks(struct drm_device *dev) static inline bool intel_panel_use_ssc(struct drm_i915_private *dev_priv) { - return dev_priv->lvds_use_ssc && i915_panel_use_ssc; + return dev_priv->lvds_use_ssc && i915_panel_use_ssc + && !(dev_priv->quirks & QUIRK_LVDS_SSC_DISABLE); } static int i9xx_crtc_mode_set(struct drm_crtc *crtc, @@ -7810,6 +7811,15 @@ static void quirk_pipea_force (struct drm_device *dev) DRM_DEBUG_DRIVER("applying pipe a force quirk\n"); } +/* + * Some machines (Lenovo U160) do not work with SSC on LVDS for some reason + */ +static void quirk_ssc_force_disable(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + dev_priv->quirks |= QUIRK_LVDS_SSC_DISABLE; +} + struct intel_quirk { int device; int subsystem_vendor; @@ -7838,6 +7848,9 @@ struct intel_quirk intel_quirks[] = { /* 855 & before need to leave pipe A & dpll A up */ { 0x3582, PCI_ANY_ID, PCI_ANY_ID, quirk_pipea_force }, { 0x2562, PCI_ANY_ID, PCI_ANY_ID, quirk_pipea_force }, + + /* Lenovo U160 cannot use SSC on LVDS */ + { 0x0046, 0x17aa, 0x3920, quirk_ssc_force_disable }, }; static void intel_init_quirks(struct drm_device *dev) diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index 660f96401a05..15bd0477a3e8 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -2000,7 +2000,7 @@ static void evergreen_gpu_init(struct radeon_device *rdev) gb_backend_map = 0x66442200; break; case CHIP_JUNIPER: - gb_backend_map = 0x00006420; + gb_backend_map = 0x00002200; break; default: gb_backend_map = diff --git a/drivers/gpu/drm/radeon/evergreen_blit_kms.c b/drivers/gpu/drm/radeon/evergreen_blit_kms.c index 57f3bc17b87e..2eb251858e72 100644 --- a/drivers/gpu/drm/radeon/evergreen_blit_kms.c +++ b/drivers/gpu/drm/radeon/evergreen_blit_kms.c @@ -252,7 +252,7 @@ draw_auto(struct radeon_device *rdev) } -/* emits 36 */ +/* emits 39 */ static void set_default_state(struct radeon_device *rdev) { @@ -531,6 +531,11 @@ set_default_state(struct radeon_device *rdev) radeon_ring_write(rdev, (SQ_DYN_GPR_CNTL_PS_FLUSH_REQ - PACKET3_SET_CONFIG_REG_START) >> 2); radeon_ring_write(rdev, 0); + /* setup LDS */ + radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 1)); + radeon_ring_write(rdev, (SQ_LDS_RESOURCE_MGMT - PACKET3_SET_CONFIG_REG_START) >> 2); + radeon_ring_write(rdev, 0x10001000); + /* SQ config */ radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONFIG_REG, 11)); radeon_ring_write(rdev, (SQ_CONFIG - PACKET3_SET_CONFIG_REG_START) >> 2); @@ -773,7 +778,7 @@ int evergreen_blit_prepare_copy(struct radeon_device *rdev, int size_bytes) /* calculate number of loops correctly */ ring_size = num_loops * dwords_per_loop; /* set default + shaders */ - ring_size += 52; /* shaders + def state */ + ring_size += 55; /* shaders + def state */ ring_size += 10; /* fence emit for VB IB */ ring_size += 5; /* done copy */ ring_size += 10; /* fence emit for done copy */ diff --git a/drivers/gpu/drm/radeon/radeon_bios.c b/drivers/gpu/drm/radeon/radeon_bios.c index 3fc5fa1aefd0..229a20f10e2b 100644 --- a/drivers/gpu/drm/radeon/radeon_bios.c +++ b/drivers/gpu/drm/radeon/radeon_bios.c @@ -331,7 +331,7 @@ static bool avivo_read_disabled_bios(struct radeon_device *rdev) seprom_cntl1 = RREG32(RADEON_SEPROM_CNTL1); viph_control = RREG32(RADEON_VIPH_CONTROL); - bus_cntl = RREG32(RADEON_BUS_CNTL); + bus_cntl = RREG32(RV370_BUS_CNTL); d1vga_control = RREG32(AVIVO_D1VGA_CONTROL); d2vga_control = RREG32(AVIVO_D2VGA_CONTROL); vga_render_control = RREG32(AVIVO_VGA_RENDER_CONTROL); @@ -350,7 +350,7 @@ static bool avivo_read_disabled_bios(struct radeon_device *rdev) WREG32(RADEON_VIPH_CONTROL, (viph_control & ~RADEON_VIPH_EN)); /* enable the rom */ - WREG32(RADEON_BUS_CNTL, (bus_cntl & ~RADEON_BUS_BIOS_DIS_ROM)); + WREG32(RV370_BUS_CNTL, (bus_cntl & ~RV370_BUS_BIOS_DIS_ROM)); /* Disable VGA mode */ WREG32(AVIVO_D1VGA_CONTROL, @@ -367,7 +367,7 @@ static bool avivo_read_disabled_bios(struct radeon_device *rdev) /* restore regs */ WREG32(RADEON_SEPROM_CNTL1, seprom_cntl1); WREG32(RADEON_VIPH_CONTROL, viph_control); - WREG32(RADEON_BUS_CNTL, bus_cntl); + WREG32(RV370_BUS_CNTL, bus_cntl); WREG32(AVIVO_D1VGA_CONTROL, d1vga_control); WREG32(AVIVO_D2VGA_CONTROL, d2vga_control); WREG32(AVIVO_VGA_RENDER_CONTROL, vga_render_control); @@ -390,7 +390,10 @@ static bool legacy_read_disabled_bios(struct radeon_device *rdev) seprom_cntl1 = RREG32(RADEON_SEPROM_CNTL1); viph_control = RREG32(RADEON_VIPH_CONTROL); - bus_cntl = RREG32(RADEON_BUS_CNTL); + if (rdev->flags & RADEON_IS_PCIE) + bus_cntl = RREG32(RV370_BUS_CNTL); + else + bus_cntl = RREG32(RADEON_BUS_CNTL); crtc_gen_cntl = RREG32(RADEON_CRTC_GEN_CNTL); crtc2_gen_cntl = 0; crtc_ext_cntl = RREG32(RADEON_CRTC_EXT_CNTL); @@ -412,7 +415,10 @@ static bool legacy_read_disabled_bios(struct radeon_device *rdev) WREG32(RADEON_VIPH_CONTROL, (viph_control & ~RADEON_VIPH_EN)); /* enable the rom */ - WREG32(RADEON_BUS_CNTL, (bus_cntl & ~RADEON_BUS_BIOS_DIS_ROM)); + if (rdev->flags & RADEON_IS_PCIE) + WREG32(RV370_BUS_CNTL, (bus_cntl & ~RV370_BUS_BIOS_DIS_ROM)); + else + WREG32(RADEON_BUS_CNTL, (bus_cntl & ~RADEON_BUS_BIOS_DIS_ROM)); /* Turn off mem requests and CRTC for both controllers */ WREG32(RADEON_CRTC_GEN_CNTL, @@ -439,7 +445,10 @@ static bool legacy_read_disabled_bios(struct radeon_device *rdev) /* restore regs */ WREG32(RADEON_SEPROM_CNTL1, seprom_cntl1); WREG32(RADEON_VIPH_CONTROL, viph_control); - WREG32(RADEON_BUS_CNTL, bus_cntl); + if (rdev->flags & RADEON_IS_PCIE) + WREG32(RV370_BUS_CNTL, bus_cntl); + else + WREG32(RADEON_BUS_CNTL, bus_cntl); WREG32(RADEON_CRTC_GEN_CNTL, crtc_gen_cntl); if (!(rdev->flags & RADEON_SINGLE_CRTC)) { WREG32(RADEON_CRTC2_GEN_CNTL, crtc2_gen_cntl); diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index cbfca3a24fdf..9792d4ffdc86 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -52,6 +52,12 @@ void radeon_connector_hotplug(struct drm_connector *connector) struct radeon_device *rdev = dev->dev_private; struct radeon_connector *radeon_connector = to_radeon_connector(connector); + /* bail if the connector does not have hpd pin, e.g., + * VGA, TV, etc. + */ + if (radeon_connector->hpd.hpd == RADEON_HPD_NONE) + return; + radeon_hpd_set_polarity(rdev, radeon_connector->hpd.hpd); /* powering up/down the eDP panel generates hpd events which diff --git a/drivers/gpu/drm/radeon/radeon_reg.h b/drivers/gpu/drm/radeon/radeon_reg.h index ec93a75369e6..bc44a3d35ec6 100644 --- a/drivers/gpu/drm/radeon/radeon_reg.h +++ b/drivers/gpu/drm/radeon/radeon_reg.h @@ -300,6 +300,8 @@ # define RADEON_BUS_READ_BURST (1 << 30) #define RADEON_BUS_CNTL1 0x0034 # define RADEON_BUS_WAIT_ON_LOCK_EN (1 << 4) +#define RV370_BUS_CNTL 0x004c +# define RV370_BUS_BIOS_DIS_ROM (1 << 2) /* rv370/rv380, rv410, r423/r430/r480, r5xx */ #define RADEON_MSI_REARM_EN 0x0160 # define RV370_MSI_REARM_EN (1 << 0) diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c index 6e3b11e5abbe..1f5850e473cc 100644 --- a/drivers/gpu/drm/radeon/rs600.c +++ b/drivers/gpu/drm/radeon/rs600.c @@ -426,7 +426,7 @@ int rs600_gart_init(struct radeon_device *rdev) return radeon_gart_table_vram_alloc(rdev); } -int rs600_gart_enable(struct radeon_device *rdev) +static int rs600_gart_enable(struct radeon_device *rdev) { u32 tmp; int r, i; @@ -440,8 +440,8 @@ int rs600_gart_enable(struct radeon_device *rdev) return r; radeon_gart_restore(rdev); /* Enable bus master */ - tmp = RREG32(R_00004C_BUS_CNTL) & C_00004C_BUS_MASTER_DIS; - WREG32(R_00004C_BUS_CNTL, tmp); + tmp = RREG32(RADEON_BUS_CNTL) & ~RS600_BUS_MASTER_DIS; + WREG32(RADEON_BUS_CNTL, tmp); /* FIXME: setup default page */ WREG32_MC(R_000100_MC_PT0_CNTL, (S_000100_EFFECTIVE_L2_CACHE_SIZE(6) | diff --git a/drivers/hwmon/adm1275.c b/drivers/hwmon/adm1275.c index b9b7caf4a1d2..8bc1bd663721 100644 --- a/drivers/hwmon/adm1275.c +++ b/drivers/hwmon/adm1275.c @@ -53,23 +53,23 @@ static int adm1275_probe(struct i2c_client *client, info->direct[PSC_VOLTAGE_IN] = true; info->direct[PSC_VOLTAGE_OUT] = true; info->direct[PSC_CURRENT_OUT] = true; - info->m[PSC_CURRENT_OUT] = 800; + info->m[PSC_CURRENT_OUT] = 807; info->b[PSC_CURRENT_OUT] = 20475; info->R[PSC_CURRENT_OUT] = -1; info->func[0] = PMBUS_HAVE_IOUT | PMBUS_HAVE_STATUS_IOUT; if (config & ADM1275_VRANGE) { - info->m[PSC_VOLTAGE_IN] = 19045; + info->m[PSC_VOLTAGE_IN] = 19199; info->b[PSC_VOLTAGE_IN] = 0; info->R[PSC_VOLTAGE_IN] = -2; - info->m[PSC_VOLTAGE_OUT] = 19045; + info->m[PSC_VOLTAGE_OUT] = 19199; info->b[PSC_VOLTAGE_OUT] = 0; info->R[PSC_VOLTAGE_OUT] = -2; } else { - info->m[PSC_VOLTAGE_IN] = 6666; + info->m[PSC_VOLTAGE_IN] = 6720; info->b[PSC_VOLTAGE_IN] = 0; info->R[PSC_VOLTAGE_IN] = -1; - info->m[PSC_VOLTAGE_OUT] = 6666; + info->m[PSC_VOLTAGE_OUT] = 6720; info->b[PSC_VOLTAGE_OUT] = 0; info->R[PSC_VOLTAGE_OUT] = -1; } diff --git a/drivers/hwmon/asus_atk0110.c b/drivers/hwmon/asus_atk0110.c index dcb78a7a8047..00e98517f94c 100644 --- a/drivers/hwmon/asus_atk0110.c +++ b/drivers/hwmon/asus_atk0110.c @@ -674,6 +674,7 @@ static int atk_debugfs_gitm_get(void *p, u64 *val) else err = -EIO; + ACPI_FREE(ret); return err; } diff --git a/drivers/hwmon/it87.c b/drivers/hwmon/it87.c index bb6405b92007..5f5247750430 100644 --- a/drivers/hwmon/it87.c +++ b/drivers/hwmon/it87.c @@ -1538,7 +1538,7 @@ static struct attribute *it87_attributes_label[] = { }; static const struct attribute_group it87_group_label = { - .attrs = it87_attributes_vid, + .attrs = it87_attributes_label, }; /* SuperIO detection - will change isa_address if a chip is found */ diff --git a/drivers/hwmon/max1111.c b/drivers/hwmon/max1111.c index 12a54aa29776..14335bbc9bdc 100644 --- a/drivers/hwmon/max1111.c +++ b/drivers/hwmon/max1111.c @@ -40,6 +40,8 @@ struct max1111_data { struct spi_transfer xfer[2]; uint8_t *tx_buf; uint8_t *rx_buf; + struct mutex drvdata_lock; + /* protect msg, xfer and buffers from multiple access */ }; static int max1111_read(struct device *dev, int channel) @@ -48,6 +50,9 @@ static int max1111_read(struct device *dev, int channel) uint8_t v1, v2; int err; + /* writing to drvdata struct is not thread safe, wait on mutex */ + mutex_lock(&data->drvdata_lock); + data->tx_buf[0] = (channel << MAX1111_CTRL_SEL_SH) | MAX1111_CTRL_PD0 | MAX1111_CTRL_PD1 | MAX1111_CTRL_SGL | MAX1111_CTRL_UNI | MAX1111_CTRL_STR; @@ -55,12 +60,15 @@ static int max1111_read(struct device *dev, int channel) err = spi_sync(data->spi, &data->msg); if (err < 0) { dev_err(dev, "spi_sync failed with %d\n", err); + mutex_unlock(&data->drvdata_lock); return err; } v1 = data->rx_buf[0]; v2 = data->rx_buf[1]; + mutex_unlock(&data->drvdata_lock); + if ((v1 & 0xc0) || (v2 & 0x3f)) return -EINVAL; @@ -176,6 +184,8 @@ static int __devinit max1111_probe(struct spi_device *spi) if (err) goto err_free_data; + mutex_init(&data->drvdata_lock); + data->spi = spi; spi_set_drvdata(spi, data); @@ -213,6 +223,7 @@ static int __devexit max1111_remove(struct spi_device *spi) hwmon_device_unregister(data->hwmon_dev); sysfs_remove_group(&spi->dev.kobj, &max1111_attr_group); + mutex_destroy(&data->drvdata_lock); kfree(data->rx_buf); kfree(data->tx_buf); kfree(data); diff --git a/drivers/hwmon/pmbus_core.c b/drivers/hwmon/pmbus_core.c index 744672c1f26d..8e31a8e2c746 100644 --- a/drivers/hwmon/pmbus_core.c +++ b/drivers/hwmon/pmbus_core.c @@ -362,8 +362,8 @@ static struct pmbus_data *pmbus_update_device(struct device *dev) * Convert linear sensor values to milli- or micro-units * depending on sensor type. */ -static int pmbus_reg2data_linear(struct pmbus_data *data, - struct pmbus_sensor *sensor) +static long pmbus_reg2data_linear(struct pmbus_data *data, + struct pmbus_sensor *sensor) { s16 exponent; s32 mantissa; @@ -397,15 +397,15 @@ static int pmbus_reg2data_linear(struct pmbus_data *data, else val >>= -exponent; - return (int)val; + return val; } /* * Convert direct sensor values to milli- or micro-units * depending on sensor type. */ -static int pmbus_reg2data_direct(struct pmbus_data *data, - struct pmbus_sensor *sensor) +static long pmbus_reg2data_direct(struct pmbus_data *data, + struct pmbus_sensor *sensor) { long val = (s16) sensor->data; long m, b, R; @@ -440,12 +440,12 @@ static int pmbus_reg2data_direct(struct pmbus_data *data, R++; } - return (int)((val - b) / m); + return (val - b) / m; } -static int pmbus_reg2data(struct pmbus_data *data, struct pmbus_sensor *sensor) +static long pmbus_reg2data(struct pmbus_data *data, struct pmbus_sensor *sensor) { - int val; + long val; if (data->info->direct[sensor->class]) val = pmbus_reg2data_direct(data, sensor); @@ -619,7 +619,7 @@ static int pmbus_get_boolean(struct pmbus_data *data, int index, int *val) if (!s1 && !s2) *val = !!regval; else { - int v1, v2; + long v1, v2; struct pmbus_sensor *sensor1, *sensor2; sensor1 = &data->sensors[s1]; @@ -661,7 +661,7 @@ static ssize_t pmbus_show_sensor(struct device *dev, if (sensor->data < 0) return sensor->data; - return snprintf(buf, PAGE_SIZE, "%d\n", pmbus_reg2data(data, sensor)); + return snprintf(buf, PAGE_SIZE, "%ld\n", pmbus_reg2data(data, sensor)); } static ssize_t pmbus_set_sensor(struct device *dev, diff --git a/drivers/media/dvb/dvb-core/dvb_frontend.c b/drivers/media/dvb/dvb-core/dvb_frontend.c index 98278041d75f..5b6b451d4694 100644 --- a/drivers/media/dvb/dvb-core/dvb_frontend.c +++ b/drivers/media/dvb/dvb-core/dvb_frontend.c @@ -1988,6 +1988,14 @@ static int dvb_frontend_open(struct inode *inode, struct file *file) if (dvbdev->users == -1 && fe->ops.ts_bus_ctrl) { if ((ret = fe->ops.ts_bus_ctrl(fe, 1)) < 0) goto err0; + + /* If we took control of the bus, we need to force + reinitialization. This is because many ts_bus_ctrl() + functions strobe the RESET pin on the demod, and if the + frontend thread already exists then the dvb_init() routine + won't get called (which is what usually does initial + register configuration). */ + fepriv->reinitialise = 1; } if ((ret = dvb_generic_open (inode, file)) < 0) diff --git a/drivers/media/radio/Kconfig b/drivers/media/radio/Kconfig index e4c97fd6f05a..52798a111e16 100644 --- a/drivers/media/radio/Kconfig +++ b/drivers/media/radio/Kconfig @@ -168,7 +168,7 @@ config RADIO_MAXIRADIO config RADIO_MIROPCM20 tristate "miroSOUND PCM20 radio" - depends on ISA && VIDEO_V4L2 && SND + depends on ISA && ISA_DMA_API && VIDEO_V4L2 && SND select SND_ISA select SND_MIRO ---help--- @@ -201,7 +201,7 @@ config RADIO_SF16FMI config RADIO_SF16FMR2 tristate "SF16FMR2 Radio" - depends on ISA && VIDEO_V4L2 + depends on ISA && VIDEO_V4L2 && SND ---help--- Choose Y here if you have one of these FM radio cards. diff --git a/drivers/media/radio/si4713-i2c.c b/drivers/media/radio/si4713-i2c.c index deca2e06ff22..c9f4a8e65dc4 100644 --- a/drivers/media/radio/si4713-i2c.c +++ b/drivers/media/radio/si4713-i2c.c @@ -1033,7 +1033,7 @@ static int si4713_write_econtrol_string(struct si4713_device *sdev, char ps_name[MAX_RDS_PS_NAME + 1]; len = control->size - 1; - if (len > MAX_RDS_PS_NAME) { + if (len < 0 || len > MAX_RDS_PS_NAME) { rval = -ERANGE; goto exit; } @@ -1057,7 +1057,7 @@ static int si4713_write_econtrol_string(struct si4713_device *sdev, char radio_text[MAX_RDS_RADIO_TEXT + 1]; len = control->size - 1; - if (len > MAX_RDS_RADIO_TEXT) { + if (len < 0 || len > MAX_RDS_RADIO_TEXT) { rval = -ERANGE; goto exit; } diff --git a/drivers/media/rc/mceusb.c b/drivers/media/rc/mceusb.c index 06dfe0957b5e..ec972dc25790 100644 --- a/drivers/media/rc/mceusb.c +++ b/drivers/media/rc/mceusb.c @@ -558,9 +558,10 @@ static void mceusb_dev_printdata(struct mceusb_dev *ir, char *buf, inout, data1); break; case MCE_CMD_S_TIMEOUT: - /* value is in units of 50us, so x*50/100 or x/2 ms */ + /* value is in units of 50us, so x*50/1000 ms */ dev_info(dev, "%s receive timeout of %d ms\n", - inout, ((data1 << 8) | data2) / 2); + inout, + ((data1 << 8) | data2) * MCE_TIME_UNIT / 1000); break; case MCE_CMD_G_TIMEOUT: dev_info(dev, "Get receive timeout\n"); @@ -847,7 +848,7 @@ static void mceusb_handle_command(struct mceusb_dev *ir, int index) switch (ir->buf_in[index]) { /* 2-byte return value commands */ case MCE_CMD_S_TIMEOUT: - ir->rc->timeout = US_TO_NS((hi << 8 | lo) / 2); + ir->rc->timeout = US_TO_NS((hi << 8 | lo) * MCE_TIME_UNIT); break; /* 1-byte return value commands */ @@ -1078,7 +1079,7 @@ static struct rc_dev *mceusb_init_rc_dev(struct mceusb_dev *ir) rc->priv = ir; rc->driver_type = RC_DRIVER_IR_RAW; rc->allowed_protos = RC_TYPE_ALL; - rc->timeout = US_TO_NS(1000); + rc->timeout = MS_TO_NS(100); if (!ir->flags.no_tx) { rc->s_tx_mask = mceusb_set_tx_mask; rc->s_tx_carrier = mceusb_set_tx_carrier; diff --git a/drivers/media/rc/nuvoton-cir.c b/drivers/media/rc/nuvoton-cir.c index 565f24c20d77..ce595f9ab4c7 100644 --- a/drivers/media/rc/nuvoton-cir.c +++ b/drivers/media/rc/nuvoton-cir.c @@ -1110,7 +1110,7 @@ static int nvt_probe(struct pnp_dev *pdev, const struct pnp_device_id *dev_id) rdev->dev.parent = &pdev->dev; rdev->driver_name = NVT_DRIVER_NAME; rdev->map_name = RC_MAP_RC6_MCE; - rdev->timeout = US_TO_NS(1000); + rdev->timeout = MS_TO_NS(100); /* rx resolution is hardwired to 50us atm, 1, 25, 100 also possible */ rdev->rx_resolution = US_TO_NS(CIR_SAMPLE_PERIOD); #if 0 diff --git a/drivers/media/video/cx23885/cx23885-core.c b/drivers/media/video/cx23885/cx23885-core.c index 64d9b2136ff6..419777a832ee 100644 --- a/drivers/media/video/cx23885/cx23885-core.c +++ b/drivers/media/video/cx23885/cx23885-core.c @@ -2060,12 +2060,8 @@ static int __devinit cx23885_initdev(struct pci_dev *pci_dev, goto fail_irq; } - if (!pci_enable_msi(pci_dev)) - err = request_irq(pci_dev->irq, cx23885_irq, - IRQF_DISABLED, dev->name, dev); - else - err = request_irq(pci_dev->irq, cx23885_irq, - IRQF_SHARED | IRQF_DISABLED, dev->name, dev); + err = request_irq(pci_dev->irq, cx23885_irq, + IRQF_SHARED | IRQF_DISABLED, dev->name, dev); if (err < 0) { printk(KERN_ERR "%s: can't get IRQ %d\n", dev->name, pci_dev->irq); @@ -2114,7 +2110,6 @@ static void __devexit cx23885_finidev(struct pci_dev *pci_dev) /* unregister stuff */ free_irq(pci_dev->irq, dev); - pci_disable_msi(pci_dev); cx23885_dev_unregister(dev); v4l2_device_unregister(v4l2_dev); diff --git a/drivers/media/video/tuner-core.c b/drivers/media/video/tuner-core.c index cfa9f7efe93d..a03945ab9f08 100644 --- a/drivers/media/video/tuner-core.c +++ b/drivers/media/video/tuner-core.c @@ -714,10 +714,19 @@ static int tuner_remove(struct i2c_client *client) * returns 0. * This function is needed for boards that have a separate tuner for * radio (like devices with tea5767). + * NOTE: mt20xx uses V4L2_TUNER_DIGITAL_TV and calls set_tv_freq to + * select a TV frequency. So, t_mode = T_ANALOG_TV could actually + * be used to represent a Digital TV too. */ static inline int check_mode(struct tuner *t, enum v4l2_tuner_type mode) { - if ((1 << mode & t->mode_mask) == 0) + int t_mode; + if (mode == V4L2_TUNER_RADIO) + t_mode = T_RADIO; + else + t_mode = T_ANALOG_TV; + + if ((t_mode & t->mode_mask) == 0) return -EINVAL; return 0; @@ -984,7 +993,7 @@ static void tuner_status(struct dvb_frontend *fe) case V4L2_TUNER_RADIO: p = "radio"; break; - case V4L2_TUNER_DIGITAL_TV: + case V4L2_TUNER_DIGITAL_TV: /* Used by mt20xx */ p = "digital TV"; break; case V4L2_TUNER_ANALOG_TV: @@ -1135,9 +1144,8 @@ static int tuner_g_tuner(struct v4l2_subdev *sd, struct v4l2_tuner *vt) return 0; if (vt->type == t->mode && analog_ops->get_afc) vt->afc = analog_ops->get_afc(&t->fe); - if (vt->type == V4L2_TUNER_ANALOG_TV) + if (t->mode != V4L2_TUNER_RADIO) { vt->capability |= V4L2_TUNER_CAP_NORM; - if (vt->type != V4L2_TUNER_RADIO) { vt->rangelow = tv_range[0] * 16; vt->rangehigh = tv_range[1] * 16; return 0; diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c index 2a7e43bc796d..aa7d1d79b8c5 100644 --- a/drivers/mmc/core/mmc.c +++ b/drivers/mmc/core/mmc.c @@ -247,12 +247,12 @@ static int mmc_read_ext_csd(struct mmc_card *card, u8 *ext_csd) return 0; /* Version is coded in the CSD_STRUCTURE byte in the EXT_CSD register */ + card->ext_csd.raw_ext_csd_structure = ext_csd[EXT_CSD_STRUCTURE]; if (card->csd.structure == 3) { - int ext_csd_struct = ext_csd[EXT_CSD_STRUCTURE]; - if (ext_csd_struct > 2) { + if (card->ext_csd.raw_ext_csd_structure > 2) { printk(KERN_ERR "%s: unrecognised EXT_CSD structure " "version %d\n", mmc_hostname(card->host), - ext_csd_struct); + card->ext_csd.raw_ext_csd_structure); err = -EINVAL; goto out; } @@ -266,6 +266,10 @@ static int mmc_read_ext_csd(struct mmc_card *card, u8 *ext_csd) goto out; } + card->ext_csd.raw_sectors[0] = ext_csd[EXT_CSD_SEC_CNT + 0]; + card->ext_csd.raw_sectors[1] = ext_csd[EXT_CSD_SEC_CNT + 1]; + card->ext_csd.raw_sectors[2] = ext_csd[EXT_CSD_SEC_CNT + 2]; + card->ext_csd.raw_sectors[3] = ext_csd[EXT_CSD_SEC_CNT + 3]; if (card->ext_csd.rev >= 2) { card->ext_csd.sectors = ext_csd[EXT_CSD_SEC_CNT + 0] << 0 | @@ -277,7 +281,7 @@ static int mmc_read_ext_csd(struct mmc_card *card, u8 *ext_csd) if (card->ext_csd.sectors > (2u * 1024 * 1024 * 1024) / 512) mmc_card_set_blockaddr(card); } - + card->ext_csd.raw_card_type = ext_csd[EXT_CSD_CARD_TYPE]; switch (ext_csd[EXT_CSD_CARD_TYPE] & EXT_CSD_CARD_TYPE_MASK) { case EXT_CSD_CARD_TYPE_DDR_52 | EXT_CSD_CARD_TYPE_52 | EXT_CSD_CARD_TYPE_26: @@ -307,6 +311,11 @@ static int mmc_read_ext_csd(struct mmc_card *card, u8 *ext_csd) mmc_hostname(card->host)); } + card->ext_csd.raw_s_a_timeout = ext_csd[EXT_CSD_S_A_TIMEOUT]; + card->ext_csd.raw_erase_timeout_mult = + ext_csd[EXT_CSD_ERASE_TIMEOUT_MULT]; + card->ext_csd.raw_hc_erase_grp_size = + ext_csd[EXT_CSD_HC_ERASE_GRP_SIZE]; if (card->ext_csd.rev >= 3) { u8 sa_shift = ext_csd[EXT_CSD_S_A_TIMEOUT]; card->ext_csd.part_config = ext_csd[EXT_CSD_PART_CONFIG]; @@ -334,6 +343,16 @@ static int mmc_read_ext_csd(struct mmc_card *card, u8 *ext_csd) card->ext_csd.boot_size = ext_csd[EXT_CSD_BOOT_MULT] << 17; } + card->ext_csd.raw_hc_erase_gap_size = + ext_csd[EXT_CSD_PARTITION_ATTRIBUTE]; + card->ext_csd.raw_sec_trim_mult = + ext_csd[EXT_CSD_SEC_TRIM_MULT]; + card->ext_csd.raw_sec_erase_mult = + ext_csd[EXT_CSD_SEC_ERASE_MULT]; + card->ext_csd.raw_sec_feature_support = + ext_csd[EXT_CSD_SEC_FEATURE_SUPPORT]; + card->ext_csd.raw_trim_mult = + ext_csd[EXT_CSD_TRIM_MULT]; if (card->ext_csd.rev >= 4) { /* * Enhanced area feature support -- check whether the eMMC @@ -341,7 +360,7 @@ static int mmc_read_ext_csd(struct mmc_card *card, u8 *ext_csd) * area offset and size to user by adding sysfs interface. */ if ((ext_csd[EXT_CSD_PARTITION_SUPPORT] & 0x2) && - (ext_csd[EXT_CSD_PARTITION_ATTRIBUTE] & 0x1)) { + (ext_csd[EXT_CSD_PARTITION_ATTRIBUTE] & 0x1)) { u8 hc_erase_grp_sz = ext_csd[EXT_CSD_HC_ERASE_GRP_SIZE]; u8 hc_wp_grp_sz = @@ -401,17 +420,17 @@ static inline void mmc_free_ext_csd(u8 *ext_csd) } -static int mmc_compare_ext_csds(struct mmc_card *card, u8 *ext_csd, - unsigned bus_width) +static int mmc_compare_ext_csds(struct mmc_card *card, unsigned bus_width) { u8 *bw_ext_csd; int err; + if (bus_width == MMC_BUS_WIDTH_1) + return 0; + err = mmc_get_ext_csd(card, &bw_ext_csd); - if (err) - return err; - if ((ext_csd == NULL || bw_ext_csd == NULL)) { + if (err || bw_ext_csd == NULL) { if (bus_width != MMC_BUS_WIDTH_1) err = -EINVAL; goto out; @@ -421,35 +440,40 @@ static int mmc_compare_ext_csds(struct mmc_card *card, u8 *ext_csd, goto out; /* only compare read only fields */ - err = (!(ext_csd[EXT_CSD_PARTITION_SUPPORT] == + err = (!(card->ext_csd.raw_partition_support == bw_ext_csd[EXT_CSD_PARTITION_SUPPORT]) && - (ext_csd[EXT_CSD_ERASED_MEM_CONT] == + (card->ext_csd.raw_erased_mem_count == bw_ext_csd[EXT_CSD_ERASED_MEM_CONT]) && - (ext_csd[EXT_CSD_REV] == + (card->ext_csd.rev == bw_ext_csd[EXT_CSD_REV]) && - (ext_csd[EXT_CSD_STRUCTURE] == + (card->ext_csd.raw_ext_csd_structure == bw_ext_csd[EXT_CSD_STRUCTURE]) && - (ext_csd[EXT_CSD_CARD_TYPE] == + (card->ext_csd.raw_card_type == bw_ext_csd[EXT_CSD_CARD_TYPE]) && - (ext_csd[EXT_CSD_S_A_TIMEOUT] == + (card->ext_csd.raw_s_a_timeout == bw_ext_csd[EXT_CSD_S_A_TIMEOUT]) && - (ext_csd[EXT_CSD_HC_WP_GRP_SIZE] == + (card->ext_csd.raw_hc_erase_gap_size == bw_ext_csd[EXT_CSD_HC_WP_GRP_SIZE]) && - (ext_csd[EXT_CSD_ERASE_TIMEOUT_MULT] == + (card->ext_csd.raw_erase_timeout_mult == bw_ext_csd[EXT_CSD_ERASE_TIMEOUT_MULT]) && - (ext_csd[EXT_CSD_HC_ERASE_GRP_SIZE] == + (card->ext_csd.raw_hc_erase_grp_size == bw_ext_csd[EXT_CSD_HC_ERASE_GRP_SIZE]) && - (ext_csd[EXT_CSD_SEC_TRIM_MULT] == + (card->ext_csd.raw_sec_trim_mult == bw_ext_csd[EXT_CSD_SEC_TRIM_MULT]) && - (ext_csd[EXT_CSD_SEC_ERASE_MULT] == + (card->ext_csd.raw_sec_erase_mult == bw_ext_csd[EXT_CSD_SEC_ERASE_MULT]) && - (ext_csd[EXT_CSD_SEC_FEATURE_SUPPORT] == + (card->ext_csd.raw_sec_feature_support == bw_ext_csd[EXT_CSD_SEC_FEATURE_SUPPORT]) && - (ext_csd[EXT_CSD_TRIM_MULT] == + (card->ext_csd.raw_trim_mult == bw_ext_csd[EXT_CSD_TRIM_MULT]) && - memcmp(&ext_csd[EXT_CSD_SEC_CNT], - &bw_ext_csd[EXT_CSD_SEC_CNT], - 4) != 0); + (card->ext_csd.raw_sectors[0] == + bw_ext_csd[EXT_CSD_SEC_CNT + 0]) && + (card->ext_csd.raw_sectors[1] == + bw_ext_csd[EXT_CSD_SEC_CNT + 1]) && + (card->ext_csd.raw_sectors[2] == + bw_ext_csd[EXT_CSD_SEC_CNT + 2]) && + (card->ext_csd.raw_sectors[3] == + bw_ext_csd[EXT_CSD_SEC_CNT + 3])); if (err) err = -EINVAL; @@ -770,7 +794,6 @@ static int mmc_init_card(struct mmc_host *host, u32 ocr, */ if (!(host->caps & MMC_CAP_BUS_WIDTH_TEST)) err = mmc_compare_ext_csds(card, - ext_csd, bus_width); else err = mmc_bus_test(card, bus_width); diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index eafe44a528ac..63c22b0bb5ad 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1428,9 +1428,9 @@ out: return features; } -#define BOND_VLAN_FEATURES (NETIF_F_ALL_TX_OFFLOADS | \ - NETIF_F_SOFT_FEATURES | \ - NETIF_F_LRO) +#define BOND_VLAN_FEATURES (NETIF_F_ALL_CSUM | NETIF_F_SG | \ + NETIF_F_FRAGLIST | NETIF_F_ALL_TSO | \ + NETIF_F_HIGHDMA | NETIF_F_LRO) static void bond_compute_features(struct bonding *bond) { diff --git a/drivers/net/gianfar.c b/drivers/net/gianfar.c index 2dfcc8047847..dfa55f94ba7f 100644 --- a/drivers/net/gianfar.c +++ b/drivers/net/gianfar.c @@ -2289,6 +2289,23 @@ static int gfar_set_mac_address(struct net_device *dev) return 0; } +/* Check if rx parser should be activated */ +void gfar_check_rx_parser_mode(struct gfar_private *priv) +{ + struct gfar __iomem *regs; + u32 tempval; + + regs = priv->gfargrp[0].regs; + + tempval = gfar_read(®s->rctrl); + /* If parse is no longer required, then disable parser */ + if (tempval & RCTRL_REQ_PARSER) + tempval |= RCTRL_PRSDEP_INIT; + else + tempval &= ~RCTRL_PRSDEP_INIT; + gfar_write(®s->rctrl, tempval); +} + /* Enables and disables VLAN insertion/extraction */ static void gfar_vlan_rx_register(struct net_device *dev, @@ -2325,12 +2342,9 @@ static void gfar_vlan_rx_register(struct net_device *dev, /* Disable VLAN tag extraction */ tempval = gfar_read(®s->rctrl); tempval &= ~RCTRL_VLEX; - /* If parse is no longer required, then disable parser */ - if (tempval & RCTRL_REQ_PARSER) - tempval |= RCTRL_PRSDEP_INIT; - else - tempval &= ~RCTRL_PRSDEP_INIT; gfar_write(®s->rctrl, tempval); + + gfar_check_rx_parser_mode(priv); } gfar_change_mtu(dev, dev->mtu); diff --git a/drivers/net/gianfar.h b/drivers/net/gianfar.h index ba36dc7a3435..440e69d8beff 100644 --- a/drivers/net/gianfar.h +++ b/drivers/net/gianfar.h @@ -274,7 +274,7 @@ extern const char gfar_driver_version[]; #define RCTRL_PROM 0x00000008 #define RCTRL_EMEN 0x00000002 #define RCTRL_REQ_PARSER (RCTRL_VLEX | RCTRL_IPCSEN | \ - RCTRL_TUCSEN) + RCTRL_TUCSEN | RCTRL_FILREN) #define RCTRL_CHECKSUMMING (RCTRL_IPCSEN | RCTRL_TUCSEN | \ RCTRL_PRSDEP_INIT) #define RCTRL_EXTHASH (RCTRL_GHTX) @@ -1156,6 +1156,7 @@ extern void gfar_configure_coalescing(struct gfar_private *priv, unsigned long tx_mask, unsigned long rx_mask); void gfar_init_sysfs(struct net_device *dev); int gfar_set_features(struct net_device *dev, u32 features); +extern void gfar_check_rx_parser_mode(struct gfar_private *priv); extern const struct ethtool_ops gfar_ethtool_ops; diff --git a/drivers/net/natsemi.c b/drivers/net/natsemi.c index 8f8b65af9ed5..60f46bc2bf64 100644 --- a/drivers/net/natsemi.c +++ b/drivers/net/natsemi.c @@ -140,7 +140,7 @@ MODULE_LICENSE("GPL"); module_param(mtu, int, 0); module_param(debug, int, 0); module_param(rx_copybreak, int, 0); -module_param(dspcfg_workaround, int, 1); +module_param(dspcfg_workaround, int, 0); module_param_array(options, int, NULL, 0); module_param_array(full_duplex, int, NULL, 0); MODULE_PARM_DESC(mtu, "DP8381x MTU (all boards)"); @@ -2028,8 +2028,8 @@ static void drain_rx(struct net_device *dev) np->rx_ring[i].cmd_status = 0; np->rx_ring[i].addr = cpu_to_le32(0xBADF00D0); /* An invalid address. */ if (np->rx_skbuff[i]) { - pci_unmap_single(np->pci_dev, - np->rx_dma[i], buflen, + pci_unmap_single(np->pci_dev, np->rx_dma[i], + buflen + NATSEMI_PADDING, PCI_DMA_FROMDEVICE); dev_kfree_skb(np->rx_skbuff[i]); } diff --git a/drivers/net/pppoe.c b/drivers/net/pppoe.c index 718879b35b7d..bc9a4bb31980 100644 --- a/drivers/net/pppoe.c +++ b/drivers/net/pppoe.c @@ -348,8 +348,9 @@ static int pppoe_device_event(struct notifier_block *this, /* Only look at sockets that are using this specific device. */ switch (event) { + case NETDEV_CHANGEADDR: case NETDEV_CHANGEMTU: - /* A change in mtu is a bad thing, requiring + /* A change in mtu or address is a bad thing, requiring * LCP re-negotiation. */ diff --git a/drivers/net/r6040.c b/drivers/net/r6040.c index 200a363c3bf5..0ffec4608441 100644 --- a/drivers/net/r6040.c +++ b/drivers/net/r6040.c @@ -677,9 +677,11 @@ static irqreturn_t r6040_interrupt(int irq, void *dev_id) if (status & RX_FIFO_FULL) dev->stats.rx_fifo_errors++; - /* Mask off RX interrupt */ - misr &= ~RX_INTS; - napi_schedule(&lp->napi); + if (likely(napi_schedule_prep(&lp->napi))) { + /* Mask off RX interrupt */ + misr &= ~RX_INTS; + __napi_schedule(&lp->napi); + } } /* TX interrupt request */ diff --git a/drivers/net/slip.c b/drivers/net/slip.c index 8ec1a9a0bb9a..2f110fb30daa 100644 --- a/drivers/net/slip.c +++ b/drivers/net/slip.c @@ -182,11 +182,11 @@ static int sl_alloc_bufs(struct slip *sl, int mtu) #ifdef SL_INCLUDE_CSLIP cbuff = xchg(&sl->cbuff, cbuff); slcomp = xchg(&sl->slcomp, slcomp); +#endif #ifdef CONFIG_SLIP_MODE_SLIP6 sl->xdata = 0; sl->xbits = 0; #endif -#endif spin_unlock_bh(&sl->lock); err = 0; diff --git a/drivers/net/tulip/dmfe.c b/drivers/net/tulip/dmfe.c index 468512731966..9a21ca3873fc 100644 --- a/drivers/net/tulip/dmfe.c +++ b/drivers/net/tulip/dmfe.c @@ -879,7 +879,6 @@ static void dmfe_free_tx_pkt(struct DEVICE *dev, struct dmfe_board_info * db) txptr = db->tx_remove_ptr; while(db->tx_packet_cnt) { tdes0 = le32_to_cpu(txptr->tdes0); - pr_debug("tdes0=%x\n", tdes0); if (tdes0 & 0x80000000) break; @@ -889,7 +888,6 @@ static void dmfe_free_tx_pkt(struct DEVICE *dev, struct dmfe_board_info * db) /* Transmit statistic counter */ if ( tdes0 != 0x7fffffff ) { - pr_debug("tdes0=%x\n", tdes0); dev->stats.collisions += (tdes0 >> 3) & 0xf; dev->stats.tx_bytes += le32_to_cpu(txptr->tdes1) & 0x7ff; if (tdes0 & TDES0_ERR_MASK) { @@ -986,7 +984,6 @@ static void dmfe_rx_packet(struct DEVICE *dev, struct dmfe_board_info * db) /* error summary bit check */ if (rdes0 & 0x8000) { /* This is a error packet */ - pr_debug("rdes0: %x\n", rdes0); dev->stats.rx_errors++; if (rdes0 & 1) dev->stats.rx_fifo_errors++; @@ -1638,7 +1635,6 @@ static u8 dmfe_sense_speed(struct dmfe_board_info * db) else /* DM9102/DM9102A */ phy_mode = phy_read(db->ioaddr, db->phy_addr, 17, db->chip_id) & 0xf000; - pr_debug("Phy_mode %x\n", phy_mode); switch (phy_mode) { case 0x1000: db->op_mode = DMFE_10MHF; break; case 0x2000: db->op_mode = DMFE_10MFD; break; diff --git a/drivers/net/usb/hso.c b/drivers/net/usb/hso.c index 387ca43f26f4..304fe78ff60e 100644 --- a/drivers/net/usb/hso.c +++ b/drivers/net/usb/hso.c @@ -2421,10 +2421,8 @@ static void hso_free_net_device(struct hso_device *hso_dev) remove_net_device(hso_net->parent); - if (hso_net->net) { + if (hso_net->net) unregister_netdev(hso_net->net); - free_netdev(hso_net->net); - } /* start freeing */ for (i = 0; i < MUX_BULK_RX_BUF_COUNT; i++) { @@ -2436,6 +2434,9 @@ static void hso_free_net_device(struct hso_device *hso_dev) kfree(hso_net->mux_bulk_tx_buf); hso_net->mux_bulk_tx_buf = NULL; + if (hso_net->net) + free_netdev(hso_net->net); + kfree(hso_dev); } diff --git a/drivers/net/wireless/ath/ath5k/pci.c b/drivers/net/wireless/ath/ath5k/pci.c index 296c316a8341..f2c0c236392f 100644 --- a/drivers/net/wireless/ath/ath5k/pci.c +++ b/drivers/net/wireless/ath/ath5k/pci.c @@ -297,7 +297,9 @@ ath5k_pci_remove(struct pci_dev *pdev) #ifdef CONFIG_PM_SLEEP static int ath5k_pci_suspend(struct device *dev) { - struct ath5k_softc *sc = pci_get_drvdata(to_pci_dev(dev)); + struct pci_dev *pdev = to_pci_dev(dev); + struct ieee80211_hw *hw = pci_get_drvdata(pdev); + struct ath5k_softc *sc = hw->priv; ath5k_led_off(sc); return 0; @@ -306,7 +308,8 @@ static int ath5k_pci_suspend(struct device *dev) static int ath5k_pci_resume(struct device *dev) { struct pci_dev *pdev = to_pci_dev(dev); - struct ath5k_softc *sc = pci_get_drvdata(pdev); + struct ieee80211_hw *hw = pci_get_drvdata(pdev); + struct ath5k_softc *sc = hw->priv; /* * Suspend/Resume resets the PCI configuration space, so we have to diff --git a/drivers/net/wireless/ath/ath5k/sysfs.c b/drivers/net/wireless/ath/ath5k/sysfs.c index 929c68cdf8ab..a073cdce1f15 100644 --- a/drivers/net/wireless/ath/ath5k/sysfs.c +++ b/drivers/net/wireless/ath/ath5k/sysfs.c @@ -10,7 +10,8 @@ static ssize_t ath5k_attr_show_##name(struct device *dev, \ struct device_attribute *attr, \ char *buf) \ { \ - struct ath5k_softc *sc = dev_get_drvdata(dev); \ + struct ieee80211_hw *hw = dev_get_drvdata(dev); \ + struct ath5k_softc *sc = hw->priv; \ return snprintf(buf, PAGE_SIZE, "%d\n", get); \ } \ \ @@ -18,7 +19,8 @@ static ssize_t ath5k_attr_store_##name(struct device *dev, \ struct device_attribute *attr, \ const char *buf, size_t count) \ { \ - struct ath5k_softc *sc = dev_get_drvdata(dev); \ + struct ieee80211_hw *hw = dev_get_drvdata(dev); \ + struct ath5k_softc *sc = hw->priv; \ int val; \ \ val = (int)simple_strtoul(buf, NULL, 10); \ @@ -33,7 +35,8 @@ static ssize_t ath5k_attr_show_##name(struct device *dev, \ struct device_attribute *attr, \ char *buf) \ { \ - struct ath5k_softc *sc = dev_get_drvdata(dev); \ + struct ieee80211_hw *hw = dev_get_drvdata(dev); \ + struct ath5k_softc *sc = hw->priv; \ return snprintf(buf, PAGE_SIZE, "%d\n", get); \ } \ static DEVICE_ATTR(name, S_IRUGO, ath5k_attr_show_##name, NULL) diff --git a/drivers/net/wireless/ath/ath9k/xmit.c b/drivers/net/wireless/ath/ath9k/xmit.c index 3779b8977d47..33443bcaa8d9 100644 --- a/drivers/net/wireless/ath/ath9k/xmit.c +++ b/drivers/net/wireless/ath/ath9k/xmit.c @@ -671,7 +671,8 @@ static int ath_compute_num_delims(struct ath_softc *sc, struct ath_atx_tid *tid, * TODO - this could be improved to be dependent on the rate. * The hardware can keep up at lower rates, but not higher rates */ - if (fi->keyix != ATH9K_TXKEYIX_INVALID) + if ((fi->keyix != ATH9K_TXKEYIX_INVALID) && + !(sc->sc_ah->caps.hw_caps & ATH9K_HW_CAP_EDMA)) ndelim += ATH_AGGR_ENCRYPTDELIM; /* diff --git a/drivers/net/wireless/ath/carl9170/usb.c b/drivers/net/wireless/ath/carl9170/usb.c index 2fb53d067512..333b69ef2ae2 100644 --- a/drivers/net/wireless/ath/carl9170/usb.c +++ b/drivers/net/wireless/ath/carl9170/usb.c @@ -112,6 +112,8 @@ static struct usb_device_id carl9170_usb_ids[] = { { USB_DEVICE(0x04bb, 0x093f) }, /* NEC WL300NU-G */ { USB_DEVICE(0x0409, 0x0249) }, + /* NEC WL300NU-AG */ + { USB_DEVICE(0x0409, 0x02b4) }, /* AVM FRITZ!WLAN USB Stick N */ { USB_DEVICE(0x057c, 0x8401) }, /* AVM FRITZ!WLAN USB Stick N 2.4 */ diff --git a/drivers/net/wireless/rtlwifi/rtl8192cu/sw.c b/drivers/net/wireless/rtlwifi/rtl8192cu/sw.c index 092e342c19df..942f7a3969a7 100644 --- a/drivers/net/wireless/rtlwifi/rtl8192cu/sw.c +++ b/drivers/net/wireless/rtlwifi/rtl8192cu/sw.c @@ -298,6 +298,7 @@ static struct usb_device_id rtl8192c_usb_ids[] = { {RTL_USB_DEVICE(0x06f8, 0xe033, rtl92cu_hal_cfg)}, /*Hercules - Edimax*/ {RTL_USB_DEVICE(0x07b8, 0x8188, rtl92cu_hal_cfg)}, /*Abocom - Abocom*/ {RTL_USB_DEVICE(0x07b8, 0x8189, rtl92cu_hal_cfg)}, /*Funai - Abocom*/ + {RTL_USB_DEVICE(0x0846, 0x9041, rtl92cu_hal_cfg)}, /*NetGear WNA1000M*/ {RTL_USB_DEVICE(0x0Df6, 0x0052, rtl92cu_hal_cfg)}, /*Sitecom - Edimax*/ {RTL_USB_DEVICE(0x0eb0, 0x9071, rtl92cu_hal_cfg)}, /*NO Brand - Etop*/ /* HP - Lite-On ,8188CUS Slim Combo */ diff --git a/drivers/ssb/driver_pcicore.c b/drivers/ssb/driver_pcicore.c index 2a20dabec76d..d6620ad309ce 100644 --- a/drivers/ssb/driver_pcicore.c +++ b/drivers/ssb/driver_pcicore.c @@ -516,8 +516,17 @@ static void ssb_pcicore_pcie_setup_workarounds(struct ssb_pcicore *pc) static void ssb_pcicore_init_clientmode(struct ssb_pcicore *pc) { + ssb_pcicore_fix_sprom_core_index(pc); + /* Disable PCI interrupts. */ ssb_write32(pc->dev, SSB_INTVEC, 0); + + /* Additional PCIe always once-executed workarounds */ + if (pc->dev->id.coreid == SSB_DEV_PCIE) { + ssb_pcicore_serdes_workaround(pc); + /* TODO: ASPM */ + /* TODO: Clock Request Update */ + } } void ssb_pcicore_init(struct ssb_pcicore *pc) @@ -529,8 +538,6 @@ void ssb_pcicore_init(struct ssb_pcicore *pc) if (!ssb_device_is_enabled(dev)) ssb_device_enable(dev, 0); - ssb_pcicore_fix_sprom_core_index(pc); - #ifdef CONFIG_SSB_PCICORE_HOSTMODE pc->hostmode = pcicore_is_in_hostmode(pc); if (pc->hostmode) @@ -538,13 +545,6 @@ void ssb_pcicore_init(struct ssb_pcicore *pc) #endif /* CONFIG_SSB_PCICORE_HOSTMODE */ if (!pc->hostmode) ssb_pcicore_init_clientmode(pc); - - /* Additional PCIe always once-executed workarounds */ - if (dev->id.coreid == SSB_DEV_PCIE) { - ssb_pcicore_serdes_workaround(pc); - /* TODO: ASPM */ - /* TODO: Clock Request Update */ - } } static u32 ssb_pcie_read(struct ssb_pcicore *pc, u32 address) diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig index 9536d386bb38..21d816e9dfa5 100644 --- a/drivers/watchdog/Kconfig +++ b/drivers/watchdog/Kconfig @@ -599,8 +599,7 @@ config IT87_WDT config HP_WATCHDOG tristate "HP ProLiant iLO2+ Hardware Watchdog Timer" - depends on X86 - default m + depends on X86 && PCI help A software monitoring watchdog and NMI sourcing driver. This driver will detect lockups and provide a stack trace. This is a driver that diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h index 52d7eca8c7bf..502b9e988679 100644 --- a/fs/btrfs/btrfs_inode.h +++ b/fs/btrfs/btrfs_inode.h @@ -34,6 +34,9 @@ struct btrfs_inode { */ struct btrfs_key location; + /* Lock for counters */ + spinlock_t lock; + /* the extent_tree has caches of all the extent mappings to disk */ struct extent_map_tree extent_tree; @@ -134,8 +137,8 @@ struct btrfs_inode { * items we think we'll end up using, and reserved_extents is the number * of extent items we've reserved metadata for. */ - atomic_t outstanding_extents; - atomic_t reserved_extents; + unsigned outstanding_extents; + unsigned reserved_extents; /* * ordered_data_close is set by truncate when a file that used @@ -184,4 +187,13 @@ static inline void btrfs_i_size_write(struct inode *inode, u64 size) BTRFS_I(inode)->disk_i_size = size; } +static inline bool btrfs_is_free_space_inode(struct btrfs_root *root, + struct inode *inode) +{ + if (root == root->fs_info->tree_root || + BTRFS_I(inode)->location.objectid == BTRFS_FREE_INO_OBJECTID) + return true; + return false; +} + #endif diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index 2e667868e0d2..011cab3aca8d 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -54,8 +54,13 @@ noinline void btrfs_set_path_blocking(struct btrfs_path *p) { int i; for (i = 0; i < BTRFS_MAX_LEVEL; i++) { - if (p->nodes[i] && p->locks[i]) - btrfs_set_lock_blocking(p->nodes[i]); + if (!p->nodes[i] || !p->locks[i]) + continue; + btrfs_set_lock_blocking_rw(p->nodes[i], p->locks[i]); + if (p->locks[i] == BTRFS_READ_LOCK) + p->locks[i] = BTRFS_READ_LOCK_BLOCKING; + else if (p->locks[i] == BTRFS_WRITE_LOCK) + p->locks[i] = BTRFS_WRITE_LOCK_BLOCKING; } } @@ -68,7 +73,7 @@ noinline void btrfs_set_path_blocking(struct btrfs_path *p) * for held */ noinline void btrfs_clear_path_blocking(struct btrfs_path *p, - struct extent_buffer *held) + struct extent_buffer *held, int held_rw) { int i; @@ -79,19 +84,29 @@ noinline void btrfs_clear_path_blocking(struct btrfs_path *p, * really sure by forcing the path to blocking before we clear * the path blocking. */ - if (held) - btrfs_set_lock_blocking(held); + if (held) { + btrfs_set_lock_blocking_rw(held, held_rw); + if (held_rw == BTRFS_WRITE_LOCK) + held_rw = BTRFS_WRITE_LOCK_BLOCKING; + else if (held_rw == BTRFS_READ_LOCK) + held_rw = BTRFS_READ_LOCK_BLOCKING; + } btrfs_set_path_blocking(p); #endif for (i = BTRFS_MAX_LEVEL - 1; i >= 0; i--) { - if (p->nodes[i] && p->locks[i]) - btrfs_clear_lock_blocking(p->nodes[i]); + if (p->nodes[i] && p->locks[i]) { + btrfs_clear_lock_blocking_rw(p->nodes[i], p->locks[i]); + if (p->locks[i] == BTRFS_WRITE_LOCK_BLOCKING) + p->locks[i] = BTRFS_WRITE_LOCK; + else if (p->locks[i] == BTRFS_READ_LOCK_BLOCKING) + p->locks[i] = BTRFS_READ_LOCK; + } } #ifdef CONFIG_DEBUG_LOCK_ALLOC if (held) - btrfs_clear_lock_blocking(held); + btrfs_clear_lock_blocking_rw(held, held_rw); #endif } @@ -119,7 +134,7 @@ noinline void btrfs_release_path(struct btrfs_path *p) if (!p->nodes[i]) continue; if (p->locks[i]) { - btrfs_tree_unlock(p->nodes[i]); + btrfs_tree_unlock_rw(p->nodes[i], p->locks[i]); p->locks[i] = 0; } free_extent_buffer(p->nodes[i]); @@ -167,6 +182,25 @@ struct extent_buffer *btrfs_lock_root_node(struct btrfs_root *root) return eb; } +/* loop around taking references on and locking the root node of the + * tree until you end up with a lock on the root. A locked buffer + * is returned, with a reference held. + */ +struct extent_buffer *btrfs_read_lock_root_node(struct btrfs_root *root) +{ + struct extent_buffer *eb; + + while (1) { + eb = btrfs_root_node(root); + btrfs_tree_read_lock(eb); + if (eb == root->node) + break; + btrfs_tree_read_unlock(eb); + free_extent_buffer(eb); + } + return eb; +} + /* cowonly root (everything not a reference counted cow subvolume), just get * put onto a simple dirty list. transaction.c walks this to make sure they * get properly updated on disk. @@ -626,14 +660,6 @@ int btrfs_realloc_node(struct btrfs_trans_handle *trans, for (i = start_slot; i < end_slot; i++) { int close = 1; - if (!parent->map_token) { - map_extent_buffer(parent, - btrfs_node_key_ptr_offset(i), - sizeof(struct btrfs_key_ptr), - &parent->map_token, &parent->kaddr, - &parent->map_start, &parent->map_len, - KM_USER1); - } btrfs_node_key(parent, &disk_key, i); if (!progress_passed && comp_keys(&disk_key, progress) < 0) continue; @@ -656,11 +682,6 @@ int btrfs_realloc_node(struct btrfs_trans_handle *trans, last_block = blocknr; continue; } - if (parent->map_token) { - unmap_extent_buffer(parent, parent->map_token, - KM_USER1); - parent->map_token = NULL; - } cur = btrfs_find_tree_block(root, blocknr, blocksize); if (cur) @@ -701,11 +722,6 @@ int btrfs_realloc_node(struct btrfs_trans_handle *trans, btrfs_tree_unlock(cur); free_extent_buffer(cur); } - if (parent->map_token) { - unmap_extent_buffer(parent, parent->map_token, - KM_USER1); - parent->map_token = NULL; - } return err; } @@ -746,7 +762,6 @@ static noinline int generic_bin_search(struct extent_buffer *eb, struct btrfs_disk_key *tmp = NULL; struct btrfs_disk_key unaligned; unsigned long offset; - char *map_token = NULL; char *kaddr = NULL; unsigned long map_start = 0; unsigned long map_len = 0; @@ -756,18 +771,13 @@ static noinline int generic_bin_search(struct extent_buffer *eb, mid = (low + high) / 2; offset = p + mid * item_size; - if (!map_token || offset < map_start || + if (!kaddr || offset < map_start || (offset + sizeof(struct btrfs_disk_key)) > map_start + map_len) { - if (map_token) { - unmap_extent_buffer(eb, map_token, KM_USER0); - map_token = NULL; - } err = map_private_extent_buffer(eb, offset, sizeof(struct btrfs_disk_key), - &map_token, &kaddr, - &map_start, &map_len, KM_USER0); + &kaddr, &map_start, &map_len); if (!err) { tmp = (struct btrfs_disk_key *)(kaddr + offset - @@ -790,14 +800,10 @@ static noinline int generic_bin_search(struct extent_buffer *eb, high = mid; else { *slot = mid; - if (map_token) - unmap_extent_buffer(eb, map_token, KM_USER0); return 0; } } *slot = low; - if (map_token) - unmap_extent_buffer(eb, map_token, KM_USER0); return 1; } @@ -890,7 +896,8 @@ static noinline int balance_level(struct btrfs_trans_handle *trans, mid = path->nodes[level]; - WARN_ON(!path->locks[level]); + WARN_ON(path->locks[level] != BTRFS_WRITE_LOCK && + path->locks[level] != BTRFS_WRITE_LOCK_BLOCKING); WARN_ON(btrfs_header_generation(mid) != trans->transid); orig_ptr = btrfs_node_blockptr(mid, orig_slot); @@ -1228,7 +1235,6 @@ static void reada_for_search(struct btrfs_root *root, u32 nr; u32 blocksize; u32 nscan = 0; - bool map = true; if (level != 1) return; @@ -1250,19 +1256,8 @@ static void reada_for_search(struct btrfs_root *root, nritems = btrfs_header_nritems(node); nr = slot; - if (node->map_token || path->skip_locking) - map = false; while (1) { - if (map && !node->map_token) { - unsigned long offset = btrfs_node_key_ptr_offset(nr); - map_private_extent_buffer(node, offset, - sizeof(struct btrfs_key_ptr), - &node->map_token, - &node->kaddr, - &node->map_start, - &node->map_len, KM_USER1); - } if (direction < 0) { if (nr == 0) break; @@ -1281,11 +1276,6 @@ static void reada_for_search(struct btrfs_root *root, if ((search <= target && target - search <= 65536) || (search > target && search - target <= 65536)) { gen = btrfs_node_ptr_generation(node, nr); - if (map && node->map_token) { - unmap_extent_buffer(node, node->map_token, - KM_USER1); - node->map_token = NULL; - } readahead_tree_block(root, search, blocksize, gen); nread += blocksize; } @@ -1293,10 +1283,6 @@ static void reada_for_search(struct btrfs_root *root, if ((nread > 65536 || nscan > 32)) break; } - if (map && node->map_token) { - unmap_extent_buffer(node, node->map_token, KM_USER1); - node->map_token = NULL; - } } /* @@ -1409,7 +1395,7 @@ static noinline void unlock_up(struct btrfs_path *path, int level, t = path->nodes[i]; if (i >= lowest_unlock && i > skip_level && path->locks[i]) { - btrfs_tree_unlock(t); + btrfs_tree_unlock_rw(t, path->locks[i]); path->locks[i] = 0; } } @@ -1436,7 +1422,7 @@ noinline void btrfs_unlock_up_safe(struct btrfs_path *path, int level) continue; if (!path->locks[i]) continue; - btrfs_tree_unlock(path->nodes[i]); + btrfs_tree_unlock_rw(path->nodes[i], path->locks[i]); path->locks[i] = 0; } } @@ -1485,6 +1471,8 @@ read_block_for_search(struct btrfs_trans_handle *trans, * we can trust our generation number */ free_extent_buffer(tmp); + btrfs_set_path_blocking(p); + tmp = read_tree_block(root, blocknr, blocksize, gen); if (tmp && btrfs_buffer_uptodate(tmp, gen)) { *eb_ret = tmp; @@ -1540,20 +1528,27 @@ read_block_for_search(struct btrfs_trans_handle *trans, static int setup_nodes_for_search(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct btrfs_path *p, - struct extent_buffer *b, int level, int ins_len) + struct extent_buffer *b, int level, int ins_len, + int *write_lock_level) { int ret; if ((p->search_for_split || ins_len > 0) && btrfs_header_nritems(b) >= BTRFS_NODEPTRS_PER_BLOCK(root) - 3) { int sret; + if (*write_lock_level < level + 1) { + *write_lock_level = level + 1; + btrfs_release_path(p); + goto again; + } + sret = reada_for_balance(root, p, level); if (sret) goto again; btrfs_set_path_blocking(p); sret = split_node(trans, root, p, level); - btrfs_clear_path_blocking(p, NULL); + btrfs_clear_path_blocking(p, NULL, 0); BUG_ON(sret > 0); if (sret) { @@ -1565,13 +1560,19 @@ setup_nodes_for_search(struct btrfs_trans_handle *trans, BTRFS_NODEPTRS_PER_BLOCK(root) / 2) { int sret; + if (*write_lock_level < level + 1) { + *write_lock_level = level + 1; + btrfs_release_path(p); + goto again; + } + sret = reada_for_balance(root, p, level); if (sret) goto again; btrfs_set_path_blocking(p); sret = balance_level(trans, root, p, level); - btrfs_clear_path_blocking(p, NULL); + btrfs_clear_path_blocking(p, NULL, 0); if (sret) { ret = sret; @@ -1615,27 +1616,78 @@ int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root int err; int level; int lowest_unlock = 1; + int root_lock; + /* everything at write_lock_level or lower must be write locked */ + int write_lock_level = 0; u8 lowest_level = 0; lowest_level = p->lowest_level; WARN_ON(lowest_level && ins_len > 0); WARN_ON(p->nodes[0] != NULL); - if (ins_len < 0) + if (ins_len < 0) { lowest_unlock = 2; + /* when we are removing items, we might have to go up to level + * two as we update tree pointers Make sure we keep write + * for those levels as well + */ + write_lock_level = 2; + } else if (ins_len > 0) { + /* + * for inserting items, make sure we have a write lock on + * level 1 so we can update keys + */ + write_lock_level = 1; + } + + if (!cow) + write_lock_level = -1; + + if (cow && (p->keep_locks || p->lowest_level)) + write_lock_level = BTRFS_MAX_LEVEL; + again: + /* + * we try very hard to do read locks on the root + */ + root_lock = BTRFS_READ_LOCK; + level = 0; if (p->search_commit_root) { + /* + * the commit roots are read only + * so we always do read locks + */ b = root->commit_root; extent_buffer_get(b); + level = btrfs_header_level(b); if (!p->skip_locking) - btrfs_tree_lock(b); + btrfs_tree_read_lock(b); } else { - if (p->skip_locking) + if (p->skip_locking) { b = btrfs_root_node(root); - else - b = btrfs_lock_root_node(root); + level = btrfs_header_level(b); + } else { + /* we don't know the level of the root node + * until we actually have it read locked + */ + b = btrfs_read_lock_root_node(root); + level = btrfs_header_level(b); + if (level <= write_lock_level) { + /* whoops, must trade for write lock */ + btrfs_tree_read_unlock(b); + free_extent_buffer(b); + b = btrfs_lock_root_node(root); + root_lock = BTRFS_WRITE_LOCK; + + /* the level might have changed, check again */ + level = btrfs_header_level(b); + } + } } + p->nodes[level] = b; + if (!p->skip_locking) + p->locks[level] = root_lock; while (b) { level = btrfs_header_level(b); @@ -1644,10 +1696,6 @@ again: * setup the path here so we can release it under lock * contention with the cow code */ - p->nodes[level] = b; - if (!p->skip_locking) - p->locks[level] = 1; - if (cow) { /* * if we don't really need to cow this block @@ -1659,6 +1707,16 @@ again: btrfs_set_path_blocking(p); + /* + * must have write locks on this node and the + * parent + */ + if (level + 1 > write_lock_level) { + write_lock_level = level + 1; + btrfs_release_path(p); + goto again; + } + err = btrfs_cow_block(trans, root, b, p->nodes[level + 1], p->slots[level + 1], &b); @@ -1671,10 +1729,7 @@ cow_done: BUG_ON(!cow && ins_len); p->nodes[level] = b; - if (!p->skip_locking) - p->locks[level] = 1; - - btrfs_clear_path_blocking(p, NULL); + btrfs_clear_path_blocking(p, NULL, 0); /* * we have a lock on b and as long as we aren't changing @@ -1700,7 +1755,7 @@ cow_done: } p->slots[level] = slot; err = setup_nodes_for_search(trans, root, p, b, level, - ins_len); + ins_len, &write_lock_level); if (err == -EAGAIN) goto again; if (err) { @@ -1710,6 +1765,19 @@ cow_done: b = p->nodes[level]; slot = p->slots[level]; + /* + * slot 0 is special, if we change the key + * we have to update the parent pointer + * which means we must have a write lock + * on the parent + */ + if (slot == 0 && cow && + write_lock_level < level + 1) { + write_lock_level = level + 1; + btrfs_release_path(p); + goto again; + } + unlock_up(p, level, lowest_unlock); if (level == lowest_level) { @@ -1728,23 +1796,42 @@ cow_done: } if (!p->skip_locking) { - btrfs_clear_path_blocking(p, NULL); - err = btrfs_try_spin_lock(b); - - if (!err) { - btrfs_set_path_blocking(p); - btrfs_tree_lock(b); - btrfs_clear_path_blocking(p, b); + level = btrfs_header_level(b); + if (level <= write_lock_level) { + err = btrfs_try_tree_write_lock(b); + if (!err) { + btrfs_set_path_blocking(p); + btrfs_tree_lock(b); + btrfs_clear_path_blocking(p, b, + BTRFS_WRITE_LOCK); + } + p->locks[level] = BTRFS_WRITE_LOCK; + } else { + err = btrfs_try_tree_read_lock(b); + if (!err) { + btrfs_set_path_blocking(p); + btrfs_tree_read_lock(b); + btrfs_clear_path_blocking(p, b, + BTRFS_READ_LOCK); + } + p->locks[level] = BTRFS_READ_LOCK; } + p->nodes[level] = b; } } else { p->slots[level] = slot; if (ins_len > 0 && btrfs_leaf_free_space(root, b) < ins_len) { + if (write_lock_level < 1) { + write_lock_level = 1; + btrfs_release_path(p); + goto again; + } + btrfs_set_path_blocking(p); err = split_leaf(trans, root, key, p, ins_len, ret == 0); - btrfs_clear_path_blocking(p, NULL); + btrfs_clear_path_blocking(p, NULL, 0); BUG_ON(err > 0); if (err) { @@ -2025,7 +2112,7 @@ static noinline int insert_new_root(struct btrfs_trans_handle *trans, add_root_to_dirty_list(root); extent_buffer_get(c); path->nodes[level] = c; - path->locks[level] = 1; + path->locks[level] = BTRFS_WRITE_LOCK; path->slots[level] = 0; return 0; } @@ -2253,14 +2340,6 @@ static noinline int __push_leaf_right(struct btrfs_trans_handle *trans, if (path->slots[0] == i) push_space += data_size; - if (!left->map_token) { - map_extent_buffer(left, (unsigned long)item, - sizeof(struct btrfs_item), - &left->map_token, &left->kaddr, - &left->map_start, &left->map_len, - KM_USER1); - } - this_item_size = btrfs_item_size(left, item); if (this_item_size + sizeof(*item) + push_space > free_space) break; @@ -2271,10 +2350,6 @@ static noinline int __push_leaf_right(struct btrfs_trans_handle *trans, break; i--; } - if (left->map_token) { - unmap_extent_buffer(left, left->map_token, KM_USER1); - left->map_token = NULL; - } if (push_items == 0) goto out_unlock; @@ -2316,21 +2391,10 @@ static noinline int __push_leaf_right(struct btrfs_trans_handle *trans, push_space = BTRFS_LEAF_DATA_SIZE(root); for (i = 0; i < right_nritems; i++) { item = btrfs_item_nr(right, i); - if (!right->map_token) { - map_extent_buffer(right, (unsigned long)item, - sizeof(struct btrfs_item), - &right->map_token, &right->kaddr, - &right->map_start, &right->map_len, - KM_USER1); - } push_space -= btrfs_item_size(right, item); btrfs_set_item_offset(right, item, push_space); } - if (right->map_token) { - unmap_extent_buffer(right, right->map_token, KM_USER1); - right->map_token = NULL; - } left_nritems -= push_items; btrfs_set_header_nritems(left, left_nritems); @@ -2467,13 +2531,6 @@ static noinline int __push_leaf_left(struct btrfs_trans_handle *trans, for (i = 0; i < nr; i++) { item = btrfs_item_nr(right, i); - if (!right->map_token) { - map_extent_buffer(right, (unsigned long)item, - sizeof(struct btrfs_item), - &right->map_token, &right->kaddr, - &right->map_start, &right->map_len, - KM_USER1); - } if (!empty && push_items > 0) { if (path->slots[0] < i) @@ -2496,11 +2553,6 @@ static noinline int __push_leaf_left(struct btrfs_trans_handle *trans, push_space += this_item_size + sizeof(*item); } - if (right->map_token) { - unmap_extent_buffer(right, right->map_token, KM_USER1); - right->map_token = NULL; - } - if (push_items == 0) { ret = 1; goto out; @@ -2530,23 +2582,12 @@ static noinline int __push_leaf_left(struct btrfs_trans_handle *trans, u32 ioff; item = btrfs_item_nr(left, i); - if (!left->map_token) { - map_extent_buffer(left, (unsigned long)item, - sizeof(struct btrfs_item), - &left->map_token, &left->kaddr, - &left->map_start, &left->map_len, - KM_USER1); - } ioff = btrfs_item_offset(left, item); btrfs_set_item_offset(left, item, ioff - (BTRFS_LEAF_DATA_SIZE(root) - old_left_item_size)); } btrfs_set_header_nritems(left, old_left_nritems + push_items); - if (left->map_token) { - unmap_extent_buffer(left, left->map_token, KM_USER1); - left->map_token = NULL; - } /* fixup right node */ if (push_items > right_nritems) { @@ -2574,21 +2615,9 @@ static noinline int __push_leaf_left(struct btrfs_trans_handle *trans, for (i = 0; i < right_nritems; i++) { item = btrfs_item_nr(right, i); - if (!right->map_token) { - map_extent_buffer(right, (unsigned long)item, - sizeof(struct btrfs_item), - &right->map_token, &right->kaddr, - &right->map_start, &right->map_len, - KM_USER1); - } - push_space = push_space - btrfs_item_size(right, item); btrfs_set_item_offset(right, item, push_space); } - if (right->map_token) { - unmap_extent_buffer(right, right->map_token, KM_USER1); - right->map_token = NULL; - } btrfs_mark_buffer_dirty(left); if (right_nritems) @@ -2729,23 +2758,10 @@ static noinline int copy_for_split(struct btrfs_trans_handle *trans, struct btrfs_item *item = btrfs_item_nr(right, i); u32 ioff; - if (!right->map_token) { - map_extent_buffer(right, (unsigned long)item, - sizeof(struct btrfs_item), - &right->map_token, &right->kaddr, - &right->map_start, &right->map_len, - KM_USER1); - } - ioff = btrfs_item_offset(right, item); btrfs_set_item_offset(right, item, ioff + rt_data_off); } - if (right->map_token) { - unmap_extent_buffer(right, right->map_token, KM_USER1); - right->map_token = NULL; - } - btrfs_set_header_nritems(l, mid); ret = 0; btrfs_item_key(right, &disk_key, 0); @@ -3264,23 +3280,10 @@ int btrfs_truncate_item(struct btrfs_trans_handle *trans, u32 ioff; item = btrfs_item_nr(leaf, i); - if (!leaf->map_token) { - map_extent_buffer(leaf, (unsigned long)item, - sizeof(struct btrfs_item), - &leaf->map_token, &leaf->kaddr, - &leaf->map_start, &leaf->map_len, - KM_USER1); - } - ioff = btrfs_item_offset(leaf, item); btrfs_set_item_offset(leaf, item, ioff + size_diff); } - if (leaf->map_token) { - unmap_extent_buffer(leaf, leaf->map_token, KM_USER1); - leaf->map_token = NULL; - } - /* shift the data */ if (from_end) { memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) + @@ -3377,22 +3380,10 @@ int btrfs_extend_item(struct btrfs_trans_handle *trans, u32 ioff; item = btrfs_item_nr(leaf, i); - if (!leaf->map_token) { - map_extent_buffer(leaf, (unsigned long)item, - sizeof(struct btrfs_item), - &leaf->map_token, &leaf->kaddr, - &leaf->map_start, &leaf->map_len, - KM_USER1); - } ioff = btrfs_item_offset(leaf, item); btrfs_set_item_offset(leaf, item, ioff - data_size); } - if (leaf->map_token) { - unmap_extent_buffer(leaf, leaf->map_token, KM_USER1); - leaf->map_token = NULL; - } - /* shift the data */ memmove_extent_buffer(leaf, btrfs_leaf_data(leaf) + data_end - data_size, btrfs_leaf_data(leaf) + @@ -3494,27 +3485,13 @@ int btrfs_insert_some_items(struct btrfs_trans_handle *trans, * item0..itemN ... dataN.offset..dataN.size .. data0.size */ /* first correct the data pointers */ - WARN_ON(leaf->map_token); for (i = slot; i < nritems; i++) { u32 ioff; item = btrfs_item_nr(leaf, i); - if (!leaf->map_token) { - map_extent_buffer(leaf, (unsigned long)item, - sizeof(struct btrfs_item), - &leaf->map_token, &leaf->kaddr, - &leaf->map_start, &leaf->map_len, - KM_USER1); - } - ioff = btrfs_item_offset(leaf, item); btrfs_set_item_offset(leaf, item, ioff - total_data); } - if (leaf->map_token) { - unmap_extent_buffer(leaf, leaf->map_token, KM_USER1); - leaf->map_token = NULL; - } - /* shift the items */ memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot + nr), btrfs_item_nr_offset(slot), @@ -3608,27 +3585,13 @@ int setup_items_for_insert(struct btrfs_trans_handle *trans, * item0..itemN ... dataN.offset..dataN.size .. data0.size */ /* first correct the data pointers */ - WARN_ON(leaf->map_token); for (i = slot; i < nritems; i++) { u32 ioff; item = btrfs_item_nr(leaf, i); - if (!leaf->map_token) { - map_extent_buffer(leaf, (unsigned long)item, - sizeof(struct btrfs_item), - &leaf->map_token, &leaf->kaddr, - &leaf->map_start, &leaf->map_len, - KM_USER1); - } - ioff = btrfs_item_offset(leaf, item); btrfs_set_item_offset(leaf, item, ioff - total_data); } - if (leaf->map_token) { - unmap_extent_buffer(leaf, leaf->map_token, KM_USER1); - leaf->map_token = NULL; - } - /* shift the items */ memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot + nr), btrfs_item_nr_offset(slot), @@ -3840,22 +3803,10 @@ int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root, u32 ioff; item = btrfs_item_nr(leaf, i); - if (!leaf->map_token) { - map_extent_buffer(leaf, (unsigned long)item, - sizeof(struct btrfs_item), - &leaf->map_token, &leaf->kaddr, - &leaf->map_start, &leaf->map_len, - KM_USER1); - } ioff = btrfs_item_offset(leaf, item); btrfs_set_item_offset(leaf, item, ioff + dsize); } - if (leaf->map_token) { - unmap_extent_buffer(leaf, leaf->map_token, KM_USER1); - leaf->map_token = NULL; - } - memmove_extent_buffer(leaf, btrfs_item_nr_offset(slot), btrfs_item_nr_offset(slot + nr), sizeof(struct btrfs_item) * @@ -4004,11 +3955,11 @@ int btrfs_search_forward(struct btrfs_root *root, struct btrfs_key *min_key, WARN_ON(!path->keep_locks); again: - cur = btrfs_lock_root_node(root); + cur = btrfs_read_lock_root_node(root); level = btrfs_header_level(cur); WARN_ON(path->nodes[level]); path->nodes[level] = cur; - path->locks[level] = 1; + path->locks[level] = BTRFS_READ_LOCK; if (btrfs_header_generation(cur) < min_trans) { ret = 1; @@ -4098,12 +4049,12 @@ find_next_key: cur = read_node_slot(root, cur, slot); BUG_ON(!cur); - btrfs_tree_lock(cur); + btrfs_tree_read_lock(cur); - path->locks[level - 1] = 1; + path->locks[level - 1] = BTRFS_READ_LOCK; path->nodes[level - 1] = cur; unlock_up(path, level, 1); - btrfs_clear_path_blocking(path, NULL); + btrfs_clear_path_blocking(path, NULL, 0); } out: if (ret == 0) @@ -4218,30 +4169,21 @@ int btrfs_next_leaf(struct btrfs_root *root, struct btrfs_path *path) u32 nritems; int ret; int old_spinning = path->leave_spinning; - int force_blocking = 0; + int next_rw_lock = 0; nritems = btrfs_header_nritems(path->nodes[0]); if (nritems == 0) return 1; - /* - * we take the blocks in an order that upsets lockdep. Using - * blocking mode is the only way around it. - */ -#ifdef CONFIG_DEBUG_LOCK_ALLOC - force_blocking = 1; -#endif - btrfs_item_key_to_cpu(path->nodes[0], &key, nritems - 1); again: level = 1; next = NULL; + next_rw_lock = 0; btrfs_release_path(path); path->keep_locks = 1; - - if (!force_blocking) - path->leave_spinning = 1; + path->leave_spinning = 1; ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); path->keep_locks = 0; @@ -4281,11 +4223,12 @@ again: } if (next) { - btrfs_tree_unlock(next); + btrfs_tree_unlock_rw(next, next_rw_lock); free_extent_buffer(next); } next = c; + next_rw_lock = path->locks[level]; ret = read_block_for_search(NULL, root, path, &next, level, slot, &key); if (ret == -EAGAIN) @@ -4297,15 +4240,14 @@ again: } if (!path->skip_locking) { - ret = btrfs_try_spin_lock(next); + ret = btrfs_try_tree_read_lock(next); if (!ret) { btrfs_set_path_blocking(path); - btrfs_tree_lock(next); - if (!force_blocking) - btrfs_clear_path_blocking(path, next); + btrfs_tree_read_lock(next); + btrfs_clear_path_blocking(path, next, + BTRFS_READ_LOCK); } - if (force_blocking) - btrfs_set_lock_blocking(next); + next_rw_lock = BTRFS_READ_LOCK; } break; } @@ -4314,14 +4256,13 @@ again: level--; c = path->nodes[level]; if (path->locks[level]) - btrfs_tree_unlock(c); + btrfs_tree_unlock_rw(c, path->locks[level]); free_extent_buffer(c); path->nodes[level] = next; path->slots[level] = 0; if (!path->skip_locking) - path->locks[level] = 1; - + path->locks[level] = next_rw_lock; if (!level) break; @@ -4336,16 +4277,14 @@ again: } if (!path->skip_locking) { - btrfs_assert_tree_locked(path->nodes[level]); - ret = btrfs_try_spin_lock(next); + ret = btrfs_try_tree_read_lock(next); if (!ret) { btrfs_set_path_blocking(path); - btrfs_tree_lock(next); - if (!force_blocking) - btrfs_clear_path_blocking(path, next); + btrfs_tree_read_lock(next); + btrfs_clear_path_blocking(path, next, + BTRFS_READ_LOCK); } - if (force_blocking) - btrfs_set_lock_blocking(next); + next_rw_lock = BTRFS_READ_LOCK; } } ret = 0; diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 3b859a3e6a0e..3be57c611040 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -755,6 +755,8 @@ struct btrfs_space_info { chunks for this space */ unsigned int chunk_alloc:1; /* set if we are allocating a chunk */ + unsigned int flush:1; /* set if we are trying to make space */ + unsigned int force_alloc; /* set if we need to force a chunk alloc for this space */ @@ -764,7 +766,7 @@ struct btrfs_space_info { struct list_head block_groups[BTRFS_NR_RAID_TYPES]; spinlock_t lock; struct rw_semaphore groups_sem; - atomic_t caching_threads; + wait_queue_head_t wait; }; struct btrfs_block_rsv { @@ -824,6 +826,7 @@ struct btrfs_caching_control { struct list_head list; struct mutex mutex; wait_queue_head_t wait; + struct btrfs_work work; struct btrfs_block_group_cache *block_group; u64 progress; atomic_t count; @@ -1032,6 +1035,8 @@ struct btrfs_fs_info { struct btrfs_workers endio_write_workers; struct btrfs_workers endio_freespace_worker; struct btrfs_workers submit_workers; + struct btrfs_workers caching_workers; + /* * fixup workers take dirty pages that didn't properly go through * the cow mechanism and make them safe to write. It happens @@ -2128,7 +2133,7 @@ static inline bool btrfs_mixed_space_info(struct btrfs_space_info *space_info) /* extent-tree.c */ static inline u64 btrfs_calc_trans_metadata_size(struct btrfs_root *root, - int num_items) + unsigned num_items) { return (root->leafsize + root->nodesize * (BTRFS_MAX_LEVEL - 1)) * 3 * num_items; @@ -2222,9 +2227,6 @@ void btrfs_set_inode_space_info(struct btrfs_root *root, struct inode *ionde); void btrfs_clear_space_info_full(struct btrfs_fs_info *info); int btrfs_check_data_free_space(struct inode *inode, u64 bytes); void btrfs_free_reserved_data_space(struct inode *inode, u64 bytes); -int btrfs_trans_reserve_metadata(struct btrfs_trans_handle *trans, - struct btrfs_root *root, - int num_items); void btrfs_trans_release_metadata(struct btrfs_trans_handle *trans, struct btrfs_root *root); int btrfs_orphan_reserve_metadata(struct btrfs_trans_handle *trans, @@ -2330,7 +2332,7 @@ struct btrfs_path *btrfs_alloc_path(void); void btrfs_free_path(struct btrfs_path *p); void btrfs_set_path_blocking(struct btrfs_path *p); void btrfs_clear_path_blocking(struct btrfs_path *p, - struct extent_buffer *held); + struct extent_buffer *held, int held_rw); void btrfs_unlock_up_safe(struct btrfs_path *p, int level); int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root, diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index 98c68e658a9b..b52c672f4c18 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -735,7 +735,7 @@ static int btrfs_batch_insert_items(struct btrfs_trans_handle *trans, } /* reset all the locked nodes in the patch to spinning locks. */ - btrfs_clear_path_blocking(path, NULL); + btrfs_clear_path_blocking(path, NULL, 0); /* insert the keys of the items */ ret = setup_items_for_insert(trans, root, path, keys, data_size, diff --git a/fs/btrfs/dir-item.c b/fs/btrfs/dir-item.c index 685f2593c4f0..c360a848d97f 100644 --- a/fs/btrfs/dir-item.c +++ b/fs/btrfs/dir-item.c @@ -89,13 +89,8 @@ int btrfs_insert_xattr_item(struct btrfs_trans_handle *trans, data_size = sizeof(*dir_item) + name_len + data_len; dir_item = insert_with_overflow(trans, root, path, &key, data_size, name, name_len); - /* - * FIXME: at some point we should handle xattr's that are larger than - * what we can fit in our leaf. We set location to NULL b/c we arent - * pointing at anything else, that will change if we store the xattr - * data in a separate inode. - */ - BUG_ON(IS_ERR(dir_item)); + if (IS_ERR(dir_item)) + return PTR_ERR(dir_item); memset(&location, 0, sizeof(location)); leaf = path->nodes[0]; diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 1ac8db5dc0a3..94ecac33cf2d 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -100,38 +100,83 @@ struct async_submit_bio { struct btrfs_work work; }; -/* These are used to set the lockdep class on the extent buffer locks. - * The class is set by the readpage_end_io_hook after the buffer has - * passed csum validation but before the pages are unlocked. +/* + * Lockdep class keys for extent_buffer->lock's in this root. For a given + * eb, the lockdep key is determined by the btrfs_root it belongs to and + * the level the eb occupies in the tree. + * + * Different roots are used for different purposes and may nest inside each + * other and they require separate keysets. As lockdep keys should be + * static, assign keysets according to the purpose of the root as indicated + * by btrfs_root->objectid. This ensures that all special purpose roots + * have separate keysets. * - * The lockdep class is also set by btrfs_init_new_buffer on freshly - * allocated blocks. + * Lock-nesting across peer nodes is always done with the immediate parent + * node locked thus preventing deadlock. As lockdep doesn't know this, use + * subclass to avoid triggering lockdep warning in such cases. * - * The class is based on the level in the tree block, which allows lockdep - * to know that lower nodes nest inside the locks of higher nodes. + * The key is set by the readpage_end_io_hook after the buffer has passed + * csum validation but before the pages are unlocked. It is also set by + * btrfs_init_new_buffer on freshly allocated blocks. * - * We also add a check to make sure the highest level of the tree is - * the same as our lockdep setup here. If BTRFS_MAX_LEVEL changes, this - * code needs update as well. + * We also add a check to make sure the highest level of the tree is the + * same as our lockdep setup here. If BTRFS_MAX_LEVEL changes, this code + * needs update as well. */ #ifdef CONFIG_DEBUG_LOCK_ALLOC # if BTRFS_MAX_LEVEL != 8 # error # endif -static struct lock_class_key btrfs_eb_class[BTRFS_MAX_LEVEL + 1]; -static const char *btrfs_eb_name[BTRFS_MAX_LEVEL + 1] = { - /* leaf */ - "btrfs-extent-00", - "btrfs-extent-01", - "btrfs-extent-02", - "btrfs-extent-03", - "btrfs-extent-04", - "btrfs-extent-05", - "btrfs-extent-06", - "btrfs-extent-07", - /* highest possible level */ - "btrfs-extent-08", + +static struct btrfs_lockdep_keyset { + u64 id; /* root objectid */ + const char *name_stem; /* lock name stem */ + char names[BTRFS_MAX_LEVEL + 1][20]; + struct lock_class_key keys[BTRFS_MAX_LEVEL + 1]; +} btrfs_lockdep_keysets[] = { + { .id = BTRFS_ROOT_TREE_OBJECTID, .name_stem = "root" }, + { .id = BTRFS_EXTENT_TREE_OBJECTID, .name_stem = "extent" }, + { .id = BTRFS_CHUNK_TREE_OBJECTID, .name_stem = "chunk" }, + { .id = BTRFS_DEV_TREE_OBJECTID, .name_stem = "dev" }, + { .id = BTRFS_FS_TREE_OBJECTID, .name_stem = "fs" }, + { .id = BTRFS_CSUM_TREE_OBJECTID, .name_stem = "csum" }, + { .id = BTRFS_ORPHAN_OBJECTID, .name_stem = "orphan" }, + { .id = BTRFS_TREE_LOG_OBJECTID, .name_stem = "log" }, + { .id = BTRFS_TREE_RELOC_OBJECTID, .name_stem = "treloc" }, + { .id = BTRFS_DATA_RELOC_TREE_OBJECTID, .name_stem = "dreloc" }, + { .id = 0, .name_stem = "tree" }, }; + +void __init btrfs_init_lockdep(void) +{ + int i, j; + + /* initialize lockdep class names */ + for (i = 0; i < ARRAY_SIZE(btrfs_lockdep_keysets); i++) { + struct btrfs_lockdep_keyset *ks = &btrfs_lockdep_keysets[i]; + + for (j = 0; j < ARRAY_SIZE(ks->names); j++) + snprintf(ks->names[j], sizeof(ks->names[j]), + "btrfs-%s-%02d", ks->name_stem, j); + } +} + +void btrfs_set_buffer_lockdep_class(u64 objectid, struct extent_buffer *eb, + int level) +{ + struct btrfs_lockdep_keyset *ks; + + BUG_ON(level >= ARRAY_SIZE(ks->keys)); + + /* find the matching keyset, id 0 is the default entry */ + for (ks = btrfs_lockdep_keysets; ks->id; ks++) + if (ks->id == objectid) + break; + + lockdep_set_class_and_name(&eb->lock, + &ks->keys[level], ks->names[level]); +} + #endif /* @@ -217,7 +262,6 @@ static int csum_tree_block(struct btrfs_root *root, struct extent_buffer *buf, unsigned long len; unsigned long cur_len; unsigned long offset = BTRFS_CSUM_SIZE; - char *map_token = NULL; char *kaddr; unsigned long map_start; unsigned long map_len; @@ -228,8 +272,7 @@ static int csum_tree_block(struct btrfs_root *root, struct extent_buffer *buf, len = buf->len - offset; while (len > 0) { err = map_private_extent_buffer(buf, offset, 32, - &map_token, &kaddr, - &map_start, &map_len, KM_USER0); + &kaddr, &map_start, &map_len); if (err) return 1; cur_len = min(len, map_len - (offset - map_start)); @@ -237,7 +280,6 @@ static int csum_tree_block(struct btrfs_root *root, struct extent_buffer *buf, crc, cur_len); len -= cur_len; offset += cur_len; - unmap_extent_buffer(buf, map_token, KM_USER0); } if (csum_size > sizeof(inline_result)) { result = kzalloc(csum_size * sizeof(char), GFP_NOFS); @@ -494,15 +536,6 @@ static noinline int check_leaf(struct btrfs_root *root, return 0; } -#ifdef CONFIG_DEBUG_LOCK_ALLOC -void btrfs_set_buffer_lockdep_class(struct extent_buffer *eb, int level) -{ - lockdep_set_class_and_name(&eb->lock, - &btrfs_eb_class[level], - btrfs_eb_name[level]); -} -#endif - static int btree_readpage_end_io_hook(struct page *page, u64 start, u64 end, struct extent_state *state) { @@ -553,7 +586,8 @@ static int btree_readpage_end_io_hook(struct page *page, u64 start, u64 end, } found_level = btrfs_header_level(eb); - btrfs_set_buffer_lockdep_class(eb, found_level); + btrfs_set_buffer_lockdep_class(btrfs_header_owner(eb), + eb, found_level); ret = csum_tree_block(root, eb, 1); if (ret) { @@ -1603,7 +1637,7 @@ struct btrfs_root *open_ctree(struct super_block *sb, goto fail_bdi; } - fs_info->btree_inode->i_mapping->flags &= ~__GFP_FS; + mapping_set_gfp_mask(fs_info->btree_inode->i_mapping, GFP_NOFS); INIT_RADIX_TREE(&fs_info->fs_roots_radix, GFP_ATOMIC); INIT_LIST_HEAD(&fs_info->trans_list); @@ -1807,6 +1841,9 @@ struct btrfs_root *open_ctree(struct super_block *sb, fs_info->thread_pool_size), &fs_info->generic_worker); + btrfs_init_workers(&fs_info->caching_workers, "cache", + 2, &fs_info->generic_worker); + /* a higher idle thresh on the submit workers makes it much more * likely that bios will be send down in a sane order to the * devices @@ -1860,6 +1897,7 @@ struct btrfs_root *open_ctree(struct super_block *sb, btrfs_start_workers(&fs_info->endio_write_workers, 1); btrfs_start_workers(&fs_info->endio_freespace_worker, 1); btrfs_start_workers(&fs_info->delayed_workers, 1); + btrfs_start_workers(&fs_info->caching_workers, 1); fs_info->bdi.ra_pages *= btrfs_super_num_devices(disk_super); fs_info->bdi.ra_pages = max(fs_info->bdi.ra_pages, @@ -2117,6 +2155,7 @@ fail_sb_buffer: btrfs_stop_workers(&fs_info->endio_freespace_worker); btrfs_stop_workers(&fs_info->submit_workers); btrfs_stop_workers(&fs_info->delayed_workers); + btrfs_stop_workers(&fs_info->caching_workers); fail_alloc: kfree(fs_info->delayed_root); fail_iput: @@ -2584,6 +2623,7 @@ int close_ctree(struct btrfs_root *root) btrfs_stop_workers(&fs_info->endio_freespace_worker); btrfs_stop_workers(&fs_info->submit_workers); btrfs_stop_workers(&fs_info->delayed_workers); + btrfs_stop_workers(&fs_info->caching_workers); btrfs_close_devices(fs_info->fs_devices); btrfs_mapping_tree_free(&fs_info->mapping_tree); diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h index a0b610a67aae..bec3ea4bd67f 100644 --- a/fs/btrfs/disk-io.h +++ b/fs/btrfs/disk-io.h @@ -87,10 +87,14 @@ int btree_lock_page_hook(struct page *page); #ifdef CONFIG_DEBUG_LOCK_ALLOC -void btrfs_set_buffer_lockdep_class(struct extent_buffer *eb, int level); +void btrfs_init_lockdep(void); +void btrfs_set_buffer_lockdep_class(u64 objectid, + struct extent_buffer *eb, int level); #else -static inline void btrfs_set_buffer_lockdep_class(struct extent_buffer *eb, - int level) +static inline void btrfs_init_lockdep(void) +{ } +static inline void btrfs_set_buffer_lockdep_class(u64 objectid, + struct extent_buffer *eb, int level) { } #endif diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 6bce721e7bbc..55bddffede73 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -320,12 +320,12 @@ static u64 add_new_free_space(struct btrfs_block_group_cache *block_group, return total_added; } -static int caching_kthread(void *data) +static noinline void caching_thread(struct btrfs_work *work) { - struct btrfs_block_group_cache *block_group = data; - struct btrfs_fs_info *fs_info = block_group->fs_info; - struct btrfs_caching_control *caching_ctl = block_group->caching_ctl; - struct btrfs_root *extent_root = fs_info->extent_root; + struct btrfs_block_group_cache *block_group; + struct btrfs_fs_info *fs_info; + struct btrfs_caching_control *caching_ctl; + struct btrfs_root *extent_root; struct btrfs_path *path; struct extent_buffer *leaf; struct btrfs_key key; @@ -334,9 +334,14 @@ static int caching_kthread(void *data) u32 nritems; int ret = 0; + caching_ctl = container_of(work, struct btrfs_caching_control, work); + block_group = caching_ctl->block_group; + fs_info = block_group->fs_info; + extent_root = fs_info->extent_root; + path = btrfs_alloc_path(); if (!path) - return -ENOMEM; + goto out; last = max_t(u64, block_group->key.objectid, BTRFS_SUPER_INFO_OFFSET); @@ -433,13 +438,11 @@ err: free_excluded_extents(extent_root, block_group); mutex_unlock(&caching_ctl->mutex); +out: wake_up(&caching_ctl->wait); put_caching_control(caching_ctl); - atomic_dec(&block_group->space_info->caching_threads); btrfs_put_block_group(block_group); - - return 0; } static int cache_block_group(struct btrfs_block_group_cache *cache, @@ -449,7 +452,6 @@ static int cache_block_group(struct btrfs_block_group_cache *cache, { struct btrfs_fs_info *fs_info = cache->fs_info; struct btrfs_caching_control *caching_ctl; - struct task_struct *tsk; int ret = 0; smp_mb(); @@ -501,6 +503,7 @@ static int cache_block_group(struct btrfs_block_group_cache *cache, caching_ctl->progress = cache->key.objectid; /* one for caching kthread, one for caching block group list */ atomic_set(&caching_ctl->count, 2); + caching_ctl->work.func = caching_thread; spin_lock(&cache->lock); if (cache->cached != BTRFS_CACHE_NO) { @@ -516,16 +519,9 @@ static int cache_block_group(struct btrfs_block_group_cache *cache, list_add_tail(&caching_ctl->list, &fs_info->caching_block_groups); up_write(&fs_info->extent_commit_sem); - atomic_inc(&cache->space_info->caching_threads); btrfs_get_block_group(cache); - tsk = kthread_run(caching_kthread, cache, "btrfs-cache-%llu\n", - cache->key.objectid); - if (IS_ERR(tsk)) { - ret = PTR_ERR(tsk); - printk(KERN_ERR "error running thread %d\n", ret); - BUG(); - } + btrfs_queue_worker(&fs_info->caching_workers, &caching_ctl->work); return ret; } @@ -2934,9 +2930,10 @@ static int update_space_info(struct btrfs_fs_info *info, u64 flags, found->full = 0; found->force_alloc = CHUNK_ALLOC_NO_FORCE; found->chunk_alloc = 0; + found->flush = 0; + init_waitqueue_head(&found->wait); *space_info = found; list_add_rcu(&found->list, &info->space_info); - atomic_set(&found->caching_threads, 0); return 0; } @@ -3320,6 +3317,14 @@ static int shrink_delalloc(struct btrfs_trans_handle *trans, if (reserved == 0) return 0; + smp_mb(); + if (root->fs_info->delalloc_bytes == 0) { + if (trans) + return 0; + btrfs_wait_ordered_extents(root, 0, 0); + return 0; + } + max_reclaim = min(reserved, to_reclaim); while (loops < 1024) { @@ -3362,6 +3367,8 @@ static int shrink_delalloc(struct btrfs_trans_handle *trans, } } + if (reclaimed >= to_reclaim && !trans) + btrfs_wait_ordered_extents(root, 0, 0); return reclaimed >= to_reclaim; } @@ -3386,15 +3393,36 @@ static int reserve_metadata_bytes(struct btrfs_trans_handle *trans, u64 num_bytes = orig_bytes; int retries = 0; int ret = 0; - bool reserved = false; bool committed = false; + bool flushing = false; again: - ret = -ENOSPC; - if (reserved) - num_bytes = 0; - + ret = 0; spin_lock(&space_info->lock); + /* + * We only want to wait if somebody other than us is flushing and we are + * actually alloed to flush. + */ + while (flush && !flushing && space_info->flush) { + spin_unlock(&space_info->lock); + /* + * If we have a trans handle we can't wait because the flusher + * may have to commit the transaction, which would mean we would + * deadlock since we are waiting for the flusher to finish, but + * hold the current transaction open. + */ + if (trans) + return -EAGAIN; + ret = wait_event_interruptible(space_info->wait, + !space_info->flush); + /* Must have been interrupted, return */ + if (ret) + return -EINTR; + + spin_lock(&space_info->lock); + } + + ret = -ENOSPC; unused = space_info->bytes_used + space_info->bytes_reserved + space_info->bytes_pinned + space_info->bytes_readonly + space_info->bytes_may_use; @@ -3409,8 +3437,7 @@ again: if (unused <= space_info->total_bytes) { unused = space_info->total_bytes - unused; if (unused >= num_bytes) { - if (!reserved) - space_info->bytes_reserved += orig_bytes; + space_info->bytes_reserved += orig_bytes; ret = 0; } else { /* @@ -3435,17 +3462,14 @@ again: * to reclaim space we can actually use it instead of somebody else * stealing it from us. */ - if (ret && !reserved) { - space_info->bytes_reserved += orig_bytes; - reserved = true; + if (ret && flush) { + flushing = true; + space_info->flush = 1; } spin_unlock(&space_info->lock); - if (!ret) - return 0; - - if (!flush) + if (!ret || !flush) goto out; /* @@ -3453,11 +3477,11 @@ again: * metadata until after the IO is completed. */ ret = shrink_delalloc(trans, root, num_bytes, 1); - if (ret > 0) - return 0; - else if (ret < 0) + if (ret < 0) goto out; + ret = 0; + /* * So if we were overcommitted it's possible that somebody else flushed * out enough space and we simply didn't have enough space to reclaim, @@ -3468,11 +3492,11 @@ again: goto again; } - spin_lock(&space_info->lock); /* * Not enough space to be reclaimed, don't bother committing the * transaction. */ + spin_lock(&space_info->lock); if (space_info->bytes_pinned < orig_bytes) ret = -ENOSPC; spin_unlock(&space_info->lock); @@ -3480,10 +3504,13 @@ again: goto out; ret = -EAGAIN; - if (trans || committed) + if (trans) goto out; ret = -ENOSPC; + if (committed) + goto out; + trans = btrfs_join_transaction(root); if (IS_ERR(trans)) goto out; @@ -3495,12 +3522,12 @@ again: } out: - if (reserved) { + if (flushing) { spin_lock(&space_info->lock); - space_info->bytes_reserved -= orig_bytes; + space_info->flush = 0; + wake_up_all(&space_info->wait); spin_unlock(&space_info->lock); } - return ret; } @@ -3710,7 +3737,6 @@ int btrfs_block_rsv_check(struct btrfs_trans_handle *trans, if (commit_trans) { if (trans) return -EAGAIN; - trans = btrfs_join_transaction(root); BUG_ON(IS_ERR(trans)); ret = btrfs_commit_transaction(trans, root); @@ -3880,26 +3906,6 @@ int btrfs_truncate_reserve_metadata(struct btrfs_trans_handle *trans, return 0; } -int btrfs_trans_reserve_metadata(struct btrfs_trans_handle *trans, - struct btrfs_root *root, - int num_items) -{ - u64 num_bytes; - int ret; - - if (num_items == 0 || root->fs_info->chunk_root == root) - return 0; - - num_bytes = btrfs_calc_trans_metadata_size(root, num_items); - ret = btrfs_block_rsv_add(trans, root, &root->fs_info->trans_block_rsv, - num_bytes); - if (!ret) { - trans->bytes_reserved += num_bytes; - trans->block_rsv = &root->fs_info->trans_block_rsv; - } - return ret; -} - void btrfs_trans_release_metadata(struct btrfs_trans_handle *trans, struct btrfs_root *root) { @@ -3950,6 +3956,30 @@ int btrfs_snap_reserve_metadata(struct btrfs_trans_handle *trans, return block_rsv_migrate_bytes(src_rsv, dst_rsv, num_bytes); } +static unsigned drop_outstanding_extent(struct inode *inode) +{ + unsigned dropped_extents = 0; + + spin_lock(&BTRFS_I(inode)->lock); + BUG_ON(!BTRFS_I(inode)->outstanding_extents); + BTRFS_I(inode)->outstanding_extents--; + + /* + * If we have more or the same amount of outsanding extents than we have + * reserved then we need to leave the reserved extents count alone. + */ + if (BTRFS_I(inode)->outstanding_extents >= + BTRFS_I(inode)->reserved_extents) + goto out; + + dropped_extents = BTRFS_I(inode)->reserved_extents - + BTRFS_I(inode)->outstanding_extents; + BTRFS_I(inode)->reserved_extents -= dropped_extents; +out: + spin_unlock(&BTRFS_I(inode)->lock); + return dropped_extents; +} + static u64 calc_csum_metadata_size(struct inode *inode, u64 num_bytes) { return num_bytes >>= 3; @@ -3959,9 +3989,8 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes) { struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_block_rsv *block_rsv = &root->fs_info->delalloc_block_rsv; - u64 to_reserve; - int nr_extents; - int reserved_extents; + u64 to_reserve = 0; + unsigned nr_extents = 0; int ret; if (btrfs_transaction_in_commit(root->fs_info)) @@ -3969,66 +3998,49 @@ int btrfs_delalloc_reserve_metadata(struct inode *inode, u64 num_bytes) num_bytes = ALIGN(num_bytes, root->sectorsize); - nr_extents = atomic_read(&BTRFS_I(inode)->outstanding_extents) + 1; - reserved_extents = atomic_read(&BTRFS_I(inode)->reserved_extents); + spin_lock(&BTRFS_I(inode)->lock); + BTRFS_I(inode)->outstanding_extents++; + + if (BTRFS_I(inode)->outstanding_extents > + BTRFS_I(inode)->reserved_extents) { + nr_extents = BTRFS_I(inode)->outstanding_extents - + BTRFS_I(inode)->reserved_extents; + BTRFS_I(inode)->reserved_extents += nr_extents; - if (nr_extents > reserved_extents) { - nr_extents -= reserved_extents; to_reserve = btrfs_calc_trans_metadata_size(root, nr_extents); - } else { - nr_extents = 0; - to_reserve = 0; } + spin_unlock(&BTRFS_I(inode)->lock); to_reserve += calc_csum_metadata_size(inode, num_bytes); ret = reserve_metadata_bytes(NULL, root, block_rsv, to_reserve, 1); - if (ret) + if (ret) { + unsigned dropped; + /* + * We don't need the return value since our reservation failed, + * we just need to clean up our counter. + */ + dropped = drop_outstanding_extent(inode); + WARN_ON(dropped > 1); return ret; - - atomic_add(nr_extents, &BTRFS_I(inode)->reserved_extents); - atomic_inc(&BTRFS_I(inode)->outstanding_extents); + } block_rsv_add_bytes(block_rsv, to_reserve, 1); - if (block_rsv->size > 512 * 1024 * 1024) - shrink_delalloc(NULL, root, to_reserve, 0); - return 0; } void btrfs_delalloc_release_metadata(struct inode *inode, u64 num_bytes) { struct btrfs_root *root = BTRFS_I(inode)->root; - u64 to_free; - int nr_extents; - int reserved_extents; + u64 to_free = 0; + unsigned dropped; num_bytes = ALIGN(num_bytes, root->sectorsize); - atomic_dec(&BTRFS_I(inode)->outstanding_extents); - WARN_ON(atomic_read(&BTRFS_I(inode)->outstanding_extents) < 0); - - reserved_extents = atomic_read(&BTRFS_I(inode)->reserved_extents); - do { - int old, new; - - nr_extents = atomic_read(&BTRFS_I(inode)->outstanding_extents); - if (nr_extents >= reserved_extents) { - nr_extents = 0; - break; - } - old = reserved_extents; - nr_extents = reserved_extents - nr_extents; - new = reserved_extents - nr_extents; - old = atomic_cmpxchg(&BTRFS_I(inode)->reserved_extents, - reserved_extents, new); - if (likely(old == reserved_extents)) - break; - reserved_extents = old; - } while (1); + dropped = drop_outstanding_extent(inode); to_free = calc_csum_metadata_size(inode, num_bytes); - if (nr_extents > 0) - to_free += btrfs_calc_trans_metadata_size(root, nr_extents); + if (dropped > 0) + to_free += btrfs_calc_trans_metadata_size(root, dropped); btrfs_block_rsv_release(root, &root->fs_info->delalloc_block_rsv, to_free); @@ -4996,14 +5008,10 @@ have_block_group: } /* - * We only want to start kthread caching if we are at - * the point where we will wait for caching to make - * progress, or if our ideal search is over and we've - * found somebody to start caching. + * The caching workers are limited to 2 threads, so we + * can queue as much work as we care to. */ - if (loop > LOOP_CACHING_NOWAIT || - (loop > LOOP_FIND_IDEAL && - atomic_read(&space_info->caching_threads) < 2)) { + if (loop > LOOP_FIND_IDEAL) { ret = cache_block_group(block_group, trans, orig_root, 0); BUG_ON(ret); @@ -5225,8 +5233,7 @@ loop: if (loop == LOOP_FIND_IDEAL && found_uncached_bg) { found_uncached_bg = false; loop++; - if (!ideal_cache_percent && - atomic_read(&space_info->caching_threads)) + if (!ideal_cache_percent) goto search; /* @@ -5630,7 +5637,7 @@ struct extent_buffer *btrfs_init_new_buffer(struct btrfs_trans_handle *trans, if (!buf) return ERR_PTR(-ENOMEM); btrfs_set_header_generation(buf, trans->transid); - btrfs_set_buffer_lockdep_class(buf, level); + btrfs_set_buffer_lockdep_class(root->root_key.objectid, buf, level); btrfs_tree_lock(buf); clean_tree_block(trans, root, buf); @@ -5917,7 +5924,7 @@ static noinline int walk_down_proc(struct btrfs_trans_handle *trans, return 1; if (path->locks[level] && !wc->keep_locks) { - btrfs_tree_unlock(eb); + btrfs_tree_unlock_rw(eb, path->locks[level]); path->locks[level] = 0; } return 0; @@ -5941,7 +5948,7 @@ static noinline int walk_down_proc(struct btrfs_trans_handle *trans, * keep the tree lock */ if (path->locks[level] && level > 0) { - btrfs_tree_unlock(eb); + btrfs_tree_unlock_rw(eb, path->locks[level]); path->locks[level] = 0; } return 0; @@ -6054,7 +6061,7 @@ static noinline int do_walk_down(struct btrfs_trans_handle *trans, BUG_ON(level != btrfs_header_level(next)); path->nodes[level] = next; path->slots[level] = 0; - path->locks[level] = 1; + path->locks[level] = BTRFS_WRITE_LOCK_BLOCKING; wc->level = level; if (wc->level == 1) wc->reada_slot = 0; @@ -6125,7 +6132,7 @@ static noinline int walk_up_proc(struct btrfs_trans_handle *trans, BUG_ON(level == 0); btrfs_tree_lock(eb); btrfs_set_lock_blocking(eb); - path->locks[level] = 1; + path->locks[level] = BTRFS_WRITE_LOCK_BLOCKING; ret = btrfs_lookup_extent_info(trans, root, eb->start, eb->len, @@ -6134,8 +6141,7 @@ static noinline int walk_up_proc(struct btrfs_trans_handle *trans, BUG_ON(ret); BUG_ON(wc->refs[level] == 0); if (wc->refs[level] == 1) { - btrfs_tree_unlock(eb); - path->locks[level] = 0; + btrfs_tree_unlock_rw(eb, path->locks[level]); return 1; } } @@ -6157,7 +6163,7 @@ static noinline int walk_up_proc(struct btrfs_trans_handle *trans, btrfs_header_generation(eb) == trans->transid) { btrfs_tree_lock(eb); btrfs_set_lock_blocking(eb); - path->locks[level] = 1; + path->locks[level] = BTRFS_WRITE_LOCK_BLOCKING; } clean_tree_block(trans, root, eb); } @@ -6236,7 +6242,8 @@ static noinline int walk_up_tree(struct btrfs_trans_handle *trans, return 0; if (path->locks[level]) { - btrfs_tree_unlock(path->nodes[level]); + btrfs_tree_unlock_rw(path->nodes[level], + path->locks[level]); path->locks[level] = 0; } free_extent_buffer(path->nodes[level]); @@ -6292,7 +6299,7 @@ int btrfs_drop_snapshot(struct btrfs_root *root, path->nodes[level] = btrfs_lock_root_node(root); btrfs_set_lock_blocking(path->nodes[level]); path->slots[level] = 0; - path->locks[level] = 1; + path->locks[level] = BTRFS_WRITE_LOCK_BLOCKING; memset(&wc->update_progress, 0, sizeof(wc->update_progress)); } else { @@ -6460,7 +6467,7 @@ int btrfs_drop_subtree(struct btrfs_trans_handle *trans, level = btrfs_header_level(node); path->nodes[level] = node; path->slots[level] = 0; - path->locks[level] = 1; + path->locks[level] = BTRFS_WRITE_LOCK_BLOCKING; wc->refs[parent_level] = 1; wc->flags[parent_level] = BTRFS_BLOCK_FLAG_FULL_BACKREF; @@ -6535,15 +6542,28 @@ static u64 update_block_group_flags(struct btrfs_root *root, u64 flags) return flags; } -static int set_block_group_ro(struct btrfs_block_group_cache *cache) +static int set_block_group_ro(struct btrfs_block_group_cache *cache, int force) { struct btrfs_space_info *sinfo = cache->space_info; u64 num_bytes; + u64 min_allocable_bytes; int ret = -ENOSPC; if (cache->ro) return 0; + /* + * We need some metadata space and system metadata space for + * allocating chunks in some corner cases until we force to set + * it to be readonly. + */ + if ((sinfo->flags & + (BTRFS_BLOCK_GROUP_SYSTEM | BTRFS_BLOCK_GROUP_METADATA)) && + !force) + min_allocable_bytes = 1 * 1024 * 1024; + else + min_allocable_bytes = 0; + spin_lock(&sinfo->lock); spin_lock(&cache->lock); num_bytes = cache->key.offset - cache->reserved - cache->pinned - @@ -6551,7 +6571,8 @@ static int set_block_group_ro(struct btrfs_block_group_cache *cache) if (sinfo->bytes_used + sinfo->bytes_reserved + sinfo->bytes_pinned + sinfo->bytes_may_use + sinfo->bytes_readonly + - cache->reserved_pinned + num_bytes <= sinfo->total_bytes) { + cache->reserved_pinned + num_bytes + min_allocable_bytes <= + sinfo->total_bytes) { sinfo->bytes_readonly += num_bytes; sinfo->bytes_reserved += cache->reserved_pinned; cache->reserved_pinned = 0; @@ -6582,7 +6603,7 @@ int btrfs_set_block_group_ro(struct btrfs_root *root, do_chunk_alloc(trans, root, 2 * 1024 * 1024, alloc_flags, CHUNK_ALLOC_FORCE); - ret = set_block_group_ro(cache); + ret = set_block_group_ro(cache, 0); if (!ret) goto out; alloc_flags = get_alloc_profile(root, cache->space_info->flags); @@ -6590,7 +6611,7 @@ int btrfs_set_block_group_ro(struct btrfs_root *root, CHUNK_ALLOC_FORCE); if (ret < 0) goto out; - ret = set_block_group_ro(cache); + ret = set_block_group_ro(cache, 0); out: btrfs_end_transaction(trans, root); return ret; @@ -7027,7 +7048,7 @@ int btrfs_read_block_groups(struct btrfs_root *root) set_avail_alloc_bits(root->fs_info, cache->flags); if (btrfs_chunk_readonly(root, cache->key.objectid)) - set_block_group_ro(cache); + set_block_group_ro(cache, 1); } list_for_each_entry_rcu(space_info, &root->fs_info->space_info, list) { @@ -7041,9 +7062,9 @@ int btrfs_read_block_groups(struct btrfs_root *root) * mirrored block groups. */ list_for_each_entry(cache, &space_info->block_groups[3], list) - set_block_group_ro(cache); + set_block_group_ro(cache, 1); list_for_each_entry(cache, &space_info->block_groups[4], list) - set_block_group_ro(cache); + set_block_group_ro(cache, 1); } init_global_block_rsv(info); diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 7055d11c1efd..5bbdb243bb6f 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -281,11 +281,10 @@ static int merge_state(struct extent_io_tree *tree, if (other->start == state->end + 1 && other->state == state->state) { merge_cb(tree, state, other); - other->start = state->start; - state->tree = NULL; - rb_erase(&state->rb_node, &tree->state); - free_extent_state(state); - state = NULL; + state->end = other->end; + other->tree = NULL; + rb_erase(&other->rb_node, &tree->state); + free_extent_state(other); } } @@ -351,7 +350,6 @@ static int insert_state(struct extent_io_tree *tree, "%llu %llu\n", (unsigned long long)found->start, (unsigned long long)found->end, (unsigned long long)start, (unsigned long long)end); - free_extent_state(state); return -EEXIST; } state->tree = tree; @@ -500,7 +498,8 @@ again: cached_state = NULL; } - if (cached && cached->tree && cached->start == start) { + if (cached && cached->tree && cached->start <= start && + cached->end > start) { if (clear) atomic_dec(&cached->refs); state = cached; @@ -742,7 +741,8 @@ again: spin_lock(&tree->lock); if (cached_state && *cached_state) { state = *cached_state; - if (state->start == start && state->tree) { + if (state->start <= start && state->end > start && + state->tree) { node = &state->rb_node; goto hit_next; } @@ -783,13 +783,13 @@ hit_next: if (err) goto out; - next_node = rb_next(node); cache_state(state, cached_state); merge_state(tree, state); if (last_end == (u64)-1) goto out; start = last_end + 1; + next_node = rb_next(&state->rb_node); if (next_node && start < end && prealloc && !need_resched()) { state = rb_entry(next_node, struct extent_state, rb_node); @@ -862,7 +862,6 @@ hit_next: * Avoid to free 'prealloc' if it can be merged with * the later extent. */ - atomic_inc(&prealloc->refs); err = insert_state(tree, prealloc, start, this_end, &bits); BUG_ON(err == -EEXIST); @@ -872,7 +871,6 @@ hit_next: goto out; } cache_state(prealloc, cached_state); - free_extent_state(prealloc); prealloc = NULL; start = this_end + 1; goto search_again; @@ -1564,7 +1562,8 @@ int test_range_bit(struct extent_io_tree *tree, u64 start, u64 end, int bitset = 0; spin_lock(&tree->lock); - if (cached && cached->tree && cached->start == start) + if (cached && cached->tree && cached->start <= start && + cached->end > start) node = &cached->rb_node; else node = tree_search(tree, start); @@ -2432,6 +2431,7 @@ static int extent_write_cache_pages(struct extent_io_tree *tree, pgoff_t index; pgoff_t end; /* Inclusive */ int scanned = 0; + int tag; pagevec_init(&pvec, 0); if (wbc->range_cyclic) { @@ -2442,11 +2442,16 @@ static int extent_write_cache_pages(struct extent_io_tree *tree, end = wbc->range_end >> PAGE_CACHE_SHIFT; scanned = 1; } + if (wbc->sync_mode == WB_SYNC_ALL) + tag = PAGECACHE_TAG_TOWRITE; + else + tag = PAGECACHE_TAG_DIRTY; retry: + if (wbc->sync_mode == WB_SYNC_ALL) + tag_pages_for_writeback(mapping, index, end); while (!done && !nr_to_write_done && (index <= end) && - (nr_pages = pagevec_lookup_tag(&pvec, mapping, &index, - PAGECACHE_TAG_DIRTY, min(end - index, - (pgoff_t)PAGEVEC_SIZE-1) + 1))) { + (nr_pages = pagevec_lookup_tag(&pvec, mapping, &index, tag, + min(end - index, (pgoff_t)PAGEVEC_SIZE-1) + 1))) { unsigned i; scanned = 1; @@ -3022,8 +3027,15 @@ static struct extent_buffer *__alloc_extent_buffer(struct extent_io_tree *tree, return NULL; eb->start = start; eb->len = len; - spin_lock_init(&eb->lock); - init_waitqueue_head(&eb->lock_wq); + rwlock_init(&eb->lock); + atomic_set(&eb->write_locks, 0); + atomic_set(&eb->read_locks, 0); + atomic_set(&eb->blocking_readers, 0); + atomic_set(&eb->blocking_writers, 0); + atomic_set(&eb->spinning_readers, 0); + atomic_set(&eb->spinning_writers, 0); + init_waitqueue_head(&eb->write_lock_wq); + init_waitqueue_head(&eb->read_lock_wq); #if LEAK_DEBUG spin_lock_irqsave(&leak_lock, flags); @@ -3119,7 +3131,7 @@ struct extent_buffer *alloc_extent_buffer(struct extent_io_tree *tree, i = 0; } for (; i < num_pages; i++, index++) { - p = find_or_create_page(mapping, index, GFP_NOFS | __GFP_HIGHMEM); + p = find_or_create_page(mapping, index, GFP_NOFS); if (!p) { WARN_ON(1); goto free_eb; @@ -3266,6 +3278,22 @@ int set_extent_buffer_dirty(struct extent_io_tree *tree, return was_dirty; } +static int __eb_straddles_pages(u64 start, u64 len) +{ + if (len < PAGE_CACHE_SIZE) + return 1; + if (start & (PAGE_CACHE_SIZE - 1)) + return 1; + if ((start + len) & (PAGE_CACHE_SIZE - 1)) + return 1; + return 0; +} + +static int eb_straddles_pages(struct extent_buffer *eb) +{ + return __eb_straddles_pages(eb->start, eb->len); +} + int clear_extent_buffer_uptodate(struct extent_io_tree *tree, struct extent_buffer *eb, struct extent_state **cached_state) @@ -3277,8 +3305,10 @@ int clear_extent_buffer_uptodate(struct extent_io_tree *tree, num_pages = num_extent_pages(eb->start, eb->len); clear_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags); - clear_extent_uptodate(tree, eb->start, eb->start + eb->len - 1, - cached_state, GFP_NOFS); + if (eb_straddles_pages(eb)) { + clear_extent_uptodate(tree, eb->start, eb->start + eb->len - 1, + cached_state, GFP_NOFS); + } for (i = 0; i < num_pages; i++) { page = extent_buffer_page(eb, i); if (page) @@ -3296,8 +3326,10 @@ int set_extent_buffer_uptodate(struct extent_io_tree *tree, num_pages = num_extent_pages(eb->start, eb->len); - set_extent_uptodate(tree, eb->start, eb->start + eb->len - 1, - NULL, GFP_NOFS); + if (eb_straddles_pages(eb)) { + set_extent_uptodate(tree, eb->start, eb->start + eb->len - 1, + NULL, GFP_NOFS); + } for (i = 0; i < num_pages; i++) { page = extent_buffer_page(eb, i); if ((i == 0 && (eb->start & (PAGE_CACHE_SIZE - 1))) || @@ -3320,9 +3352,12 @@ int extent_range_uptodate(struct extent_io_tree *tree, int uptodate; unsigned long index; - ret = test_range_bit(tree, start, end, EXTENT_UPTODATE, 1, NULL); - if (ret) - return 1; + if (__eb_straddles_pages(start, end - start + 1)) { + ret = test_range_bit(tree, start, end, + EXTENT_UPTODATE, 1, NULL); + if (ret) + return 1; + } while (start <= end) { index = start >> PAGE_CACHE_SHIFT; page = find_get_page(tree->mapping, index); @@ -3350,10 +3385,12 @@ int extent_buffer_uptodate(struct extent_io_tree *tree, if (test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags)) return 1; - ret = test_range_bit(tree, eb->start, eb->start + eb->len - 1, - EXTENT_UPTODATE, 1, cached_state); - if (ret) - return ret; + if (eb_straddles_pages(eb)) { + ret = test_range_bit(tree, eb->start, eb->start + eb->len - 1, + EXTENT_UPTODATE, 1, cached_state); + if (ret) + return ret; + } num_pages = num_extent_pages(eb->start, eb->len); for (i = 0; i < num_pages; i++) { @@ -3386,9 +3423,11 @@ int read_extent_buffer_pages(struct extent_io_tree *tree, if (test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags)) return 0; - if (test_range_bit(tree, eb->start, eb->start + eb->len - 1, - EXTENT_UPTODATE, 1, NULL)) { - return 0; + if (eb_straddles_pages(eb)) { + if (test_range_bit(tree, eb->start, eb->start + eb->len - 1, + EXTENT_UPTODATE, 1, NULL)) { + return 0; + } } if (start) { @@ -3492,9 +3531,8 @@ void read_extent_buffer(struct extent_buffer *eb, void *dstv, page = extent_buffer_page(eb, i); cur = min(len, (PAGE_CACHE_SIZE - offset)); - kaddr = kmap_atomic(page, KM_USER1); + kaddr = page_address(page); memcpy(dst, kaddr + offset, cur); - kunmap_atomic(kaddr, KM_USER1); dst += cur; len -= cur; @@ -3504,9 +3542,9 @@ void read_extent_buffer(struct extent_buffer *eb, void *dstv, } int map_private_extent_buffer(struct extent_buffer *eb, unsigned long start, - unsigned long min_len, char **token, char **map, + unsigned long min_len, char **map, unsigned long *map_start, - unsigned long *map_len, int km) + unsigned long *map_len) { size_t offset = start & (PAGE_CACHE_SIZE - 1); char *kaddr; @@ -3536,42 +3574,12 @@ int map_private_extent_buffer(struct extent_buffer *eb, unsigned long start, } p = extent_buffer_page(eb, i); - kaddr = kmap_atomic(p, km); - *token = kaddr; + kaddr = page_address(p); *map = kaddr + offset; *map_len = PAGE_CACHE_SIZE - offset; return 0; } -int map_extent_buffer(struct extent_buffer *eb, unsigned long start, - unsigned long min_len, - char **token, char **map, - unsigned long *map_start, - unsigned long *map_len, int km) -{ - int err; - int save = 0; - if (eb->map_token) { - unmap_extent_buffer(eb, eb->map_token, km); - eb->map_token = NULL; - save = 1; - } - err = map_private_extent_buffer(eb, start, min_len, token, map, - map_start, map_len, km); - if (!err && save) { - eb->map_token = *token; - eb->kaddr = *map; - eb->map_start = *map_start; - eb->map_len = *map_len; - } - return err; -} - -void unmap_extent_buffer(struct extent_buffer *eb, char *token, int km) -{ - kunmap_atomic(token, km); -} - int memcmp_extent_buffer(struct extent_buffer *eb, const void *ptrv, unsigned long start, unsigned long len) @@ -3595,9 +3603,8 @@ int memcmp_extent_buffer(struct extent_buffer *eb, const void *ptrv, cur = min(len, (PAGE_CACHE_SIZE - offset)); - kaddr = kmap_atomic(page, KM_USER0); + kaddr = page_address(page); ret = memcmp(ptr, kaddr + offset, cur); - kunmap_atomic(kaddr, KM_USER0); if (ret) break; @@ -3630,9 +3637,8 @@ void write_extent_buffer(struct extent_buffer *eb, const void *srcv, WARN_ON(!PageUptodate(page)); cur = min(len, PAGE_CACHE_SIZE - offset); - kaddr = kmap_atomic(page, KM_USER1); + kaddr = page_address(page); memcpy(kaddr + offset, src, cur); - kunmap_atomic(kaddr, KM_USER1); src += cur; len -= cur; @@ -3661,9 +3667,8 @@ void memset_extent_buffer(struct extent_buffer *eb, char c, WARN_ON(!PageUptodate(page)); cur = min(len, PAGE_CACHE_SIZE - offset); - kaddr = kmap_atomic(page, KM_USER0); + kaddr = page_address(page); memset(kaddr + offset, c, cur); - kunmap_atomic(kaddr, KM_USER0); len -= cur; offset = 0; @@ -3694,9 +3699,8 @@ void copy_extent_buffer(struct extent_buffer *dst, struct extent_buffer *src, cur = min(len, (unsigned long)(PAGE_CACHE_SIZE - offset)); - kaddr = kmap_atomic(page, KM_USER0); + kaddr = page_address(page); read_extent_buffer(src, kaddr + offset, src_offset, cur); - kunmap_atomic(kaddr, KM_USER0); src_offset += cur; len -= cur; @@ -3709,20 +3713,17 @@ static void move_pages(struct page *dst_page, struct page *src_page, unsigned long dst_off, unsigned long src_off, unsigned long len) { - char *dst_kaddr = kmap_atomic(dst_page, KM_USER0); + char *dst_kaddr = page_address(dst_page); if (dst_page == src_page) { memmove(dst_kaddr + dst_off, dst_kaddr + src_off, len); } else { - char *src_kaddr = kmap_atomic(src_page, KM_USER1); + char *src_kaddr = page_address(src_page); char *p = dst_kaddr + dst_off + len; char *s = src_kaddr + src_off + len; while (len--) *--p = *--s; - - kunmap_atomic(src_kaddr, KM_USER1); } - kunmap_atomic(dst_kaddr, KM_USER0); } static inline bool areas_overlap(unsigned long src, unsigned long dst, unsigned long len) @@ -3735,20 +3736,17 @@ static void copy_pages(struct page *dst_page, struct page *src_page, unsigned long dst_off, unsigned long src_off, unsigned long len) { - char *dst_kaddr = kmap_atomic(dst_page, KM_USER0); + char *dst_kaddr = page_address(dst_page); char *src_kaddr; if (dst_page != src_page) { - src_kaddr = kmap_atomic(src_page, KM_USER1); + src_kaddr = page_address(src_page); } else { src_kaddr = dst_kaddr; BUG_ON(areas_overlap(src_off, dst_off, len)); } memcpy(dst_kaddr + dst_off, src_kaddr + src_off, len); - kunmap_atomic(dst_kaddr, KM_USER0); - if (dst_page != src_page) - kunmap_atomic(src_kaddr, KM_USER1); } void memcpy_extent_buffer(struct extent_buffer *dst, unsigned long dst_offset, diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index a11a92ee2d30..21a7ca9e7282 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -120,8 +120,6 @@ struct extent_state { struct extent_buffer { u64 start; unsigned long len; - char *map_token; - char *kaddr; unsigned long map_start; unsigned long map_len; struct page *first_page; @@ -130,14 +128,26 @@ struct extent_buffer { struct rcu_head rcu_head; atomic_t refs; - /* the spinlock is used to protect most operations */ - spinlock_t lock; + /* count of read lock holders on the extent buffer */ + atomic_t write_locks; + atomic_t read_locks; + atomic_t blocking_writers; + atomic_t blocking_readers; + atomic_t spinning_readers; + atomic_t spinning_writers; + + /* protects write locks */ + rwlock_t lock; - /* - * when we keep the lock held while blocking, waiters go onto - * the wq + /* readers use lock_wq while they wait for the write + * lock holders to unlock */ - wait_queue_head_t lock_wq; + wait_queue_head_t write_lock_wq; + + /* writers use read_lock_wq while they wait for readers + * to unlock + */ + wait_queue_head_t read_lock_wq; }; static inline void extent_set_compress_type(unsigned long *bio_flags, @@ -279,15 +289,10 @@ int clear_extent_buffer_uptodate(struct extent_io_tree *tree, int extent_buffer_uptodate(struct extent_io_tree *tree, struct extent_buffer *eb, struct extent_state *cached_state); -int map_extent_buffer(struct extent_buffer *eb, unsigned long offset, - unsigned long min_len, char **token, char **map, - unsigned long *map_start, - unsigned long *map_len, int km); int map_private_extent_buffer(struct extent_buffer *eb, unsigned long offset, - unsigned long min_len, char **token, char **map, + unsigned long min_len, char **map, unsigned long *map_start, - unsigned long *map_len, int km); -void unmap_extent_buffer(struct extent_buffer *eb, char *token, int km); + unsigned long *map_len); int extent_range_uptodate(struct extent_io_tree *tree, u64 start, u64 end); int extent_clear_unlock_delalloc(struct inode *inode, diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index f92ff0ed6e03..b910694f61ed 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -177,6 +177,15 @@ static int __btrfs_lookup_bio_sums(struct btrfs_root *root, WARN_ON(bio->bi_vcnt <= 0); + /* + * the free space stuff is only read when it hasn't been + * updated in the current transaction. So, we can safely + * read from the commit root and sidestep a nasty deadlock + * between reading the free space cache and updating the csum tree. + */ + if (btrfs_is_free_space_inode(root, inode)) + path->search_commit_root = 1; + disk_bytenr = (u64)bio->bi_sector << 9; if (dio) offset = logical_offset; @@ -665,10 +674,6 @@ int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans, struct btrfs_sector_sum *sector_sum; u32 nritems; u32 ins_size; - char *eb_map; - char *eb_token; - unsigned long map_len; - unsigned long map_start; u16 csum_size = btrfs_super_csum_size(&root->fs_info->super_copy); @@ -817,30 +822,9 @@ found: item_end = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_csum_item); item_end = (struct btrfs_csum_item *)((unsigned char *)item_end + btrfs_item_size_nr(leaf, path->slots[0])); - eb_token = NULL; next_sector: - if (!eb_token || - (unsigned long)item + csum_size >= map_start + map_len) { - int err; - - if (eb_token) - unmap_extent_buffer(leaf, eb_token, KM_USER1); - eb_token = NULL; - err = map_private_extent_buffer(leaf, (unsigned long)item, - csum_size, - &eb_token, &eb_map, - &map_start, &map_len, KM_USER1); - if (err) - eb_token = NULL; - } - if (eb_token) { - memcpy(eb_token + ((unsigned long)item & (PAGE_CACHE_SIZE - 1)), - §or_sum->sum, csum_size); - } else { - write_extent_buffer(leaf, §or_sum->sum, - (unsigned long)item, csum_size); - } + write_extent_buffer(leaf, §or_sum->sum, (unsigned long)item, csum_size); total_bytes += root->sectorsize; sector_sum++; @@ -853,10 +837,7 @@ next_sector: goto next_sector; } } - if (eb_token) { - unmap_extent_buffer(leaf, eb_token, KM_USER1); - eb_token = NULL; - } + btrfs_mark_buffer_dirty(path->nodes[0]); if (total_bytes < sums->len) { btrfs_release_path(path); diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 23d1d811e2b2..41ca5fdaee6c 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1082,7 +1082,8 @@ static noinline int prepare_pages(struct btrfs_root *root, struct file *file, again: for (i = 0; i < num_pages; i++) { - pages[i] = grab_cache_page(inode->i_mapping, index + i); + pages[i] = find_or_create_page(inode->i_mapping, index + i, + GFP_NOFS); if (!pages[i]) { faili = i - 1; err = -ENOMEM; @@ -1239,9 +1240,11 @@ static noinline ssize_t __btrfs_buffered_write(struct file *file, * managed to copy. */ if (num_pages > dirty_pages) { - if (copied > 0) - atomic_inc( - &BTRFS_I(inode)->outstanding_extents); + if (copied > 0) { + spin_lock(&BTRFS_I(inode)->lock); + BTRFS_I(inode)->outstanding_extents++; + spin_unlock(&BTRFS_I(inode)->lock); + } btrfs_delalloc_release_space(inode, (num_pages - dirty_pages) << PAGE_CACHE_SHIFT); diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index bf0d61567f3d..6377713f639c 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -98,6 +98,12 @@ struct inode *lookup_free_space_inode(struct btrfs_root *root, return inode; spin_lock(&block_group->lock); + if (BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM) { + printk(KERN_INFO "Old style space inode found, converting.\n"); + BTRFS_I(inode)->flags &= ~BTRFS_INODE_NODATASUM; + block_group->disk_cache_state = BTRFS_DC_CLEAR; + } + if (!btrfs_fs_closing(root->fs_info)) { block_group->inode = igrab(inode); block_group->iref = 1; @@ -135,7 +141,7 @@ int __create_free_space_inode(struct btrfs_root *root, btrfs_set_inode_gid(leaf, inode_item, 0); btrfs_set_inode_mode(leaf, inode_item, S_IFREG | 0600); btrfs_set_inode_flags(leaf, inode_item, BTRFS_INODE_NOCOMPRESS | - BTRFS_INODE_PREALLOC | BTRFS_INODE_NODATASUM); + BTRFS_INODE_PREALLOC); btrfs_set_inode_nlink(leaf, inode_item, 1); btrfs_set_inode_transid(leaf, inode_item, trans->transid); btrfs_set_inode_block_group(leaf, inode_item, offset); @@ -239,17 +245,12 @@ int __load_free_space_cache(struct btrfs_root *root, struct inode *inode, struct btrfs_free_space_header *header; struct extent_buffer *leaf; struct page *page; - u32 *checksums = NULL, *crc; - char *disk_crcs = NULL; struct btrfs_key key; struct list_head bitmaps; u64 num_entries; u64 num_bitmaps; u64 generation; - u32 cur_crc = ~(u32)0; pgoff_t index = 0; - unsigned long first_page_offset; - int num_checksums; int ret = 0; INIT_LIST_HEAD(&bitmaps); @@ -292,16 +293,6 @@ int __load_free_space_cache(struct btrfs_root *root, struct inode *inode, if (!num_entries) goto out; - /* Setup everything for doing checksumming */ - num_checksums = i_size_read(inode) / PAGE_CACHE_SIZE; - checksums = crc = kzalloc(sizeof(u32) * num_checksums, GFP_NOFS); - if (!checksums) - goto out; - first_page_offset = (sizeof(u32) * num_checksums) + sizeof(u64); - disk_crcs = kzalloc(first_page_offset, GFP_NOFS); - if (!disk_crcs) - goto out; - ret = readahead_cache(inode); if (ret) goto out; @@ -311,18 +302,12 @@ int __load_free_space_cache(struct btrfs_root *root, struct inode *inode, struct btrfs_free_space *e; void *addr; unsigned long offset = 0; - unsigned long start_offset = 0; int need_loop = 0; if (!num_entries && !num_bitmaps) break; - if (index == 0) { - start_offset = first_page_offset; - offset = start_offset; - } - - page = grab_cache_page(inode->i_mapping, index); + page = find_or_create_page(inode->i_mapping, index, GFP_NOFS); if (!page) goto free_cache; @@ -342,8 +327,15 @@ int __load_free_space_cache(struct btrfs_root *root, struct inode *inode, if (index == 0) { u64 *gen; - memcpy(disk_crcs, addr, first_page_offset); - gen = addr + (sizeof(u32) * num_checksums); + /* + * We put a bogus crc in the front of the first page in + * case old kernels try to mount a fs with the new + * format to make sure they discard the cache. + */ + addr += sizeof(u64); + offset += sizeof(u64); + + gen = addr; if (*gen != BTRFS_I(inode)->generation) { printk(KERN_ERR "btrfs: space cache generation" " (%llu) does not match inode (%llu)\n", @@ -355,24 +347,10 @@ int __load_free_space_cache(struct btrfs_root *root, struct inode *inode, page_cache_release(page); goto free_cache; } - crc = (u32 *)disk_crcs; - } - entry = addr + start_offset; - - /* First lets check our crc before we do anything fun */ - cur_crc = ~(u32)0; - cur_crc = btrfs_csum_data(root, addr + start_offset, cur_crc, - PAGE_CACHE_SIZE - start_offset); - btrfs_csum_final(cur_crc, (char *)&cur_crc); - if (cur_crc != *crc) { - printk(KERN_ERR "btrfs: crc mismatch for page %lu\n", - index); - kunmap(page); - unlock_page(page); - page_cache_release(page); - goto free_cache; + addr += sizeof(u64); + offset += sizeof(u64); } - crc++; + entry = addr; while (1) { if (!num_entries) @@ -470,8 +448,6 @@ next: ret = 1; out: - kfree(checksums); - kfree(disk_crcs); return ret; free_cache: __btrfs_remove_free_space_cache(ctl); @@ -569,8 +545,7 @@ int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, struct btrfs_key key; u64 start, end, len; u64 bytes = 0; - u32 *crc, *checksums; - unsigned long first_page_offset; + u32 crc = ~(u32)0; int index = 0, num_pages = 0; int entries = 0; int bitmaps = 0; @@ -590,34 +565,13 @@ int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, num_pages = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; - /* Since the first page has all of our checksums and our generation we - * need to calculate the offset into the page that we can start writing - * our entries. - */ - first_page_offset = (sizeof(u32) * num_pages) + sizeof(u64); - filemap_write_and_wait(inode->i_mapping); btrfs_wait_ordered_range(inode, inode->i_size & ~(root->sectorsize - 1), (u64)-1); - /* make sure we don't overflow that first page */ - if (first_page_offset + sizeof(struct btrfs_free_space_entry) >= PAGE_CACHE_SIZE) { - /* this is really the same as running out of space, where we also return 0 */ - printk(KERN_CRIT "Btrfs: free space cache was too big for the crc page\n"); - ret = 0; - goto out_update; - } - - /* We need a checksum per page. */ - crc = checksums = kzalloc(sizeof(u32) * num_pages, GFP_NOFS); - if (!crc) - return -1; - pages = kzalloc(sizeof(struct page *) * num_pages, GFP_NOFS); - if (!pages) { - kfree(crc); + if (!pages) return -1; - } /* Get the cluster for this block_group if it exists */ if (block_group && !list_empty(&block_group->cluster_list)) @@ -640,7 +594,7 @@ int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, * know and don't freak out. */ while (index < num_pages) { - page = grab_cache_page(inode->i_mapping, index); + page = find_or_create_page(inode->i_mapping, index, GFP_NOFS); if (!page) { int i; @@ -648,7 +602,7 @@ int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, unlock_page(pages[i]); page_cache_release(pages[i]); } - goto out_free; + goto out; } pages[index] = page; index++; @@ -668,17 +622,11 @@ int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, /* Write out the extent entries */ do { struct btrfs_free_space_entry *entry; - void *addr; + void *addr, *orig; unsigned long offset = 0; - unsigned long start_offset = 0; next_page = false; - if (index == 0) { - start_offset = first_page_offset; - offset = start_offset; - } - if (index >= num_pages) { out_of_space = true; break; @@ -686,10 +634,26 @@ int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, page = pages[index]; - addr = kmap(page); - entry = addr + start_offset; + orig = addr = kmap(page); + if (index == 0) { + u64 *gen; - memset(addr, 0, PAGE_CACHE_SIZE); + /* + * We're going to put in a bogus crc for this page to + * make sure that old kernels who aren't aware of this + * format will be sure to discard the cache. + */ + addr += sizeof(u64); + offset += sizeof(u64); + + gen = addr; + *gen = trans->transid; + addr += sizeof(u64); + offset += sizeof(u64); + } + entry = addr; + + memset(addr, 0, PAGE_CACHE_SIZE - offset); while (node && !next_page) { struct btrfs_free_space *e; @@ -752,13 +716,19 @@ int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, next_page = true; entry++; } - *crc = ~(u32)0; - *crc = btrfs_csum_data(root, addr + start_offset, *crc, - PAGE_CACHE_SIZE - start_offset); - kunmap(page); - btrfs_csum_final(*crc, (char *)crc); - crc++; + /* Generate bogus crc value */ + if (index == 0) { + u32 *tmp; + crc = btrfs_csum_data(root, orig + sizeof(u64), crc, + PAGE_CACHE_SIZE - sizeof(u64)); + btrfs_csum_final(crc, (char *)&crc); + crc++; + tmp = orig; + *tmp = crc; + } + + kunmap(page); bytes += PAGE_CACHE_SIZE; @@ -779,11 +749,7 @@ int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, addr = kmap(page); memcpy(addr, entry->bitmap, PAGE_CACHE_SIZE); - *crc = ~(u32)0; - *crc = btrfs_csum_data(root, addr, *crc, PAGE_CACHE_SIZE); kunmap(page); - btrfs_csum_final(*crc, (char *)crc); - crc++; bytes += PAGE_CACHE_SIZE; list_del_init(&entry->list); @@ -796,7 +762,7 @@ int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, i_size_read(inode) - 1, &cached_state, GFP_NOFS); ret = 0; - goto out_free; + goto out; } /* Zero out the rest of the pages just to make sure */ @@ -811,20 +777,6 @@ int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, index++; } - /* Write the checksums and trans id to the first page */ - { - void *addr; - u64 *gen; - - page = pages[0]; - - addr = kmap(page); - memcpy(addr, checksums, sizeof(u32) * num_pages); - gen = addr + (sizeof(u32) * num_pages); - *gen = trans->transid; - kunmap(page); - } - ret = btrfs_dirty_pages(root, inode, pages, num_pages, 0, bytes, &cached_state); btrfs_drop_pages(pages, num_pages); @@ -833,7 +785,7 @@ int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, if (ret) { ret = 0; - goto out_free; + goto out; } BTRFS_I(inode)->generation = trans->transid; @@ -850,7 +802,7 @@ int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, clear_extent_bit(&BTRFS_I(inode)->io_tree, 0, bytes - 1, EXTENT_DIRTY | EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING, 0, 0, NULL, GFP_NOFS); - goto out_free; + goto out; } leaf = path->nodes[0]; if (ret > 0) { @@ -866,7 +818,7 @@ int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, EXTENT_DO_ACCOUNTING, 0, 0, NULL, GFP_NOFS); btrfs_release_path(path); - goto out_free; + goto out; } } header = btrfs_item_ptr(leaf, path->slots[0], @@ -879,11 +831,8 @@ int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, ret = 1; -out_free: - kfree(checksums); +out: kfree(pages); - -out_update: if (ret != 1) { invalidate_inode_pages2_range(inode->i_mapping, 0, index); BTRFS_I(inode)->generation = 0; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 88829993db6c..4360ccb191b1 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -750,15 +750,6 @@ static u64 get_extent_allocation_hint(struct inode *inode, u64 start, return alloc_hint; } -static inline bool is_free_space_inode(struct btrfs_root *root, - struct inode *inode) -{ - if (root == root->fs_info->tree_root || - BTRFS_I(inode)->location.objectid == BTRFS_FREE_INO_OBJECTID) - return true; - return false; -} - /* * when extent_io.c finds a delayed allocation range in the file, * the call backs end up in this code. The basic idea is to @@ -791,7 +782,7 @@ static noinline int cow_file_range(struct inode *inode, struct extent_map_tree *em_tree = &BTRFS_I(inode)->extent_tree; int ret = 0; - BUG_ON(is_free_space_inode(root, inode)); + BUG_ON(btrfs_is_free_space_inode(root, inode)); trans = btrfs_join_transaction(root); BUG_ON(IS_ERR(trans)); trans->block_rsv = &root->fs_info->delalloc_block_rsv; @@ -1073,7 +1064,7 @@ static noinline int run_delalloc_nocow(struct inode *inode, if (!path) return -ENOMEM; - nolock = is_free_space_inode(root, inode); + nolock = btrfs_is_free_space_inode(root, inode); if (nolock) trans = btrfs_join_transaction_nolock(root); @@ -1299,7 +1290,9 @@ static int btrfs_split_extent_hook(struct inode *inode, if (!(orig->state & EXTENT_DELALLOC)) return 0; - atomic_inc(&BTRFS_I(inode)->outstanding_extents); + spin_lock(&BTRFS_I(inode)->lock); + BTRFS_I(inode)->outstanding_extents++; + spin_unlock(&BTRFS_I(inode)->lock); return 0; } @@ -1317,7 +1310,9 @@ static int btrfs_merge_extent_hook(struct inode *inode, if (!(other->state & EXTENT_DELALLOC)) return 0; - atomic_dec(&BTRFS_I(inode)->outstanding_extents); + spin_lock(&BTRFS_I(inode)->lock); + BTRFS_I(inode)->outstanding_extents--; + spin_unlock(&BTRFS_I(inode)->lock); return 0; } @@ -1338,12 +1333,15 @@ static int btrfs_set_bit_hook(struct inode *inode, if (!(state->state & EXTENT_DELALLOC) && (*bits & EXTENT_DELALLOC)) { struct btrfs_root *root = BTRFS_I(inode)->root; u64 len = state->end + 1 - state->start; - bool do_list = !is_free_space_inode(root, inode); + bool do_list = !btrfs_is_free_space_inode(root, inode); - if (*bits & EXTENT_FIRST_DELALLOC) + if (*bits & EXTENT_FIRST_DELALLOC) { *bits &= ~EXTENT_FIRST_DELALLOC; - else - atomic_inc(&BTRFS_I(inode)->outstanding_extents); + } else { + spin_lock(&BTRFS_I(inode)->lock); + BTRFS_I(inode)->outstanding_extents++; + spin_unlock(&BTRFS_I(inode)->lock); + } spin_lock(&root->fs_info->delalloc_lock); BTRFS_I(inode)->delalloc_bytes += len; @@ -1371,12 +1369,15 @@ static int btrfs_clear_bit_hook(struct inode *inode, if ((state->state & EXTENT_DELALLOC) && (*bits & EXTENT_DELALLOC)) { struct btrfs_root *root = BTRFS_I(inode)->root; u64 len = state->end + 1 - state->start; - bool do_list = !is_free_space_inode(root, inode); + bool do_list = !btrfs_is_free_space_inode(root, inode); - if (*bits & EXTENT_FIRST_DELALLOC) + if (*bits & EXTENT_FIRST_DELALLOC) { *bits &= ~EXTENT_FIRST_DELALLOC; - else if (!(*bits & EXTENT_DO_ACCOUNTING)) - atomic_dec(&BTRFS_I(inode)->outstanding_extents); + } else if (!(*bits & EXTENT_DO_ACCOUNTING)) { + spin_lock(&BTRFS_I(inode)->lock); + BTRFS_I(inode)->outstanding_extents--; + spin_unlock(&BTRFS_I(inode)->lock); + } if (*bits & EXTENT_DO_ACCOUNTING) btrfs_delalloc_release_metadata(inode, len); @@ -1478,7 +1479,7 @@ static int btrfs_submit_bio_hook(struct inode *inode, int rw, struct bio *bio, skip_sum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM; - if (is_free_space_inode(root, inode)) + if (btrfs_is_free_space_inode(root, inode)) ret = btrfs_bio_wq_end_io(root->fs_info, bio, 2); else ret = btrfs_bio_wq_end_io(root->fs_info, bio, 0); @@ -1728,7 +1729,7 @@ static int btrfs_finish_ordered_io(struct inode *inode, u64 start, u64 end) return 0; BUG_ON(!ordered_extent); - nolock = is_free_space_inode(root, inode); + nolock = btrfs_is_free_space_inode(root, inode); if (test_bit(BTRFS_ORDERED_NOCOW, &ordered_extent->flags)) { BUG_ON(!list_empty(&ordered_extent->list)); @@ -2535,13 +2536,6 @@ static void btrfs_read_locked_inode(struct inode *inode) inode_item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_inode_item); - if (!leaf->map_token) - map_private_extent_buffer(leaf, (unsigned long)inode_item, - sizeof(struct btrfs_inode_item), - &leaf->map_token, &leaf->kaddr, - &leaf->map_start, &leaf->map_len, - KM_USER1); - inode->i_mode = btrfs_inode_mode(leaf, inode_item); inode->i_nlink = btrfs_inode_nlink(leaf, inode_item); inode->i_uid = btrfs_inode_uid(leaf, inode_item); @@ -2579,11 +2573,6 @@ cache_acl: if (!maybe_acls) cache_no_acl(inode); - if (leaf->map_token) { - unmap_extent_buffer(leaf, leaf->map_token, KM_USER1); - leaf->map_token = NULL; - } - btrfs_free_path(path); switch (inode->i_mode & S_IFMT) { @@ -2628,13 +2617,6 @@ static void fill_inode_item(struct btrfs_trans_handle *trans, struct btrfs_inode_item *item, struct inode *inode) { - if (!leaf->map_token) - map_private_extent_buffer(leaf, (unsigned long)item, - sizeof(struct btrfs_inode_item), - &leaf->map_token, &leaf->kaddr, - &leaf->map_start, &leaf->map_len, - KM_USER1); - btrfs_set_inode_uid(leaf, item, inode->i_uid); btrfs_set_inode_gid(leaf, item, inode->i_gid); btrfs_set_inode_size(leaf, item, BTRFS_I(inode)->disk_i_size); @@ -2663,11 +2645,6 @@ static void fill_inode_item(struct btrfs_trans_handle *trans, btrfs_set_inode_rdev(leaf, item, inode->i_rdev); btrfs_set_inode_flags(leaf, item, BTRFS_I(inode)->flags); btrfs_set_inode_block_group(leaf, item, 0); - - if (leaf->map_token) { - unmap_extent_buffer(leaf, leaf->map_token, KM_USER1); - leaf->map_token = NULL; - } } /* @@ -2688,7 +2665,7 @@ noinline int btrfs_update_inode(struct btrfs_trans_handle *trans, * The data relocation inode should also be directly updated * without delay */ - if (!is_free_space_inode(root, inode) + if (!btrfs_is_free_space_inode(root, inode) && root->root_key.objectid != BTRFS_DATA_RELOC_TREE_OBJECTID) { ret = btrfs_delayed_update_inode(trans, root, inode); if (!ret) @@ -3403,7 +3380,7 @@ static int btrfs_truncate_page(struct address_space *mapping, loff_t from) ret = -ENOMEM; again: - page = grab_cache_page(mapping, index); + page = find_or_create_page(mapping, index, GFP_NOFS); if (!page) { btrfs_delalloc_release_space(inode, PAGE_CACHE_SIZE); goto out; @@ -3639,7 +3616,7 @@ void btrfs_evict_inode(struct inode *inode) truncate_inode_pages(&inode->i_data, 0); if (inode->i_nlink && (btrfs_root_refs(&root->root_item) != 0 || - is_free_space_inode(root, inode))) + btrfs_is_free_space_inode(root, inode))) goto no_delete; if (is_bad_inode(inode)) { @@ -4293,7 +4270,7 @@ int btrfs_write_inode(struct inode *inode, struct writeback_control *wbc) if (BTRFS_I(inode)->dummy_inode) return 0; - if (btrfs_fs_closing(root->fs_info) && is_free_space_inode(root, inode)) + if (btrfs_fs_closing(root->fs_info) && btrfs_is_free_space_inode(root, inode)) nolock = true; if (wbc->sync_mode == WB_SYNC_ALL) { @@ -6752,8 +6729,9 @@ struct inode *btrfs_alloc_inode(struct super_block *sb) ei->index_cnt = (u64)-1; ei->last_unlink_trans = 0; - atomic_set(&ei->outstanding_extents, 0); - atomic_set(&ei->reserved_extents, 0); + spin_lock_init(&ei->lock); + ei->outstanding_extents = 0; + ei->reserved_extents = 0; ei->ordered_data_close = 0; ei->orphan_meta_reserved = 0; @@ -6791,8 +6769,8 @@ void btrfs_destroy_inode(struct inode *inode) WARN_ON(!list_empty(&inode->i_dentry)); WARN_ON(inode->i_data.nrpages); - WARN_ON(atomic_read(&BTRFS_I(inode)->outstanding_extents)); - WARN_ON(atomic_read(&BTRFS_I(inode)->reserved_extents)); + WARN_ON(BTRFS_I(inode)->outstanding_extents); + WARN_ON(BTRFS_I(inode)->reserved_extents); /* * This can happen where we create an inode, but somebody else also @@ -6847,7 +6825,7 @@ int btrfs_drop_inode(struct inode *inode) struct btrfs_root *root = BTRFS_I(inode)->root; if (btrfs_root_refs(&root->root_item) == 0 && - !is_free_space_inode(root, inode)) + !btrfs_is_free_space_inode(root, inode)) return 1; else return generic_drop_inode(inode); diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index a3c4751e07db..fd252fff4c66 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -867,8 +867,8 @@ again: /* step one, lock all the pages */ for (i = 0; i < num_pages; i++) { struct page *page; - page = grab_cache_page(inode->i_mapping, - start_index + i); + page = find_or_create_page(inode->i_mapping, + start_index + i, GFP_NOFS); if (!page) break; @@ -938,7 +938,9 @@ again: GFP_NOFS); if (i_done != num_pages) { - atomic_inc(&BTRFS_I(inode)->outstanding_extents); + spin_lock(&BTRFS_I(inode)->lock); + BTRFS_I(inode)->outstanding_extents++; + spin_unlock(&BTRFS_I(inode)->lock); btrfs_delalloc_release_space(inode, (num_pages - i_done) << PAGE_CACHE_SHIFT); } diff --git a/fs/btrfs/locking.c b/fs/btrfs/locking.c index 66fa43dc3f0f..d77b67c4b275 100644 --- a/fs/btrfs/locking.c +++ b/fs/btrfs/locking.c @@ -24,185 +24,197 @@ #include "extent_io.h" #include "locking.h" -static inline void spin_nested(struct extent_buffer *eb) -{ - spin_lock(&eb->lock); -} +void btrfs_assert_tree_read_locked(struct extent_buffer *eb); /* - * Setting a lock to blocking will drop the spinlock and set the - * flag that forces other procs who want the lock to wait. After - * this you can safely schedule with the lock held. + * if we currently have a spinning reader or writer lock + * (indicated by the rw flag) this will bump the count + * of blocking holders and drop the spinlock. */ -void btrfs_set_lock_blocking(struct extent_buffer *eb) +void btrfs_set_lock_blocking_rw(struct extent_buffer *eb, int rw) { - if (!test_bit(EXTENT_BUFFER_BLOCKING, &eb->bflags)) { - set_bit(EXTENT_BUFFER_BLOCKING, &eb->bflags); - spin_unlock(&eb->lock); + if (rw == BTRFS_WRITE_LOCK) { + if (atomic_read(&eb->blocking_writers) == 0) { + WARN_ON(atomic_read(&eb->spinning_writers) != 1); + atomic_dec(&eb->spinning_writers); + btrfs_assert_tree_locked(eb); + atomic_inc(&eb->blocking_writers); + write_unlock(&eb->lock); + } + } else if (rw == BTRFS_READ_LOCK) { + btrfs_assert_tree_read_locked(eb); + atomic_inc(&eb->blocking_readers); + WARN_ON(atomic_read(&eb->spinning_readers) == 0); + atomic_dec(&eb->spinning_readers); + read_unlock(&eb->lock); } - /* exit with the spin lock released and the bit set */ + return; } /* - * clearing the blocking flag will take the spinlock again. - * After this you can't safely schedule + * if we currently have a blocking lock, take the spinlock + * and drop our blocking count */ -void btrfs_clear_lock_blocking(struct extent_buffer *eb) +void btrfs_clear_lock_blocking_rw(struct extent_buffer *eb, int rw) { - if (test_bit(EXTENT_BUFFER_BLOCKING, &eb->bflags)) { - spin_nested(eb); - clear_bit(EXTENT_BUFFER_BLOCKING, &eb->bflags); - smp_mb__after_clear_bit(); + if (rw == BTRFS_WRITE_LOCK_BLOCKING) { + BUG_ON(atomic_read(&eb->blocking_writers) != 1); + write_lock(&eb->lock); + WARN_ON(atomic_read(&eb->spinning_writers)); + atomic_inc(&eb->spinning_writers); + if (atomic_dec_and_test(&eb->blocking_writers)) + wake_up(&eb->write_lock_wq); + } else if (rw == BTRFS_READ_LOCK_BLOCKING) { + BUG_ON(atomic_read(&eb->blocking_readers) == 0); + read_lock(&eb->lock); + atomic_inc(&eb->spinning_readers); + if (atomic_dec_and_test(&eb->blocking_readers)) + wake_up(&eb->read_lock_wq); } - /* exit with the spin lock held */ + return; } /* - * unfortunately, many of the places that currently set a lock to blocking - * don't end up blocking for very long, and often they don't block - * at all. For a dbench 50 run, if we don't spin on the blocking bit - * at all, the context switch rate can jump up to 400,000/sec or more. - * - * So, we're still stuck with this crummy spin on the blocking bit, - * at least until the most common causes of the short blocks - * can be dealt with. + * take a spinning read lock. This will wait for any blocking + * writers */ -static int btrfs_spin_on_block(struct extent_buffer *eb) +void btrfs_tree_read_lock(struct extent_buffer *eb) { - int i; - - for (i = 0; i < 512; i++) { - if (!test_bit(EXTENT_BUFFER_BLOCKING, &eb->bflags)) - return 1; - if (need_resched()) - break; - cpu_relax(); +again: + wait_event(eb->write_lock_wq, atomic_read(&eb->blocking_writers) == 0); + read_lock(&eb->lock); + if (atomic_read(&eb->blocking_writers)) { + read_unlock(&eb->lock); + wait_event(eb->write_lock_wq, + atomic_read(&eb->blocking_writers) == 0); + goto again; } - return 0; + atomic_inc(&eb->read_locks); + atomic_inc(&eb->spinning_readers); } /* - * This is somewhat different from trylock. It will take the - * spinlock but if it finds the lock is set to blocking, it will - * return without the lock held. - * - * returns 1 if it was able to take the lock and zero otherwise - * - * After this call, scheduling is not safe without first calling - * btrfs_set_lock_blocking() + * returns 1 if we get the read lock and 0 if we don't + * this won't wait for blocking writers */ -int btrfs_try_spin_lock(struct extent_buffer *eb) +int btrfs_try_tree_read_lock(struct extent_buffer *eb) { - int i; + if (atomic_read(&eb->blocking_writers)) + return 0; - if (btrfs_spin_on_block(eb)) { - spin_nested(eb); - if (!test_bit(EXTENT_BUFFER_BLOCKING, &eb->bflags)) - return 1; - spin_unlock(&eb->lock); + read_lock(&eb->lock); + if (atomic_read(&eb->blocking_writers)) { + read_unlock(&eb->lock); + return 0; } - /* spin for a bit on the BLOCKING flag */ - for (i = 0; i < 2; i++) { - cpu_relax(); - if (!btrfs_spin_on_block(eb)) - break; - - spin_nested(eb); - if (!test_bit(EXTENT_BUFFER_BLOCKING, &eb->bflags)) - return 1; - spin_unlock(&eb->lock); - } - return 0; + atomic_inc(&eb->read_locks); + atomic_inc(&eb->spinning_readers); + return 1; } /* - * the autoremove wake function will return 0 if it tried to wake up - * a process that was already awake, which means that process won't - * count as an exclusive wakeup. The waitq code will continue waking - * procs until it finds one that was actually sleeping. - * - * For btrfs, this isn't quite what we want. We want a single proc - * to be notified that the lock is ready for taking. If that proc - * already happen to be awake, great, it will loop around and try for - * the lock. - * - * So, btrfs_wake_function always returns 1, even when the proc that we - * tried to wake up was already awake. + * returns 1 if we get the read lock and 0 if we don't + * this won't wait for blocking writers or readers */ -static int btrfs_wake_function(wait_queue_t *wait, unsigned mode, - int sync, void *key) +int btrfs_try_tree_write_lock(struct extent_buffer *eb) { - autoremove_wake_function(wait, mode, sync, key); + if (atomic_read(&eb->blocking_writers) || + atomic_read(&eb->blocking_readers)) + return 0; + write_lock(&eb->lock); + if (atomic_read(&eb->blocking_writers) || + atomic_read(&eb->blocking_readers)) { + write_unlock(&eb->lock); + return 0; + } + atomic_inc(&eb->write_locks); + atomic_inc(&eb->spinning_writers); return 1; } /* - * returns with the extent buffer spinlocked. - * - * This will spin and/or wait as required to take the lock, and then - * return with the spinlock held. - * - * After this call, scheduling is not safe without first calling - * btrfs_set_lock_blocking() + * drop a spinning read lock + */ +void btrfs_tree_read_unlock(struct extent_buffer *eb) +{ + btrfs_assert_tree_read_locked(eb); + WARN_ON(atomic_read(&eb->spinning_readers) == 0); + atomic_dec(&eb->spinning_readers); + atomic_dec(&eb->read_locks); + read_unlock(&eb->lock); +} + +/* + * drop a blocking read lock + */ +void btrfs_tree_read_unlock_blocking(struct extent_buffer *eb) +{ + btrfs_assert_tree_read_locked(eb); + WARN_ON(atomic_read(&eb->blocking_readers) == 0); + if (atomic_dec_and_test(&eb->blocking_readers)) + wake_up(&eb->read_lock_wq); + atomic_dec(&eb->read_locks); +} + +/* + * take a spinning write lock. This will wait for both + * blocking readers or writers */ int btrfs_tree_lock(struct extent_buffer *eb) { - DEFINE_WAIT(wait); - wait.func = btrfs_wake_function; - - if (!btrfs_spin_on_block(eb)) - goto sleep; - - while(1) { - spin_nested(eb); - - /* nobody is blocking, exit with the spinlock held */ - if (!test_bit(EXTENT_BUFFER_BLOCKING, &eb->bflags)) - return 0; - - /* - * we have the spinlock, but the real owner is blocking. - * wait for them - */ - spin_unlock(&eb->lock); - - /* - * spin for a bit, and if the blocking flag goes away, - * loop around - */ - cpu_relax(); - if (btrfs_spin_on_block(eb)) - continue; -sleep: - prepare_to_wait_exclusive(&eb->lock_wq, &wait, - TASK_UNINTERRUPTIBLE); - - if (test_bit(EXTENT_BUFFER_BLOCKING, &eb->bflags)) - schedule(); - - finish_wait(&eb->lock_wq, &wait); +again: + wait_event(eb->read_lock_wq, atomic_read(&eb->blocking_readers) == 0); + wait_event(eb->write_lock_wq, atomic_read(&eb->blocking_writers) == 0); + write_lock(&eb->lock); + if (atomic_read(&eb->blocking_readers)) { + write_unlock(&eb->lock); + wait_event(eb->read_lock_wq, + atomic_read(&eb->blocking_readers) == 0); + goto again; } + if (atomic_read(&eb->blocking_writers)) { + write_unlock(&eb->lock); + wait_event(eb->write_lock_wq, + atomic_read(&eb->blocking_writers) == 0); + goto again; + } + WARN_ON(atomic_read(&eb->spinning_writers)); + atomic_inc(&eb->spinning_writers); + atomic_inc(&eb->write_locks); return 0; } +/* + * drop a spinning or a blocking write lock. + */ int btrfs_tree_unlock(struct extent_buffer *eb) { - /* - * if we were a blocking owner, we don't have the spinlock held - * just clear the bit and look for waiters - */ - if (test_and_clear_bit(EXTENT_BUFFER_BLOCKING, &eb->bflags)) - smp_mb__after_clear_bit(); - else - spin_unlock(&eb->lock); - - if (waitqueue_active(&eb->lock_wq)) - wake_up(&eb->lock_wq); + int blockers = atomic_read(&eb->blocking_writers); + + BUG_ON(blockers > 1); + + btrfs_assert_tree_locked(eb); + atomic_dec(&eb->write_locks); + + if (blockers) { + WARN_ON(atomic_read(&eb->spinning_writers)); + atomic_dec(&eb->blocking_writers); + smp_wmb(); + wake_up(&eb->write_lock_wq); + } else { + WARN_ON(atomic_read(&eb->spinning_writers) != 1); + atomic_dec(&eb->spinning_writers); + write_unlock(&eb->lock); + } return 0; } void btrfs_assert_tree_locked(struct extent_buffer *eb) { - if (!test_bit(EXTENT_BUFFER_BLOCKING, &eb->bflags)) - assert_spin_locked(&eb->lock); + BUG_ON(!atomic_read(&eb->write_locks)); +} + +void btrfs_assert_tree_read_locked(struct extent_buffer *eb) +{ + BUG_ON(!atomic_read(&eb->read_locks)); } diff --git a/fs/btrfs/locking.h b/fs/btrfs/locking.h index 5c33a560a2f1..17247ddb81a0 100644 --- a/fs/btrfs/locking.h +++ b/fs/btrfs/locking.h @@ -19,11 +19,43 @@ #ifndef __BTRFS_LOCKING_ #define __BTRFS_LOCKING_ +#define BTRFS_WRITE_LOCK 1 +#define BTRFS_READ_LOCK 2 +#define BTRFS_WRITE_LOCK_BLOCKING 3 +#define BTRFS_READ_LOCK_BLOCKING 4 + int btrfs_tree_lock(struct extent_buffer *eb); int btrfs_tree_unlock(struct extent_buffer *eb); int btrfs_try_spin_lock(struct extent_buffer *eb); -void btrfs_set_lock_blocking(struct extent_buffer *eb); -void btrfs_clear_lock_blocking(struct extent_buffer *eb); +void btrfs_tree_read_lock(struct extent_buffer *eb); +void btrfs_tree_read_unlock(struct extent_buffer *eb); +void btrfs_tree_read_unlock_blocking(struct extent_buffer *eb); +void btrfs_set_lock_blocking_rw(struct extent_buffer *eb, int rw); +void btrfs_clear_lock_blocking_rw(struct extent_buffer *eb, int rw); void btrfs_assert_tree_locked(struct extent_buffer *eb); +int btrfs_try_tree_read_lock(struct extent_buffer *eb); +int btrfs_try_tree_write_lock(struct extent_buffer *eb); + +static inline void btrfs_tree_unlock_rw(struct extent_buffer *eb, int rw) +{ + if (rw == BTRFS_WRITE_LOCK || rw == BTRFS_WRITE_LOCK_BLOCKING) + btrfs_tree_unlock(eb); + else if (rw == BTRFS_READ_LOCK_BLOCKING) + btrfs_tree_read_unlock_blocking(eb); + else if (rw == BTRFS_READ_LOCK) + btrfs_tree_read_unlock(eb); + else + BUG(); +} + +static inline void btrfs_set_lock_blocking(struct extent_buffer *eb) +{ + btrfs_set_lock_blocking_rw(eb, BTRFS_WRITE_LOCK); +} + +static inline void btrfs_clear_lock_blocking(struct extent_buffer *eb) +{ + btrfs_clear_lock_blocking_rw(eb, BTRFS_WRITE_LOCK_BLOCKING); +} #endif diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 5e0a3dc79a45..59bb1764273d 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -2955,7 +2955,8 @@ static int relocate_file_extent_cluster(struct inode *inode, page_cache_sync_readahead(inode->i_mapping, ra, NULL, index, last_index + 1 - index); - page = grab_cache_page(inode->i_mapping, index); + page = find_or_create_page(inode->i_mapping, index, + GFP_NOFS); if (!page) { btrfs_delalloc_release_metadata(inode, PAGE_CACHE_SIZE); diff --git a/fs/btrfs/struct-funcs.c b/fs/btrfs/struct-funcs.c index c0f7ecaf1e79..bc1f6ad18442 100644 --- a/fs/btrfs/struct-funcs.c +++ b/fs/btrfs/struct-funcs.c @@ -50,36 +50,22 @@ u##bits btrfs_##name(struct extent_buffer *eb, \ unsigned long part_offset = (unsigned long)s; \ unsigned long offset = part_offset + offsetof(type, member); \ type *p; \ - /* ugly, but we want the fast path here */ \ - if (eb->map_token && offset >= eb->map_start && \ - offset + sizeof(((type *)0)->member) <= eb->map_start + \ - eb->map_len) { \ - p = (type *)(eb->kaddr + part_offset - eb->map_start); \ - return le##bits##_to_cpu(p->member); \ - } \ - { \ - int err; \ - char *map_token; \ - char *kaddr; \ - int unmap_on_exit = (eb->map_token == NULL); \ - unsigned long map_start; \ - unsigned long map_len; \ - u##bits res; \ - err = map_extent_buffer(eb, offset, \ - sizeof(((type *)0)->member), \ - &map_token, &kaddr, \ - &map_start, &map_len, KM_USER1); \ - if (err) { \ - __le##bits leres; \ - read_eb_member(eb, s, type, member, &leres); \ - return le##bits##_to_cpu(leres); \ - } \ - p = (type *)(kaddr + part_offset - map_start); \ - res = le##bits##_to_cpu(p->member); \ - if (unmap_on_exit) \ - unmap_extent_buffer(eb, map_token, KM_USER1); \ - return res; \ - } \ + int err; \ + char *kaddr; \ + unsigned long map_start; \ + unsigned long map_len; \ + u##bits res; \ + err = map_private_extent_buffer(eb, offset, \ + sizeof(((type *)0)->member), \ + &kaddr, &map_start, &map_len); \ + if (err) { \ + __le##bits leres; \ + read_eb_member(eb, s, type, member, &leres); \ + return le##bits##_to_cpu(leres); \ + } \ + p = (type *)(kaddr + part_offset - map_start); \ + res = le##bits##_to_cpu(p->member); \ + return res; \ } \ void btrfs_set_##name(struct extent_buffer *eb, \ type *s, u##bits val) \ @@ -87,36 +73,21 @@ void btrfs_set_##name(struct extent_buffer *eb, \ unsigned long part_offset = (unsigned long)s; \ unsigned long offset = part_offset + offsetof(type, member); \ type *p; \ - /* ugly, but we want the fast path here */ \ - if (eb->map_token && offset >= eb->map_start && \ - offset + sizeof(((type *)0)->member) <= eb->map_start + \ - eb->map_len) { \ - p = (type *)(eb->kaddr + part_offset - eb->map_start); \ - p->member = cpu_to_le##bits(val); \ - return; \ - } \ - { \ - int err; \ - char *map_token; \ - char *kaddr; \ - int unmap_on_exit = (eb->map_token == NULL); \ - unsigned long map_start; \ - unsigned long map_len; \ - err = map_extent_buffer(eb, offset, \ - sizeof(((type *)0)->member), \ - &map_token, &kaddr, \ - &map_start, &map_len, KM_USER1); \ - if (err) { \ - __le##bits val2; \ - val2 = cpu_to_le##bits(val); \ - write_eb_member(eb, s, type, member, &val2); \ - return; \ - } \ - p = (type *)(kaddr + part_offset - map_start); \ - p->member = cpu_to_le##bits(val); \ - if (unmap_on_exit) \ - unmap_extent_buffer(eb, map_token, KM_USER1); \ - } \ + int err; \ + char *kaddr; \ + unsigned long map_start; \ + unsigned long map_len; \ + err = map_private_extent_buffer(eb, offset, \ + sizeof(((type *)0)->member), \ + &kaddr, &map_start, &map_len); \ + if (err) { \ + __le##bits val2; \ + val2 = cpu_to_le##bits(val); \ + write_eb_member(eb, s, type, member, &val2); \ + return; \ + } \ + p = (type *)(kaddr + part_offset - map_start); \ + p->member = cpu_to_le##bits(val); \ } #include "ctree.h" @@ -125,15 +96,6 @@ void btrfs_node_key(struct extent_buffer *eb, struct btrfs_disk_key *disk_key, int nr) { unsigned long ptr = btrfs_node_key_ptr_offset(nr); - if (eb->map_token && ptr >= eb->map_start && - ptr + sizeof(*disk_key) <= eb->map_start + eb->map_len) { - memcpy(disk_key, eb->kaddr + ptr - eb->map_start, - sizeof(*disk_key)); - return; - } else if (eb->map_token) { - unmap_extent_buffer(eb, eb->map_token, KM_USER1); - eb->map_token = NULL; - } read_eb_member(eb, (struct btrfs_key_ptr *)ptr, struct btrfs_key_ptr, key, disk_key); } diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 51dcec86757f..eb55863bb4ae 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -260,7 +260,7 @@ static struct btrfs_trans_handle *start_transaction(struct btrfs_root *root, { struct btrfs_trans_handle *h; struct btrfs_transaction *cur_trans; - int retries = 0; + u64 num_bytes = 0; int ret; if (root->fs_info->fs_state & BTRFS_SUPER_FLAG_ERROR) @@ -274,6 +274,19 @@ static struct btrfs_trans_handle *start_transaction(struct btrfs_root *root, h->block_rsv = NULL; goto got_it; } + + /* + * Do the reservation before we join the transaction so we can do all + * the appropriate flushing if need be. + */ + if (num_items > 0 && root != root->fs_info->chunk_root) { + num_bytes = btrfs_calc_trans_metadata_size(root, num_items); + ret = btrfs_block_rsv_add(NULL, root, + &root->fs_info->trans_block_rsv, + num_bytes); + if (ret) + return ERR_PTR(ret); + } again: h = kmem_cache_alloc(btrfs_trans_handle_cachep, GFP_NOFS); if (!h) @@ -310,24 +323,9 @@ again: goto again; } - if (num_items > 0) { - ret = btrfs_trans_reserve_metadata(h, root, num_items); - if (ret == -EAGAIN && !retries) { - retries++; - btrfs_commit_transaction(h, root); - goto again; - } else if (ret == -EAGAIN) { - /* - * We have already retried and got EAGAIN, so really we - * don't have space, so set ret to -ENOSPC. - */ - ret = -ENOSPC; - } - - if (ret < 0) { - btrfs_end_transaction(h, root); - return ERR_PTR(ret); - } + if (num_bytes) { + h->block_rsv = &root->fs_info->trans_block_rsv; + h->bytes_reserved = num_bytes; } got_it: @@ -499,10 +497,17 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans, } if (lock && cur_trans->blocked && !cur_trans->in_commit) { - if (throttle) + if (throttle) { + /* + * We may race with somebody else here so end up having + * to call end_transaction on ourselves again, so inc + * our use_count. + */ + trans->use_count++; return btrfs_commit_transaction(trans, root); - else + } else { wake_up_process(info->transaction_kthread); + } } WARN_ON(cur_trans != info->running_transaction); diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index f3cacc079102..babee65f8eda 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -1733,8 +1733,8 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans, btrfs_read_buffer(next, ptr_gen); btrfs_tree_lock(next); - clean_tree_block(trans, root, next); btrfs_set_lock_blocking(next); + clean_tree_block(trans, root, next); btrfs_wait_tree_block_writeback(next); btrfs_tree_unlock(next); @@ -1802,8 +1802,8 @@ static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans, next = path->nodes[*level]; btrfs_tree_lock(next); - clean_tree_block(trans, root, next); btrfs_set_lock_blocking(next); + clean_tree_block(trans, root, next); btrfs_wait_tree_block_writeback(next); btrfs_tree_unlock(next); @@ -1870,8 +1870,8 @@ static int walk_log_tree(struct btrfs_trans_handle *trans, next = path->nodes[orig_level]; btrfs_tree_lock(next); - clean_tree_block(trans, log, next); btrfs_set_lock_blocking(next); + clean_tree_block(trans, log, next); btrfs_wait_tree_block_writeback(next); btrfs_tree_unlock(next); diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 90d956c17d92..53875ae73ad4 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -3599,7 +3599,7 @@ int btrfs_read_sys_array(struct btrfs_root *root) if (!sb) return -ENOMEM; btrfs_set_buffer_uptodate(sb); - btrfs_set_buffer_lockdep_class(sb, 0); + btrfs_set_buffer_lockdep_class(root->root_key.objectid, sb, 0); write_extent_buffer(sb, super_copy, 0, BTRFS_SUPER_INFO_SIZE); array_size = btrfs_super_sys_array_size(super_copy); diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c index 5366fe452ab0..d733b9cfea34 100644 --- a/fs/btrfs/xattr.c +++ b/fs/btrfs/xattr.c @@ -102,43 +102,57 @@ static int do_setxattr(struct btrfs_trans_handle *trans, if (!path) return -ENOMEM; - /* first lets see if we already have this xattr */ - di = btrfs_lookup_xattr(trans, root, path, btrfs_ino(inode), name, - strlen(name), -1); - if (IS_ERR(di)) { - ret = PTR_ERR(di); - goto out; - } - - /* ok we already have this xattr, lets remove it */ - if (di) { - /* if we want create only exit */ - if (flags & XATTR_CREATE) { - ret = -EEXIST; + if (flags & XATTR_REPLACE) { + di = btrfs_lookup_xattr(trans, root, path, btrfs_ino(inode), name, + name_len, -1); + if (IS_ERR(di)) { + ret = PTR_ERR(di); + goto out; + } else if (!di) { + ret = -ENODATA; goto out; } - ret = btrfs_delete_one_dir_name(trans, root, path, di); - BUG_ON(ret); + if (ret) + goto out; btrfs_release_path(path); + } - /* if we don't have a value then we are removing the xattr */ - if (!value) +again: + ret = btrfs_insert_xattr_item(trans, root, path, btrfs_ino(inode), + name, name_len, value, size); + if (ret == -EEXIST) { + if (flags & XATTR_CREATE) goto out; - } else { + /* + * We can't use the path we already have since we won't have the + * proper locking for a delete, so release the path and + * re-lookup to delete the thing. + */ btrfs_release_path(path); + di = btrfs_lookup_xattr(trans, root, path, btrfs_ino(inode), + name, name_len, -1); + if (IS_ERR(di)) { + ret = PTR_ERR(di); + goto out; + } else if (!di) { + /* Shouldn't happen but just in case... */ + btrfs_release_path(path); + goto again; + } - if (flags & XATTR_REPLACE) { - /* we couldn't find the attr to replace */ - ret = -ENODATA; + ret = btrfs_delete_one_dir_name(trans, root, path, di); + if (ret) goto out; + + /* + * We have a value to set, so go back and try to insert it now. + */ + if (value) { + btrfs_release_path(path); + goto again; } } - - /* ok we have to create a completely new xattr */ - ret = btrfs_insert_xattr_item(trans, root, path, btrfs_ino(inode), - name, name_len, value, size); - BUG_ON(ret); out: btrfs_free_path(path); return ret; diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index 79743d146be6..0c1d91756528 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -1438,12 +1438,15 @@ char *ceph_mdsc_build_path(struct dentry *dentry, int *plen, u64 *base, struct dentry *temp; char *path; int len, pos; + unsigned seq; if (dentry == NULL) return ERR_PTR(-EINVAL); retry: len = 0; + seq = read_seqbegin(&rename_lock); + rcu_read_lock(); for (temp = dentry; !IS_ROOT(temp);) { struct inode *inode = temp->d_inode; if (inode && ceph_snap(inode) == CEPH_SNAPDIR) @@ -1455,10 +1458,12 @@ retry: len += 1 + temp->d_name.len; temp = temp->d_parent; if (temp == NULL) { + rcu_read_unlock(); pr_err("build_path corrupt dentry %p\n", dentry); return ERR_PTR(-EINVAL); } } + rcu_read_unlock(); if (len) len--; /* no leading '/' */ @@ -1467,9 +1472,12 @@ retry: return ERR_PTR(-ENOMEM); pos = len; path[pos] = 0; /* trailing null */ + rcu_read_lock(); for (temp = dentry; !IS_ROOT(temp) && pos != 0; ) { - struct inode *inode = temp->d_inode; + struct inode *inode; + spin_lock(&temp->d_lock); + inode = temp->d_inode; if (inode && ceph_snap(inode) == CEPH_SNAPDIR) { dout("build_path path+%d: %p SNAPDIR\n", pos, temp); @@ -1478,21 +1486,26 @@ retry: break; } else { pos -= temp->d_name.len; - if (pos < 0) + if (pos < 0) { + spin_unlock(&temp->d_lock); break; + } strncpy(path + pos, temp->d_name.name, temp->d_name.len); } + spin_unlock(&temp->d_lock); if (pos) path[--pos] = '/'; temp = temp->d_parent; if (temp == NULL) { + rcu_read_unlock(); pr_err("build_path corrupt dentry\n"); kfree(path); return ERR_PTR(-EINVAL); } } - if (pos != 0) { + rcu_read_unlock(); + if (pos != 0 || read_seqretry(&rename_lock, seq)) { pr_err("build_path did not end path lookup where " "expected, namelen is %d, pos is %d\n", len, pos); /* presumably this is only possible if racing with a diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 3e2989976297..bc4b12ca537b 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -35,6 +35,7 @@ #include <linux/delay.h> #include <linux/kthread.h> #include <linux/freezer.h> +#include <linux/namei.h> #include <net/ipv6.h> #include "cifsfs.h" #include "cifspdu.h" @@ -542,14 +543,12 @@ static const struct super_operations cifs_super_ops = { static struct dentry * cifs_get_root(struct smb_vol *vol, struct super_block *sb) { - int xid, rc; - struct inode *inode; - struct qstr name; - struct dentry *dparent = NULL, *dchild = NULL, *alias; + struct dentry *dentry; struct cifs_sb_info *cifs_sb = CIFS_SB(sb); - unsigned int i, full_len, len; - char *full_path = NULL, *pstart; + char *full_path = NULL; + char *s, *p; char sep; + int xid; full_path = cifs_build_path_to_root(vol, cifs_sb, cifs_sb_master_tcon(cifs_sb)); @@ -560,73 +559,32 @@ cifs_get_root(struct smb_vol *vol, struct super_block *sb) xid = GetXid(); sep = CIFS_DIR_SEP(cifs_sb); - dparent = dget(sb->s_root); - full_len = strlen(full_path); - full_path[full_len] = sep; - pstart = full_path + 1; - - for (i = 1, len = 0; i <= full_len; i++) { - if (full_path[i] != sep || !len) { - len++; - continue; - } - - full_path[i] = 0; - cFYI(1, "get dentry for %s", pstart); - - name.name = pstart; - name.len = len; - name.hash = full_name_hash(pstart, len); - dchild = d_lookup(dparent, &name); - if (dchild == NULL) { - cFYI(1, "not exists"); - dchild = d_alloc(dparent, &name); - if (dchild == NULL) { - dput(dparent); - dparent = ERR_PTR(-ENOMEM); - goto out; - } - } - - cFYI(1, "get inode"); - if (dchild->d_inode == NULL) { - cFYI(1, "not exists"); - inode = NULL; - if (cifs_sb_master_tcon(CIFS_SB(sb))->unix_ext) - rc = cifs_get_inode_info_unix(&inode, full_path, - sb, xid); - else - rc = cifs_get_inode_info(&inode, full_path, - NULL, sb, xid, NULL); - if (rc) { - dput(dchild); - dput(dparent); - dparent = ERR_PTR(rc); - goto out; - } - alias = d_materialise_unique(dchild, inode); - if (alias != NULL) { - dput(dchild); - if (IS_ERR(alias)) { - dput(dparent); - dparent = ERR_PTR(-EINVAL); /* XXX */ - goto out; - } - dchild = alias; - } - } - cFYI(1, "parent %p, child %p", dparent, dchild); - - dput(dparent); - dparent = dchild; - len = 0; - pstart = full_path + i + 1; - full_path[i] = sep; - } -out: + dentry = dget(sb->s_root); + p = s = full_path; + + do { + struct inode *dir = dentry->d_inode; + struct dentry *child; + + /* skip separators */ + while (*s == sep) + s++; + if (!*s) + break; + p = s++; + /* next separator */ + while (*s && *s != sep) + s++; + + mutex_lock(&dir->i_mutex); + child = lookup_one_len(p, dentry, s - p); + mutex_unlock(&dir->i_mutex); + dput(dentry); + dentry = child; + } while (!IS_ERR(dentry)); _FreeXid(xid); kfree(full_path); - return dparent; + return dentry; } static int cifs_set_super(struct super_block *sb, void *data) diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 0900e1658c96..036ca83e5f46 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -129,5 +129,5 @@ extern long cifs_ioctl(struct file *filep, unsigned int cmd, unsigned long arg); extern const struct export_operations cifs_export_ops; #endif /* CIFS_NFSD_EXPORT */ -#define CIFS_VERSION "1.73" +#define CIFS_VERSION "1.74" #endif /* _CIFSFS_H */ diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index dbd669cc5bc7..ccc1afa0bf3b 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -3485,7 +3485,7 @@ cifs_construct_tcon(struct cifs_sb_info *cifs_sb, uid_t fsuid) goto out; } - snprintf(username, MAX_USERNAME_SIZE, "krb50x%x", fsuid); + snprintf(username, sizeof(username), "krb50x%x", fsuid); vol_info->username = username; vol_info->local_nls = cifs_sb->local_nls; vol_info->linux_uid = fsuid; diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index 81914df47ef1..fa8c21d913bc 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -55,6 +55,7 @@ build_path_from_dentry(struct dentry *direntry) char dirsep; struct cifs_sb_info *cifs_sb = CIFS_SB(direntry->d_sb); struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); + unsigned seq; if (direntry == NULL) return NULL; /* not much we can do if dentry is freed and @@ -68,22 +69,29 @@ build_path_from_dentry(struct dentry *direntry) dfsplen = 0; cifs_bp_rename_retry: namelen = dfsplen; + seq = read_seqbegin(&rename_lock); + rcu_read_lock(); for (temp = direntry; !IS_ROOT(temp);) { namelen += (1 + temp->d_name.len); temp = temp->d_parent; if (temp == NULL) { cERROR(1, "corrupt dentry"); + rcu_read_unlock(); return NULL; } } + rcu_read_unlock(); full_path = kmalloc(namelen+1, GFP_KERNEL); if (full_path == NULL) return full_path; full_path[namelen] = 0; /* trailing null */ + rcu_read_lock(); for (temp = direntry; !IS_ROOT(temp);) { + spin_lock(&temp->d_lock); namelen -= 1 + temp->d_name.len; if (namelen < 0) { + spin_unlock(&temp->d_lock); break; } else { full_path[namelen] = dirsep; @@ -91,14 +99,17 @@ cifs_bp_rename_retry: temp->d_name.len); cFYI(0, "name: %s", full_path + namelen); } + spin_unlock(&temp->d_lock); temp = temp->d_parent; if (temp == NULL) { cERROR(1, "corrupt dentry"); + rcu_read_unlock(); kfree(full_path); return NULL; } } - if (namelen != dfsplen) { + rcu_read_unlock(); + if (namelen != dfsplen || read_seqretry(&rename_lock, seq)) { cERROR(1, "did not end path lookup where expected namelen is %d", namelen); /* presumably this is only possible if racing with a rename diff --git a/fs/cifs/file.c b/fs/cifs/file.c index bb71471a4d9d..a9b4a24f2a16 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -1737,7 +1737,7 @@ cifs_iovec_read(struct file *file, const struct iovec *iov, io_parms.pid = pid; io_parms.tcon = pTcon; io_parms.offset = *poffset; - io_parms.length = len; + io_parms.length = cur_len; rc = CIFSSMBRead(xid, &io_parms, &bytes_read, &read_data, &buf_type); pSMBr = (struct smb_com_read_rsp *)read_data; diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index 3892ab817a36..d3e619692ee0 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -428,8 +428,7 @@ static void build_ntlmssp_negotiate_blob(unsigned char *pbuffer, (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) { flags |= NTLMSSP_NEGOTIATE_SIGN; if (!ses->server->session_estab) - flags |= NTLMSSP_NEGOTIATE_KEY_XCH | - NTLMSSP_NEGOTIATE_EXTENDED_SEC; + flags |= NTLMSSP_NEGOTIATE_KEY_XCH; } sec_blob->NegotiateFlags = cpu_to_le32(flags); @@ -465,10 +464,11 @@ static int build_ntlmssp_auth_blob(unsigned char *pbuffer, NTLMSSP_NEGOTIATE_128 | NTLMSSP_NEGOTIATE_UNICODE | NTLMSSP_NEGOTIATE_NTLM | NTLMSSP_NEGOTIATE_EXTENDED_SEC; if (ses->server->sec_mode & - (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) + (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) { flags |= NTLMSSP_NEGOTIATE_SIGN; - if (ses->server->sec_mode & SECMODE_SIGN_REQUIRED) - flags |= NTLMSSP_NEGOTIATE_ALWAYS_SIGN; + if (!ses->server->session_estab) + flags |= NTLMSSP_NEGOTIATE_KEY_XCH; + } tmp = pbuffer + sizeof(AUTHENTICATE_MESSAGE); sec_blob->NegotiateFlags = cpu_to_le32(flags); diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index e141939080f0..739fb59bcdc2 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -37,7 +37,7 @@ static DEFINE_MUTEX(read_mutex); /* These macros may change in future, to provide better st_ino semantics. */ #define OFFSET(x) ((x)->i_ino) -static unsigned long cramino(struct cramfs_inode *cino, unsigned int offset) +static unsigned long cramino(const struct cramfs_inode *cino, unsigned int offset) { if (!cino->offset) return offset + 1; @@ -61,7 +61,7 @@ static unsigned long cramino(struct cramfs_inode *cino, unsigned int offset) } static struct inode *get_cramfs_inode(struct super_block *sb, - struct cramfs_inode *cramfs_inode, unsigned int offset) + const struct cramfs_inode *cramfs_inode, unsigned int offset) { struct inode *inode; static struct timespec zerotime; @@ -317,7 +317,7 @@ static int cramfs_fill_super(struct super_block *sb, void *data, int silent) /* Set it all up.. */ sb->s_op = &cramfs_ops; root = get_cramfs_inode(sb, &super.root, 0); - if (!root) + if (IS_ERR(root)) goto out; sb->s_root = d_alloc_root(root); if (!sb->s_root) { @@ -423,6 +423,7 @@ static int cramfs_readdir(struct file *filp, void *dirent, filldir_t filldir) static struct dentry * cramfs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) { unsigned int offset = 0; + struct inode *inode = NULL; int sorted; mutex_lock(&read_mutex); @@ -449,8 +450,8 @@ static struct dentry * cramfs_lookup(struct inode *dir, struct dentry *dentry, s for (;;) { if (!namelen) { - mutex_unlock(&read_mutex); - return ERR_PTR(-EIO); + inode = ERR_PTR(-EIO); + goto out; } if (name[namelen-1]) break; @@ -462,17 +463,18 @@ static struct dentry * cramfs_lookup(struct inode *dir, struct dentry *dentry, s if (retval > 0) continue; if (!retval) { - struct cramfs_inode entry = *de; - mutex_unlock(&read_mutex); - d_add(dentry, get_cramfs_inode(dir->i_sb, &entry, dir_off)); - return NULL; + inode = get_cramfs_inode(dir->i_sb, de, dir_off); + break; } /* else (retval < 0) */ if (sorted) break; } +out: mutex_unlock(&read_mutex); - d_add(dentry, NULL); + if (IS_ERR(inode)) + return ERR_CAST(inode); + d_add(dentry, inode); return NULL; } diff --git a/fs/dcache.c b/fs/dcache.c index 37f72ee5bf7c..fbdcbca40725 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -1813,8 +1813,6 @@ seqretry: tname = dentry->d_name.name; i = dentry->d_inode; prefetch(tname); - if (i) - prefetch(i); /* * This seqcount check is required to ensure name and * len are loaded atomically, so as not to walk off the @@ -2213,14 +2211,15 @@ static void dentry_unlock_parents_for_move(struct dentry *dentry, * The hash value has to match the hash queue that the dentry is on.. */ /* - * d_move - move a dentry + * __d_move - move a dentry * @dentry: entry to move * @target: new dentry * * Update the dcache to reflect the move of a file name. Negative - * dcache entries should not be moved in this way. + * dcache entries should not be moved in this way. Caller hold + * rename_lock. */ -void d_move(struct dentry * dentry, struct dentry * target) +static void __d_move(struct dentry * dentry, struct dentry * target) { if (!dentry->d_inode) printk(KERN_WARNING "VFS: moving negative dcache entry\n"); @@ -2228,8 +2227,6 @@ void d_move(struct dentry * dentry, struct dentry * target) BUG_ON(d_ancestor(dentry, target)); BUG_ON(d_ancestor(target, dentry)); - write_seqlock(&rename_lock); - dentry_lock_for_move(dentry, target); write_seqcount_begin(&dentry->d_seq); @@ -2275,6 +2272,20 @@ void d_move(struct dentry * dentry, struct dentry * target) spin_unlock(&target->d_lock); fsnotify_d_move(dentry); spin_unlock(&dentry->d_lock); +} + +/* + * d_move - move a dentry + * @dentry: entry to move + * @target: new dentry + * + * Update the dcache to reflect the move of a file name. Negative + * dcache entries should not be moved in this way. + */ +void d_move(struct dentry *dentry, struct dentry *target) +{ + write_seqlock(&rename_lock); + __d_move(dentry, target); write_sequnlock(&rename_lock); } EXPORT_SYMBOL(d_move); @@ -2302,7 +2313,7 @@ struct dentry *d_ancestor(struct dentry *p1, struct dentry *p2) * This helper attempts to cope with remotely renamed directories * * It assumes that the caller is already holding - * dentry->d_parent->d_inode->i_mutex and the inode->i_lock + * dentry->d_parent->d_inode->i_mutex, inode->i_lock and rename_lock * * Note: If ever the locking in lock_rename() changes, then please * remember to update this too... @@ -2317,11 +2328,6 @@ static struct dentry *__d_unalias(struct inode *inode, if (alias->d_parent == dentry->d_parent) goto out_unalias; - /* Check for loops */ - ret = ERR_PTR(-ELOOP); - if (d_ancestor(alias, dentry)) - goto out_err; - /* See lock_rename() */ ret = ERR_PTR(-EBUSY); if (!mutex_trylock(&dentry->d_sb->s_vfs_rename_mutex)) @@ -2331,7 +2337,7 @@ static struct dentry *__d_unalias(struct inode *inode, goto out_err; m2 = &alias->d_parent->d_inode->i_mutex; out_unalias: - d_move(alias, dentry); + __d_move(alias, dentry); ret = alias; out_err: spin_unlock(&inode->i_lock); @@ -2416,15 +2422,24 @@ struct dentry *d_materialise_unique(struct dentry *dentry, struct inode *inode) alias = __d_find_alias(inode, 0); if (alias) { actual = alias; - /* Is this an anonymous mountpoint that we could splice - * into our tree? */ - if (IS_ROOT(alias)) { + write_seqlock(&rename_lock); + + if (d_ancestor(alias, dentry)) { + /* Check for loops */ + actual = ERR_PTR(-ELOOP); + } else if (IS_ROOT(alias)) { + /* Is this an anonymous mountpoint that we + * could splice into our tree? */ __d_materialise_dentry(dentry, alias); + write_sequnlock(&rename_lock); __d_drop(alias); goto found; + } else { + /* Nope, but we must(!) avoid directory + * aliasing */ + actual = __d_unalias(inode, dentry, alias); } - /* Nope, but we must(!) avoid directory aliasing */ - actual = __d_unalias(inode, dentry, alias); + write_sequnlock(&rename_lock); if (IS_ERR(actual)) dput(alias); goto out_nolock; diff --git a/fs/exofs/super.c b/fs/exofs/super.c index 06065bd37fc3..c57beddcc217 100644 --- a/fs/exofs/super.c +++ b/fs/exofs/super.c @@ -913,7 +913,7 @@ struct dentry *exofs_get_parent(struct dentry *child) unsigned long ino = exofs_parent_ino(child); if (!ino) - return NULL; + return ERR_PTR(-ESTALE); return d_obtain_alias(exofs_iget(child->d_inode->i_sb, ino)); } diff --git a/fs/fscache/page.c b/fs/fscache/page.c index 2f343b4d7a7d..3f7a59bfa7ad 100644 --- a/fs/fscache/page.c +++ b/fs/fscache/page.c @@ -976,16 +976,12 @@ void __fscache_uncache_all_inode_pages(struct fscache_cookie *cookie, pagevec_init(&pvec, 0); next = 0; - while (next <= (loff_t)-1 && - pagevec_lookup(&pvec, mapping, next, PAGEVEC_SIZE) - ) { + do { + if (!pagevec_lookup(&pvec, mapping, next, PAGEVEC_SIZE)) + break; for (i = 0; i < pagevec_count(&pvec); i++) { struct page *page = pvec.pages[i]; - pgoff_t page_index = page->index; - - ASSERTCMP(page_index, >=, next); - next = page_index + 1; - + next = page->index; if (PageFsCache(page)) { __fscache_wait_on_page_write(cookie, page); __fscache_uncache_page(cookie, page); @@ -993,7 +989,7 @@ void __fscache_uncache_all_inode_pages(struct fscache_cookie *cookie, } pagevec_release(&pvec); cond_resched(); - } + } while (++next); _leave(""); } diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c index 802ac5eeba28..f9fbbe96c222 100644 --- a/fs/gfs2/aops.c +++ b/fs/gfs2/aops.c @@ -1069,6 +1069,7 @@ int gfs2_releasepage(struct page *page, gfp_t gfp_mask) return 0; gfs2_log_lock(sdp); + spin_lock(&sdp->sd_ail_lock); head = bh = page_buffers(page); do { if (atomic_read(&bh->b_count)) @@ -1080,6 +1081,7 @@ int gfs2_releasepage(struct page *page, gfp_t gfp_mask) goto not_possible; bh = bh->b_this_page; } while(bh != head); + spin_unlock(&sdp->sd_ail_lock); gfs2_log_unlock(sdp); head = bh = page_buffers(page); @@ -1112,6 +1114,7 @@ not_possible: /* Should never happen */ WARN_ON(buffer_dirty(bh)); WARN_ON(buffer_pinned(bh)); cannot_release: + spin_unlock(&sdp->sd_ail_lock); gfs2_log_unlock(sdp); return 0; } diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c index 8ef70f464731..2cca29316bd6 100644 --- a/fs/gfs2/glops.c +++ b/fs/gfs2/glops.c @@ -47,10 +47,10 @@ static void __gfs2_ail_flush(struct gfs2_glock *gl) bd_ail_gl_list); bh = bd->bd_bh; gfs2_remove_from_ail(bd); - spin_unlock(&sdp->sd_ail_lock); - bd->bd_bh = NULL; bh->b_private = NULL; + spin_unlock(&sdp->sd_ail_lock); + bd->bd_blkno = bh->b_blocknr; gfs2_log_lock(sdp); gfs2_assert_withdraw(sdp, !buffer_busy(bh)); @@ -221,8 +221,10 @@ static void inode_go_inval(struct gfs2_glock *gl, int flags) } } - if (ip == GFS2_I(gl->gl_sbd->sd_rindex)) + if (ip == GFS2_I(gl->gl_sbd->sd_rindex)) { + gfs2_log_flush(gl->gl_sbd, NULL); gl->gl_sbd->sd_rindex_uptodate = 0; + } if (ip && S_ISREG(ip->i_inode.i_mode)) truncate_inode_pages(ip->i_inode.i_mapping, 0); } diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index 0a064e91ac70..81206e70cbf6 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -17,6 +17,7 @@ #include <linux/buffer_head.h> #include <linux/rcupdate.h> #include <linux/rculist_bl.h> +#include <linux/completion.h> #define DIO_WAIT 0x00000010 #define DIO_METADATA 0x00000020 @@ -546,6 +547,7 @@ struct gfs2_sbd { struct gfs2_glock *sd_trans_gl; wait_queue_head_t sd_glock_wait; atomic_t sd_glock_disposal; + struct completion sd_locking_init; /* Inode Stuff */ diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index 903115f2bb34..85c62923ee29 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -903,6 +903,7 @@ void gfs2_meta_syncfs(struct gfs2_sbd *sdp) if (gfs2_ail1_empty(sdp)) break; } + gfs2_log_flush(sdp, NULL); } static inline int gfs2_jrnl_flush_reqd(struct gfs2_sbd *sdp) diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 8ac9ae189b53..2a77071fb7b6 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -72,6 +72,7 @@ static struct gfs2_sbd *init_sbd(struct super_block *sb) init_waitqueue_head(&sdp->sd_glock_wait); atomic_set(&sdp->sd_glock_disposal, 0); + init_completion(&sdp->sd_locking_init); spin_lock_init(&sdp->sd_statfs_spin); spin_lock_init(&sdp->sd_rindex_spin); @@ -1017,11 +1018,13 @@ hostdata_error: fsname++; if (lm->lm_mount == NULL) { fs_info(sdp, "Now mounting FS...\n"); + complete(&sdp->sd_locking_init); return 0; } ret = lm->lm_mount(sdp, fsname); if (ret == 0) fs_info(sdp, "Joined cluster. Now mounting FS...\n"); + complete(&sdp->sd_locking_init); return ret; } diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index ed540e7018be..fb0edf735483 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -757,13 +757,17 @@ static int gfs2_write_inode(struct inode *inode, struct writeback_control *wbc) struct timespec atime; struct gfs2_dinode *di; int ret = -EAGAIN; + int unlock_required = 0; /* Skip timestamp update, if this is from a memalloc */ if (current->flags & PF_MEMALLOC) goto do_flush; - ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); - if (ret) - goto do_flush; + if (!gfs2_glock_is_locked_by_me(ip->i_gl)) { + ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); + if (ret) + goto do_flush; + unlock_required = 1; + } ret = gfs2_trans_begin(sdp, RES_DINODE, 0); if (ret) goto do_unlock; @@ -780,7 +784,8 @@ static int gfs2_write_inode(struct inode *inode, struct writeback_control *wbc) } gfs2_trans_end(sdp); do_unlock: - gfs2_glock_dq_uninit(&gh); + if (unlock_required) + gfs2_glock_dq_uninit(&gh); do_flush: if (wbc->sync_mode == WB_SYNC_ALL) gfs2_log_flush(GFS2_SB(inode), ip->i_gl); @@ -1427,7 +1432,20 @@ out: return error; } -/* +/** + * gfs2_evict_inode - Remove an inode from cache + * @inode: The inode to evict + * + * There are three cases to consider: + * 1. i_nlink == 0, we are final opener (and must deallocate) + * 2. i_nlink == 0, we are not the final opener (and cannot deallocate) + * 3. i_nlink > 0 + * + * If the fs is read only, then we have to treat all cases as per #3 + * since we are unable to do any deallocation. The inode will be + * deallocated by the next read/write node to attempt an allocation + * in the same resource group + * * We have to (at the moment) hold the inodes main lock to cover * the gap between unlocking the shared lock on the iopen lock and * taking the exclusive lock. I'd rather do a shared -> exclusive @@ -1470,6 +1488,8 @@ static void gfs2_evict_inode(struct inode *inode) if (error) goto out_truncate; + /* Case 1 starts here */ + if (S_ISDIR(inode->i_mode) && (ip->i_diskflags & GFS2_DIF_EXHASH)) { error = gfs2_dir_exhash_dealloc(ip); @@ -1493,13 +1513,16 @@ static void gfs2_evict_inode(struct inode *inode) goto out_unlock; out_truncate: + /* Case 2 starts here */ error = gfs2_trans_begin(sdp, 0, sdp->sd_jdesc->jd_blocks); if (error) goto out_unlock; - gfs2_final_release_pages(ip); + /* Needs to be done before glock release & also in a transaction */ + truncate_inode_pages(&inode->i_data, 0); gfs2_trans_end(sdp); out_unlock: + /* Error path for case 1 */ if (test_bit(HIF_HOLDER, &ip->i_iopen_gh.gh_iflags)) gfs2_glock_dq(&ip->i_iopen_gh); gfs2_holder_uninit(&ip->i_iopen_gh); @@ -1507,6 +1530,7 @@ out_unlock: if (error && error != GLR_TRYFAILED && error != -EROFS) fs_warn(sdp, "gfs2_evict_inode: %d\n", error); out: + /* Case 3 starts here */ truncate_inode_pages(&inode->i_data, 0); end_writeback(inode); diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c index e20eab37bc80..443cabcfcd23 100644 --- a/fs/gfs2/sys.c +++ b/fs/gfs2/sys.c @@ -338,6 +338,9 @@ static ssize_t lkfirst_store(struct gfs2_sbd *sdp, const char *buf, size_t len) rv = sscanf(buf, "%u", &first); if (rv != 1 || first > 1) return -EINVAL; + rv = wait_for_completion_killable(&sdp->sd_locking_init); + if (rv) + return rv; spin_lock(&sdp->sd_jindex_spin); rv = -EBUSY; if (test_bit(SDF_NOJOURNALID, &sdp->sd_flags) == 0) @@ -414,7 +417,9 @@ static ssize_t jid_store(struct gfs2_sbd *sdp, const char *buf, size_t len) rv = sscanf(buf, "%d", &jid); if (rv != 1) return -EINVAL; - + rv = wait_for_completion_killable(&sdp->sd_locking_init); + if (rv) + return rv; spin_lock(&sdp->sd_jindex_spin); rv = -EINVAL; if (sdp->sd_lockstruct.ls_ops->lm_mount == NULL) diff --git a/fs/hppfs/hppfs.c b/fs/hppfs/hppfs.c index 87ed48e0343d..85c098a499f3 100644 --- a/fs/hppfs/hppfs.c +++ b/fs/hppfs/hppfs.c @@ -139,7 +139,8 @@ static int file_removed(struct dentry *dentry, const char *file) static struct dentry *hppfs_lookup(struct inode *ino, struct dentry *dentry, struct nameidata *nd) { - struct dentry *proc_dentry, *new, *parent; + struct dentry *proc_dentry, *parent; + struct qstr *name = &dentry->d_name; struct inode *inode; int err, deleted; @@ -149,23 +150,9 @@ static struct dentry *hppfs_lookup(struct inode *ino, struct dentry *dentry, else if (deleted) return ERR_PTR(-ENOENT); - err = -ENOMEM; parent = HPPFS_I(ino)->proc_dentry; mutex_lock(&parent->d_inode->i_mutex); - proc_dentry = d_lookup(parent, &dentry->d_name); - if (proc_dentry == NULL) { - proc_dentry = d_alloc(parent, &dentry->d_name); - if (proc_dentry == NULL) { - mutex_unlock(&parent->d_inode->i_mutex); - goto out; - } - new = (*parent->d_inode->i_op->lookup)(parent->d_inode, - proc_dentry, NULL); - if (new) { - dput(proc_dentry); - proc_dentry = new; - } - } + proc_dentry = lookup_one_len(name->name, parent, name->len); mutex_unlock(&parent->d_inode->i_mutex); if (IS_ERR(proc_dentry)) @@ -174,13 +161,11 @@ static struct dentry *hppfs_lookup(struct inode *ino, struct dentry *dentry, err = -ENOMEM; inode = get_inode(ino->i_sb, proc_dentry); if (!inode) - goto out_dput; + goto out; d_add(dentry, inode); return NULL; - out_dput: - dput(proc_dentry); out: return ERR_PTR(err); } @@ -690,8 +675,10 @@ static struct inode *get_inode(struct super_block *sb, struct dentry *dentry) struct inode *proc_ino = dentry->d_inode; struct inode *inode = new_inode(sb); - if (!inode) + if (!inode) { + dput(dentry); return ERR_PTR(-ENOMEM); + } if (S_ISDIR(dentry->d_inode->i_mode)) { inode->i_op = &hppfs_dir_iops; @@ -704,7 +691,7 @@ static struct inode *get_inode(struct super_block *sb, struct dentry *dentry) inode->i_fop = &hppfs_file_fops; } - HPPFS_I(inode)->proc_dentry = dget(dentry); + HPPFS_I(inode)->proc_dentry = dentry; inode->i_uid = proc_ino->i_uid; inode->i_gid = proc_ino->i_gid; @@ -737,7 +724,7 @@ static int hppfs_fill_super(struct super_block *sb, void *d, int silent) sb->s_fs_info = proc_mnt; err = -ENOMEM; - root_inode = get_inode(sb, proc_mnt->mnt_sb->s_root); + root_inode = get_inode(sb, dget(proc_mnt->mnt_sb->s_root)); if (!root_inode) goto out_mntput; diff --git a/fs/libfs.c b/fs/libfs.c index c88eab55aec9..275ca4749a2e 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -822,7 +822,7 @@ ssize_t simple_attr_write(struct file *file, const char __user *buf, goto out; attr->set_buf[size] = '\0'; - val = simple_strtol(attr->set_buf, NULL, 0); + val = simple_strtoll(attr->set_buf, NULL, 0); ret = attr->set(attr->data, val); if (ret == 0) ret = len; /* on success, claim we got the whole input */ diff --git a/fs/namei.c b/fs/namei.c index 0223c41fb114..14ab8d3f2f0c 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -433,6 +433,8 @@ static int unlazy_walk(struct nameidata *nd, struct dentry *dentry) goto err_parent; BUG_ON(nd->inode != parent->d_inode); } else { + if (dentry->d_parent != parent) + goto err_parent; spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); if (!__d_rcu_to_refcount(dentry, nd->seq)) goto err_child; @@ -940,7 +942,6 @@ static bool __follow_mount_rcu(struct nameidata *nd, struct path *path, * Don't forget we might have a non-mountpoint managed dentry * that wants to block transit. */ - *inode = path->dentry->d_inode; if (unlikely(managed_dentry_might_block(path->dentry))) return false; @@ -953,6 +954,12 @@ static bool __follow_mount_rcu(struct nameidata *nd, struct path *path, path->mnt = mounted; path->dentry = mounted->mnt_root; nd->seq = read_seqcount_begin(&path->dentry->d_seq); + /* + * Update the inode too. We don't need to re-check the + * dentry sequence number here after this d_inode read, + * because a mount-point is always pinned. + */ + *inode = path->dentry->d_inode; } return true; } diff --git a/fs/nfs/nfs4filelayout.c b/fs/nfs/nfs4filelayout.c index 0bafcc91c27f..f9d03abcd04c 100644 --- a/fs/nfs/nfs4filelayout.c +++ b/fs/nfs/nfs4filelayout.c @@ -398,7 +398,6 @@ filelayout_write_pagelist(struct nfs_write_data *data, int sync) * this offset and save the original offset. */ data->args.offset = filelayout_get_dserver_offset(lseg, offset); - data->mds_offset = offset; /* Perform an asynchronous write */ status = nfs_initiate_write(data, ds->ds_clp->cl_rpcclient, diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 6870bc61ceec..e6e8f3b9a1de 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -91,7 +91,7 @@ static int nfs4_stat_to_errno(int); #define encode_getfh_maxsz (op_encode_hdr_maxsz) #define decode_getfh_maxsz (op_decode_hdr_maxsz + 1 + \ ((3+NFS4_FHSIZE) >> 2)) -#define nfs4_fattr_bitmap_maxsz 3 +#define nfs4_fattr_bitmap_maxsz 4 #define encode_getattr_maxsz (op_encode_hdr_maxsz + nfs4_fattr_bitmap_maxsz) #define nfs4_name_maxsz (1 + ((3 + NFS4_MAXNAMLEN) >> 2)) #define nfs4_path_maxsz (1 + ((3 + NFS4_MAXPATHLEN) >> 2)) diff --git a/fs/nfs/write.c b/fs/nfs/write.c index e268e3b23497..727168059684 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -864,6 +864,8 @@ static int nfs_write_rpcsetup(struct nfs_page *req, data->args.fh = NFS_FH(inode); data->args.offset = req_offset(req) + offset; + /* pnfs_set_layoutcommit needs this */ + data->mds_offset = data->args.offset; data->args.pgbase = req->wb_pgbase + offset; data->args.pages = data->pagevec; data->args.count = count; diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c index 29309e25417f..b57aab9a1184 100644 --- a/fs/ufs/namei.c +++ b/fs/ufs/namei.c @@ -56,16 +56,12 @@ static struct dentry *ufs_lookup(struct inode * dir, struct dentry *dentry, stru lock_ufs(dir->i_sb); ino = ufs_inode_by_name(dir, &dentry->d_name); - if (ino) { + if (ino) inode = ufs_iget(dir->i_sb, ino); - if (IS_ERR(inode)) { - unlock_ufs(dir->i_sb); - return ERR_CAST(inode); - } - } unlock_ufs(dir->i_sb); - d_add(dentry, inode); - return NULL; + if (IS_ERR(inode)) + return ERR_CAST(inode); + return d_splice_alias(inode, dentry); } /* diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h index 3a10ef5914eb..6cd5b6403a7b 100644 --- a/include/acpi/acpi_bus.h +++ b/include/acpi/acpi_bus.h @@ -210,7 +210,7 @@ struct acpi_device_power_state { struct acpi_device_power { int state; /* Current state */ struct acpi_device_power_flags flags; - struct acpi_device_power_state states[4]; /* Power states (D0-D3) */ + struct acpi_device_power_state states[ACPI_D_STATE_COUNT]; /* Power states (D0-D3Cold) */ }; /* Performance Management */ diff --git a/include/acpi/acpiosxf.h b/include/acpi/acpiosxf.h index a756bc8d866d..4543b6f75867 100644 --- a/include/acpi/acpiosxf.h +++ b/include/acpi/acpiosxf.h @@ -98,8 +98,11 @@ acpi_os_table_override(struct acpi_table_header *existing_table, /* * Spinlock primitives */ + +#ifndef acpi_os_create_lock acpi_status acpi_os_create_lock(acpi_spinlock *out_handle); +#endif void acpi_os_delete_lock(acpi_spinlock handle); diff --git a/include/acpi/platform/aclinux.h b/include/acpi/platform/aclinux.h index 5d2a5e9544d9..2ce1be9f6291 100644 --- a/include/acpi/platform/aclinux.h +++ b/include/acpi/platform/aclinux.h @@ -159,6 +159,24 @@ static inline void *acpi_os_acquire_object(acpi_cache_t * cache) } while (0) #endif +/* + * When lockdep is enabled, the spin_lock_init() macro stringifies it's + * argument and uses that as a name for the lock in debugging. + * By executing spin_lock_init() in a macro the key changes from "lock" for + * all locks to the name of the argument of acpi_os_create_lock(), which + * prevents lockdep from reporting false positives for ACPICA locks. + */ +#define acpi_os_create_lock(__handle) \ +({ \ + spinlock_t *lock = ACPI_ALLOCATE(sizeof(*lock)); \ + \ + if (lock) { \ + *(__handle) = lock; \ + spin_lock_init(*(__handle)); \ + } \ + lock ? AE_OK : AE_NO_MEMORY; \ +}) + #endif /* __KERNEL__ */ #endif /* __ACLINUX_H__ */ diff --git a/include/drm/drm_pciids.h b/include/drm/drm_pciids.h index e08f344c6cff..3d53efd25ab9 100644 --- a/include/drm/drm_pciids.h +++ b/include/drm/drm_pciids.h @@ -182,6 +182,7 @@ {0x1002, 0x6750, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_NEW_MEMMAP}, \ {0x1002, 0x6758, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_NEW_MEMMAP}, \ {0x1002, 0x6759, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_NEW_MEMMAP}, \ + {0x1002, 0x675F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TURKS|RADEON_NEW_MEMMAP}, \ {0x1002, 0x6760, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CAICOS|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \ {0x1002, 0x6761, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CAICOS|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \ {0x1002, 0x6762, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CAICOS|RADEON_NEW_MEMMAP}, \ @@ -192,6 +193,7 @@ {0x1002, 0x6767, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CAICOS|RADEON_NEW_MEMMAP}, \ {0x1002, 0x6768, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CAICOS|RADEON_NEW_MEMMAP}, \ {0x1002, 0x6770, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CAICOS|RADEON_NEW_MEMMAP}, \ + {0x1002, 0x6778, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CAICOS|RADEON_NEW_MEMMAP}, \ {0x1002, 0x6779, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CAICOS|RADEON_NEW_MEMMAP}, \ {0x1002, 0x6880, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CYPRESS|RADEON_IS_MOBILITY|RADEON_NEW_MEMMAP}, \ {0x1002, 0x6888, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_CYPRESS|RADEON_NEW_MEMMAP}, \ diff --git a/include/linux/mmc/card.h b/include/linux/mmc/card.h index c6927a4d157f..6ad43554ac05 100644 --- a/include/linux/mmc/card.h +++ b/include/linux/mmc/card.h @@ -64,6 +64,19 @@ struct mmc_ext_csd { unsigned long long enhanced_area_offset; /* Units: Byte */ unsigned int enhanced_area_size; /* Units: KB */ unsigned int boot_size; /* in bytes */ + u8 raw_partition_support; /* 160 */ + u8 raw_erased_mem_count; /* 181 */ + u8 raw_ext_csd_structure; /* 194 */ + u8 raw_card_type; /* 196 */ + u8 raw_s_a_timeout; /* 217 */ + u8 raw_hc_erase_gap_size; /* 221 */ + u8 raw_erase_timeout_mult; /* 223 */ + u8 raw_hc_erase_grp_size; /* 224 */ + u8 raw_sec_trim_mult; /* 229 */ + u8 raw_sec_erase_mult; /* 230 */ + u8 raw_sec_feature_support;/* 231 */ + u8 raw_trim_mult; /* 232 */ + u8 raw_sectors[4]; /* 212 - 4 bytes */ }; struct sd_scr { diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index 54b8b4d7b68f..9e19477991ad 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -1097,12 +1097,6 @@ struct net_device { #define NETIF_F_ALL_FCOE (NETIF_F_FCOE_CRC | NETIF_F_FCOE_MTU | \ NETIF_F_FSO) -#define NETIF_F_ALL_TX_OFFLOADS (NETIF_F_ALL_CSUM | NETIF_F_SG | \ - NETIF_F_FRAGLIST | NETIF_F_ALL_TSO | \ - NETIF_F_HIGHDMA | \ - NETIF_F_SCTP_CSUM | \ - NETIF_F_ALL_FCOE) - /* * If one device supports one of these features, then enable them * for all in netdev_increment_features. diff --git a/include/linux/sched.h b/include/linux/sched.h index 496770a96487..14a6c7b545de 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -844,6 +844,7 @@ enum cpu_idle_type { #define SD_SERIALIZE 0x0400 /* Only a single load balancing instance */ #define SD_ASYM_PACKING 0x0800 /* Place busy groups earlier in the domain */ #define SD_PREFER_SIBLING 0x1000 /* Prefer to place tasks in a sibling domain */ +#define SD_OVERLAP 0x2000 /* sched_domains of this level overlap */ enum powersavings_balance_level { POWERSAVINGS_BALANCE_NONE = 0, /* No power saving load balance */ @@ -893,16 +894,21 @@ static inline int sd_power_saving_flags(void) return 0; } -struct sched_group { - struct sched_group *next; /* Must be a circular list */ +struct sched_group_power { atomic_t ref; - /* * CPU power of this group, SCHED_LOAD_SCALE being max power for a * single CPU. */ - unsigned int cpu_power, cpu_power_orig; + unsigned int power, power_orig; +}; + +struct sched_group { + struct sched_group *next; /* Must be a circular list */ + atomic_t ref; + unsigned int group_weight; + struct sched_group_power *sgp; /* * The CPUs this group covers. @@ -1254,6 +1260,9 @@ struct task_struct { #ifdef CONFIG_PREEMPT_RCU int rcu_read_lock_nesting; char rcu_read_unlock_special; +#if defined(CONFIG_RCU_BOOST) && defined(CONFIG_TREE_PREEMPT_RCU) + int rcu_boosted; +#endif /* #if defined(CONFIG_RCU_BOOST) && defined(CONFIG_TREE_PREEMPT_RCU) */ struct list_head rcu_node_entry; #endif /* #ifdef CONFIG_PREEMPT_RCU */ #ifdef CONFIG_TREE_PREEMPT_RCU diff --git a/include/linux/sdla.h b/include/linux/sdla.h index 564acd3a71c1..9995c7fc3f60 100644 --- a/include/linux/sdla.h +++ b/include/linux/sdla.h @@ -112,11 +112,7 @@ struct sdla_dlci_conf { short Tb_max; }; -#ifndef __KERNEL__ - -void sdla(void *cfg_info, char *dev, struct frad_conf *conf, int quiet); - -#else +#ifdef __KERNEL__ /* important Z80 window addresses */ #define SDLA_CONTROL_WND 0xE000 diff --git a/include/net/sctp/command.h b/include/net/sctp/command.h index dd6847e5d6e4..6506458ccd33 100644 --- a/include/net/sctp/command.h +++ b/include/net/sctp/command.h @@ -63,6 +63,7 @@ typedef enum { SCTP_CMD_ECN_ECNE, /* Do delayed ECNE processing. */ SCTP_CMD_ECN_CWR, /* Do delayed CWR processing. */ SCTP_CMD_TIMER_START, /* Start a timer. */ + SCTP_CMD_TIMER_START_ONCE, /* Start a timer once */ SCTP_CMD_TIMER_RESTART, /* Restart a timer. */ SCTP_CMD_TIMER_STOP, /* Stop a timer. */ SCTP_CMD_INIT_CHOOSE_TRANSPORT, /* Choose transport for an INIT. */ diff --git a/include/net/sctp/ulpevent.h b/include/net/sctp/ulpevent.h index 99b027b2adce..ca4693b4e09e 100644 --- a/include/net/sctp/ulpevent.h +++ b/include/net/sctp/ulpevent.h @@ -80,7 +80,7 @@ static inline struct sctp_ulpevent *sctp_skb2event(struct sk_buff *skb) void sctp_ulpevent_free(struct sctp_ulpevent *); int sctp_ulpevent_is_notification(const struct sctp_ulpevent *); -void sctp_queue_purge_ulpevents(struct sk_buff_head *list); +unsigned int sctp_queue_purge_ulpevents(struct sk_buff_head *list); struct sctp_ulpevent *sctp_ulpevent_make_assoc_change( const struct sctp_association *asoc, diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 7e59ffb3d0ba..ba06207b1dd3 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -84,9 +84,32 @@ DEFINE_PER_CPU(struct rcu_data, rcu_bh_data); static struct rcu_state *rcu_state; +/* + * The rcu_scheduler_active variable transitions from zero to one just + * before the first task is spawned. So when this variable is zero, RCU + * can assume that there is but one task, allowing RCU to (for example) + * optimized synchronize_sched() to a simple barrier(). When this variable + * is one, RCU must actually do all the hard work required to detect real + * grace periods. This variable is also used to suppress boot-time false + * positives from lockdep-RCU error checking. + */ int rcu_scheduler_active __read_mostly; EXPORT_SYMBOL_GPL(rcu_scheduler_active); +/* + * The rcu_scheduler_fully_active variable transitions from zero to one + * during the early_initcall() processing, which is after the scheduler + * is capable of creating new tasks. So RCU processing (for example, + * creating tasks for RCU priority boosting) must be delayed until after + * rcu_scheduler_fully_active transitions from zero to one. We also + * currently delay invocation of any RCU callbacks until after this point. + * + * It might later prove better for people registering RCU callbacks during + * early boot to take responsibility for these callbacks, but one step at + * a time. + */ +static int rcu_scheduler_fully_active __read_mostly; + #ifdef CONFIG_RCU_BOOST /* @@ -98,7 +121,6 @@ DEFINE_PER_CPU(unsigned int, rcu_cpu_kthread_status); DEFINE_PER_CPU(int, rcu_cpu_kthread_cpu); DEFINE_PER_CPU(unsigned int, rcu_cpu_kthread_loops); DEFINE_PER_CPU(char, rcu_cpu_has_work); -static char rcu_kthreads_spawnable; #endif /* #ifdef CONFIG_RCU_BOOST */ @@ -1467,6 +1489,8 @@ static void rcu_process_callbacks(struct softirq_action *unused) */ static void invoke_rcu_callbacks(struct rcu_state *rsp, struct rcu_data *rdp) { + if (unlikely(!ACCESS_ONCE(rcu_scheduler_fully_active))) + return; if (likely(!rsp->boost)) { rcu_do_batch(rsp, rdp); return; diff --git a/kernel/rcutree_plugin.h b/kernel/rcutree_plugin.h index 14dc7dd00902..8aafbb80b8b0 100644 --- a/kernel/rcutree_plugin.h +++ b/kernel/rcutree_plugin.h @@ -68,6 +68,7 @@ struct rcu_state rcu_preempt_state = RCU_STATE_INITIALIZER(rcu_preempt_state); DEFINE_PER_CPU(struct rcu_data, rcu_preempt_data); static struct rcu_state *rcu_state = &rcu_preempt_state; +static void rcu_read_unlock_special(struct task_struct *t); static int rcu_preempted_readers_exp(struct rcu_node *rnp); /* @@ -147,7 +148,7 @@ static void rcu_preempt_note_context_switch(int cpu) struct rcu_data *rdp; struct rcu_node *rnp; - if (t->rcu_read_lock_nesting && + if (t->rcu_read_lock_nesting > 0 && (t->rcu_read_unlock_special & RCU_READ_UNLOCK_BLOCKED) == 0) { /* Possibly blocking in an RCU read-side critical section. */ @@ -190,6 +191,14 @@ static void rcu_preempt_note_context_switch(int cpu) rnp->gp_tasks = &t->rcu_node_entry; } raw_spin_unlock_irqrestore(&rnp->lock, flags); + } else if (t->rcu_read_lock_nesting < 0 && + t->rcu_read_unlock_special) { + + /* + * Complete exit from RCU read-side critical section on + * behalf of preempted instance of __rcu_read_unlock(). + */ + rcu_read_unlock_special(t); } /* @@ -284,7 +293,7 @@ static struct list_head *rcu_next_node_entry(struct task_struct *t, * notify RCU core processing or task having blocked during the RCU * read-side critical section. */ -static void rcu_read_unlock_special(struct task_struct *t) +static noinline void rcu_read_unlock_special(struct task_struct *t) { int empty; int empty_exp; @@ -309,7 +318,7 @@ static void rcu_read_unlock_special(struct task_struct *t) } /* Hardware IRQ handlers cannot block. */ - if (in_irq()) { + if (in_irq() || in_serving_softirq()) { local_irq_restore(flags); return; } @@ -342,6 +351,11 @@ static void rcu_read_unlock_special(struct task_struct *t) #ifdef CONFIG_RCU_BOOST if (&t->rcu_node_entry == rnp->boost_tasks) rnp->boost_tasks = np; + /* Snapshot and clear ->rcu_boosted with rcu_node lock held. */ + if (t->rcu_boosted) { + special |= RCU_READ_UNLOCK_BOOSTED; + t->rcu_boosted = 0; + } #endif /* #ifdef CONFIG_RCU_BOOST */ t->rcu_blocked_node = NULL; @@ -358,7 +372,6 @@ static void rcu_read_unlock_special(struct task_struct *t) #ifdef CONFIG_RCU_BOOST /* Unboost if we were boosted. */ if (special & RCU_READ_UNLOCK_BOOSTED) { - t->rcu_read_unlock_special &= ~RCU_READ_UNLOCK_BOOSTED; rt_mutex_unlock(t->rcu_boost_mutex); t->rcu_boost_mutex = NULL; } @@ -387,13 +400,22 @@ void __rcu_read_unlock(void) struct task_struct *t = current; barrier(); /* needed if we ever invoke rcu_read_unlock in rcutree.c */ - --t->rcu_read_lock_nesting; - barrier(); /* decrement before load of ->rcu_read_unlock_special */ - if (t->rcu_read_lock_nesting == 0 && - unlikely(ACCESS_ONCE(t->rcu_read_unlock_special))) - rcu_read_unlock_special(t); + if (t->rcu_read_lock_nesting != 1) + --t->rcu_read_lock_nesting; + else { + t->rcu_read_lock_nesting = INT_MIN; + barrier(); /* assign before ->rcu_read_unlock_special load */ + if (unlikely(ACCESS_ONCE(t->rcu_read_unlock_special))) + rcu_read_unlock_special(t); + barrier(); /* ->rcu_read_unlock_special load before assign */ + t->rcu_read_lock_nesting = 0; + } #ifdef CONFIG_PROVE_LOCKING - WARN_ON_ONCE(ACCESS_ONCE(t->rcu_read_lock_nesting) < 0); + { + int rrln = ACCESS_ONCE(t->rcu_read_lock_nesting); + + WARN_ON_ONCE(rrln < 0 && rrln > INT_MIN / 2); + } #endif /* #ifdef CONFIG_PROVE_LOCKING */ } EXPORT_SYMBOL_GPL(__rcu_read_unlock); @@ -589,7 +611,8 @@ static void rcu_preempt_check_callbacks(int cpu) rcu_preempt_qs(cpu); return; } - if (per_cpu(rcu_preempt_data, cpu).qs_pending) + if (t->rcu_read_lock_nesting > 0 && + per_cpu(rcu_preempt_data, cpu).qs_pending) t->rcu_read_unlock_special |= RCU_READ_UNLOCK_NEED_QS; } @@ -695,9 +718,12 @@ static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp) raw_spin_lock_irqsave(&rnp->lock, flags); for (;;) { - if (!sync_rcu_preempt_exp_done(rnp)) + if (!sync_rcu_preempt_exp_done(rnp)) { + raw_spin_unlock_irqrestore(&rnp->lock, flags); break; + } if (rnp->parent == NULL) { + raw_spin_unlock_irqrestore(&rnp->lock, flags); wake_up(&sync_rcu_preempt_exp_wq); break; } @@ -707,7 +733,6 @@ static void rcu_report_exp_rnp(struct rcu_state *rsp, struct rcu_node *rnp) raw_spin_lock(&rnp->lock); /* irqs already disabled */ rnp->expmask &= ~mask; } - raw_spin_unlock_irqrestore(&rnp->lock, flags); } /* @@ -1174,7 +1199,7 @@ static int rcu_boost(struct rcu_node *rnp) t = container_of(tb, struct task_struct, rcu_node_entry); rt_mutex_init_proxy_locked(&mtx, t); t->rcu_boost_mutex = &mtx; - t->rcu_read_unlock_special |= RCU_READ_UNLOCK_BOOSTED; + t->rcu_boosted = 1; raw_spin_unlock_irqrestore(&rnp->lock, flags); rt_mutex_lock(&mtx); /* Side effect: boosts task t's priority. */ rt_mutex_unlock(&mtx); /* Keep lockdep happy. */ @@ -1532,7 +1557,7 @@ static int __cpuinit rcu_spawn_one_cpu_kthread(int cpu) struct sched_param sp; struct task_struct *t; - if (!rcu_kthreads_spawnable || + if (!rcu_scheduler_fully_active || per_cpu(rcu_cpu_kthread_task, cpu) != NULL) return 0; t = kthread_create(rcu_cpu_kthread, (void *)(long)cpu, "rcuc%d", cpu); @@ -1639,7 +1664,7 @@ static int __cpuinit rcu_spawn_one_node_kthread(struct rcu_state *rsp, struct sched_param sp; struct task_struct *t; - if (!rcu_kthreads_spawnable || + if (!rcu_scheduler_fully_active || rnp->qsmaskinit == 0) return 0; if (rnp->node_kthread_task == NULL) { @@ -1665,7 +1690,7 @@ static int __init rcu_spawn_kthreads(void) int cpu; struct rcu_node *rnp; - rcu_kthreads_spawnable = 1; + rcu_scheduler_fully_active = 1; for_each_possible_cpu(cpu) { per_cpu(rcu_cpu_has_work, cpu) = 0; if (cpu_online(cpu)) @@ -1687,7 +1712,7 @@ static void __cpuinit rcu_prepare_kthreads(int cpu) struct rcu_node *rnp = rdp->mynode; /* Fire up the incoming CPU's kthread and leaf rcu_node kthread. */ - if (rcu_kthreads_spawnable) { + if (rcu_scheduler_fully_active) { (void)rcu_spawn_one_cpu_kthread(cpu); if (rnp->node_kthread_task == NULL) (void)rcu_spawn_one_node_kthread(rcu_state, rnp); @@ -1726,6 +1751,13 @@ static void rcu_cpu_kthread_setrt(int cpu, int to_rt) { } +static int __init rcu_scheduler_really_started(void) +{ + rcu_scheduler_fully_active = 1; + return 0; +} +early_initcall(rcu_scheduler_really_started); + static void __cpuinit rcu_prepare_kthreads(int cpu) { } diff --git a/kernel/sched.c b/kernel/sched.c index 9769c756ad66..fde6ff903525 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -2544,13 +2544,9 @@ static int ttwu_remote(struct task_struct *p, int wake_flags) } #ifdef CONFIG_SMP -static void sched_ttwu_pending(void) +static void sched_ttwu_do_pending(struct task_struct *list) { struct rq *rq = this_rq(); - struct task_struct *list = xchg(&rq->wake_list, NULL); - - if (!list) - return; raw_spin_lock(&rq->lock); @@ -2563,9 +2559,45 @@ static void sched_ttwu_pending(void) raw_spin_unlock(&rq->lock); } +#ifdef CONFIG_HOTPLUG_CPU + +static void sched_ttwu_pending(void) +{ + struct rq *rq = this_rq(); + struct task_struct *list = xchg(&rq->wake_list, NULL); + + if (!list) + return; + + sched_ttwu_do_pending(list); +} + +#endif /* CONFIG_HOTPLUG_CPU */ + void scheduler_ipi(void) { - sched_ttwu_pending(); + struct rq *rq = this_rq(); + struct task_struct *list = xchg(&rq->wake_list, NULL); + + if (!list) + return; + + /* + * Not all reschedule IPI handlers call irq_enter/irq_exit, since + * traditionally all their work was done from the interrupt return + * path. Now that we actually do some work, we need to make sure + * we do call them. + * + * Some archs already do call them, luckily irq_enter/exit nest + * properly. + * + * Arguably we should visit all archs and update all handlers, + * however a fair share of IPIs are still resched only so this would + * somewhat pessimize the simple resched case. + */ + irq_enter(); + sched_ttwu_do_pending(list); + irq_exit(); } static void ttwu_queue_remote(struct task_struct *p, int cpu) @@ -6557,7 +6589,7 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, break; } - if (!group->cpu_power) { + if (!group->sgp->power) { printk(KERN_CONT "\n"); printk(KERN_ERR "ERROR: domain->cpu_power not " "set\n"); @@ -6581,9 +6613,9 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, cpulist_scnprintf(str, sizeof(str), sched_group_cpus(group)); printk(KERN_CONT " %s", str); - if (group->cpu_power != SCHED_POWER_SCALE) { + if (group->sgp->power != SCHED_POWER_SCALE) { printk(KERN_CONT " (cpu_power = %d)", - group->cpu_power); + group->sgp->power); } group = group->next; @@ -6774,11 +6806,39 @@ static struct root_domain *alloc_rootdomain(void) return rd; } +static void free_sched_groups(struct sched_group *sg, int free_sgp) +{ + struct sched_group *tmp, *first; + + if (!sg) + return; + + first = sg; + do { + tmp = sg->next; + + if (free_sgp && atomic_dec_and_test(&sg->sgp->ref)) + kfree(sg->sgp); + + kfree(sg); + sg = tmp; + } while (sg != first); +} + static void free_sched_domain(struct rcu_head *rcu) { struct sched_domain *sd = container_of(rcu, struct sched_domain, rcu); - if (atomic_dec_and_test(&sd->groups->ref)) + + /* + * If its an overlapping domain it has private groups, iterate and + * nuke them all. + */ + if (sd->flags & SD_OVERLAP) { + free_sched_groups(sd->groups, 1); + } else if (atomic_dec_and_test(&sd->groups->ref)) { + kfree(sd->groups->sgp); kfree(sd->groups); + } kfree(sd); } @@ -6945,6 +7005,7 @@ int sched_smt_power_savings = 0, sched_mc_power_savings = 0; struct sd_data { struct sched_domain **__percpu sd; struct sched_group **__percpu sg; + struct sched_group_power **__percpu sgp; }; struct s_data { @@ -6964,15 +7025,73 @@ struct sched_domain_topology_level; typedef struct sched_domain *(*sched_domain_init_f)(struct sched_domain_topology_level *tl, int cpu); typedef const struct cpumask *(*sched_domain_mask_f)(int cpu); +#define SDTL_OVERLAP 0x01 + struct sched_domain_topology_level { sched_domain_init_f init; sched_domain_mask_f mask; + int flags; struct sd_data data; }; -/* - * Assumes the sched_domain tree is fully constructed - */ +static int +build_overlap_sched_groups(struct sched_domain *sd, int cpu) +{ + struct sched_group *first = NULL, *last = NULL, *groups = NULL, *sg; + const struct cpumask *span = sched_domain_span(sd); + struct cpumask *covered = sched_domains_tmpmask; + struct sd_data *sdd = sd->private; + struct sched_domain *child; + int i; + + cpumask_clear(covered); + + for_each_cpu(i, span) { + struct cpumask *sg_span; + + if (cpumask_test_cpu(i, covered)) + continue; + + sg = kzalloc_node(sizeof(struct sched_group) + cpumask_size(), + GFP_KERNEL, cpu_to_node(i)); + + if (!sg) + goto fail; + + sg_span = sched_group_cpus(sg); + + child = *per_cpu_ptr(sdd->sd, i); + if (child->child) { + child = child->child; + cpumask_copy(sg_span, sched_domain_span(child)); + } else + cpumask_set_cpu(i, sg_span); + + cpumask_or(covered, covered, sg_span); + + sg->sgp = *per_cpu_ptr(sdd->sgp, cpumask_first(sg_span)); + atomic_inc(&sg->sgp->ref); + + if (cpumask_test_cpu(cpu, sg_span)) + groups = sg; + + if (!first) + first = sg; + if (last) + last->next = sg; + last = sg; + last->next = first; + } + sd->groups = groups; + + return 0; + +fail: + free_sched_groups(first, 0); + + return -ENOMEM; +} + static int get_group(int cpu, struct sd_data *sdd, struct sched_group **sg) { struct sched_domain *sd = *per_cpu_ptr(sdd->sd, cpu); @@ -6981,24 +7100,24 @@ static int get_group(int cpu, struct sd_data *sdd, struct sched_group **sg) if (child) cpu = cpumask_first(sched_domain_span(child)); - if (sg) + if (sg) { *sg = *per_cpu_ptr(sdd->sg, cpu); + (*sg)->sgp = *per_cpu_ptr(sdd->sgp, cpu); + atomic_set(&(*sg)->sgp->ref, 1); /* for claim_allocations */ + } return cpu; } /* - * build_sched_groups takes the cpumask we wish to span, and a pointer - * to a function which identifies what group(along with sched group) a CPU - * belongs to. The return value of group_fn must be a >= 0 and < nr_cpu_ids - * (due to the fact that we keep track of groups covered with a struct cpumask). - * * build_sched_groups will build a circular linked list of the groups * covered by the given span, and will set each group's ->cpumask correctly, * and ->cpu_power to 0. + * + * Assumes the sched_domain tree is fully constructed */ -static void -build_sched_groups(struct sched_domain *sd) +static int +build_sched_groups(struct sched_domain *sd, int cpu) { struct sched_group *first = NULL, *last = NULL; struct sd_data *sdd = sd->private; @@ -7006,6 +7125,12 @@ build_sched_groups(struct sched_domain *sd) struct cpumask *covered; int i; + get_group(cpu, sdd, &sd->groups); + atomic_inc(&sd->groups->ref); + + if (cpu != cpumask_first(sched_domain_span(sd))) + return 0; + lockdep_assert_held(&sched_domains_mutex); covered = sched_domains_tmpmask; @@ -7020,7 +7145,7 @@ build_sched_groups(struct sched_domain *sd) continue; cpumask_clear(sched_group_cpus(sg)); - sg->cpu_power = 0; + sg->sgp->power = 0; for_each_cpu(j, span) { if (get_group(j, sdd, NULL) != group) @@ -7037,6 +7162,8 @@ build_sched_groups(struct sched_domain *sd) last = sg; } last->next = first; + + return 0; } /* @@ -7051,12 +7178,17 @@ build_sched_groups(struct sched_domain *sd) */ static void init_sched_groups_power(int cpu, struct sched_domain *sd) { - WARN_ON(!sd || !sd->groups); + struct sched_group *sg = sd->groups; - if (cpu != group_first_cpu(sd->groups)) - return; + WARN_ON(!sd || !sg); + + do { + sg->group_weight = cpumask_weight(sched_group_cpus(sg)); + sg = sg->next; + } while (sg != sd->groups); - sd->groups->group_weight = cpumask_weight(sched_group_cpus(sd->groups)); + if (cpu != group_first_cpu(sg)) + return; update_group_power(sd, cpu); } @@ -7177,15 +7309,15 @@ static enum s_alloc __visit_domain_allocation_hell(struct s_data *d, static void claim_allocations(int cpu, struct sched_domain *sd) { struct sd_data *sdd = sd->private; - struct sched_group *sg = sd->groups; WARN_ON_ONCE(*per_cpu_ptr(sdd->sd, cpu) != sd); *per_cpu_ptr(sdd->sd, cpu) = NULL; - if (cpu == cpumask_first(sched_group_cpus(sg))) { - WARN_ON_ONCE(*per_cpu_ptr(sdd->sg, cpu) != sg); + if (atomic_read(&(*per_cpu_ptr(sdd->sg, cpu))->ref)) *per_cpu_ptr(sdd->sg, cpu) = NULL; - } + + if (atomic_read(&(*per_cpu_ptr(sdd->sgp, cpu))->ref)) + *per_cpu_ptr(sdd->sgp, cpu) = NULL; } #ifdef CONFIG_SCHED_SMT @@ -7210,7 +7342,7 @@ static struct sched_domain_topology_level default_topology[] = { #endif { sd_init_CPU, cpu_cpu_mask, }, #ifdef CONFIG_NUMA - { sd_init_NODE, cpu_node_mask, }, + { sd_init_NODE, cpu_node_mask, SDTL_OVERLAP, }, { sd_init_ALLNODES, cpu_allnodes_mask, }, #endif { NULL, }, @@ -7234,9 +7366,14 @@ static int __sdt_alloc(const struct cpumask *cpu_map) if (!sdd->sg) return -ENOMEM; + sdd->sgp = alloc_percpu(struct sched_group_power *); + if (!sdd->sgp) + return -ENOMEM; + for_each_cpu(j, cpu_map) { struct sched_domain *sd; struct sched_group *sg; + struct sched_group_power *sgp; sd = kzalloc_node(sizeof(struct sched_domain) + cpumask_size(), GFP_KERNEL, cpu_to_node(j)); @@ -7251,6 +7388,13 @@ static int __sdt_alloc(const struct cpumask *cpu_map) return -ENOMEM; *per_cpu_ptr(sdd->sg, j) = sg; + + sgp = kzalloc_node(sizeof(struct sched_group_power), + GFP_KERNEL, cpu_to_node(j)); + if (!sgp) + return -ENOMEM; + + *per_cpu_ptr(sdd->sgp, j) = sgp; } } @@ -7266,11 +7410,15 @@ static void __sdt_free(const struct cpumask *cpu_map) struct sd_data *sdd = &tl->data; for_each_cpu(j, cpu_map) { - kfree(*per_cpu_ptr(sdd->sd, j)); + struct sched_domain *sd = *per_cpu_ptr(sdd->sd, j); + if (sd && (sd->flags & SD_OVERLAP)) + free_sched_groups(sd->groups, 0); kfree(*per_cpu_ptr(sdd->sg, j)); + kfree(*per_cpu_ptr(sdd->sgp, j)); } free_percpu(sdd->sd); free_percpu(sdd->sg); + free_percpu(sdd->sgp); } } @@ -7316,8 +7464,13 @@ static int build_sched_domains(const struct cpumask *cpu_map, struct sched_domain_topology_level *tl; sd = NULL; - for (tl = sched_domain_topology; tl->init; tl++) + for (tl = sched_domain_topology; tl->init; tl++) { sd = build_sched_domain(tl, &d, cpu_map, attr, sd, i); + if (tl->flags & SDTL_OVERLAP || sched_feat(FORCE_SD_OVERLAP)) + sd->flags |= SD_OVERLAP; + if (cpumask_equal(cpu_map, sched_domain_span(sd))) + break; + } while (sd->child) sd = sd->child; @@ -7329,13 +7482,13 @@ static int build_sched_domains(const struct cpumask *cpu_map, for_each_cpu(i, cpu_map) { for (sd = *per_cpu_ptr(d.sd, i); sd; sd = sd->parent) { sd->span_weight = cpumask_weight(sched_domain_span(sd)); - get_group(i, sd->private, &sd->groups); - atomic_inc(&sd->groups->ref); - - if (i != cpumask_first(sched_domain_span(sd))) - continue; - - build_sched_groups(sd); + if (sd->flags & SD_OVERLAP) { + if (build_overlap_sched_groups(sd, i)) + goto error; + } else { + if (build_sched_groups(sd, i)) + goto error; + } } } @@ -7757,6 +7910,9 @@ static void init_cfs_rq(struct cfs_rq *cfs_rq, struct rq *rq) #endif #endif cfs_rq->min_vruntime = (u64)(-(1LL << 20)); +#ifndef CONFIG_64BIT + cfs_rq->min_vruntime_copy = cfs_rq->min_vruntime; +#endif } static void init_rt_rq(struct rt_rq *rt_rq, struct rq *rq) diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 433491c2dc8f..c768588e180b 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1585,7 +1585,7 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, } /* Adjust by relative CPU power of the group */ - avg_load = (avg_load * SCHED_POWER_SCALE) / group->cpu_power; + avg_load = (avg_load * SCHED_POWER_SCALE) / group->sgp->power; if (local_group) { this_load = avg_load; @@ -2631,7 +2631,7 @@ static void update_cpu_power(struct sched_domain *sd, int cpu) power >>= SCHED_POWER_SHIFT; } - sdg->cpu_power_orig = power; + sdg->sgp->power_orig = power; if (sched_feat(ARCH_POWER)) power *= arch_scale_freq_power(sd, cpu); @@ -2647,7 +2647,7 @@ static void update_cpu_power(struct sched_domain *sd, int cpu) power = 1; cpu_rq(cpu)->cpu_power = power; - sdg->cpu_power = power; + sdg->sgp->power = power; } static void update_group_power(struct sched_domain *sd, int cpu) @@ -2665,11 +2665,11 @@ static void update_group_power(struct sched_domain *sd, int cpu) group = child->groups; do { - power += group->cpu_power; + power += group->sgp->power; group = group->next; } while (group != child->groups); - sdg->cpu_power = power; + sdg->sgp->power = power; } /* @@ -2691,7 +2691,7 @@ fix_small_capacity(struct sched_domain *sd, struct sched_group *group) /* * If ~90% of the cpu_power is still there, we're good. */ - if (group->cpu_power * 32 > group->cpu_power_orig * 29) + if (group->sgp->power * 32 > group->sgp->power_orig * 29) return 1; return 0; @@ -2771,7 +2771,7 @@ static inline void update_sg_lb_stats(struct sched_domain *sd, } /* Adjust by relative CPU power of the group */ - sgs->avg_load = (sgs->group_load*SCHED_POWER_SCALE) / group->cpu_power; + sgs->avg_load = (sgs->group_load*SCHED_POWER_SCALE) / group->sgp->power; /* * Consider the group unbalanced when the imbalance is larger @@ -2788,7 +2788,7 @@ static inline void update_sg_lb_stats(struct sched_domain *sd, if ((max_cpu_load - min_cpu_load) >= avg_load_per_task && max_nr_running > 1) sgs->group_imb = 1; - sgs->group_capacity = DIV_ROUND_CLOSEST(group->cpu_power, + sgs->group_capacity = DIV_ROUND_CLOSEST(group->sgp->power, SCHED_POWER_SCALE); if (!sgs->group_capacity) sgs->group_capacity = fix_small_capacity(sd, group); @@ -2877,7 +2877,7 @@ static inline void update_sd_lb_stats(struct sched_domain *sd, int this_cpu, return; sds->total_load += sgs.group_load; - sds->total_pwr += sg->cpu_power; + sds->total_pwr += sg->sgp->power; /* * In case the child domain prefers tasks go to siblings @@ -2962,7 +2962,7 @@ static int check_asym_packing(struct sched_domain *sd, if (this_cpu > busiest_cpu) return 0; - *imbalance = DIV_ROUND_CLOSEST(sds->max_load * sds->busiest->cpu_power, + *imbalance = DIV_ROUND_CLOSEST(sds->max_load * sds->busiest->sgp->power, SCHED_POWER_SCALE); return 1; } @@ -2993,7 +2993,7 @@ static inline void fix_small_imbalance(struct sd_lb_stats *sds, scaled_busy_load_per_task = sds->busiest_load_per_task * SCHED_POWER_SCALE; - scaled_busy_load_per_task /= sds->busiest->cpu_power; + scaled_busy_load_per_task /= sds->busiest->sgp->power; if (sds->max_load - sds->this_load + scaled_busy_load_per_task >= (scaled_busy_load_per_task * imbn)) { @@ -3007,28 +3007,28 @@ static inline void fix_small_imbalance(struct sd_lb_stats *sds, * moving them. */ - pwr_now += sds->busiest->cpu_power * + pwr_now += sds->busiest->sgp->power * min(sds->busiest_load_per_task, sds->max_load); - pwr_now += sds->this->cpu_power * + pwr_now += sds->this->sgp->power * min(sds->this_load_per_task, sds->this_load); pwr_now /= SCHED_POWER_SCALE; /* Amount of load we'd subtract */ tmp = (sds->busiest_load_per_task * SCHED_POWER_SCALE) / - sds->busiest->cpu_power; + sds->busiest->sgp->power; if (sds->max_load > tmp) - pwr_move += sds->busiest->cpu_power * + pwr_move += sds->busiest->sgp->power * min(sds->busiest_load_per_task, sds->max_load - tmp); /* Amount of load we'd add */ - if (sds->max_load * sds->busiest->cpu_power < + if (sds->max_load * sds->busiest->sgp->power < sds->busiest_load_per_task * SCHED_POWER_SCALE) - tmp = (sds->max_load * sds->busiest->cpu_power) / - sds->this->cpu_power; + tmp = (sds->max_load * sds->busiest->sgp->power) / + sds->this->sgp->power; else tmp = (sds->busiest_load_per_task * SCHED_POWER_SCALE) / - sds->this->cpu_power; - pwr_move += sds->this->cpu_power * + sds->this->sgp->power; + pwr_move += sds->this->sgp->power * min(sds->this_load_per_task, sds->this_load + tmp); pwr_move /= SCHED_POWER_SCALE; @@ -3074,7 +3074,7 @@ static inline void calculate_imbalance(struct sd_lb_stats *sds, int this_cpu, load_above_capacity *= (SCHED_LOAD_SCALE * SCHED_POWER_SCALE); - load_above_capacity /= sds->busiest->cpu_power; + load_above_capacity /= sds->busiest->sgp->power; } /* @@ -3090,8 +3090,8 @@ static inline void calculate_imbalance(struct sd_lb_stats *sds, int this_cpu, max_pull = min(sds->max_load - sds->avg_load, load_above_capacity); /* How much load to actually move to equalise the imbalance */ - *imbalance = min(max_pull * sds->busiest->cpu_power, - (sds->avg_load - sds->this_load) * sds->this->cpu_power) + *imbalance = min(max_pull * sds->busiest->sgp->power, + (sds->avg_load - sds->this_load) * sds->this->sgp->power) / SCHED_POWER_SCALE; /* diff --git a/kernel/sched_features.h b/kernel/sched_features.h index be40f7371ee1..1e7066d76c26 100644 --- a/kernel/sched_features.h +++ b/kernel/sched_features.h @@ -70,3 +70,5 @@ SCHED_FEAT(NONIRQ_POWER, 1) * using the scheduler IPI. Reduces rq->lock contention/bounces. */ SCHED_FEAT(TTWU_QUEUE, 1) + +SCHED_FEAT(FORCE_SD_OVERLAP, 0) diff --git a/kernel/signal.c b/kernel/signal.c index ff7678603328..415d85d6f6c6 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -1178,18 +1178,25 @@ struct sighand_struct *__lock_task_sighand(struct task_struct *tsk, { struct sighand_struct *sighand; - rcu_read_lock(); for (;;) { + local_irq_save(*flags); + rcu_read_lock(); sighand = rcu_dereference(tsk->sighand); - if (unlikely(sighand == NULL)) + if (unlikely(sighand == NULL)) { + rcu_read_unlock(); + local_irq_restore(*flags); break; + } - spin_lock_irqsave(&sighand->siglock, *flags); - if (likely(sighand == tsk->sighand)) + spin_lock(&sighand->siglock); + if (likely(sighand == tsk->sighand)) { + rcu_read_unlock(); break; - spin_unlock_irqrestore(&sighand->siglock, *flags); + } + spin_unlock(&sighand->siglock); + rcu_read_unlock(); + local_irq_restore(*flags); } - rcu_read_unlock(); return sighand; } diff --git a/kernel/softirq.c b/kernel/softirq.c index 40cf63ddd4b3..fca82c32042b 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -315,16 +315,24 @@ static inline void invoke_softirq(void) { if (!force_irqthreads) __do_softirq(); - else + else { + __local_bh_disable((unsigned long)__builtin_return_address(0), + SOFTIRQ_OFFSET); wakeup_softirqd(); + __local_bh_enable(SOFTIRQ_OFFSET); + } } #else static inline void invoke_softirq(void) { if (!force_irqthreads) do_softirq(); - else + else { + __local_bh_disable((unsigned long)__builtin_return_address(0), + SOFTIRQ_OFFSET); wakeup_softirqd(); + __local_bh_enable(SOFTIRQ_OFFSET); + } } #endif diff --git a/mm/vmscan.c b/mm/vmscan.c index 5ed24b94c5e6..d036e59d302b 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2310,7 +2310,8 @@ static bool pgdat_balanced(pg_data_t *pgdat, unsigned long balanced_pages, for (i = 0; i <= classzone_idx; i++) present_pages += pgdat->node_zones[i].present_pages; - return balanced_pages > (present_pages >> 2); + /* A special case here: if zone has no page, we think it's balanced */ + return balanced_pages >= (present_pages >> 2); } /* is kswapd sleeping prematurely? */ diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c index 86bff9b1ac47..6e82148edfc8 100644 --- a/net/8021q/vlan_dev.c +++ b/net/8021q/vlan_dev.c @@ -528,7 +528,11 @@ static int vlan_dev_init(struct net_device *dev) (1<<__LINK_STATE_DORMANT))) | (1<<__LINK_STATE_PRESENT); - dev->hw_features = NETIF_F_ALL_TX_OFFLOADS; + dev->hw_features = NETIF_F_ALL_CSUM | NETIF_F_SG | + NETIF_F_FRAGLIST | NETIF_F_ALL_TSO | + NETIF_F_HIGHDMA | NETIF_F_SCTP_CSUM | + NETIF_F_ALL_FCOE; + dev->features |= real_dev->vlan_features | NETIF_F_LLTX; dev->gso_max_size = real_dev->gso_max_size; diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index d3a05b9ade7a..bcd158f40bb9 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -393,6 +393,9 @@ int hci_conn_del(struct hci_conn *conn) hci_dev_put(hdev); + if (conn->handle == 0) + kfree(conn); + return 0; } diff --git a/net/bluetooth/hidp/core.c b/net/bluetooth/hidp/core.c index c405a954a603..43b4c2deb7cc 100644 --- a/net/bluetooth/hidp/core.c +++ b/net/bluetooth/hidp/core.c @@ -464,7 +464,8 @@ static void hidp_idle_timeout(unsigned long arg) { struct hidp_session *session = (struct hidp_session *) arg; - kthread_stop(session->task); + atomic_inc(&session->terminate); + wake_up_process(session->task); } static void hidp_set_timer(struct hidp_session *session) @@ -535,7 +536,8 @@ static void hidp_process_hid_control(struct hidp_session *session, skb_queue_purge(&session->ctrl_transmit); skb_queue_purge(&session->intr_transmit); - kthread_stop(session->task); + atomic_inc(&session->terminate); + wake_up_process(current); } } @@ -706,9 +708,8 @@ static int hidp_session(void *arg) add_wait_queue(sk_sleep(intr_sk), &intr_wait); session->waiting_for_startup = 0; wake_up_interruptible(&session->startup_queue); - while (!kthread_should_stop()) { - set_current_state(TASK_INTERRUPTIBLE); - + set_current_state(TASK_INTERRUPTIBLE); + while (!atomic_read(&session->terminate)) { if (ctrl_sk->sk_state != BT_CONNECTED || intr_sk->sk_state != BT_CONNECTED) break; @@ -726,6 +727,7 @@ static int hidp_session(void *arg) hidp_process_transmit(session); schedule(); + set_current_state(TASK_INTERRUPTIBLE); } set_current_state(TASK_RUNNING); remove_wait_queue(sk_sleep(intr_sk), &intr_wait); @@ -1060,7 +1062,8 @@ int hidp_add_connection(struct hidp_connadd_req *req, struct socket *ctrl_sock, err_add_device: hid_destroy_device(session->hid); session->hid = NULL; - kthread_stop(session->task); + atomic_inc(&session->terminate); + wake_up_process(session->task); unlink: hidp_del_timer(session); @@ -1111,7 +1114,8 @@ int hidp_del_connection(struct hidp_conndel_req *req) skb_queue_purge(&session->ctrl_transmit); skb_queue_purge(&session->intr_transmit); - kthread_stop(session->task); + atomic_inc(&session->terminate); + wake_up_process(session->task); } } else err = -ENOENT; diff --git a/net/bluetooth/hidp/hidp.h b/net/bluetooth/hidp/hidp.h index 19e95004b286..af1bcc823f26 100644 --- a/net/bluetooth/hidp/hidp.h +++ b/net/bluetooth/hidp/hidp.h @@ -142,6 +142,7 @@ struct hidp_session { uint ctrl_mtu; uint intr_mtu; + atomic_t terminate; struct task_struct *task; unsigned char keys[8]; diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 56fdd9162da9..7705e26e699f 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -620,7 +620,8 @@ static void l2cap_conn_start(struct l2cap_conn *conn) struct sock *parent = bt_sk(sk)->parent; rsp.result = cpu_to_le16(L2CAP_CR_PEND); rsp.status = cpu_to_le16(L2CAP_CS_AUTHOR_PEND); - parent->sk_data_ready(parent, 0); + if (parent) + parent->sk_data_ready(parent, 0); } else { sk->sk_state = BT_CONFIG; @@ -2323,7 +2324,7 @@ static inline int l2cap_config_req(struct l2cap_conn *conn, struct l2cap_cmd_hdr sk = chan->sk; - if (sk->sk_state != BT_CONFIG) { + if (sk->sk_state != BT_CONFIG && sk->sk_state != BT_CONNECT2) { struct l2cap_cmd_rej rej; rej.reason = cpu_to_le16(0x0002); @@ -2334,7 +2335,7 @@ static inline int l2cap_config_req(struct l2cap_conn *conn, struct l2cap_cmd_hdr /* Reject if config buffer is too small. */ len = cmd_len - sizeof(*req); - if (chan->conf_len + len > sizeof(chan->conf_req)) { + if (len < 0 || chan->conf_len + len > sizeof(chan->conf_req)) { l2cap_send_cmd(conn, cmd->ident, L2CAP_CONF_RSP, l2cap_build_conf_rsp(chan, rsp, L2CAP_CONF_REJECT, flags), rsp); @@ -4009,7 +4010,8 @@ static int l2cap_security_cfm(struct hci_conn *hcon, u8 status, u8 encrypt) struct sock *parent = bt_sk(sk)->parent; res = L2CAP_CR_PEND; stat = L2CAP_CS_AUTHOR_PEND; - parent->sk_data_ready(parent, 0); + if (parent) + parent->sk_data_ready(parent, 0); } else { sk->sk_state = BT_CONFIG; res = L2CAP_CR_SUCCESS; diff --git a/net/ceph/ceph_fs.c b/net/ceph/ceph_fs.c index a3a3a31d3c37..41466ccb972a 100644 --- a/net/ceph/ceph_fs.c +++ b/net/ceph/ceph_fs.c @@ -36,16 +36,19 @@ int ceph_flags_to_mode(int flags) if ((flags & O_DIRECTORY) == O_DIRECTORY) return CEPH_FILE_MODE_PIN; #endif - if ((flags & O_APPEND) == O_APPEND) - flags |= O_WRONLY; - if ((flags & O_ACCMODE) == O_RDWR) - mode = CEPH_FILE_MODE_RDWR; - else if ((flags & O_ACCMODE) == O_WRONLY) + switch (flags & O_ACCMODE) { + case O_WRONLY: mode = CEPH_FILE_MODE_WR; - else + break; + case O_RDONLY: mode = CEPH_FILE_MODE_RD; - + break; + case O_RDWR: + case O_ACCMODE: /* this is what the VFS does */ + mode = CEPH_FILE_MODE_RDWR; + break; + } #ifdef O_LAZY if (flags & O_LAZY) mode |= CEPH_FILE_MODE_LAZY; diff --git a/net/mac80211/scan.c b/net/mac80211/scan.c index 58ffa7d069c7..669d2e32efb6 100644 --- a/net/mac80211/scan.c +++ b/net/mac80211/scan.c @@ -877,7 +877,8 @@ int ieee80211_request_sched_scan_start(struct ieee80211_sub_if_data *sdata, for (i = 0; i < IEEE80211_NUM_BANDS; i++) { local->sched_scan_ies.ie[i] = kzalloc(2 + IEEE80211_MAX_SSID_LEN + - local->scan_ies_len, + local->scan_ies_len + + req->ie_len, GFP_KERNEL); if (!local->sched_scan_ies.ie[i]) { ret = -ENOMEM; diff --git a/net/mac80211/wpa.c b/net/mac80211/wpa.c index d91c1a26630d..8f6a302d2ac3 100644 --- a/net/mac80211/wpa.c +++ b/net/mac80211/wpa.c @@ -86,6 +86,11 @@ ieee80211_rx_h_michael_mic_verify(struct ieee80211_rx_data *rx) struct sk_buff *skb = rx->skb; struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(skb); struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data; + int queue = rx->queue; + + /* otherwise, TKIP is vulnerable to TID 0 vs. non-QoS replays */ + if (rx->queue == NUM_RX_DATA_QUEUES - 1) + queue = 0; /* * it makes no sense to check for MIC errors on anything other @@ -148,8 +153,8 @@ ieee80211_rx_h_michael_mic_verify(struct ieee80211_rx_data *rx) update_iv: /* update IV in key information to be able to detect replays */ - rx->key->u.tkip.rx[rx->queue].iv32 = rx->tkip_iv32; - rx->key->u.tkip.rx[rx->queue].iv16 = rx->tkip_iv16; + rx->key->u.tkip.rx[queue].iv32 = rx->tkip_iv32; + rx->key->u.tkip.rx[queue].iv16 = rx->tkip_iv16; return RX_CONTINUE; @@ -241,6 +246,11 @@ ieee80211_crypto_tkip_decrypt(struct ieee80211_rx_data *rx) struct ieee80211_key *key = rx->key; struct sk_buff *skb = rx->skb; struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(skb); + int queue = rx->queue; + + /* otherwise, TKIP is vulnerable to TID 0 vs. non-QoS replays */ + if (rx->queue == NUM_RX_DATA_QUEUES - 1) + queue = 0; hdrlen = ieee80211_hdrlen(hdr->frame_control); @@ -261,7 +271,7 @@ ieee80211_crypto_tkip_decrypt(struct ieee80211_rx_data *rx) res = ieee80211_tkip_decrypt_data(rx->local->wep_rx_tfm, key, skb->data + hdrlen, skb->len - hdrlen, rx->sta->sta.addr, - hdr->addr1, hwaccel, rx->queue, + hdr->addr1, hwaccel, queue, &rx->tkip_iv32, &rx->tkip_iv16); if (res != TKIP_DECRYPT_OK) diff --git a/net/sctp/output.c b/net/sctp/output.c index b4f3cf06d8da..08b3cead6503 100644 --- a/net/sctp/output.c +++ b/net/sctp/output.c @@ -500,23 +500,20 @@ int sctp_packet_transmit(struct sctp_packet *packet) * Note: Adler-32 is no longer applicable, as has been replaced * by CRC32-C as described in <draft-ietf-tsvwg-sctpcsum-02.txt>. */ - if (!sctp_checksum_disable && - !(dst->dev->features & (NETIF_F_NO_CSUM | NETIF_F_SCTP_CSUM))) { - __u32 crc32 = sctp_start_cksum((__u8 *)sh, cksum_buf_len); + if (!sctp_checksum_disable) { + if (!(dst->dev->features & NETIF_F_SCTP_CSUM)) { + __u32 crc32 = sctp_start_cksum((__u8 *)sh, cksum_buf_len); - /* 3) Put the resultant value into the checksum field in the - * common header, and leave the rest of the bits unchanged. - */ - sh->checksum = sctp_end_cksum(crc32); - } else { - if (dst->dev->features & NETIF_F_SCTP_CSUM) { + /* 3) Put the resultant value into the checksum field in the + * common header, and leave the rest of the bits unchanged. + */ + sh->checksum = sctp_end_cksum(crc32); + } else { /* no need to seed pseudo checksum for SCTP */ nskb->ip_summed = CHECKSUM_PARTIAL; nskb->csum_start = (skb_transport_header(nskb) - nskb->head); nskb->csum_offset = offsetof(struct sctphdr, checksum); - } else { - nskb->ip_summed = CHECKSUM_UNNECESSARY; } } diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c index 1c88c8911dc5..d03682109b7a 100644 --- a/net/sctp/outqueue.c +++ b/net/sctp/outqueue.c @@ -1582,6 +1582,8 @@ static void sctp_check_transmitted(struct sctp_outq *q, #endif /* SCTP_DEBUG */ if (transport) { if (bytes_acked) { + struct sctp_association *asoc = transport->asoc; + /* We may have counted DATA that was migrated * to this transport due to DEL-IP operation. * Subtract those bytes, since the were never @@ -1600,6 +1602,17 @@ static void sctp_check_transmitted(struct sctp_outq *q, transport->error_count = 0; transport->asoc->overall_error_count = 0; + /* + * While in SHUTDOWN PENDING, we may have started + * the T5 shutdown guard timer after reaching the + * retransmission limit. Stop that timer as soon + * as the receiver acknowledged any data. + */ + if (asoc->state == SCTP_STATE_SHUTDOWN_PENDING && + del_timer(&asoc->timers + [SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD])) + sctp_association_put(asoc); + /* Mark the destination transport address as * active if it is not so marked. */ @@ -1629,10 +1642,15 @@ static void sctp_check_transmitted(struct sctp_outq *q, * A sender is doing zero window probing when the * receiver's advertised window is zero, and there is * only one data chunk in flight to the receiver. + * + * Allow the association to timeout while in SHUTDOWN + * PENDING or SHUTDOWN RECEIVED in case the receiver + * stays in zero window mode forever. */ if (!q->asoc->peer.rwnd && !list_empty(&tlist) && - (sack_ctsn+2 == q->asoc->next_tsn)) { + (sack_ctsn+2 == q->asoc->next_tsn) && + q->asoc->state < SCTP_STATE_SHUTDOWN_PENDING) { SCTP_DEBUG_PRINTK("%s: SACK received for zero " "window probe: %u\n", __func__, sack_ctsn); diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c index 534c2e5feb05..6e0f88295aaf 100644 --- a/net/sctp/sm_sideeffect.c +++ b/net/sctp/sm_sideeffect.c @@ -670,10 +670,19 @@ static void sctp_cmd_transport_on(sctp_cmd_seq_t *cmds, /* 8.3 Upon the receipt of the HEARTBEAT ACK, the sender of the * HEARTBEAT should clear the error counter of the destination * transport address to which the HEARTBEAT was sent. - * The association's overall error count is also cleared. */ t->error_count = 0; - t->asoc->overall_error_count = 0; + + /* + * Although RFC4960 specifies that the overall error count must + * be cleared when a HEARTBEAT ACK is received, we make an + * exception while in SHUTDOWN PENDING. If the peer keeps its + * window shut forever, we may never be able to transmit our + * outstanding data and rely on the retransmission limit be reached + * to shutdown the association. + */ + if (t->asoc->state != SCTP_STATE_SHUTDOWN_PENDING) + t->asoc->overall_error_count = 0; /* Clear the hb_sent flag to signal that we had a good * acknowledgement. @@ -1437,6 +1446,13 @@ static int sctp_cmd_interpreter(sctp_event_t event_type, sctp_cmd_setup_t2(commands, asoc, cmd->obj.ptr); break; + case SCTP_CMD_TIMER_START_ONCE: + timer = &asoc->timers[cmd->obj.to]; + + if (timer_pending(timer)) + break; + /* fall through */ + case SCTP_CMD_TIMER_START: timer = &asoc->timers[cmd->obj.to]; timeout = asoc->timeouts[cmd->obj.to]; diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index a297283154d5..246117142b5c 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -5154,7 +5154,7 @@ sctp_disposition_t sctp_sf_do_9_2_start_shutdown( * The sender of the SHUTDOWN MAY also start an overall guard timer * 'T5-shutdown-guard' to bound the overall time for shutdown sequence. */ - sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_START, + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_RESTART, SCTP_TO(SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD)); if (asoc->autoclose) @@ -5299,14 +5299,28 @@ sctp_disposition_t sctp_sf_do_6_3_3_rtx(const struct sctp_endpoint *ep, SCTP_INC_STATS(SCTP_MIB_T3_RTX_EXPIREDS); if (asoc->overall_error_count >= asoc->max_retrans) { - sctp_add_cmd_sf(commands, SCTP_CMD_SET_SK_ERR, - SCTP_ERROR(ETIMEDOUT)); - /* CMD_ASSOC_FAILED calls CMD_DELETE_TCB. */ - sctp_add_cmd_sf(commands, SCTP_CMD_ASSOC_FAILED, - SCTP_PERR(SCTP_ERROR_NO_ERROR)); - SCTP_INC_STATS(SCTP_MIB_ABORTEDS); - SCTP_DEC_STATS(SCTP_MIB_CURRESTAB); - return SCTP_DISPOSITION_DELETE_TCB; + if (asoc->state == SCTP_STATE_SHUTDOWN_PENDING) { + /* + * We are here likely because the receiver had its rwnd + * closed for a while and we have not been able to + * transmit the locally queued data within the maximum + * retransmission attempts limit. Start the T5 + * shutdown guard timer to give the receiver one last + * chance and some additional time to recover before + * aborting. + */ + sctp_add_cmd_sf(commands, SCTP_CMD_TIMER_START_ONCE, + SCTP_TO(SCTP_EVENT_TIMEOUT_T5_SHUTDOWN_GUARD)); + } else { + sctp_add_cmd_sf(commands, SCTP_CMD_SET_SK_ERR, + SCTP_ERROR(ETIMEDOUT)); + /* CMD_ASSOC_FAILED calls CMD_DELETE_TCB. */ + sctp_add_cmd_sf(commands, SCTP_CMD_ASSOC_FAILED, + SCTP_PERR(SCTP_ERROR_NO_ERROR)); + SCTP_INC_STATS(SCTP_MIB_ABORTEDS); + SCTP_DEC_STATS(SCTP_MIB_CURRESTAB); + return SCTP_DISPOSITION_DELETE_TCB; + } } /* E1) For the destination address for which the timer diff --git a/net/sctp/sm_statetable.c b/net/sctp/sm_statetable.c index 0338dc6fdc9d..7c211a7f90f4 100644 --- a/net/sctp/sm_statetable.c +++ b/net/sctp/sm_statetable.c @@ -827,7 +827,7 @@ static const sctp_sm_table_entry_t other_event_table[SCTP_NUM_OTHER_TYPES][SCTP_ /* SCTP_STATE_ESTABLISHED */ \ TYPE_SCTP_FUNC(sctp_sf_timer_ignore), \ /* SCTP_STATE_SHUTDOWN_PENDING */ \ - TYPE_SCTP_FUNC(sctp_sf_timer_ignore), \ + TYPE_SCTP_FUNC(sctp_sf_t5_timer_expire), \ /* SCTP_STATE_SHUTDOWN_SENT */ \ TYPE_SCTP_FUNC(sctp_sf_t5_timer_expire), \ /* SCTP_STATE_SHUTDOWN_RECEIVED */ \ diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 08c6238802de..d3ccf7973c59 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -1384,6 +1384,7 @@ SCTP_STATIC void sctp_close(struct sock *sk, long timeout) struct sctp_endpoint *ep; struct sctp_association *asoc; struct list_head *pos, *temp; + unsigned int data_was_unread; SCTP_DEBUG_PRINTK("sctp_close(sk: 0x%p, timeout:%ld)\n", sk, timeout); @@ -1393,6 +1394,10 @@ SCTP_STATIC void sctp_close(struct sock *sk, long timeout) ep = sctp_sk(sk)->ep; + /* Clean up any skbs sitting on the receive queue. */ + data_was_unread = sctp_queue_purge_ulpevents(&sk->sk_receive_queue); + data_was_unread += sctp_queue_purge_ulpevents(&sctp_sk(sk)->pd_lobby); + /* Walk all associations on an endpoint. */ list_for_each_safe(pos, temp, &ep->asocs) { asoc = list_entry(pos, struct sctp_association, asocs); @@ -1410,7 +1415,9 @@ SCTP_STATIC void sctp_close(struct sock *sk, long timeout) } } - if (sock_flag(sk, SOCK_LINGER) && !sk->sk_lingertime) { + if (data_was_unread || !skb_queue_empty(&asoc->ulpq.lobby) || + !skb_queue_empty(&asoc->ulpq.reasm) || + (sock_flag(sk, SOCK_LINGER) && !sk->sk_lingertime)) { struct sctp_chunk *chunk; chunk = sctp_make_abort_user(asoc, NULL, 0); @@ -1420,10 +1427,6 @@ SCTP_STATIC void sctp_close(struct sock *sk, long timeout) sctp_primitive_SHUTDOWN(asoc, NULL); } - /* Clean up any skbs sitting on the receive queue. */ - sctp_queue_purge_ulpevents(&sk->sk_receive_queue); - sctp_queue_purge_ulpevents(&sctp_sk(sk)->pd_lobby); - /* On a TCP-style socket, block for at most linger_time if set. */ if (sctp_style(sk, TCP) && timeout) sctp_wait_for_close(sk, timeout); diff --git a/net/sctp/ulpevent.c b/net/sctp/ulpevent.c index e70e5fc87890..8a84017834c2 100644 --- a/net/sctp/ulpevent.c +++ b/net/sctp/ulpevent.c @@ -1081,9 +1081,19 @@ void sctp_ulpevent_free(struct sctp_ulpevent *event) } /* Purge the skb lists holding ulpevents. */ -void sctp_queue_purge_ulpevents(struct sk_buff_head *list) +unsigned int sctp_queue_purge_ulpevents(struct sk_buff_head *list) { struct sk_buff *skb; - while ((skb = skb_dequeue(list)) != NULL) - sctp_ulpevent_free(sctp_skb2event(skb)); + unsigned int data_unread = 0; + + while ((skb = skb_dequeue(list)) != NULL) { + struct sctp_ulpevent *event = sctp_skb2event(skb); + + if (!sctp_ulpevent_is_notification(event)) + data_unread += skb->len; + + sctp_ulpevent_free(event); + } + + return data_unread; } diff --git a/net/sunrpc/rpcb_clnt.c b/net/sunrpc/rpcb_clnt.c index 9a80a922c527..e45d2fbbe5a8 100644 --- a/net/sunrpc/rpcb_clnt.c +++ b/net/sunrpc/rpcb_clnt.c @@ -597,7 +597,7 @@ void rpcb_getport_async(struct rpc_task *task) u32 bind_version; struct rpc_xprt *xprt; struct rpc_clnt *rpcb_clnt; - static struct rpcbind_args *map; + struct rpcbind_args *map; struct rpc_task *child; struct sockaddr_storage addr; struct sockaddr *sap = (struct sockaddr *)&addr; diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index a27406b1654f..4814e246a874 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -616,30 +616,25 @@ static void __rpc_execute(struct rpc_task *task) BUG_ON(RPC_IS_QUEUED(task)); for (;;) { + void (*do_action)(struct rpc_task *); /* - * Execute any pending callback. + * Execute any pending callback first. */ - if (task->tk_callback) { - void (*save_callback)(struct rpc_task *); - - /* - * We set tk_callback to NULL before calling it, - * in case it sets the tk_callback field itself: - */ - save_callback = task->tk_callback; - task->tk_callback = NULL; - save_callback(task); - } else { + do_action = task->tk_callback; + task->tk_callback = NULL; + if (do_action == NULL) { /* * Perform the next FSM step. - * tk_action may be NULL when the task has been killed - * by someone else. + * tk_action may be NULL if the task has been killed. + * In particular, note that rpc_killall_tasks may + * do this at any time, so beware when dereferencing. */ - if (task->tk_action == NULL) + do_action = task->tk_action; + if (do_action == NULL) break; - task->tk_action(task); } + do_action(task); /* * Lockless check for whether task is sleeping or not. diff --git a/net/wireless/core.c b/net/wireless/core.c index c22ef3492ee6..880dbe2e6f94 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -366,6 +366,7 @@ struct wiphy *wiphy_new(const struct cfg80211_ops *ops, int sizeof_priv) mutex_init(&rdev->mtx); mutex_init(&rdev->devlist_mtx); + mutex_init(&rdev->sched_scan_mtx); INIT_LIST_HEAD(&rdev->netdev_list); spin_lock_init(&rdev->bss_lock); INIT_LIST_HEAD(&rdev->bss_list); @@ -701,6 +702,7 @@ void cfg80211_dev_free(struct cfg80211_registered_device *rdev) rfkill_destroy(rdev->rfkill); mutex_destroy(&rdev->mtx); mutex_destroy(&rdev->devlist_mtx); + mutex_destroy(&rdev->sched_scan_mtx); list_for_each_entry_safe(scan, tmp, &rdev->bss_list, list) cfg80211_put_bss(&scan->pub); cfg80211_rdev_free_wowlan(rdev); @@ -737,12 +739,16 @@ static void wdev_cleanup_work(struct work_struct *work) ___cfg80211_scan_done(rdev, true); } + cfg80211_unlock_rdev(rdev); + + mutex_lock(&rdev->sched_scan_mtx); + if (WARN_ON(rdev->sched_scan_req && rdev->sched_scan_req->dev == wdev->netdev)) { __cfg80211_stop_sched_scan(rdev, false); } - cfg80211_unlock_rdev(rdev); + mutex_unlock(&rdev->sched_scan_mtx); mutex_lock(&rdev->devlist_mtx); rdev->opencount--; @@ -830,9 +836,9 @@ static int cfg80211_netdev_notifier_call(struct notifier_block * nb, break; case NL80211_IFTYPE_P2P_CLIENT: case NL80211_IFTYPE_STATION: - cfg80211_lock_rdev(rdev); + mutex_lock(&rdev->sched_scan_mtx); __cfg80211_stop_sched_scan(rdev, false); - cfg80211_unlock_rdev(rdev); + mutex_unlock(&rdev->sched_scan_mtx); wdev_lock(wdev); #ifdef CONFIG_CFG80211_WEXT diff --git a/net/wireless/core.h b/net/wireless/core.h index 3dce1f167eba..a570ff9214ec 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -65,6 +65,8 @@ struct cfg80211_registered_device { struct work_struct scan_done_wk; struct work_struct sched_scan_results_wk; + struct mutex sched_scan_mtx; + #ifdef CONFIG_NL80211_TESTMODE struct genl_info *testmode_info; #endif diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index f07602d7bf68..cea338150d05 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -3461,9 +3461,6 @@ static int nl80211_start_sched_scan(struct sk_buff *skb, if (!is_valid_ie_attr(info->attrs[NL80211_ATTR_IE])) return -EINVAL; - if (rdev->sched_scan_req) - return -EINPROGRESS; - if (!info->attrs[NL80211_ATTR_SCHED_SCAN_INTERVAL]) return -EINVAL; @@ -3502,12 +3499,21 @@ static int nl80211_start_sched_scan(struct sk_buff *skb, if (ie_len > wiphy->max_scan_ie_len) return -EINVAL; + mutex_lock(&rdev->sched_scan_mtx); + + if (rdev->sched_scan_req) { + err = -EINPROGRESS; + goto out; + } + request = kzalloc(sizeof(*request) + sizeof(*request->ssids) * n_ssids + sizeof(*request->channels) * n_channels + ie_len, GFP_KERNEL); - if (!request) - return -ENOMEM; + if (!request) { + err = -ENOMEM; + goto out; + } if (n_ssids) request->ssids = (void *)&request->channels[n_channels]; @@ -3605,6 +3611,7 @@ static int nl80211_start_sched_scan(struct sk_buff *skb, out_free: kfree(request); out: + mutex_unlock(&rdev->sched_scan_mtx); return err; } @@ -3612,12 +3619,17 @@ static int nl80211_stop_sched_scan(struct sk_buff *skb, struct genl_info *info) { struct cfg80211_registered_device *rdev = info->user_ptr[0]; + int err; if (!(rdev->wiphy.flags & WIPHY_FLAG_SUPPORTS_SCHED_SCAN) || !rdev->ops->sched_scan_stop) return -EOPNOTSUPP; - return __cfg80211_stop_sched_scan(rdev, false); + mutex_lock(&rdev->sched_scan_mtx); + err = __cfg80211_stop_sched_scan(rdev, false); + mutex_unlock(&rdev->sched_scan_mtx); + + return err; } static int nl80211_send_bss(struct sk_buff *msg, u32 pid, u32 seq, int flags, diff --git a/net/wireless/scan.c b/net/wireless/scan.c index 7a6c67667d70..ae0c2256ba3b 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -100,14 +100,14 @@ void __cfg80211_sched_scan_results(struct work_struct *wk) rdev = container_of(wk, struct cfg80211_registered_device, sched_scan_results_wk); - cfg80211_lock_rdev(rdev); + mutex_lock(&rdev->sched_scan_mtx); /* we don't have sched_scan_req anymore if the scan is stopping */ if (rdev->sched_scan_req) nl80211_send_sched_scan_results(rdev, rdev->sched_scan_req->dev); - cfg80211_unlock_rdev(rdev); + mutex_unlock(&rdev->sched_scan_mtx); } void cfg80211_sched_scan_results(struct wiphy *wiphy) @@ -123,9 +123,9 @@ void cfg80211_sched_scan_stopped(struct wiphy *wiphy) { struct cfg80211_registered_device *rdev = wiphy_to_dev(wiphy); - cfg80211_lock_rdev(rdev); + mutex_lock(&rdev->sched_scan_mtx); __cfg80211_stop_sched_scan(rdev, true); - cfg80211_unlock_rdev(rdev); + mutex_unlock(&rdev->sched_scan_mtx); } EXPORT_SYMBOL(cfg80211_sched_scan_stopped); @@ -135,7 +135,7 @@ int __cfg80211_stop_sched_scan(struct cfg80211_registered_device *rdev, int err; struct net_device *dev; - ASSERT_RDEV_LOCK(rdev); + lockdep_assert_held(&rdev->sched_scan_mtx); if (!rdev->sched_scan_req) return 0; diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index d70f85eb7864..9414b9c5b1e4 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -1345,6 +1345,8 @@ out: xfrm_state_check_expire(x1); err = 0; + x->km.state = XFRM_STATE_DEAD; + __xfrm_state_put(x); } spin_unlock_bh(&x1->lock); diff --git a/scripts/depmod.sh b/scripts/depmod.sh index 3b029cba2baf..a27235685949 100755 --- a/scripts/depmod.sh +++ b/scripts/depmod.sh @@ -21,13 +21,15 @@ fi # older versions of depmod require the version string to start with three # numbers, so we cheat with a symlink here depmod_hack_needed=true -mkdir -p .tmp_depmod/lib/modules/$KERNELRELEASE -if "$DEPMOD" -b .tmp_depmod $KERNELRELEASE 2>/dev/null; then - if test -e .tmp_depmod/lib/modules/$KERNELRELEASE/modules.dep -o \ - -e .tmp_depmod/lib/modules/$KERNELRELEASE/modules.dep.bin; then +tmp_dir=$(mktemp -d ${TMPDIR:-/tmp}/depmod.XXXXXX) +mkdir -p "$tmp_dir/lib/modules/$KERNELRELEASE" +if "$DEPMOD" -b "$tmp_dir" $KERNELRELEASE 2>/dev/null; then + if test -e "$tmp_dir/lib/modules/$KERNELRELEASE/modules.dep" -o \ + -e "$tmp_dir/lib/modules/$KERNELRELEASE/modules.dep.bin"; then depmod_hack_needed=false fi fi +rm -rf "$tmp_dir" if $depmod_hack_needed; then symlink="$INSTALL_MOD_PATH/lib/modules/99.98.$KERNELRELEASE" ln -s "$KERNELRELEASE" "$symlink" diff --git a/sound/soc/codecs/wm8994.c b/sound/soc/codecs/wm8994.c index c2fc0356c2a4..83014a7c2e14 100644 --- a/sound/soc/codecs/wm8994.c +++ b/sound/soc/codecs/wm8994.c @@ -1190,7 +1190,6 @@ SND_SOC_DAPM_INPUT("DMIC1DAT"), SND_SOC_DAPM_INPUT("DMIC2DAT"), SND_SOC_DAPM_INPUT("Clock"), -SND_SOC_DAPM_MICBIAS("MICBIAS", WM8994_MICBIAS, 2, 0), SND_SOC_DAPM_SUPPLY_S("MICBIAS Supply", 1, SND_SOC_NOPM, 0, 0, micbias_ev, SND_SOC_DAPM_PRE_PMU), @@ -1509,8 +1508,10 @@ static const struct snd_soc_dapm_route wm8994_revd_intercon[] = { { "AIF2DACDAT", NULL, "AIF1DACDAT" }, { "AIF1ADCDAT", NULL, "AIF2ADCDAT" }, { "AIF2ADCDAT", NULL, "AIF1ADCDAT" }, - { "MICBIAS", NULL, "CLK_SYS" }, - { "MICBIAS", NULL, "MICBIAS Supply" }, + { "MICBIAS1", NULL, "CLK_SYS" }, + { "MICBIAS1", NULL, "MICBIAS Supply" }, + { "MICBIAS2", NULL, "CLK_SYS" }, + { "MICBIAS2", NULL, "MICBIAS Supply" }, }; static const struct snd_soc_dapm_route wm8994_intercon[] = { @@ -2763,7 +2764,7 @@ static void wm8958_default_micdet(u16 status, void *data) report = SND_JACK_MICROPHONE; /* Everything else is buttons; just assign slots */ - if (status & 0x1c0) + if (status & 0x1c) report |= SND_JACK_BTN_0; done: diff --git a/sound/soc/sh/fsi-ak4642.c b/sound/soc/sh/fsi-ak4642.c index d6f4703b3c07..770a71a15366 100644 --- a/sound/soc/sh/fsi-ak4642.c +++ b/sound/soc/sh/fsi-ak4642.c @@ -97,7 +97,7 @@ static int fsi_ak4642_remove(struct platform_device *pdev) static struct fsi_ak4642_data fsi_a_ak4642 = { .name = "AK4642", - .card = "FSIA (AK4642)", + .card = "FSIA-AK4642", .cpu_dai = "fsia-dai", .codec = "ak4642-codec.0-0012", .platform = "sh_fsi.0", @@ -106,7 +106,7 @@ static struct fsi_ak4642_data fsi_a_ak4642 = { static struct fsi_ak4642_data fsi_b_ak4642 = { .name = "AK4642", - .card = "FSIB (AK4642)", + .card = "FSIB-AK4642", .cpu_dai = "fsib-dai", .codec = "ak4642-codec.0-0012", .platform = "sh_fsi.0", @@ -115,7 +115,7 @@ static struct fsi_ak4642_data fsi_b_ak4642 = { static struct fsi_ak4642_data fsi_a_ak4643 = { .name = "AK4643", - .card = "FSIA (AK4643)", + .card = "FSIA-AK4643", .cpu_dai = "fsia-dai", .codec = "ak4642-codec.0-0013", .platform = "sh_fsi.0", @@ -124,7 +124,7 @@ static struct fsi_ak4642_data fsi_a_ak4643 = { static struct fsi_ak4642_data fsi_b_ak4643 = { .name = "AK4643", - .card = "FSIB (AK4643)", + .card = "FSIB-AK4643", .cpu_dai = "fsib-dai", .codec = "ak4642-codec.0-0013", .platform = "sh_fsi.0", @@ -133,7 +133,7 @@ static struct fsi_ak4642_data fsi_b_ak4643 = { static struct fsi_ak4642_data fsi2_a_ak4642 = { .name = "AK4642", - .card = "FSI2A (AK4642)", + .card = "FSI2A-AK4642", .cpu_dai = "fsia-dai", .codec = "ak4642-codec.0-0012", .platform = "sh_fsi2", @@ -142,7 +142,7 @@ static struct fsi_ak4642_data fsi2_a_ak4642 = { static struct fsi_ak4642_data fsi2_b_ak4642 = { .name = "AK4642", - .card = "FSI2B (AK4642)", + .card = "FSI2B-AK4642", .cpu_dai = "fsib-dai", .codec = "ak4642-codec.0-0012", .platform = "sh_fsi2", @@ -151,7 +151,7 @@ static struct fsi_ak4642_data fsi2_b_ak4642 = { static struct fsi_ak4642_data fsi2_a_ak4643 = { .name = "AK4643", - .card = "FSI2A (AK4643)", + .card = "FSI2A-AK4643", .cpu_dai = "fsia-dai", .codec = "ak4642-codec.0-0013", .platform = "sh_fsi2", @@ -160,7 +160,7 @@ static struct fsi_ak4642_data fsi2_a_ak4643 = { static struct fsi_ak4642_data fsi2_b_ak4643 = { .name = "AK4643", - .card = "FSI2B (AK4643)", + .card = "FSI2B-AK4643", .cpu_dai = "fsib-dai", .codec = "ak4642-codec.0-0013", .platform = "sh_fsi2", diff --git a/sound/soc/sh/fsi-da7210.c b/sound/soc/sh/fsi-da7210.c index dbafd7ac5590..59553fd8c2fb 100644 --- a/sound/soc/sh/fsi-da7210.c +++ b/sound/soc/sh/fsi-da7210.c @@ -42,7 +42,7 @@ static struct snd_soc_dai_link fsi_da7210_dai = { }; static struct snd_soc_card fsi_soc_card = { - .name = "FSI (DA7210)", + .name = "FSI-DA7210", .dai_link = &fsi_da7210_dai, .num_links = 1, }; diff --git a/sound/soc/sh/fsi-hdmi.c b/sound/soc/sh/fsi-hdmi.c index 9719985eb82d..d3d9fd880680 100644 --- a/sound/soc/sh/fsi-hdmi.c +++ b/sound/soc/sh/fsi-hdmi.c @@ -83,13 +83,13 @@ static int fsi_hdmi_remove(struct platform_device *pdev) static struct fsi_hdmi_data fsi2_a_hdmi = { .cpu_dai = "fsia-dai", - .card = "FSI2A (SH MOBILE HDMI)", + .card = "FSI2A-HDMI", .id = FSI_PORT_A, }; static struct fsi_hdmi_data fsi2_b_hdmi = { .cpu_dai = "fsib-dai", - .card = "FSI2B (SH MOBILE HDMI)", + .card = "FSI2B-HDMI", .id = FSI_PORT_B, }; |