From 23a75eee070f1370bee803a34f285cf81eb5f331 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Szilveszter=20=C3=96rd=C3=B6g?= Date: Fri, 6 Aug 2010 09:26:38 +0800 Subject: crypto: hash - Fix handling of small unaligned buffers MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If a scatterwalk chain contains an entry with an unaligned offset then hash_walk_next() will cut off the next step at the next alignment point. However, if the entry ends before the next alignment point then we a loop, which leads to a kernel oops. Fix this by checking whether the next aligment point is before the end of the current entry. Signed-off-by: Szilveszter Ördög Acked-by: David S. Miller Signed-off-by: Herbert Xu diff --git a/crypto/ahash.c b/crypto/ahash.c index b8c59b8..f669822 100644 --- a/crypto/ahash.c +++ b/crypto/ahash.c @@ -47,8 +47,11 @@ static int hash_walk_next(struct crypto_hash_walk *walk) walk->data = crypto_kmap(walk->pg, 0); walk->data += offset; - if (offset & alignmask) - nbytes = alignmask + 1 - (offset & alignmask); + if (offset & alignmask) { + unsigned int unaligned = alignmask + 1 - (offset & alignmask); + if (nbytes > unaligned) + nbytes = unaligned; + } walk->entrylen -= nbytes; return nbytes; -- cgit v0.10.2 From 326a6346ffb5b19eb593530d9d3096d409e46f62 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Fri, 6 Aug 2010 09:40:28 +0800 Subject: crypto: testmgr - Fix test disabling option This patch fixes a serious bug in the test disabling patch where it can cause an spurious load of the cryptomgr module even when it's compiled in. It also negates the test disabling option so that its absence causes tests to be enabled. The Kconfig option is also now behind EMBEDDED. Signed-off-by: Herbert Xu diff --git a/crypto/Kconfig b/crypto/Kconfig index 1cd497d..6f5c50f 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -101,13 +101,12 @@ config CRYPTO_MANAGER2 select CRYPTO_BLKCIPHER2 select CRYPTO_PCOMP2 -config CRYPTO_MANAGER_TESTS - bool "Run algolithms' self-tests" - default y - depends on CRYPTO_MANAGER2 +config CRYPTO_MANAGER_DISABLE_TESTS + bool "Disable run-time self tests" + depends on CRYPTO_MANAGER2 && EMBEDDED help - Run cryptomanager's tests for the new crypto algorithms being - registered. + Disable run-time self tests that normally take place at + algorithm registration. config CRYPTO_GF128MUL tristate "GF(2^128) multiplication functions (EXPERIMENTAL)" diff --git a/crypto/algboss.c b/crypto/algboss.c index 40bd391..791d194 100644 --- a/crypto/algboss.c +++ b/crypto/algboss.c @@ -206,13 +206,16 @@ err: return NOTIFY_OK; } -#ifdef CONFIG_CRYPTO_MANAGER_TESTS static int cryptomgr_test(void *data) { struct crypto_test_param *param = data; u32 type = param->type; int err = 0; +#ifdef CONFIG_CRYPTO_MANAGER_DISABLE_TESTS + goto skiptest; +#endif + if (type & CRYPTO_ALG_TESTED) goto skiptest; @@ -267,7 +270,6 @@ err_put_module: err: return NOTIFY_OK; } -#endif /* CONFIG_CRYPTO_MANAGER_TESTS */ static int cryptomgr_notify(struct notifier_block *this, unsigned long msg, void *data) @@ -275,10 +277,8 @@ static int cryptomgr_notify(struct notifier_block *this, unsigned long msg, switch (msg) { case CRYPTO_MSG_ALG_REQUEST: return cryptomgr_schedule_probe(data); -#ifdef CONFIG_CRYPTO_MANAGER_TESTS case CRYPTO_MSG_ALG_REGISTER: return cryptomgr_schedule_test(data); -#endif } return NOTIFY_DONE; diff --git a/crypto/testmgr.c b/crypto/testmgr.c index abd980c..fa8c8f7 100644 --- a/crypto/testmgr.c +++ b/crypto/testmgr.c @@ -23,7 +23,7 @@ #include "internal.h" -#ifndef CONFIG_CRYPTO_MANAGER_TESTS +#ifdef CONFIG_CRYPTO_MANAGER_DISABLE_TESTS /* a perfect nop */ int alg_test(const char *driver, const char *alg, u32 type, u32 mask) @@ -2542,6 +2542,6 @@ non_fips_alg: return -EINVAL; } -#endif /* CONFIG_CRYPTO_MANAGER_TESTS */ +#endif /* CONFIG_CRYPTO_MANAGER_DISABLE_TESTS */ EXPORT_SYMBOL_GPL(alg_test); -- cgit v0.10.2 From 00ca28a507b215dcd121735f16764ea4173c4ff9 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Fri, 6 Aug 2010 10:34:00 +0800 Subject: crypto: testmgr - Default to no tests On Thu, Aug 05, 2010 at 07:01:21PM -0700, Linus Torvalds wrote: > On Thu, Aug 5, 2010 at 6:40 PM, Herbert Xu wrote: > > > > -config CRYPTO_MANAGER_TESTS > > - bool "Run algolithms' self-tests" > > - default y > > - depends on CRYPTO_MANAGER2 > > +config CRYPTO_MANAGER_DISABLE_TESTS > > + bool "Disable run-time self tests" > > + depends on CRYPTO_MANAGER2 && EMBEDDED > > Why do you still want to force-enable those tests? I was going to > complain about the "default y" anyway, now I'm _really_ complaining, > because you've now made it impossible to disable those tests. Why? As requested, this patch sets the default to y and removes the EMBEDDED dependency. Signed-off-by: Herbert Xu diff --git a/crypto/Kconfig b/crypto/Kconfig index 6f5c50f..e573077 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -103,7 +103,8 @@ config CRYPTO_MANAGER2 config CRYPTO_MANAGER_DISABLE_TESTS bool "Disable run-time self tests" - depends on CRYPTO_MANAGER2 && EMBEDDED + default y + depends on CRYPTO_MANAGER2 help Disable run-time self tests that normally take place at algorithm registration. -- cgit v0.10.2 From 18cb2aef91b37dbce2bec2f39bb1dddd0e9dd838 Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Sat, 7 Aug 2010 03:26:23 +0900 Subject: percpu: handle __percpu notations in UP accessors UP accessors didn't take care of __percpu notations leading to a lot of spurious sparse warnings on UP configurations. Fix it. Signed-off-by: Namhyung Kim Signed-off-by: Tejun Heo diff --git a/include/asm-generic/percpu.h b/include/asm-generic/percpu.h index b5043a9..08923b6 100644 --- a/include/asm-generic/percpu.h +++ b/include/asm-generic/percpu.h @@ -70,11 +70,16 @@ extern void setup_per_cpu_areas(void); #else /* ! SMP */ -#define per_cpu(var, cpu) (*((void)(cpu), &(var))) -#define __get_cpu_var(var) (var) -#define __raw_get_cpu_var(var) (var) -#define this_cpu_ptr(ptr) per_cpu_ptr(ptr, 0) -#define __this_cpu_ptr(ptr) this_cpu_ptr(ptr) +#define VERIFY_PERCPU_PTR(__p) ({ \ + __verify_pcpu_ptr((__p)); \ + (typeof(*(__p)) __kernel __force *)(__p); \ +}) + +#define per_cpu(var, cpu) (*((void)(cpu), VERIFY_PERCPU_PTR(&(var)))) +#define __get_cpu_var(var) (*VERIFY_PERCPU_PTR(&(var))) +#define __raw_get_cpu_var(var) (*VERIFY_PERCPU_PTR(&(var))) +#define this_cpu_ptr(ptr) per_cpu_ptr(ptr, 0) +#define __this_cpu_ptr(ptr) this_cpu_ptr(ptr) #endif /* SMP */ diff --git a/include/linux/percpu.h b/include/linux/percpu.h index b8b9084..49466b1 100644 --- a/include/linux/percpu.h +++ b/include/linux/percpu.h @@ -149,7 +149,7 @@ extern void __init percpu_init_late(void); #else /* CONFIG_SMP */ -#define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); (ptr); }) +#define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR((ptr)); }) /* can't distinguish from other static vars, always false */ static inline bool is_kernel_percpu_address(unsigned long addr) -- cgit v0.10.2 From 94cb121c9483f1ec9b1ef0c249fbfc49c628fa6b Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Sat, 7 Aug 2010 03:26:24 +0900 Subject: percpu: add __percpu notations to UP allocator Add __percpu notations to UP percpu allocator. Signed-off-by: Namhyung Kim Signed-off-by: Tejun Heo diff --git a/mm/percpu_up.c b/mm/percpu_up.c index c4351c7..db884fa 100644 --- a/mm/percpu_up.c +++ b/mm/percpu_up.c @@ -14,13 +14,13 @@ void __percpu *__alloc_percpu(size_t size, size_t align) * percpu sections on SMP for which this path isn't used. */ WARN_ON_ONCE(align > SMP_CACHE_BYTES); - return kzalloc(size, GFP_KERNEL); + return (void __percpu __force *)kzalloc(size, GFP_KERNEL); } EXPORT_SYMBOL_GPL(__alloc_percpu); void free_percpu(void __percpu *p) { - kfree(p); + kfree(this_cpu_ptr(p)); } EXPORT_SYMBOL_GPL(free_percpu); -- cgit v0.10.2 From 50b6e71ae83714be509b80727dbf90fa8b1c0717 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Mon, 9 Aug 2010 10:29:28 -0400 Subject: Revert "hwrng: n2-drv - remove casts from void*" This reverts commit 8b9cfdca9c52f7d39c3ccfac1668e31c20c9f42e. This patch needs to wait for the HWRNG API to start using void * for priv first. Signed-off-by: Herbert Xu diff --git a/drivers/char/hw_random/n2-drv.c b/drivers/char/hw_random/n2-drv.c index 7a4f080..d8a4ca8 100644 --- a/drivers/char/hw_random/n2-drv.c +++ b/drivers/char/hw_random/n2-drv.c @@ -387,7 +387,7 @@ static int n2rng_init_control(struct n2rng *np) static int n2rng_data_read(struct hwrng *rng, u32 *data) { - struct n2rng *np = rng->priv; + struct n2rng *np = (struct n2rng *) rng->priv; unsigned long ra = __pa(&np->test_data); int len; -- cgit v0.10.2 From 92298e668372f2f6c8a79fb272f13d65161a4876 Mon Sep 17 00:00:00 2001 From: Dave Airlie Date: Fri, 13 Aug 2010 10:22:17 +1000 Subject: PCI: provide stub pci_domain_nr function for !CONFIG_PCI configs Allows the new PCI domain aware DRM code to compile on m68k. Reported-by: Geert Uytterhoeven Signed-off-by: Dave Airlie Signed-off-by: Jesse Barnes diff --git a/include/linux/pci.h b/include/linux/pci.h index b1d1795..c8d95e3 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -1214,6 +1214,9 @@ static inline struct pci_dev *pci_get_bus_and_slot(unsigned int bus, unsigned int devfn) { return NULL; } +static inline int pci_domain_nr(struct pci_bus *bus) +{ return 0; } + #define dev_is_pci(d) (false) #define dev_is_pf(d) (false) #define dev_num_vf(d) (0) -- cgit v0.10.2 From f14c4f141ad88290bb23d597a0a47833e7891fcd Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Tue, 27 Jul 2010 08:14:35 +0000 Subject: ARM: mach-shmobile: Runtime PM V3 This is the SH-Mobile ARM Runtime PM implementation V3. Only clocks are managed at this point. If Runtime PM is disabled then the clocks will be enabled before ->probe() and disabled after ->remove(). Signed-off-by: Magnus Damm Signed-off-by: Paul Mundt diff --git a/arch/arm/mach-shmobile/Makefile b/arch/arm/mach-shmobile/Makefile index 5e16b4c..ae416fe 100644 --- a/arch/arm/mach-shmobile/Makefile +++ b/arch/arm/mach-shmobile/Makefile @@ -3,7 +3,7 @@ # # Common objects -obj-y := timer.o console.o clock.o +obj-y := timer.o console.o clock.o pm_runtime.o # CPU objects obj-$(CONFIG_ARCH_SH7367) += setup-sh7367.o clock-sh7367.o intc-sh7367.o diff --git a/arch/arm/mach-shmobile/clock-sh7372.c b/arch/arm/mach-shmobile/clock-sh7372.c index fb4e9b1..4a1d019 100644 --- a/arch/arm/mach-shmobile/clock-sh7372.c +++ b/arch/arm/mach-shmobile/clock-sh7372.c @@ -428,7 +428,7 @@ static struct clk mstp_clks[MSTP_NR] = { [MSTP201] = MSTP(&div6_clks[DIV6_SUB], SMSTPCR2, 1, 0), /* SCIFA3 */ [MSTP200] = MSTP(&div6_clks[DIV6_SUB], SMSTPCR2, 0, 0), /* SCIFA4 */ [MSTP329] = MSTP(&r_clk, SMSTPCR3, 29, 0), /* CMT10 */ - [MSTP328] = MSTP(&div6_clks[DIV6_SPU], SMSTPCR3, 28, CLK_ENABLE_ON_INIT), /* FSIA */ + [MSTP328] = MSTP(&div6_clks[DIV6_SPU], SMSTPCR3, 28, 0), /* FSIA */ [MSTP323] = MSTP(&div6_clks[DIV6_SUB], SMSTPCR3, 23, 0), /* IIC1 */ [MSTP322] = MSTP(&div6_clks[DIV6_SUB], SMSTPCR3, 22, 0), /* USB0 */ [MSTP314] = MSTP(&div4_clks[DIV4_HP], SMSTPCR3, 14, 0), /* SDHI0 */ diff --git a/arch/arm/mach-shmobile/pm_runtime.c b/arch/arm/mach-shmobile/pm_runtime.c new file mode 100644 index 0000000..94912d3 --- /dev/null +++ b/arch/arm/mach-shmobile/pm_runtime.c @@ -0,0 +1,169 @@ +/* + * arch/arm/mach-shmobile/pm_runtime.c + * + * Runtime PM support code for SuperH Mobile ARM + * + * Copyright (C) 2009-2010 Magnus Damm + * + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + */ + +#include +#include +#include +#include +#include +#include +#include +#include + +#ifdef CONFIG_PM_RUNTIME +#define BIT_ONCE 0 +#define BIT_ACTIVE 1 +#define BIT_CLK_ENABLED 2 + +struct pm_runtime_data { + unsigned long flags; + struct clk *clk; +}; + +static void __devres_release(struct device *dev, void *res) +{ + struct pm_runtime_data *prd = res; + + dev_dbg(dev, "__devres_release()\n"); + + if (test_bit(BIT_CLK_ENABLED, &prd->flags)) + clk_disable(prd->clk); + + if (test_bit(BIT_ACTIVE, &prd->flags)) + clk_put(prd->clk); +} + +static struct pm_runtime_data *__to_prd(struct device *dev) +{ + return devres_find(dev, __devres_release, NULL, NULL); +} + +static void platform_pm_runtime_init(struct device *dev, + struct pm_runtime_data *prd) +{ + if (prd && !test_and_set_bit(BIT_ONCE, &prd->flags)) { + prd->clk = clk_get(dev, NULL); + if (!IS_ERR(prd->clk)) { + set_bit(BIT_ACTIVE, &prd->flags); + dev_info(dev, "clocks managed by runtime pm\n"); + } + } +} + +static void platform_pm_runtime_bug(struct device *dev, + struct pm_runtime_data *prd) +{ + if (prd && !test_and_set_bit(BIT_ONCE, &prd->flags)) + dev_err(dev, "runtime pm suspend before resume\n"); +} + +int platform_pm_runtime_suspend(struct device *dev) +{ + struct pm_runtime_data *prd = __to_prd(dev); + + dev_dbg(dev, "platform_pm_runtime_suspend()\n"); + + platform_pm_runtime_bug(dev, prd); + + if (prd && test_bit(BIT_ACTIVE, &prd->flags)) { + clk_disable(prd->clk); + clear_bit(BIT_CLK_ENABLED, &prd->flags); + } + + return 0; +} + +int platform_pm_runtime_resume(struct device *dev) +{ + struct pm_runtime_data *prd = __to_prd(dev); + + dev_dbg(dev, "platform_pm_runtime_resume()\n"); + + platform_pm_runtime_init(dev, prd); + + if (prd && test_bit(BIT_ACTIVE, &prd->flags)) { + clk_enable(prd->clk); + set_bit(BIT_CLK_ENABLED, &prd->flags); + } + + return 0; +} + +int platform_pm_runtime_idle(struct device *dev) +{ + /* suspend synchronously to disable clocks immediately */ + return pm_runtime_suspend(dev); +} + +static int platform_bus_notify(struct notifier_block *nb, + unsigned long action, void *data) +{ + struct device *dev = data; + struct pm_runtime_data *prd; + + dev_dbg(dev, "platform_bus_notify() %ld !\n", action); + + if (action == BUS_NOTIFY_BIND_DRIVER) { + prd = devres_alloc(__devres_release, sizeof(*prd), GFP_KERNEL); + if (prd) + devres_add(dev, prd); + else + dev_err(dev, "unable to alloc memory for runtime pm\n"); + } + + return 0; +} + +#else /* CONFIG_PM_RUNTIME */ + +static int platform_bus_notify(struct notifier_block *nb, + unsigned long action, void *data) +{ + struct device *dev = data; + struct clk *clk; + + dev_dbg(dev, "platform_bus_notify() %ld !\n", action); + + switch (action) { + case BUS_NOTIFY_BIND_DRIVER: + clk = clk_get(dev, NULL); + if (!IS_ERR(clk)) { + clk_enable(clk); + clk_put(clk); + dev_info(dev, "runtime pm disabled, clock forced on\n"); + } + break; + case BUS_NOTIFY_UNBOUND_DRIVER: + clk = clk_get(dev, NULL); + if (!IS_ERR(clk)) { + clk_disable(clk); + clk_put(clk); + dev_info(dev, "runtime pm disabled, clock forced off\n"); + } + break; + } + + return 0; +} + +#endif /* CONFIG_PM_RUNTIME */ + +static struct notifier_block platform_bus_notifier = { + .notifier_call = platform_bus_notify +}; + +static int __init sh_pm_runtime_init(void) +{ + bus_register_notifier(&platform_bus_type, &platform_bus_notifier); + return 0; +} +core_initcall(sh_pm_runtime_init); -- cgit v0.10.2 From a4909b52682c1e2ac858754c8bc46f2501e67dee Mon Sep 17 00:00:00 2001 From: Guennadi Liakhovetski Date: Mon, 26 Jul 2010 16:20:53 +0000 Subject: ARM: mach-shmobile: add clock definitions for CEU and CSI2 Two more clocks to be managed by the runtime PM. Signed-off-by: Guennadi Liakhovetski Signed-off-by: Paul Mundt diff --git a/arch/arm/mach-shmobile/clock-sh7372.c b/arch/arm/mach-shmobile/clock-sh7372.c index 4a1d019..dcaac1b 100644 --- a/arch/arm/mach-shmobile/clock-sh7372.c +++ b/arch/arm/mach-shmobile/clock-sh7372.c @@ -395,7 +395,7 @@ static struct clk div6_reparent_clks[DIV6_REPARENT_NR] = { enum { MSTP001, MSTP131, MSTP130, - MSTP129, MSTP128, + MSTP129, MSTP128, MSTP127, MSTP126, MSTP118, MSTP117, MSTP116, MSTP106, MSTP101, MSTP100, MSTP223, @@ -413,6 +413,8 @@ static struct clk mstp_clks[MSTP_NR] = { [MSTP130] = MSTP(&div4_clks[DIV4_B], SMSTPCR1, 30, 0), /* VEU2 */ [MSTP129] = MSTP(&div4_clks[DIV4_B], SMSTPCR1, 29, 0), /* VEU1 */ [MSTP128] = MSTP(&div4_clks[DIV4_B], SMSTPCR1, 28, 0), /* VEU0 */ + [MSTP127] = MSTP(&div4_clks[DIV4_B], SMSTPCR1, 27, 0), /* CEU */ + [MSTP126] = MSTP(&div4_clks[DIV4_B], SMSTPCR1, 26, 0), /* CSI2 */ [MSTP118] = MSTP(&div4_clks[DIV4_B], SMSTPCR1, 18, 0), /* DSITX */ [MSTP117] = MSTP(&div4_clks[DIV4_B], SMSTPCR1, 17, 0), /* LCDC1 */ [MSTP116] = MSTP(&div6_clks[DIV6_SUB], SMSTPCR1, 16, 0), /* IIC0 */ @@ -498,6 +500,8 @@ static struct clk_lookup lookups[] = { CLKDEV_DEV_ID("uio_pdrv_genirq.3", &mstp_clks[MSTP130]), /* VEU2 */ CLKDEV_DEV_ID("uio_pdrv_genirq.2", &mstp_clks[MSTP129]), /* VEU1 */ CLKDEV_DEV_ID("uio_pdrv_genirq.1", &mstp_clks[MSTP128]), /* VEU0 */ + CLKDEV_DEV_ID("sh_mobile_ceu.0", &mstp_clks[MSTP127]), /* CEU */ + CLKDEV_DEV_ID("sh-mobile-csi2.0", &mstp_clks[MSTP126]), /* CSI2 */ CLKDEV_DEV_ID("sh-mipi-dsi.0", &mstp_clks[MSTP118]), /* DSITX */ CLKDEV_DEV_ID("sh_mobile_lcdc_fb.1", &mstp_clks[MSTP117]), /* LCDC1 */ CLKDEV_DEV_ID("i2c-sh_mobile.0", &mstp_clks[MSTP116]), /* IIC0 */ -- cgit v0.10.2 From 2863e935a59b36d55522af8620112c64a56b4022 Mon Sep 17 00:00:00 2001 From: Arnd Hannemann Date: Sat, 14 Aug 2010 19:45:21 +0000 Subject: ARM: mach-shmobile: AP4EVB: use gpio-leds for leds. This allows control of the leds on the AP4EVB board by kernel events (triggers) like mmc0/mmc1. In userland, the leds won't be available anymore over /sys/class/gpio but via /sys/class/leds. Signed-off-by: Arnd Hannemann Signed-off-by: Paul Mundt diff --git a/arch/arm/mach-shmobile/board-ap4evb.c b/arch/arm/mach-shmobile/board-ap4evb.c index 23d472f..98c1cf7 100644 --- a/arch/arm/mach-shmobile/board-ap4evb.c +++ b/arch/arm/mach-shmobile/board-ap4evb.c @@ -39,6 +39,7 @@ #include #include #include +#include #include #include @@ -650,7 +651,44 @@ static struct platform_device hdmi_device = { }, }; +static struct gpio_led ap4evb_leds[] = { + { + .name = "led4", + .gpio = GPIO_PORT185, + .default_state = LEDS_GPIO_DEFSTATE_ON, + }, + { + .name = "led2", + .gpio = GPIO_PORT186, + .default_state = LEDS_GPIO_DEFSTATE_ON, + }, + { + .name = "led3", + .gpio = GPIO_PORT187, + .default_state = LEDS_GPIO_DEFSTATE_ON, + }, + { + .name = "led1", + .gpio = GPIO_PORT188, + .default_state = LEDS_GPIO_DEFSTATE_ON, + } +}; + +static struct gpio_led_platform_data ap4evb_leds_pdata = { + .num_leds = ARRAY_SIZE(ap4evb_leds), + .leds = &ap4evb_leds, +}; + +static struct platform_device leds_device = { + .name = "leds-gpio", + .id = 0, + .dev = { + .platform_data = &ap4evb_leds_pdata, + }, +}; + static struct platform_device *ap4evb_devices[] __initdata = { + &leds_device, &nor_flash_device, &smc911x_device, &sdhi0_device, @@ -840,20 +878,6 @@ static void __init ap4evb_init(void) gpio_request(GPIO_FN_CS5A, NULL); gpio_request(GPIO_FN_IRQ6_39, NULL); - /* enable LED 1 - 4 */ - gpio_request(GPIO_PORT185, NULL); - gpio_request(GPIO_PORT186, NULL); - gpio_request(GPIO_PORT187, NULL); - gpio_request(GPIO_PORT188, NULL); - gpio_direction_output(GPIO_PORT185, 1); - gpio_direction_output(GPIO_PORT186, 1); - gpio_direction_output(GPIO_PORT187, 1); - gpio_direction_output(GPIO_PORT188, 1); - gpio_export(GPIO_PORT185, 0); - gpio_export(GPIO_PORT186, 0); - gpio_export(GPIO_PORT187, 0); - gpio_export(GPIO_PORT188, 0); - /* enable Debug switch (S6) */ gpio_request(GPIO_PORT32, NULL); gpio_request(GPIO_PORT33, NULL); -- cgit v0.10.2 From 8050fbf268c3bc09752ceece40aff5d3f4e3e182 Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Mon, 16 Aug 2010 14:54:35 +0900 Subject: ARM: mach-shmobile: Fix up gpio-leds pdata definition. Signed-off-by: Paul Mundt diff --git a/arch/arm/mach-shmobile/board-ap4evb.c b/arch/arm/mach-shmobile/board-ap4evb.c index 98c1cf7..8afdb1d 100644 --- a/arch/arm/mach-shmobile/board-ap4evb.c +++ b/arch/arm/mach-shmobile/board-ap4evb.c @@ -676,7 +676,7 @@ static struct gpio_led ap4evb_leds[] = { static struct gpio_led_platform_data ap4evb_leds_pdata = { .num_leds = ARRAY_SIZE(ap4evb_leds), - .leds = &ap4evb_leds, + .leds = ap4evb_leds, }; static struct platform_device leds_device = { -- cgit v0.10.2 From 8d9df9f0844ed87541453a3ef91bfc9f487053b7 Mon Sep 17 00:00:00 2001 From: Xiaotian Feng Date: Mon, 16 Aug 2010 09:54:28 +0200 Subject: workqueue: free rescuer on destroy_workqueue wq->rescuer is not freed when wq is destroyed, leads a memory leak then. This patch also remove a redundant line. Signed-off-by: Xiaotian Feng Signed-off-by: Tejun Heo Cc: Oleg Nesterov diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 2994a0e..1001b6e 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -2782,7 +2782,6 @@ struct workqueue_struct *__alloc_workqueue_key(const char *name, if (IS_ERR(rescuer->task)) goto err; - wq->rescuer = rescuer; rescuer->task->flags |= PF_THREAD_BOUND; wake_up_process(rescuer->task); } @@ -2848,6 +2847,7 @@ void destroy_workqueue(struct workqueue_struct *wq) if (wq->flags & WQ_RESCUER) { kthread_stop(wq->rescuer->task); free_mayday_mask(wq->mayday_mask); + kfree(wq->rescuer); } free_cwqs(wq); -- cgit v0.10.2 From ec53fe3d655befce6420a1b8485af9b1145a0c90 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Eric=20B=C3=A9nard?= Date: Fri, 30 Jul 2010 22:58:40 +0200 Subject: eukrea-baseboards: fix the merge in one file MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Last change I did made a big mess in function names and CONFIG names, this patch fixes this so that the baseboard support is really built when selected. Signed-off-by: Eric Bénard Signed-off-by: Sascha Hauer diff --git a/arch/arm/mach-mx25/eukrea_mbimxsd-baseboard.c b/arch/arm/mach-mx25/eukrea_mbimxsd-baseboard.c index 91931dc..4aaadc7 100644 --- a/arch/arm/mach-mx25/eukrea_mbimxsd-baseboard.c +++ b/arch/arm/mach-mx25/eukrea_mbimxsd-baseboard.c @@ -215,7 +215,7 @@ struct imx_ssi_platform_data eukrea_mbimxsd_ssi_pdata = { * Add platform devices present on this baseboard and init * them from CPU side as far as required to use them later on */ -void __init eukrea_mbimxsd_baseboard_init(void) +void __init eukrea_mbimxsd25_baseboard_init(void) { if (mxc_iomux_v3_setup_multiple_pads(eukrea_mbimxsd_pads, ARRAY_SIZE(eukrea_mbimxsd_pads))) diff --git a/arch/arm/mach-mx25/mach-cpuimx25.c b/arch/arm/mach-mx25/mach-cpuimx25.c index 56b2e26..b5efee2 100644 --- a/arch/arm/mach-mx25/mach-cpuimx25.c +++ b/arch/arm/mach-mx25/mach-cpuimx25.c @@ -147,8 +147,8 @@ static void __init eukrea_cpuimx25_init(void) if (!otg_mode_host) mxc_register_device(&otg_udc_device, &otg_device_pdata); -#ifdef CONFIG_MACH_EUKREA_MBIMXSD_BASEBOARD - eukrea_mbimxsd_baseboard_init(); +#ifdef CONFIG_MACH_EUKREA_MBIMXSD25_BASEBOARD + eukrea_mbimxsd25_baseboard_init(); #endif } diff --git a/arch/arm/mach-mx3/eukrea_mbimxsd-baseboard.c b/arch/arm/mach-mx3/eukrea_mbimxsd-baseboard.c index 1dc5004..f8f15e3 100644 --- a/arch/arm/mach-mx3/eukrea_mbimxsd-baseboard.c +++ b/arch/arm/mach-mx3/eukrea_mbimxsd-baseboard.c @@ -216,7 +216,7 @@ struct imx_ssi_platform_data eukrea_mbimxsd_ssi_pdata = { * Add platform devices present on this baseboard and init * them from CPU side as far as required to use them later on */ -void __init eukrea_mbimxsd_baseboard_init(void) +void __init eukrea_mbimxsd35_baseboard_init(void) { if (mxc_iomux_v3_setup_multiple_pads(eukrea_mbimxsd_pads, ARRAY_SIZE(eukrea_mbimxsd_pads))) diff --git a/arch/arm/mach-mx3/mach-cpuimx35.c b/arch/arm/mach-mx3/mach-cpuimx35.c index 63f970f..e4268bb 100644 --- a/arch/arm/mach-mx3/mach-cpuimx35.c +++ b/arch/arm/mach-mx3/mach-cpuimx35.c @@ -201,8 +201,8 @@ static void __init mxc_board_init(void) if (!otg_mode_host) mxc_register_device(&mxc_otg_udc_device, &otg_device_pdata); -#ifdef CONFIG_MACH_EUKREA_MBIMXSD_BASEBOARD - eukrea_mbimxsd_baseboard_init(); +#ifdef CONFIG_MACH_EUKREA_MBIMXSD35_BASEBOARD + eukrea_mbimxsd35_baseboard_init(); #endif } diff --git a/arch/arm/plat-mxc/include/mach/eukrea-baseboards.h b/arch/arm/plat-mxc/include/mach/eukrea-baseboards.h index 634e3f4..656acb4 100644 --- a/arch/arm/plat-mxc/include/mach/eukrea-baseboards.h +++ b/arch/arm/plat-mxc/include/mach/eukrea-baseboards.h @@ -37,9 +37,9 @@ * mach-mx5/eukrea_mbimx51-baseboard.c for cpuimx51 */ -extern void eukrea_mbimx25_baseboard_init(void); +extern void eukrea_mbimxsd25_baseboard_init(void); extern void eukrea_mbimx27_baseboard_init(void); -extern void eukrea_mbimx35_baseboard_init(void); +extern void eukrea_mbimxsd35_baseboard_init(void); extern void eukrea_mbimx51_baseboard_init(void); #endif -- cgit v0.10.2 From 3d6e614952e3e4cf9e70e58893a740ffec220b24 Mon Sep 17 00:00:00 2001 From: "Hans J. Koch" Date: Thu, 12 Aug 2010 15:10:53 +0200 Subject: mx35: Fix boot ROM hang in internal boot mode MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If a watchdog reset occurs after booting in internal boot mode, the i.MX35 won't boot anymore. The boot ROM code seems to assume that some clocks are turned on (they are after a power-on reset). This patch turns on the necessary clocks. Signed-off-by: Hans J. Koch Acked-by: Uwe Kleine-König Reported-by: John Ogness Signed-off-by: Sascha Hauer diff --git a/arch/arm/mach-mx3/clock-imx35.c b/arch/arm/mach-mx3/clock-imx35.c index d3af0fd..32adf8d 100644 --- a/arch/arm/mach-mx3/clock-imx35.c +++ b/arch/arm/mach-mx3/clock-imx35.c @@ -485,10 +485,10 @@ static struct clk_lookup lookups[] = { int __init mx35_clocks_init() { - unsigned int ll = 0; + unsigned int cgr2 = 3 << 26, cgr3 = 0; #if defined(CONFIG_DEBUG_LL) && !defined(CONFIG_DEBUG_ICEDCC) - ll = (3 << 16); + cgr2 |= 3 << 16; #endif clkdev_add_table(lookups, ARRAY_SIZE(lookups)); @@ -499,8 +499,20 @@ int __init mx35_clocks_init() __raw_writel((3 << 18), CCM_BASE + CCM_CGR0); __raw_writel((3 << 2) | (3 << 4) | (3 << 6) | (3 << 8) | (3 << 16), CCM_BASE + CCM_CGR1); - __raw_writel((3 << 26) | ll, CCM_BASE + CCM_CGR2); - __raw_writel(0, CCM_BASE + CCM_CGR3); + + /* + * Check if we came up in internal boot mode. If yes, we need some + * extra clocks turned on, otherwise the MX35 boot ROM code will + * hang after a watchdog reset. + */ + if (!(__raw_readl(CCM_BASE + CCM_RCSR) & (3 << 10))) { + /* Additionally turn on UART1, SCC, and IIM clocks */ + cgr2 |= 3 << 16 | 3 << 4; + cgr3 |= 3 << 2; + } + + __raw_writel(cgr2, CCM_BASE + CCM_CGR2); + __raw_writel(cgr3, CCM_BASE + CCM_CGR3); mxc_timer_init(&gpt_clk, MX35_IO_ADDRESS(MX35_GPT1_BASE_ADDR), MX35_INT_GPT); -- cgit v0.10.2 From 9a6f17fd1cd84811962735ff47111f84a8f05073 Mon Sep 17 00:00:00 2001 From: Sascha Hauer Date: Thu, 22 Apr 2010 12:37:13 +0200 Subject: i.MX35: remove get_3_3_div helper function In the v2 reference manual there are no dividers combined of two dividers. Instead, all dividers are simple 6bit dividers. I assume the combined dividers only exist in preliminary hardware. Signed-off-by: Sascha Hauer diff --git a/arch/arm/mach-mx3/clock-imx35.c b/arch/arm/mach-mx3/clock-imx35.c index 32adf8d..d3a8d3f 100644 --- a/arch/arm/mach-mx3/clock-imx35.c +++ b/arch/arm/mach-mx3/clock-imx35.c @@ -176,16 +176,11 @@ static unsigned long get_rate_ipg(struct clk *clk) return get_rate_ahb(NULL) >> 1; } -static unsigned long get_3_3_div(unsigned long in) -{ - return (((in >> 3) & 0x7) + 1) * ((in & 0x7) + 1); -} - static unsigned long get_rate_uart(struct clk *clk) { unsigned long pdr3 = __raw_readl(CCM_BASE + CCM_PDR3); unsigned long pdr4 = __raw_readl(CCM_BASE + CCM_PDR4); - unsigned long div = get_3_3_div(pdr4 >> 10); + unsigned long div = ((pdr4 >> 10) & 0x3f) + 1; if (pdr3 & (1 << 14)) return get_rate_arm() / div; @@ -216,7 +211,7 @@ static unsigned long get_rate_sdhc(struct clk *clk) break; } - return rate / get_3_3_div(div); + return rate / (div + 1); } static unsigned long get_rate_mshc(struct clk *clk) @@ -270,7 +265,7 @@ static unsigned long get_rate_csi(struct clk *clk) else rate = get_rate_ppll(); - return rate / get_3_3_div((pdr2 >> 16) & 0x3f); + return rate / (((pdr2 >> 16) & 0x3f) + 1); } static unsigned long get_rate_otg(struct clk *clk) @@ -283,22 +278,21 @@ static unsigned long get_rate_otg(struct clk *clk) else rate = get_rate_ppll(); - return rate / get_3_3_div((pdr4 >> 22) & 0x3f); + return rate / (((pdr4 >> 22) & 0x3f) + 1); } static unsigned long get_rate_ipg_per(struct clk *clk) { unsigned long pdr0 = __raw_readl(CCM_BASE + CCM_PDR0); unsigned long pdr4 = __raw_readl(CCM_BASE + CCM_PDR4); - unsigned long div1, div2; + unsigned long div; if (pdr0 & (1 << 26)) { - div1 = (pdr4 >> 19) & 0x7; - div2 = (pdr4 >> 16) & 0x7; - return get_rate_arm() / ((div1 + 1) * (div2 + 1)); + div = (pdr4 >> 16) & 0x3f; + return get_rate_arm() / (div + 1); } else { - div1 = (pdr0 >> 12) & 0x7; - return get_rate_ahb(NULL) / div1; + div = (pdr0 >> 12) & 0x7; + return get_rate_ahb(NULL) / div; } } -- cgit v0.10.2 From c96c1e394fa72bd995154af7143873945a3fa89d Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Mon, 16 Aug 2010 15:54:05 +0200 Subject: clock-imx35: fix divider if ahb is source for ipg_per According to the Datasheet: "i.MX35 (MCIMX35) Multimedia Applications Processor Reference Manual, Rev. 2" "Table 14-6. PDR0 Field Descriptions" the divider is CCM_PER_AHB[3:0] + 1. This patch adds the missing + 1. Signed-off-by: Marc Kleine-Budde Signed-off-by: Michael Grzeschik Signed-off-by: Sascha Hauer diff --git a/arch/arm/mach-mx3/clock-imx35.c b/arch/arm/mach-mx3/clock-imx35.c index d3a8d3f..e582ceb 100644 --- a/arch/arm/mach-mx3/clock-imx35.c +++ b/arch/arm/mach-mx3/clock-imx35.c @@ -292,7 +292,7 @@ static unsigned long get_rate_ipg_per(struct clk *clk) return get_rate_arm() / (div + 1); } else { div = (pdr0 >> 12) & 0x7; - return get_rate_ahb(NULL) / div; + return get_rate_ahb(NULL) / (div + 1); } } -- cgit v0.10.2 From c56702de106b4fde301282f7ea60412276481442 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Mon, 16 Aug 2010 15:54:06 +0200 Subject: clock-imx35: correct arm and ahb clock calculation According to the Datasheet: "i.MX35 (MCIMX35) Multimedia Applications Processor Reference Manual, Rev. 2" "Figure 14-24. Clock Control And Gating" change the result of get_rate_ahb based on the frequency returned by get_rate_arm to calculate the proper rate. Signed-off-by: Marc Kleine-Budde Signed-off-by: Michael Grzeschik Signed-off-by: Sascha Hauer diff --git a/arch/arm/mach-mx3/clock-imx35.c b/arch/arm/mach-mx3/clock-imx35.c index e582ceb..1d311b7 100644 --- a/arch/arm/mach-mx3/clock-imx35.c +++ b/arch/arm/mach-mx3/clock-imx35.c @@ -155,7 +155,7 @@ static unsigned long get_rate_arm(void) aad = &clk_consumer[(pdr0 >> 16) & 0xf]; if (aad->sel) - fref = fref * 2 / 3; + fref = fref * 3 / 4; return fref / aad->arm; } @@ -164,7 +164,7 @@ static unsigned long get_rate_ahb(struct clk *clk) { unsigned long pdr0 = __raw_readl(CCM_BASE + CCM_PDR0); struct arm_ahb_div *aad; - unsigned long fref = get_rate_mpll(); + unsigned long fref = get_rate_arm(); aad = &clk_consumer[(pdr0 >> 16) & 0xf]; -- cgit v0.10.2 From c735c40587cea45f9d2f83339f374595073c1923 Mon Sep 17 00:00:00 2001 From: Juergen Beisert Date: Mon, 16 Aug 2010 15:54:07 +0200 Subject: clock-imx35: Calculate the base clock rate for the IPU unit The mx3fb driver needs the clock the IPU runs in order to calculate the divider for the LCD clock. This patch adds the clock rate calculation routine for the i.MX35 CPU. Signed-off-by: Juergen Beisert Signed-off-by: Michael Grzeschik Signed-off-by: Sascha Hauer diff --git a/arch/arm/mach-mx3/clock-imx35.c b/arch/arm/mach-mx3/clock-imx35.c index 1d311b7..7a62e74 100644 --- a/arch/arm/mach-mx3/clock-imx35.c +++ b/arch/arm/mach-mx3/clock-imx35.c @@ -296,6 +296,33 @@ static unsigned long get_rate_ipg_per(struct clk *clk) } } +static unsigned long get_rate_hsp(struct clk *clk) +{ + unsigned long hsp_podf = (__raw_readl(CCM_BASE + CCM_PDR0) >> 20) & 0x03; + unsigned long fref = get_rate_mpll(); + + if (fref > 400 * 1000 * 1000) { + switch (hsp_podf) { + case 0: + return fref >> 2; + case 1: + return fref >> 3; + case 2: + return fref / 3; + } + } else { + switch (hsp_podf) { + case 0: + case 2: + return fref / 3; + case 1: + return fref / 6; + } + } + + return 0; +} + static int clk_cgr_enable(struct clk *clk) { u32 reg; @@ -353,7 +380,7 @@ DEFINE_CLOCK(i2c1_clk, 0, CCM_CGR1, 10, get_rate_ipg_per, NULL); DEFINE_CLOCK(i2c2_clk, 1, CCM_CGR1, 12, get_rate_ipg_per, NULL); DEFINE_CLOCK(i2c3_clk, 2, CCM_CGR1, 14, get_rate_ipg_per, NULL); DEFINE_CLOCK(iomuxc_clk, 0, CCM_CGR1, 16, NULL, NULL); -DEFINE_CLOCK(ipu_clk, 0, CCM_CGR1, 18, get_rate_ahb, NULL); +DEFINE_CLOCK(ipu_clk, 0, CCM_CGR1, 18, get_rate_hsp, NULL); DEFINE_CLOCK(kpp_clk, 0, CCM_CGR1, 20, get_rate_ipg, NULL); DEFINE_CLOCK(mlb_clk, 0, CCM_CGR1, 22, get_rate_ahb, NULL); DEFINE_CLOCK(mshc_clk, 0, CCM_CGR1, 24, get_rate_mshc, NULL); -- cgit v0.10.2 From 23b90cfd7bcf8ae1c0711df1cae7dfec3c01aec8 Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Tue, 17 Aug 2010 09:15:24 -0700 Subject: x86/PCI: only define pci_domain_nr if PCI and PCI_DOMAINS are set Otherwise we'll duplicate definitions with the pci.h stubs. Reported-by: Randy Dunlap Acked-by: Randy Dunlap Signed-off-by: Jesse Barnes diff --git a/arch/x86/include/asm/pci.h b/arch/x86/include/asm/pci.h index 404a880..d395540 100644 --- a/arch/x86/include/asm/pci.h +++ b/arch/x86/include/asm/pci.h @@ -27,6 +27,9 @@ extern struct pci_bus *pci_scan_bus_on_node(int busno, struct pci_ops *ops, int node); extern struct pci_bus *pci_scan_bus_with_sysdata(int busno); +#ifdef CONFIG_PCI + +#ifdef CONFIG_PCI_DOMAINS static inline int pci_domain_nr(struct pci_bus *bus) { struct pci_sysdata *sd = bus->sysdata; @@ -37,13 +40,12 @@ static inline int pci_proc_domain(struct pci_bus *bus) { return pci_domain_nr(bus); } - +#endif /* Can be used to override the logic in pci_scan_bus for skipping already-configured bus numbers - to be used for buggy BIOSes or architectures with incomplete PCI setup by the loader */ -#ifdef CONFIG_PCI extern unsigned int pcibios_assign_all_busses(void); extern int pci_legacy_init(void); # ifdef CONFIG_ACPI -- cgit v0.10.2 From 3572b33efb044f76c224197408a812cda7d54fe4 Mon Sep 17 00:00:00 2001 From: Magnus Damm Date: Mon, 16 Aug 2010 10:40:49 +0000 Subject: ARM: mach-shmobile: Clock framework comment fix Update clock framework header comment for SH-Mobile ARM. Signed-off-by: Magnus Damm Signed-off-by: Paul Mundt diff --git a/arch/arm/mach-shmobile/clock.c b/arch/arm/mach-shmobile/clock.c index b7c705a..6b7c7c4 100644 --- a/arch/arm/mach-shmobile/clock.c +++ b/arch/arm/mach-shmobile/clock.c @@ -1,8 +1,10 @@ /* - * SH-Mobile Timer + * SH-Mobile Clock Framework * * Copyright (C) 2010 Magnus Damm * + * Used together with arch/arm/common/clkdev.c and drivers/sh/clk.c. + * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by * the Free Software Foundation; version 2 of the License. -- cgit v0.10.2 From 4e1774f3a5de0885be2e59a8e8ef0e99c5f79e7a Mon Sep 17 00:00:00 2001 From: Guennadi Liakhovetski Date: Tue, 17 Aug 2010 22:53:58 +0000 Subject: ARM: mach-shmobile: do not enable the PLLC2 clock on init AP4EVB has to reparent PLLC2 to provide a precise HDMI clock, this is only possible, if PLLC2 hasn't been enabled yet. Since no other driver currently uses PLLC2 we can safely remove the CLK_ENABLE_ON_INIT flag. This fixes the "Cannot set PLLC2 parent: -16, 1 users" error message, when trying to use HDMI on AP4EVB. Signed-off-by: Guennadi Liakhovetski Signed-off-by: Paul Mundt diff --git a/arch/arm/mach-shmobile/clock-sh7372.c b/arch/arm/mach-shmobile/clock-sh7372.c index dcaac1b..7594689 100644 --- a/arch/arm/mach-shmobile/clock-sh7372.c +++ b/arch/arm/mach-shmobile/clock-sh7372.c @@ -286,7 +286,6 @@ static struct clk_ops pllc2_clk_ops = { struct clk pllc2_clk = { .ops = &pllc2_clk_ops, - .flags = CLK_ENABLE_ON_INIT, .parent = &extal1_div2_clk, .freq_table = pllc2_freq_table, .parent_table = pllc2_parent, -- cgit v0.10.2 From 9f6f11b65ea2da8fab4d6902595d770e89e64d93 Mon Sep 17 00:00:00 2001 From: Kuninori Morimoto Date: Wed, 18 Aug 2010 03:37:19 +0000 Subject: ARM: mach-shmobile: ap4evb: modify FSI2 ID Signed-off-by: Kuninori Morimoto Signed-off-by: Paul Mundt diff --git a/arch/arm/mach-shmobile/board-ap4evb.c b/arch/arm/mach-shmobile/board-ap4evb.c index 8afdb1d..1b95f12 100644 --- a/arch/arm/mach-shmobile/board-ap4evb.c +++ b/arch/arm/mach-shmobile/board-ap4evb.c @@ -559,7 +559,7 @@ static struct resource fsi_resources[] = { static struct platform_device fsi_device = { .name = "sh_fsi2", - .id = 0, + .id = -1, .num_resources = ARRAY_SIZE(fsi_resources), .resource = fsi_resources, .dev = { -- cgit v0.10.2 From 410d878bd05315a19894b8b4685107f2cd0712a5 Mon Sep 17 00:00:00 2001 From: Arnd Hannemann Date: Wed, 18 Aug 2010 16:24:13 +0000 Subject: ARM: mach-shmobile: ap4evb: fix write protect for SDHI1 SDHI1 on AP4EVB has no pin to detect write protection of a card. This had the result that a card inserted in this slot was always detected as read only. This patch adds the corresponding flag to disable write protection detection for SDHI1. Signed-off-by: Arnd Hannemann Signed-off-by: Paul Mundt diff --git a/arch/arm/mach-shmobile/board-ap4evb.c b/arch/arm/mach-shmobile/board-ap4evb.c index 1b95f12..95935c8 100644 --- a/arch/arm/mach-shmobile/board-ap4evb.c +++ b/arch/arm/mach-shmobile/board-ap4evb.c @@ -25,6 +25,7 @@ #include #include #include +#include #include #include #include @@ -308,6 +309,7 @@ static struct sh_mobile_sdhi_info sdhi1_info = { .dma_slave_tx = SHDMA_SLAVE_SDHI1_TX, .dma_slave_rx = SHDMA_SLAVE_SDHI1_RX, .tmio_ocr_mask = MMC_VDD_165_195, + .tmio_flags = TMIO_MMC_WRPROTECT_DISABLE, }; static struct resource sdhi1_resources[] = { -- cgit v0.10.2 From 51e3c1b558b31b11bf5fc66d3c6f5adacf3573f7 Mon Sep 17 00:00:00 2001 From: Sergey Senozhatsky Date: Fri, 20 Aug 2010 10:36:34 +0300 Subject: x86, hwmon: Fix unsafe smp_processor_id() in thermal_throttle_add_dev Fix BUG: using smp_processor_id() in preemptible thermal_throttle_add_dev. We know the cpu number when calling thermal_throttle_add_dev, so we can remove smp_processor_id call in thermal_throttle_add_dev by supplying the cpu number as argument. This should resolve kernel bugzilla 16615/16629. Signed-off-by: Sergey Senozhatsky LKML-Reference: <20100820073634.GB5209@swordfish.minsk.epam.com> Cc: Fenghua Yu Cc: Joerg Roedel Cc: Maciej Rutecki Signed-off-by: H. Peter Anvin diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c index c2a8b26..d9368ee 100644 --- a/arch/x86/kernel/cpu/mcheck/therm_throt.c +++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c @@ -202,10 +202,11 @@ static int therm_throt_process(bool new_event, int event, int level) #ifdef CONFIG_SYSFS /* Add/Remove thermal_throttle interface for CPU device: */ -static __cpuinit int thermal_throttle_add_dev(struct sys_device *sys_dev) +static __cpuinit int thermal_throttle_add_dev(struct sys_device *sys_dev, + unsigned int cpu) { int err; - struct cpuinfo_x86 *c = &cpu_data(smp_processor_id()); + struct cpuinfo_x86 *c = &cpu_data(cpu); err = sysfs_create_group(&sys_dev->kobj, &thermal_attr_group); if (err) @@ -251,7 +252,7 @@ thermal_throttle_cpu_callback(struct notifier_block *nfb, case CPU_UP_PREPARE: case CPU_UP_PREPARE_FROZEN: mutex_lock(&therm_cpu_lock); - err = thermal_throttle_add_dev(sys_dev); + err = thermal_throttle_add_dev(sys_dev, cpu); mutex_unlock(&therm_cpu_lock); WARN_ON(err); break; @@ -287,7 +288,7 @@ static __init int thermal_throttle_init_device(void) #endif /* connect live CPUs to sysfs */ for_each_online_cpu(cpu) { - err = thermal_throttle_add_dev(get_cpu_sysdev(cpu)); + err = thermal_throttle_add_dev(get_cpu_sysdev(cpu), cpu); WARN_ON(err); } #ifdef CONFIG_HOTPLUG_CPU -- cgit v0.10.2 From a38b372fc2a3d8fe7ddbeab94d4c8bcb721f2403 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Sat, 21 Aug 2010 16:24:03 +0800 Subject: mxc/tzic: add base address when accessing TZIC registers When we call tzic_enable_wake function, the kernel will crash because of access to an unmapped address. This is because two register access operations forgot to add base address. Signed-off-by: Jason Wang Signed-off-by: Sascha Hauer diff --git a/arch/arm/plat-mxc/tzic.c b/arch/arm/plat-mxc/tzic.c index b3da9aa..3703ab2 100644 --- a/arch/arm/plat-mxc/tzic.c +++ b/arch/arm/plat-mxc/tzic.c @@ -164,8 +164,9 @@ int tzic_enable_wake(int is_idle) return -EAGAIN; for (i = 0; i < 4; i++) { - v = is_idle ? __raw_readl(TZIC_ENSET0(i)) : wakeup_intr[i]; - __raw_writel(v, TZIC_WAKEUP0(i)); + v = is_idle ? __raw_readl(tzic_base + TZIC_ENSET0(i)) : + wakeup_intr[i]; + __raw_writel(v, tzic_base + TZIC_WAKEUP0(i)); } return 0; -- cgit v0.10.2 From 7f67bff58662604874778e6853e5c394f944e554 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Sat, 21 Aug 2010 16:24:04 +0800 Subject: mx5/clock: fix clear bit fields issue in _clk_ccgr_disable function We can see MXC_CCM_CCGRx_MOD_OFF is defined as 0 while MXC_CCM_CCGRx_CG_MASK is defined as 0x3 in crm_regs.h, here in the _clk_ccgr_disable function, we want to clear the corresponding enable bit fields to disable this clock, so we should choose MASK instead of OFF otherwise clocks can't be disabled. Signed-off-by: Jason Wang Signed-off-by: Sascha Hauer diff --git a/arch/arm/mach-mx5/clock-mx51.c b/arch/arm/mach-mx5/clock-mx51.c index 6af69de..57c10a9 100644 --- a/arch/arm/mach-mx5/clock-mx51.c +++ b/arch/arm/mach-mx5/clock-mx51.c @@ -56,7 +56,7 @@ static void _clk_ccgr_disable(struct clk *clk) { u32 reg; reg = __raw_readl(clk->enable_reg); - reg &= ~(MXC_CCM_CCGRx_MOD_OFF << clk->enable_shift); + reg &= ~(MXC_CCM_CCGRx_CG_MASK << clk->enable_shift); __raw_writel(reg, clk->enable_reg); } -- cgit v0.10.2 From 14f0f512ba6f4eccc9b7cf67be47439ed2917afb Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Sat, 21 Aug 2010 16:24:05 +0800 Subject: ARM: imx: set cache line size to 64 bytes for i.MX5 The core of i.MX5 series is cortex-A8, its cache line size is 64 bytes instead of 32 bytes. Refer to the OMAP3's selection, we choose 64 bytes for i.MX5, this can increase a little bit performance when perform cache operations. Signed-off-by: Jason Wang Signed-off-by: Sascha Hauer diff --git a/arch/arm/plat-mxc/Kconfig b/arch/arm/plat-mxc/Kconfig index 0527e65..6785db4 100644 --- a/arch/arm/plat-mxc/Kconfig +++ b/arch/arm/plat-mxc/Kconfig @@ -43,6 +43,7 @@ config ARCH_MXC91231 config ARCH_MX5 bool "MX5-based" select CPU_V7 + select ARM_L1_CACHE_SHIFT_6 help This enables support for systems based on the Freescale i.MX51 family -- cgit v0.10.2 From 96aa1b419d47286db446f292cf898bb1a8b27f24 Mon Sep 17 00:00:00 2001 From: Ciju Rajan K Date: Mon, 23 Aug 2010 10:56:30 +0200 Subject: blkio: Fix return code for mkdir calls If the cgroup hierarchy for blkio control groups is deeper than two levels, kernel should not allow the creation of further levels. mkdir system call does not except EINVAL as a return value. This patch replaces EINVAL with more appropriate EPERM Signed-off-by: Ciju Rajan K Reviewed-by: KAMEZAWA Hiroyuki Signed-off-by: Jens Axboe diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c index a680964..2fef1ef 100644 --- a/block/blk-cgroup.c +++ b/block/blk-cgroup.c @@ -966,7 +966,7 @@ blkiocg_create(struct cgroup_subsys *subsys, struct cgroup *cgroup) /* Currently we do not support hierarchy deeper than two level (0,1) */ if (parent != cgroup->top_cgroup) - return ERR_PTR(-EINVAL); + return ERR_PTR(-EPERM); blkcg = kzalloc(sizeof(*blkcg), GFP_KERNEL); if (!blkcg) -- cgit v0.10.2 From 75230ff2751e88d594a13a70eae2c146f45e323b Mon Sep 17 00:00:00 2001 From: "Stephen M. Cameron" Date: Mon, 23 Aug 2010 11:02:17 +0200 Subject: cciss: disable doorbell reset on reset_devices The doorbell reset initially appears to work correctly, the controller resets, comes up, some i/o can even be done, but on at least some Smart Arrays in some servers, it eventually causes a subsequent controller lockup due to some kind of PCIe error, and kdump can end up leaving the root filesystem in an unbootable state. For this reason, until the problem is fixed, or at least isolated to certain hardware enough to be avoided, the doorbell reset should not be used at all. Signed-off-by: Stephen M. Cameron Signed-off-by: Jens Axboe diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c index 31064df..ce1a75d 100644 --- a/drivers/block/cciss.c +++ b/drivers/block/cciss.c @@ -4519,6 +4519,12 @@ static __devinit int cciss_kdump_hard_reset_controller(struct pci_dev *pdev) misc_fw_support = readl(&cfgtable->misc_fw_support); use_doorbell = misc_fw_support & MISC_FW_DOORBELL_RESET; + /* The doorbell reset seems to cause lockups on some Smart + * Arrays (e.g. P410, P410i, maybe others). Until this is + * fixed or at least isolated, avoid the doorbell reset. + */ + use_doorbell = 0; + rc = cciss_controller_hard_reset(pdev, vaddr, use_doorbell); if (rc) goto unmap_cfgtable; -- cgit v0.10.2 From 06bd6ebffae36d3b105677598c48e8bd0a10b205 Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Sun, 22 Aug 2010 23:19:42 +0900 Subject: workqueue: annotate lock context change Some of internal functions called within gcwq->lock context releases and regrabs the lock but were missing proper annotations. Add it. Signed-off-by: Namhyung Kim Signed-off-by: Tejun Heo diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 1001b6e..7415f27a 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -1485,6 +1485,8 @@ static void gcwq_mayday_timeout(unsigned long __gcwq) * otherwise. */ static bool maybe_create_worker(struct global_cwq *gcwq) +__releases(&gcwq->lock) +__acquires(&gcwq->lock) { if (!need_to_create_worker(gcwq)) return false; @@ -1722,6 +1724,8 @@ static void cwq_dec_nr_in_flight(struct cpu_workqueue_struct *cwq, int color) * spin_lock_irq(gcwq->lock) which is released and regrabbed. */ static void process_one_work(struct worker *worker, struct work_struct *work) +__releases(&gcwq->lock) +__acquires(&gcwq->lock) { struct cpu_workqueue_struct *cwq = get_work_cwq(work); struct global_cwq *gcwq = cwq->gcwq; @@ -3230,6 +3234,8 @@ static int __cpuinit trustee_thread(void *__gcwq) * multiple times. To be used by cpu_callback. */ static void __cpuinit wait_trustee_state(struct global_cwq *gcwq, int state) +__releases(&gcwq->lock) +__acquires(&gcwq->lock) { if (!(gcwq->trustee_state == state || gcwq->trustee_state == TRUSTEE_DONE)) { -- cgit v0.10.2 From 972fa1c5316d18c8297123e08e9b6930ca34f888 Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Sun, 22 Aug 2010 23:19:43 +0900 Subject: workqueue: mark lock acquisition on worker_maybe_bind_and_lock() worker_maybe_bind_and_lock() actually grabs gcwq->lock but was missing proper annotation. Add it. So this patch will remove following sparse warnings: kernel/workqueue.c:1214:13: warning: context imbalance in 'worker_maybe_bind_and_lock' - wrong count at exit arch/x86/include/asm/irqflags.h:44:9: warning: context imbalance in 'worker_rebind_fn' - unexpected unlock kernel/workqueue.c:1991:17: warning: context imbalance in 'rescuer_thread' - unexpected unlock Signed-off-by: Namhyung Kim Signed-off-by: Tejun Heo diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 7415f27a..cc3456f 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -1212,6 +1212,7 @@ static void worker_leave_idle(struct worker *worker) * bound), %false if offline. */ static bool worker_maybe_bind_and_lock(struct worker *worker) +__acquires(&gcwq->lock) { struct global_cwq *gcwq = worker->gcwq; struct task_struct *task = worker->task; -- cgit v0.10.2 From b6508c1618e7aab085f191efb41b7b019a94ea38 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Mon, 23 Aug 2010 12:23:33 +0200 Subject: cfq-iosched: Do not idle if slice_idle=0 Do not idle either on cfq queue or service tree if slice_idle=0. User does not want any queue or service tree idling. Currently even if slice_idle=0, we were waiting for request to finish before expiring the queue and that can lead to lower queue depths. Acked-by: Jeff Moyer Signed-off-by: Vivek Goyal Signed-off-by: Jens Axboe diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index eb4086f..8830569 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -1839,6 +1839,9 @@ static bool cfq_should_idle(struct cfq_data *cfqd, struct cfq_queue *cfqq) BUG_ON(!service_tree); BUG_ON(!service_tree->count); + if (!cfqd->cfq_slice_idle) + return false; + /* We never do for idle class queues. */ if (prio == IDLE_WORKLOAD) return false; @@ -1879,7 +1882,7 @@ static void cfq_arm_slice_timer(struct cfq_data *cfqd) /* * idle is disabled, either manually or by past process history */ - if (!cfqd->cfq_slice_idle || !cfq_should_idle(cfqd, cfqq)) + if (!cfq_should_idle(cfqd, cfqq)) return; /* -- cgit v0.10.2 From 02b35081fc98f681411586d3acf9eaad8b8f6e07 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Mon, 23 Aug 2010 12:23:53 +0200 Subject: cfq-iosched: Do group share accounting in IOPS when slice_idle=0 o Implement another CFQ mode where we charge group in terms of number of requests dispatched instead of measuring the time. Measuring in terms of time is not possible when we are driving deeper queue depths and there are requests from multiple cfq queues in the request queue. o This mode currently gets activated if one sets slice_idle=0 and associated disk supports NCQ. Again the idea is that on an NCQ disk with idling disabled most of the queues will dispatch 1 or more requests and then cfq queue expiry happens and we don't have a way to measure time. So start providing fairness in terms of IOPS. o Currently IOPS mode works only with cfq group scheduling. CFQ is following different scheduling algorithms for queue and group scheduling. These IOPS stats are used only for group scheduling hence in non-croup mode nothing should change. o For CFQ group scheduling one can disable slice idling so that we don't idle on queue and drive deeper request queue depths (achieving better throughput), at the same time group idle is enabled so one should get service differentiation among groups. Signed-off-by: Vivek Goyal Acked-by: Jeff Moyer Signed-off-by: Jens Axboe diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 8830569..3fc6be1 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -378,6 +378,21 @@ CFQ_CFQQ_FNS(wait_busy); &cfqg->service_trees[i][j]: NULL) \ +static inline bool iops_mode(struct cfq_data *cfqd) +{ + /* + * If we are not idling on queues and it is a NCQ drive, parallel + * execution of requests is on and measuring time is not possible + * in most of the cases until and unless we drive shallower queue + * depths and that becomes a performance bottleneck. In such cases + * switch to start providing fairness in terms of number of IOs. + */ + if (!cfqd->cfq_slice_idle && cfqd->hw_tag) + return true; + else + return false; +} + static inline enum wl_prio_t cfqq_prio(struct cfq_queue *cfqq) { if (cfq_class_idle(cfqq)) @@ -906,7 +921,6 @@ static inline unsigned int cfq_cfqq_slice_usage(struct cfq_queue *cfqq) slice_used = cfqq->allocated_slice; } - cfq_log_cfqq(cfqq->cfqd, cfqq, "sl_used=%u", slice_used); return slice_used; } @@ -914,19 +928,21 @@ static void cfq_group_served(struct cfq_data *cfqd, struct cfq_group *cfqg, struct cfq_queue *cfqq) { struct cfq_rb_root *st = &cfqd->grp_service_tree; - unsigned int used_sl, charge_sl; + unsigned int used_sl, charge; int nr_sync = cfqg->nr_cfqq - cfqg_busy_async_queues(cfqd, cfqg) - cfqg->service_tree_idle.count; BUG_ON(nr_sync < 0); - used_sl = charge_sl = cfq_cfqq_slice_usage(cfqq); + used_sl = charge = cfq_cfqq_slice_usage(cfqq); - if (!cfq_cfqq_sync(cfqq) && !nr_sync) - charge_sl = cfqq->allocated_slice; + if (iops_mode(cfqd)) + charge = cfqq->slice_dispatch; + else if (!cfq_cfqq_sync(cfqq) && !nr_sync) + charge = cfqq->allocated_slice; /* Can't update vdisktime while group is on service tree */ cfq_rb_erase(&cfqg->rb_node, st); - cfqg->vdisktime += cfq_scale_slice(charge_sl, cfqg); + cfqg->vdisktime += cfq_scale_slice(charge, cfqg); __cfq_group_service_tree_add(st, cfqg); /* This group is being expired. Save the context */ @@ -940,6 +956,8 @@ static void cfq_group_served(struct cfq_data *cfqd, struct cfq_group *cfqg, cfq_log_cfqg(cfqd, cfqg, "served: vt=%llu min_vt=%llu", cfqg->vdisktime, st->min_vdisktime); + cfq_log_cfqq(cfqq->cfqd, cfqq, "sl_used=%u disp=%u charge=%u iops=%u", + used_sl, cfqq->slice_dispatch, charge, iops_mode(cfqd)); cfq_blkiocg_update_timeslice_used(&cfqg->blkg, used_sl); cfq_blkiocg_set_start_empty_time(&cfqg->blkg); } -- cgit v0.10.2 From 80bdf0c78fff075dfa21576273c8b0c7db22bdfe Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Mon, 23 Aug 2010 12:24:26 +0200 Subject: cfq-iosched: Implement tunable group_idle o Implement a new tunable group_idle, which allows idling on the group instead of a cfq queue. Hence one can set slice_idle = 0 and not idle on the individual queues but idle on the group. This way on fast storage we can get fairness between groups at the same time overall throughput improves. Signed-off-by: Vivek Goyal Acked-by: Jeff Moyer Signed-off-by: Jens Axboe diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 3fc6be1..85e4819 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -30,6 +30,7 @@ static const int cfq_slice_sync = HZ / 10; static int cfq_slice_async = HZ / 25; static const int cfq_slice_async_rq = 2; static int cfq_slice_idle = HZ / 125; +static int cfq_group_idle = HZ / 125; static const int cfq_target_latency = HZ * 3/10; /* 300 ms */ static const int cfq_hist_divisor = 4; @@ -198,6 +199,8 @@ struct cfq_group { struct hlist_node cfqd_node; atomic_t ref; #endif + /* number of requests that are on the dispatch list or inside driver */ + int dispatched; }; /* @@ -271,6 +274,7 @@ struct cfq_data { unsigned int cfq_slice[2]; unsigned int cfq_slice_async_rq; unsigned int cfq_slice_idle; + unsigned int cfq_group_idle; unsigned int cfq_latency; unsigned int cfq_group_isolation; @@ -1884,7 +1888,7 @@ static void cfq_arm_slice_timer(struct cfq_data *cfqd) { struct cfq_queue *cfqq = cfqd->active_queue; struct cfq_io_context *cic; - unsigned long sl; + unsigned long sl, group_idle = 0; /* * SSD device without seek penalty, disable idling. But only do so @@ -1900,8 +1904,13 @@ static void cfq_arm_slice_timer(struct cfq_data *cfqd) /* * idle is disabled, either manually or by past process history */ - if (!cfq_should_idle(cfqd, cfqq)) - return; + if (!cfq_should_idle(cfqd, cfqq)) { + /* no queue idling. Check for group idling */ + if (cfqd->cfq_group_idle) + group_idle = cfqd->cfq_group_idle; + else + return; + } /* * still active requests from this queue, don't idle @@ -1928,13 +1937,21 @@ static void cfq_arm_slice_timer(struct cfq_data *cfqd) return; } + /* There are other queues in the group, don't do group idle */ + if (group_idle && cfqq->cfqg->nr_cfqq > 1) + return; + cfq_mark_cfqq_wait_request(cfqq); - sl = cfqd->cfq_slice_idle; + if (group_idle) + sl = cfqd->cfq_group_idle; + else + sl = cfqd->cfq_slice_idle; mod_timer(&cfqd->idle_slice_timer, jiffies + sl); cfq_blkiocg_update_set_idle_time_stats(&cfqq->cfqg->blkg); - cfq_log_cfqq(cfqd, cfqq, "arm_idle: %lu", sl); + cfq_log_cfqq(cfqd, cfqq, "arm_idle: %lu group_idle: %d", sl, + group_idle ? 1 : 0); } /* @@ -1950,6 +1967,7 @@ static void cfq_dispatch_insert(struct request_queue *q, struct request *rq) cfqq->next_rq = cfq_find_next_rq(cfqd, cfqq, rq); cfq_remove_request(rq); cfqq->dispatched++; + (RQ_CFQG(rq))->dispatched++; elv_dispatch_sort(q, rq); cfqd->rq_in_flight[cfq_cfqq_sync(cfqq)]++; @@ -2219,7 +2237,7 @@ static struct cfq_queue *cfq_select_queue(struct cfq_data *cfqd) cfqq = NULL; goto keep_queue; } else - goto expire; + goto check_group_idle; } /* @@ -2247,8 +2265,23 @@ static struct cfq_queue *cfq_select_queue(struct cfq_data *cfqd) * flight or is idling for a new request, allow either of these * conditions to happen (or time out) before selecting a new queue. */ - if (timer_pending(&cfqd->idle_slice_timer) || - (cfqq->dispatched && cfq_should_idle(cfqd, cfqq))) { + if (timer_pending(&cfqd->idle_slice_timer)) { + cfqq = NULL; + goto keep_queue; + } + + if (cfqq->dispatched && cfq_should_idle(cfqd, cfqq)) { + cfqq = NULL; + goto keep_queue; + } + + /* + * If group idle is enabled and there are requests dispatched from + * this group, wait for requests to complete. + */ +check_group_idle: + if (cfqd->cfq_group_idle && cfqq->cfqg->nr_cfqq == 1 + && cfqq->cfqg->dispatched) { cfqq = NULL; goto keep_queue; } @@ -3396,6 +3429,7 @@ static void cfq_completed_request(struct request_queue *q, struct request *rq) WARN_ON(!cfqq->dispatched); cfqd->rq_in_driver--; cfqq->dispatched--; + (RQ_CFQG(rq))->dispatched--; cfq_blkiocg_update_completion_stats(&cfqq->cfqg->blkg, rq_start_time_ns(rq), rq_io_start_time_ns(rq), rq_data_dir(rq), rq_is_sync(rq)); @@ -3425,7 +3459,10 @@ static void cfq_completed_request(struct request_queue *q, struct request *rq) * the queue. */ if (cfq_should_wait_busy(cfqd, cfqq)) { - cfqq->slice_end = jiffies + cfqd->cfq_slice_idle; + unsigned long extend_sl = cfqd->cfq_slice_idle; + if (!cfqd->cfq_slice_idle) + extend_sl = cfqd->cfq_group_idle; + cfqq->slice_end = jiffies + extend_sl; cfq_mark_cfqq_wait_busy(cfqq); cfq_log_cfqq(cfqd, cfqq, "will busy wait"); } @@ -3871,6 +3908,7 @@ static void *cfq_init_queue(struct request_queue *q) cfqd->cfq_slice[1] = cfq_slice_sync; cfqd->cfq_slice_async_rq = cfq_slice_async_rq; cfqd->cfq_slice_idle = cfq_slice_idle; + cfqd->cfq_group_idle = cfq_group_idle; cfqd->cfq_latency = 1; cfqd->cfq_group_isolation = 0; cfqd->hw_tag = -1; @@ -3943,6 +3981,7 @@ SHOW_FUNCTION(cfq_fifo_expire_async_show, cfqd->cfq_fifo_expire[0], 1); SHOW_FUNCTION(cfq_back_seek_max_show, cfqd->cfq_back_max, 0); SHOW_FUNCTION(cfq_back_seek_penalty_show, cfqd->cfq_back_penalty, 0); SHOW_FUNCTION(cfq_slice_idle_show, cfqd->cfq_slice_idle, 1); +SHOW_FUNCTION(cfq_group_idle_show, cfqd->cfq_group_idle, 1); SHOW_FUNCTION(cfq_slice_sync_show, cfqd->cfq_slice[1], 1); SHOW_FUNCTION(cfq_slice_async_show, cfqd->cfq_slice[0], 1); SHOW_FUNCTION(cfq_slice_async_rq_show, cfqd->cfq_slice_async_rq, 0); @@ -3975,6 +4014,7 @@ STORE_FUNCTION(cfq_back_seek_max_store, &cfqd->cfq_back_max, 0, UINT_MAX, 0); STORE_FUNCTION(cfq_back_seek_penalty_store, &cfqd->cfq_back_penalty, 1, UINT_MAX, 0); STORE_FUNCTION(cfq_slice_idle_store, &cfqd->cfq_slice_idle, 0, UINT_MAX, 1); +STORE_FUNCTION(cfq_group_idle_store, &cfqd->cfq_group_idle, 0, UINT_MAX, 1); STORE_FUNCTION(cfq_slice_sync_store, &cfqd->cfq_slice[1], 1, UINT_MAX, 1); STORE_FUNCTION(cfq_slice_async_store, &cfqd->cfq_slice[0], 1, UINT_MAX, 1); STORE_FUNCTION(cfq_slice_async_rq_store, &cfqd->cfq_slice_async_rq, 1, @@ -3996,6 +4036,7 @@ static struct elv_fs_entry cfq_attrs[] = { CFQ_ATTR(slice_async), CFQ_ATTR(slice_async_rq), CFQ_ATTR(slice_idle), + CFQ_ATTR(group_idle), CFQ_ATTR(low_latency), CFQ_ATTR(group_isolation), __ATTR_NULL @@ -4049,6 +4090,12 @@ static int __init cfq_init(void) if (!cfq_slice_idle) cfq_slice_idle = 1; +#ifdef CONFIG_CFQ_GROUP_IOSCHED + if (!cfq_group_idle) + cfq_group_idle = 1; +#else + cfq_group_idle = 0; +#endif if (cfq_slab_setup()) return -ENOMEM; -- cgit v0.10.2 From c4e7893ebc3a5c507b53f59b9de448db20849944 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Mon, 23 Aug 2010 12:25:03 +0200 Subject: cfq-iosched: blktrace print per slice sector stats o Divyesh had gotten rid of this code in the past. I want to re-introduce it back as it helps me a lot during debugging. Reviewed-by: Jeff Moyer Reviewed-by: Divyesh Shah Signed-off-by: Vivek Goyal Signed-off-by: Jens Axboe diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c index 85e4819..f65c6f0 100644 --- a/block/cfq-iosched.c +++ b/block/cfq-iosched.c @@ -148,6 +148,8 @@ struct cfq_queue { struct cfq_queue *new_cfqq; struct cfq_group *cfqg; struct cfq_group *orig_cfqg; + /* Number of sectors dispatched from queue in single dispatch round */ + unsigned long nr_sectors; }; /* @@ -960,8 +962,9 @@ static void cfq_group_served(struct cfq_data *cfqd, struct cfq_group *cfqg, cfq_log_cfqg(cfqd, cfqg, "served: vt=%llu min_vt=%llu", cfqg->vdisktime, st->min_vdisktime); - cfq_log_cfqq(cfqq->cfqd, cfqq, "sl_used=%u disp=%u charge=%u iops=%u", - used_sl, cfqq->slice_dispatch, charge, iops_mode(cfqd)); + cfq_log_cfqq(cfqq->cfqd, cfqq, "sl_used=%u disp=%u charge=%u iops=%u" + " sect=%u", used_sl, cfqq->slice_dispatch, charge, + iops_mode(cfqd), cfqq->nr_sectors); cfq_blkiocg_update_timeslice_used(&cfqg->blkg, used_sl); cfq_blkiocg_set_start_empty_time(&cfqg->blkg); } @@ -1609,6 +1612,7 @@ static void __cfq_set_active_queue(struct cfq_data *cfqd, cfqq->allocated_slice = 0; cfqq->slice_end = 0; cfqq->slice_dispatch = 0; + cfqq->nr_sectors = 0; cfq_clear_cfqq_wait_request(cfqq); cfq_clear_cfqq_must_dispatch(cfqq); @@ -1971,6 +1975,7 @@ static void cfq_dispatch_insert(struct request_queue *q, struct request *rq) elv_dispatch_sort(q, rq); cfqd->rq_in_flight[cfq_cfqq_sync(cfqq)]++; + cfqq->nr_sectors += blk_rq_sectors(rq); cfq_blkiocg_update_dispatch_stats(&cfqq->cfqg->blkg, blk_rq_bytes(rq), rq_data_dir(rq), rq_is_sync(rq)); } -- cgit v0.10.2 From 6d6ac1c1a3d4f95953aa3b085e8f16692d3a7179 Mon Sep 17 00:00:00 2001 From: Vivek Goyal Date: Mon, 23 Aug 2010 12:25:29 +0200 Subject: cfq-iosched: Documentation help for new tunables Some documentation to provide help with tunables. Signed-off-by: Vivek Goyal Acked-by: Jeff Moyer Signed-off-by: Jens Axboe diff --git a/Documentation/block/cfq-iosched.txt b/Documentation/block/cfq-iosched.txt new file mode 100644 index 0000000..e578fee --- /dev/null +++ b/Documentation/block/cfq-iosched.txt @@ -0,0 +1,45 @@ +CFQ ioscheduler tunables +======================== + +slice_idle +---------- +This specifies how long CFQ should idle for next request on certain cfq queues +(for sequential workloads) and service trees (for random workloads) before +queue is expired and CFQ selects next queue to dispatch from. + +By default slice_idle is a non-zero value. That means by default we idle on +queues/service trees. This can be very helpful on highly seeky media like +single spindle SATA/SAS disks where we can cut down on overall number of +seeks and see improved throughput. + +Setting slice_idle to 0 will remove all the idling on queues/service tree +level and one should see an overall improved throughput on faster storage +devices like multiple SATA/SAS disks in hardware RAID configuration. The down +side is that isolation provided from WRITES also goes down and notion of +IO priority becomes weaker. + +So depending on storage and workload, it might be useful to set slice_idle=0. +In general I think for SATA/SAS disks and software RAID of SATA/SAS disks +keeping slice_idle enabled should be useful. For any configurations where +there are multiple spindles behind single LUN (Host based hardware RAID +controller or for storage arrays), setting slice_idle=0 might end up in better +throughput and acceptable latencies. + +CFQ IOPS Mode for group scheduling +=================================== +Basic CFQ design is to provide priority based time slices. Higher priority +process gets bigger time slice and lower priority process gets smaller time +slice. Measuring time becomes harder if storage is fast and supports NCQ and +it would be better to dispatch multiple requests from multiple cfq queues in +request queue at a time. In such scenario, it is not possible to measure time +consumed by single queue accurately. + +What is possible though is to measure number of requests dispatched from a +single queue and also allow dispatch from multiple cfq queue at the same time. +This effectively becomes the fairness in terms of IOPS (IO operations per +second). + +If one sets slice_idle=0 and if storage supports NCQ, CFQ internally switches +to IOPS mode and starts providing fairness in terms of number of requests +dispatched. Note that this mode switching takes effect only for group +scheduling. For non-cgroup users nothing should change. diff --git a/Documentation/cgroups/blkio-controller.txt b/Documentation/cgroups/blkio-controller.txt index 48e0b21..6919d62 100644 --- a/Documentation/cgroups/blkio-controller.txt +++ b/Documentation/cgroups/blkio-controller.txt @@ -217,6 +217,7 @@ Details of cgroup files CFQ sysfs tunable ================= /sys/block//queue/iosched/group_isolation +----------------------------------------------- If group_isolation=1, it provides stronger isolation between groups at the expense of throughput. By default group_isolation is 0. In general that @@ -243,6 +244,33 @@ By default one should run with group_isolation=0. If that is not sufficient and one wants stronger isolation between groups, then set group_isolation=1 but this will come at cost of reduced throughput. +/sys/block//queue/iosched/slice_idle +------------------------------------------ +On a faster hardware CFQ can be slow, especially with sequential workload. +This happens because CFQ idles on a single queue and single queue might not +drive deeper request queue depths to keep the storage busy. In such scenarios +one can try setting slice_idle=0 and that would switch CFQ to IOPS +(IO operations per second) mode on NCQ supporting hardware. + +That means CFQ will not idle between cfq queues of a cfq group and hence be +able to driver higher queue depth and achieve better throughput. That also +means that cfq provides fairness among groups in terms of IOPS and not in +terms of disk time. + +/sys/block//queue/iosched/group_idle +------------------------------------------ +If one disables idling on individual cfq queues and cfq service trees by +setting slice_idle=0, group_idle kicks in. That means CFQ will still idle +on the group in an attempt to provide fairness among groups. + +By default group_idle is same as slice_idle and does not do anything if +slice_idle is enabled. + +One can experience an overall throughput drop if you have created multiple +groups and put applications in that group which are not driving enough +IO to keep disk busy. In that case set group_idle=0, and CFQ will not idle +on individual groups and throughput should improve. + What works ========== - Currently only sync IO queues are support. All the buffered writes are -- cgit v0.10.2 From 4ee69851cd4880f574d22f5ce08bec35b01c94e3 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 23 Aug 2010 12:28:15 +0200 Subject: cciss: handle allocation failure If kmalloc() fails then cleanup and return failure (-1). Signed-off-by: Dan Carpenter Acked-by: Stephen M. Cameron Signed-off-by: Jens Axboe diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c index ce1a75d..7191c16 100644 --- a/drivers/block/cciss.c +++ b/drivers/block/cciss.c @@ -4718,6 +4718,9 @@ static int __devinit cciss_init_one(struct pci_dev *pdev, h->scatter_list = kmalloc(h->max_commands * sizeof(struct scatterlist *), GFP_KERNEL); + if (!h->scatter_list) + goto clean4; + for (k = 0; k < h->nr_cmds; k++) { h->scatter_list[k] = kmalloc(sizeof(struct scatterlist) * h->maxsgentries, -- cgit v0.10.2 From c87ffbb812cf6150097a5095b031f4013a8f78a5 Mon Sep 17 00:00:00 2001 From: Xiaotian Feng Date: Mon, 23 Aug 2010 12:30:29 +0200 Subject: block: put dev->kobj in blk_register_queue fail path kernel needs to kobject_put on dev->kobj if elv_register_queue fails. Signed-off-by: Xiaotian Feng Cc: "Martin K. Petersen" Cc: Stephen Hemminger Cc: Nikanth Karthikesan Cc: David Teigland Signed-off-by: Jens Axboe diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c index 001ab18..0749b89 100644 --- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -511,6 +511,7 @@ int blk_register_queue(struct gendisk *disk) kobject_uevent(&q->kobj, KOBJ_REMOVE); kobject_del(&q->kobj); blk_trace_remove_sysfs(disk_to_dev(disk)); + kobject_put(&dev->kobj); return ret; } -- cgit v0.10.2 From 5e00d1b5b4c10fb839afd5ce61db8e24339454b0 Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Thu, 12 Aug 2010 14:31:06 +0200 Subject: BLOCK: fix bio.bi_rw handling Return of the bi_rw tests is no longer bool after commit 74450be1. But results of such tests are stored in bools. This doesn't fit in there for some compilers (gcc 4.5 here), so either use !! magic to get real bools or use ulong where the result is assigned somewhere. Signed-off-by: Jiri Slaby Cc: Christoph Hellwig Reviewed-by: Jeff Moyer Signed-off-by: Jens Axboe diff --git a/block/blk-core.c b/block/blk-core.c index ee1a1e7..32a1c12 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -1198,9 +1198,9 @@ static int __make_request(struct request_queue *q, struct bio *bio) int el_ret; unsigned int bytes = bio->bi_size; const unsigned short prio = bio_prio(bio); - const bool sync = (bio->bi_rw & REQ_SYNC); - const bool unplug = (bio->bi_rw & REQ_UNPLUG); - const unsigned int ff = bio->bi_rw & REQ_FAILFAST_MASK; + const bool sync = !!(bio->bi_rw & REQ_SYNC); + const bool unplug = !!(bio->bi_rw & REQ_UNPLUG); + const unsigned long ff = bio->bi_rw & REQ_FAILFAST_MASK; int rw_flags; if ((bio->bi_rw & REQ_HARDBARRIER) && diff --git a/drivers/block/loop.c b/drivers/block/loop.c index f3c636d..91797bb 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -477,7 +477,7 @@ static int do_bio_filebacked(struct loop_device *lo, struct bio *bio) pos = ((loff_t) bio->bi_sector << 9) + lo->lo_offset; if (bio_rw(bio) == WRITE) { - bool barrier = (bio->bi_rw & REQ_HARDBARRIER); + bool barrier = !!(bio->bi_rw & REQ_HARDBARRIER); struct file *file = lo->lo_backing_file; if (barrier) { -- cgit v0.10.2 From 72f465033702ebfe20db8f50edaad59f0f38b0f5 Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Mon, 23 Aug 2010 13:35:09 +0200 Subject: bio-integrity.c: remove dependency on __GFP_NOFAIL The kmalloc() in bio_integrity_prep() is failable, so remove __GFP_NOFAIL from its mask. Signed-off-by: David Rientjes Signed-off-by: Andrew Morton Signed-off-by: Jens Axboe diff --git a/fs/bio-integrity.c b/fs/bio-integrity.c index 612a5c3..a8f4cc6 100644 --- a/fs/bio-integrity.c +++ b/fs/bio-integrity.c @@ -413,7 +413,7 @@ int bio_integrity_prep(struct bio *bio) /* Allocate kernel buffer for protection data */ len = sectors * blk_integrity_tuple_size(bi); - buf = kmalloc(len, GFP_NOIO | __GFP_NOFAIL | q->bounce_gfp); + buf = kmalloc(len, GFP_NOIO | q->bounce_gfp); if (unlikely(buf == NULL)) { printk(KERN_ERR "could not allocate integrity buffer\n"); return -EIO; -- cgit v0.10.2 From 220eb7fd984bfc7e6b4005fdf32efe9cd8af7cf2 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 23 Aug 2010 13:36:20 +0200 Subject: fs/bio-integrity.c: return -ENOMEM on kmalloc failure Cc: David Rientjes Signed-off-by: Andrew Morton Signed-off-by: Jens Axboe diff --git a/fs/bio-integrity.c b/fs/bio-integrity.c index a8f4cc6..4d0ff5e 100644 --- a/fs/bio-integrity.c +++ b/fs/bio-integrity.c @@ -416,7 +416,7 @@ int bio_integrity_prep(struct bio *bio) buf = kmalloc(len, GFP_NOIO | q->bounce_gfp); if (unlikely(buf == NULL)) { printk(KERN_ERR "could not allocate integrity buffer\n"); - return -EIO; + return -ENOMEM; } end = (((unsigned long) buf) + len + PAGE_SIZE - 1) >> PAGE_SHIFT; -- cgit v0.10.2 From 5dd531a03ad721b41911ddb32e6e0481404e7aaf Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 23 Aug 2010 13:52:19 +0200 Subject: block: add function call to switch the IO scheduler from a driver Currently drivers must do an elevator_exit() + elevator_init() to switch IO schedulers. There are a few problems with this: - Since commit 1abec4fdbb142e3ccb6ce99832fae42129134a96, elevator_init() requires a zeroed out q->elevator pointer. The two existing in-kernel users don't do that. - It will only work at initialization time, since using the above two-staged construct does not properly quisce the queue. So add elevator_change() which takes care of this, and convert the elv_iosched_store() sysfs interface to use this helper as well. Reported-by: Peter Oberparleiter Reported-by: Kevin Vigor Signed-off-by: Jens Axboe diff --git a/block/elevator.c b/block/elevator.c index ec585c9..205b09a 100644 --- a/block/elevator.c +++ b/block/elevator.c @@ -1009,18 +1009,19 @@ static int elevator_switch(struct request_queue *q, struct elevator_type *new_e) { struct elevator_queue *old_elevator, *e; void *data; + int err; /* * Allocate new elevator */ e = elevator_alloc(q, new_e); if (!e) - return 0; + return -ENOMEM; data = elevator_init_queue(q, e); if (!data) { kobject_put(&e->kobj); - return 0; + return -ENOMEM; } /* @@ -1043,7 +1044,8 @@ static int elevator_switch(struct request_queue *q, struct elevator_type *new_e) __elv_unregister_queue(old_elevator); - if (elv_register_queue(q)) + err = elv_register_queue(q); + if (err) goto fail_register; /* @@ -1056,7 +1058,7 @@ static int elevator_switch(struct request_queue *q, struct elevator_type *new_e) blk_add_trace_msg(q, "elv switch: %s", e->elevator_type->elevator_name); - return 1; + return 0; fail_register: /* @@ -1071,17 +1073,19 @@ fail_register: queue_flag_clear(QUEUE_FLAG_ELVSWITCH, q); spin_unlock_irq(q->queue_lock); - return 0; + return err; } -ssize_t elv_iosched_store(struct request_queue *q, const char *name, - size_t count) +/* + * Switch this queue to the given IO scheduler. + */ +int elevator_change(struct request_queue *q, const char *name) { char elevator_name[ELV_NAME_MAX]; struct elevator_type *e; if (!q->elevator) - return count; + return -ENXIO; strlcpy(elevator_name, name, sizeof(elevator_name)); e = elevator_get(strstrip(elevator_name)); @@ -1092,13 +1096,27 @@ ssize_t elv_iosched_store(struct request_queue *q, const char *name, if (!strcmp(elevator_name, q->elevator->elevator_type->elevator_name)) { elevator_put(e); - return count; + return 0; } - if (!elevator_switch(q, e)) - printk(KERN_ERR "elevator: switch to %s failed\n", - elevator_name); - return count; + return elevator_switch(q, e); +} +EXPORT_SYMBOL(elevator_change); + +ssize_t elv_iosched_store(struct request_queue *q, const char *name, + size_t count) +{ + int ret; + + if (!q->elevator) + return count; + + ret = elevator_change(q, name); + if (!ret) + return count; + + printk(KERN_ERR "elevator: switch to %s failed\n", name); + return ret; } ssize_t elv_iosched_show(struct request_queue *q, char *name) diff --git a/include/linux/elevator.h b/include/linux/elevator.h index 2c958f4..926b503 100644 --- a/include/linux/elevator.h +++ b/include/linux/elevator.h @@ -136,6 +136,7 @@ extern ssize_t elv_iosched_store(struct request_queue *, const char *, size_t); extern int elevator_init(struct request_queue *, char *); extern void elevator_exit(struct elevator_queue *); +extern int elevator_change(struct request_queue *, const char *); extern int elv_rq_merge_ok(struct request *, struct bio *); /* -- cgit v0.10.2 From 52cc2eef31587b22ce9fbe77b064a031a9613ab0 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 23 Aug 2010 14:02:44 +0200 Subject: block: switch s390 tape_block and mg_disk to elevator_change() Now that we have this API, switch the two in-kernel users to it. Resolves an oops introduced by commit 1abec4fdbb142e3ccb6ce99832fae42129134a96. Signed-off-by: Jens Axboe diff --git a/drivers/block/mg_disk.c b/drivers/block/mg_disk.c index b82c5ce..76fa3de 100644 --- a/drivers/block/mg_disk.c +++ b/drivers/block/mg_disk.c @@ -974,8 +974,7 @@ static int mg_probe(struct platform_device *plat_dev) host->breq->queuedata = host; /* mflash is random device, thanx for the noop */ - elevator_exit(host->breq->elevator); - err = elevator_init(host->breq, "noop"); + err = elevator_change(host->breq, "noop"); if (err) { printk(KERN_ERR "%s:%d (elevator_init) fail\n", __func__, __LINE__); diff --git a/drivers/s390/char/tape_block.c b/drivers/s390/char/tape_block.c index b7de025..85cf607 100644 --- a/drivers/s390/char/tape_block.c +++ b/drivers/s390/char/tape_block.c @@ -217,8 +217,7 @@ tapeblock_setup_device(struct tape_device * device) if (!blkdat->request_queue) return -ENOMEM; - elevator_exit(blkdat->request_queue->elevator); - rc = elevator_init(blkdat->request_queue, "noop"); + rc = elevator_change(blkdat->request_queue, "noop"); if (rc) goto cleanup_queue; -- cgit v0.10.2 From e41e704bc4f49057fc68b643108366e6e6781aa3 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 24 Aug 2010 14:22:47 +0200 Subject: workqueue: improve destroy_workqueue() debuggability Now that the worklist is global, having works pending after wq destruction can easily lead to oops and destroy_workqueue() have several BUG_ON()s to catch these cases. Unfortunately, BUG_ON() doesn't tell much about how the work became pending after the final flush_workqueue(). This patch adds WQ_DYING which is set before the final flush begins. If a work is requested to be queued on a dying workqueue, WARN_ON_ONCE() is triggered and the request is ignored. This clearly indicates which caller is trying to queue a work on a dying workqueue and keeps the system working in most cases. Locking rule comment is updated such that the 'I' rule includes modifying the field from destruction path. Signed-off-by: Tejun Heo diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h index 4f9d277..c959666 100644 --- a/include/linux/workqueue.h +++ b/include/linux/workqueue.h @@ -241,6 +241,8 @@ enum { WQ_HIGHPRI = 1 << 4, /* high priority */ WQ_CPU_INTENSIVE = 1 << 5, /* cpu instensive workqueue */ + WQ_DYING = 1 << 6, /* internal: workqueue is dying */ + WQ_MAX_ACTIVE = 512, /* I like 512, better ideas? */ WQ_MAX_UNBOUND_PER_CPU = 4, /* 4 * #cpus for unbound wq */ WQ_DFL_ACTIVE = WQ_MAX_ACTIVE / 2, diff --git a/kernel/workqueue.c b/kernel/workqueue.c index cc3456f..362b50d 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -87,7 +87,8 @@ enum { /* * Structure fields follow one of the following exclusion rules. * - * I: Set during initialization and read-only afterwards. + * I: Modifiable by initialization/destruction paths and read-only for + * everyone else. * * P: Preemption protected. Disabling preemption is enough and should * only be modified and accessed from the local cpu. @@ -944,6 +945,9 @@ static void __queue_work(unsigned int cpu, struct workqueue_struct *wq, debug_work_activate(work); + if (WARN_ON_ONCE(wq->flags & WQ_DYING)) + return; + /* determine gcwq to use */ if (!(wq->flags & WQ_UNBOUND)) { struct global_cwq *last_gcwq; @@ -2828,6 +2832,7 @@ void destroy_workqueue(struct workqueue_struct *wq) { unsigned int cpu; + wq->flags |= WQ_DYING; flush_workqueue(wq); /* -- cgit v0.10.2 From f1a7bfaf6bb9cb195577e674c0ab2fd0a55d9014 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 21 Aug 2010 01:50:52 +0200 Subject: PCI: PCIe AER: Introduce pci_aer_available() Introduce a function allowing the caller to check whether to try to enable PCIe AER. Signed-off-by: Rafael J. Wysocki Signed-off-by: Jesse Barnes diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 679c39d..7754a67 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -140,8 +140,10 @@ static inline void pci_msi_init_pci_dev(struct pci_dev *dev) { } #ifdef CONFIG_PCIEAER void pci_no_aer(void); +bool pci_aer_available(void); #else static inline void pci_no_aer(void) { } +static inline bool pci_aer_available(void) { return false; } #endif static inline int pci_no_d1d2(struct pci_dev *dev) diff --git a/drivers/pci/pcie/aer/aerdrv.c b/drivers/pci/pcie/aer/aerdrv.c index 484cc55..f409948 100644 --- a/drivers/pci/pcie/aer/aerdrv.c +++ b/drivers/pci/pcie/aer/aerdrv.c @@ -72,6 +72,11 @@ void pci_no_aer(void) pcie_aer_disable = 1; /* has priority over 'forceload' */ } +bool pci_aer_available(void) +{ + return !pcie_aer_disable && pci_msi_enabled(); +} + static int set_device_error_reporting(struct pci_dev *dev, void *data) { bool enable = *((bool *)data); @@ -411,9 +416,7 @@ static void aer_error_resume(struct pci_dev *dev) */ static int __init aer_service_init(void) { - if (pcie_aer_disable) - return -ENXIO; - if (!pci_msi_enabled()) + if (!pci_aer_available()) return -ENXIO; return pcie_port_service_register(&aerdriver); } -- cgit v0.10.2 From 79dd9182db2072d63ccf160bb9a3463b1c952723 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 21 Aug 2010 01:51:44 +0200 Subject: PCI: PCIe: Introduce commad line switch for disabling port services Introduce kernel command line switch pcie_ports= allowing one to disable all of the native PCIe port services, so that PCIe ports are treated like PCI-to-PCI bridges. Signed-off-by: Rafael J. Wysocki Signed-off-by: Jesse Barnes diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 873b680..b9008fc 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -2042,6 +2042,10 @@ and is between 256 and 4096 characters. It is defined in the file force Enable ASPM even on devices that claim not to support it. WARNING: Forcing ASPM on may cause system lockups. + pcie_ports= [PCIE] PCIe ports handling: + compat Treat PCIe ports as PCI-to-PCI bridges, disable the PCIe + ports driver. + pcie_pme= [PCIE,PM] Native PCIe PME signaling options: Format: {auto|force}[,nomsi] auto Use native PCIe PME signaling if the BIOS allows the diff --git a/drivers/pci/pcie/portdrv.h b/drivers/pci/pcie/portdrv.h index 813a5c3..966f6e9 100644 --- a/drivers/pci/pcie/portdrv.h +++ b/drivers/pci/pcie/portdrv.h @@ -20,6 +20,8 @@ #define get_descriptor_id(type, service) (((type - 4) << 4) | service) +extern bool pcie_ports_disabled; + extern struct bus_type pcie_port_bus_type; extern int pcie_port_device_register(struct pci_dev *dev); #ifdef CONFIG_PM diff --git a/drivers/pci/pcie/portdrv_core.c b/drivers/pci/pcie/portdrv_core.c index e73effb..2bf2fe5 100644 --- a/drivers/pci/pcie/portdrv_core.c +++ b/drivers/pci/pcie/portdrv_core.c @@ -494,6 +494,9 @@ static void pcie_port_shutdown_service(struct device *dev) {} */ int pcie_port_service_register(struct pcie_port_service_driver *new) { + if (pcie_ports_disabled) + return -ENODEV; + new->driver.name = (char *)new->name; new->driver.bus = &pcie_port_bus_type; new->driver.probe = pcie_port_probe_service; diff --git a/drivers/pci/pcie/portdrv_pci.c b/drivers/pci/pcie/portdrv_pci.c index 3debed2..a04392d 100644 --- a/drivers/pci/pcie/portdrv_pci.c +++ b/drivers/pci/pcie/portdrv_pci.c @@ -29,6 +29,18 @@ MODULE_AUTHOR(DRIVER_AUTHOR); MODULE_DESCRIPTION(DRIVER_DESC); MODULE_LICENSE("GPL"); +/* If this switch is set, PCIe port native services should not be enabled. */ +bool pcie_ports_disabled; + +static int __init pcie_port_setup(char *str) +{ + if (!strncmp(str, "compat", 6)) + pcie_ports_disabled = true; + + return 1; +} +__setup("pcie_ports=", pcie_port_setup); + /* global data */ static int pcie_portdrv_restore_config(struct pci_dev *dev) @@ -301,6 +313,9 @@ static int __init pcie_portdrv_init(void) { int retval; + if (pcie_ports_disabled) + return -EACCES; + dmi_check_system(pcie_portdrv_dmi_table); retval = pcie_port_bus_register(); -- cgit v0.10.2 From b879dc4b3e81069e3f715b7569bb0f43eed76c76 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 21 Aug 2010 01:52:37 +0200 Subject: ACPI/PCI: Reorder checks in acpi_pci_osc_control_set() Make acpi_pci_osc_control_set() attempt to find the handle of the _OSC object under the given PCI root bridge object after verifying that its second argument is correct and that there is a struct acpi_pci_root object for the given root bridge handle, which is more logical than the old code. Signed-off-by: Rafael J. Wysocki Reviewed-by: Hidetoshi Seto Signed-off-by: Jesse Barnes diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c index 1f67057..e10dbaf 100644 --- a/drivers/acpi/pci_root.c +++ b/drivers/acpi/pci_root.c @@ -378,10 +378,6 @@ acpi_status acpi_pci_osc_control_set(acpi_handle handle, u32 flags) acpi_handle tmp; struct acpi_pci_root *root; - status = acpi_get_handle(handle, "_OSC", &tmp); - if (ACPI_FAILURE(status)) - return status; - control_req = (flags & OSC_PCI_CONTROL_MASKS); if (!control_req) return AE_TYPE; @@ -390,6 +386,10 @@ acpi_status acpi_pci_osc_control_set(acpi_handle handle, u32 flags) if (!root) return AE_NOT_EXIST; + status = acpi_get_handle(handle, "_OSC", &tmp); + if (ACPI_FAILURE(status)) + return status; + mutex_lock(&osc_lock); /* No need to evaluate _OSC if the control was already granted. */ if ((root->osc_control_set & control_req) == control_req) -- cgit v0.10.2 From ab8e8957a2ae21c0f036476c6db13e949be730ac Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 21 Aug 2010 01:53:27 +0200 Subject: ACPI/PCI: Make acpi_pci_query_osc() return control bits Make acpi_pci_query_osc() use an additional pointer argument to return the mask of control bits obtained from the BIOS to the caller. Signed-off-by: Rafael J. Wysocki Signed-off-by: Jesse Barnes diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c index e10dbaf..d2ae816 100644 --- a/drivers/acpi/pci_root.c +++ b/drivers/acpi/pci_root.c @@ -226,22 +226,35 @@ static acpi_status acpi_pci_run_osc(acpi_handle handle, return status; } -static acpi_status acpi_pci_query_osc(struct acpi_pci_root *root, u32 flags) +static acpi_status acpi_pci_query_osc(struct acpi_pci_root *root, + u32 support, + u32 *control) { acpi_status status; - u32 support_set, result, capbuf[3]; + u32 result, capbuf[3]; + + support &= OSC_PCI_SUPPORT_MASKS; + support |= root->osc_support_set; - /* do _OSC query for all possible controls */ - support_set = root->osc_support_set | (flags & OSC_PCI_SUPPORT_MASKS); capbuf[OSC_QUERY_TYPE] = OSC_QUERY_ENABLE; - capbuf[OSC_SUPPORT_TYPE] = support_set; - capbuf[OSC_CONTROL_TYPE] = OSC_PCI_CONTROL_MASKS; + capbuf[OSC_SUPPORT_TYPE] = support; + if (control) { + *control &= OSC_PCI_CONTROL_MASKS; + capbuf[OSC_CONTROL_TYPE] = *control | root->osc_control_set; + } else { + /* Run _OSC query for all possible controls. */ + capbuf[OSC_CONTROL_TYPE] = OSC_PCI_CONTROL_MASKS; + } status = acpi_pci_run_osc(root->device->handle, capbuf, &result); if (ACPI_SUCCESS(status)) { - root->osc_support_set = support_set; - root->osc_control_qry = result; - root->osc_queried = 1; + root->osc_support_set = support; + if (control) { + *control = result; + } else { + root->osc_control_qry = result; + root->osc_queried = 1; + } } return status; } @@ -255,7 +268,7 @@ static acpi_status acpi_pci_osc_support(struct acpi_pci_root *root, u32 flags) if (ACPI_FAILURE(status)) return status; mutex_lock(&osc_lock); - status = acpi_pci_query_osc(root, flags); + status = acpi_pci_query_osc(root, flags, NULL); mutex_unlock(&osc_lock); return status; } @@ -397,7 +410,7 @@ acpi_status acpi_pci_osc_control_set(acpi_handle handle, u32 flags) /* Need to query controls first before requesting them */ if (!root->osc_queried) { - status = acpi_pci_query_osc(root, root->osc_support_set); + status = acpi_pci_query_osc(root, root->osc_support_set, NULL); if (ACPI_FAILURE(status)) goto out; } -- cgit v0.10.2 From 2b8fd9186d9275b07aef43e5bb4e98cd571f9a7d Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 23 Aug 2010 23:55:59 +0200 Subject: ACPI/PCI: Do not preserve _OSC control bits returned by a query There is the assumption in acpi_pci_osc_control_set() that it is always sufficient to compare the mask of _OSC control bits to be requested with the result of an _OSC query where all of the known control bits have been checked. However, in general, that need not be the case. For example, if an _OSC feature A depends on an _OSC feature B and control of A, B plus another _OSC feature C is requested simultaneously, the BIOS may return A, B, C, while it would only return C if A and C were requested without B. That may result in passing a wrong mask of _OSC control bits to an _OSC control request, in which case the BIOS may only grant control of a subset of the requested features. Moreover, acpi_pci_run_osc() will return error code if that happens and the caller of acpi_pci_osc_control_set() will not know that it's been granted control of some _OSC features. Consequently, the system will generally not work as expected. Apart from this acpi_pci_osc_control_set() always uses the mask of _OSC control bits returned by the very first invocation of acpi_pci_query_osc(), but that is done with the second argument equal to OSC_PCI_SEGMENT_GROUPS_SUPPORT which generally happens to affect the returned _OSC control bits. For these reasons, make acpi_pci_osc_control_set() always check if control of the requested _OSC features will be granted before making the final control request. As a result, the osc_control_qry and osc_queried members of struct acpi_pci_root are not necessary any more, so drop them and remove the remaining code referring to them. Signed-off-by: Rafael J. Wysocki Signed-off-by: Jesse Barnes diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c index d2ae816..77cd196 100644 --- a/drivers/acpi/pci_root.c +++ b/drivers/acpi/pci_root.c @@ -249,12 +249,8 @@ static acpi_status acpi_pci_query_osc(struct acpi_pci_root *root, status = acpi_pci_run_osc(root->device->handle, capbuf, &result); if (ACPI_SUCCESS(status)) { root->osc_support_set = support; - if (control) { + if (control) *control = result; - } else { - root->osc_control_qry = result; - root->osc_queried = 1; - } } return status; } @@ -409,14 +405,12 @@ acpi_status acpi_pci_osc_control_set(acpi_handle handle, u32 flags) goto out; /* Need to query controls first before requesting them */ - if (!root->osc_queried) { - status = acpi_pci_query_osc(root, root->osc_support_set, NULL); - if (ACPI_FAILURE(status)) - goto out; - } - if ((root->osc_control_qry & control_req) != control_req) { - printk(KERN_DEBUG - "Firmware did not grant requested _OSC control\n"); + flags = control_req; + status = acpi_pci_query_osc(root, root->osc_support_set, &flags); + if (ACPI_FAILURE(status)) + goto out; + + if (flags != control_req) { status = AE_SUPPORT; goto out; } diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h index baacd98..4de84ce 100644 --- a/include/acpi/acpi_bus.h +++ b/include/acpi/acpi_bus.h @@ -377,9 +377,6 @@ struct acpi_pci_root { u32 osc_support_set; /* _OSC state of support bits */ u32 osc_control_set; /* _OSC state of control bits */ - u32 osc_control_qry; /* the latest _OSC query result */ - - u32 osc_queried:1; /* has _OSC control been queried? */ }; /* helper */ -- cgit v0.10.2 From 75fb60f26befb59dbfa05cb122972642b7bdd219 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 23 Aug 2010 23:53:11 +0200 Subject: ACPI/PCI: Negotiate _OSC control bits before requesting them It is possible that the BIOS will not grant control of all _OSC features requested via acpi_pci_osc_control_set(), so it is recommended to negotiate the final set of _OSC features with the query flag set before calling _OSC to request control of these features. To implement it, rework acpi_pci_osc_control_set() so that the caller can specify the mask of _OSC control bits to negotiate and the mask of _OSC control bits that are absolutely necessary to it. Then, acpi_pci_osc_control_set() will run _OSC queries in a loop until the mask of _OSC control bits returned by the BIOS is equal to the mask passed to it. Also, before running the _OSC request acpi_pci_osc_control_set() will check if the caller's required control bits are present in the final mask. Using this mechanism we will be able to avoid situations in which the BIOS doesn't grant control of certain _OSC features, because they depend on some other _OSC features that have not been requested. Signed-off-by: Rafael J. Wysocki Signed-off-by: Jesse Barnes diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c index 77cd196..c347131 100644 --- a/drivers/acpi/pci_root.c +++ b/drivers/acpi/pci_root.c @@ -374,21 +374,32 @@ out: EXPORT_SYMBOL_GPL(acpi_get_pci_dev); /** - * acpi_pci_osc_control_set - commit requested control to Firmware - * @handle: acpi_handle for the target ACPI object - * @flags: driver's requested control bits + * acpi_pci_osc_control_set - Request control of PCI root _OSC features. + * @handle: ACPI handle of a PCI root bridge (or PCIe Root Complex). + * @mask: Mask of _OSC bits to request control of, place to store control mask. + * @req: Mask of _OSC bits the control of is essential to the caller. * - * Attempt to take control from Firmware on requested control bits. + * Run _OSC query for @mask and if that is successful, compare the returned + * mask of control bits with @req. If all of the @req bits are set in the + * returned mask, run _OSC request for it. + * + * The variable at the @mask address may be modified regardless of whether or + * not the function returns success. On success it will contain the mask of + * _OSC bits the BIOS has granted control of, but its contents are meaningless + * on failure. **/ -acpi_status acpi_pci_osc_control_set(acpi_handle handle, u32 flags) +acpi_status acpi_pci_osc_control_set(acpi_handle handle, u32 *mask, u32 req) { + struct acpi_pci_root *root; acpi_status status; - u32 control_req, result, capbuf[3]; + u32 ctrl, capbuf[3]; acpi_handle tmp; - struct acpi_pci_root *root; - control_req = (flags & OSC_PCI_CONTROL_MASKS); - if (!control_req) + if (!mask) + return AE_BAD_PARAMETER; + + ctrl = *mask & OSC_PCI_CONTROL_MASKS; + if ((ctrl & req) != req) return AE_TYPE; root = acpi_pci_find_root(handle); @@ -400,27 +411,33 @@ acpi_status acpi_pci_osc_control_set(acpi_handle handle, u32 flags) return status; mutex_lock(&osc_lock); + + *mask = ctrl | root->osc_control_set; /* No need to evaluate _OSC if the control was already granted. */ - if ((root->osc_control_set & control_req) == control_req) + if ((root->osc_control_set & ctrl) == ctrl) goto out; - /* Need to query controls first before requesting them */ - flags = control_req; - status = acpi_pci_query_osc(root, root->osc_support_set, &flags); - if (ACPI_FAILURE(status)) - goto out; + /* Need to check the available controls bits before requesting them. */ + while (*mask) { + status = acpi_pci_query_osc(root, root->osc_support_set, mask); + if (ACPI_FAILURE(status)) + goto out; + if (ctrl == *mask) + break; + ctrl = *mask; + } - if (flags != control_req) { + if ((ctrl & req) != req) { status = AE_SUPPORT; goto out; } capbuf[OSC_QUERY_TYPE] = 0; capbuf[OSC_SUPPORT_TYPE] = root->osc_support_set; - capbuf[OSC_CONTROL_TYPE] = root->osc_control_set | control_req; - status = acpi_pci_run_osc(handle, capbuf, &result); + capbuf[OSC_CONTROL_TYPE] = ctrl; + status = acpi_pci_run_osc(handle, capbuf, mask); if (ACPI_SUCCESS(status)) - root->osc_control_set = result; + root->osc_control_set = *mask; out: mutex_unlock(&osc_lock); return status; @@ -551,8 +568,8 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device) if (flags != base_flags) acpi_pci_osc_support(root, flags); - status = acpi_pci_osc_control_set(root->device->handle, - OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL); + flags = OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL; + status = acpi_pci_osc_control_set(root->device->handle, &flags, flags); if (ACPI_FAILURE(status)) { printk(KERN_INFO "Unable to assume PCIe control: Disabling ASPM\n"); diff --git a/drivers/pci/hotplug/acpi_pcihp.c b/drivers/pci/hotplug/acpi_pcihp.c index 45fcc1e..3d93d52 100644 --- a/drivers/pci/hotplug/acpi_pcihp.c +++ b/drivers/pci/hotplug/acpi_pcihp.c @@ -360,7 +360,7 @@ int acpi_get_hp_hw_control_from_firmware(struct pci_dev *pdev, u32 flags) acpi_get_name(handle, ACPI_FULL_PATHNAME, &string); dbg("Trying to get hotplug control for %s\n", (char *)string.pointer); - status = acpi_pci_osc_control_set(handle, flags); + status = acpi_pci_osc_control_set(handle, &flags, flags); if (ACPI_SUCCESS(status)) goto got_one; if (status == AE_SUPPORT) diff --git a/drivers/pci/pcie/aer/aerdrv_acpi.c b/drivers/pci/pcie/aer/aerdrv_acpi.c index f278d7b..3a276a0 100644 --- a/drivers/pci/pcie/aer/aerdrv_acpi.c +++ b/drivers/pci/pcie/aer/aerdrv_acpi.c @@ -39,9 +39,9 @@ int aer_osc_setup(struct pcie_device *pciedev) handle = acpi_find_root_bridge_handle(pdev); if (handle) { - status = acpi_pci_osc_control_set(handle, - OSC_PCI_EXPRESS_AER_CONTROL | - OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL); + u32 flags = OSC_PCI_EXPRESS_AER_CONTROL | + OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL; + status = acpi_pci_osc_control_set(handle, &flags, flags); } if (ACPI_FAILURE(status)) { diff --git a/drivers/pci/pcie/pme/pcie_pme_acpi.c b/drivers/pci/pcie/pme/pcie_pme_acpi.c index 83ab228..be20222 100644 --- a/drivers/pci/pcie/pme/pcie_pme_acpi.c +++ b/drivers/pci/pcie/pme/pcie_pme_acpi.c @@ -28,6 +28,7 @@ int pcie_pme_acpi_setup(struct pcie_device *srv) acpi_status status = AE_NOT_FOUND; struct pci_dev *port = srv->port; acpi_handle handle; + u32 flags; int error = 0; if (acpi_pci_disabled) @@ -39,9 +40,10 @@ int pcie_pme_acpi_setup(struct pcie_device *srv) if (!handle) return -EINVAL; - status = acpi_pci_osc_control_set(handle, - OSC_PCI_EXPRESS_PME_CONTROL | - OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL); + flags = OSC_PCI_EXPRESS_PME_CONTROL | + OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL; + + status = acpi_pci_osc_control_set(handle, &flags, flags); if (ACPI_FAILURE(status)) { dev_info(&port->dev, "Failed to receive control of PCIe PME service: %s\n", diff --git a/include/linux/acpi.h b/include/linux/acpi.h index ccf94dc..c227757 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -304,8 +304,8 @@ acpi_status acpi_run_osc(acpi_handle handle, struct acpi_osc_context *context); OSC_PCI_EXPRESS_PME_CONTROL | \ OSC_PCI_EXPRESS_AER_CONTROL | \ OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL) - -extern acpi_status acpi_pci_osc_control_set(acpi_handle handle, u32 flags); +extern acpi_status acpi_pci_osc_control_set(acpi_handle handle, + u32 *mask, u32 req); extern void acpi_early_init(void); #else /* !CONFIG_ACPI */ -- cgit v0.10.2 From 28eb5f274a305bf3a13b2c80c4804d4515d05c64 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 21 Aug 2010 22:02:38 +0200 Subject: PCI: PCIe: Ask BIOS for control of all native services at once After commit 852972acff8f10f3a15679be2059bb94916cba5d (ACPI: Disable ASPM if the platform won't provide _OSC control for PCIe) control of the PCIe Capability Structure is unconditionally requested by acpi_pci_root_add(), which in principle may cause problems to happen in two ways. First, the BIOS may refuse to give control of the PCIe Capability Structure if it is not asked for any of the _OSC features depending on it at the same time. Second, the BIOS may assume that control of the _OSC features depending on the PCIe Capability Structure will be requested in the future and may behave incorrectly if that doesn't happen. For this reason, control of the PCIe Capability Structure should always be requested along with control of any other _OSC features that may depend on it (ie. PCIe native PME, PCIe native hot-plug, PCIe AER). Rework the PCIe port driver so that (1) it checks which native PCIe port services can be enabled, according to the BIOS, and (2) it requests control of all these services simultaneously. In particular, this causes pcie_portdrv_probe() to fail if the BIOS refuses to grant control of the PCIe Capability Structure, which means that no native PCIe port services can be enabled for the PCIe Root Complex the given port belongs to. If that happens, ASPM is disabled to avoid problems with mishandling it by the part of the PCIe hierarchy for which control of the PCIe Capability Structure has not been received. Make it possible to override this behavior using 'pcie_ports=native' (use the PCIe native services regardless of the BIOS response to the control request), or 'pcie_ports=compat' (do not use the PCIe native services at all). Accordingly, rework the existing PCIe port service drivers so that they don't request control of the services directly. Signed-off-by: Rafael J. Wysocki Signed-off-by: Jesse Barnes diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index b9008fc..8f126e8 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -2043,18 +2043,17 @@ and is between 256 and 4096 characters. It is defined in the file WARNING: Forcing ASPM on may cause system lockups. pcie_ports= [PCIE] PCIe ports handling: + auto Ask the BIOS whether or not to use native PCIe services + associated with PCIe ports (PME, hot-plug, AER). Use + them only if that is allowed by the BIOS. + native Use native PCIe services associated with PCIe ports + unconditionally. compat Treat PCIe ports as PCI-to-PCI bridges, disable the PCIe ports driver. pcie_pme= [PCIE,PM] Native PCIe PME signaling options: - Format: {auto|force}[,nomsi] - auto Use native PCIe PME signaling if the BIOS allows the - kernel to control PCIe config registers of root ports. - force Use native PCIe PME signaling even if the BIOS refuses - to allow the kernel to control the relevant PCIe config - registers. nomsi Do not use MSI for native PCIe PME signaling (this makes - all PCIe root ports use INTx for everything). + all PCIe root ports use INTx for all services). pcmv= [HW,PCMCIA] BadgePAD 4 diff --git a/drivers/acpi/pci_root.c b/drivers/acpi/pci_root.c index c347131..3ba8d1f 100644 --- a/drivers/acpi/pci_root.c +++ b/drivers/acpi/pci_root.c @@ -33,7 +33,6 @@ #include #include #include -#include #include #include #include @@ -568,14 +567,6 @@ static int __devinit acpi_pci_root_add(struct acpi_device *device) if (flags != base_flags) acpi_pci_osc_support(root, flags); - flags = OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL; - status = acpi_pci_osc_control_set(root->device->handle, &flags, flags); - - if (ACPI_FAILURE(status)) { - printk(KERN_INFO "Unable to assume PCIe control: Disabling ASPM\n"); - pcie_no_aspm(); - } - pci_acpi_add_bus_pm_notifier(device, root->bus); if (device->wakeup.flags.run_wake) device_set_run_wake(root->bus->bridge, true); diff --git a/drivers/pci/hotplug/acpi_pcihp.c b/drivers/pci/hotplug/acpi_pcihp.c index 3d93d52..3bc72d1 100644 --- a/drivers/pci/hotplug/acpi_pcihp.c +++ b/drivers/pci/hotplug/acpi_pcihp.c @@ -338,9 +338,7 @@ int acpi_get_hp_hw_control_from_firmware(struct pci_dev *pdev, u32 flags) acpi_handle chandle, handle; struct acpi_buffer string = { ACPI_ALLOCATE_BUFFER, NULL }; - flags &= (OSC_PCI_EXPRESS_NATIVE_HP_CONTROL | - OSC_SHPC_NATIVE_HP_CONTROL | - OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL); + flags &= OSC_SHPC_NATIVE_HP_CONTROL; if (!flags) { err("Invalid flags %u specified!\n", flags); return -EINVAL; diff --git a/drivers/pci/hotplug/pciehp.h b/drivers/pci/hotplug/pciehp.h index 4ed76b4..653de6f 100644 --- a/drivers/pci/hotplug/pciehp.h +++ b/drivers/pci/hotplug/pciehp.h @@ -176,19 +176,7 @@ static inline void pciehp_firmware_init(void) { pciehp_acpi_slot_detection_init(); } - -static inline int pciehp_get_hp_hw_control_from_firmware(struct pci_dev *dev) -{ - int retval; - u32 flags = (OSC_PCI_EXPRESS_NATIVE_HP_CONTROL | - OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL); - retval = acpi_get_hp_hw_control_from_firmware(dev, flags); - if (retval) - return retval; - return pciehp_acpi_slot_detection_check(dev); -} #else #define pciehp_firmware_init() do {} while (0) -#define pciehp_get_hp_hw_control_from_firmware(dev) 0 #endif /* CONFIG_ACPI */ #endif /* _PCIEHP_H */ diff --git a/drivers/pci/hotplug/pciehp_acpi.c b/drivers/pci/hotplug/pciehp_acpi.c index 1f4000a..2574700 100644 --- a/drivers/pci/hotplug/pciehp_acpi.c +++ b/drivers/pci/hotplug/pciehp_acpi.c @@ -85,9 +85,7 @@ static int __init dummy_probe(struct pcie_device *dev) acpi_handle handle; struct dummy_slot *slot, *tmp; struct pci_dev *pdev = dev->port; - /* Note: pciehp_detect_mode != PCIEHP_DETECT_ACPI here */ - if (pciehp_get_hp_hw_control_from_firmware(pdev)) - return -ENODEV; + pos = pci_pcie_cap(pdev); if (!pos) return -ENODEV; diff --git a/drivers/pci/hotplug/pciehp_core.c b/drivers/pci/hotplug/pciehp_core.c index 3588ea6..aa5f3ff 100644 --- a/drivers/pci/hotplug/pciehp_core.c +++ b/drivers/pci/hotplug/pciehp_core.c @@ -59,7 +59,7 @@ module_param(pciehp_force, bool, 0644); MODULE_PARM_DESC(pciehp_debug, "Debugging mode enabled or not"); MODULE_PARM_DESC(pciehp_poll_mode, "Using polling mechanism for hot-plug events or not"); MODULE_PARM_DESC(pciehp_poll_time, "Polling mechanism frequency, in seconds"); -MODULE_PARM_DESC(pciehp_force, "Force pciehp, even if _OSC and OSHP are missing"); +MODULE_PARM_DESC(pciehp_force, "Force pciehp, even if OSHP is missing"); #define PCIE_MODULE_NAME "pciehp" @@ -235,7 +235,7 @@ static int pciehp_probe(struct pcie_device *dev) dev_info(&dev->device, "Bypassing BIOS check for pciehp use on %s\n", pci_name(dev->port)); - else if (pciehp_get_hp_hw_control_from_firmware(dev->port)) + else if (pciehp_acpi_slot_detection_check(dev->port)) goto err_out_none; ctrl = pcie_init(dev); diff --git a/drivers/pci/pcie/Makefile b/drivers/pci/pcie/Makefile index ea65454..4d2b187 100644 --- a/drivers/pci/pcie/Makefile +++ b/drivers/pci/pcie/Makefile @@ -6,6 +6,7 @@ obj-$(CONFIG_PCIEASPM) += aspm.o pcieportdrv-y := portdrv_core.o portdrv_pci.o portdrv_bus.o +pcieportdrv-$(CONFIG_ACPI) += portdrv_acpi.o obj-$(CONFIG_PCIEPORTBUS) += pcieportdrv.o diff --git a/drivers/pci/pcie/aer/aerdrv_acpi.c b/drivers/pci/pcie/aer/aerdrv_acpi.c index 3a276a0..2bb9b89 100644 --- a/drivers/pci/pcie/aer/aerdrv_acpi.c +++ b/drivers/pci/pcie/aer/aerdrv_acpi.c @@ -19,42 +19,6 @@ #include #include "aerdrv.h" -/** - * aer_osc_setup - run ACPI _OSC method - * @pciedev: pcie_device which AER is being enabled on - * - * @return: Zero on success. Nonzero otherwise. - * - * Invoked when PCIe bus loads AER service driver. To avoid conflict with - * BIOS AER support requires BIOS to yield AER control to OS native driver. - **/ -int aer_osc_setup(struct pcie_device *pciedev) -{ - acpi_status status = AE_NOT_FOUND; - struct pci_dev *pdev = pciedev->port; - acpi_handle handle = NULL; - - if (acpi_pci_disabled) - return -1; - - handle = acpi_find_root_bridge_handle(pdev); - if (handle) { - u32 flags = OSC_PCI_EXPRESS_AER_CONTROL | - OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL; - status = acpi_pci_osc_control_set(handle, &flags, flags); - } - - if (ACPI_FAILURE(status)) { - dev_printk(KERN_DEBUG, &pciedev->device, "AER service couldn't " - "init device: %s\n", - (status == AE_SUPPORT || status == AE_NOT_FOUND) ? - "no _OSC support" : "_OSC failed"); - return -1; - } - - return 0; -} - #ifdef CONFIG_ACPI_APEI static inline int hest_match_pci(struct acpi_hest_aer_common *p, struct pci_dev *pci) diff --git a/drivers/pci/pcie/aer/aerdrv_core.c b/drivers/pci/pcie/aer/aerdrv_core.c index fc0b5a9..29e268f 100644 --- a/drivers/pci/pcie/aer/aerdrv_core.c +++ b/drivers/pci/pcie/aer/aerdrv_core.c @@ -772,22 +772,10 @@ void aer_isr(struct work_struct *work) */ int aer_init(struct pcie_device *dev) { - if (pcie_aer_get_firmware_first(dev->port)) { - dev_printk(KERN_DEBUG, &dev->device, - "PCIe errors handled by platform firmware.\n"); - goto out; - } - - if (aer_osc_setup(dev)) - goto out; - - return 0; -out: if (forceload) { dev_printk(KERN_DEBUG, &dev->device, "aerdrv forceload requested.\n"); pcie_aer_force_firmware_first(dev->port, 0); - return 0; } - return -ENXIO; + return 0; } diff --git a/drivers/pci/pcie/pme/Makefile b/drivers/pci/pcie/pme/Makefile index 8b92380..3c67bf4 100644 --- a/drivers/pci/pcie/pme/Makefile +++ b/drivers/pci/pcie/pme/Makefile @@ -2,7 +2,4 @@ # Makefile for PCI-Express Root Port PME signaling driver # -obj-$(CONFIG_PCIE_PME) += pmedriver.o - -pmedriver-objs := pcie_pme.o -pmedriver-$(CONFIG_ACPI) += pcie_pme_acpi.o +obj-$(CONFIG_PCIE_PME) += pcie_pme.o diff --git a/drivers/pci/pcie/pme/pcie_pme.c b/drivers/pci/pcie/pme/pcie_pme.c index bbdea18..9d1361e 100644 --- a/drivers/pci/pcie/pme/pcie_pme.c +++ b/drivers/pci/pcie/pme/pcie_pme.c @@ -24,37 +24,12 @@ #include #include "../../pci.h" -#include "pcie_pme.h" +#include "../portdrv.h" #define PCI_EXP_RTSTA_PME 0x10000 /* PME status */ #define PCI_EXP_RTSTA_PENDING 0x20000 /* PME pending */ /* - * If set, this switch will prevent the PCIe root port PME service driver from - * being registered. Consequently, the interrupt-based PCIe PME signaling will - * not be used by any PCIe root ports in that case. - */ -static bool pcie_pme_disabled = true; - -/* - * The PCI Express Base Specification 2.0, Section 6.1.8, states the following: - * "In order to maintain compatibility with non-PCI Express-aware system - * software, system power management logic must be configured by firmware to use - * the legacy mechanism of signaling PME by default. PCI Express-aware system - * software must notify the firmware prior to enabling native, interrupt-based - * PME signaling." However, if the platform doesn't provide us with a suitable - * notification mechanism or the notification fails, it is not clear whether or - * not we are supposed to use the interrupt-based PCIe PME signaling. The - * switch below can be used to indicate the desired behaviour. When set, it - * will make the kernel use the interrupt-based PCIe PME signaling regardless of - * the platform notification status, although the kernel will attempt to notify - * the platform anyway. When unset, it will prevent the kernel from using the - * the interrupt-based PCIe PME signaling if the platform notification fails, - * which is the default. - */ -static bool pcie_pme_force_enable; - -/* * If this switch is set, MSI will not be used for PCIe PME signaling. This * causes the PCIe port driver to use INTx interrupts only, but it turns out * that using MSI for PCIe PME signaling doesn't play well with PCIe PME-based @@ -64,38 +39,13 @@ bool pcie_pme_msi_disabled; static int __init pcie_pme_setup(char *str) { - if (!strncmp(str, "auto", 4)) - pcie_pme_disabled = false; - else if (!strncmp(str, "force", 5)) - pcie_pme_force_enable = true; - - str = strchr(str, ','); - if (str) { - str++; - str += strspn(str, " \t"); - if (*str && !strcmp(str, "nomsi")) - pcie_pme_msi_disabled = true; - } + if (!strncmp(str, "nomsi", 5)) + pcie_pme_msi_disabled = true; return 1; } __setup("pcie_pme=", pcie_pme_setup); -/** - * pcie_pme_platform_setup - Ensure that the kernel controls the PCIe PME. - * @srv: PCIe PME root port service to use for carrying out the check. - * - * Notify the platform that the native PCIe PME is going to be used and return - * 'true' if the control of the PCIe PME registers has been acquired from the - * platform. - */ -static bool pcie_pme_platform_setup(struct pcie_device *srv) -{ - if (!pcie_pme_platform_notify(srv)) - return true; - return pcie_pme_force_enable; -} - struct pcie_pme_service_data { spinlock_t lock; struct pcie_device *srv; @@ -108,7 +58,7 @@ struct pcie_pme_service_data { * @dev: PCIe root port or event collector. * @enable: Enable or disable the interrupt. */ -static void pcie_pme_interrupt_enable(struct pci_dev *dev, bool enable) +void pcie_pme_interrupt_enable(struct pci_dev *dev, bool enable) { int rtctl_pos; u16 rtctl; @@ -417,9 +367,6 @@ static int pcie_pme_probe(struct pcie_device *srv) struct pcie_pme_service_data *data; int ret; - if (!pcie_pme_platform_setup(srv)) - return -EACCES; - data = kzalloc(sizeof(*data), GFP_KERNEL); if (!data) return -ENOMEM; @@ -509,8 +456,7 @@ static struct pcie_port_service_driver pcie_pme_driver = { */ static int __init pcie_pme_service_init(void) { - return pcie_pme_disabled ? - -ENODEV : pcie_port_service_register(&pcie_pme_driver); + return pcie_port_service_register(&pcie_pme_driver); } module_init(pcie_pme_service_init); diff --git a/drivers/pci/pcie/pme/pcie_pme.h b/drivers/pci/pcie/pme/pcie_pme.h deleted file mode 100644 index b30d2b7..0000000 --- a/drivers/pci/pcie/pme/pcie_pme.h +++ /dev/null @@ -1,28 +0,0 @@ -/* - * drivers/pci/pcie/pme/pcie_pme.h - * - * PCI Express Root Port PME signaling support - * - * Copyright (C) 2009 Rafael J. Wysocki , Novell Inc. - */ - -#ifndef _PCIE_PME_H_ -#define _PCIE_PME_H_ - -struct pcie_device; - -#ifdef CONFIG_ACPI -extern int pcie_pme_acpi_setup(struct pcie_device *srv); - -static inline int pcie_pme_platform_notify(struct pcie_device *srv) -{ - return pcie_pme_acpi_setup(srv); -} -#else /* !CONFIG_ACPI */ -static inline int pcie_pme_platform_notify(struct pcie_device *srv) -{ - return 0; -} -#endif /* !CONFIG_ACPI */ - -#endif diff --git a/drivers/pci/pcie/pme/pcie_pme_acpi.c b/drivers/pci/pcie/pme/pcie_pme_acpi.c deleted file mode 100644 index be20222..0000000 --- a/drivers/pci/pcie/pme/pcie_pme_acpi.c +++ /dev/null @@ -1,56 +0,0 @@ -/* - * PCIe Native PME support, ACPI-related part - * - * Copyright (C) 2009 Rafael J. Wysocki , Novell Inc. - * - * This file is subject to the terms and conditions of the GNU General Public - * License V2. See the file "COPYING" in the main directory of this archive - * for more details. - */ - -#include -#include -#include -#include -#include -#include - -/** - * pcie_pme_acpi_setup - Request the ACPI BIOS to release control over PCIe PME. - * @srv - PCIe PME service for a root port or event collector. - * - * Invoked when the PCIe bus type loads PCIe PME service driver. To avoid - * conflict with the BIOS PCIe support requires the BIOS to yield PCIe PME - * control to the kernel. - */ -int pcie_pme_acpi_setup(struct pcie_device *srv) -{ - acpi_status status = AE_NOT_FOUND; - struct pci_dev *port = srv->port; - acpi_handle handle; - u32 flags; - int error = 0; - - if (acpi_pci_disabled) - return -ENOSYS; - - dev_info(&port->dev, "Requesting control of PCIe PME from ACPI BIOS\n"); - - handle = acpi_find_root_bridge_handle(port); - if (!handle) - return -EINVAL; - - flags = OSC_PCI_EXPRESS_PME_CONTROL | - OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL; - - status = acpi_pci_osc_control_set(handle, &flags, flags); - if (ACPI_FAILURE(status)) { - dev_info(&port->dev, - "Failed to receive control of PCIe PME service: %s\n", - (status == AE_SUPPORT || status == AE_NOT_FOUND) ? - "no _OSC support" : "ACPI _OSC failed"); - error = -ENODEV; - } - - return error; -} diff --git a/drivers/pci/pcie/portdrv.h b/drivers/pci/pcie/portdrv.h index 966f6e9..7b5aba0 100644 --- a/drivers/pci/pcie/portdrv.h +++ b/drivers/pci/pcie/portdrv.h @@ -21,6 +21,7 @@ #define get_descriptor_id(type, service) (((type - 4) << 4) | service) extern bool pcie_ports_disabled; +extern bool pcie_ports_auto; extern struct bus_type pcie_port_bus_type; extern int pcie_port_device_register(struct pci_dev *dev); @@ -32,6 +33,8 @@ extern void pcie_port_device_remove(struct pci_dev *dev); extern int __must_check pcie_port_bus_register(void); extern void pcie_port_bus_unregister(void); +struct pci_dev; + #ifdef CONFIG_PCIE_PME extern bool pcie_pme_msi_disabled; @@ -44,9 +47,26 @@ static inline bool pcie_pme_no_msi(void) { return pcie_pme_msi_disabled; } + +extern void pcie_pme_interrupt_enable(struct pci_dev *dev, bool enable); #else /* !CONFIG_PCIE_PME */ static inline void pcie_pme_disable_msi(void) {} static inline bool pcie_pme_no_msi(void) { return false; } +static inline void pcie_pme_interrupt_enable(struct pci_dev *dev, bool en) {} #endif /* !CONFIG_PCIE_PME */ +#ifdef CONFIG_ACPI +extern int pcie_port_acpi_setup(struct pci_dev *port, int *mask); + +static inline int pcie_port_platform_notify(struct pci_dev *port, int *mask) +{ + return pcie_port_acpi_setup(port, mask); +} +#else /* !CONFIG_ACPI */ +static inline int pcie_port_platform_notify(struct pci_dev *port, int *mask) +{ + return 0; +} +#endif /* !CONFIG_ACPI */ + #endif /* _PORTDRV_H_ */ diff --git a/drivers/pci/pcie/portdrv_acpi.c b/drivers/pci/pcie/portdrv_acpi.c new file mode 100644 index 0000000..b7c4cb1 --- /dev/null +++ b/drivers/pci/pcie/portdrv_acpi.c @@ -0,0 +1,77 @@ +/* + * PCIe Port Native Services Support, ACPI-Related Part + * + * Copyright (C) 2010 Rafael J. Wysocki , Novell Inc. + * + * This file is subject to the terms and conditions of the GNU General Public + * License V2. See the file "COPYING" in the main directory of this archive + * for more details. + */ + +#include +#include +#include +#include +#include +#include + +#include "aer/aerdrv.h" +#include "../pci.h" + +/** + * pcie_port_acpi_setup - Request the BIOS to release control of PCIe services. + * @port: PCIe Port service for a root port or event collector. + * @srv_mask: Bit mask of services that can be enabled for @port. + * + * Invoked when @port is identified as a PCIe port device. To avoid conflicts + * with the BIOS PCIe port native services support requires the BIOS to yield + * control of these services to the kernel. The mask of services that the BIOS + * allows to be enabled for @port is written to @srv_mask. + * + * NOTE: It turns out that we cannot do that for individual port services + * separately, because that would make some systems work incorrectly. + */ +int pcie_port_acpi_setup(struct pci_dev *port, int *srv_mask) +{ + acpi_status status; + acpi_handle handle; + u32 flags; + + if (acpi_pci_disabled) + return 0; + + handle = acpi_find_root_bridge_handle(port); + if (!handle) + return -EINVAL; + + flags = OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL + | OSC_PCI_EXPRESS_NATIVE_HP_CONTROL + | OSC_PCI_EXPRESS_PME_CONTROL; + + if (pci_aer_available()) { + if (pcie_aer_get_firmware_first(port)) + dev_dbg(&port->dev, "PCIe errors handled by BIOS.\n"); + else + flags |= OSC_PCI_EXPRESS_AER_CONTROL; + } + + status = acpi_pci_osc_control_set(handle, &flags, + OSC_PCI_EXPRESS_CAP_STRUCTURE_CONTROL); + if (ACPI_FAILURE(status)) { + dev_dbg(&port->dev, "ACPI _OSC request failed (code %d)\n", + status); + return -ENODEV; + } + + dev_info(&port->dev, "ACPI _OSC control granted for 0x%02x\n", flags); + + *srv_mask = PCIE_PORT_SERVICE_VC; + if (flags & OSC_PCI_EXPRESS_NATIVE_HP_CONTROL) + *srv_mask |= PCIE_PORT_SERVICE_HP; + if (flags & OSC_PCI_EXPRESS_PME_CONTROL) + *srv_mask |= PCIE_PORT_SERVICE_PME; + if (flags & OSC_PCI_EXPRESS_AER_CONTROL) + *srv_mask |= PCIE_PORT_SERVICE_AER; + + return 0; +} diff --git a/drivers/pci/pcie/portdrv_core.c b/drivers/pci/pcie/portdrv_core.c index 2bf2fe5..d0245c8 100644 --- a/drivers/pci/pcie/portdrv_core.c +++ b/drivers/pci/pcie/portdrv_core.c @@ -14,6 +14,8 @@ #include #include #include +#include +#include #include "../pci.h" #include "portdrv.h" @@ -236,23 +238,40 @@ static int get_port_device_capability(struct pci_dev *dev) int services = 0, pos; u16 reg16; u32 reg32; + int cap_mask; + int err; + + err = pcie_port_platform_notify(dev, &cap_mask); + if (pcie_ports_auto) { + if (err) { + pcie_no_aspm(); + return 0; + } + } else { + cap_mask = PCIE_PORT_SERVICE_PME | PCIE_PORT_SERVICE_HP + | PCIE_PORT_SERVICE_VC; + if (pci_aer_available()) + cap_mask |= PCIE_PORT_SERVICE_AER; + } pos = pci_pcie_cap(dev); pci_read_config_word(dev, pos + PCI_EXP_FLAGS, ®16); /* Hot-Plug Capable */ - if (reg16 & PCI_EXP_FLAGS_SLOT) { + if ((cap_mask & PCIE_PORT_SERVICE_HP) && (reg16 & PCI_EXP_FLAGS_SLOT)) { pci_read_config_dword(dev, pos + PCI_EXP_SLTCAP, ®32); if (reg32 & PCI_EXP_SLTCAP_HPC) services |= PCIE_PORT_SERVICE_HP; } /* AER capable */ - if (pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR)) + if ((cap_mask & PCIE_PORT_SERVICE_AER) + && pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR)) services |= PCIE_PORT_SERVICE_AER; /* VC support */ if (pci_find_ext_capability(dev, PCI_EXT_CAP_ID_VC)) services |= PCIE_PORT_SERVICE_VC; /* Root ports are capable of generating PME too */ - if (dev->pcie_type == PCI_EXP_TYPE_ROOT_PORT) + if ((cap_mask & PCIE_PORT_SERVICE_PME) + && dev->pcie_type == PCI_EXP_TYPE_ROOT_PORT) services |= PCIE_PORT_SERVICE_PME; return services; diff --git a/drivers/pci/pcie/portdrv_pci.c b/drivers/pci/pcie/portdrv_pci.c index a04392d..8f1338d 100644 --- a/drivers/pci/pcie/portdrv_pci.c +++ b/drivers/pci/pcie/portdrv_pci.c @@ -15,6 +15,7 @@ #include #include #include +#include #include "portdrv.h" #include "aer/aerdrv.h" @@ -32,10 +33,23 @@ MODULE_LICENSE("GPL"); /* If this switch is set, PCIe port native services should not be enabled. */ bool pcie_ports_disabled; +/* + * If this switch is set, ACPI _OSC will be used to determine whether or not to + * enable PCIe port native services. + */ +bool pcie_ports_auto = true; + static int __init pcie_port_setup(char *str) { - if (!strncmp(str, "compat", 6)) + if (!strncmp(str, "compat", 6)) { pcie_ports_disabled = true; + } else if (!strncmp(str, "native", 6)) { + pcie_ports_disabled = false; + pcie_ports_auto = false; + } else if (!strncmp(str, "auto", 4)) { + pcie_ports_disabled = false; + pcie_ports_auto = true; + } return 1; } @@ -313,8 +327,10 @@ static int __init pcie_portdrv_init(void) { int retval; - if (pcie_ports_disabled) + if (pcie_ports_disabled) { + pcie_no_aspm(); return -EACCES; + } dmi_check_system(pcie_portdrv_dmi_table); -- cgit v0.10.2 From 2bd50dd800b52245294cfceb56be62020cdc7515 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 21 Aug 2010 01:57:39 +0200 Subject: PCI: PCIe: Disable PCIe port services during port initialization In principle PCIe port services may be enabled by the BIOS, so it's better to disable them during port initialization to avoid spurious events from being generated. Signed-off-by: Rafael J. Wysocki Reviewed-by: Hidetoshi Seto Signed-off-by: Jesse Barnes diff --git a/drivers/pci/pcie/portdrv_core.c b/drivers/pci/pcie/portdrv_core.c index d0245c8..a9c222d 100644 --- a/drivers/pci/pcie/portdrv_core.c +++ b/drivers/pci/pcie/portdrv_core.c @@ -259,20 +259,43 @@ static int get_port_device_capability(struct pci_dev *dev) /* Hot-Plug Capable */ if ((cap_mask & PCIE_PORT_SERVICE_HP) && (reg16 & PCI_EXP_FLAGS_SLOT)) { pci_read_config_dword(dev, pos + PCI_EXP_SLTCAP, ®32); - if (reg32 & PCI_EXP_SLTCAP_HPC) + if (reg32 & PCI_EXP_SLTCAP_HPC) { services |= PCIE_PORT_SERVICE_HP; + /* + * Disable hot-plug interrupts in case they have been + * enabled by the BIOS and the hot-plug service driver + * is not loaded. + */ + pos += PCI_EXP_SLTCTL; + pci_read_config_word(dev, pos, ®16); + reg16 &= ~(PCI_EXP_SLTCTL_CCIE | PCI_EXP_SLTCTL_HPIE); + pci_write_config_word(dev, pos, reg16); + } } /* AER capable */ if ((cap_mask & PCIE_PORT_SERVICE_AER) - && pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR)) + && pci_find_ext_capability(dev, PCI_EXT_CAP_ID_ERR)) { services |= PCIE_PORT_SERVICE_AER; + /* + * Disable AER on this port in case it's been enabled by the + * BIOS (the AER service driver will enable it when necessary). + */ + pci_disable_pcie_error_reporting(dev); + } /* VC support */ if (pci_find_ext_capability(dev, PCI_EXT_CAP_ID_VC)) services |= PCIE_PORT_SERVICE_VC; /* Root ports are capable of generating PME too */ if ((cap_mask & PCIE_PORT_SERVICE_PME) - && dev->pcie_type == PCI_EXP_TYPE_ROOT_PORT) + && dev->pcie_type == PCI_EXP_TYPE_ROOT_PORT) { services |= PCIE_PORT_SERVICE_PME; + /* + * Disable PME interrupt on this port in case it's been enabled + * by the BIOS (the PME service driver will enable it when + * necessary). + */ + pcie_pme_interrupt_enable(dev, false); + } return services; } -- cgit v0.10.2 From 271fb719cc472af3b1e96d8c527bb0da7060a172 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 21 Aug 2010 01:58:22 +0200 Subject: PCI: PCIe: Move PCIe PME code to the pcie directory The PCIe PME code only consists of one file, so it doesn't need to occupy its own directory. Move it to drivers/pci/pcie/pme.c and remove the contents of drivers/pci/pcie/pme . Signed-off-by: Rafael J. Wysocki Signed-off-by: Jesse Barnes diff --git a/drivers/pci/pcie/Makefile b/drivers/pci/pcie/Makefile index 4d2b187..00c62df 100644 --- a/drivers/pci/pcie/Makefile +++ b/drivers/pci/pcie/Makefile @@ -13,4 +13,4 @@ obj-$(CONFIG_PCIEPORTBUS) += pcieportdrv.o # Build PCI Express AER if needed obj-$(CONFIG_PCIEAER) += aer/ -obj-$(CONFIG_PCIE_PME) += pme/ +obj-$(CONFIG_PCIE_PME) += pme.o diff --git a/drivers/pci/pcie/pme.c b/drivers/pci/pcie/pme.c new file mode 100644 index 0000000..2f3c904 --- /dev/null +++ b/drivers/pci/pcie/pme.c @@ -0,0 +1,462 @@ +/* + * PCIe Native PME support + * + * Copyright (C) 2007 - 2009 Intel Corp + * Copyright (C) 2007 - 2009 Shaohua Li + * Copyright (C) 2009 Rafael J. Wysocki , Novell Inc. + * + * This file is subject to the terms and conditions of the GNU General Public + * License V2. See the file "COPYING" in the main directory of this archive + * for more details. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "../pci.h" +#include "portdrv.h" + +#define PCI_EXP_RTSTA_PME 0x10000 /* PME status */ +#define PCI_EXP_RTSTA_PENDING 0x20000 /* PME pending */ + +/* + * If this switch is set, MSI will not be used for PCIe PME signaling. This + * causes the PCIe port driver to use INTx interrupts only, but it turns out + * that using MSI for PCIe PME signaling doesn't play well with PCIe PME-based + * wake-up from system sleep states. + */ +bool pcie_pme_msi_disabled; + +static int __init pcie_pme_setup(char *str) +{ + if (!strncmp(str, "nomsi", 5)) + pcie_pme_msi_disabled = true; + + return 1; +} +__setup("pcie_pme=", pcie_pme_setup); + +struct pcie_pme_service_data { + spinlock_t lock; + struct pcie_device *srv; + struct work_struct work; + bool noirq; /* Don't enable the PME interrupt used by this service. */ +}; + +/** + * pcie_pme_interrupt_enable - Enable/disable PCIe PME interrupt generation. + * @dev: PCIe root port or event collector. + * @enable: Enable or disable the interrupt. + */ +void pcie_pme_interrupt_enable(struct pci_dev *dev, bool enable) +{ + int rtctl_pos; + u16 rtctl; + + rtctl_pos = pci_pcie_cap(dev) + PCI_EXP_RTCTL; + + pci_read_config_word(dev, rtctl_pos, &rtctl); + if (enable) + rtctl |= PCI_EXP_RTCTL_PMEIE; + else + rtctl &= ~PCI_EXP_RTCTL_PMEIE; + pci_write_config_word(dev, rtctl_pos, rtctl); +} + +/** + * pcie_pme_clear_status - Clear root port PME interrupt status. + * @dev: PCIe root port or event collector. + */ +static void pcie_pme_clear_status(struct pci_dev *dev) +{ + int rtsta_pos; + u32 rtsta; + + rtsta_pos = pci_pcie_cap(dev) + PCI_EXP_RTSTA; + + pci_read_config_dword(dev, rtsta_pos, &rtsta); + rtsta |= PCI_EXP_RTSTA_PME; + pci_write_config_dword(dev, rtsta_pos, rtsta); +} + +/** + * pcie_pme_walk_bus - Scan a PCI bus for devices asserting PME#. + * @bus: PCI bus to scan. + * + * Scan given PCI bus and all buses under it for devices asserting PME#. + */ +static bool pcie_pme_walk_bus(struct pci_bus *bus) +{ + struct pci_dev *dev; + bool ret = false; + + list_for_each_entry(dev, &bus->devices, bus_list) { + /* Skip PCIe devices in case we started from a root port. */ + if (!pci_is_pcie(dev) && pci_check_pme_status(dev)) { + pm_request_resume(&dev->dev); + pci_wakeup_event(dev); + ret = true; + } + + if (dev->subordinate && pcie_pme_walk_bus(dev->subordinate)) + ret = true; + } + + return ret; +} + +/** + * pcie_pme_from_pci_bridge - Check if PCIe-PCI bridge generated a PME. + * @bus: Secondary bus of the bridge. + * @devfn: Device/function number to check. + * + * PME from PCI devices under a PCIe-PCI bridge may be converted to an in-band + * PCIe PME message. In such that case the bridge should use the Requester ID + * of device/function number 0 on its secondary bus. + */ +static bool pcie_pme_from_pci_bridge(struct pci_bus *bus, u8 devfn) +{ + struct pci_dev *dev; + bool found = false; + + if (devfn) + return false; + + dev = pci_dev_get(bus->self); + if (!dev) + return false; + + if (pci_is_pcie(dev) && dev->pcie_type == PCI_EXP_TYPE_PCI_BRIDGE) { + down_read(&pci_bus_sem); + if (pcie_pme_walk_bus(bus)) + found = true; + up_read(&pci_bus_sem); + } + + pci_dev_put(dev); + return found; +} + +/** + * pcie_pme_handle_request - Find device that generated PME and handle it. + * @port: Root port or event collector that generated the PME interrupt. + * @req_id: PCIe Requester ID of the device that generated the PME. + */ +static void pcie_pme_handle_request(struct pci_dev *port, u16 req_id) +{ + u8 busnr = req_id >> 8, devfn = req_id & 0xff; + struct pci_bus *bus; + struct pci_dev *dev; + bool found = false; + + /* First, check if the PME is from the root port itself. */ + if (port->devfn == devfn && port->bus->number == busnr) { + if (pci_check_pme_status(port)) { + pm_request_resume(&port->dev); + found = true; + } else { + /* + * Apparently, the root port generated the PME on behalf + * of a non-PCIe device downstream. If this is done by + * a root port, the Requester ID field in its status + * register may contain either the root port's, or the + * source device's information (PCI Express Base + * Specification, Rev. 2.0, Section 6.1.9). + */ + down_read(&pci_bus_sem); + found = pcie_pme_walk_bus(port->subordinate); + up_read(&pci_bus_sem); + } + goto out; + } + + /* Second, find the bus the source device is on. */ + bus = pci_find_bus(pci_domain_nr(port->bus), busnr); + if (!bus) + goto out; + + /* Next, check if the PME is from a PCIe-PCI bridge. */ + found = pcie_pme_from_pci_bridge(bus, devfn); + if (found) + goto out; + + /* Finally, try to find the PME source on the bus. */ + down_read(&pci_bus_sem); + list_for_each_entry(dev, &bus->devices, bus_list) { + pci_dev_get(dev); + if (dev->devfn == devfn) { + found = true; + break; + } + pci_dev_put(dev); + } + up_read(&pci_bus_sem); + + if (found) { + /* The device is there, but we have to check its PME status. */ + found = pci_check_pme_status(dev); + if (found) { + pm_request_resume(&dev->dev); + pci_wakeup_event(dev); + } + pci_dev_put(dev); + } else if (devfn) { + /* + * The device is not there, but we can still try to recover by + * assuming that the PME was reported by a PCIe-PCI bridge that + * used devfn different from zero. + */ + dev_dbg(&port->dev, "PME interrupt generated for " + "non-existent device %02x:%02x.%d\n", + busnr, PCI_SLOT(devfn), PCI_FUNC(devfn)); + found = pcie_pme_from_pci_bridge(bus, 0); + } + + out: + if (!found) + dev_dbg(&port->dev, "Spurious native PME interrupt!\n"); +} + +/** + * pcie_pme_work_fn - Work handler for PCIe PME interrupt. + * @work: Work structure giving access to service data. + */ +static void pcie_pme_work_fn(struct work_struct *work) +{ + struct pcie_pme_service_data *data = + container_of(work, struct pcie_pme_service_data, work); + struct pci_dev *port = data->srv->port; + int rtsta_pos; + u32 rtsta; + + rtsta_pos = pci_pcie_cap(port) + PCI_EXP_RTSTA; + + spin_lock_irq(&data->lock); + + for (;;) { + if (data->noirq) + break; + + pci_read_config_dword(port, rtsta_pos, &rtsta); + if (rtsta & PCI_EXP_RTSTA_PME) { + /* + * Clear PME status of the port. If there are other + * pending PMEs, the status will be set again. + */ + pcie_pme_clear_status(port); + + spin_unlock_irq(&data->lock); + pcie_pme_handle_request(port, rtsta & 0xffff); + spin_lock_irq(&data->lock); + + continue; + } + + /* No need to loop if there are no more PMEs pending. */ + if (!(rtsta & PCI_EXP_RTSTA_PENDING)) + break; + + spin_unlock_irq(&data->lock); + cpu_relax(); + spin_lock_irq(&data->lock); + } + + if (!data->noirq) + pcie_pme_interrupt_enable(port, true); + + spin_unlock_irq(&data->lock); +} + +/** + * pcie_pme_irq - Interrupt handler for PCIe root port PME interrupt. + * @irq: Interrupt vector. + * @context: Interrupt context pointer. + */ +static irqreturn_t pcie_pme_irq(int irq, void *context) +{ + struct pci_dev *port; + struct pcie_pme_service_data *data; + int rtsta_pos; + u32 rtsta; + unsigned long flags; + + port = ((struct pcie_device *)context)->port; + data = get_service_data((struct pcie_device *)context); + + rtsta_pos = pci_pcie_cap(port) + PCI_EXP_RTSTA; + + spin_lock_irqsave(&data->lock, flags); + pci_read_config_dword(port, rtsta_pos, &rtsta); + + if (!(rtsta & PCI_EXP_RTSTA_PME)) { + spin_unlock_irqrestore(&data->lock, flags); + return IRQ_NONE; + } + + pcie_pme_interrupt_enable(port, false); + spin_unlock_irqrestore(&data->lock, flags); + + /* We don't use pm_wq, because it's freezable. */ + schedule_work(&data->work); + + return IRQ_HANDLED; +} + +/** + * pcie_pme_set_native - Set the PME interrupt flag for given device. + * @dev: PCI device to handle. + * @ign: Ignored. + */ +static int pcie_pme_set_native(struct pci_dev *dev, void *ign) +{ + dev_info(&dev->dev, "Signaling PME through PCIe PME interrupt\n"); + + device_set_run_wake(&dev->dev, true); + dev->pme_interrupt = true; + return 0; +} + +/** + * pcie_pme_mark_devices - Set the PME interrupt flag for devices below a port. + * @port: PCIe root port or event collector to handle. + * + * For each device below given root port, including the port itself (or for each + * root complex integrated endpoint if @port is a root complex event collector) + * set the flag indicating that it can signal run-time wake-up events via PCIe + * PME interrupts. + */ +static void pcie_pme_mark_devices(struct pci_dev *port) +{ + pcie_pme_set_native(port, NULL); + if (port->subordinate) { + pci_walk_bus(port->subordinate, pcie_pme_set_native, NULL); + } else { + struct pci_bus *bus = port->bus; + struct pci_dev *dev; + + /* Check if this is a root port event collector. */ + if (port->pcie_type != PCI_EXP_TYPE_RC_EC || !bus) + return; + + down_read(&pci_bus_sem); + list_for_each_entry(dev, &bus->devices, bus_list) + if (pci_is_pcie(dev) + && dev->pcie_type == PCI_EXP_TYPE_RC_END) + pcie_pme_set_native(dev, NULL); + up_read(&pci_bus_sem); + } +} + +/** + * pcie_pme_probe - Initialize PCIe PME service for given root port. + * @srv: PCIe service to initialize. + */ +static int pcie_pme_probe(struct pcie_device *srv) +{ + struct pci_dev *port; + struct pcie_pme_service_data *data; + int ret; + + data = kzalloc(sizeof(*data), GFP_KERNEL); + if (!data) + return -ENOMEM; + + spin_lock_init(&data->lock); + INIT_WORK(&data->work, pcie_pme_work_fn); + data->srv = srv; + set_service_data(srv, data); + + port = srv->port; + pcie_pme_interrupt_enable(port, false); + pcie_pme_clear_status(port); + + ret = request_irq(srv->irq, pcie_pme_irq, IRQF_SHARED, "PCIe PME", srv); + if (ret) { + kfree(data); + } else { + pcie_pme_mark_devices(port); + pcie_pme_interrupt_enable(port, true); + } + + return ret; +} + +/** + * pcie_pme_suspend - Suspend PCIe PME service device. + * @srv: PCIe service device to suspend. + */ +static int pcie_pme_suspend(struct pcie_device *srv) +{ + struct pcie_pme_service_data *data = get_service_data(srv); + struct pci_dev *port = srv->port; + + spin_lock_irq(&data->lock); + pcie_pme_interrupt_enable(port, false); + pcie_pme_clear_status(port); + data->noirq = true; + spin_unlock_irq(&data->lock); + + synchronize_irq(srv->irq); + + return 0; +} + +/** + * pcie_pme_resume - Resume PCIe PME service device. + * @srv - PCIe service device to resume. + */ +static int pcie_pme_resume(struct pcie_device *srv) +{ + struct pcie_pme_service_data *data = get_service_data(srv); + struct pci_dev *port = srv->port; + + spin_lock_irq(&data->lock); + data->noirq = false; + pcie_pme_clear_status(port); + pcie_pme_interrupt_enable(port, true); + spin_unlock_irq(&data->lock); + + return 0; +} + +/** + * pcie_pme_remove - Prepare PCIe PME service device for removal. + * @srv - PCIe service device to resume. + */ +static void pcie_pme_remove(struct pcie_device *srv) +{ + pcie_pme_suspend(srv); + free_irq(srv->irq, srv); + kfree(get_service_data(srv)); +} + +static struct pcie_port_service_driver pcie_pme_driver = { + .name = "pcie_pme", + .port_type = PCI_EXP_TYPE_ROOT_PORT, + .service = PCIE_PORT_SERVICE_PME, + + .probe = pcie_pme_probe, + .suspend = pcie_pme_suspend, + .resume = pcie_pme_resume, + .remove = pcie_pme_remove, +}; + +/** + * pcie_pme_service_init - Register the PCIe PME service driver. + */ +static int __init pcie_pme_service_init(void) +{ + return pcie_port_service_register(&pcie_pme_driver); +} + +module_init(pcie_pme_service_init); diff --git a/drivers/pci/pcie/pme/Makefile b/drivers/pci/pcie/pme/Makefile deleted file mode 100644 index 3c67bf4..0000000 --- a/drivers/pci/pcie/pme/Makefile +++ /dev/null @@ -1,5 +0,0 @@ -# -# Makefile for PCI-Express Root Port PME signaling driver -# - -obj-$(CONFIG_PCIE_PME) += pcie_pme.o diff --git a/drivers/pci/pcie/pme/pcie_pme.c b/drivers/pci/pcie/pme/pcie_pme.c deleted file mode 100644 index 9d1361e..0000000 --- a/drivers/pci/pcie/pme/pcie_pme.c +++ /dev/null @@ -1,462 +0,0 @@ -/* - * PCIe Native PME support - * - * Copyright (C) 2007 - 2009 Intel Corp - * Copyright (C) 2007 - 2009 Shaohua Li - * Copyright (C) 2009 Rafael J. Wysocki , Novell Inc. - * - * This file is subject to the terms and conditions of the GNU General Public - * License V2. See the file "COPYING" in the main directory of this archive - * for more details. - */ - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include "../../pci.h" -#include "../portdrv.h" - -#define PCI_EXP_RTSTA_PME 0x10000 /* PME status */ -#define PCI_EXP_RTSTA_PENDING 0x20000 /* PME pending */ - -/* - * If this switch is set, MSI will not be used for PCIe PME signaling. This - * causes the PCIe port driver to use INTx interrupts only, but it turns out - * that using MSI for PCIe PME signaling doesn't play well with PCIe PME-based - * wake-up from system sleep states. - */ -bool pcie_pme_msi_disabled; - -static int __init pcie_pme_setup(char *str) -{ - if (!strncmp(str, "nomsi", 5)) - pcie_pme_msi_disabled = true; - - return 1; -} -__setup("pcie_pme=", pcie_pme_setup); - -struct pcie_pme_service_data { - spinlock_t lock; - struct pcie_device *srv; - struct work_struct work; - bool noirq; /* Don't enable the PME interrupt used by this service. */ -}; - -/** - * pcie_pme_interrupt_enable - Enable/disable PCIe PME interrupt generation. - * @dev: PCIe root port or event collector. - * @enable: Enable or disable the interrupt. - */ -void pcie_pme_interrupt_enable(struct pci_dev *dev, bool enable) -{ - int rtctl_pos; - u16 rtctl; - - rtctl_pos = pci_pcie_cap(dev) + PCI_EXP_RTCTL; - - pci_read_config_word(dev, rtctl_pos, &rtctl); - if (enable) - rtctl |= PCI_EXP_RTCTL_PMEIE; - else - rtctl &= ~PCI_EXP_RTCTL_PMEIE; - pci_write_config_word(dev, rtctl_pos, rtctl); -} - -/** - * pcie_pme_clear_status - Clear root port PME interrupt status. - * @dev: PCIe root port or event collector. - */ -static void pcie_pme_clear_status(struct pci_dev *dev) -{ - int rtsta_pos; - u32 rtsta; - - rtsta_pos = pci_pcie_cap(dev) + PCI_EXP_RTSTA; - - pci_read_config_dword(dev, rtsta_pos, &rtsta); - rtsta |= PCI_EXP_RTSTA_PME; - pci_write_config_dword(dev, rtsta_pos, rtsta); -} - -/** - * pcie_pme_walk_bus - Scan a PCI bus for devices asserting PME#. - * @bus: PCI bus to scan. - * - * Scan given PCI bus and all buses under it for devices asserting PME#. - */ -static bool pcie_pme_walk_bus(struct pci_bus *bus) -{ - struct pci_dev *dev; - bool ret = false; - - list_for_each_entry(dev, &bus->devices, bus_list) { - /* Skip PCIe devices in case we started from a root port. */ - if (!pci_is_pcie(dev) && pci_check_pme_status(dev)) { - pm_request_resume(&dev->dev); - pci_wakeup_event(dev); - ret = true; - } - - if (dev->subordinate && pcie_pme_walk_bus(dev->subordinate)) - ret = true; - } - - return ret; -} - -/** - * pcie_pme_from_pci_bridge - Check if PCIe-PCI bridge generated a PME. - * @bus: Secondary bus of the bridge. - * @devfn: Device/function number to check. - * - * PME from PCI devices under a PCIe-PCI bridge may be converted to an in-band - * PCIe PME message. In such that case the bridge should use the Requester ID - * of device/function number 0 on its secondary bus. - */ -static bool pcie_pme_from_pci_bridge(struct pci_bus *bus, u8 devfn) -{ - struct pci_dev *dev; - bool found = false; - - if (devfn) - return false; - - dev = pci_dev_get(bus->self); - if (!dev) - return false; - - if (pci_is_pcie(dev) && dev->pcie_type == PCI_EXP_TYPE_PCI_BRIDGE) { - down_read(&pci_bus_sem); - if (pcie_pme_walk_bus(bus)) - found = true; - up_read(&pci_bus_sem); - } - - pci_dev_put(dev); - return found; -} - -/** - * pcie_pme_handle_request - Find device that generated PME and handle it. - * @port: Root port or event collector that generated the PME interrupt. - * @req_id: PCIe Requester ID of the device that generated the PME. - */ -static void pcie_pme_handle_request(struct pci_dev *port, u16 req_id) -{ - u8 busnr = req_id >> 8, devfn = req_id & 0xff; - struct pci_bus *bus; - struct pci_dev *dev; - bool found = false; - - /* First, check if the PME is from the root port itself. */ - if (port->devfn == devfn && port->bus->number == busnr) { - if (pci_check_pme_status(port)) { - pm_request_resume(&port->dev); - found = true; - } else { - /* - * Apparently, the root port generated the PME on behalf - * of a non-PCIe device downstream. If this is done by - * a root port, the Requester ID field in its status - * register may contain either the root port's, or the - * source device's information (PCI Express Base - * Specification, Rev. 2.0, Section 6.1.9). - */ - down_read(&pci_bus_sem); - found = pcie_pme_walk_bus(port->subordinate); - up_read(&pci_bus_sem); - } - goto out; - } - - /* Second, find the bus the source device is on. */ - bus = pci_find_bus(pci_domain_nr(port->bus), busnr); - if (!bus) - goto out; - - /* Next, check if the PME is from a PCIe-PCI bridge. */ - found = pcie_pme_from_pci_bridge(bus, devfn); - if (found) - goto out; - - /* Finally, try to find the PME source on the bus. */ - down_read(&pci_bus_sem); - list_for_each_entry(dev, &bus->devices, bus_list) { - pci_dev_get(dev); - if (dev->devfn == devfn) { - found = true; - break; - } - pci_dev_put(dev); - } - up_read(&pci_bus_sem); - - if (found) { - /* The device is there, but we have to check its PME status. */ - found = pci_check_pme_status(dev); - if (found) { - pm_request_resume(&dev->dev); - pci_wakeup_event(dev); - } - pci_dev_put(dev); - } else if (devfn) { - /* - * The device is not there, but we can still try to recover by - * assuming that the PME was reported by a PCIe-PCI bridge that - * used devfn different from zero. - */ - dev_dbg(&port->dev, "PME interrupt generated for " - "non-existent device %02x:%02x.%d\n", - busnr, PCI_SLOT(devfn), PCI_FUNC(devfn)); - found = pcie_pme_from_pci_bridge(bus, 0); - } - - out: - if (!found) - dev_dbg(&port->dev, "Spurious native PME interrupt!\n"); -} - -/** - * pcie_pme_work_fn - Work handler for PCIe PME interrupt. - * @work: Work structure giving access to service data. - */ -static void pcie_pme_work_fn(struct work_struct *work) -{ - struct pcie_pme_service_data *data = - container_of(work, struct pcie_pme_service_data, work); - struct pci_dev *port = data->srv->port; - int rtsta_pos; - u32 rtsta; - - rtsta_pos = pci_pcie_cap(port) + PCI_EXP_RTSTA; - - spin_lock_irq(&data->lock); - - for (;;) { - if (data->noirq) - break; - - pci_read_config_dword(port, rtsta_pos, &rtsta); - if (rtsta & PCI_EXP_RTSTA_PME) { - /* - * Clear PME status of the port. If there are other - * pending PMEs, the status will be set again. - */ - pcie_pme_clear_status(port); - - spin_unlock_irq(&data->lock); - pcie_pme_handle_request(port, rtsta & 0xffff); - spin_lock_irq(&data->lock); - - continue; - } - - /* No need to loop if there are no more PMEs pending. */ - if (!(rtsta & PCI_EXP_RTSTA_PENDING)) - break; - - spin_unlock_irq(&data->lock); - cpu_relax(); - spin_lock_irq(&data->lock); - } - - if (!data->noirq) - pcie_pme_interrupt_enable(port, true); - - spin_unlock_irq(&data->lock); -} - -/** - * pcie_pme_irq - Interrupt handler for PCIe root port PME interrupt. - * @irq: Interrupt vector. - * @context: Interrupt context pointer. - */ -static irqreturn_t pcie_pme_irq(int irq, void *context) -{ - struct pci_dev *port; - struct pcie_pme_service_data *data; - int rtsta_pos; - u32 rtsta; - unsigned long flags; - - port = ((struct pcie_device *)context)->port; - data = get_service_data((struct pcie_device *)context); - - rtsta_pos = pci_pcie_cap(port) + PCI_EXP_RTSTA; - - spin_lock_irqsave(&data->lock, flags); - pci_read_config_dword(port, rtsta_pos, &rtsta); - - if (!(rtsta & PCI_EXP_RTSTA_PME)) { - spin_unlock_irqrestore(&data->lock, flags); - return IRQ_NONE; - } - - pcie_pme_interrupt_enable(port, false); - spin_unlock_irqrestore(&data->lock, flags); - - /* We don't use pm_wq, because it's freezable. */ - schedule_work(&data->work); - - return IRQ_HANDLED; -} - -/** - * pcie_pme_set_native - Set the PME interrupt flag for given device. - * @dev: PCI device to handle. - * @ign: Ignored. - */ -static int pcie_pme_set_native(struct pci_dev *dev, void *ign) -{ - dev_info(&dev->dev, "Signaling PME through PCIe PME interrupt\n"); - - device_set_run_wake(&dev->dev, true); - dev->pme_interrupt = true; - return 0; -} - -/** - * pcie_pme_mark_devices - Set the PME interrupt flag for devices below a port. - * @port: PCIe root port or event collector to handle. - * - * For each device below given root port, including the port itself (or for each - * root complex integrated endpoint if @port is a root complex event collector) - * set the flag indicating that it can signal run-time wake-up events via PCIe - * PME interrupts. - */ -static void pcie_pme_mark_devices(struct pci_dev *port) -{ - pcie_pme_set_native(port, NULL); - if (port->subordinate) { - pci_walk_bus(port->subordinate, pcie_pme_set_native, NULL); - } else { - struct pci_bus *bus = port->bus; - struct pci_dev *dev; - - /* Check if this is a root port event collector. */ - if (port->pcie_type != PCI_EXP_TYPE_RC_EC || !bus) - return; - - down_read(&pci_bus_sem); - list_for_each_entry(dev, &bus->devices, bus_list) - if (pci_is_pcie(dev) - && dev->pcie_type == PCI_EXP_TYPE_RC_END) - pcie_pme_set_native(dev, NULL); - up_read(&pci_bus_sem); - } -} - -/** - * pcie_pme_probe - Initialize PCIe PME service for given root port. - * @srv: PCIe service to initialize. - */ -static int pcie_pme_probe(struct pcie_device *srv) -{ - struct pci_dev *port; - struct pcie_pme_service_data *data; - int ret; - - data = kzalloc(sizeof(*data), GFP_KERNEL); - if (!data) - return -ENOMEM; - - spin_lock_init(&data->lock); - INIT_WORK(&data->work, pcie_pme_work_fn); - data->srv = srv; - set_service_data(srv, data); - - port = srv->port; - pcie_pme_interrupt_enable(port, false); - pcie_pme_clear_status(port); - - ret = request_irq(srv->irq, pcie_pme_irq, IRQF_SHARED, "PCIe PME", srv); - if (ret) { - kfree(data); - } else { - pcie_pme_mark_devices(port); - pcie_pme_interrupt_enable(port, true); - } - - return ret; -} - -/** - * pcie_pme_suspend - Suspend PCIe PME service device. - * @srv: PCIe service device to suspend. - */ -static int pcie_pme_suspend(struct pcie_device *srv) -{ - struct pcie_pme_service_data *data = get_service_data(srv); - struct pci_dev *port = srv->port; - - spin_lock_irq(&data->lock); - pcie_pme_interrupt_enable(port, false); - pcie_pme_clear_status(port); - data->noirq = true; - spin_unlock_irq(&data->lock); - - synchronize_irq(srv->irq); - - return 0; -} - -/** - * pcie_pme_resume - Resume PCIe PME service device. - * @srv - PCIe service device to resume. - */ -static int pcie_pme_resume(struct pcie_device *srv) -{ - struct pcie_pme_service_data *data = get_service_data(srv); - struct pci_dev *port = srv->port; - - spin_lock_irq(&data->lock); - data->noirq = false; - pcie_pme_clear_status(port); - pcie_pme_interrupt_enable(port, true); - spin_unlock_irq(&data->lock); - - return 0; -} - -/** - * pcie_pme_remove - Prepare PCIe PME service device for removal. - * @srv - PCIe service device to resume. - */ -static void pcie_pme_remove(struct pcie_device *srv) -{ - pcie_pme_suspend(srv); - free_irq(srv->irq, srv); - kfree(get_service_data(srv)); -} - -static struct pcie_port_service_driver pcie_pme_driver = { - .name = "pcie_pme", - .port_type = PCI_EXP_TYPE_ROOT_PORT, - .service = PCIE_PORT_SERVICE_PME, - - .probe = pcie_pme_probe, - .suspend = pcie_pme_suspend, - .resume = pcie_pme_resume, - .remove = pcie_pme_remove, -}; - -/** - * pcie_pme_service_init - Register the PCIe PME service driver. - */ -static int __init pcie_pme_service_init(void) -{ - return pcie_port_service_register(&pcie_pme_driver); -} - -module_init(pcie_pme_service_init); -- cgit v0.10.2 From a9d2a6df11f5b9dc19ad4147374e8b67c4438158 Mon Sep 17 00:00:00 2001 From: Kenji Kaneshige Date: Sat, 21 Aug 2010 01:59:10 +0200 Subject: PCI: PCIe: Remove the port driver module exit routine The PCIe port driver's module exit routine is never used, so drop it. Signed-off-by: Kenji Kaneshige Signed-off-by: Rafael J. Wysocki Reviewed-by: Hidetoshi Seto Signed-off-by: Jesse Barnes diff --git a/drivers/pci/pcie/portdrv_pci.c b/drivers/pci/pcie/portdrv_pci.c index 8f1338d..f9033e1 100644 --- a/drivers/pci/pcie/portdrv_pci.c +++ b/drivers/pci/pcie/portdrv_pci.c @@ -346,11 +346,4 @@ static int __init pcie_portdrv_init(void) return retval; } -static void __exit pcie_portdrv_exit(void) -{ - pci_unregister_driver(&pcie_portdriver); - pcie_port_bus_unregister(); -} - module_init(pcie_portdrv_init); -module_exit(pcie_portdrv_exit); -- cgit v0.10.2 From b7d460897739e02f186425b7276e3fdb1595cea7 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Tue, 24 Aug 2010 22:44:12 -0700 Subject: x86, mm: Fix CONFIG_VMSPLIT_1G and 2G_OPT trampoline rc2 kernel crashes when booting second cpu on this CONFIG_VMSPLIT_2G_OPT laptop: whereas cloning from kernel to low mappings pgd range does need to limit by both KERNEL_PGD_PTRS and KERNEL_PGD_BOUNDARY, cloning kernel pgd range itself must not be limited by the smaller KERNEL_PGD_BOUNDARY. Signed-off-by: Hugh Dickins LKML-Reference: Signed-off-by: H. Peter Anvin diff --git a/arch/x86/kernel/trampoline.c b/arch/x86/kernel/trampoline.c index a874495..e2a5952 100644 --- a/arch/x86/kernel/trampoline.c +++ b/arch/x86/kernel/trampoline.c @@ -45,8 +45,7 @@ void __init setup_trampoline_page_table(void) /* Copy kernel address range */ clone_pgd_range(trampoline_pg_dir + KERNEL_PGD_BOUNDARY, swapper_pg_dir + KERNEL_PGD_BOUNDARY, - min_t(unsigned long, KERNEL_PGD_PTRS, - KERNEL_PGD_BOUNDARY)); + KERNEL_PGD_PTRS); /* Initialize low mappings */ clone_pgd_range(trampoline_pg_dir, -- cgit v0.10.2 From 750d857c682f4db60d14722d430c7ccc35070962 Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Fri, 13 Aug 2010 16:29:04 +0200 Subject: oprofile: fix crash when accessing freed task structs This patch fixes a crash during shutdown reported below. The crash is caused by accessing already freed task structs. The fix changes the order for registering and unregistering notifier callbacks. All notifiers must be initialized before buffers start working. To stop buffer synchronization we cancel all workqueues, unregister the notifier callback and then flush all buffers. After all of this we finally can free all tasks listed. This should avoid accessing freed tasks. On 22.07.10 01:14:40, Benjamin Herrenschmidt wrote: > So the initial observation is a spinlock bad magic followed by a crash > in the spinlock debug code: > > [ 1541.586531] BUG: spinlock bad magic on CPU#5, events/5/136 > [ 1541.597564] Unable to handle kernel paging request for data at address 0x6b6b6b6b6b6b6d03 > > Backtrace looks like: > > spin_bug+0x74/0xd4 > ._raw_spin_lock+0x48/0x184 > ._spin_lock+0x10/0x24 > .get_task_mm+0x28/0x8c > .sync_buffer+0x1b4/0x598 > .wq_sync_buffer+0xa0/0xdc > .worker_thread+0x1d8/0x2a8 > .kthread+0xa8/0xb4 > .kernel_thread+0x54/0x70 > > So we are accessing a freed task struct in the work queue when > processing the samples. Reported-by: Benjamin Herrenschmidt Cc: stable@kernel.org Signed-off-by: Robert Richter diff --git a/drivers/oprofile/buffer_sync.c b/drivers/oprofile/buffer_sync.c index a9352b2..b7e755f 100644 --- a/drivers/oprofile/buffer_sync.c +++ b/drivers/oprofile/buffer_sync.c @@ -141,16 +141,6 @@ static struct notifier_block module_load_nb = { .notifier_call = module_load_notify, }; - -static void end_sync(void) -{ - end_cpu_work(); - /* make sure we don't leak task structs */ - process_task_mortuary(); - process_task_mortuary(); -} - - int sync_start(void) { int err; @@ -158,7 +148,7 @@ int sync_start(void) if (!zalloc_cpumask_var(&marked_cpus, GFP_KERNEL)) return -ENOMEM; - start_cpu_work(); + mutex_lock(&buffer_mutex); err = task_handoff_register(&task_free_nb); if (err) @@ -173,7 +163,10 @@ int sync_start(void) if (err) goto out4; + start_cpu_work(); + out: + mutex_unlock(&buffer_mutex); return err; out4: profile_event_unregister(PROFILE_MUNMAP, &munmap_nb); @@ -182,7 +175,6 @@ out3: out2: task_handoff_unregister(&task_free_nb); out1: - end_sync(); free_cpumask_var(marked_cpus); goto out; } @@ -190,11 +182,20 @@ out1: void sync_stop(void) { + /* flush buffers */ + mutex_lock(&buffer_mutex); + end_cpu_work(); unregister_module_notifier(&module_load_nb); profile_event_unregister(PROFILE_MUNMAP, &munmap_nb); profile_event_unregister(PROFILE_TASK_EXIT, &task_exit_nb); task_handoff_unregister(&task_free_nb); - end_sync(); + mutex_unlock(&buffer_mutex); + flush_scheduled_work(); + + /* make sure we don't leak task structs */ + process_task_mortuary(); + process_task_mortuary(); + free_cpumask_var(marked_cpus); } diff --git a/drivers/oprofile/cpu_buffer.c b/drivers/oprofile/cpu_buffer.c index 219f79e..f179ac2 100644 --- a/drivers/oprofile/cpu_buffer.c +++ b/drivers/oprofile/cpu_buffer.c @@ -120,8 +120,6 @@ void end_cpu_work(void) cancel_delayed_work(&b->work); } - - flush_scheduled_work(); } /* -- cgit v0.10.2 From 8a2e8e5dec7e29c56a46ba176c664ab6a3d04118 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 25 Aug 2010 10:33:56 +0200 Subject: workqueue: fix cwq->nr_active underflow cwq->nr_active is used to keep track of how many work items are active for the cpu workqueue, where 'active' is defined as either pending on global worklist or executing. This is used to implement the max_active limit and workqueue freezing. If a work item is queued after nr_active has already reached max_active, the work item doesn't increment nr_active and is put on the delayed queue and gets activated later as previous active work items retire. try_to_grab_pending() which is used in the cancellation path unconditionally decremented nr_active whether the work item being cancelled is currently active or delayed, so cancelling a delayed work item makes nr_active underflow. This breaks max_active enforcement and triggers BUG_ON() in destroy_workqueue() later on. This patch fixes this bug by adding a flag WORK_STRUCT_DELAYED, which is set while a work item in on the delayed list and making try_to_grab_pending() decrement nr_active iff the work item is currently active. The addition of the flag enlarges cwq alignment to 256 bytes which is getting a bit too large. It's scheduled to be reduced back to 128 bytes by merging WORK_STRUCT_PENDING and WORK_STRUCT_CWQ in the next devel cycle. Signed-off-by: Tejun Heo Reported-by: Johannes Berg diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h index c959666..f11100f 100644 --- a/include/linux/workqueue.h +++ b/include/linux/workqueue.h @@ -25,18 +25,20 @@ typedef void (*work_func_t)(struct work_struct *work); enum { WORK_STRUCT_PENDING_BIT = 0, /* work item is pending execution */ - WORK_STRUCT_CWQ_BIT = 1, /* data points to cwq */ - WORK_STRUCT_LINKED_BIT = 2, /* next work is linked to this one */ + WORK_STRUCT_DELAYED_BIT = 1, /* work item is delayed */ + WORK_STRUCT_CWQ_BIT = 2, /* data points to cwq */ + WORK_STRUCT_LINKED_BIT = 3, /* next work is linked to this one */ #ifdef CONFIG_DEBUG_OBJECTS_WORK - WORK_STRUCT_STATIC_BIT = 3, /* static initializer (debugobjects) */ - WORK_STRUCT_COLOR_SHIFT = 4, /* color for workqueue flushing */ + WORK_STRUCT_STATIC_BIT = 4, /* static initializer (debugobjects) */ + WORK_STRUCT_COLOR_SHIFT = 5, /* color for workqueue flushing */ #else - WORK_STRUCT_COLOR_SHIFT = 3, /* color for workqueue flushing */ + WORK_STRUCT_COLOR_SHIFT = 4, /* color for workqueue flushing */ #endif WORK_STRUCT_COLOR_BITS = 4, WORK_STRUCT_PENDING = 1 << WORK_STRUCT_PENDING_BIT, + WORK_STRUCT_DELAYED = 1 << WORK_STRUCT_DELAYED_BIT, WORK_STRUCT_CWQ = 1 << WORK_STRUCT_CWQ_BIT, WORK_STRUCT_LINKED = 1 << WORK_STRUCT_LINKED_BIT, #ifdef CONFIG_DEBUG_OBJECTS_WORK @@ -59,8 +61,8 @@ enum { /* * Reserve 7 bits off of cwq pointer w/ debugobjects turned - * off. This makes cwqs aligned to 128 bytes which isn't too - * excessive while allowing 15 workqueue flush colors. + * off. This makes cwqs aligned to 256 bytes and allows 15 + * workqueue flush colors. */ WORK_STRUCT_FLAG_BITS = WORK_STRUCT_COLOR_SHIFT + WORK_STRUCT_COLOR_BITS, diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 362b50d..a2dccfc 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -941,6 +941,7 @@ static void __queue_work(unsigned int cpu, struct workqueue_struct *wq, struct global_cwq *gcwq; struct cpu_workqueue_struct *cwq; struct list_head *worklist; + unsigned int work_flags; unsigned long flags; debug_work_activate(work); @@ -990,14 +991,17 @@ static void __queue_work(unsigned int cpu, struct workqueue_struct *wq, BUG_ON(!list_empty(&work->entry)); cwq->nr_in_flight[cwq->work_color]++; + work_flags = work_color_to_flags(cwq->work_color); if (likely(cwq->nr_active < cwq->max_active)) { cwq->nr_active++; worklist = gcwq_determine_ins_pos(gcwq, cwq); - } else + } else { + work_flags |= WORK_STRUCT_DELAYED; worklist = &cwq->delayed_works; + } - insert_work(cwq, work, worklist, work_color_to_flags(cwq->work_color)); + insert_work(cwq, work, worklist, work_flags); spin_unlock_irqrestore(&gcwq->lock, flags); } @@ -1666,6 +1670,7 @@ static void cwq_activate_first_delayed(struct cpu_workqueue_struct *cwq) struct list_head *pos = gcwq_determine_ins_pos(cwq->gcwq, cwq); move_linked_works(work, pos, NULL); + __clear_bit(WORK_STRUCT_DELAYED_BIT, work_data_bits(work)); cwq->nr_active++; } @@ -1673,6 +1678,7 @@ static void cwq_activate_first_delayed(struct cpu_workqueue_struct *cwq) * cwq_dec_nr_in_flight - decrement cwq's nr_in_flight * @cwq: cwq of interest * @color: color of work which left the queue + * @delayed: for a delayed work * * A work either has completed or is removed from pending queue, * decrement nr_in_flight of its cwq and handle workqueue flushing. @@ -1680,19 +1686,22 @@ static void cwq_activate_first_delayed(struct cpu_workqueue_struct *cwq) * CONTEXT: * spin_lock_irq(gcwq->lock). */ -static void cwq_dec_nr_in_flight(struct cpu_workqueue_struct *cwq, int color) +static void cwq_dec_nr_in_flight(struct cpu_workqueue_struct *cwq, int color, + bool delayed) { /* ignore uncolored works */ if (color == WORK_NO_COLOR) return; cwq->nr_in_flight[color]--; - cwq->nr_active--; - if (!list_empty(&cwq->delayed_works)) { - /* one down, submit a delayed one */ - if (cwq->nr_active < cwq->max_active) - cwq_activate_first_delayed(cwq); + if (!delayed) { + cwq->nr_active--; + if (!list_empty(&cwq->delayed_works)) { + /* one down, submit a delayed one */ + if (cwq->nr_active < cwq->max_active) + cwq_activate_first_delayed(cwq); + } } /* is flush in progress and are we at the flushing tip? */ @@ -1823,7 +1832,7 @@ __acquires(&gcwq->lock) hlist_del_init(&worker->hentry); worker->current_work = NULL; worker->current_cwq = NULL; - cwq_dec_nr_in_flight(cwq, work_color); + cwq_dec_nr_in_flight(cwq, work_color, false); } /** @@ -2388,7 +2397,8 @@ static int try_to_grab_pending(struct work_struct *work) debug_work_deactivate(work); list_del_init(&work->entry); cwq_dec_nr_in_flight(get_work_cwq(work), - get_work_color(work)); + get_work_color(work), + *work_data_bits(work) & WORK_STRUCT_DELAYED); ret = 1; } } -- cgit v0.10.2 From 2a643ec67f9efc4b6921a3dd6e257f3b5360622b Mon Sep 17 00:00:00 2001 From: "Stephen M. Cameron" Date: Wed, 25 Aug 2010 19:58:53 +0200 Subject: cciss: fix reporting of max queue depth since init The ioctl path and the scsi tape path were not accounting for their additions to the queue depth. Signed-off-by: Stephen M. Cameron Signed-off-by: Jens Axboe diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c index 7191c16..6124c2f 100644 --- a/drivers/block/cciss.c +++ b/drivers/block/cciss.c @@ -297,6 +297,8 @@ static void enqueue_cmd_and_start_io(ctlr_info_t *h, spin_lock_irqsave(&h->lock, flags); addQ(&h->reqQ, c); h->Qdepth++; + if (h->Qdepth > h->maxQsinceinit) + h->maxQsinceinit = h->Qdepth; start_io(h); spin_unlock_irqrestore(&h->lock, flags); } -- cgit v0.10.2 From 6cb6cbeaa9d6573c674842b9471c0467967aae99 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Wed, 25 Aug 2010 08:58:50 -0700 Subject: MAINTAINERS: Add RCU's public git tree RCU now has a public git tree, so add it via the T: line in the MAINTAINERS file. Signed-off-by: Paul E. McKenney Cc: joe@perches.com LKML-Reference: <20100825155850.GA6601@linux.vnet.ibm.com> Signed-off-by: Ingo Molnar diff --git a/MAINTAINERS b/MAINTAINERS index 5fa8451..614861a 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4799,6 +4799,7 @@ RCUTORTURE MODULE M: Josh Triplett M: "Paul E. McKenney" S: Supported +T: git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-2.6-rcu.git F: Documentation/RCU/torture.txt F: kernel/rcutorture.c @@ -4823,6 +4824,7 @@ M: Dipankar Sarma M: "Paul E. McKenney" W: http://www.rdrop.com/users/paulmck/rclock/ S: Supported +T: git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-2.6-rcu.git F: Documentation/RCU/ F: include/linux/rcu* F: include/linux/srcu* -- cgit v0.10.2 From 6e63e80d88521a176989ed14b420f42dc418e46a Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 25 Aug 2010 21:33:29 +0200 Subject: PCI hotplug: Fix build with CONFIG_ACPI unset One of the recent changes caused complilation of drivers/pci/hotplug/pciehp_core.c to fail. Fix this issue. Signed-off-by: Rafael J. Wysocki Signed-off-by: Jesse Barnes diff --git a/drivers/pci/hotplug/pciehp.h b/drivers/pci/hotplug/pciehp.h index 653de6f..73d5139 100644 --- a/drivers/pci/hotplug/pciehp.h +++ b/drivers/pci/hotplug/pciehp.h @@ -178,5 +178,9 @@ static inline void pciehp_firmware_init(void) } #else #define pciehp_firmware_init() do {} while (0) +static inline int pciehp_acpi_slot_detection_check(struct pci_dev *dev) +{ + return 0; +} #endif /* CONFIG_ACPI */ #endif /* _PCIEHP_H */ -- cgit v0.10.2 From 5225c45899e872383ca39f5533d28ec63c54b39e Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Sun, 22 Aug 2010 04:29:17 +0200 Subject: perf: Initialize callchains roots's childen hits Each histogram entry has a callchain root that stores the callchain samples. However we forgot to initialize the tracking of children hits of these roots, which then got random values on their creation. The root children hits is multiplied by the minimum percentage of hits provided by the user, and the result becomes the minimum hits expected from children branches. If the random value due to the uninitialization is big enough, then this minimum number of hits can be huge and eventually filter every children branches. The end result was invisible callchains. All we need to fix this is to initialize the children hits of the root. Reported-by: Christoph Hellwig Signed-off-by: Frederic Weisbecker Cc: Ingo Molnar Cc: Peter Zijlstra Cc: Arnaldo Carvalho de Melo Cc: Paul Mackerras Cc: 2.6.32.x-2.6.35.y diff --git a/tools/perf/util/callchain.h b/tools/perf/util/callchain.h index 624a96c..6de4313 100644 --- a/tools/perf/util/callchain.h +++ b/tools/perf/util/callchain.h @@ -50,6 +50,7 @@ static inline void callchain_init(struct callchain_node *node) INIT_LIST_HEAD(&node->children); INIT_LIST_HEAD(&node->val); + node->children_hit = 0; node->parent = NULL; node->hit = 0; } -- cgit v0.10.2 From 6628bc74f1aa9c35dd386320bf7ec04f12edb1b3 Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Fri, 27 Aug 2010 09:15:09 +0200 Subject: writeback: do not lose wakeup events when forking bdi threads This patch fixes the following issue: INFO: task mount.nfs4:1120 blocked for more than 120 seconds. "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. mount.nfs4 D 00000000fffc6a21 0 1120 1119 0x00000000 ffff880235643948 0000000000000046 ffffffff00000000 ffffffff00000000 ffff880235643fd8 ffff880235314760 00000000001d44c0 ffff880235643fd8 00000000001d44c0 00000000001d44c0 00000000001d44c0 00000000001d44c0 Call Trace: [] schedule_timeout+0x34/0xf1 [] ? wait_for_common+0x3f/0x130 [] ? trace_hardirqs_on+0xd/0xf [] wait_for_common+0xd2/0x130 [] ? default_wake_function+0x0/0xf [] ? _raw_spin_unlock+0x26/0x2a [] wait_for_completion+0x18/0x1a [] sync_inodes_sb+0xca/0x1bc [] __sync_filesystem+0x47/0x7e [] sync_filesystem+0x47/0x4b [] generic_shutdown_super+0x22/0xd2 [] kill_anon_super+0x11/0x4f [] nfs4_kill_super+0x3f/0x72 [nfs] [] deactivate_locked_super+0x21/0x41 [] deactivate_super+0x40/0x45 [] mntput_no_expire+0xb8/0xed [] release_mounts+0x9a/0xb0 [] put_mnt_ns+0x6a/0x7b [] nfs_follow_remote_path+0x19a/0x296 [nfs] [] nfs4_try_mount+0x75/0xaf [nfs] [] nfs4_get_sb+0x276/0x2ff [nfs] [] vfs_kern_mount+0xb8/0x196 [] do_kern_mount+0x48/0xe8 [] do_mount+0x771/0x7e8 [] sys_mount+0x83/0xbd [] system_call_fastpath+0x16/0x1b The reason of this hang was a race condition: when the flusher thread is forking a bdi thread, we use 'kthread_run()', so we run it _before_ we make it visible in 'bdi->wb.task'. The bdi thread runs, does all works, and goes sleep. 'bdi->wb.task' is still NULL. And this is a dangerous time window. If at this time someone queues a work for this bdi, he does not see the bdi thread and wakes up the forker thread instead! But the forker has already forked this bdi thread, but just did not make it visible yet! The result is that we lose the wake up event for this bdi thread and the NFS4 code waits forever. To fix the problem, we should use 'ktrhead_create()' for creating bdi threads, then make them visible in 'bdi->wb.task', and only after this wake them up. This is exactly what this patch does. Signed-off-by: Artem Bityutskiy Signed-off-by: Jens Axboe diff --git a/mm/backing-dev.c b/mm/backing-dev.c index eaa4a5b..c2bf86f 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -445,8 +445,8 @@ static int bdi_forker_thread(void *ptr) switch (action) { case FORK_THREAD: __set_current_state(TASK_RUNNING); - task = kthread_run(bdi_writeback_thread, &bdi->wb, "flush-%s", - dev_name(bdi->dev)); + task = kthread_create(bdi_writeback_thread, &bdi->wb, + "flush-%s", dev_name(bdi->dev)); if (IS_ERR(task)) { /* * If thread creation fails, force writeout of @@ -457,10 +457,13 @@ static int bdi_forker_thread(void *ptr) /* * The spinlock makes sure we do not lose * wake-ups when racing with 'bdi_queue_work()'. + * And as soon as the bdi thread is visible, we + * can start it. */ spin_lock_bh(&bdi->wb_lock); bdi->wb.task = task; spin_unlock_bh(&bdi->wb_lock); + wake_up_process(task); } break; -- cgit v0.10.2 From a002d148426f40bc2b7dc066982eb177cdebeaaa Mon Sep 17 00:00:00 2001 From: Huang Shijie Date: Sun, 8 Aug 2010 14:39:07 +0200 Subject: percpu: fix a memory leak in pcpu_extend_area_map() The original code did not free the old map. This patch fixes it. tj: use @old as memcpy source instead of @chunk->map, and indentation and description update Signed-off-by: Huang Shijie Signed-off-by: Tejun Heo Cc: stable@kernel.org diff --git a/mm/percpu.c b/mm/percpu.c index e61dc2c..a1830d8 100644 --- a/mm/percpu.c +++ b/mm/percpu.c @@ -393,7 +393,9 @@ static int pcpu_extend_area_map(struct pcpu_chunk *chunk, int new_alloc) goto out_unlock; old_size = chunk->map_alloc * sizeof(chunk->map[0]); - memcpy(new, chunk->map, old_size); + old = chunk->map; + + memcpy(new, old, old_size); chunk->map_alloc = new_alloc; chunk->map = new; -- cgit v0.10.2 From 54157c44471f5e266508ac08d270f2bc5857e8bb Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Wed, 11 Aug 2010 11:19:19 +0900 Subject: percpu: fix a mismatch between code and comment When pcpu_build_alloc_info() searches best_upa value, it ignores current value if the number of waste units exceeds 1/3 of the number of total cpus. But the comment on the code says that it will ignore if wastage is over 25%. Modify the comment. Signed-off-by: Namhyung Kim Signed-off-by: Tejun Heo diff --git a/mm/percpu.c b/mm/percpu.c index a1830d8..58c572b 100644 --- a/mm/percpu.c +++ b/mm/percpu.c @@ -1164,7 +1164,7 @@ static struct pcpu_alloc_info * __init pcpu_build_alloc_info( } /* - * Don't accept if wastage is over 25%. The + * Don't accept if wastage is over 1/3. The * greater-than comparison ensures upa==1 always * passes the following check. */ -- cgit v0.10.2 From 7c38875a0d0a9b90eee66be79e36995c86acc70c Mon Sep 17 00:00:00 2001 From: Jason Baron Date: Thu, 26 Aug 2010 10:35:43 -0400 Subject: workqueue: Add a workqueue chapter to the tracepoint docbook Add a workqueue chapter to the traceopint docbook. Current book is at: http://www.kernel.org/doc/htmldocs/tracepoint/ Signed-off-by: Jason Baron Signed-off-by: Tejun Heo diff --git a/Documentation/DocBook/tracepoint.tmpl b/Documentation/DocBook/tracepoint.tmpl index e8473ea..b57a9ed 100644 --- a/Documentation/DocBook/tracepoint.tmpl +++ b/Documentation/DocBook/tracepoint.tmpl @@ -104,4 +104,9 @@ Block IO !Iinclude/trace/events/block.h + + + Workqueue +!Iinclude/trace/events/workqueue.h + -- cgit v0.10.2 From 557de5eb29caeab56c2f97b49eec02e32a2cbe32 Mon Sep 17 00:00:00 2001 From: Mike Rapoport Date: Sun, 22 Aug 2010 14:22:40 +0300 Subject: libertas: if_sdio: fix buffer alignment in struct if_sdio_card The commit 886275ce41a9751117367fb387ed171049eb6148 (param: lock if_sdio's lbs_helper_name and lbs_fw_name against sysfs changes) introduced new fields into the if_sdio_card structure. It caused missalignment of the if_sdio_card.buffer field and failure at driver load time: ~# modprobe libertas_sdio [ 62.315124] libertas_sdio: Libertas SDIO driver [ 62.319976] libertas_sdio: Copyright Pierre Ossman [ 63.020629] DMA misaligned error with device 48 [ 63.025207] mmci-omap-hs mmci-omap-hs.1: unexpected dma status 800 [ 66.005035] libertas: command 0x0003 timed out [ 66.009826] libertas: Timeout submitting command 0x0003 [ 66.016296] libertas: PREP_CMD: command 0x0003 failed: -110 Adding explicit alignment attribute for the if_sdio_card.buffer field fixes this problem. Signed-off-by: Mike Rapoport Acked-by: Marek Vasut Acked-by: Dan Williams Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/libertas/if_sdio.c b/drivers/net/wireless/libertas/if_sdio.c index 6e71346..382f237 100644 --- a/drivers/net/wireless/libertas/if_sdio.c +++ b/drivers/net/wireless/libertas/if_sdio.c @@ -126,7 +126,7 @@ struct if_sdio_card { const char *helper; const char *firmware; - u8 buffer[65536]; + u8 buffer[65536] __attribute__((aligned(4))); spinlock_t lock; struct if_sdio_packet *packets; -- cgit v0.10.2 From b76b4014f9d988d2412b873e4d4c13c7f9afc4e4 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Sat, 28 Aug 2010 08:52:10 +0200 Subject: writeback: Fix lost wake-up shutting down writeback thread Setting the task state here may cause us to miss the wake up from kthread_stop(), so we need to recheck kthread_should_stop() or risk sleeping forever in the following schedule(). Symptom was an indefinite hang on an NFSv4 mount. (NFSv4 may create multiple mounts in a temporary namespace while traversing the mount path, and since the temporary namespace is immediately destroyed, it may end up destroying a mount very soon after it was created, possibly making this race more likely.) INFO: task mount.nfs4:4314 blocked for more than 120 seconds. "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. mount.nfs4 D 0000000000000000 2880 4314 4313 0x00000000 ffff88001ed6da28 0000000000000046 ffff88001ed6dfd8 ffff88001ed6dfd8 ffff88001ed6c000 ffff88001ed6c000 ffff88001ed6c000 ffff88001e5003a0 ffff88001ed6dfd8 ffff88001e5003a8 ffff88001ed6c000 ffff88001ed6dfd8 Call Trace: [] schedule_timeout+0x1cd/0x2e0 [] ? mark_held_locks+0x6c/0xa0 [] ? _raw_spin_unlock_irq+0x30/0x60 [] ? trace_hardirqs_on_caller+0x14d/0x190 [] ? sub_preempt_count+0xe/0xd0 [] wait_for_common+0x120/0x190 [] ? default_wake_function+0x0/0x20 [] wait_for_completion+0x1d/0x20 [] kthread_stop+0x4a/0x150 [] ? thaw_process+0x70/0x80 [] bdi_unregister+0x10a/0x1a0 [] nfs_put_super+0x19/0x20 [] generic_shutdown_super+0x54/0xe0 [] kill_anon_super+0x16/0x60 [] nfs4_kill_super+0x39/0x90 [] deactivate_locked_super+0x45/0x60 [] deactivate_super+0x49/0x70 [] mntput_no_expire+0x84/0xe0 [] release_mounts+0x9f/0xc0 [] put_mnt_ns+0x65/0x80 [] nfs_follow_remote_path+0x1e6/0x420 [] nfs4_try_mount+0x6f/0xd0 [] nfs4_get_sb+0xa2/0x360 [] vfs_kern_mount+0x88/0x1f0 [] do_kern_mount+0x52/0x130 [] ? _lock_kernel+0x6a/0x170 [] do_mount+0x26e/0x7f0 [] ? copy_mount_options+0xea/0x190 [] sys_mount+0x98/0xf0 [] system_call_fastpath+0x16/0x1b 1 lock held by mount.nfs4/4314: #0: (&type->s_umount_key#24){+.+...}, at: [] deactivate_super+0x41/0x70 Signed-off-by: J. Bruce Fields Signed-off-by: Jens Axboe Acked-by: Artem Bityutskiy diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 7d9d06b..81e086d 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -808,7 +808,7 @@ int bdi_writeback_thread(void *data) wb->last_active = jiffies; set_current_state(TASK_INTERRUPTIBLE); - if (!list_empty(&bdi->work_list)) { + if (!list_empty(&bdi->work_list) || kthread_should_stop()) { __set_current_state(TASK_RUNNING); continue; } -- cgit v0.10.2 From 7619b1b2e2b96842459f499fd48ec32e3d223d02 Mon Sep 17 00:00:00 2001 From: Ken Kawasaki Date: Sat, 28 Aug 2010 12:45:01 +0000 Subject: pcnet_cs: add new_id pcnet_cs: add new_id: "KENTRONICS KEP-230" 10Base-T PCMCIA card. Signed-off-by: Ken Kawasaki Signed-off-by: David S. Miller diff --git a/drivers/net/pcmcia/pcnet_cs.c b/drivers/net/pcmcia/pcnet_cs.c index bfdef72..7fd8c55 100644 --- a/drivers/net/pcmcia/pcnet_cs.c +++ b/drivers/net/pcmcia/pcnet_cs.c @@ -1644,6 +1644,7 @@ static struct pcmcia_device_id pcnet_ids[] = { PCMCIA_DEVICE_PROD_ID12("IO DATA", "PCETTX", 0x547e66dc, 0x6fc5459b), PCMCIA_DEVICE_PROD_ID12("iPort", "10/100 Ethernet Card", 0x56c538d2, 0x11b0ffc0), PCMCIA_DEVICE_PROD_ID12("KANSAI ELECTRIC CO.,LTD", "KLA-PCM/T", 0xb18dc3b4, 0xcc51a956), + PCMCIA_DEVICE_PROD_ID12("KENTRONICS", "KEP-230", 0xaf8144c9, 0x868f6616), PCMCIA_DEVICE_PROD_ID12("KCI", "PE520 PCMCIA Ethernet Adapter", 0xa89b87d3, 0x1eb88e64), PCMCIA_DEVICE_PROD_ID12("KINGMAX", "EN10T2T", 0x7bcb459a, 0xa5c81fa5), PCMCIA_DEVICE_PROD_ID12("Kingston", "KNE-PC2", 0x1128e633, 0xce2a89b3), -- cgit v0.10.2 From d9f66c1a46163c7c83411058516a69da547262f8 Mon Sep 17 00:00:00 2001 From: Mike Auty Date: Sat, 28 Aug 2010 20:35:17 -0700 Subject: Input: wacom - fix mousewheel handling for old wacom tablets This fixes a regression introduced in 3b57ca0f80c5c8994b5b1e3d3f904cfe727951f2. The data[6] byte contains either 1 or -1 depending on the whether the mouse wheel on older wacom tablets is moved down (1) or up (-1). The patch introduced in the above commit changed the cast from (signed char) to (signed). When cast as a signed integer and negated, the value of -1 (stored in the byte as 0xff) became -255 rather than 1. This patch reverts the cast to a (signed char) and also removes an unnecessary (signed) cast, as all the values operated on are bitmasked. Signed-off-by: Mike Auty Reviewed-by: Ping Cheng Cc; stable@kernel.org Signed-off-by: Dmitry Torokhov diff --git a/drivers/input/tablet/wacom_wac.c b/drivers/input/tablet/wacom_wac.c index 40d77ba..6e29bad 100644 --- a/drivers/input/tablet/wacom_wac.c +++ b/drivers/input/tablet/wacom_wac.c @@ -243,10 +243,10 @@ static int wacom_graphire_irq(struct wacom_wac *wacom) if (features->type == WACOM_G4 || features->type == WACOM_MO) { input_report_abs(input, ABS_DISTANCE, data[6] & 0x3f); - rw = (signed)(data[7] & 0x04) - (data[7] & 0x03); + rw = (data[7] & 0x04) - (data[7] & 0x03); } else { input_report_abs(input, ABS_DISTANCE, data[7] & 0x3f); - rw = -(signed)data[6]; + rw = -(signed char)data[6]; } input_report_rel(input, REL_WHEEL, rw); } -- cgit v0.10.2 From 2c4e9671edfef534e9726366707d64e63d44e7e6 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Sat, 28 Aug 2010 20:35:17 -0700 Subject: Input: use PIT_TICK_RATE in vt beep ioctl The KIOCSOUND and KDMKTONE ioctls are based on the CLOCK_TICK_RATE, which is architecture and sometimes configuration specific. In practice, most user applications assume that it is actually defined as the i8253 PIT base clock of 1193182 Hz, which is true on some architectures but not on others. This patch makes the vt code use the PIT frequency on all architectures, which is much more well-defined. It will change the behavior of user applications sending the beep ioctl on all architectures that define CLOCK_TICK_RATE different from PIT_TICK_RATE. The original breakage was introduced in commit bcc8ca099 "Adapt drivers/char/vt_ioctl.c to non-x86". Hopefully, reverting this change will make the frequency correct in more cases than it will make it incorrect. Signed-off-by: Arnd Bergmann Acked-by: Alan Cox Signed-off-by: Dmitry Torokhov diff --git a/drivers/char/vt_ioctl.c b/drivers/char/vt_ioctl.c index 2bbeaae..38df8c1 100644 --- a/drivers/char/vt_ioctl.c +++ b/drivers/char/vt_ioctl.c @@ -533,11 +533,14 @@ int vt_ioctl(struct tty_struct *tty, struct file * file, case KIOCSOUND: if (!perm) goto eperm; - /* FIXME: This is an old broken API but we need to keep it - supported and somehow separate the historic advertised - tick rate from any real one */ + /* + * The use of PIT_TICK_RATE is historic, it used to be + * the platform-dependent CLOCK_TICK_RATE between 2.6.12 + * and 2.6.36, which was a minor but unfortunate ABI + * change. + */ if (arg) - arg = CLOCK_TICK_RATE / arg; + arg = PIT_TICK_RATE / arg; kd_mksound(arg, 0); break; @@ -553,11 +556,8 @@ int vt_ioctl(struct tty_struct *tty, struct file * file, */ ticks = HZ * ((arg >> 16) & 0xffff) / 1000; count = ticks ? (arg & 0xffff) : 0; - /* FIXME: This is an old broken API but we need to keep it - supported and somehow separate the historic advertised - tick rate from any real one */ if (count) - count = CLOCK_TICK_RATE / count; + count = PIT_TICK_RATE / count; kd_mksound(count, ticks); break; } -- cgit v0.10.2 From ba4d695a90c9176fca8e45d6c872bbf4e8bed315 Mon Sep 17 00:00:00 2001 From: Henrik Rydberg Date: Sat, 28 Aug 2010 21:33:50 -0700 Subject: Input: MT - initialize slots to unused For MT slots, the ABS_MT_TRACKING_ID determines whether a slot is in use, but currently leaves initialization up to the drivers. This patch sets the slot state to unused upon creation. Signed-off-by: Henrik Rydberg Signed-off-by: Dmitry Torokhov diff --git a/drivers/input/input.c b/drivers/input/input.c index a9b025f..ab69820 100644 --- a/drivers/input/input.c +++ b/drivers/input/input.c @@ -1599,11 +1599,14 @@ EXPORT_SYMBOL(input_free_device); * @dev: input device supporting MT events and finger tracking * @num_slots: number of slots used by the device * - * This function allocates all necessary memory for MT slot handling - * in the input device, and adds ABS_MT_SLOT to the device capabilities. + * This function allocates all necessary memory for MT slot handling in the + * input device, and adds ABS_MT_SLOT to the device capabilities. All slots + * are initially marked as unused iby setting ABS_MT_TRACKING_ID to -1. */ int input_mt_create_slots(struct input_dev *dev, unsigned int num_slots) { + int i; + if (!num_slots) return 0; @@ -1614,6 +1617,10 @@ int input_mt_create_slots(struct input_dev *dev, unsigned int num_slots) dev->mtsize = num_slots; input_set_abs_params(dev, ABS_MT_SLOT, 0, num_slots - 1, 0, 0); + /* Mark slots as 'unused' */ + for (i = 0; i < num_slots; i++) + dev->mt[i].abs[ABS_MT_TRACKING_ID - ABS_MT_FIRST] = -1; + return 0; } EXPORT_SYMBOL(input_mt_create_slots); -- cgit v0.10.2 From 8807286e569c4f12fa2bc980187f3e2abc606d11 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 10 Aug 2010 12:33:20 -0400 Subject: drm/radeon/kms: use tracked values for sclk and mclk Rather than calling get_memory_clock and get_engine_clock, used the tracked values from the pm code. Calling the tables adds additional latency in the modesetting and pm paths. Signed-off-by: Alex Deucher Signed-off-by: Dave Airlie diff --git a/drivers/gpu/drm/radeon/radeon_clocks.c b/drivers/gpu/drm/radeon/radeon_clocks.c index 14448a7..690d8907 100644 --- a/drivers/gpu/drm/radeon/radeon_clocks.c +++ b/drivers/gpu/drm/radeon/radeon_clocks.c @@ -327,6 +327,14 @@ void radeon_get_clock_info(struct drm_device *dev) mpll->max_feedback_div = 0xff; mpll->best_vco = 0; + if (!rdev->clock.default_sclk) + rdev->clock.default_sclk = radeon_get_engine_clock(rdev); + if ((!rdev->clock.default_mclk) && rdev->asic->get_memory_clock) + rdev->clock.default_mclk = radeon_get_memory_clock(rdev); + + rdev->pm.current_sclk = rdev->clock.default_sclk; + rdev->pm.current_mclk = rdev->clock.default_mclk; + } /* 10 khz */ diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index 69b3c22..256d204 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -293,30 +293,20 @@ bool radeon_card_posted(struct radeon_device *rdev) void radeon_update_bandwidth_info(struct radeon_device *rdev) { fixed20_12 a; - u32 sclk, mclk; + u32 sclk = rdev->pm.current_sclk; + u32 mclk = rdev->pm.current_mclk; - if (rdev->flags & RADEON_IS_IGP) { - sclk = radeon_get_engine_clock(rdev); - mclk = rdev->clock.default_mclk; - - a.full = dfixed_const(100); - rdev->pm.sclk.full = dfixed_const(sclk); - rdev->pm.sclk.full = dfixed_div(rdev->pm.sclk, a); - rdev->pm.mclk.full = dfixed_const(mclk); - rdev->pm.mclk.full = dfixed_div(rdev->pm.mclk, a); + /* sclk/mclk in Mhz */ + a.full = dfixed_const(100); + rdev->pm.sclk.full = dfixed_const(sclk); + rdev->pm.sclk.full = dfixed_div(rdev->pm.sclk, a); + rdev->pm.mclk.full = dfixed_const(mclk); + rdev->pm.mclk.full = dfixed_div(rdev->pm.mclk, a); + if (rdev->flags & RADEON_IS_IGP) { a.full = dfixed_const(16); /* core_bandwidth = sclk(Mhz) * 16 */ rdev->pm.core_bandwidth.full = dfixed_div(rdev->pm.sclk, a); - } else { - sclk = radeon_get_engine_clock(rdev); - mclk = radeon_get_memory_clock(rdev); - - a.full = dfixed_const(100); - rdev->pm.sclk.full = dfixed_const(sclk); - rdev->pm.sclk.full = dfixed_div(rdev->pm.sclk, a); - rdev->pm.mclk.full = dfixed_const(mclk); - rdev->pm.mclk.full = dfixed_div(rdev->pm.mclk, a); } } diff --git a/drivers/gpu/drm/radeon/radeon_i2c.c b/drivers/gpu/drm/radeon/radeon_i2c.c index 0416804..6a13ee3 100644 --- a/drivers/gpu/drm/radeon/radeon_i2c.c +++ b/drivers/gpu/drm/radeon/radeon_i2c.c @@ -213,7 +213,7 @@ static void post_xfer(struct i2c_adapter *i2c_adap) static u32 radeon_get_i2c_prescale(struct radeon_device *rdev) { - u32 sclk = radeon_get_engine_clock(rdev); + u32 sclk = rdev->pm.current_sclk; u32 prescale = 0; u32 nm; u8 n, m, loop; -- cgit v0.10.2 From 87cbf8f2c5d1b1fc4642c3dc0bb6efc587479603 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Fri, 27 Aug 2010 13:59:54 -0400 Subject: drm/radeon/kms: fix a regression on r7xx AGP due to the HDP flush fix commit: 812d046915f48236657f02c06d7dc47140e9ceda drm/radeon/kms/r7xx: add workaround for hw issue with HDP flush breaks on AGP boards since there is no VRAM gart table. This patch fixes the issue by creating a VRAM scratch page so that can be used on both AGP and PCIE. Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=29834 Signed-off-by: Alex Deucher Cc: stable@kernel.org Signed-off-by: Dave Airlie diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index d0ebae9..aee8376 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -3541,7 +3541,7 @@ void r600_ioctl_wait_idle(struct radeon_device *rdev, struct radeon_bo *bo) * rather than write to HDP_REG_COHERENCY_FLUSH_CNTL */ if ((rdev->family >= CHIP_RV770) && (rdev->family <= CHIP_RV740)) { - void __iomem *ptr = (void *)rdev->gart.table.vram.ptr; + void __iomem *ptr = (void *)rdev->vram_scratch.ptr; u32 tmp; WREG32(HDP_DEBUG1, 0); diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index 3dfcfa3..e90d9e3 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -1013,6 +1013,11 @@ int radeon_gem_set_tiling_ioctl(struct drm_device *dev, void *data, int radeon_gem_get_tiling_ioctl(struct drm_device *dev, void *data, struct drm_file *filp); +/* VRAM scratch page for HDP bug */ +struct r700_vram_scratch { + struct radeon_bo *robj; + volatile uint32_t *ptr; +}; /* * Core structure, functions and helpers. @@ -1079,6 +1084,7 @@ struct radeon_device { const struct firmware *pfp_fw; /* r6/700 PFP firmware */ const struct firmware *rlc_fw; /* r6/700 RLC firmware */ struct r600_blit r600_blit; + struct r700_vram_scratch vram_scratch; int msi_enabled; /* msi enabled */ struct r600_ih ih; /* r6/700 interrupt ring */ struct workqueue_struct *wq; diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c index f1c79681..a84e386 100644 --- a/drivers/gpu/drm/radeon/rv770.c +++ b/drivers/gpu/drm/radeon/rv770.c @@ -905,6 +905,54 @@ static void rv770_gpu_init(struct radeon_device *rdev) } +static int rv770_vram_scratch_init(struct radeon_device *rdev) +{ + int r; + u64 gpu_addr; + + if (rdev->vram_scratch.robj == NULL) { + r = radeon_bo_create(rdev, NULL, RADEON_GPU_PAGE_SIZE, + true, RADEON_GEM_DOMAIN_VRAM, + &rdev->vram_scratch.robj); + if (r) { + return r; + } + } + + r = radeon_bo_reserve(rdev->vram_scratch.robj, false); + if (unlikely(r != 0)) + return r; + r = radeon_bo_pin(rdev->vram_scratch.robj, + RADEON_GEM_DOMAIN_VRAM, &gpu_addr); + if (r) { + radeon_bo_unreserve(rdev->vram_scratch.robj); + return r; + } + r = radeon_bo_kmap(rdev->vram_scratch.robj, + (void **)&rdev->vram_scratch.ptr); + if (r) + radeon_bo_unpin(rdev->vram_scratch.robj); + radeon_bo_unreserve(rdev->vram_scratch.robj); + + return r; +} + +static void rv770_vram_scratch_fini(struct radeon_device *rdev) +{ + int r; + + if (rdev->vram_scratch.robj == NULL) { + return; + } + r = radeon_bo_reserve(rdev->vram_scratch.robj, false); + if (likely(r == 0)) { + radeon_bo_kunmap(rdev->vram_scratch.robj); + radeon_bo_unpin(rdev->vram_scratch.robj); + radeon_bo_unreserve(rdev->vram_scratch.robj); + } + radeon_bo_unref(&rdev->vram_scratch.robj); +} + int rv770_mc_init(struct radeon_device *rdev) { u32 tmp; @@ -970,6 +1018,9 @@ static int rv770_startup(struct radeon_device *rdev) if (r) return r; } + r = rv770_vram_scratch_init(rdev); + if (r) + return r; rv770_gpu_init(rdev); r = r600_blit_init(rdev); if (r) { @@ -1195,6 +1246,7 @@ void rv770_fini(struct radeon_device *rdev) r600_irq_fini(rdev); radeon_irq_kms_fini(rdev); rv770_pcie_gart_fini(rdev); + rv770_vram_scratch_fini(rdev); radeon_gem_fini(rdev); radeon_fence_driver_fini(rdev); radeon_clocks_fini(rdev); -- cgit v0.10.2 From 4afc31345e5f543e5d89a47aeadaaad1d91a5bc8 Mon Sep 17 00:00:00 2001 From: Ryusuke Konishi Date: Sun, 29 Aug 2010 01:55:38 +0900 Subject: nilfs2: fix leak of shadow dat inode in error path of load_nilfs If load_nilfs() gets an error while doing recovery, it will fail to free the shadow inode of dat (nilfs->ns_gc_dat). This fixes the leak issue. Signed-off-by: Ryusuke Konishi diff --git a/fs/nilfs2/the_nilfs.c b/fs/nilfs2/the_nilfs.c index 4317f17..ba7c10c 100644 --- a/fs/nilfs2/the_nilfs.c +++ b/fs/nilfs2/the_nilfs.c @@ -446,6 +446,7 @@ int load_nilfs(struct the_nilfs *nilfs, struct nilfs_sb_info *sbi) nilfs_mdt_destroy(nilfs->ns_cpfile); nilfs_mdt_destroy(nilfs->ns_sufile); nilfs_mdt_destroy(nilfs->ns_dat); + nilfs_mdt_destroy(nilfs->ns_gc_dat); failed: nilfs_clear_recovery_info(&ri); -- cgit v0.10.2 From 50e77fcd790e02a095f906df0f6fa72afef7e11c Mon Sep 17 00:00:00 2001 From: Eric Miao Date: Wed, 18 Aug 2010 11:51:13 +0800 Subject: ARM: pxa: remove __init from cpufreq_driver->init() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This is causing section mismatches when linking, as cpufreq_driver->init() is not supposed to be in init section. Reported-by: Tomáš 'Sleep_Walker' Čech Signed-off-by: Eric Miao diff --git a/arch/arm/mach-pxa/cpufreq-pxa2xx.c b/arch/arm/mach-pxa/cpufreq-pxa2xx.c index 268a9bc..50d5939 100644 --- a/arch/arm/mach-pxa/cpufreq-pxa2xx.c +++ b/arch/arm/mach-pxa/cpufreq-pxa2xx.c @@ -398,7 +398,7 @@ static int pxa_set_target(struct cpufreq_policy *policy, return 0; } -static __init int pxa_cpufreq_init(struct cpufreq_policy *policy) +static int pxa_cpufreq_init(struct cpufreq_policy *policy) { int i; unsigned int freq; diff --git a/arch/arm/mach-pxa/cpufreq-pxa3xx.c b/arch/arm/mach-pxa/cpufreq-pxa3xx.c index 27fa329..0a0d0fe 100644 --- a/arch/arm/mach-pxa/cpufreq-pxa3xx.c +++ b/arch/arm/mach-pxa/cpufreq-pxa3xx.c @@ -204,7 +204,7 @@ static int pxa3xx_cpufreq_set(struct cpufreq_policy *policy, return 0; } -static __init int pxa3xx_cpufreq_init(struct cpufreq_policy *policy) +static int pxa3xx_cpufreq_init(struct cpufreq_policy *policy) { int ret = -EINVAL; -- cgit v0.10.2 From be7a6d9dd67251675b83a29f4b15392eae62e70b Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Fri, 13 Aug 2010 11:39:07 +0800 Subject: ARM: pxa: fix CI_HSYNC and CI_VSYNC MFP defines for pxa300 According to PXA3xx Processor Family Developer Manuall Vol1. section "Pin Descriptions and Control", PXA30x and PXA31x Processor Alternate Function Table shows the Alt FN 0 for GPIO51 is CI_HSYNC and for GPIO52 is CI_VSYNC. This patch fixes the MFP defines and also corrects the order of MFD defines. Signed-off-by: Axel Lin Signed-off-by: Eric Miao diff --git a/arch/arm/mach-pxa/include/mach/mfp-pxa300.h b/arch/arm/mach-pxa/include/mach/mfp-pxa300.h index 7139e0d..4e12870 100644 --- a/arch/arm/mach-pxa/include/mach/mfp-pxa300.h +++ b/arch/arm/mach-pxa/include/mach/mfp-pxa300.h @@ -71,10 +71,10 @@ #define GPIO46_CI_DD_7 MFP_CFG_DRV(GPIO46, AF0, DS04X) #define GPIO47_CI_DD_8 MFP_CFG_DRV(GPIO47, AF1, DS04X) #define GPIO48_CI_DD_9 MFP_CFG_DRV(GPIO48, AF1, DS04X) -#define GPIO52_CI_HSYNC MFP_CFG_DRV(GPIO52, AF0, DS04X) -#define GPIO51_CI_VSYNC MFP_CFG_DRV(GPIO51, AF0, DS04X) #define GPIO49_CI_MCLK MFP_CFG_DRV(GPIO49, AF0, DS04X) #define GPIO50_CI_PCLK MFP_CFG_DRV(GPIO50, AF0, DS04X) +#define GPIO51_CI_HSYNC MFP_CFG_DRV(GPIO51, AF0, DS04X) +#define GPIO52_CI_VSYNC MFP_CFG_DRV(GPIO52, AF0, DS04X) /* KEYPAD */ #define GPIO3_KP_DKIN_6 MFP_CFG_LPM(GPIO3, AF2, FLOAT) -- cgit v0.10.2 From b3282ab116cd7ebd2829ac8088524b5b279330e5 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= Date: Thu, 12 Aug 2010 08:18:28 +0200 Subject: ARM: pxa: Make id const in pwm_probe() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This fixes the compiler warning: arch/arm/plat-pxa/pwm.c: In function 'pwm_probe': arch/arm/plat-pxa/pwm.c:179: warning: initialization discards qualifiers from pointer target type Signed-off-by: Uwe Kleine-König Signed-off-by: Eric Miao diff --git a/arch/arm/plat-pxa/pwm.c b/arch/arm/plat-pxa/pwm.c index 0732c6c..ef32686 100644 --- a/arch/arm/plat-pxa/pwm.c +++ b/arch/arm/plat-pxa/pwm.c @@ -176,7 +176,7 @@ static inline void __add_pwm(struct pwm_device *pwm) static int __devinit pwm_probe(struct platform_device *pdev) { - struct platform_device_id *id = platform_get_device_id(pdev); + const struct platform_device_id *id = platform_get_device_id(pdev); struct pwm_device *pwm, *secondary = NULL; struct resource *r; int ret = 0; -- cgit v0.10.2 From ffb287c9daabc3734a92def0d59796f065ad29eb Mon Sep 17 00:00:00 2001 From: Marek Vasut Date: Thu, 26 Aug 2010 05:51:57 +0200 Subject: ARM: pxa168fb: fix section mismatch Signed-off-by: Marek Vasut Acked-by: Haojian Zhuang Signed-off-by: Eric Miao diff --git a/drivers/video/pxa168fb.c b/drivers/video/pxa168fb.c index c91a7f7..5d786bd 100644 --- a/drivers/video/pxa168fb.c +++ b/drivers/video/pxa168fb.c @@ -559,7 +559,7 @@ static struct fb_ops pxa168fb_ops = { .fb_imageblit = cfb_imageblit, }; -static int __init pxa168fb_init_mode(struct fb_info *info, +static int __devinit pxa168fb_init_mode(struct fb_info *info, struct pxa168fb_mach_info *mi) { struct pxa168fb_info *fbi = info->par; @@ -599,7 +599,7 @@ static int __init pxa168fb_init_mode(struct fb_info *info, return ret; } -static int __init pxa168fb_probe(struct platform_device *pdev) +static int __devinit pxa168fb_probe(struct platform_device *pdev) { struct pxa168fb_mach_info *mi; struct fb_info *info = 0; @@ -792,7 +792,7 @@ static struct platform_driver pxa168fb_driver = { .probe = pxa168fb_probe, }; -static int __devinit pxa168fb_init(void) +static int __init pxa168fb_init(void) { return platform_driver_register(&pxa168fb_driver); } -- cgit v0.10.2 From 01ebc12f5f2e88a1c6a5436b71a506ac2bf66d6b Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Sat, 7 Aug 2010 11:09:29 +0200 Subject: UBI: eliminate update of list_for_each_entry loop cursor list_for_each_entry uses its first argument to move from one element to the next, so modifying it can break the iteration. The variable re1 is already used within the loop as a temporary variable, and is not live here. The semantic match that finds this problem is as follows: (http://coccinelle.lip6.fr/) // @r@ iterator name list_for_each_entry; expression x,E; position p1,p2; @@ list_for_each_entry@p1(x,...) { <... x =@p2 E ...> } @@ expression x,E; position r.p1,r.p2; statement S; @@ *x =@p2 E ... list_for_each_entry@p1(x,...) S // Signed-off-by: Julia Lawall Signed-off-by: Artem Bityutskiy diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c index 4dfa6b9..3d2d1a6 100644 --- a/drivers/mtd/ubi/cdev.c +++ b/drivers/mtd/ubi/cdev.c @@ -798,18 +798,18 @@ static int rename_volumes(struct ubi_device *ubi, goto out_free; } - re = kzalloc(sizeof(struct ubi_rename_entry), GFP_KERNEL); - if (!re) { + re1 = kzalloc(sizeof(struct ubi_rename_entry), GFP_KERNEL); + if (!re1) { err = -ENOMEM; ubi_close_volume(desc); goto out_free; } - re->remove = 1; - re->desc = desc; - list_add(&re->list, &rename_list); + re1->remove = 1; + re1->desc = desc; + list_add(&re1->list, &rename_list); dbg_msg("will remove volume %d, name \"%s\"", - re->desc->vol->vol_id, re->desc->vol->name); + re1->desc->vol->vol_id, re1->desc->vol->name); } mutex_lock(&ubi->device_mutex); -- cgit v0.10.2 From 80c1c16fb89e7055acb560e6d90c881b5c7496c1 Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Sat, 28 Aug 2010 12:11:40 +0300 Subject: UBI: fix forward compatibility Commit 0798cea8c2e1afee59686c51d27d0e96b05e42d1 "UBI: improve corrupted flash handling" broke delet-compatible volumes handling - it introduced a limit of 8 eraseblocks which may be corrupted. And delete-compatible eraseblocks are added to the "corrupted" list, so if we'd have a large delete-compatible volume, UBI would refuse it. The fix is to add delete-compatible volumes to the erase list instead. Indeed, they are corrupted, we just have to erase them. Signed-off-by: Artem Bityutskiy diff --git a/drivers/mtd/ubi/scan.c b/drivers/mtd/ubi/scan.c index 372a15a..69b52e9 100644 --- a/drivers/mtd/ubi/scan.c +++ b/drivers/mtd/ubi/scan.c @@ -843,7 +843,7 @@ static int process_eb(struct ubi_device *ubi, struct ubi_scan_info *si, case UBI_COMPAT_DELETE: ubi_msg("\"delete\" compatible internal volume %d:%d" " found, will remove it", vol_id, lnum); - err = add_to_list(si, pnum, ec, &si->corr); + err = add_to_list(si, pnum, ec, &si->erase); if (err) return err; return 0; -- cgit v0.10.2 From 7c44ece988a3419e51c833cf14be186a22224dd1 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 30 Aug 2010 17:35:52 +1000 Subject: Move .gitignore from drivers/md to lib/raid6 Another missing bit of the raid6 -> /lib move. Reported-by: Andreas Schwab Signed-off-by: NeilBrown diff --git a/drivers/md/.gitignore b/drivers/md/.gitignore deleted file mode 100644 index a7afec6..0000000 --- a/drivers/md/.gitignore +++ /dev/null @@ -1,4 +0,0 @@ -mktables -raid6altivec*.c -raid6int*.c -raid6tables.c diff --git a/lib/raid6/.gitignore b/lib/raid6/.gitignore new file mode 100644 index 0000000..162beca --- /dev/null +++ b/lib/raid6/.gitignore @@ -0,0 +1,4 @@ +mktables +altivec*.c +int*.c +tables.c -- cgit v0.10.2 From bd52b746262c8d77e73903d6608014fb2fcdcd9d Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Mon, 30 Aug 2010 17:33:33 +1000 Subject: md: don't clear MD_CHANGE_CLEAN in md_update_sb() for external arrays If this bit is cleared in md_update_sb() the kernel will allow writes to the array if userspace triggers md_allow_write(), e.g. through stripe_cache_size, when mdmon is not active. When mdmon is active the array transitions to active-idle bypassing write-pending, setting up a race for mdmon to set the array clean before a write arrives. Signed-off-by: Dan Williams Signed-off-by: NeilBrown diff --git a/drivers/md/md.c b/drivers/md/md.c index c148b63..a1f6b59 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -2167,9 +2167,10 @@ repeat: rdev->recovery_offset = mddev->curr_resync_completed; } - if (mddev->external || !mddev->persistent) { + if (!mddev->persistent) { + if (!mddev->external) + clear_bit(MD_CHANGE_CLEAN, &mddev->flags); clear_bit(MD_CHANGE_DEVS, &mddev->flags); - clear_bit(MD_CHANGE_CLEAN, &mddev->flags); wake_up(&mddev->sb_wait); return; } -- cgit v0.10.2 From 070dc6dd7103b6b3f7e4d46e754354a5c15f366e Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 30 Aug 2010 17:33:34 +1000 Subject: md: resolve confusion of MD_CHANGE_CLEAN MD_CHANGE_CLEAN is used for two different purposes and this leads to confusion. One of the purposes is largely mirrored by MD_CHANGE_PENDING which is not used for anything else, so have MD_CHANGE_PENDING take over that purpose fully. The two purposes are: 1/ tell md_update_sb that an update is needed and that it is just a clean/dirty transition. 2/ tell user-space that an transition from clean to dirty is pending (something wants to write), and tell te kernel (by clearin the flag) that the transition is OK. The first purpose remains wit MD_CHANGE_CLEAN, the second is moved fully to MD_CHANGE_PENDING. This means that various places which conditionally set or cleared MD_CHANGE_CLEAN no longer need to be conditional. Signed-off-by: NeilBrown diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c index 1ba1e12..ed4900a 100644 --- a/drivers/md/bitmap.c +++ b/drivers/md/bitmap.c @@ -1542,8 +1542,7 @@ void bitmap_cond_end_sync(struct bitmap *bitmap, sector_t sector) atomic_read(&bitmap->mddev->recovery_active) == 0); bitmap->mddev->curr_resync_completed = bitmap->mddev->curr_resync; - if (bitmap->mddev->persistent) - set_bit(MD_CHANGE_CLEAN, &bitmap->mddev->flags); + set_bit(MD_CHANGE_CLEAN, &bitmap->mddev->flags); sector &= ~((1ULL << CHUNK_BLOCK_SHIFT(bitmap)) - 1); s = 0; while (s < sector && s < bitmap->mddev->resync_max_sectors) { diff --git a/drivers/md/md.c b/drivers/md/md.c index a1f6b59..43cf9cc 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -2168,8 +2168,7 @@ repeat: } if (!mddev->persistent) { - if (!mddev->external) - clear_bit(MD_CHANGE_CLEAN, &mddev->flags); + clear_bit(MD_CHANGE_CLEAN, &mddev->flags); clear_bit(MD_CHANGE_DEVS, &mddev->flags); wake_up(&mddev->sb_wait); return; @@ -2179,7 +2178,6 @@ repeat: mddev->utime = get_seconds(); - set_bit(MD_CHANGE_PENDING, &mddev->flags); if (test_and_clear_bit(MD_CHANGE_DEVS, &mddev->flags)) force_change = 1; if (test_and_clear_bit(MD_CHANGE_CLEAN, &mddev->flags)) @@ -3372,7 +3370,7 @@ array_state_show(mddev_t *mddev, char *page) case 0: if (mddev->in_sync) st = clean; - else if (test_bit(MD_CHANGE_CLEAN, &mddev->flags)) + else if (test_bit(MD_CHANGE_PENDING, &mddev->flags)) st = write_pending; else if (mddev->safemode) st = active_idle; @@ -3453,9 +3451,7 @@ array_state_store(mddev_t *mddev, const char *buf, size_t len) mddev->in_sync = 1; if (mddev->safemode == 1) mddev->safemode = 0; - if (mddev->persistent) - set_bit(MD_CHANGE_CLEAN, - &mddev->flags); + set_bit(MD_CHANGE_CLEAN, &mddev->flags); } err = 0; } else @@ -3467,8 +3463,7 @@ array_state_store(mddev_t *mddev, const char *buf, size_t len) case active: if (mddev->pers) { restart_array(mddev); - if (mddev->external) - clear_bit(MD_CHANGE_CLEAN, &mddev->flags); + clear_bit(MD_CHANGE_PENDING, &mddev->flags); wake_up(&mddev->sb_wait); err = 0; } else { @@ -6573,6 +6568,7 @@ void md_write_start(mddev_t *mddev, struct bio *bi) if (mddev->in_sync) { mddev->in_sync = 0; set_bit(MD_CHANGE_CLEAN, &mddev->flags); + set_bit(MD_CHANGE_PENDING, &mddev->flags); md_wakeup_thread(mddev->thread); did_change = 1; } @@ -6581,7 +6577,6 @@ void md_write_start(mddev_t *mddev, struct bio *bi) if (did_change) sysfs_notify_dirent_safe(mddev->sysfs_state); wait_event(mddev->sb_wait, - !test_bit(MD_CHANGE_CLEAN, &mddev->flags) && !test_bit(MD_CHANGE_PENDING, &mddev->flags)); } @@ -6617,6 +6612,7 @@ int md_allow_write(mddev_t *mddev) if (mddev->in_sync) { mddev->in_sync = 0; set_bit(MD_CHANGE_CLEAN, &mddev->flags); + set_bit(MD_CHANGE_PENDING, &mddev->flags); if (mddev->safemode_delay && mddev->safemode == 0) mddev->safemode = 1; @@ -6626,7 +6622,7 @@ int md_allow_write(mddev_t *mddev) } else spin_unlock_irq(&mddev->write_lock); - if (test_bit(MD_CHANGE_CLEAN, &mddev->flags)) + if (test_bit(MD_CHANGE_PENDING, &mddev->flags)) return -EAGAIN; else return 0; @@ -6824,8 +6820,7 @@ void md_do_sync(mddev_t *mddev) atomic_read(&mddev->recovery_active) == 0); mddev->curr_resync_completed = mddev->curr_resync; - if (mddev->persistent) - set_bit(MD_CHANGE_CLEAN, &mddev->flags); + set_bit(MD_CHANGE_CLEAN, &mddev->flags); sysfs_notify(&mddev->kobj, NULL, "sync_completed"); } @@ -7104,8 +7099,7 @@ void md_check_recovery(mddev_t *mddev) mddev->recovery_cp == MaxSector) { mddev->in_sync = 1; did_change = 1; - if (mddev->persistent) - set_bit(MD_CHANGE_CLEAN, &mddev->flags); + set_bit(MD_CHANGE_CLEAN, &mddev->flags); } if (mddev->safemode == 1) mddev->safemode = 0; diff --git a/drivers/md/md.h b/drivers/md/md.h index a953fe2..3931299 100644 --- a/drivers/md/md.h +++ b/drivers/md/md.h @@ -140,7 +140,7 @@ struct mddev_s unsigned long flags; #define MD_CHANGE_DEVS 0 /* Some device status has changed */ #define MD_CHANGE_CLEAN 1 /* transition to or from 'clean' */ -#define MD_CHANGE_PENDING 2 /* superblock update in progress */ +#define MD_CHANGE_PENDING 2 /* switch from 'clean' to 'active' in progress */ int suspended; atomic_t active_io; -- cgit v0.10.2 From fa66f07aa1f0950e1dc78b7ab39728b3f8aa77a1 Mon Sep 17 00:00:00 2001 From: Stephane Eranian Date: Thu, 26 Aug 2010 16:40:01 +0200 Subject: perf_events: Fix time tracking for events with pid != -1 and cpu != -1 Per-thread events with a cpu filter, i.e., cpu != -1, were not reporting correct timings when the thread never ran on the monitored cpu. The time enabled was reported as a negative value. This patch fixes the problem by updating tstamp_stopped, tstamp_running in event_sched_out() for events with filters and which are marked as INACTIVE. The function group_sched_out() is modified to systematically call into event_sched_out() to avoid duplicating the timing adjustment code twice. With the patch, I now get: $ task_cpu -i -e unhalted_core_cycles,unhalted_core_cycles noploop 2 noploop for 2 seconds CPU0 0 unhalted_core_cycles (ena=1,991,136,594, run=0) CPU0 0 unhalted_core_cycles (ena=1,991,136,594, run=0) CPU1 0 unhalted_core_cycles (ena=1,991,136,594, run=0) CPU1 0 unhalted_core_cycles (ena=1,991,136,594, run=0) CPU2 0 unhalted_core_cycles (ena=1,991,136,594, run=0) CPU2 0 unhalted_core_cycles (ena=1,991,136,594, run=0) CPU3 4,747,990,931 unhalted_core_cycles (ena=1,991,136,594, run=1,991,136,594) CPU3 4,747,990,931 unhalted_core_cycles (ena=1,991,136,594, run=1,991,136,594) Signed-off-by: Stephane Eranian Acked-by: Peter Zijlstra Cc: paulus@samba.org Cc: davem@davemloft.net Cc: fweisbec@gmail.com Cc: perfmon2-devel@lists.sf.net Cc: eranian@google.com LKML-Reference: <4c76802d.aae9d80a.115d.70fe@mx.google.com> Signed-off-by: Ingo Molnar diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 403d180..657555a 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -402,11 +402,31 @@ static void perf_group_detach(struct perf_event *event) } } +static inline int +event_filter_match(struct perf_event *event) +{ + return event->cpu == -1 || event->cpu == smp_processor_id(); +} + static void event_sched_out(struct perf_event *event, struct perf_cpu_context *cpuctx, struct perf_event_context *ctx) { + u64 delta; + /* + * An event which could not be activated because of + * filter mismatch still needs to have its timings + * maintained, otherwise bogus information is return + * via read() for time_enabled, time_running: + */ + if (event->state == PERF_EVENT_STATE_INACTIVE + && !event_filter_match(event)) { + delta = ctx->time - event->tstamp_stopped; + event->tstamp_running += delta; + event->tstamp_stopped = ctx->time; + } + if (event->state != PERF_EVENT_STATE_ACTIVE) return; @@ -432,9 +452,7 @@ group_sched_out(struct perf_event *group_event, struct perf_event_context *ctx) { struct perf_event *event; - - if (group_event->state != PERF_EVENT_STATE_ACTIVE) - return; + int state = group_event->state; event_sched_out(group_event, cpuctx, ctx); @@ -444,7 +462,7 @@ group_sched_out(struct perf_event *group_event, list_for_each_entry(event, &group_event->sibling_list, group_entry) event_sched_out(event, cpuctx, ctx); - if (group_event->attr.exclusive) + if (state == PERF_EVENT_STATE_ACTIVE && group_event->attr.exclusive) cpuctx->exclusive = 0; } -- cgit v0.10.2 From 1deacd7a1d4bb22f812908c34e3254de76e4faa8 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 12 Aug 2010 12:10:59 -0700 Subject: UBI: fix kconfig unmet dependency warning: (OPTPROBES && KPROBES && HAVE_OPTPROBES && !PREEMPT && DEBUG_KERNEL || MTD_UBI_DEBUG && MTD && SYSFS && MTD_UBI || UBIFS_FS_DEBUG && MISC_FILESYSTEMS && UBIFS_FS || LOCKDEP && DEBUG_KERNEL && TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT || LATENCYTOP && HAVE_LATENCYTOP_SUPPORT && DEBUG_KERNEL && STACKTRACE_SUPPORT && PROC_FS) selects KALLSYMS_ALL which has unmet direct dependencies (DEBUG_KERNEL && KALLSYMS) Signed-off-by: Randy Dunlap Signed-off-by: Artem Bityutskiy diff --git a/drivers/mtd/ubi/Kconfig.debug b/drivers/mtd/ubi/Kconfig.debug index 2246f15..61f6e5e 100644 --- a/drivers/mtd/ubi/Kconfig.debug +++ b/drivers/mtd/ubi/Kconfig.debug @@ -6,7 +6,7 @@ config MTD_UBI_DEBUG depends on SYSFS depends on MTD_UBI select DEBUG_FS - select KALLSYMS_ALL + select KALLSYMS_ALL if KALLSYMS && DEBUG_KERNEL help This option enables UBI debugging. -- cgit v0.10.2 From d3f6e6c666c0f68991d785177c4c62fcd1d651f2 Mon Sep 17 00:00:00 2001 From: Artem Bityutskiy Date: Sun, 29 Aug 2010 23:34:44 +0300 Subject: UBI: do not oops when erroneous PEB is scheduled for scrubbing When an erroneous PEB is scheduling for scrubbing, we end up with the following oops: [] (prot_queue_del+0x0/0x50) from [] (ubi_wl_scrub_peb+0xec/0x13c) [] (ubi_wl_scrub_peb+0x0/0x13c) from [] (ubi_eba_read_leb+0x200/0x428) [] (ubi_eba_read_leb+0x0/0x428) from [] (ubi_leb_read+0xe8/0x138) [] (ubi_leb_read+0x0/0x138) from [] (ubifs_start_scan+0x7c/0xf4) [] (ubifs_start_scan+0x0/0xf4) from [] (ubifs_recover_leb+0x3c/0x730) [] (ubifs_recover_leb+0x0/0x730) from [] (ubifs_recover_log_leb+0xc8/0x2dc) [] (ubifs_recover_log_leb+0x0/0x2dc) from [] (ubifs_replay_journal+0xb90/0x13a4) [] (ubifs_replay_journal+0x0/0x13a4) from [] (ubifs_fill_super+0xb84/0x1054) [] (ubifs_fill_super+0x0/0x1054) from [] (ubifs_get_sb+0xc4/0x2ac) [] (ubifs_get_sb+0x0/0x2ac) from [] (vfs_kern_mount+0x58/0x94) [] (vfs_kern_mount+0x0/0x94) from [] (do_kern_mount+0x40/0xe8) [] (do_kern_mount+0x0/0xe8) from [] (do_new_mount+0x68/0x8c) [] (do_new_mount+0x0/0x8c) from [] (do_mount+0x15c/0x1b8) [] (do_mount+0x0/0x1b8) from [] (sys_mount+0x8c/0xd4) [] (sys_mount+0x0/0xd4) from [] (ret_fast_syscall+0x0/0x2c) Kernel panic - not syncing: Fatal exception The problem is that 'ubi_wl_scrub_peb()' does not expect that PEBs may be in the erroneous tree, which is a bug. This patch fixes the bug and adds corresponding check to 'ubi_wl_scrub_peb()'. Now it will simply ignore erroneous PEBs, instead of causing an oops. Reported-by: Matthieu CASTET Signed-off-by: Artem Bityutskiy diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c index ee7b1d8..97a4356 100644 --- a/drivers/mtd/ubi/wl.c +++ b/drivers/mtd/ubi/wl.c @@ -1212,7 +1212,8 @@ int ubi_wl_scrub_peb(struct ubi_device *ubi, int pnum) retry: spin_lock(&ubi->wl_lock); e = ubi->lookuptbl[pnum]; - if (e == ubi->move_from || in_wl_tree(e, &ubi->scrub)) { + if (e == ubi->move_from || in_wl_tree(e, &ubi->scrub) || + in_wl_tree(e, &ubi->erroneous)) { spin_unlock(&ubi->wl_lock); return 0; } -- cgit v0.10.2 From 8f587df479c3cea14ba1a9b9d58f34fd2fd6d58b Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 4 Aug 2010 16:27:45 +0000 Subject: 9p: potential ERR_PTR() dereference p9_client_walk() can return error values if we run out of space or there is a problem with the network. Signed-off-by: Dan Carpenter Signed-off-by: Eric Van Hensbergen diff --git a/fs/9p/fid.c b/fs/9p/fid.c index 3585636..6406f89 100644 --- a/fs/9p/fid.c +++ b/fs/9p/fid.c @@ -242,7 +242,8 @@ struct p9_fid *v9fs_fid_lookup(struct dentry *dentry) } kfree(wnames); fid_out: - v9fs_fid_add(dentry, fid); + if (!IS_ERR(fid)) + v9fs_fid_add(dentry, fid); err_out: up_read(&v9ses->rename_sem); return fid; -- cgit v0.10.2 From edce6820a9fdda85521211cb334a183e34cc455e Mon Sep 17 00:00:00 2001 From: Jeffrey Carlyle Date: Mon, 30 Aug 2010 19:55:09 +0200 Subject: scatterlist: prevent invalid free when alloc fails When alloc fails, free_table is being called. Depending on the number of bytes requested, we determine if we are going to call _get_free_page() or kmalloc(). When alloc fails, our math is wrong (due to sg_size - 1), and the last buffer is wrongfully assumed to have been allocated by kmalloc. Hence, kfree gets called and a panic occurs. Signed-off-by: Jeffrey Carlyle Signed-off-by: Olusanya Soyannwo Acked-by: Tejun Heo Signed-off-by: Jens Axboe diff --git a/lib/scatterlist.c b/lib/scatterlist.c index a5ec428..4ceb05d 100644 --- a/lib/scatterlist.c +++ b/lib/scatterlist.c @@ -248,8 +248,18 @@ int __sg_alloc_table(struct sg_table *table, unsigned int nents, left -= sg_size; sg = alloc_fn(alloc_size, gfp_mask); - if (unlikely(!sg)) - return -ENOMEM; + if (unlikely(!sg)) { + /* + * Adjust entry count to reflect that the last + * entry of the previous table won't be used for + * linkage. Without this, sg_kfree() may get + * confused. + */ + if (prv) + table->nents = ++table->orig_nents; + + return -ENOMEM; + } sg_init_table(sg, alloc_size); table->nents = table->orig_nents += sg_size; -- cgit v0.10.2 From d8e1ba76d619dbc0be8fbeee4e6c683b5c812d3a Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Tue, 24 Aug 2010 15:27:34 -0400 Subject: ath5k: check return value of ieee80211_get_tx_rate This avoids a NULL pointer dereference as reported here: https://bugzilla.redhat.com/show_bug.cgi?id=625889 When the WARN condition is hit in ieee80211_get_tx_rate, it will return NULL. So, we need to check the return value and avoid dereferencing it in that case. Signed-off-by: John W. Linville Cc: stable@kernel.org Acked-by: Bob Copeland diff --git a/drivers/net/wireless/ath/ath5k/base.c b/drivers/net/wireless/ath/ath5k/base.c index 373dcfe..d77ce99 100644 --- a/drivers/net/wireless/ath/ath5k/base.c +++ b/drivers/net/wireless/ath/ath5k/base.c @@ -1327,6 +1327,10 @@ ath5k_txbuf_setup(struct ath5k_softc *sc, struct ath5k_buf *bf, PCI_DMA_TODEVICE); rate = ieee80211_get_tx_rate(sc->hw, info); + if (!rate) { + ret = -EINVAL; + goto err_unmap; + } if (info->flags & IEEE80211_TX_CTL_NO_ACK) flags |= AR5K_TXDESC_NOACK; -- cgit v0.10.2 From f880c2050f30b23c9b6f80028c09f76e693bf309 Mon Sep 17 00:00:00 2001 From: Christian Lamparter Date: Tue, 24 Aug 2010 22:54:05 +0200 Subject: p54: fix tx feedback status flag check Michael reported that p54* never really entered power save mode, even tough it was enabled. It turned out that upon a power save mode change the firmware will set a special flag onto the last outgoing frame tx status (which in this case is almost always the designated PSM nullfunc frame). This flag confused the driver; It erroneously reported transmission failures to the stack, which then generated the next nullfunc. and so on... Cc: Reported-by: Michael Buesch Tested-by: Michael Buesch Signed-off-by: Christian Lamparter Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/p54/txrx.c b/drivers/net/wireless/p54/txrx.c index 173aec3..0e937dc 100644 --- a/drivers/net/wireless/p54/txrx.c +++ b/drivers/net/wireless/p54/txrx.c @@ -446,7 +446,7 @@ static void p54_rx_frame_sent(struct p54_common *priv, struct sk_buff *skb) } if (!(info->flags & IEEE80211_TX_CTL_NO_ACK) && - (!payload->status)) + !(payload->status & P54_TX_FAILED)) info->flags |= IEEE80211_TX_STAT_ACK; if (payload->status & P54_TX_PSM_CANCELLED) info->flags |= IEEE80211_TX_STAT_TX_FILTERED; -- cgit v0.10.2 From 071249b1d501b1f31a6b1af3fbcbe03158a84e5c Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 25 Aug 2010 14:47:38 +0200 Subject: mac80211: delete work timer The new workqueue changes helped me find this bug that's been lingering since the changes to the work processing in mac80211 -- the work timer is never deleted properly. Do that to avoid having it fire after all data structures have been freed. It can't be re-armed because all it will do, if running, is schedule the work, but that gets flushed later and won't have anything to do since all work items are gone by now (by way of interface removal). Cc: stable@kernel.org [2.6.34+] Signed-off-by: Johannes Berg Signed-off-by: John W. Linville diff --git a/net/mac80211/main.c b/net/mac80211/main.c index 798a91b..ded5c38 100644 --- a/net/mac80211/main.c +++ b/net/mac80211/main.c @@ -732,6 +732,12 @@ void ieee80211_unregister_hw(struct ieee80211_hw *hw) rtnl_unlock(); + /* + * Now all work items will be gone, but the + * timer might still be armed, so delete it + */ + del_timer_sync(&local->work_timer); + cancel_work_sync(&local->reconfig_filter); ieee80211_clear_tx_pending(local); -- cgit v0.10.2 From 9ef808048564928a83f3a52c65c5725688cf5cbe Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Fri, 27 Aug 2010 09:44:12 -0400 Subject: MAINTAINERS: change broken url for prism54 Reported-by: Joe Perches Signed-off-by: John W. Linville diff --git a/MAINTAINERS b/MAINTAINERS index 44620ae..8e85429 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4549,7 +4549,7 @@ F: include/linux/preempt.h PRISM54 WIRELESS DRIVER M: "Luis R. Rodriguez" L: linux-wireless@vger.kernel.org -W: http://prism54.org +W: http://wireless.kernel.org/en/users/Drivers/p54 S: Obsolete F: drivers/net/wireless/prism54/ -- cgit v0.10.2 From 42da2f948d949efd0111309f5827bf0298bcc9a4 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 30 Aug 2010 12:24:54 +0200 Subject: wireless extensions: fix kernel heap content leak Wireless extensions have an unfortunate, undocumented requirement which requires drivers to always fill iwp->length when returning a successful status. When a driver doesn't do this, it leads to a kernel heap content leak when userspace offers a larger buffer than would have been necessary. Arguably, this is a driver bug, as it should, if it returns 0, fill iwp->length, even if it separately indicated that the buffer contents was not valid. However, we can also at least avoid the memory content leak if the driver doesn't do this by setting the iwp length to max_tokens, which then reflects how big the buffer is that the driver may fill, regardless of how big the userspace buffer is. To illustrate the point, this patch also fixes a corresponding cfg80211 bug (since this requirement isn't documented nor was ever pointed out by anyone during code review, I don't trust all drivers nor all cfg80211 handlers to implement it correctly). Cc: stable@kernel.org [all the way back] Signed-off-by: Johannes Berg Signed-off-by: John W. Linville diff --git a/net/wireless/wext-compat.c b/net/wireless/wext-compat.c index bb5e0a5..7e5c3a4 100644 --- a/net/wireless/wext-compat.c +++ b/net/wireless/wext-compat.c @@ -1420,6 +1420,9 @@ int cfg80211_wext_giwessid(struct net_device *dev, { struct wireless_dev *wdev = dev->ieee80211_ptr; + data->flags = 0; + data->length = 0; + switch (wdev->iftype) { case NL80211_IFTYPE_ADHOC: return cfg80211_ibss_wext_giwessid(dev, info, data, ssid); diff --git a/net/wireless/wext-core.c b/net/wireless/wext-core.c index 0ef17bc..8f5116f 100644 --- a/net/wireless/wext-core.c +++ b/net/wireless/wext-core.c @@ -782,6 +782,22 @@ static int ioctl_standard_iw_point(struct iw_point *iwp, unsigned int cmd, } } + if (IW_IS_GET(cmd) && !(descr->flags & IW_DESCR_FLAG_NOMAX)) { + /* + * If this is a GET, but not NOMAX, it means that the extra + * data is not bounded by userspace, but by max_tokens. Thus + * set the length to max_tokens. This matches the extra data + * allocation. + * The driver should fill it with the number of tokens it + * provided, and it may check iwp->length rather than having + * knowledge of max_tokens. If the driver doesn't change the + * iwp->length, this ioctl just copies back max_token tokens + * filled with zeroes. Hopefully the driver isn't claiming + * them to be valid data. + */ + iwp->length = descr->max_tokens; + } + err = handler(dev, info, (union iwreq_data *) iwp, extra); iwp->length += essid_compat; -- cgit v0.10.2 From 4562c986f0d694124de20815adf7e1aad8a94668 Mon Sep 17 00:00:00 2001 From: Matthew McClintock Date: Fri, 27 Aug 2010 11:58:21 +0000 Subject: powerpc/kexec: Adds correct calling convention for kexec purgatory Call kexec purgatory code correctly. We were getting lucky before. If you examine the powerpc 32bit kexec "purgatory" code you will see it expects the following: >From kexec-tools: purgatory/arch/ppc/v2wrap_32.S -> calling convention: -> r3 = physical number of this cpu (all cpus) -> r4 = address of this chunk (master only) As such, we need to set r3 to the current core, r4 happens to be unused by purgatory at the moment but we go ahead and set it here as well Signed-off-by: Matthew McClintock Signed-off-by: Benjamin Herrenschmidt diff --git a/arch/powerpc/kernel/misc_32.S b/arch/powerpc/kernel/misc_32.S index 6bbd7a6..a7a570d 100644 --- a/arch/powerpc/kernel/misc_32.S +++ b/arch/powerpc/kernel/misc_32.S @@ -810,6 +810,9 @@ relocate_new_kernel: isync sync + mfspr r3, SPRN_PIR /* current core we are running on */ + mr r4, r5 /* load physical address of chunk called */ + /* jump to the entry point, usually the setup routine */ mtlr r5 blrl -- cgit v0.10.2 From b0d278b7d3ae9115939ddcea93f516308cc367e2 Mon Sep 17 00:00:00 2001 From: Paul Mackerras Date: Tue, 10 Aug 2010 20:38:23 +0000 Subject: powerpc/perf_event: Reduce latency of calling perf_event_do_pending Commit 0fe1ac48 ("powerpc/perf_event: Fix oops due to perf_event_do_pending call") moved the call to perf_event_do_pending in timer_interrupt() down so that it was after the irq_enter() call. Unfortunately this moved it after the code that checks whether it is time for the next decrementer clock event. The result is that the call to perf_event_do_pending() won't happen until the next decrementer clock event is due. This was pointed out by Milton Miller. This fixes it by moving the check for whether it's time for the next decrementer clock event down to the point where we're about to call the event handler, after we've called perf_event_do_pending. This has the side effect that on old pre-Core99 Powermacs where we use the ppc_n_lost_interrupts mechanism to replay interrupts, a replayed interrupt will incur a little more latency since it will now do the code from the irq_enter down to the irq_exit, that it used to skip. However, these machines are now old and rare enough that this doesn't matter. To make it clear that ppc_n_lost_interrupts is only used on Powermacs, and to speed up the code slightly on non-Powermac ppc32 machines, the code that tests ppc_n_lost_interrupts is now conditional on CONFIG_PMAC as well as CONFIG_PPC32. Signed-off-by: Paul Mackerras Cc: stable@kernel.org Signed-off-by: Benjamin Herrenschmidt diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c index ce53dfa..8533b3b 100644 --- a/arch/powerpc/kernel/time.c +++ b/arch/powerpc/kernel/time.c @@ -577,20 +577,11 @@ void timer_interrupt(struct pt_regs * regs) * some CPUs will continuue to take decrementer exceptions */ set_dec(DECREMENTER_MAX); -#ifdef CONFIG_PPC32 +#if defined(CONFIG_PPC32) && defined(CONFIG_PMAC) if (atomic_read(&ppc_n_lost_interrupts) != 0) do_IRQ(regs); #endif - now = get_tb_or_rtc(); - if (now < decrementer->next_tb) { - /* not time for this event yet */ - now = decrementer->next_tb - now; - if (now <= DECREMENTER_MAX) - set_dec((int)now); - trace_timer_interrupt_exit(regs); - return; - } old_regs = set_irq_regs(regs); irq_enter(); @@ -606,8 +597,16 @@ void timer_interrupt(struct pt_regs * regs) get_lppaca()->int_dword.fields.decr_int = 0; #endif - if (evt->event_handler) - evt->event_handler(evt); + now = get_tb_or_rtc(); + if (now >= decrementer->next_tb) { + decrementer->next_tb = ~(u64)0; + if (evt->event_handler) + evt->event_handler(evt); + } else { + now = decrementer->next_tb - now; + if (now <= DECREMENTER_MAX) + set_dec((int)now); + } #ifdef CONFIG_PPC_ISERIES if (firmware_has_feature(FW_FEATURE_ISERIES) && hvlpevent_is_pending()) -- cgit v0.10.2 From 54a834043314c257210db2a9d59f8cc605571639 Mon Sep 17 00:00:00 2001 From: Michael Neuling Date: Wed, 25 Aug 2010 21:04:25 +0000 Subject: powerpc: Don't use kernel stack with translation off In f761622e59433130bc33ad086ce219feee9eb961 we changed early_setup_secondary so it's called using the proper kernel stack rather than the emergency one. Unfortunately, this stack pointer can't be used when translation is off on PHYP as this stack pointer might be outside the RMO. This results in the following on all non zero cpus: cpu 0x1: Vector: 300 (Data Access) at [c00000001639fd10] pc: 000000000001c50c lr: 000000000000821c sp: c00000001639ff90 msr: 8000000000001000 dar: c00000001639ffa0 dsisr: 42000000 current = 0xc000000016393540 paca = 0xc000000006e00200 pid = 0, comm = swapper The original patch was only tested on bare metal system, so it never caught this problem. This changes __secondary_start so that we calculate the new stack pointer but only start using it after we've called early_setup_secondary. With this patch, the above problem goes away. Signed-off-by: Michael Neuling Signed-off-by: Benjamin Herrenschmidt diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S index 4d6681d..c571cd3 100644 --- a/arch/powerpc/kernel/head_64.S +++ b/arch/powerpc/kernel/head_64.S @@ -575,13 +575,19 @@ __secondary_start: /* Initialize the kernel stack. Just a repeat for iSeries. */ LOAD_REG_ADDR(r3, current_set) sldi r28,r24,3 /* get current_set[cpu#] */ - ldx r1,r3,r28 - addi r1,r1,THREAD_SIZE-STACK_FRAME_OVERHEAD - std r1,PACAKSAVE(r13) + ldx r14,r3,r28 + addi r14,r14,THREAD_SIZE-STACK_FRAME_OVERHEAD + std r14,PACAKSAVE(r13) /* Do early setup for that CPU (stab, slb, hash table pointer) */ bl .early_setup_secondary + /* + * setup the new stack pointer, but *don't* use this until + * translation is on. + */ + mr r1, r14 + /* Clear backchain so we get nice backtraces */ li r7,0 mtlr r7 -- cgit v0.10.2 From 628e300cccaa628d8fb92aa28cb7530a3d5f2257 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 30 Aug 2010 18:35:24 -0700 Subject: irda: Correctly clean up self->ias_obj on irda_bind() failure. If irda_open_tsap() fails, the irda_bind() code tries to destroy the ->ias_obj object by hand, but does so wrongly. In particular, it fails to a) release the hashbin attached to the object and b) reset the self->ias_obj pointer to NULL. Fix both problems by using irias_delete_object() and explicitly setting self->ias_obj to NULL, just as irda_release() does. Reported-by: Tavis Ormandy Signed-off-by: David S. Miller diff --git a/net/irda/af_irda.c b/net/irda/af_irda.c index 79986a6..fd55b51 100644 --- a/net/irda/af_irda.c +++ b/net/irda/af_irda.c @@ -824,8 +824,8 @@ static int irda_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len) err = irda_open_tsap(self, addr->sir_lsap_sel, addr->sir_name); if (err < 0) { - kfree(self->ias_obj->name); - kfree(self->ias_obj); + irias_delete_object(self->ias_obj); + self->ias_obj = NULL; goto out; } -- cgit v0.10.2 From 10f0412f57f2a76a90eff4376f59cbb0a39e4e18 Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Mon, 30 Aug 2010 10:56:18 +0200 Subject: oprofile, x86: fix init_sysfs error handling On failure init_sysfs() might not properly free resources. The error code of the function is not checked. And, when reinitializing the exit function might be called twice. This patch fixes all this. Cc: stable@kernel.org Signed-off-by: Robert Richter diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c index f6b48f6..73a41d3 100644 --- a/arch/x86/oprofile/nmi_int.c +++ b/arch/x86/oprofile/nmi_int.c @@ -568,8 +568,13 @@ static int __init init_sysfs(void) int error; error = sysdev_class_register(&oprofile_sysclass); - if (!error) - error = sysdev_register(&device_oprofile); + if (error) + return error; + + error = sysdev_register(&device_oprofile); + if (error) + sysdev_class_unregister(&oprofile_sysclass); + return error; } @@ -695,6 +700,8 @@ int __init op_nmi_init(struct oprofile_operations *ops) char *cpu_type = NULL; int ret = 0; + using_nmi = 0; + if (!cpu_has_apic) return -ENODEV; @@ -774,7 +781,10 @@ int __init op_nmi_init(struct oprofile_operations *ops) mux_init(ops); - init_sysfs(); + ret = init_sysfs(); + if (ret) + return ret; + using_nmi = 1; printk(KERN_INFO "oprofile: using NMI interrupt.\n"); return 0; -- cgit v0.10.2 From 477a3c33d1efa0342a74bd02da2e049191993e2c Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 31 Aug 2010 10:54:35 +0200 Subject: workqueue: fix GCWQ_DISASSOCIATED initialization init_workqueues() incorrectly marks workqueues for all possible CPUs associated. Combined with mayday_mask initialization bug, this can make rescuers keep trying to bind to an offline gcwq indefinitely. Fix init_workqueues() such that only online CPUs have their gcwqs have GCWQ_DISASSOCIATED cleared. Signed-off-by: Tejun Heo Reported-by: CAI Qian diff --git a/kernel/workqueue.c b/kernel/workqueue.c index a2dccfc..c8183b2 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -3558,8 +3558,7 @@ static int __init init_workqueues(void) spin_lock_init(&gcwq->lock); INIT_LIST_HEAD(&gcwq->worklist); gcwq->cpu = cpu; - if (cpu == WORK_CPU_UNBOUND) - gcwq->flags |= GCWQ_DISASSOCIATED; + gcwq->flags |= GCWQ_DISASSOCIATED; INIT_LIST_HEAD(&gcwq->idle_list); for (i = 0; i < BUSY_WORKER_HASH_SIZE; i++) @@ -3583,6 +3582,8 @@ static int __init init_workqueues(void) struct global_cwq *gcwq = get_gcwq(cpu); struct worker *worker; + if (cpu != WORK_CPU_UNBOUND) + gcwq->flags &= ~GCWQ_DISASSOCIATED; worker = create_worker(gcwq, true); BUG_ON(!worker); spin_lock_irq(&gcwq->lock); -- cgit v0.10.2 From 9c37547ab62f88aac3e1e3c2065b611f811de9b5 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 31 Aug 2010 11:18:34 +0200 Subject: workqueue: use zalloc_cpumask_var() for gcwq->mayday_mask alloc_mayday_mask() was using alloc_cpumask_var() making gcwq->mayday_mask contain garbage after initialization on CONFIG_CPUMASK_OFFSTACK=y configurations. This combined with the previously fixed GCWQ_DISASSOCIATED initialization bug could make rescuers fall into infinite loop trying to bind to an offline cpu. Signed-off-by: Tejun Heo Reported-by: CAI Qian diff --git a/kernel/workqueue.c b/kernel/workqueue.c index c8183b2..7855429 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -196,7 +196,7 @@ typedef cpumask_var_t mayday_mask_t; cpumask_test_and_set_cpu((cpu), (mask)) #define mayday_clear_cpu(cpu, mask) cpumask_clear_cpu((cpu), (mask)) #define for_each_mayday_cpu(cpu, mask) for_each_cpu((cpu), (mask)) -#define alloc_mayday_mask(maskp, gfp) alloc_cpumask_var((maskp), (gfp)) +#define alloc_mayday_mask(maskp, gfp) zalloc_cpumask_var((maskp), (gfp)) #define free_mayday_mask(mask) free_cpumask_var((mask)) #else typedef unsigned long mayday_mask_t; -- cgit v0.10.2 From 6d4f2fb0860c36e7ef16cc41126242a8b6542771 Mon Sep 17 00:00:00 2001 From: Alexander Graf Date: Tue, 31 Aug 2010 04:15:22 +0200 Subject: powerpc/85xx: Fix compilation of mpc85xx_mds.c Commit 99d8238f berobbed the for_each loop of its iterator! Let's be nice and give it back, so it compiles for us. CC: Anton Vorontsov Signed-off-by: Alexander Graf Signed-off-by: Kumar Gala diff --git a/arch/powerpc/platforms/85xx/mpc85xx_mds.c b/arch/powerpc/platforms/85xx/mpc85xx_mds.c index da64be1..aa34cac 100644 --- a/arch/powerpc/platforms/85xx/mpc85xx_mds.c +++ b/arch/powerpc/platforms/85xx/mpc85xx_mds.c @@ -357,6 +357,7 @@ static void __init mpc85xx_mds_setup_arch(void) { #ifdef CONFIG_PCI struct pci_controller *hose; + struct device_node *np; #endif dma_addr_t max = 0xffffffff; -- cgit v0.10.2 From dc1c41f450a5f201f1d8c19aef32319f3b84c273 Mon Sep 17 00:00:00 2001 From: Kumar Gala Date: Tue, 31 Aug 2010 11:39:25 -0500 Subject: powerpc/85xx: Fix compile issue with p1022_ds due to lmb rename to memblock arch/powerpc/platforms/85xx/p1022_ds.c:22:23: error: linux/lmb.h: No such file or directory arch/powerpc/platforms/85xx/p1022_ds.c: In function 'p1022_ds_setup_arch': arch/powerpc/platforms/85xx/p1022_ds.c:100: error: implicit declaration of function 'memblock_end_of_DRAM' arch/powerpc/platforms/85xx/p1022_ds.c: At top level: arch/powerpc/platforms/85xx/p1022_ds.c:147: error: 'udbg_progress' undeclared here (not in a function) make[2]: *** [arch/powerpc/platforms/85xx/p1022_ds.o] Error 1 Signed-off-by: Kumar Gala diff --git a/arch/powerpc/platforms/85xx/p1022_ds.c b/arch/powerpc/platforms/85xx/p1022_ds.c index e1467c9..34e0090 100644 --- a/arch/powerpc/platforms/85xx/p1022_ds.c +++ b/arch/powerpc/platforms/85xx/p1022_ds.c @@ -19,7 +19,7 @@ #include #include -#include +#include #include #include @@ -97,7 +97,7 @@ static void __init p1022_ds_setup_arch(void) #endif #ifdef CONFIG_SWIOTLB - if (lmb_end_of_DRAM() > max) { + if (memblock_end_of_DRAM() > max) { ppc_swiotlb_enable = 1; set_pci_dma_ops(&swiotlb_dma_ops); ppc_md.pci_dma_dev_setup = pci_dma_dev_setup_swiotlb; -- cgit v0.10.2 From b963ea89f00f13c648af4082e5efb2172d369727 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 30 Aug 2010 19:08:01 -0700 Subject: netlink: Make NETLINK_USERSOCK work again. Once we started enforcing the a nl_table[] entry exist for a protocol, NETLINK_USERSOCK stopped working. Add a dummy table entry so that it works again. Reported-by: Thomas Voegtle Tested-by: Thomas Voegtle Signed-off-by: David S. Miller diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 980fe4a..cd96ed3 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -2102,6 +2102,26 @@ static void __net_exit netlink_net_exit(struct net *net) #endif } +static void __init netlink_add_usersock_entry(void) +{ + unsigned long *listeners; + int groups = 32; + + listeners = kzalloc(NLGRPSZ(groups) + sizeof(struct listeners_rcu_head), + GFP_KERNEL); + if (!listeners) + panic("netlink_add_usersock_entry: Cannot allocate listneres\n"); + + netlink_table_grab(); + + nl_table[NETLINK_USERSOCK].groups = groups; + nl_table[NETLINK_USERSOCK].listeners = listeners; + nl_table[NETLINK_USERSOCK].module = THIS_MODULE; + nl_table[NETLINK_USERSOCK].registered = 1; + + netlink_table_ungrab(); +} + static struct pernet_operations __net_initdata netlink_net_ops = { .init = netlink_net_init, .exit = netlink_net_exit, @@ -2150,6 +2170,8 @@ static int __init netlink_proto_init(void) hash->rehash_time = jiffies; } + netlink_add_usersock_entry(); + sock_register(&netlink_family_ops); register_pernet_subsys(&netlink_net_ops); /* The netlink device handler may be needed early. */ -- cgit v0.10.2 From c3d34d5d9654ec9c2510f9341bfb1030b8f029d1 Mon Sep 17 00:00:00 2001 From: "John W. Linville" Date: Mon, 30 Aug 2010 17:36:40 -0400 Subject: wireless: register wiphy rfkill w/o holding cfg80211_mutex Otherwise lockdep complains... https://bugzilla.kernel.org/show_bug.cgi?id=17311 [ INFO: possible circular locking dependency detected ] 2.6.36-rc2-git4 #12 ------------------------------------------------------- kworker/0:3/3630 is trying to acquire lock: (rtnl_mutex){+.+.+.}, at: [] rtnl_lock+0x12/0x14 but task is already holding lock: (rfkill_global_mutex){+.+.+.}, at: [] rfkill_switch_all+0x24/0x49 [rfkill] which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #2 (rfkill_global_mutex){+.+.+.}: [] lock_acquire+0x120/0x15b [] __mutex_lock_common+0x54/0x52e [] mutex_lock_nested+0x34/0x39 [] rfkill_register+0x2b/0x29c [rfkill] [] wiphy_register+0x1ae/0x270 [cfg80211] [] ieee80211_register_hw+0x1b4/0x3cf [mac80211] [] iwl_ucode_callback+0x9e9/0xae3 [iwlagn] [] request_firmware_work_func+0x54/0x6f [] kthread+0x8c/0x94 [] kernel_thread_helper+0x4/0x10 -> #1 (cfg80211_mutex){+.+.+.}: [] lock_acquire+0x120/0x15b [] __mutex_lock_common+0x54/0x52e [] mutex_lock_nested+0x34/0x39 [] cfg80211_get_dev_from_ifindex+0x1b/0x7c [cfg80211] [] cfg80211_wext_giwscan+0x58/0x990 [cfg80211] [] ioctl_standard_iw_point+0x1a8/0x272 [] ioctl_standard_call+0x91/0xa7 [] T.723+0xbd/0x12c [] wext_handle_ioctl+0x31/0x6d [] dev_ioctl+0x63d/0x67a [] sock_ioctl+0x48/0x21d [] do_vfs_ioctl+0x4ba/0x509 [] sys_ioctl+0x51/0x74 [] system_call_fastpath+0x16/0x1b -> #0 (rtnl_mutex){+.+.+.}: [] __lock_acquire+0xa93/0xd9a [] lock_acquire+0x120/0x15b [] __mutex_lock_common+0x54/0x52e [] mutex_lock_nested+0x34/0x39 [] rtnl_lock+0x12/0x14 [] cfg80211_rfkill_set_block+0x1a/0x7b [cfg80211] [] rfkill_set_block+0x80/0xd5 [rfkill] [] __rfkill_switch_all+0x3f/0x6f [rfkill] [] rfkill_switch_all+0x38/0x49 [rfkill] [] rfkill_op_handler+0x105/0x136 [rfkill] [] process_one_work+0x248/0x403 [] worker_thread+0x139/0x214 [] kthread+0x8c/0x94 [] kernel_thread_helper+0x4/0x10 Signed-off-by: John W. Linville Acked-by: Johannes Berg diff --git a/net/wireless/core.c b/net/wireless/core.c index 541e2ff..d6d046b 100644 --- a/net/wireless/core.c +++ b/net/wireless/core.c @@ -475,12 +475,10 @@ int wiphy_register(struct wiphy *wiphy) mutex_lock(&cfg80211_mutex); res = device_add(&rdev->wiphy.dev); - if (res) - goto out_unlock; - - res = rfkill_register(rdev->rfkill); - if (res) - goto out_rm_dev; + if (res) { + mutex_unlock(&cfg80211_mutex); + return res; + } /* set up regulatory info */ wiphy_update_regulatory(wiphy, NL80211_REGDOM_SET_BY_CORE); @@ -509,13 +507,18 @@ int wiphy_register(struct wiphy *wiphy) cfg80211_debugfs_rdev_add(rdev); mutex_unlock(&cfg80211_mutex); + /* + * due to a locking dependency this has to be outside of the + * cfg80211_mutex lock + */ + res = rfkill_register(rdev->rfkill); + if (res) + goto out_rm_dev; + return 0; out_rm_dev: device_del(&rdev->wiphy.dev); - -out_unlock: - mutex_unlock(&cfg80211_mutex); return res; } EXPORT_SYMBOL(wiphy_register); -- cgit v0.10.2 From 803288e61e346ba367373bc7d5eeb6e11c81a33c Mon Sep 17 00:00:00 2001 From: "Luis R. Rodriguez" Date: Mon, 30 Aug 2010 19:26:32 -0400 Subject: ath9k_hw: Fix EEPROM uncompress block reading on AR9003 The EEPROM is compressed on AR9003, upon decompression the wrong upper limit was being used for the block which prevented the 5 GHz CTL indexes from being used, which are stored towards the end of the EEPROM block. This fix allows the actual intended regulatory limits to be used on AR9003 hardware. Cc: stable@kernel.org [2.6.36+] Signed-off-by: Luis R. Rodriguez Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c b/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c index b883b17..057fb69 100644 --- a/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c +++ b/drivers/net/wireless/ath/ath9k/ar9003_eeprom.c @@ -797,7 +797,7 @@ static bool ar9300_uncompress_block(struct ath_hw *ah, length = block[it+1]; length &= 0xff; - if (length > 0 && spot >= 0 && spot+length < mdataSize) { + if (length > 0 && spot >= 0 && spot+length <= mdataSize) { ath_print(common, ATH_DBG_EEPROM, "Restore at %d: spot=%d " "offset=%d length=%d\n", -- cgit v0.10.2 From 904879748d7439a6dabdc6be9aad983e216b027d Mon Sep 17 00:00:00 2001 From: "Luis R. Rodriguez" Date: Mon, 30 Aug 2010 19:26:33 -0400 Subject: ath9k_hw: fix parsing of HT40 5 GHz CTLs The 5 GHz CTL indexes were not being read for all hardware devices due to the masking out through the CTL_MODE_M mask being one bit too short. Without this the calibrated regulatory maximum values were not being picked up when devices operate on 5 GHz in HT40 mode. The final output power used for Atheros devices is the minimum between the calibrated CTL values and what CRDA provides. Cc: stable@kernel.org [2.6.27+] Signed-off-by: Luis R. Rodriguez Signed-off-by: John W. Linville diff --git a/drivers/net/wireless/ath/ath9k/eeprom.h b/drivers/net/wireless/ath/ath9k/eeprom.h index 7f48df1..0b09db0 100644 --- a/drivers/net/wireless/ath/ath9k/eeprom.h +++ b/drivers/net/wireless/ath/ath9k/eeprom.h @@ -62,7 +62,7 @@ #define SD_NO_CTL 0xE0 #define NO_CTL 0xff -#define CTL_MODE_M 7 +#define CTL_MODE_M 0xf #define CTL_11A 0 #define CTL_11B 1 #define CTL_11G 2 diff --git a/drivers/net/wireless/ath/regd.h b/drivers/net/wireless/ath/regd.h index a1c3952..345dd97 100644 --- a/drivers/net/wireless/ath/regd.h +++ b/drivers/net/wireless/ath/regd.h @@ -31,7 +31,6 @@ enum ctl_group { #define NO_CTL 0xff #define SD_NO_CTL 0xE0 #define NO_CTL 0xff -#define CTL_MODE_M 7 #define CTL_11A 0 #define CTL_11B 1 #define CTL_11G 2 -- cgit v0.10.2 From 3aaba20f26f58843e8f20611e5c0b1c06954310f Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Mon, 23 Aug 2010 16:50:12 +0800 Subject: tracing: Fix a race in function profile While we are reading trace_stat/functionX and someone just disabled function_profile at that time, we can trigger this: divide error: 0000 [#1] PREEMPT SMP ... EIP is at function_stat_show+0x90/0x230 ... This fix just takes the ftrace_profile_lock and checks if rec->counter is 0. If it's 0, we know the profile buffer has been reset. Signed-off-by: Li Zefan Cc: stable@kernel.org LKML-Reference: <4C723644.4040708@cn.fujitsu.com> Signed-off-by: Steven Rostedt diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 0d88ce9..7cb1f45 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -381,12 +381,19 @@ static int function_stat_show(struct seq_file *m, void *v) { struct ftrace_profile *rec = v; char str[KSYM_SYMBOL_LEN]; + int ret = 0; #ifdef CONFIG_FUNCTION_GRAPH_TRACER - static DEFINE_MUTEX(mutex); static struct trace_seq s; unsigned long long avg; unsigned long long stddev; #endif + mutex_lock(&ftrace_profile_lock); + + /* we raced with function_profile_reset() */ + if (unlikely(rec->counter == 0)) { + ret = -EBUSY; + goto out; + } kallsyms_lookup(rec->ip, NULL, NULL, NULL, str); seq_printf(m, " %-30.30s %10lu", str, rec->counter); @@ -408,7 +415,6 @@ static int function_stat_show(struct seq_file *m, void *v) do_div(stddev, (rec->counter - 1) * 1000); } - mutex_lock(&mutex); trace_seq_init(&s); trace_print_graph_duration(rec->time, &s); trace_seq_puts(&s, " "); @@ -416,11 +422,12 @@ static int function_stat_show(struct seq_file *m, void *v) trace_seq_puts(&s, " "); trace_print_graph_duration(stddev, &s); trace_print_seq(m, &s); - mutex_unlock(&mutex); #endif seq_putc(m, '\n'); +out: + mutex_unlock(&ftrace_profile_lock); - return 0; + return ret; } static void ftrace_profile_reset(struct ftrace_profile_stat *stat) -- cgit v0.10.2 From ff33f1821259d00d3fd85d86f59783e2ca3c9ee1 Mon Sep 17 00:00:00 2001 From: Li Yang Date: Fri, 18 Jun 2010 14:24:20 +0800 Subject: fsl_rio: fix compile errors Fixes the following compile problem on E500 platforms: arch/powerpc/sysdev/fsl_rio.c: In function 'fsl_rio_mcheck_exception': arch/powerpc/sysdev/fsl_rio.c:248: error: 'MCSR_MASK' undeclared (first use in this function) Also fixes the compile problem on non-E500 platforms. Signed-off-by: Li Yang Signed-off-by: Kumar Gala diff --git a/arch/powerpc/sysdev/fsl_rio.c b/arch/powerpc/sysdev/fsl_rio.c index 6425abe..3017532 100644 --- a/arch/powerpc/sysdev/fsl_rio.c +++ b/arch/powerpc/sysdev/fsl_rio.c @@ -240,12 +240,13 @@ struct rio_priv { static void __iomem *rio_regs_win; +#ifdef CONFIG_E500 static int (*saved_mcheck_exception)(struct pt_regs *regs); static int fsl_rio_mcheck_exception(struct pt_regs *regs) { const struct exception_table_entry *entry = NULL; - unsigned long reason = (mfspr(SPRN_MCSR) & MCSR_MASK); + unsigned long reason = mfspr(SPRN_MCSR); if (reason & MCSR_BUS_RBERR) { reason = in_be32((u32 *)(rio_regs_win + RIO_LTLEDCSR)); @@ -269,6 +270,7 @@ static int fsl_rio_mcheck_exception(struct pt_regs *regs) else return cur_cpu_spec->machine_check(regs); } +#endif /** * fsl_rio_doorbell_send - Send a MPC85xx doorbell message @@ -1517,8 +1519,10 @@ int fsl_rio_setup(struct platform_device *dev) fsl_rio_doorbell_init(port); fsl_rio_port_write_init(port); +#ifdef CONFIG_E500 saved_mcheck_exception = ppc_md.machine_check_exception; ppc_md.machine_check_exception = fsl_rio_mcheck_exception; +#endif /* Ensure that RFXE is set */ mtspr(SPRN_HID1, (mfspr(SPRN_HID1) | 0x20000)); -- cgit v0.10.2 From fa9fc821f8892590211a9aa9e855eb83152b9870 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Sun, 29 Aug 2010 21:47:18 +0200 Subject: arch/powerpc/platforms/83xx/mpc837x_mds.c: Add missing iounmap The function of_iomap returns the result of calling ioremap, so iounmap should be called on the result in the error handling code, as done in the normal exit of the function. The sematic match that finds this problem is as follows: (http://coccinelle.lip6.fr/) // @r exists@ local idexpression x; expression E,E1; identifier l; statement S; @@ *x = of_iomap(...); ... when != iounmap(x) when != if (...) { ... iounmap(x); ... } when != E = x when any ( if (x == NULL) S | if (...) { ... when != iounmap(x) when != if (...) { ... iounmap(x); ... } ( return <+...x...+>; | * return ...; ) } ) ... when != x = E1 when any iounmap(x); // Signed-off-by: Julia Lawall Signed-off-by: Kumar Gala diff --git a/arch/powerpc/platforms/83xx/mpc837x_mds.c b/arch/powerpc/platforms/83xx/mpc837x_mds.c index f9751c8..8306832 100644 --- a/arch/powerpc/platforms/83xx/mpc837x_mds.c +++ b/arch/powerpc/platforms/83xx/mpc837x_mds.c @@ -48,8 +48,10 @@ static int mpc837xmds_usb_cfg(void) return -1; np = of_find_node_by_name(NULL, "usb"); - if (!np) - return -ENODEV; + if (!np) { + ret = -ENODEV; + goto out; + } phy_type = of_get_property(np, "phy_type", NULL); if (phy_type && !strcmp(phy_type, "ulpi")) { clrbits8(bcsr_regs + 12, BCSR12_USB_SER_PIN); @@ -65,8 +67,9 @@ static int mpc837xmds_usb_cfg(void) } of_node_put(np); +out: iounmap(bcsr_regs); - return 0; + return ret; } /* ************************************************************************ -- cgit v0.10.2 From 5aac4d73dc234fe7be91679b547af2436d7f8d1e Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Sun, 29 Aug 2010 11:52:44 +0200 Subject: arch/powerpc/sysdev/qe_lib/qe.c: Add of_node_put to avoid memory leak Add a call to of_node_put in the error handling code following a call to of_find_compatible_node. The semantic match that finds this problem is as follows: (http://coccinelle.lip6.fr/) // @r exists@ local idexpression x; expression E,E1; statement S; @@ *x = (of_find_node_by_path |of_find_node_by_name |of_find_node_by_phandle |of_get_parent |of_get_next_parent |of_get_next_child |of_find_compatible_node |of_match_node )(...); ... if (x == NULL) S <... when != x = E *if (...) { ... when != of_node_put(x) when != if (...) { ... of_node_put(x); ... } ( return <+...x...+>; | * return ...; ) } ...> of_node_put(x); // Signed-off-by: Julia Lawall Acked-by: Timur Tabi Signed-off-by: Kumar Gala diff --git a/arch/powerpc/sysdev/qe_lib/qe.c b/arch/powerpc/sysdev/qe_lib/qe.c index 3da8014..90020de 100644 --- a/arch/powerpc/sysdev/qe_lib/qe.c +++ b/arch/powerpc/sysdev/qe_lib/qe.c @@ -640,6 +640,7 @@ unsigned int qe_get_num_of_snums(void) if ((num_of_snums < 28) || (num_of_snums > QE_NUM_OF_SNUM)) { /* No QE ever has fewer than 28 SNUMs */ pr_err("QE: number of snum is invalid\n"); + of_node_put(qe); return -EINVAL; } } -- cgit v0.10.2 From a28dec2f26013aad89446b1f708f948617bc28a2 Mon Sep 17 00:00:00 2001 From: Anton Vorontsov Date: Sun, 8 Aug 2010 18:03:33 +0400 Subject: powerpc/85xx: Add P1021 PCI IDs and quirks This is needed for proper PCI-E support on P1021 SoCs. Signed-off-by: Anton Vorontsov Signed-off-by: Kumar Gala diff --git a/arch/powerpc/sysdev/fsl_pci.c b/arch/powerpc/sysdev/fsl_pci.c index 209384b..4ae9332 100644 --- a/arch/powerpc/sysdev/fsl_pci.c +++ b/arch/powerpc/sysdev/fsl_pci.c @@ -399,6 +399,8 @@ DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P1013E, quirk_fsl_pcie_header); DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P1013, quirk_fsl_pcie_header); DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P1020E, quirk_fsl_pcie_header); DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P1020, quirk_fsl_pcie_header); +DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P1021E, quirk_fsl_pcie_header); +DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P1021, quirk_fsl_pcie_header); DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P1022E, quirk_fsl_pcie_header); DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P1022, quirk_fsl_pcie_header); DECLARE_PCI_FIXUP_HEADER(0x1957, PCI_DEVICE_ID_P2010E, quirk_fsl_pcie_header); diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index f6a3b2d..10d3330 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -2300,6 +2300,8 @@ #define PCI_DEVICE_ID_P2010 0x0079 #define PCI_DEVICE_ID_P1020E 0x0100 #define PCI_DEVICE_ID_P1020 0x0101 +#define PCI_DEVICE_ID_P1021E 0x0102 +#define PCI_DEVICE_ID_P1021 0x0103 #define PCI_DEVICE_ID_P1011E 0x0108 #define PCI_DEVICE_ID_P1011 0x0109 #define PCI_DEVICE_ID_P1022E 0x0110 -- cgit v0.10.2 From 17134d96735115644cc2f0e2b1bab51ca6e3ab95 Mon Sep 17 00:00:00 2001 From: Stephen Hemminger Date: Tue, 31 Aug 2010 15:25:55 -0700 Subject: PCI: bus speed strings should be const Signed-off-by: Stephen Hemminger Signed-off-by: Jesse Barnes diff --git a/drivers/pci/slot.c b/drivers/pci/slot.c index 659eaa0..968cfea 100644 --- a/drivers/pci/slot.c +++ b/drivers/pci/slot.c @@ -49,7 +49,7 @@ static ssize_t address_read_file(struct pci_slot *slot, char *buf) } /* these strings match up with the values in pci_bus_speed */ -static char *pci_bus_speed_strings[] = { +static const char *pci_bus_speed_strings[] = { "33 MHz PCI", /* 0x00 */ "66 MHz PCI", /* 0x01 */ "66 MHz PCI-X", /* 0x02 */ -- cgit v0.10.2 From 950eaaca681c44aab87a46225c9e44f902c080aa Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 31 Aug 2010 17:00:18 -0700 Subject: pid: make setpgid() system call use RCU read-side critical section [ 23.584719] [ 23.584720] =================================================== [ 23.585059] [ INFO: suspicious rcu_dereference_check() usage. ] [ 23.585176] --------------------------------------------------- [ 23.585176] kernel/pid.c:419 invoked rcu_dereference_check() without protection! [ 23.585176] [ 23.585176] other info that might help us debug this: [ 23.585176] [ 23.585176] [ 23.585176] rcu_scheduler_active = 1, debug_locks = 1 [ 23.585176] 1 lock held by rc.sysinit/728: [ 23.585176] #0: (tasklist_lock){.+.+..}, at: [] sys_setpgid+0x5f/0x193 [ 23.585176] [ 23.585176] stack backtrace: [ 23.585176] Pid: 728, comm: rc.sysinit Not tainted 2.6.36-rc2 #2 [ 23.585176] Call Trace: [ 23.585176] [] lockdep_rcu_dereference+0x99/0xa2 [ 23.585176] [] find_task_by_pid_ns+0x50/0x6a [ 23.585176] [] find_task_by_vpid+0x1d/0x1f [ 23.585176] [] sys_setpgid+0x67/0x193 [ 23.585176] [] system_call_fastpath+0x16/0x1b [ 24.959669] type=1400 audit(1282938522.956:4): avc: denied { module_request } for pid=766 comm="hwclock" kmod="char-major-10-135" scontext=system_u:system_r:hwclock_t:s0 tcontext=system_u:system_r:kernel_t:s0 tclas It turns out that the setpgid() system call fails to enter an RCU read-side critical section before doing a PID-to-task_struct translation. This commit therefore does rcu_read_lock() before the translation, and also does rcu_read_unlock() after the last use of the returned pointer. Reported-by: Andrew Morton Signed-off-by: Paul E. McKenney Acked-by: David Howells diff --git a/kernel/sys.c b/kernel/sys.c index e9ad444..7f5a0cd 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -931,6 +931,7 @@ SYSCALL_DEFINE2(setpgid, pid_t, pid, pid_t, pgid) pgid = pid; if (pgid < 0) return -EINVAL; + rcu_read_lock(); /* From this point forward we keep holding onto the tasklist lock * so that our parent does not change from under us. -DaveM @@ -984,6 +985,7 @@ SYSCALL_DEFINE2(setpgid, pid_t, pid, pid_t, pgid) out: /* All paths lead to here, thus we are safe. -DaveM */ write_unlock_irq(&tasklist_lock); + rcu_read_unlock(); return err; } -- cgit v0.10.2 From 57157becdd1d23e6c2b8661ffe6c78d7d605d121 Mon Sep 17 00:00:00 2001 From: Henrik Rydberg Date: Tue, 31 Aug 2010 17:27:02 -0700 Subject: Input: bcm5974 - adjust major/minor to scale By visual inspection, the reported touch_major and touch_minor axes are a factor of two too small. Presumably the device actually reports the semi-major and semi-minor axes. Corrected with this patch. Signed-off-by: Henrik Rydberg Signed-off-by: Dmitry Torokhov diff --git a/drivers/input/mouse/bcm5974.c b/drivers/input/mouse/bcm5974.c index ea67c49..b952317 100644 --- a/drivers/input/mouse/bcm5974.c +++ b/drivers/input/mouse/bcm5974.c @@ -337,10 +337,14 @@ static void report_finger_data(struct input_dev *input, const struct bcm5974_config *cfg, const struct tp_finger *f) { - input_report_abs(input, ABS_MT_TOUCH_MAJOR, raw2int(f->force_major)); - input_report_abs(input, ABS_MT_TOUCH_MINOR, raw2int(f->force_minor)); - input_report_abs(input, ABS_MT_WIDTH_MAJOR, raw2int(f->size_major)); - input_report_abs(input, ABS_MT_WIDTH_MINOR, raw2int(f->size_minor)); + input_report_abs(input, ABS_MT_TOUCH_MAJOR, + raw2int(f->force_major) << 1); + input_report_abs(input, ABS_MT_TOUCH_MINOR, + raw2int(f->force_minor) << 1); + input_report_abs(input, ABS_MT_WIDTH_MAJOR, + raw2int(f->size_major) << 1); + input_report_abs(input, ABS_MT_WIDTH_MINOR, + raw2int(f->size_minor) << 1); input_report_abs(input, ABS_MT_ORIENTATION, MAX_FINGER_ORIENTATION - raw2int(f->orientation)); input_report_abs(input, ABS_MT_POSITION_X, raw2int(f->abs_x)); -- cgit v0.10.2 From af045b86662f17bf130239a65995c61a34f00a6b Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Tue, 31 Aug 2010 17:27:02 -0700 Subject: Input: i8042 - fix device removal on unload We need to call platform_device_unregister(i8042_platform_device) before calling platform_driver_unregister() because i8042_remove() resets i8042_platform_device to NULL. This leaves the platform device instance behind and prevents driver reload. Fixes https://bugzilla.kernel.org/show_bug.cgi?id=16613 Reported-by: Seryodkin Victor Cc: stable@kernel.org Signed-off-by: Dmitry Torokhov diff --git a/drivers/input/serio/i8042.c b/drivers/input/serio/i8042.c index 46e4ba0..f585131 100644 --- a/drivers/input/serio/i8042.c +++ b/drivers/input/serio/i8042.c @@ -1485,8 +1485,8 @@ static int __init i8042_init(void) static void __exit i8042_exit(void) { - platform_driver_unregister(&i8042_driver); platform_device_unregister(i8042_platform_device); + platform_driver_unregister(&i8042_driver); i8042_platform_exit(); panic_blink = NULL; -- cgit v0.10.2 From 6445671b00b3c4bcb26c8e0fc74abe780a67b901 Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Tue, 31 Aug 2010 22:26:15 -0400 Subject: alpha: remove homegrown L1_CACHE_ALIGN macro Let's use the standard L1_CACHE_ALIGN macro instead. Signed-off-by: FUJITA Tomonori Signed-off-by: Matt Turner diff --git a/arch/alpha/include/asm/cache.h b/arch/alpha/include/asm/cache.h index f199e69..ad368a9 100644 --- a/arch/alpha/include/asm/cache.h +++ b/arch/alpha/include/asm/cache.h @@ -17,7 +17,6 @@ # define L1_CACHE_SHIFT 5 #endif -#define L1_CACHE_ALIGN(x) (((x)+(L1_CACHE_BYTES-1))&~(L1_CACHE_BYTES-1)) #define SMP_CACHE_BYTES L1_CACHE_BYTES #endif -- cgit v0.10.2 From 59b25ed91400ace98d6cf0d59b1cb6928ad5cd37 Mon Sep 17 00:00:00 2001 From: "Morten H. Larsen" Date: Tue, 31 Aug 2010 22:29:13 -0400 Subject: Fix call to replaced SuperIO functions This patch fixes the failure to compile Alpha Generic because of previously overlooked calls to ns87312_enable_ide(). The function has been replaced by newer SuperIO code. Tested-by: Michael Cree Signed-off-by: Morten H. Larsen Signed-off-by: Matt Turner diff --git a/arch/alpha/kernel/proto.h b/arch/alpha/kernel/proto.h index 3d2627e..d3e52d3 100644 --- a/arch/alpha/kernel/proto.h +++ b/arch/alpha/kernel/proto.h @@ -156,9 +156,6 @@ extern void SMC669_Init(int); /* es1888.c */ extern void es1888_init(void); -/* ns87312.c */ -extern void ns87312_enable_ide(long ide_base); - /* ../lib/fpreg.c */ extern void alpha_write_fp_reg (unsigned long reg, unsigned long val); extern unsigned long alpha_read_fp_reg (unsigned long reg); diff --git a/arch/alpha/kernel/sys_cabriolet.c b/arch/alpha/kernel/sys_cabriolet.c index affd0f3..14c8898 100644 --- a/arch/alpha/kernel/sys_cabriolet.c +++ b/arch/alpha/kernel/sys_cabriolet.c @@ -33,7 +33,7 @@ #include "irq_impl.h" #include "pci_impl.h" #include "machvec_impl.h" - +#include "pc873xx.h" /* Note mask bit is true for DISABLED irqs. */ static unsigned long cached_irq_mask = ~0UL; @@ -236,17 +236,30 @@ cabriolet_map_irq(struct pci_dev *dev, u8 slot, u8 pin) } static inline void __init +cabriolet_enable_ide(void) +{ + if (pc873xx_probe() == -1) { + printk(KERN_ERR "Probing for PC873xx Super IO chip failed.\n"); + } else { + printk(KERN_INFO "Found %s Super IO chip at 0x%x\n", + pc873xx_get_model(), pc873xx_get_base()); + + pc873xx_enable_ide(); + } +} + +static inline void __init cabriolet_init_pci(void) { common_init_pci(); - ns87312_enable_ide(0x398); + cabriolet_enable_ide(); } static inline void __init cia_cab_init_pci(void) { cia_init_pci(); - ns87312_enable_ide(0x398); + cabriolet_enable_ide(); } /* diff --git a/arch/alpha/kernel/sys_takara.c b/arch/alpha/kernel/sys_takara.c index 2304648..4da596b 100644 --- a/arch/alpha/kernel/sys_takara.c +++ b/arch/alpha/kernel/sys_takara.c @@ -29,7 +29,7 @@ #include "irq_impl.h" #include "pci_impl.h" #include "machvec_impl.h" - +#include "pc873xx.h" /* Note mask bit is true for DISABLED irqs. */ static unsigned long cached_irq_mask[2] = { -1, -1 }; @@ -264,7 +264,14 @@ takara_init_pci(void) alpha_mv.pci_map_irq = takara_map_irq_srm; cia_init_pci(); - ns87312_enable_ide(0x26e); + + if (pc873xx_probe() == -1) { + printk(KERN_ERR "Probing for PC873xx Super IO chip failed.\n"); + } else { + printk(KERN_INFO "Found %s Super IO chip at 0x%x\n", + pc873xx_get_model(), pc873xx_get_base()); + pc873xx_enable_ide(); + } } -- cgit v0.10.2 From 7b598cdd03e35a4361f4ff7ebc630aa508c35be6 Mon Sep 17 00:00:00 2001 From: Michael Cree Date: Tue, 31 Aug 2010 22:46:04 -0400 Subject: alpha: convert perf_event to use local_t Updates the Alpha perf_event code to match the changes recently made to the core perf_event code in commit e78505958cf123048fb48cb56b79cebb8edd15fb. Signed-off-by: Michael Cree Signed-off-by: Matt Turner diff --git a/arch/alpha/kernel/perf_event.c b/arch/alpha/kernel/perf_event.c index 51c39fa..85d8e4f 100644 --- a/arch/alpha/kernel/perf_event.c +++ b/arch/alpha/kernel/perf_event.c @@ -241,20 +241,20 @@ static inline unsigned long alpha_read_pmc(int idx) static int alpha_perf_event_set_period(struct perf_event *event, struct hw_perf_event *hwc, int idx) { - long left = atomic64_read(&hwc->period_left); + long left = local64_read(&hwc->period_left); long period = hwc->sample_period; int ret = 0; if (unlikely(left <= -period)) { left = period; - atomic64_set(&hwc->period_left, left); + local64_set(&hwc->period_left, left); hwc->last_period = period; ret = 1; } if (unlikely(left <= 0)) { left += period; - atomic64_set(&hwc->period_left, left); + local64_set(&hwc->period_left, left); hwc->last_period = period; ret = 1; } @@ -269,7 +269,7 @@ static int alpha_perf_event_set_period(struct perf_event *event, if (left > (long)alpha_pmu->pmc_max_period[idx]) left = alpha_pmu->pmc_max_period[idx]; - atomic64_set(&hwc->prev_count, (unsigned long)(-left)); + local64_set(&hwc->prev_count, (unsigned long)(-left)); alpha_write_pmc(idx, (unsigned long)(-left)); @@ -300,10 +300,10 @@ static unsigned long alpha_perf_event_update(struct perf_event *event, long delta; again: - prev_raw_count = atomic64_read(&hwc->prev_count); + prev_raw_count = local64_read(&hwc->prev_count); new_raw_count = alpha_read_pmc(idx); - if (atomic64_cmpxchg(&hwc->prev_count, prev_raw_count, + if (local64_cmpxchg(&hwc->prev_count, prev_raw_count, new_raw_count) != prev_raw_count) goto again; @@ -316,8 +316,8 @@ again: delta += alpha_pmu->pmc_max_period[idx] + 1; } - atomic64_add(delta, &event->count); - atomic64_sub(delta, &hwc->period_left); + local64_add(delta, &event->count); + local64_sub(delta, &hwc->period_left); return new_raw_count; } @@ -636,7 +636,7 @@ static int __hw_perf_event_init(struct perf_event *event) if (!hwc->sample_period) { hwc->sample_period = alpha_pmu->pmc_max_period[0]; hwc->last_period = hwc->sample_period; - atomic64_set(&hwc->period_left, hwc->sample_period); + local64_set(&hwc->period_left, hwc->sample_period); } return 0; -- cgit v0.10.2 From 68d3f1d810500e8b975bdf0b20dd83d060076b4b Mon Sep 17 00:00:00 2001 From: Don Zickus Date: Tue, 31 Aug 2010 23:00:07 -0400 Subject: lockup_detector: Sync touch_*_watchdog back to old semantics During my rewrite, the semantics of touch_nmi_watchdog and touch_softlockup_watchdog changed enough to break some drivers (mostly over preemptable regions). These are cases where long delays on one CPU (due to print_delay for example) can cause long delays on other CPUs - so we must 'touch' the nmi_watchdog flag of those other CPUs as well. This change brings those touch_*_watchdog() functions back in line with to how they used to work. Signed-off-by: Don Zickus Acked-by: Cyrill Gorcunov Cc: peterz@infradead.org Cc: fweisbec@gmail.com LKML-Reference: <1283310009-22168-2-git-send-email-dzickus@redhat.com> Signed-off-by: Ingo Molnar diff --git a/kernel/watchdog.c b/kernel/watchdog.c index 0d53c8e..7f9c3c5 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -122,7 +122,7 @@ static void __touch_watchdog(void) void touch_softlockup_watchdog(void) { - __get_cpu_var(watchdog_touch_ts) = 0; + __raw_get_cpu_var(watchdog_touch_ts) = 0; } EXPORT_SYMBOL(touch_softlockup_watchdog); @@ -142,7 +142,14 @@ void touch_all_softlockup_watchdogs(void) #ifdef CONFIG_HARDLOCKUP_DETECTOR void touch_nmi_watchdog(void) { - __get_cpu_var(watchdog_nmi_touch) = true; + if (watchdog_enabled) { + unsigned cpu; + + for_each_present_cpu(cpu) { + if (per_cpu(watchdog_nmi_touch, cpu) != true) + per_cpu(watchdog_nmi_touch, cpu) = true; + } + } touch_softlockup_watchdog(); } EXPORT_SYMBOL(touch_nmi_watchdog); @@ -433,6 +440,9 @@ static int watchdog_enable(int cpu) wake_up_process(p); } + /* if any cpu succeeds, watchdog is considered enabled for the system */ + watchdog_enabled = 1; + return 0; } @@ -455,9 +465,6 @@ static void watchdog_disable(int cpu) per_cpu(softlockup_watchdog, cpu) = NULL; kthread_stop(p); } - - /* if any cpu succeeds, watchdog is considered enabled for the system */ - watchdog_enabled = 1; } static void watchdog_enable_all_cpus(void) -- cgit v0.10.2 From 25d3584797a39f57b69cd835722ac7c41113fb9a Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Mon, 16 Aug 2010 15:15:14 +0100 Subject: ARM: 6330/1: perf: reword comments relating to perf_event_do_pending This is purely a cosmetic change to the ARM perf backend because the current comments about the relationship between NMIs, interrupt context and perf_event_do_pending are misleading. This patch updates the comments so that they reflect what the code actually does (which is in line with other architectures). Acked-by: Jamie Iles Signed-off-by: Will Deacon Signed-off-by: Russell King diff --git a/arch/arm/include/asm/perf_event.h b/arch/arm/include/asm/perf_event.h index 48837e6..b5799a3 100644 --- a/arch/arm/include/asm/perf_event.h +++ b/arch/arm/include/asm/perf_event.h @@ -17,7 +17,7 @@ * counter interrupts are regular interrupts and not an NMI. This * means that when we receive the interrupt we can call * perf_event_do_pending() that handles all of the work with - * interrupts enabled. + * interrupts disabled. */ static inline void set_perf_event_pending(void) diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index 417c392..e213925 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -1041,8 +1041,8 @@ armv6pmu_handle_irq(int irq_num, /* * Handle the pending perf events. * - * Note: this call *must* be run with interrupts enabled. For - * platforms that can have the PMU interrupts raised as a PMI, this + * Note: this call *must* be run with interrupts disabled. For + * platforms that can have the PMU interrupts raised as an NMI, this * will not work. */ perf_event_do_pending(); @@ -2017,8 +2017,8 @@ static irqreturn_t armv7pmu_handle_irq(int irq_num, void *dev) /* * Handle the pending perf events. * - * Note: this call *must* be run with interrupts enabled. For - * platforms that can have the PMU interrupts raised as a PMI, this + * Note: this call *must* be run with interrupts disabled. For + * platforms that can have the PMU interrupts raised as an NMI, this * will not work. */ perf_event_do_pending(); -- cgit v0.10.2 From 418cf646c9999f2f7d216b1a693c2c575ab8094c Mon Sep 17 00:00:00 2001 From: Mikael Pettersson Date: Wed, 25 Aug 2010 20:49:01 +0100 Subject: ARM: 6343/1: wire up fanotify and prlimit64 syscalls on ARM The 2.6.36-rc kernel added three new system calls: fanotify_init, fanotify_mark, and prlimit64. This patch wires them up on ARM. The only non-trivial issue here is the u64 argument to sys_fanotify_mark(), but it is the 3rd argument and thus passed in r2/r3 in both kernel and user space, so it causes no problems. Tested with a 2.6.36-rc2 EABI kernel on an ixp4xx machine. Tested-by: Anand Gadiyar Signed-off-by: Mikael Pettersson Signed-off-by: Russell King diff --git a/arch/arm/include/asm/unistd.h b/arch/arm/include/asm/unistd.h index d02cfb6..c891eb7 100644 --- a/arch/arm/include/asm/unistd.h +++ b/arch/arm/include/asm/unistd.h @@ -393,6 +393,9 @@ #define __NR_perf_event_open (__NR_SYSCALL_BASE+364) #define __NR_recvmmsg (__NR_SYSCALL_BASE+365) #define __NR_accept4 (__NR_SYSCALL_BASE+366) +#define __NR_fanotify_init (__NR_SYSCALL_BASE+367) +#define __NR_fanotify_mark (__NR_SYSCALL_BASE+368) +#define __NR_prlimit64 (__NR_SYSCALL_BASE+369) /* * The following SWIs are ARM private. diff --git a/arch/arm/kernel/calls.S b/arch/arm/kernel/calls.S index afeb71f..5c26ecc 100644 --- a/arch/arm/kernel/calls.S +++ b/arch/arm/kernel/calls.S @@ -376,6 +376,9 @@ CALL(sys_perf_event_open) /* 365 */ CALL(sys_recvmmsg) CALL(sys_accept4) + CALL(sys_fanotify_init) + CALL(sys_fanotify_mark) + CALL(sys_prlimit64) #ifndef syscalls_counted .equ syscalls_padding, ((NR_syscalls + 3) & ~3) - NR_syscalls #define syscalls_counted -- cgit v0.10.2 From 3e073367a57d41e506f20aebb98e308387ce3090 Mon Sep 17 00:00:00 2001 From: Michael Cree Date: Wed, 1 Sep 2010 11:25:17 -0400 Subject: alpha: Fix printk format errors MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When compiling alpha generic build get errors such as: arch/alpha/kernel/err_marvel.c: In function ‘marvel_print_err_cyc’: arch/alpha/kernel/err_marvel.c:119: error: format ‘%ld’ expects type ‘long int’, but argument 6 has type ‘u64’ Replaced a number of %ld format specifiers with %lld since u64 is unsigned long long. Signed-off-by: Michael Cree Signed-off-by: Matt Turner diff --git a/arch/alpha/kernel/err_marvel.c b/arch/alpha/kernel/err_marvel.c index 52a79df..5c905aa 100644 --- a/arch/alpha/kernel/err_marvel.c +++ b/arch/alpha/kernel/err_marvel.c @@ -109,7 +109,7 @@ marvel_print_err_cyc(u64 err_cyc) #define IO7__ERR_CYC__CYCLE__M (0x7) printk("%s Packet In Error: %s\n" - "%s Error in %s, cycle %ld%s%s\n", + "%s Error in %s, cycle %lld%s%s\n", err_print_prefix, packet_desc[EXTRACT(err_cyc, IO7__ERR_CYC__PACKET)], err_print_prefix, @@ -313,7 +313,7 @@ marvel_print_po7_ugbge_sym(u64 ugbge_sym) } printk("%s Up Hose Garbage Symptom:\n" - "%s Source Port: %ld - Dest PID: %ld - OpCode: %s\n", + "%s Source Port: %lld - Dest PID: %lld - OpCode: %s\n", err_print_prefix, err_print_prefix, EXTRACT(ugbge_sym, IO7__PO7_UGBGE_SYM__UPH_SRC_PORT), @@ -552,7 +552,7 @@ marvel_print_pox_spl_cmplt(u64 spl_cmplt) #define IO7__POX_SPLCMPLT__REM_BYTE_COUNT__M (0xfff) printk("%s Split Completion Error:\n" - "%s Source (Bus:Dev:Func): %ld:%ld:%ld\n", + "%s Source (Bus:Dev:Func): %lld:%lld:%lld\n", err_print_prefix, err_print_prefix, EXTRACT(spl_cmplt, IO7__POX_SPLCMPLT__SOURCE_BUS), -- cgit v0.10.2 From 269f45c25028c75fe10e6d9be86e7202ab461fbc Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Wed, 1 Sep 2010 14:50:50 +0200 Subject: oprofile, x86: fix init_sysfs() function stub MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The use of the return value of init_sysfs() with commit 10f0412 oprofile, x86: fix init_sysfs error handling discovered the following build error for !CONFIG_PM: .../linux/arch/x86/oprofile/nmi_int.c: In function ‘op_nmi_init’: .../linux/arch/x86/oprofile/nmi_int.c:784: error: expected expression before ‘do’ make[2]: *** [arch/x86/oprofile/nmi_int.o] Error 1 make[1]: *** [arch/x86/oprofile] Error 2 This patch fixes this. Reported-by: Ingo Molnar Cc: stable@kernel.org Signed-off-by: Robert Richter diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c index 73a41d3..cfe4faa 100644 --- a/arch/x86/oprofile/nmi_int.c +++ b/arch/x86/oprofile/nmi_int.c @@ -585,8 +585,10 @@ static void exit_sysfs(void) } #else -#define init_sysfs() do { } while (0) -#define exit_sysfs() do { } while (0) + +static inline int init_sysfs(void) { return 0; } +static inline void exit_sysfs(void) { } + #endif /* CONFIG_PM */ static int __init p4_init(char **cpu_type) -- cgit v0.10.2 From aa8a9e25c5e865de85410be4e9269b552974f7a2 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Wed, 1 Sep 2010 13:10:53 -0700 Subject: MAINTAINERS: Add ehea driver as Supported This change just add the IBM eHEA 10Gb network drivers as supported. Signed-off-by: Breno Leitao Signed-off-by: David S. Miller diff --git a/MAINTAINERS b/MAINTAINERS index 8e85429..db6b351 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2171,6 +2171,12 @@ L: linux-rdma@vger.kernel.org S: Supported F: drivers/infiniband/hw/ehca/ +EHEA (IBM pSeries eHEA 10Gb ethernet adapter) DRIVER +M: Breno Leitao +L: netdev@vger.kernel.org +S: Maintained +F: drivers/net/ehea/ + EMBEDDED LINUX M: Paul Gortmaker M: Matt Mackall -- cgit v0.10.2 From 78b620ce9e168d08ecfac2f4bb056c511b0601ec Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 31 Aug 2010 02:05:57 +0000 Subject: vhost: stop worker only if created Its currently illegal to call kthread_stop(NULL) Reported-by: Ingo Molnar Signed-off-by: Eric Dumazet Acked-by: Tejun Heo Signed-off-by: David S. Miller diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index e05557d..4b99117 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -323,7 +323,10 @@ void vhost_dev_cleanup(struct vhost_dev *dev) dev->mm = NULL; WARN_ON(!list_empty(&dev->work_list)); - kthread_stop(dev->worker); + if (dev->worker) { + kthread_stop(dev->worker); + dev->worker = NULL; + } } static int log_access_ok(void __user *log_base, u64 addr, unsigned long sz) -- cgit v0.10.2 From 0f04cfd098fb81fded74e78ea1a1b86cc6c6c31e Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Tue, 31 Aug 2010 13:21:42 +0000 Subject: net sched: fix kernel leak in act_police While reviewing commit 1c40be12f7d8ca1d387510d39787b12e512a7ce8, I audited other users of tc_action_ops->dump for information leaks. That commit covered almost all of them but act_police still had a leak. opt.limit and opt.capab aren't zeroed out before the structure is passed out. This patch uses the C99 initializers to zero everything unused out. Signed-off-by: Jeff Mahoney Acked-by: Jeff Mahoney Signed-off-by: David S. Miller diff --git a/net/sched/act_police.c b/net/sched/act_police.c index 537a487..7ebf743 100644 --- a/net/sched/act_police.c +++ b/net/sched/act_police.c @@ -350,22 +350,19 @@ tcf_act_police_dump(struct sk_buff *skb, struct tc_action *a, int bind, int ref) { unsigned char *b = skb_tail_pointer(skb); struct tcf_police *police = a->priv; - struct tc_police opt; - - opt.index = police->tcf_index; - opt.action = police->tcf_action; - opt.mtu = police->tcfp_mtu; - opt.burst = police->tcfp_burst; - opt.refcnt = police->tcf_refcnt - ref; - opt.bindcnt = police->tcf_bindcnt - bind; + struct tc_police opt = { + .index = police->tcf_index, + .action = police->tcf_action, + .mtu = police->tcfp_mtu, + .burst = police->tcfp_burst, + .refcnt = police->tcf_refcnt - ref, + .bindcnt = police->tcf_bindcnt - bind, + }; + if (police->tcfp_R_tab) opt.rate = police->tcfp_R_tab->rate; - else - memset(&opt.rate, 0, sizeof(opt.rate)); if (police->tcfp_P_tab) opt.peakrate = police->tcfp_P_tab->rate; - else - memset(&opt.peakrate, 0, sizeof(opt.peakrate)); NLA_PUT(skb, TCA_POLICE_TBF, sizeof(opt), &opt); if (police->tcfp_result) NLA_PUT_U32(skb, TCA_POLICE_RESULT, police->tcfp_result); -- cgit v0.10.2 From 9c01ae58d4fee39e2af5b1379ee5431dd585cf62 Mon Sep 17 00:00:00 2001 From: Denis Kirjanov Date: Sun, 29 Aug 2010 21:21:38 +0000 Subject: pxa168_eth: fix a mdiobus leak mdiobus resources must be released on exit Signed-off-by: Denis Kirjanov Acked-by: Dan Carpenter Signed-off-by: David S. Miller diff --git a/drivers/net/pxa168_eth.c b/drivers/net/pxa168_eth.c index 410ea0a..85eddda 100644 --- a/drivers/net/pxa168_eth.c +++ b/drivers/net/pxa168_eth.c @@ -1606,6 +1606,8 @@ static int pxa168_eth_remove(struct platform_device *pdev) iounmap(pep->base); pep->base = NULL; + mdiobus_unregister(pep->smi_bus); + mdiobus_free(pep->smi_bus); unregister_netdev(dev); flush_scheduled_work(); free_netdev(dev); -- cgit v0.10.2 From 3b2eb6131e2f6ff646abb0fc69648179b8b70216 Mon Sep 17 00:00:00 2001 From: Michal Soltys Date: Mon, 30 Aug 2010 11:34:10 +0000 Subject: net/sched/sch_hfsc.c: initialize parent's cl_cfmin properly in init_vf() This patch fixes init_vf() function, so on each new backlog period parent's cl_cfmin is properly updated (including further propgation towards the root), even if the activated leaf has no upperlimit curve defined. Signed-off-by: Michal Soltys Signed-off-by: David S. Miller diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c index abd904b..4749609 100644 --- a/net/sched/sch_hfsc.c +++ b/net/sched/sch_hfsc.c @@ -761,8 +761,8 @@ init_vf(struct hfsc_class *cl, unsigned int len) if (f != cl->cl_f) { cl->cl_f = f; cftree_update(cl); - update_cfmin(cl->cl_parent); } + update_cfmin(cl->cl_parent); } } -- cgit v0.10.2 From 928497f0209027ccd649480a38a499fab9c3f6f6 Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Tue, 31 Aug 2010 05:54:00 +0000 Subject: xfrm_user: avoid a warning with some compiler Attached is a small patch to remove a warning ("warning: ISO C90 forbids mixed declarations and code" with gcc 4.3.2). Signed-off-by: Nicolas Dichtel Signed-off-by: David S. Miller diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index b14ed4b..8bae6b2 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -1801,7 +1801,7 @@ static int xfrm_add_sa_expire(struct sk_buff *skb, struct nlmsghdr *nlh, struct xfrm_user_expire *ue = nlmsg_data(nlh); struct xfrm_usersa_info *p = &ue->state; struct xfrm_mark m; - u32 mark = xfrm_mark_get(attrs, &m);; + u32 mark = xfrm_mark_get(attrs, &m); x = xfrm_state_lookup(net, mark, &p->id.daddr, p->id.spi, p->id.proto, p->family); -- cgit v0.10.2 From 750e9fad8c7f44e0005ffb7195f72dd76978c2cf Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Tue, 31 Aug 2010 05:50:43 +0000 Subject: ipv4: minor fix about RPF in help of Kconfig Signed-off-by: Nicolas Dichtel Signed-off-by: David S. Miller diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig index 7c3a7d1..571f895 100644 --- a/net/ipv4/Kconfig +++ b/net/ipv4/Kconfig @@ -46,7 +46,7 @@ config IP_ADVANCED_ROUTER rp_filter on use: echo 1 > /proc/sys/net/ipv4/conf//rp_filter - and + or echo 1 > /proc/sys/net/ipv4/conf/all/rp_filter Note that some distributions enable it in startup scripts. -- cgit v0.10.2 From de6be6c1f77798c4da38301693d33aff1cd76e84 Mon Sep 17 00:00:00 2001 From: stephen hemminger Date: Mon, 30 Aug 2010 07:51:17 +0000 Subject: sky2: don't do GRO on second port There's something very important I forgot to tell you. What? Don't cross the GRO streams. Why? It would be bad. I'm fuzzy on the whole good/bad thing. What do you mean, "bad"? Try to imagine all the Internet as you know it stopping instantaneously and every bit in every packet swapping at the speed of light. Total packet reordering. Right. That's bad. Okay. All right. Important safety tip. Thanks, Hubert The simplest way to stop this is just avoid doing GRO on the second port. Very few Marvell boards support two ports per ring, and GRO is just an optimization. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index 194e5cf..a6ff113 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -2520,24 +2520,27 @@ static inline void sky2_tx_done(struct net_device *dev, u16 last) } } -static inline void sky2_skb_rx(const struct sky2_port *sky2, +static inline void sky2_skb_rx(struct napi_struct *napi, + const struct sky2_port *sky2, u32 status, struct sk_buff *skb) { #ifdef SKY2_VLAN_TAG_USED - u16 vlan_tag = be16_to_cpu(sky2->rx_tag); if (sky2->vlgrp && (status & GMR_FS_VLAN)) { - if (skb->ip_summed == CHECKSUM_NONE) + u16 vlan_tag = be16_to_cpu(sky2->rx_tag); + + if (skb->ip_summed == CHECKSUM_NONE || + sky2->netdev != napi->dev) vlan_hwaccel_receive_skb(skb, sky2->vlgrp, vlan_tag); else - vlan_gro_receive(&sky2->hw->napi, sky2->vlgrp, - vlan_tag, skb); + vlan_gro_receive(napi, sky2->vlgrp, vlan_tag, skb); return; } #endif - if (skb->ip_summed == CHECKSUM_NONE) + if (skb->ip_summed == CHECKSUM_NONE || + sky2->netdev != napi->dev) netif_receive_skb(skb); else - napi_gro_receive(&sky2->hw->napi, skb); + napi_gro_receive(napi, skb); } static inline void sky2_rx_done(struct sky2_hw *hw, unsigned port, @@ -2638,7 +2641,7 @@ static int sky2_status_intr(struct sky2_hw *hw, int to_do, u16 idx) skb->protocol = eth_type_trans(skb, dev); - sky2_skb_rx(sky2, status, skb); + sky2_skb_rx(&hw->napi, sky2, status, skb); /* Stop after net poll weight */ if (++work_done >= to_do) -- cgit v0.10.2 From a9d8fb4d09ec8823045f44ebcde53a563fce2feb Mon Sep 17 00:00:00 2001 From: Erik Gilling Date: Fri, 27 Aug 2010 12:57:39 -0700 Subject: [ARM] tegra: Add ZRELADDR default for ARCH_TEGRA fixes the warning: .config:369:warning: symbol value '' invalid for ZRELADDR and the prompt for ZRELADDR on make Signed-off-by: Erik Gilling diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index a7ed21f..16bc8eb 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -1601,6 +1601,7 @@ config ZRELADDR ARCH_ORION5X ||\ ARCH_SPEAR3XX ||\ ARCH_SPEAR6XX ||\ + ARCH_TEGRA ||\ ARCH_U8500 ||\ ARCH_VERSATILE ||\ ARCH_W90X900 -- cgit v0.10.2 From 93f68f1ef787d97ab688f78a01f446e85bb9a496 Mon Sep 17 00:00:00 2001 From: Nathan Fontenot Date: Wed, 18 Aug 2010 09:58:46 +0000 Subject: powerpc/pseries: Correct rtas_data_buf locking in dlpar code The dlpar code can cause a deadlock to occur when making the RTAS configure-connector call. This occurs because we make kmalloc calls, which can block, while parsing the rtas_data_buf and holding the rtas_data_buf_lock. This an cause issues if someone else attempts to grab the rtas_data_bug_lock. This patch alleviates this issue by copying the contents of the rtas_data_buf to a local buffer before parsing. This allows us to only hold the rtas_data_buf_lock around the RTAS configure-connector calls. Signed-off-by: Nathan Fontenot Signed-off-by: Benjamin Herrenschmidt diff --git a/arch/powerpc/platforms/pseries/dlpar.c b/arch/powerpc/platforms/pseries/dlpar.c index 227c1c3..72d8054 100644 --- a/arch/powerpc/platforms/pseries/dlpar.c +++ b/arch/powerpc/platforms/pseries/dlpar.c @@ -129,20 +129,35 @@ struct device_node *dlpar_configure_connector(u32 drc_index) struct property *property; struct property *last_property = NULL; struct cc_workarea *ccwa; + char *data_buf; int cc_token; - int rc; + int rc = -1; cc_token = rtas_token("ibm,configure-connector"); if (cc_token == RTAS_UNKNOWN_SERVICE) return NULL; - spin_lock(&rtas_data_buf_lock); - ccwa = (struct cc_workarea *)&rtas_data_buf[0]; + data_buf = kzalloc(RTAS_DATA_BUF_SIZE, GFP_KERNEL); + if (!data_buf) + return NULL; + + ccwa = (struct cc_workarea *)&data_buf[0]; ccwa->drc_index = drc_index; ccwa->zero = 0; - rc = rtas_call(cc_token, 2, 1, NULL, rtas_data_buf, NULL); - while (rc) { + do { + /* Since we release the rtas_data_buf lock between configure + * connector calls we want to re-populate the rtas_data_buffer + * with the contents of the previous call. + */ + spin_lock(&rtas_data_buf_lock); + + memcpy(rtas_data_buf, data_buf, RTAS_DATA_BUF_SIZE); + rc = rtas_call(cc_token, 2, 1, NULL, rtas_data_buf, NULL); + memcpy(data_buf, rtas_data_buf, RTAS_DATA_BUF_SIZE); + + spin_unlock(&rtas_data_buf_lock); + switch (rc) { case NEXT_SIBLING: dn = dlpar_parse_cc_node(ccwa); @@ -197,18 +212,19 @@ struct device_node *dlpar_configure_connector(u32 drc_index) "returned from configure-connector\n", rc); goto cc_error; } + } while (rc); - rc = rtas_call(cc_token, 2, 1, NULL, rtas_data_buf, NULL); +cc_error: + kfree(data_buf); + + if (rc) { + if (first_dn) + dlpar_free_cc_nodes(first_dn); + + return NULL; } - spin_unlock(&rtas_data_buf_lock); return first_dn; - -cc_error: - if (first_dn) - dlpar_free_cc_nodes(first_dn); - spin_unlock(&rtas_data_buf_lock); - return NULL; } static struct device_node *derive_parent(const char *path) -- cgit v0.10.2 From 24cd804d1dc60a74c53da983094df3516500c276 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Mon, 30 Aug 2010 14:15:33 +0000 Subject: 3c59x: Remove incorrect locking; correct documented lock hierarchy vortex_ioctl() was grabbing vortex_private::lock around its call to generic_mii_ioctl(). This is no longer necessary since there are more specific locks which the mdio_{read,write}() functions will obtain. Worse, those functions do not save and restore IRQ flags when locking the MII state, so interrupts will be enabled when generic_mii_ioctl() returns. Since there is currently no need for any function to call mdio_{read,write}() while holding another spinlock, do not change them to save and restore IRQ flags but remove the specification of ordering between vortex_private::lock and vortex_private::mii_lock. Signed-off-by: Ben Hutchings Signed-off-by: David S. Miller diff --git a/drivers/net/3c59x.c b/drivers/net/3c59x.c index c685a55..a045559 100644 --- a/drivers/net/3c59x.c +++ b/drivers/net/3c59x.c @@ -647,7 +647,7 @@ struct vortex_private { u16 io_size; /* Size of PCI region (for release_region) */ /* Serialises access to hardware other than MII and variables below. - * The lock hierarchy is rtnl_lock > lock > mii_lock > window_lock. */ + * The lock hierarchy is rtnl_lock > {lock, mii_lock} > window_lock. */ spinlock_t lock; spinlock_t mii_lock; /* Serialises access to MII */ @@ -2984,7 +2984,6 @@ static int vortex_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) { int err; struct vortex_private *vp = netdev_priv(dev); - unsigned long flags; pci_power_t state = 0; if(VORTEX_PCI(vp)) @@ -2994,9 +2993,7 @@ static int vortex_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) if(state != 0) pci_set_power_state(VORTEX_PCI(vp), PCI_D0); - spin_lock_irqsave(&vp->lock, flags); err = generic_mii_ioctl(&vp->mii, if_mii(rq), cmd, NULL); - spin_unlock_irqrestore(&vp->lock, flags); if(state != 0) pci_set_power_state(VORTEX_PCI(vp), state); -- cgit v0.10.2 From 0b3b4fea0a50cc669acc8634806c2ecd6474f68c Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 1 Sep 2010 11:24:42 -0400 Subject: drm/radeon/kms: remove useless clock code This code was originally for forcing some clocks on certain asics. However, this code was later moved to asic specific functions for all of the affected asics. The only users of the original code at this point were r600, rv770, and evergreen and the code was not relevant for those asics. So, remove it. Signed-off-by: Alex Deucher Signed-off-by: Dave Airlie diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index 957d506..fe31b0d 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -2054,11 +2054,6 @@ int evergreen_resume(struct radeon_device *rdev) */ /* post card */ atom_asic_init(rdev->mode_info.atom_context); - /* Initialize clocks */ - r = radeon_clocks_init(rdev); - if (r) { - return r; - } r = evergreen_startup(rdev); if (r) { @@ -2164,9 +2159,6 @@ int evergreen_init(struct radeon_device *rdev) radeon_surface_init(rdev); /* Initialize clocks */ radeon_get_clock_info(rdev->ddev); - r = radeon_clocks_init(rdev); - if (r) - return r; /* Fence driver */ r = radeon_fence_driver_init(rdev); if (r) @@ -2236,7 +2228,6 @@ void evergreen_fini(struct radeon_device *rdev) evergreen_pcie_gart_fini(rdev); radeon_gem_fini(rdev); radeon_fence_driver_fini(rdev); - radeon_clocks_fini(rdev); radeon_agp_fini(rdev); radeon_bo_fini(rdev); radeon_atombios_fini(rdev); diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index aee8376..04f134d 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -2489,11 +2489,6 @@ int r600_resume(struct radeon_device *rdev) */ /* post card */ atom_asic_init(rdev->mode_info.atom_context); - /* Initialize clocks */ - r = radeon_clocks_init(rdev); - if (r) { - return r; - } r = r600_startup(rdev); if (r) { @@ -2586,9 +2581,6 @@ int r600_init(struct radeon_device *rdev) radeon_surface_init(rdev); /* Initialize clocks */ radeon_get_clock_info(rdev->ddev); - r = radeon_clocks_init(rdev); - if (r) - return r; /* Fence driver */ r = radeon_fence_driver_init(rdev); if (r) @@ -2663,7 +2655,6 @@ void r600_fini(struct radeon_device *rdev) radeon_agp_fini(rdev); radeon_gem_fini(rdev); radeon_fence_driver_fini(rdev); - radeon_clocks_fini(rdev); radeon_bo_fini(rdev); radeon_atombios_fini(rdev); kfree(rdev->bios); diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index e90d9e3..a168d64 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -1339,8 +1339,6 @@ extern bool radeon_card_posted(struct radeon_device *rdev); extern void radeon_update_bandwidth_info(struct radeon_device *rdev); extern void radeon_update_display_priority(struct radeon_device *rdev); extern bool radeon_boot_test_post_card(struct radeon_device *rdev); -extern int radeon_clocks_init(struct radeon_device *rdev); -extern void radeon_clocks_fini(struct radeon_device *rdev); extern void radeon_scratch_init(struct radeon_device *rdev); extern void radeon_surface_init(struct radeon_device *rdev); extern int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data); diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c index a21bf88..25e1dd1 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.c +++ b/drivers/gpu/drm/radeon/radeon_asic.c @@ -858,21 +858,3 @@ int radeon_asic_init(struct radeon_device *rdev) return 0; } -/* - * Wrapper around modesetting bits. Move to radeon_clocks.c? - */ -int radeon_clocks_init(struct radeon_device *rdev) -{ - int r; - - r = radeon_static_clocks_init(rdev->ddev); - if (r) { - return r; - } - DRM_INFO("Clocks initialized !\n"); - return 0; -} - -void radeon_clocks_fini(struct radeon_device *rdev) -{ -} diff --git a/drivers/gpu/drm/radeon/radeon_clocks.c b/drivers/gpu/drm/radeon/radeon_clocks.c index 690d8907..5249af8 100644 --- a/drivers/gpu/drm/radeon/radeon_clocks.c +++ b/drivers/gpu/drm/radeon/radeon_clocks.c @@ -905,53 +905,3 @@ void radeon_legacy_set_clock_gating(struct radeon_device *rdev, int enable) } } -static void radeon_apply_clock_quirks(struct radeon_device *rdev) -{ - uint32_t tmp; - - /* XXX make sure engine is idle */ - - if (rdev->family < CHIP_RS600) { - tmp = RREG32_PLL(RADEON_SCLK_CNTL); - if (ASIC_IS_R300(rdev) || ASIC_IS_RV100(rdev)) - tmp |= RADEON_SCLK_FORCE_CP | RADEON_SCLK_FORCE_VIP; - if ((rdev->family == CHIP_RV250) - || (rdev->family == CHIP_RV280)) - tmp |= - RADEON_SCLK_FORCE_DISP1 | RADEON_SCLK_FORCE_DISP2; - if ((rdev->family == CHIP_RV350) - || (rdev->family == CHIP_RV380)) - tmp |= R300_SCLK_FORCE_VAP; - if (rdev->family == CHIP_R420) - tmp |= R300_SCLK_FORCE_PX | R300_SCLK_FORCE_TX; - WREG32_PLL(RADEON_SCLK_CNTL, tmp); - } else if (rdev->family < CHIP_R600) { - tmp = RREG32_PLL(AVIVO_CP_DYN_CNTL); - tmp |= AVIVO_CP_FORCEON; - WREG32_PLL(AVIVO_CP_DYN_CNTL, tmp); - - tmp = RREG32_PLL(AVIVO_E2_DYN_CNTL); - tmp |= AVIVO_E2_FORCEON; - WREG32_PLL(AVIVO_E2_DYN_CNTL, tmp); - - tmp = RREG32_PLL(AVIVO_IDCT_DYN_CNTL); - tmp |= AVIVO_IDCT_FORCEON; - WREG32_PLL(AVIVO_IDCT_DYN_CNTL, tmp); - } -} - -int radeon_static_clocks_init(struct drm_device *dev) -{ - struct radeon_device *rdev = dev->dev_private; - - /* XXX make sure engine is idle */ - - if (radeon_dynclks != -1) { - if (radeon_dynclks) { - if (rdev->asic->set_clock_gating) - radeon_set_clock_gating(rdev, 1); - } - } - radeon_apply_clock_quirks(rdev); - return 0; -} diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h index 8f93e2b..efbe975 100644 --- a/drivers/gpu/drm/radeon/radeon_mode.h +++ b/drivers/gpu/drm/radeon/radeon_mode.h @@ -600,7 +600,6 @@ extern bool radeon_get_atom_connector_info_from_supported_devices_table(struct d void radeon_enc_destroy(struct drm_encoder *encoder); void radeon_copy_fb(struct drm_device *dev, struct drm_gem_object *dst_obj); void radeon_combios_asic_init(struct drm_device *dev); -extern int radeon_static_clocks_init(struct drm_device *dev); bool radeon_crtc_scaling_mode_fixup(struct drm_crtc *crtc, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode); diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c index a84e386..bfa59db 100644 --- a/drivers/gpu/drm/radeon/rv770.c +++ b/drivers/gpu/drm/radeon/rv770.c @@ -1074,11 +1074,6 @@ int rv770_resume(struct radeon_device *rdev) */ /* post card */ atom_asic_init(rdev->mode_info.atom_context); - /* Initialize clocks */ - r = radeon_clocks_init(rdev); - if (r) { - return r; - } r = rv770_startup(rdev); if (r) { @@ -1169,9 +1164,6 @@ int rv770_init(struct radeon_device *rdev) radeon_surface_init(rdev); /* Initialize clocks */ radeon_get_clock_info(rdev->ddev); - r = radeon_clocks_init(rdev); - if (r) - return r; /* Fence driver */ r = radeon_fence_driver_init(rdev); if (r) @@ -1249,7 +1241,6 @@ void rv770_fini(struct radeon_device *rdev) rv770_vram_scratch_fini(rdev); radeon_gem_fini(rdev); radeon_fence_driver_fini(rdev); - radeon_clocks_fini(rdev); radeon_agp_fini(rdev); radeon_bo_fini(rdev); radeon_atombios_fini(rdev); -- cgit v0.10.2 From 0d9958b18e10d7426d94cc3dd024920a40db3ee2 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 1 Sep 2010 12:03:37 -0400 Subject: drm/radeon/kms: force legacy pll algo for RV515 LVDS There has been periodic evidence that LVDS, on at least some panels, prefers the dividers selected by the legacy pll algo. This patch forces the use of the legacy pll algo on RV515 LVDS panels. The old behavior (new pll algo) can be selected by setting the new_pll module parameter to 1. Signed-off-by: Alex Deucher Cc: stable@kernel.org Signed-off-by: Dave Airlie diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c index 577239a..977f5c2 100644 --- a/drivers/gpu/drm/radeon/atombios_crtc.c +++ b/drivers/gpu/drm/radeon/atombios_crtc.c @@ -534,6 +534,20 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc, pll->algo = PLL_ALGO_LEGACY; pll->flags |= RADEON_PLL_PREFER_CLOSEST_LOWER; } + /* There is some evidence (often anecdotal) that RV515 LVDS + * (on some boards at least) prefers the legacy algo. I'm not + * sure whether this should handled generically or on a + * case-by-case quirk basis. Both algos should work fine in the + * majority of cases. + */ + if ((radeon_encoder->active_device & (ATOM_DEVICE_LCD_SUPPORT)) && + (rdev->family == CHIP_RV515)) { + /* allow the user to overrride just in case */ + if (radeon_new_pll == 1) + pll->algo = PLL_ALGO_NEW; + else + pll->algo = PLL_ALGO_LEGACY; + } } else { if (encoder->encoder_type != DRM_MODE_ENCODER_DAC) pll->flags |= RADEON_PLL_NO_ODD_POST_DIV; -- cgit v0.10.2 From cf4c12f9a2289e3679722590e1226ae8deb14385 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 1 Sep 2010 17:15:06 -0400 Subject: drm/radeon/kms: fix tv module parameter The tv parameter was added to disable the tv-out connector, however, it caused a crash if it was set to 0 due to drm_connector_init not getting called. If tv=0, don't attempt to add the connector. Might fix: https://bugzilla.kernel.org/show_bug.cgi?id=17241 Signed-off-by: Alex Deucher Signed-off-by: Dave Airlie diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index 1a5ee39..a9dd784 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -1051,10 +1051,16 @@ radeon_add_atom_connector(struct drm_device *dev, uint32_t subpixel_order = SubPixelNone; bool shared_ddc = false; - /* fixme - tv/cv/din */ if (connector_type == DRM_MODE_CONNECTOR_Unknown) return; + /* if the user selected tv=0 don't try and add the connector */ + if (((connector_type == DRM_MODE_CONNECTOR_SVIDEO) || + (connector_type == DRM_MODE_CONNECTOR_Composite) || + (connector_type == DRM_MODE_CONNECTOR_9PinDIN)) && + (radeon_tv == 0)) + return; + /* see if we already added it */ list_for_each_entry(connector, &dev->mode_config.connector_list, head) { radeon_connector = to_radeon_connector(connector); @@ -1209,19 +1215,17 @@ radeon_add_atom_connector(struct drm_device *dev, case DRM_MODE_CONNECTOR_SVIDEO: case DRM_MODE_CONNECTOR_Composite: case DRM_MODE_CONNECTOR_9PinDIN: - if (radeon_tv == 1) { - drm_connector_init(dev, &radeon_connector->base, &radeon_tv_connector_funcs, connector_type); - drm_connector_helper_add(&radeon_connector->base, &radeon_tv_connector_helper_funcs); - radeon_connector->dac_load_detect = true; - drm_connector_attach_property(&radeon_connector->base, - rdev->mode_info.load_detect_property, - 1); - drm_connector_attach_property(&radeon_connector->base, - rdev->mode_info.tv_std_property, - radeon_atombios_get_tv_info(rdev)); - /* no HPD on analog connectors */ - radeon_connector->hpd.hpd = RADEON_HPD_NONE; - } + drm_connector_init(dev, &radeon_connector->base, &radeon_tv_connector_funcs, connector_type); + drm_connector_helper_add(&radeon_connector->base, &radeon_tv_connector_helper_funcs); + radeon_connector->dac_load_detect = true; + drm_connector_attach_property(&radeon_connector->base, + rdev->mode_info.load_detect_property, + 1); + drm_connector_attach_property(&radeon_connector->base, + rdev->mode_info.tv_std_property, + radeon_atombios_get_tv_info(rdev)); + /* no HPD on analog connectors */ + radeon_connector->hpd.hpd = RADEON_HPD_NONE; break; case DRM_MODE_CONNECTOR_LVDS: radeon_dig_connector = kzalloc(sizeof(struct radeon_connector_atom_dig), GFP_KERNEL); @@ -1272,10 +1276,16 @@ radeon_add_legacy_connector(struct drm_device *dev, struct radeon_connector *radeon_connector; uint32_t subpixel_order = SubPixelNone; - /* fixme - tv/cv/din */ if (connector_type == DRM_MODE_CONNECTOR_Unknown) return; + /* if the user selected tv=0 don't try and add the connector */ + if (((connector_type == DRM_MODE_CONNECTOR_SVIDEO) || + (connector_type == DRM_MODE_CONNECTOR_Composite) || + (connector_type == DRM_MODE_CONNECTOR_9PinDIN)) && + (radeon_tv == 0)) + return; + /* see if we already added it */ list_for_each_entry(connector, &dev->mode_config.connector_list, head) { radeon_connector = to_radeon_connector(connector); @@ -1347,26 +1357,24 @@ radeon_add_legacy_connector(struct drm_device *dev, case DRM_MODE_CONNECTOR_SVIDEO: case DRM_MODE_CONNECTOR_Composite: case DRM_MODE_CONNECTOR_9PinDIN: - if (radeon_tv == 1) { - drm_connector_init(dev, &radeon_connector->base, &radeon_tv_connector_funcs, connector_type); - drm_connector_helper_add(&radeon_connector->base, &radeon_tv_connector_helper_funcs); - radeon_connector->dac_load_detect = true; - /* RS400,RC410,RS480 chipset seems to report a lot - * of false positive on load detect, we haven't yet - * found a way to make load detect reliable on those - * chipset, thus just disable it for TV. - */ - if (rdev->family == CHIP_RS400 || rdev->family == CHIP_RS480) - radeon_connector->dac_load_detect = false; - drm_connector_attach_property(&radeon_connector->base, - rdev->mode_info.load_detect_property, - radeon_connector->dac_load_detect); - drm_connector_attach_property(&radeon_connector->base, - rdev->mode_info.tv_std_property, - radeon_combios_get_tv_info(rdev)); - /* no HPD on analog connectors */ - radeon_connector->hpd.hpd = RADEON_HPD_NONE; - } + drm_connector_init(dev, &radeon_connector->base, &radeon_tv_connector_funcs, connector_type); + drm_connector_helper_add(&radeon_connector->base, &radeon_tv_connector_helper_funcs); + radeon_connector->dac_load_detect = true; + /* RS400,RC410,RS480 chipset seems to report a lot + * of false positive on load detect, we haven't yet + * found a way to make load detect reliable on those + * chipset, thus just disable it for TV. + */ + if (rdev->family == CHIP_RS400 || rdev->family == CHIP_RS480) + radeon_connector->dac_load_detect = false; + drm_connector_attach_property(&radeon_connector->base, + rdev->mode_info.load_detect_property, + radeon_connector->dac_load_detect); + drm_connector_attach_property(&radeon_connector->base, + rdev->mode_info.tv_std_property, + radeon_combios_get_tv_info(rdev)); + /* no HPD on analog connectors */ + radeon_connector->hpd.hpd = RADEON_HPD_NONE; break; case DRM_MODE_CONNECTOR_LVDS: drm_connector_init(dev, &radeon_connector->base, &radeon_lvds_connector_funcs, connector_type); -- cgit v0.10.2 From 95347871865ca5093c7e87a223274f7c3b5eccda Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 1 Sep 2010 17:20:42 -0400 Subject: drm/radeon/kms: properly set crtc high base on r7xx Signed-off-by: Alex Deucher Cc: stable@kernel.org Signed-off-by: Dave Airlie diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c index 977f5c2..71b3150 100644 --- a/drivers/gpu/drm/radeon/atombios_crtc.c +++ b/drivers/gpu/drm/radeon/atombios_crtc.c @@ -1070,11 +1070,11 @@ static int avivo_crtc_set_base(struct drm_crtc *crtc, int x, int y, if (rdev->family >= CHIP_RV770) { if (radeon_crtc->crtc_id) { - WREG32(R700_D2GRPH_PRIMARY_SURFACE_ADDRESS_HIGH, 0); - WREG32(R700_D2GRPH_SECONDARY_SURFACE_ADDRESS_HIGH, 0); + WREG32(R700_D2GRPH_PRIMARY_SURFACE_ADDRESS_HIGH, upper_32_bits(fb_location)); + WREG32(R700_D2GRPH_SECONDARY_SURFACE_ADDRESS_HIGH, upper_32_bits(fb_location)); } else { - WREG32(R700_D1GRPH_PRIMARY_SURFACE_ADDRESS_HIGH, 0); - WREG32(R700_D1GRPH_SECONDARY_SURFACE_ADDRESS_HIGH, 0); + WREG32(R700_D1GRPH_PRIMARY_SURFACE_ADDRESS_HIGH, upper_32_bits(fb_location)); + WREG32(R700_D1GRPH_SECONDARY_SURFACE_ADDRESS_HIGH, upper_32_bits(fb_location)); } } WREG32(AVIVO_D1GRPH_PRIMARY_SURFACE_ADDRESS + radeon_crtc->crtc_offset, -- cgit v0.10.2 From 87f94b4e91dc042620c527f3c30c37e5127ef757 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 1 Sep 2010 18:06:39 -0700 Subject: bridge: Clear INET control block of SKBs passed into ip_fragment(). In a similar vain to commit 17762060c25590bfddd68cc1131f28ec720f405f ("bridge: Clear IPCB before possible entry into IP stack") Any time we call into the IP stack we have to make sure the state there is as expected by the ipv4 code. With help from Eric Dumazet and Herbert Xu. Reported-by: Bandan Das Signed-off-by: David S. Miller diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c index 5ed00bd..137f232 100644 --- a/net/bridge/br_netfilter.c +++ b/net/bridge/br_netfilter.c @@ -761,9 +761,11 @@ static int br_nf_dev_queue_xmit(struct sk_buff *skb) { if (skb->nfct != NULL && skb->protocol == htons(ETH_P_IP) && skb->len + nf_bridge_mtu_reduction(skb) > skb->dev->mtu && - !skb_is_gso(skb)) + !skb_is_gso(skb)) { + /* BUG: Should really parse the IP options here. */ + memset(IPCB(skb), 0, sizeof(struct inet_skb_parm)); return ip_fragment(skb, br_dev_queue_push_xmit); - else + } else return br_dev_queue_push_xmit(skb); } #else -- cgit v0.10.2 From 3d3be4333fdf6faa080947b331a6a19bce1a4f57 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 1 Sep 2010 00:50:51 +0000 Subject: gro: fix different skb headrooms Packets entering GRO might have different headrooms, even for a given flow (because of implementation details in drivers, like copybreak). We cant force drivers to deliver packets with a fixed headroom. 1) fix skb_segment() skb_segment() makes the false assumption headrooms of fragments are same than the head. When CHECKSUM_PARTIAL is used, this can give csum_start errors, and crash later in skb_copy_and_csum_dev() 2) allocate a minimal skb for head of frag_list skb_gro_receive() uses netdev_alloc_skb(headroom + skb_gro_offset(p)) to allocate a fresh skb. This adds NET_SKB_PAD to a padding already provided by netdevice, depending on various things, like copybreak. Use alloc_skb() to allocate an exact padding, to reduce cache line needs: NET_SKB_PAD + NET_IP_ALIGN bugzilla : https://bugzilla.kernel.org/show_bug.cgi?id=16626 Many thanks to Plamen Petrov, testing many debugging patches ! With help of Jarek Poplawski. Reported-by: Plamen Petrov Signed-off-by: Eric Dumazet CC: Jarek Poplawski Signed-off-by: David S. Miller diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 3a2513f..26396ff 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -2573,6 +2573,10 @@ struct sk_buff *skb_segment(struct sk_buff *skb, int features) __copy_skb_header(nskb, skb); nskb->mac_len = skb->mac_len; + /* nskb and skb might have different headroom */ + if (nskb->ip_summed == CHECKSUM_PARTIAL) + nskb->csum_start += skb_headroom(nskb) - headroom; + skb_reset_mac_header(nskb); skb_set_network_header(nskb, skb->mac_len); nskb->transport_header = (nskb->network_header + @@ -2702,8 +2706,8 @@ int skb_gro_receive(struct sk_buff **head, struct sk_buff *skb) } else if (skb_gro_len(p) != pinfo->gso_size) return -E2BIG; - headroom = skb_headroom(p); - nskb = netdev_alloc_skb(p->dev, headroom + skb_gro_offset(p)); + headroom = NET_SKB_PAD + NET_IP_ALIGN; + nskb = alloc_skb(headroom + skb_gro_offset(p), GFP_ATOMIC); if (unlikely(!nskb)) return -ENOMEM; -- cgit v0.10.2 From ea39302b87b8d944d567ef29c0647c09da5743fa Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Fri, 27 Aug 2010 16:04:29 -0400 Subject: drm/radeon/kms/evergreen: work around bad data in some i2c tables The 7th entry in a lot of evergreen i2c gpio tables is partially zeroed. Fix the entry. Should fix the missing ddc entry in: https://bugs.freedesktop.org/show_bug.cgi?id=29255 Signed-off-by: Alex Deucher Signed-off-by: Dave Airlie diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c index 6114198..ebae14c 100644 --- a/drivers/gpu/drm/radeon/radeon_atombios.c +++ b/drivers/gpu/drm/radeon/radeon_atombios.c @@ -85,6 +85,19 @@ static inline struct radeon_i2c_bus_rec radeon_lookup_i2c_gpio(struct radeon_dev for (i = 0; i < num_indices; i++) { gpio = &i2c_info->asGPIO_Info[i]; + /* some evergreen boards have bad data for this entry */ + if (ASIC_IS_DCE4(rdev)) { + if ((i == 7) && + (gpio->usClkMaskRegisterIndex == 0x1936) && + (gpio->sucI2cId.ucAccess == 0)) { + gpio->sucI2cId.ucAccess = 0x97; + gpio->ucDataMaskShift = 8; + gpio->ucDataEnShift = 8; + gpio->ucDataY_Shift = 8; + gpio->ucDataA_Shift = 8; + } + } + if (gpio->sucI2cId.ucAccess == id) { i2c.mask_clk_reg = le16_to_cpu(gpio->usClkMaskRegisterIndex) * 4; i2c.mask_data_reg = le16_to_cpu(gpio->usDataMaskRegisterIndex) * 4; @@ -147,6 +160,20 @@ void radeon_atombios_i2c_init(struct radeon_device *rdev) for (i = 0; i < num_indices; i++) { gpio = &i2c_info->asGPIO_Info[i]; i2c.valid = false; + + /* some evergreen boards have bad data for this entry */ + if (ASIC_IS_DCE4(rdev)) { + if ((i == 7) && + (gpio->usClkMaskRegisterIndex == 0x1936) && + (gpio->sucI2cId.ucAccess == 0)) { + gpio->sucI2cId.ucAccess = 0x97; + gpio->ucDataMaskShift = 8; + gpio->ucDataEnShift = 8; + gpio->ucDataY_Shift = 8; + gpio->ucDataA_Shift = 8; + } + } + i2c.mask_clk_reg = le16_to_cpu(gpio->usClkMaskRegisterIndex) * 4; i2c.mask_data_reg = le16_to_cpu(gpio->usDataMaskRegisterIndex) * 4; i2c.en_clk_reg = le16_to_cpu(gpio->usClkEnRegisterIndex) * 4; -- cgit v0.10.2 From 9bc08a45fb117c696e4940cfa1208cb1cc7a2f25 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Thu, 2 Sep 2010 15:14:38 +1000 Subject: xfs: improve buffer cache hash scalability When doing large parallel file creates on a 16p machines, large amounts of time is being spent in _xfs_buf_find(). A system wide profile with perf top shows this: 1134740.00 19.3% _xfs_buf_find 733142.00 12.5% __ticket_spin_lock The problem is that the hash contains 45,000 buffers, and the hash table width is only 256 buffers. That means we've got around 200 buffers per chain, and searching it is quite expensive. The hash table size needs to increase. Secondly, every time we do a lookup, we promote the buffer we find to the head of the hash chain. This is causing cachelines to be dirtied and causes invalidation of cachelines across all CPUs that may have walked the hash chain recently. hence every walk of the hash chain is effectively a cold cache walk. Remove the promotion to avoid this invalidation. The results are: 1045043.00 21.2% __ticket_spin_lock 326184.00 6.6% _xfs_buf_find A 70% drop in the CPU usage when looking up buffers. Unfortunately that does not result in an increase in performance underthis workload as contention on the inode_lock soaks up most of the reduction in CPU usage. Signed-off-by: Dave Chinner Reviewed-by: Christoph Hellwig diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index ea79072..d72cf2b 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -440,12 +440,7 @@ _xfs_buf_find( ASSERT(btp == bp->b_target); if (bp->b_file_offset == range_base && bp->b_buffer_length == range_length) { - /* - * If we look at something, bring it to the - * front of the list for next time. - */ atomic_inc(&bp->b_hold); - list_move(&bp->b_hash_list, &hash->bh_list); goto found; } } @@ -1443,8 +1438,7 @@ xfs_alloc_bufhash( { unsigned int i; - btp->bt_hashshift = external ? 3 : 8; /* 8 or 256 buckets */ - btp->bt_hashmask = (1 << btp->bt_hashshift) - 1; + btp->bt_hashshift = external ? 3 : 12; /* 8 or 4096 buckets */ btp->bt_hash = kmem_zalloc_large((1 << btp->bt_hashshift) * sizeof(xfs_bufhash_t)); for (i = 0; i < (1 << btp->bt_hashshift); i++) { diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h index d072e5f..2a05614 100644 --- a/fs/xfs/linux-2.6/xfs_buf.h +++ b/fs/xfs/linux-2.6/xfs_buf.h @@ -137,7 +137,6 @@ typedef struct xfs_buftarg { size_t bt_smask; /* per device buffer hash table */ - uint bt_hashmask; uint bt_hashshift; xfs_bufhash_t *bt_hash; -- cgit v0.10.2 From 048e78a5bc22c27410cb5ca9680c3c7ac400607f Mon Sep 17 00:00:00 2001 From: David Henningsson Date: Thu, 2 Sep 2010 08:35:47 +0200 Subject: ALSA: hda - Add a new hp-laptop model for Conexant 5066, tested on HP G60 This new model adds the following functionality to HP G60: - Automute of internal speakers - Autoswitch of internal/external mics - Remove SPDIF not physically present BugLink: http://launchpad.net/bugs/587388 Cc: stable@kernel.org Signed-off-by: David Henningsson Signed-off-by: Takashi Iwai diff --git a/Documentation/sound/alsa/HD-Audio-Models.txt b/Documentation/sound/alsa/HD-Audio-Models.txt index ce46fa1..37c6aad 100644 --- a/Documentation/sound/alsa/HD-Audio-Models.txt +++ b/Documentation/sound/alsa/HD-Audio-Models.txt @@ -296,6 +296,7 @@ Conexant 5051 Conexant 5066 ============= laptop Basic Laptop config (default) + hp-laptop HP laptops, e g G60 dell-laptop Dell laptops dell-vostro Dell Vostro olpc-xo-1_5 OLPC XO 1.5 diff --git a/sound/pci/hda/patch_conexant.c b/sound/pci/hda/patch_conexant.c index 5cdb80e..4f06199 100644 --- a/sound/pci/hda/patch_conexant.c +++ b/sound/pci/hda/patch_conexant.c @@ -116,6 +116,7 @@ struct conexant_spec { unsigned int dell_vostro:1; unsigned int ideapad:1; unsigned int thinkpad:1; + unsigned int hp_laptop:1; unsigned int ext_mic_present; unsigned int recording; @@ -2299,6 +2300,18 @@ static void cxt5066_ideapad_automic(struct hda_codec *codec) } } +/* toggle input of built-in digital mic and mic jack appropriately */ +static void cxt5066_hp_laptop_automic(struct hda_codec *codec) +{ + unsigned int present; + + present = snd_hda_jack_detect(codec, 0x1b); + snd_printdd("CXT5066: external microphone present=%d\n", present); + snd_hda_codec_write(codec, 0x17, 0, AC_VERB_SET_CONNECT_SEL, + present ? 1 : 3); +} + + /* toggle input of built-in digital mic and mic jack appropriately order is: external mic -> dock mic -> interal mic */ static void cxt5066_thinkpad_automic(struct hda_codec *codec) @@ -2408,6 +2421,20 @@ static void cxt5066_ideapad_event(struct hda_codec *codec, unsigned int res) } /* unsolicited event for jack sensing */ +static void cxt5066_hp_laptop_event(struct hda_codec *codec, unsigned int res) +{ + snd_printdd("CXT5066_hp_laptop: unsol event %x (%x)\n", res, res >> 26); + switch (res >> 26) { + case CONEXANT_HP_EVENT: + cxt5066_hp_automute(codec); + break; + case CONEXANT_MIC_EVENT: + cxt5066_hp_laptop_automic(codec); + break; + } +} + +/* unsolicited event for jack sensing */ static void cxt5066_thinkpad_event(struct hda_codec *codec, unsigned int res) { snd_printdd("CXT5066_thinkpad: unsol event %x (%x)\n", res, res >> 26); @@ -2989,6 +3016,14 @@ static struct hda_verb cxt5066_init_verbs_portd_lo[] = { { } /* end */ }; + +static struct hda_verb cxt5066_init_verbs_hp_laptop[] = { + {0x14, AC_VERB_SET_CONNECT_SEL, 0x0}, + {0x19, AC_VERB_SET_UNSOLICITED_ENABLE, AC_USRSP_EN | CONEXANT_HP_EVENT}, + {0x1b, AC_VERB_SET_UNSOLICITED_ENABLE, AC_USRSP_EN | CONEXANT_MIC_EVENT}, + { } /* end */ +}; + /* initialize jack-sensing, too */ static int cxt5066_init(struct hda_codec *codec) { @@ -3004,6 +3039,8 @@ static int cxt5066_init(struct hda_codec *codec) cxt5066_ideapad_automic(codec); else if (spec->thinkpad) cxt5066_thinkpad_automic(codec); + else if (spec->hp_laptop) + cxt5066_hp_laptop_automic(codec); } cxt5066_set_mic_boost(codec); return 0; @@ -3031,6 +3068,7 @@ enum { CXT5066_DELL_VOSTO, /* Dell Vostro 1015i */ CXT5066_IDEAPAD, /* Lenovo IdeaPad U150 */ CXT5066_THINKPAD, /* Lenovo ThinkPad T410s, others? */ + CXT5066_HP_LAPTOP, /* HP Laptop */ CXT5066_MODELS }; @@ -3041,6 +3079,7 @@ static const char *cxt5066_models[CXT5066_MODELS] = { [CXT5066_DELL_VOSTO] = "dell-vostro", [CXT5066_IDEAPAD] = "ideapad", [CXT5066_THINKPAD] = "thinkpad", + [CXT5066_HP_LAPTOP] = "hp-laptop", }; static struct snd_pci_quirk cxt5066_cfg_tbl[] = { @@ -3052,6 +3091,7 @@ static struct snd_pci_quirk cxt5066_cfg_tbl[] = { SND_PCI_QUIRK(0x1028, 0x02d8, "Dell Vostro", CXT5066_DELL_VOSTO), SND_PCI_QUIRK(0x1028, 0x0402, "Dell Vostro", CXT5066_DELL_VOSTO), SND_PCI_QUIRK(0x1028, 0x0408, "Dell Inspiron One 19T", CXT5066_IDEAPAD), + SND_PCI_QUIRK(0x103c, 0x360b, "HP G60", CXT5066_HP_LAPTOP), SND_PCI_QUIRK(0x1179, 0xff50, "Toshiba Satellite P500-PSPGSC-01800T", CXT5066_OLPC_XO_1_5), SND_PCI_QUIRK(0x1179, 0xffe0, "Toshiba Satellite Pro T130-15F", CXT5066_OLPC_XO_1_5), SND_PCI_QUIRK(0x17aa, 0x21b2, "Thinkpad X100e", CXT5066_IDEAPAD), @@ -3116,6 +3156,23 @@ static int patch_cxt5066(struct hda_codec *codec) spec->num_init_verbs++; spec->dell_automute = 1; break; + case CXT5066_HP_LAPTOP: + codec->patch_ops.init = cxt5066_init; + codec->patch_ops.unsol_event = cxt5066_hp_laptop_event; + spec->init_verbs[spec->num_init_verbs] = + cxt5066_init_verbs_hp_laptop; + spec->num_init_verbs++; + spec->hp_laptop = 1; + spec->mixers[spec->num_mixers++] = cxt5066_mixer_master; + spec->mixers[spec->num_mixers++] = cxt5066_mixers; + /* no S/PDIF out */ + spec->multiout.dig_out_nid = 0; + /* input source automatically selected */ + spec->input_mux = NULL; + spec->port_d_mode = 0; + spec->mic_boost = 3; /* default 30dB gain */ + break; + case CXT5066_OLPC_XO_1_5: codec->patch_ops.init = cxt5066_olpc_init; codec->patch_ops.unsol_event = cxt5066_olpc_unsol_event; -- cgit v0.10.2 From 026b5ca3b65f878019a8eb0c7a702cd5c20a4104 Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Wed, 1 Sep 2010 14:33:29 +0100 Subject: ARM: 6344/1: Mark CPU_32v6K as depended on CPU_V7 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit CPU_32v6K is selected by CPU_V7 but it only depends on CPU_V6. Signed-off-by: Catalin Marinas Acked-by: Uwe Kleine-König Signed-off-by: Russell King diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig index 33c3f57..a0a2928 100644 --- a/arch/arm/mm/Kconfig +++ b/arch/arm/mm/Kconfig @@ -398,7 +398,7 @@ config CPU_V6 # ARMv6k config CPU_32v6K bool "Support ARM V6K processor extensions" if !SMP - depends on CPU_V6 + depends on CPU_V6 || CPU_V7 default y if SMP && !(ARCH_MX3 || ARCH_OMAP2) help Say Y here if your ARMv6 processor supports the 'K' extension. -- cgit v0.10.2 From 23963e54ce187ca6e907c83176c15508b0f6e60d Mon Sep 17 00:00:00 2001 From: Arkadiusz Mi?kiewicz Date: Thu, 26 Aug 2010 10:19:43 +0000 Subject: xfs: Disallow 32bit project quota id Currently on-disk structure is able to keep only 16bit project quota id, so disallow 32bit ones. This fixes a problem where parts of kernel structures holding project quota id are 32bit while parts (on-disk) are 16bit variables which causes project quota member files to be inaccessible for some operations (like mv/rm). Signed-off-by: Arkadiusz Mi?kiewicz Reviewed-by: Christoph Hellwig Signed-off-by: Alex Elder diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c index 237f5ff..4fec427 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl.c +++ b/fs/xfs/linux-2.6/xfs_ioctl.c @@ -907,6 +907,13 @@ xfs_ioctl_setattr( return XFS_ERROR(EIO); /* + * Disallow 32bit project ids because on-disk structure + * is 16bit only. + */ + if ((mask & FSX_PROJID) && (fa->fsx_projid > (__uint16_t)-1)) + return XFS_ERROR(EINVAL); + + /* * If disk quotas is on, we make sure that the dquots do exist on disk, * before we start any other transactions. Trying to do this later * is messy. We don't care to take a readlock to look at the ids -- cgit v0.10.2 From 65b4711ff513767341aa1915c822de6ec0de65cb Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Thu, 2 Sep 2010 09:32:08 +0100 Subject: ARM: 6352/1: perf: fix event validation The validate_event function in the ARM perf events backend has the following problems: 1.) Events that are disabled count towards the cost. 2.) Events associated with other PMUs [for example, software events or breakpoints] do not count towards the cost, but do fail validation, causing the group to fail. This patch changes validate_event so that it ignores events in the PERF_EVENT_STATE_OFF state or that are scheduled for other PMUs. Reported-by: Pawel Moll Acked-by: Jamie Iles Signed-off-by: Will Deacon Signed-off-by: Russell King diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index e213925..ecbb028 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -319,8 +319,8 @@ validate_event(struct cpu_hw_events *cpuc, { struct hw_perf_event fake_event = event->hw; - if (event->pmu && event->pmu != &pmu) - return 0; + if (event->pmu != &pmu || event->state <= PERF_EVENT_STATE_OFF) + return 1; return armpmu->get_event_idx(cpuc, &fake_event) >= 0; } -- cgit v0.10.2 From 5e4e7573e1ec286120109e73bf54cff465488725 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 2 Sep 2010 09:39:09 -0700 Subject: Revert "sky2: don't do GRO on second port" This reverts commit de6be6c1f77798c4da38301693d33aff1cd76e84. After some discussion with Jarek Poplawski and Eric Dumazet, we've decided that this change is incorrect. Signed-off-by: David S. Miller diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index a6ff113..194e5cf 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -2520,27 +2520,24 @@ static inline void sky2_tx_done(struct net_device *dev, u16 last) } } -static inline void sky2_skb_rx(struct napi_struct *napi, - const struct sky2_port *sky2, +static inline void sky2_skb_rx(const struct sky2_port *sky2, u32 status, struct sk_buff *skb) { #ifdef SKY2_VLAN_TAG_USED + u16 vlan_tag = be16_to_cpu(sky2->rx_tag); if (sky2->vlgrp && (status & GMR_FS_VLAN)) { - u16 vlan_tag = be16_to_cpu(sky2->rx_tag); - - if (skb->ip_summed == CHECKSUM_NONE || - sky2->netdev != napi->dev) + if (skb->ip_summed == CHECKSUM_NONE) vlan_hwaccel_receive_skb(skb, sky2->vlgrp, vlan_tag); else - vlan_gro_receive(napi, sky2->vlgrp, vlan_tag, skb); + vlan_gro_receive(&sky2->hw->napi, sky2->vlgrp, + vlan_tag, skb); return; } #endif - if (skb->ip_summed == CHECKSUM_NONE || - sky2->netdev != napi->dev) + if (skb->ip_summed == CHECKSUM_NONE) netif_receive_skb(skb); else - napi_gro_receive(napi, skb); + napi_gro_receive(&sky2->hw->napi, skb); } static inline void sky2_rx_done(struct sky2_hw *hw, unsigned port, @@ -2641,7 +2638,7 @@ static int sky2_status_intr(struct sky2_hw *hw, int to_do, u16 idx) skb->protocol = eth_type_trans(skb, dev); - sky2_skb_rx(&hw->napi, sky2, status, skb); + sky2_skb_rx(sky2, status, skb); /* Stop after net poll weight */ if (++work_done >= to_do) -- cgit v0.10.2 From 7bcbf81a2296a8f71342445560dcbe16100b567c Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Wed, 1 Sep 2010 23:07:10 +0000 Subject: ipvs: avoid oops for passive FTP Fix Passive FTP problem in ip_vs_ftp: - Do not oops in nf_nat_set_seq_adjust (adjust_tcp_sequence) when iptable_nat module is not loaded Signed-off-by: Julian Anastasov Signed-off-by: Simon Horman Signed-off-by: David S. Miller diff --git a/net/netfilter/ipvs/ip_vs_ftp.c b/net/netfilter/ipvs/ip_vs_ftp.c index f228a17..33b329b 100644 --- a/net/netfilter/ipvs/ip_vs_ftp.c +++ b/net/netfilter/ipvs/ip_vs_ftp.c @@ -45,6 +45,7 @@ #include #include #include +#include #include #include #include @@ -359,7 +360,7 @@ static int ip_vs_ftp_out(struct ip_vs_app *app, struct ip_vs_conn *cp, buf_len = strlen(buf); ct = nf_ct_get(skb, &ctinfo); - if (ct && !nf_ct_is_untracked(ct)) { + if (ct && !nf_ct_is_untracked(ct) && nfct_nat(ct)) { /* If mangling fails this function will return 0 * which will cause the packet to be dropped. * Mangling can only fail under memory pressure, -- cgit v0.10.2 From 8f34a430ac16d5fbd9d6b383184d35e152f5a963 Mon Sep 17 00:00:00 2001 From: "J. Bruce Fields" Date: Thu, 2 Sep 2010 15:23:16 -0400 Subject: nfsd4: mask out non-access bits in nfs4_access_to_omode This fixes an unnecessary BUG(). Signed-off-by: J. Bruce Fields diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 3dfef06..cf0d2ff 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -440,7 +440,7 @@ test_share(struct nfs4_stateid *stp, struct nfsd4_open *open) { static int nfs4_access_to_omode(u32 access) { - switch (access) { + switch (access & NFS4_SHARE_ACCESS_BOTH) { case NFS4_SHARE_ACCESS_READ: return O_RDONLY; case NFS4_SHARE_ACCESS_WRITE: -- cgit v0.10.2 From d842a93c4b3a20d31f4fb97357e0964210d6066f Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Thu, 12 Aug 2010 14:31:05 +0200 Subject: [SCSI] fix bio.bi_rw handling Return of the bi_rw tests is no longer bool after commit 74450be1. So testing against constants doesn't make sense anymore. Fix this bug in osd_req_read by removing "== 1" in test. This is not a problem now, where REQ_WRITE is 1, but this can change in the future and we don't want to rely on that. Signed-off-by: Jiri Slaby Signed-off-by: James Bottomley diff --git a/drivers/scsi/osd/osd_initiator.c b/drivers/scsi/osd/osd_initiator.c index fda4de3..e88bbdd 100644 --- a/drivers/scsi/osd/osd_initiator.c +++ b/drivers/scsi/osd/osd_initiator.c @@ -865,7 +865,7 @@ void osd_req_read(struct osd_request *or, { _osd_req_encode_common(or, OSD_ACT_READ, obj, offset, len); WARN_ON(or->in.bio || or->in.total_bytes); - WARN_ON(1 == (bio->bi_rw & REQ_WRITE)); + WARN_ON(bio->bi_rw & REQ_WRITE); or->in.bio = bio; or->in.total_bytes = len; } -- cgit v0.10.2 From b15d05b0d358cedf9c4d420a60d2ee2d0f530788 Mon Sep 17 00:00:00 2001 From: Jayamohan Kallickal Date: Thu, 12 Aug 2010 23:36:06 +0530 Subject: [SCSI] be2iscsi: Fix for Login failure The current code in tree has problems with Login. This patch fixes the Login Failure . Signed-off-by: Jayamohan Kallickal [mnc: Can't believe I missed that.] Reviewed-by: Mike Christie Signed-off-by: James Bottomley diff --git a/drivers/scsi/be2iscsi/be_iscsi.c b/drivers/scsi/be2iscsi/be_iscsi.c index 7d4d227..7f11f3e 100644 --- a/drivers/scsi/be2iscsi/be_iscsi.c +++ b/drivers/scsi/be2iscsi/be_iscsi.c @@ -300,8 +300,7 @@ int beiscsi_get_host_param(struct Scsi_Host *shost, enum iscsi_host_param param, char *buf) { struct beiscsi_hba *phba = (struct beiscsi_hba *)iscsi_host_priv(shost); - int len = 0; - int status; + int status = 0; SE_DEBUG(DBG_LVL_8, "In beiscsi_get_host_param, param= %d\n", param); switch (param) { @@ -315,7 +314,7 @@ int beiscsi_get_host_param(struct Scsi_Host *shost, default: return iscsi_host_get_param(shost, param, buf); } - return len; + return status; } int beiscsi_get_macaddr(char *buf, struct beiscsi_hba *phba) diff --git a/drivers/scsi/be2iscsi/be_mgmt.c b/drivers/scsi/be2iscsi/be_mgmt.c index 26350e4..877324f 100644 --- a/drivers/scsi/be2iscsi/be_mgmt.c +++ b/drivers/scsi/be2iscsi/be_mgmt.c @@ -368,7 +368,7 @@ int mgmt_open_connection(struct beiscsi_hba *phba, memset(req, 0, sizeof(*req)); wrb->tag0 |= tag; - be_wrb_hdr_prepare(wrb, sizeof(*req), true, 1); + be_wrb_hdr_prepare(wrb, sizeof(*req), false, 1); be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_ISCSI, OPCODE_COMMON_ISCSI_TCP_CONNECT_AND_OFFLOAD, sizeof(*req)); -- cgit v0.10.2 From 36ed2176fedaa180b8ea3cdacf68c958e0090a3c Mon Sep 17 00:00:00 2001 From: "Stephen M. Cameron" Date: Wed, 25 Aug 2010 10:44:14 -0500 Subject: [SCSI] hpsa: disable doorbell reset on reset_devices The doorbell reset initially appears to work correctly, the controller resets, comes up, some i/o can even be done, but on at least some Smart Arrays in some servers, it eventually causes a subsequent controller lockup due to some kind of PCIe error, and kdump can end up leaving the root filesystem in an unbootable state. For this reason, until the problem is fixed, or at least isolated to certain hardware enough to be avoided, the doorbell reset should not be used at all. Signed-off-by: Stephen M. Cameron Signed-off-by: James Bottomley diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c index 4f5551b..c5d0606 100644 --- a/drivers/scsi/hpsa.c +++ b/drivers/scsi/hpsa.c @@ -3231,6 +3231,12 @@ static __devinit int hpsa_kdump_hard_reset_controller(struct pci_dev *pdev) misc_fw_support = readl(&cfgtable->misc_fw_support); use_doorbell = misc_fw_support & MISC_FW_DOORBELL_RESET; + /* The doorbell reset seems to cause lockups on some Smart + * Arrays (e.g. P410, P410i, maybe others). Until this is + * fixed or at least isolated, avoid the doorbell reset. + */ + use_doorbell = 0; + rc = hpsa_controller_hard_reset(pdev, vaddr, use_doorbell); if (rc) goto unmap_cfgtable; -- cgit v0.10.2 From 6f131ce1dfa9b283ddc212df42b015d152c670a5 Mon Sep 17 00:00:00 2001 From: Jean Sacren Date: Wed, 25 Aug 2010 17:58:09 -0600 Subject: [SCSI] Fix warning: zero-length gnu_printf format string warning: zero-length gnu_printf format string Fix the above warning by inserting a space into the literal string. Signed-off-by: Jean Sacren Signed-off-by: James Bottomley diff --git a/drivers/scsi/constants.c b/drivers/scsi/constants.c index cd05e04..d0c8234 100644 --- a/drivers/scsi/constants.c +++ b/drivers/scsi/constants.c @@ -1404,13 +1404,13 @@ void scsi_print_sense(char *name, struct scsi_cmnd *cmd) { struct scsi_sense_hdr sshdr; - scmd_printk(KERN_INFO, cmd, ""); + scmd_printk(KERN_INFO, cmd, " "); scsi_decode_sense_buffer(cmd->sense_buffer, SCSI_SENSE_BUFFERSIZE, &sshdr); scsi_show_sense_hdr(&sshdr); scsi_decode_sense_extras(cmd->sense_buffer, SCSI_SENSE_BUFFERSIZE, &sshdr); - scmd_printk(KERN_INFO, cmd, ""); + scmd_printk(KERN_INFO, cmd, " "); scsi_show_extd_sense(sshdr.asc, sshdr.ascq); } EXPORT_SYMBOL(scsi_print_sense); @@ -1453,7 +1453,7 @@ EXPORT_SYMBOL(scsi_show_result); void scsi_print_result(struct scsi_cmnd *cmd) { - scmd_printk(KERN_INFO, cmd, ""); + scmd_printk(KERN_INFO, cmd, " "); scsi_show_result(cmd->result); } EXPORT_SYMBOL(scsi_print_result); -- cgit v0.10.2 From 0b5d404e349c0236b11466c0a4785520c0be6982 Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Thu, 2 Sep 2010 13:22:11 -0700 Subject: pkt_sched: Fix lockdep warning on est_tree_lock in gen_estimator This patch fixes a lockdep warning: [ 516.287584] ========================================================= [ 516.288386] [ INFO: possible irq lock inversion dependency detected ] [ 516.288386] 2.6.35b #7 [ 516.288386] --------------------------------------------------------- [ 516.288386] swapper/0 just changed the state of lock: [ 516.288386] (&qdisc_tx_lock){+.-...}, at: [] est_timer+0x62/0x1b4 [ 516.288386] but this lock took another, SOFTIRQ-unsafe lock in the past: [ 516.288386] (est_tree_lock){+.+...} [ 516.288386] [ 516.288386] and interrupts could create inverse lock ordering between them. ... So, est_tree_lock needs BH protection because it's taken by qdisc_tx_lock, which is used both in BH and process contexts. (Full warning with this patch at netdev, 02 Sep 2010.) Fixes commit: ae638c47dc040b8def16d05dc6acdd527628f231 ("pkt_sched: gen_estimator: add a new lock") Signed-off-by: Jarek Poplawski Signed-off-by: David S. Miller diff --git a/net/core/gen_estimator.c b/net/core/gen_estimator.c index 9fbe7f7..6743146 100644 --- a/net/core/gen_estimator.c +++ b/net/core/gen_estimator.c @@ -232,7 +232,7 @@ int gen_new_estimator(struct gnet_stats_basic_packed *bstats, est->last_packets = bstats->packets; est->avpps = rate_est->pps<<10; - spin_lock(&est_tree_lock); + spin_lock_bh(&est_tree_lock); if (!elist[idx].timer.function) { INIT_LIST_HEAD(&elist[idx].list); setup_timer(&elist[idx].timer, est_timer, idx); @@ -243,7 +243,7 @@ int gen_new_estimator(struct gnet_stats_basic_packed *bstats, list_add_rcu(&est->list, &elist[idx].list); gen_add_node(est); - spin_unlock(&est_tree_lock); + spin_unlock_bh(&est_tree_lock); return 0; } @@ -270,7 +270,7 @@ void gen_kill_estimator(struct gnet_stats_basic_packed *bstats, { struct gen_estimator *e; - spin_lock(&est_tree_lock); + spin_lock_bh(&est_tree_lock); while ((e = gen_find_node(bstats, rate_est))) { rb_erase(&e->node, &est_root); @@ -281,7 +281,7 @@ void gen_kill_estimator(struct gnet_stats_basic_packed *bstats, list_del_rcu(&e->list); call_rcu(&e->e_rcu, __gen_kill_estimator); } - spin_unlock(&est_tree_lock); + spin_unlock_bh(&est_tree_lock); } EXPORT_SYMBOL(gen_kill_estimator); @@ -320,9 +320,9 @@ bool gen_estimator_active(const struct gnet_stats_basic_packed *bstats, ASSERT_RTNL(); - spin_lock(&est_tree_lock); + spin_lock_bh(&est_tree_lock); res = gen_find_node(bstats, rate_est) != NULL; - spin_unlock(&est_tree_lock); + spin_unlock_bh(&est_tree_lock); return res; } -- cgit v0.10.2 From 2e4c332913b5d39fef686b3964098f0d8fd97ead Mon Sep 17 00:00:00 2001 From: David Miller Date: Tue, 31 Aug 2010 13:35:31 -0700 Subject: [SCSI] sd, sym53c8xx: Remove warnings after vsprintf %pV introducation. GCC warns about empty printf format strings, and after the addition of %pV these existing such cases in the scsi driver layer were exposed enough for the compiler to start seeing them. Based almost entirely upon a patch by Joe Perches. [jejb: fix up sym53c8xx msg] Signed-off-by: David S. Miller Signed-off-by: James Bottomley diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index 2714bec..cd71f46 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -2625,15 +2625,15 @@ module_exit(exit_sd); static void sd_print_sense_hdr(struct scsi_disk *sdkp, struct scsi_sense_hdr *sshdr) { - sd_printk(KERN_INFO, sdkp, ""); + sd_printk(KERN_INFO, sdkp, " "); scsi_show_sense_hdr(sshdr); - sd_printk(KERN_INFO, sdkp, ""); + sd_printk(KERN_INFO, sdkp, " "); scsi_show_extd_sense(sshdr->asc, sshdr->ascq); } static void sd_print_result(struct scsi_disk *sdkp, int result) { - sd_printk(KERN_INFO, sdkp, ""); + sd_printk(KERN_INFO, sdkp, " "); scsi_show_result(result); } diff --git a/drivers/scsi/sym53c8xx_2/sym_hipd.c b/drivers/scsi/sym53c8xx_2/sym_hipd.c index a7bc8b7..2c3e89d 100644 --- a/drivers/scsi/sym53c8xx_2/sym_hipd.c +++ b/drivers/scsi/sym53c8xx_2/sym_hipd.c @@ -72,10 +72,7 @@ static void sym_printl_hex(u_char *p, int n) static void sym_print_msg(struct sym_ccb *cp, char *label, u_char *msg) { - if (label) - sym_print_addr(cp->cmd, "%s: ", label); - else - sym_print_addr(cp->cmd, ""); + sym_print_addr(cp->cmd, "%s: ", label); spi_print_msg(msg); printf("\n"); @@ -4558,7 +4555,8 @@ static void sym_int_sir(struct sym_hcb *np) switch (np->msgin [2]) { case M_X_MODIFY_DP: if (DEBUG_FLAGS & DEBUG_POINTER) - sym_print_msg(cp, NULL, np->msgin); + sym_print_msg(cp, "extended msg ", + np->msgin); tmp = (np->msgin[3]<<24) + (np->msgin[4]<<16) + (np->msgin[5]<<8) + (np->msgin[6]); sym_modify_dp(np, tp, cp, tmp); @@ -4585,7 +4583,7 @@ static void sym_int_sir(struct sym_hcb *np) */ case M_IGN_RESIDUE: if (DEBUG_FLAGS & DEBUG_POINTER) - sym_print_msg(cp, NULL, np->msgin); + sym_print_msg(cp, "1 or 2 byte ", np->msgin); if (cp->host_flags & HF_SENSE) OUTL_DSP(np, SCRIPTA_BA(np, clrack)); else -- cgit v0.10.2 From dc4e96ce2dceb649224ee84f83592aac8c54c9b7 Mon Sep 17 00:00:00 2001 From: Steve Wise Date: Sat, 28 Aug 2010 13:35:05 +0000 Subject: RDMA/cxgb3: Don't exceed the max HW CQ depth The max depth supported by T3 is 64K entries. This fixes a bug introduced in commit 9918b28d ("RDMA/cxgb3: Increase the max CQ depth") that causes stalls and possibly crashes in large MPI clusters. Signed-off-by: Steve Wise Cc: Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/cxgb3/cxio_hal.h b/drivers/infiniband/hw/cxgb3/cxio_hal.h index 8f0caf7..78fbe9f 100644 --- a/drivers/infiniband/hw/cxgb3/cxio_hal.h +++ b/drivers/infiniband/hw/cxgb3/cxio_hal.h @@ -53,7 +53,7 @@ #define T3_MAX_PBL_SIZE 256 #define T3_MAX_RQ_SIZE 1024 #define T3_MAX_QP_DEPTH (T3_MAX_RQ_SIZE-1) -#define T3_MAX_CQ_DEPTH 262144 +#define T3_MAX_CQ_DEPTH 65536 #define T3_MAX_NUM_STAG (1<<15) #define T3_MAX_MR_SIZE 0x100000000ULL #define T3_PAGESIZE_MASK 0xffff000 /* 4KB-128MB */ -- cgit v0.10.2 From 3ba6462355c1c69dde58739a871d13bbb993e2e3 Mon Sep 17 00:00:00 2001 From: Francisco Jerez Date: Sat, 28 Aug 2010 17:56:33 +0200 Subject: drm/nouveau: Take fence spinlock before reading the last sequence. It fixes a race between the TTM delayed work queue and the GEM IOCTLs (fdo bug 29583) uncovered by the BKL removal. Signed-off-by: Francisco Jerez Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.c b/drivers/gpu/drm/nouveau/nouveau_fence.c index 6b208ff..87ac21e 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fence.c +++ b/drivers/gpu/drm/nouveau/nouveau_fence.c @@ -64,16 +64,17 @@ nouveau_fence_update(struct nouveau_channel *chan) struct nouveau_fence *fence; uint32_t sequence; + spin_lock(&chan->fence.lock); + if (USE_REFCNT) sequence = nvchan_rd32(chan, 0x48); else sequence = atomic_read(&chan->fence.last_sequence_irq); if (chan->fence.sequence_ack == sequence) - return; + goto out; chan->fence.sequence_ack = sequence; - spin_lock(&chan->fence.lock); list_for_each_safe(entry, tmp, &chan->fence.pending) { fence = list_entry(entry, struct nouveau_fence, entry); @@ -85,6 +86,7 @@ nouveau_fence_update(struct nouveau_channel *chan) if (sequence == chan->fence.sequence_ack) break; } +out: spin_unlock(&chan->fence.lock); } -- cgit v0.10.2 From 374c3af880ef260f36dfc968d9725494666dff31 Mon Sep 17 00:00:00 2001 From: Francisco Jerez Date: Sun, 29 Aug 2010 12:21:16 +0200 Subject: drm/nouveau: Don't take struct_mutex around the pushbuf IOCTL. We don't need it and it can lead to lock order inversions with respect to drm_global_mutex, potentially causing dead locks. Signed-off-by: Francisco Jerez Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c index 93711df..10cace9 100644 --- a/drivers/gpu/drm/nouveau/nouveau_gem.c +++ b/drivers/gpu/drm/nouveau/nouveau_gem.c @@ -245,7 +245,7 @@ validate_fini_list(struct list_head *list, struct nouveau_fence *fence) list_del(&nvbo->entry); nvbo->reserved_by = NULL; ttm_bo_unreserve(&nvbo->bo); - drm_gem_object_unreference(nvbo->gem); + drm_gem_object_unreference_unlocked(nvbo->gem); } } @@ -300,7 +300,7 @@ retry: validate_fini(op, NULL); if (ret == -EAGAIN) ret = ttm_bo_wait_unreserved(&nvbo->bo, false); - drm_gem_object_unreference(gem); + drm_gem_object_unreference_unlocked(gem); if (ret) { NV_ERROR(dev, "fail reserve\n"); return ret; @@ -616,8 +616,6 @@ nouveau_gem_ioctl_pushbuf(struct drm_device *dev, void *data, return PTR_ERR(bo); } - mutex_lock(&dev->struct_mutex); - /* Mark push buffers as being used on PFIFO, the validation code * will then make sure that if the pushbuf bo moves, that they * happen on the kernel channel, which will in turn cause a sync @@ -731,7 +729,6 @@ nouveau_gem_ioctl_pushbuf(struct drm_device *dev, void *data, out: validate_fini(&op, fence); nouveau_fence_unref((void**)&fence); - mutex_unlock(&dev->struct_mutex); kfree(bo); kfree(push); -- cgit v0.10.2 From 615661f3948a066fd22a36fe8ea0c528b75ee373 Mon Sep 17 00:00:00 2001 From: Marcin Slusarz Date: Sun, 22 Aug 2010 20:54:08 +0200 Subject: drm/nv50: initialize ramht_refs list for faked 0 channel We need it for PFIFO_INTR_CACHE_ERROR interrupt handling, because nouveau_fifo_swmthd looks for matching gpuobj in ramht_refs list. It fixes kernel panic in nouveau_gpuobj_ref_find. Signed-off-by: Marcin Slusarz Signed-off-by: Ben Skeggs diff --git a/drivers/gpu/drm/nouveau/nv50_instmem.c b/drivers/gpu/drm/nouveau/nv50_instmem.c index c95bf9b..91ef93c 100644 --- a/drivers/gpu/drm/nouveau/nv50_instmem.c +++ b/drivers/gpu/drm/nouveau/nv50_instmem.c @@ -139,6 +139,8 @@ nv50_instmem_init(struct drm_device *dev) chan->file_priv = (struct drm_file *)-2; dev_priv->fifos[0] = dev_priv->fifos[127] = chan; + INIT_LIST_HEAD(&chan->ramht_refs); + /* Channel's PRAMIN object + heap */ ret = nouveau_gpuobj_new_fake(dev, 0, c_offset, c_size, 0, NULL, &chan->ramin); -- cgit v0.10.2 From 72656c46f50b8dfe50e15793692982e636e3df20 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Fri, 3 Sep 2010 12:19:33 +1000 Subject: xfs: prevent 32bit overflow in space reservation If we attempt to preallocate more than 2^32 blocks of space in a single syscall, the transaction block reservation will overflow leading to a hangs in the superblock block accounting code. This is trivially reproduced with xfs_io. Fix the problem by capping the allocation reservation to the maximum number of blocks a single xfs_bmapi() call can allocate (2^21 blocks). Signed-off-by: Dave Chinner Reviewed-by: Christoph Hellwig diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 66d585c..4c7c7bf 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -2299,15 +2299,22 @@ xfs_alloc_file_space( e = allocatesize_fsb; } + /* + * The transaction reservation is limited to a 32-bit block + * count, hence we need to limit the number of blocks we are + * trying to reserve to avoid an overflow. We can't allocate + * more than @nimaps extents, and an extent is limited on disk + * to MAXEXTLEN (21 bits), so use that to enforce the limit. + */ + resblks = min_t(xfs_fileoff_t, (e - s), (MAXEXTLEN * nimaps)); if (unlikely(rt)) { - resrtextents = qblocks = (uint)(e - s); + resrtextents = qblocks = resblks; resrtextents /= mp->m_sb.sb_rextsize; resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0); quota_flag = XFS_QMOPT_RES_RTBLKS; } else { resrtextents = 0; - resblks = qblocks = \ - XFS_DIOSTRAT_SPACE_RES(mp, (uint)(e - s)); + resblks = qblocks = XFS_DIOSTRAT_SPACE_RES(mp, resblks); quota_flag = XFS_QMOPT_RES_REGBLKS; } -- cgit v0.10.2 From 2e556b5b320838fde98480a1f6cf220a5af200fc Mon Sep 17 00:00:00 2001 From: Don Zickus Date: Thu, 2 Sep 2010 15:07:47 -0400 Subject: perf, x86: Fix accidentally ack'ing a second event on intel perf counter During testing of a patch to stop having the perf subsytem swallow nmis, it was uncovered that Nehalem boxes were randomly getting unknown nmis when using the perf tool. Moving the ack'ing of the PMI closer to when we get the status allows the hardware to properly re-set the PMU bit signaling another PMI was triggered during the processing of the first PMI. This allows the new logic for dealing with the shortcomings of multiple PMIs to handle the extra NMI by 'eat'ing it later. Now one can wonder why are we getting a second PMI when we disable all the PMUs in the begining of the NMI handler to prevent such a case, for that I do not know. But I know the fix below helps deal with this quirk. Tested on multiple Nehalems where the problem was occuring. With the patch, the code now loops a second time to handle the second PMI (whereas before it was not). Signed-off-by: Don Zickus Cc: peterz@infradead.org Cc: robert.richter@amd.com Cc: gorcunov@gmail.com Cc: fweisbec@gmail.com Cc: ying.huang@intel.com Cc: ming.m.lin@intel.com Cc: eranian@google.com LKML-Reference: <1283454469-1909-2-git-send-email-dzickus@redhat.com> Signed-off-by: Ingo Molnar diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index d8d86d0..1297bf1 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -712,7 +712,7 @@ static int intel_pmu_handle_irq(struct pt_regs *regs) struct perf_sample_data data; struct cpu_hw_events *cpuc; int bit, loops; - u64 ack, status; + u64 status; perf_sample_data_init(&data, 0); @@ -728,6 +728,7 @@ static int intel_pmu_handle_irq(struct pt_regs *regs) loops = 0; again: + intel_pmu_ack_status(status); if (++loops > 100) { WARN_ONCE(1, "perfevents: irq loop stuck!\n"); perf_event_print_debug(); @@ -736,7 +737,6 @@ again: } inc_irq_stat(apic_perf_irqs); - ack = status; intel_pmu_lbr_read(); @@ -761,8 +761,6 @@ again: x86_pmu_stop(event); } - intel_pmu_ack_status(ack); - /* * Repeat if there is more work to be done: */ -- cgit v0.10.2 From de725dec9de7a7541996176d59cf8542365b8b0e Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 2 Sep 2010 15:07:49 -0400 Subject: perf, x86: Fix handle_irq return values Now that we rely on the number of handled overflows, ensure all handle_irq implementations actually return the right number. Signed-off-by: Peter Zijlstra Signed-off-by: Don Zickus Cc: peterz@infradead.org Cc: robert.richter@amd.com Cc: gorcunov@gmail.com Cc: fweisbec@gmail.com Cc: ying.huang@intel.com Cc: ming.m.lin@intel.com Cc: eranian@google.com LKML-Reference: <1283454469-1909-4-git-send-email-dzickus@redhat.com> Signed-off-by: Ingo Molnar diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index 1297bf1..ee05c90 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -713,6 +713,7 @@ static int intel_pmu_handle_irq(struct pt_regs *regs) struct cpu_hw_events *cpuc; int bit, loops; u64 status; + int handled = 0; perf_sample_data_init(&data, 0); @@ -743,12 +744,16 @@ again: /* * PEBS overflow sets bit 62 in the global status register */ - if (__test_and_clear_bit(62, (unsigned long *)&status)) + if (__test_and_clear_bit(62, (unsigned long *)&status)) { + handled++; x86_pmu.drain_pebs(regs); + } for_each_set_bit(bit, (unsigned long *)&status, X86_PMC_IDX_MAX) { struct perf_event *event = cpuc->events[bit]; + handled++; + if (!test_bit(bit, cpuc->active_mask)) continue; @@ -770,7 +775,7 @@ again: done: intel_pmu_enable_all(0); - return 1; + return handled; } static struct event_constraint * diff --git a/arch/x86/kernel/cpu/perf_event_p4.c b/arch/x86/kernel/cpu/perf_event_p4.c index 7e578e9..b560db3 100644 --- a/arch/x86/kernel/cpu/perf_event_p4.c +++ b/arch/x86/kernel/cpu/perf_event_p4.c @@ -692,7 +692,7 @@ static int p4_pmu_handle_irq(struct pt_regs *regs) inc_irq_stat(apic_perf_irqs); } - return handled > 0; + return handled; } /* -- cgit v0.10.2 From 4177c42a6301a34c20038ec2771a33dcc30bb338 Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Thu, 2 Sep 2010 15:07:48 -0400 Subject: perf, x86: Try to handle unknown nmis with an enabled PMU When the PMU is enabled it is valid to have unhandled nmis, two events could trigger 'simultaneously' raising two back-to-back NMIs. If the first NMI handles both, the latter will be empty and daze the CPU. The solution to avoid an 'unknown nmi' massage in this case was simply to stop the nmi handler chain when the PMU is enabled by stating the nmi was handled. This has the drawback that a) we can not detect unknown nmis anymore, and b) subsequent nmi handlers are not called. This patch addresses this. Now, we check this unknown NMI if it could be a PMU back-to-back NMI. Otherwise we pass it and let the kernel handle the unknown nmi. This is a debug log: cpu #6, nmi #32333, skip_nmi #32330, handled = 1, time = 1934364430 cpu #6, nmi #32334, skip_nmi #32330, handled = 1, time = 1934704616 cpu #6, nmi #32335, skip_nmi #32336, handled = 2, time = 1936032320 cpu #6, nmi #32336, skip_nmi #32336, handled = 0, time = 1936034139 cpu #6, nmi #32337, skip_nmi #32336, handled = 1, time = 1936120100 cpu #6, nmi #32338, skip_nmi #32336, handled = 1, time = 1936404607 cpu #6, nmi #32339, skip_nmi #32336, handled = 1, time = 1937983416 cpu #6, nmi #32340, skip_nmi #32341, handled = 2, time = 1938201032 cpu #6, nmi #32341, skip_nmi #32341, handled = 0, time = 1938202830 cpu #6, nmi #32342, skip_nmi #32341, handled = 1, time = 1938443743 cpu #6, nmi #32343, skip_nmi #32341, handled = 1, time = 1939956552 cpu #6, nmi #32344, skip_nmi #32341, handled = 1, time = 1940073224 cpu #6, nmi #32345, skip_nmi #32341, handled = 1, time = 1940485677 cpu #6, nmi #32346, skip_nmi #32347, handled = 2, time = 1941947772 cpu #6, nmi #32347, skip_nmi #32347, handled = 1, time = 1941949818 cpu #6, nmi #32348, skip_nmi #32347, handled = 0, time = 1941951591 Uhhuh. NMI received for unknown reason 00 on CPU 6. Do you have a strange power saving mode enabled? Dazed and confused, but trying to continue Deltas: nmi #32334 340186 nmi #32335 1327704 nmi #32336 1819 <<<< back-to-back nmi [1] nmi #32337 85961 nmi #32338 284507 nmi #32339 1578809 nmi #32340 217616 nmi #32341 1798 <<<< back-to-back nmi [2] nmi #32342 240913 nmi #32343 1512809 nmi #32344 116672 nmi #32345 412453 nmi #32346 1462095 <<<< 1st nmi (standard) handling 2 counters nmi #32347 2046 <<<< 2nd nmi (back-to-back) handling one counter nmi #32348 1773 <<<< 3rd nmi (back-to-back) handling no counter! [3] For back-to-back nmi detection there are the following rules: The PMU nmi handler was handling more than one counter and no counter was handled in the subsequent nmi (see [1] and [2] above). There is another case if there are two subsequent back-to-back nmis [3]. The 2nd is detected as back-to-back because the first handled more than one counter. If the second handles one counter and the 3rd handles nothing, we drop the 3rd nmi because it could be a back-to-back nmi. Signed-off-by: Robert Richter Signed-off-by: Peter Zijlstra [ renamed nmi variable to pmu_nmi to avoid clash with .nmi in entry.S ] Signed-off-by: Don Zickus Cc: peterz@infradead.org Cc: gorcunov@gmail.com Cc: fweisbec@gmail.com Cc: ying.huang@intel.com Cc: ming.m.lin@intel.com Cc: eranian@google.com LKML-Reference: <1283454469-1909-3-git-send-email-dzickus@redhat.com> Signed-off-by: Ingo Molnar diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index f2da20f..3efdf28 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1154,7 +1154,7 @@ static int x86_pmu_handle_irq(struct pt_regs *regs) /* * event overflow */ - handled = 1; + handled++; data.period = event->hw.last_period; if (!x86_perf_event_set_period(event)) @@ -1200,12 +1200,20 @@ void perf_events_lapic_init(void) apic_write(APIC_LVTPC, APIC_DM_NMI); } +struct pmu_nmi_state { + unsigned int marked; + int handled; +}; + +static DEFINE_PER_CPU(struct pmu_nmi_state, pmu_nmi); + static int __kprobes perf_event_nmi_handler(struct notifier_block *self, unsigned long cmd, void *__args) { struct die_args *args = __args; - struct pt_regs *regs; + unsigned int this_nmi; + int handled; if (!atomic_read(&active_events)) return NOTIFY_DONE; @@ -1214,22 +1222,47 @@ perf_event_nmi_handler(struct notifier_block *self, case DIE_NMI: case DIE_NMI_IPI: break; - + case DIE_NMIUNKNOWN: + this_nmi = percpu_read(irq_stat.__nmi_count); + if (this_nmi != __get_cpu_var(pmu_nmi).marked) + /* let the kernel handle the unknown nmi */ + return NOTIFY_DONE; + /* + * This one is a PMU back-to-back nmi. Two events + * trigger 'simultaneously' raising two back-to-back + * NMIs. If the first NMI handles both, the latter + * will be empty and daze the CPU. So, we drop it to + * avoid false-positive 'unknown nmi' messages. + */ + return NOTIFY_STOP; default: return NOTIFY_DONE; } - regs = args->regs; - apic_write(APIC_LVTPC, APIC_DM_NMI); - /* - * Can't rely on the handled return value to say it was our NMI, two - * events could trigger 'simultaneously' raising two back-to-back NMIs. - * - * If the first NMI handles both, the latter will be empty and daze - * the CPU. - */ - x86_pmu.handle_irq(regs); + + handled = x86_pmu.handle_irq(args->regs); + if (!handled) + return NOTIFY_DONE; + + this_nmi = percpu_read(irq_stat.__nmi_count); + if ((handled > 1) || + /* the next nmi could be a back-to-back nmi */ + ((__get_cpu_var(pmu_nmi).marked == this_nmi) && + (__get_cpu_var(pmu_nmi).handled > 1))) { + /* + * We could have two subsequent back-to-back nmis: The + * first handles more than one counter, the 2nd + * handles only one counter and the 3rd handles no + * counter. + * + * This is the 2nd nmi because the previous was + * handling more than one counter. We will mark the + * next (3rd) and then drop it if unhandled. + */ + __get_cpu_var(pmu_nmi).marked = this_nmi + 1; + __get_cpu_var(pmu_nmi).handled = handled; + } return NOTIFY_STOP; } -- cgit v0.10.2 From ef5dc121d5a0bb1fa477c5395277259f07d318a3 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 2 Sep 2010 15:48:16 -0700 Subject: mutex: Fix annotations to include it in kernel-locking docbook Fix kernel-doc notation in linux/mutex.h and kernel/mutex.c, then add these 2 files to the kernel-locking docbook as the Mutex API reference chapter. Add one API function to mutex-design.txt and correct a typo in that file. Signed-off-by: Randy Dunlap Cc: Rusty Russell LKML-Reference: <20100902154816.6cc2f9ad.randy.dunlap@oracle.com> Signed-off-by: Ingo Molnar diff --git a/Documentation/DocBook/kernel-locking.tmpl b/Documentation/DocBook/kernel-locking.tmpl index 0b1a3f9..a0d479d 100644 --- a/Documentation/DocBook/kernel-locking.tmpl +++ b/Documentation/DocBook/kernel-locking.tmpl @@ -1961,6 +1961,12 @@ machines due to caching. + + Mutex API reference +!Iinclude/linux/mutex.h +!Ekernel/mutex.c + + Further reading diff --git a/Documentation/mutex-design.txt b/Documentation/mutex-design.txt index c91ccc0..38c10fd 100644 --- a/Documentation/mutex-design.txt +++ b/Documentation/mutex-design.txt @@ -9,7 +9,7 @@ firstly, there's nothing wrong with semaphores. But if the simpler mutex semantics are sufficient for your code, then there are a couple of advantages of mutexes: - - 'struct mutex' is smaller on most architectures: .e.g on x86, + - 'struct mutex' is smaller on most architectures: E.g. on x86, 'struct semaphore' is 20 bytes, 'struct mutex' is 16 bytes. A smaller structure size means less RAM footprint, and better CPU-cache utilization. @@ -136,3 +136,4 @@ the APIs of 'struct mutex' have been streamlined: void mutex_lock_nested(struct mutex *lock, unsigned int subclass); int mutex_lock_interruptible_nested(struct mutex *lock, unsigned int subclass); + int atomic_dec_and_mutex_lock(atomic_t *cnt, struct mutex *lock); diff --git a/include/linux/mutex.h b/include/linux/mutex.h index 878cab4..f363bc8 100644 --- a/include/linux/mutex.h +++ b/include/linux/mutex.h @@ -78,6 +78,14 @@ struct mutex_waiter { # include #else # define __DEBUG_MUTEX_INITIALIZER(lockname) +/** + * mutex_init - initialize the mutex + * @mutex: the mutex to be initialized + * + * Initialize the mutex to unlocked state. + * + * It is not allowed to initialize an already locked mutex. + */ # define mutex_init(mutex) \ do { \ static struct lock_class_key __key; \ diff --git a/kernel/mutex.c b/kernel/mutex.c index 4c0b7b3..200407c 100644 --- a/kernel/mutex.c +++ b/kernel/mutex.c @@ -36,15 +36,6 @@ # include #endif -/*** - * mutex_init - initialize the mutex - * @lock: the mutex to be initialized - * @key: the lock_class_key for the class; used by mutex lock debugging - * - * Initialize the mutex to unlocked state. - * - * It is not allowed to initialize an already locked mutex. - */ void __mutex_init(struct mutex *lock, const char *name, struct lock_class_key *key) { @@ -68,7 +59,7 @@ EXPORT_SYMBOL(__mutex_init); static __used noinline void __sched __mutex_lock_slowpath(atomic_t *lock_count); -/*** +/** * mutex_lock - acquire the mutex * @lock: the mutex to be acquired * @@ -105,7 +96,7 @@ EXPORT_SYMBOL(mutex_lock); static __used noinline void __sched __mutex_unlock_slowpath(atomic_t *lock_count); -/*** +/** * mutex_unlock - release the mutex * @lock: the mutex to be released * @@ -364,8 +355,8 @@ __mutex_lock_killable_slowpath(atomic_t *lock_count); static noinline int __sched __mutex_lock_interruptible_slowpath(atomic_t *lock_count); -/*** - * mutex_lock_interruptible - acquire the mutex, interruptable +/** + * mutex_lock_interruptible - acquire the mutex, interruptible * @lock: the mutex to be acquired * * Lock the mutex like mutex_lock(), and return 0 if the mutex has @@ -456,15 +447,15 @@ static inline int __mutex_trylock_slowpath(atomic_t *lock_count) return prev == 1; } -/*** - * mutex_trylock - try acquire the mutex, without waiting +/** + * mutex_trylock - try to acquire the mutex, without waiting * @lock: the mutex to be acquired * * Try to acquire the mutex atomically. Returns 1 if the mutex * has been acquired successfully, and 0 on contention. * * NOTE: this function follows the spin_trylock() convention, so - * it is negated to the down_trylock() return values! Be careful + * it is negated from the down_trylock() return values! Be careful * about this when converting semaphore users to mutexes. * * This function must not be used in interrupt context. The -- cgit v0.10.2 From 9af25465081480a75824fd7a16a37a5cfebeede9 Mon Sep 17 00:00:00 2001 From: Tao Ma Date: Mon, 30 Aug 2010 02:44:03 +0000 Subject: xfs: Make fiemap work with sparse files In xfs_vn_fiemap, we set bvm_count to fi_extent_max + 1 and want to return fi_extent_max extents, but actually it won't work for a sparse file. The reason is that in xfs_getbmap we will calculate holes and set it in 'out', while out is malloced by bmv_count(fi_extent_max+1) which didn't consider holes. So in the worst case, if 'out' vector looks like [hole, extent, hole, extent, hole, ... hole, extent, hole], we will only return half of fi_extent_max extents. This patch add a new parameter BMV_IF_NO_HOLES for bvm_iflags. So with this flags, we don't use our 'out' in xfs_getbmap for a hole. The solution is a bit ugly by just don't increasing index of 'out' vector. I felt that it is not easy to skip it at the very beginning since we have the complicated check and some function like xfs_getbmapx_fix_eof_hole to adjust 'out'. Cc: Dave Chinner Signed-off-by: Tao Ma Signed-off-by: Alex Elder diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c index 68be25d..b1fc2a6 100644 --- a/fs/xfs/linux-2.6/xfs_iops.c +++ b/fs/xfs/linux-2.6/xfs_iops.c @@ -664,7 +664,7 @@ xfs_vn_fiemap( fieinfo->fi_extents_max + 1; bm.bmv_count = min_t(__s32, bm.bmv_count, (PAGE_SIZE * 16 / sizeof(struct getbmapx))); - bm.bmv_iflags = BMV_IF_PREALLOC; + bm.bmv_iflags = BMV_IF_PREALLOC | BMV_IF_NO_HOLES; if (fieinfo->fi_flags & FIEMAP_FLAG_XATTR) bm.bmv_iflags |= BMV_IF_ATTRFORK; if (!(fieinfo->fi_flags & FIEMAP_FLAG_SYNC)) diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 23f14e5..f90dadd 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -5533,12 +5533,24 @@ xfs_getbmap( map[i].br_startblock)) goto out_free_map; - nexleft--; bmv->bmv_offset = out[cur_ext].bmv_offset + out[cur_ext].bmv_length; bmv->bmv_length = max_t(__int64_t, 0, bmvend - bmv->bmv_offset); + + /* + * In case we don't want to return the hole, + * don't increase cur_ext so that we can reuse + * it in the next loop. + */ + if ((iflags & BMV_IF_NO_HOLES) && + map[i].br_startblock == HOLESTARTBLOCK) { + memset(&out[cur_ext], 0, sizeof(out[cur_ext])); + continue; + } + + nexleft--; bmv->bmv_entries++; cur_ext++; } diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h index 7cf7220..87c2e9d 100644 --- a/fs/xfs/xfs_fs.h +++ b/fs/xfs/xfs_fs.h @@ -114,8 +114,10 @@ struct getbmapx { #define BMV_IF_NO_DMAPI_READ 0x2 /* Do not generate DMAPI read event */ #define BMV_IF_PREALLOC 0x4 /* rtn status BMV_OF_PREALLOC if req */ #define BMV_IF_DELALLOC 0x8 /* rtn status BMV_OF_DELALLOC if req */ +#define BMV_IF_NO_HOLES 0x10 /* Do not return holes */ #define BMV_IF_VALID \ - (BMV_IF_ATTRFORK|BMV_IF_NO_DMAPI_READ|BMV_IF_PREALLOC|BMV_IF_DELALLOC) + (BMV_IF_ATTRFORK|BMV_IF_NO_DMAPI_READ|BMV_IF_PREALLOC| \ + BMV_IF_DELALLOC|BMV_IF_NO_HOLES) /* bmv_oflags values - returned for each non-header segment */ #define BMV_OF_PREALLOC 0x1 /* segment = unwritten pre-allocation */ -- cgit v0.10.2 From 1ef78abec6b5e9e3062e3ae6660eabaf055a718d Mon Sep 17 00:00:00 2001 From: Ajit Khaparde Date: Fri, 3 Sep 2010 06:17:10 +0000 Subject: be2net: fix net-snmp error because of wrong packet stats Wrong packet statistics for multicast Rx was causing net-snmp error messages every 15 seconds. Instead of picking the multicast stats from hardware, now maintain it in the driver itself. Signed-off-by: Ajit Khaparde Signed-off-by: David S. Miller diff --git a/drivers/net/benet/be.h b/drivers/net/benet/be.h index 99197bd..53306bf 100644 --- a/drivers/net/benet/be.h +++ b/drivers/net/benet/be.h @@ -181,6 +181,7 @@ struct be_drvr_stats { u64 be_rx_bytes_prev; u64 be_rx_pkts; u32 be_rx_rate; + u32 be_rx_mcast_pkt; /* number of non ether type II frames dropped where * frame len > length field of Mac Hdr */ u32 be_802_3_dropped_frames; diff --git a/drivers/net/benet/be_ethtool.c b/drivers/net/benet/be_ethtool.c index cd16243..13f0abb 100644 --- a/drivers/net/benet/be_ethtool.c +++ b/drivers/net/benet/be_ethtool.c @@ -60,6 +60,7 @@ static const struct be_ethtool_stat et_stats[] = { {DRVSTAT_INFO(be_rx_events)}, {DRVSTAT_INFO(be_tx_compl)}, {DRVSTAT_INFO(be_rx_compl)}, + {DRVSTAT_INFO(be_rx_mcast_pkt)}, {DRVSTAT_INFO(be_ethrx_post_fail)}, {DRVSTAT_INFO(be_802_3_dropped_frames)}, {DRVSTAT_INFO(be_802_3_malformed_frames)}, diff --git a/drivers/net/benet/be_hw.h b/drivers/net/benet/be_hw.h index 5d38046..a2ec5df 100644 --- a/drivers/net/benet/be_hw.h +++ b/drivers/net/benet/be_hw.h @@ -167,8 +167,11 @@ #define FLASH_FCoE_BIOS_START_g3 (13631488) #define FLASH_REDBOOT_START_g3 (262144) - - +/************* Rx Packet Type Encoding **************/ +#define BE_UNICAST_PACKET 0 +#define BE_MULTICAST_PACKET 1 +#define BE_BROADCAST_PACKET 2 +#define BE_RSVD_PACKET 3 /* * BE descriptors: host memory data structures whose formats diff --git a/drivers/net/benet/be_main.c b/drivers/net/benet/be_main.c index 74e146f..a5a24e6 100644 --- a/drivers/net/benet/be_main.c +++ b/drivers/net/benet/be_main.c @@ -247,6 +247,7 @@ void netdev_stats_update(struct be_adapter *adapter) dev_stats->tx_packets = drvr_stats(adapter)->be_tx_pkts; dev_stats->rx_bytes = drvr_stats(adapter)->be_rx_bytes; dev_stats->tx_bytes = drvr_stats(adapter)->be_tx_bytes; + dev_stats->multicast = drvr_stats(adapter)->be_rx_mcast_pkt; /* bad pkts received */ dev_stats->rx_errors = port_stats->rx_crc_errors + @@ -294,7 +295,6 @@ void netdev_stats_update(struct be_adapter *adapter) /* no space available in linux */ dev_stats->tx_dropped = 0; - dev_stats->multicast = port_stats->rx_multicast_frames; dev_stats->collisions = 0; /* detailed tx_errors */ @@ -848,7 +848,7 @@ static void be_rx_rate_update(struct be_adapter *adapter) } static void be_rx_stats_update(struct be_adapter *adapter, - u32 pktsize, u16 numfrags) + u32 pktsize, u16 numfrags, u8 pkt_type) { struct be_drvr_stats *stats = drvr_stats(adapter); @@ -856,6 +856,9 @@ static void be_rx_stats_update(struct be_adapter *adapter, stats->be_rx_frags += numfrags; stats->be_rx_bytes += pktsize; stats->be_rx_pkts++; + + if (pkt_type == BE_MULTICAST_PACKET) + stats->be_rx_mcast_pkt++; } static inline bool do_pkt_csum(struct be_eth_rx_compl *rxcp, bool cso) @@ -925,9 +928,11 @@ static void skb_fill_rx_data(struct be_adapter *adapter, u16 rxq_idx, i, j; u32 pktsize, hdr_len, curr_frag_len, size; u8 *start; + u8 pkt_type; rxq_idx = AMAP_GET_BITS(struct amap_eth_rx_compl, fragndx, rxcp); pktsize = AMAP_GET_BITS(struct amap_eth_rx_compl, pktsize, rxcp); + pkt_type = AMAP_GET_BITS(struct amap_eth_rx_compl, cast_enc, rxcp); page_info = get_rx_page_info(adapter, rxq_idx); @@ -993,7 +998,7 @@ static void skb_fill_rx_data(struct be_adapter *adapter, BUG_ON(j > MAX_SKB_FRAGS); done: - be_rx_stats_update(adapter, pktsize, num_rcvd); + be_rx_stats_update(adapter, pktsize, num_rcvd, pkt_type); } /* Process the RX completion indicated by rxcp when GRO is disabled */ @@ -1060,6 +1065,7 @@ static void be_rx_compl_process_gro(struct be_adapter *adapter, u32 num_rcvd, pkt_size, remaining, vlanf, curr_frag_len; u16 i, rxq_idx = 0, vid, j; u8 vtm; + u8 pkt_type; num_rcvd = AMAP_GET_BITS(struct amap_eth_rx_compl, numfrags, rxcp); /* Is it a flush compl that has no data */ @@ -1070,6 +1076,7 @@ static void be_rx_compl_process_gro(struct be_adapter *adapter, vlanf = AMAP_GET_BITS(struct amap_eth_rx_compl, vtp, rxcp); rxq_idx = AMAP_GET_BITS(struct amap_eth_rx_compl, fragndx, rxcp); vtm = AMAP_GET_BITS(struct amap_eth_rx_compl, vtm, rxcp); + pkt_type = AMAP_GET_BITS(struct amap_eth_rx_compl, cast_enc, rxcp); /* vlanf could be wrongly set in some cards. * ignore if vtm is not set */ @@ -1125,7 +1132,7 @@ static void be_rx_compl_process_gro(struct be_adapter *adapter, vlan_gro_frags(&eq_obj->napi, adapter->vlan_grp, vid); } - be_rx_stats_update(adapter, pkt_size, num_rcvd); + be_rx_stats_update(adapter, pkt_size, num_rcvd, pkt_type); } static struct be_eth_rx_compl *be_rx_compl_get(struct be_adapter *adapter) -- cgit v0.10.2 From d053de911bff69ba7cdda36d3107c1da0fba7ccd Mon Sep 17 00:00:00 2001 From: Ajit Khaparde Date: Fri, 3 Sep 2010 06:23:30 +0000 Subject: be2net: fix a bug in UE detection logic The ONLINE registers can return 0xFFFFFFFF on more than one occassion. On systems that care, reading these registers could lead to problems. So the new code decides that the ASIC has encountered and error by reading the UE_STATUS_LOW/HIGH registers. AND them with the mask values and a non-zero result indicates an error. Signed-off-by: Ajit Khaparde Signed-off-by: David S. Miller diff --git a/drivers/net/benet/be_cmds.c b/drivers/net/benet/be_cmds.c index 3d30549..78f32fe 100644 --- a/drivers/net/benet/be_cmds.c +++ b/drivers/net/benet/be_cmds.c @@ -207,7 +207,7 @@ static int be_mbox_db_ready_wait(struct be_adapter *adapter, void __iomem *db) if (msecs > 4000) { dev_err(&adapter->pdev->dev, "mbox poll timed out\n"); - be_dump_ue(adapter); + be_detect_dump_ue(adapter); return -1; } diff --git a/drivers/net/benet/be_cmds.h b/drivers/net/benet/be_cmds.h index bdc10a2..ad1e6fa 100644 --- a/drivers/net/benet/be_cmds.h +++ b/drivers/net/benet/be_cmds.h @@ -992,5 +992,5 @@ extern int be_cmd_set_loopback(struct be_adapter *adapter, u8 port_num, extern int be_cmd_get_phy_info(struct be_adapter *adapter, struct be_dma_mem *cmd); extern int be_cmd_set_qos(struct be_adapter *adapter, u32 bps, u32 domain); -extern void be_dump_ue(struct be_adapter *adapter); +extern void be_detect_dump_ue(struct be_adapter *adapter); diff --git a/drivers/net/benet/be_main.c b/drivers/net/benet/be_main.c index a5a24e6..6eda7a0 100644 --- a/drivers/net/benet/be_main.c +++ b/drivers/net/benet/be_main.c @@ -1750,26 +1750,7 @@ static int be_poll_tx_mcc(struct napi_struct *napi, int budget) return 1; } -static inline bool be_detect_ue(struct be_adapter *adapter) -{ - u32 online0 = 0, online1 = 0; - - pci_read_config_dword(adapter->pdev, PCICFG_ONLINE0, &online0); - - pci_read_config_dword(adapter->pdev, PCICFG_ONLINE1, &online1); - - if (!online0 || !online1) { - adapter->ue_detected = true; - dev_err(&adapter->pdev->dev, - "UE Detected!! online0=%d online1=%d\n", - online0, online1); - return true; - } - - return false; -} - -void be_dump_ue(struct be_adapter *adapter) +void be_detect_dump_ue(struct be_adapter *adapter) { u32 ue_status_lo, ue_status_hi, ue_status_lo_mask, ue_status_hi_mask; u32 i; @@ -1786,6 +1767,11 @@ void be_dump_ue(struct be_adapter *adapter) ue_status_lo = (ue_status_lo & (~ue_status_lo_mask)); ue_status_hi = (ue_status_hi & (~ue_status_hi_mask)); + if (ue_status_lo || ue_status_hi) { + adapter->ue_detected = true; + dev_err(&adapter->pdev->dev, "UE Detected!!\n"); + } + if (ue_status_lo) { for (i = 0; ue_status_lo; ue_status_lo >>= 1, i++) { if (ue_status_lo & 1) @@ -1821,10 +1807,8 @@ static void be_worker(struct work_struct *work) adapter->rx_post_starved = false; be_post_rx_frags(adapter); } - if (!adapter->ue_detected) { - if (be_detect_ue(adapter)) - be_dump_ue(adapter); - } + if (!adapter->ue_detected) + be_detect_dump_ue(adapter); schedule_delayed_work(&adapter->work, msecs_to_jiffies(1000)); } -- cgit v0.10.2 From 323f30b361b2e6febb9065ab4a1a5298acb66ac1 Mon Sep 17 00:00:00 2001 From: Ajit Khaparde Date: Fri, 3 Sep 2010 06:24:13 +0000 Subject: be2net: remove a BUG_ON in be_cmds.c Async notifications other than link status are possible in certain configurations. Remove the BUG_ON in the mcc completion processing path. Signed-off-by: Ajit Khaparde Signed-off-by: David S. Miller diff --git a/drivers/net/benet/be_cmds.c b/drivers/net/benet/be_cmds.c index 78f32fe..34abcc9 100644 --- a/drivers/net/benet/be_cmds.c +++ b/drivers/net/benet/be_cmds.c @@ -140,10 +140,8 @@ int be_process_mcc(struct be_adapter *adapter, int *status) while ((compl = be_mcc_compl_get(adapter))) { if (compl->flags & CQE_FLAGS_ASYNC_MASK) { /* Interpret flags as an async trailer */ - BUG_ON(!is_link_state_evt(compl->flags)); - - /* Interpret compl as a async link evt */ - be_async_link_state_process(adapter, + if (is_link_state_evt(compl->flags)) + be_async_link_state_process(adapter, (struct be_async_event_link_state *) compl); } else if (compl->flags & CQE_FLAGS_COMPLETED_MASK) { *status = be_mcc_compl_process(adapter, compl); -- cgit v0.10.2 From 3fb5a991916091a908d53608a5899240039fb51e Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Thu, 2 Sep 2010 15:42:43 +0000 Subject: cls_cgroup: Fix rcu lockdep warning Dave reported an rcu lockdep warning on 2.6.35.4 kernel task->cgroups and task->cgroups->subsys[i] are protected by RCU. So we avoid accessing invalid pointers here. This might happen, for example, when you are deref-ing those pointers while someone move @task from one cgroup to another. Reported-by: Dave Jones Signed-off-by: Li Zefan Signed-off-by: David S. Miller diff --git a/include/net/cls_cgroup.h b/include/net/cls_cgroup.h index 726cc35..ef6c24a 100644 --- a/include/net/cls_cgroup.h +++ b/include/net/cls_cgroup.h @@ -27,11 +27,17 @@ struct cgroup_cls_state #ifdef CONFIG_NET_CLS_CGROUP static inline u32 task_cls_classid(struct task_struct *p) { + int classid; + if (in_interrupt()) return 0; - return container_of(task_subsys_state(p, net_cls_subsys_id), - struct cgroup_cls_state, css)->classid; + rcu_read_lock(); + classid = container_of(task_subsys_state(p, net_cls_subsys_id), + struct cgroup_cls_state, css)->classid; + rcu_read_unlock(); + + return classid; } #else extern int net_cls_subsys_id; -- cgit v0.10.2 From 57f9bdac2510cd7fda58e4a111d250861eb1ebeb Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 25 Aug 2010 09:12:29 +0200 Subject: sysfs: checking for NULL instead of ERR_PTR d_path() returns an ERR_PTR and it doesn't return NULL. Signed-off-by: Dan Carpenter Cc: stable Reviewed-by: "Eric W. Biederman" Signed-off-by: Greg Kroah-Hartman diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index 1b27b56..da3fefe 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -340,7 +340,7 @@ static int sysfs_open_file(struct inode *inode, struct file *file) char *p; p = d_path(&file->f_path, last_sysfs_file, sizeof(last_sysfs_file)); - if (p) + if (!IS_ERR(p)) memmove(last_sysfs_file, p, strlen(p) + 1); /* need attr_sd for attr and ops, its parent for kobj */ -- cgit v0.10.2 From 9fc2b2d0cf743008d8f6be6293278f4ef61f09f3 Mon Sep 17 00:00:00 2001 From: Francisco Jerez Date: Sun, 22 Aug 2010 17:37:24 +0200 Subject: vt: Fix console corruption on driver hand-over. After 02f0777a0d6560eb995aade34a1b82f95c0452da "vc_origin" is no longer reset to the screen buffer before calling the con_init() hook of the new console driver. If the old driver wasn't using a fixed scanout buffer (e.g. the case of vgacon) "vc_origin" may be a pointer to a VRAM location, and its contents aren't guaranteed to be preserved after calling con_deinit() on the old driver and con_init() on the new driver, i.e. the subsequent console resize may fill the framebuffer with garbage. It can be reproduced in the transition from vgacon to the nouveau framebuffer driver: in that case the legacy VGA aperture "vc_origin" points to becomes unreadable after fbcon_init(). This patch reverts the mentioned commit. To avoid the problem it intended to fix, stop using "vc_scr_end" in vc_do_resize() to calculate how many rows we have to copy (actually the code looks simpler this way without the help of "vc_scr_end"). Signed-off-by: Francisco Jerez Cc: qiaochong Cc: Greg Kroah-Hartman Cc: Andrew Morton Cc: Alan Cox Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/char/vt.c b/drivers/char/vt.c index 50590c7..281aada 100644 --- a/drivers/char/vt.c +++ b/drivers/char/vt.c @@ -906,22 +906,16 @@ static int vc_do_resize(struct tty_struct *tty, struct vc_data *vc, * bottom of buffer */ old_origin += (old_rows - new_rows) * old_row_size; - end = vc->vc_scr_end; } else { /* * Cursor is in no man's land, copy 1/2 screenful * from the top and bottom of cursor position */ old_origin += (vc->vc_y - new_rows/2) * old_row_size; - end = old_origin + (old_row_size * new_rows); } - } else - /* - * Cursor near the top, copy contents from the top of buffer - */ - end = (old_rows > new_rows) ? old_origin + - (old_row_size * new_rows) : - vc->vc_scr_end; + } + + end = old_origin + old_row_size * min(old_rows, new_rows); update_attr(vc); @@ -3075,8 +3069,7 @@ static int bind_con_driver(const struct consw *csw, int first, int last, old_was_color = vc->vc_can_do_color; vc->vc_sw->con_deinit(vc); - if (!vc->vc_origin) - vc->vc_origin = (unsigned long)vc->vc_screenbuf; + vc->vc_origin = (unsigned long)vc->vc_screenbuf; visual_init(vc, i, 0); set_origin(vc); update_attr(vc); -- cgit v0.10.2 From d86b3001a1a6f9b8be15e5e060338de085d2bbbc Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Tue, 31 Aug 2010 17:08:52 +0200 Subject: MAINTAINERS: orphan isicom I do not maintain isicom anymore... Signed-off-by: Jiri Slaby Signed-off-by: Greg Kroah-Hartman diff --git a/MAINTAINERS b/MAINTAINERS index c36f5d7..c5afc93d 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3923,8 +3923,7 @@ F: Documentation/sound/oss/MultiSound F: sound/oss/msnd* MULTITECH MULTIPORT CARD (ISICOM) -M: Jiri Slaby -S: Maintained +S: Orphan F: drivers/char/isicom.c F: include/linux/isicom.h -- cgit v0.10.2 From 71cad0554956de87c3fc413b1eac9313887eb14f Mon Sep 17 00:00:00 2001 From: Philippe Langlais Date: Tue, 31 Aug 2010 14:19:09 +0200 Subject: serial: fix port type conflict between NS16550A & U6_16550A Bug seen by Dr. David Alan Gilbert with sparse Signed-off-by: Philippe Langlais Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/include/linux/serial.h b/include/linux/serial.h index 1ebc694..ef91406 100644 --- a/include/linux/serial.h +++ b/include/linux/serial.h @@ -77,8 +77,7 @@ struct serial_struct { #define PORT_16654 11 #define PORT_16850 12 #define PORT_RSA 13 /* RSA-DV II/S card */ -#define PORT_U6_16550A 14 -#define PORT_MAX 14 +#define PORT_MAX 13 #define SERIAL_IO_PORT 0 #define SERIAL_IO_HUB6 1 diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h index 64458a9..563e234 100644 --- a/include/linux/serial_core.h +++ b/include/linux/serial_core.h @@ -44,7 +44,8 @@ #define PORT_RM9000 16 /* PMC-Sierra RM9xxx internal UART */ #define PORT_OCTEON 17 /* Cavium OCTEON internal UART */ #define PORT_AR7 18 /* Texas Instruments AR7 internal UART */ -#define PORT_MAX_8250 18 /* max port ID */ +#define PORT_U6_16550A 19 /* ST-Ericsson U6xxx internal UART */ +#define PORT_MAX_8250 19 /* max port ID */ /* * ARM specific type numbers. These are not currently guaranteed -- cgit v0.10.2 From 336746918299f2ca16b31490655b4ff7c8824c87 Mon Sep 17 00:00:00 2001 From: Sonic Zhang Date: Sat, 28 Aug 2010 16:32:55 -0400 Subject: serial: bfin_sport_uart: restore transmit frame sync fix The large cleanup/rewrite of resources in commit ccf68e59e93181df9353c0cc accidentally reverted an earlier fix in commit a19e8b205915b2925aca75b. So restore it here. Signed-off-by: Sonic Zhang Signed-off-by: Mike Frysinger Cc: stable [.34 and newer] Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/serial/bfin_sport_uart.c b/drivers/serial/bfin_sport_uart.c index e57fb3d..5318dd3 100644 --- a/drivers/serial/bfin_sport_uart.c +++ b/drivers/serial/bfin_sport_uart.c @@ -121,7 +121,7 @@ static int sport_uart_setup(struct sport_uart_port *up, int size, int baud_rate) unsigned int sclk = get_sclk(); /* Set TCR1 and TCR2, TFSR is not enabled for uart */ - SPORT_PUT_TCR1(up, (ITFS | TLSBIT | ITCLK)); + SPORT_PUT_TCR1(up, (LATFS | ITFS | TFSR | TLSBIT | ITCLK)); SPORT_PUT_TCR2(up, size + 1); pr_debug("%s TCR1:%x, TCR2:%x\n", __func__, SPORT_GET_TCR1(up), SPORT_GET_TCR2(up)); -- cgit v0.10.2 From 6eb68d6f3bf1707d5d816ea9242b7d38f25b942e Mon Sep 17 00:00:00 2001 From: Nathael Pajani Date: Thu, 2 Sep 2010 16:06:16 +0200 Subject: tty: fix tty_line must not be equal to number of allocated tty pointers in tty driver I found a bug "by chance" in drivers/char/tty_io.c I mean "by chance" because I was just reading the code of the tty_find_polling_driver() to make a new tty_find_by_name() function. In tty_find_polling_driver() the driver actually test "tty_line <= p->num" while num refers to the number of struct tty_struct pointers allocated for the p->ttys (p is a tty_driver), and tty_line is scanned in a tty name, which can be for example ttyS2. Then tty_line equals 2. And if p->num is 2, we have only p->ttys[0] and p->ttys[1], but no p->ttys[2]. This is actually unharmful, for tty_find_polling_driver() is used only in drivers/serial/kgdboc.c, and there's a test over there to find a console with a matching index, which will never happen. This is still a bug anyway. Signed-off-by: Nathael Pajani Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index 949067a..613c852 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -355,7 +355,7 @@ struct tty_driver *tty_find_polling_driver(char *name, int *line) if (*stp == '\0') stp = NULL; - if (tty_line >= 0 && tty_line <= p->num && p->ops && + if (tty_line >= 0 && tty_line < p->num && p->ops && p->ops->poll_init && !p->ops->poll_init(p, tty_line, stp)) { res = tty_driver_kref_get(p); *line = tty_line; -- cgit v0.10.2 From 0f1312b260499b34bb92fc9bd8ca1560dda2c4da Mon Sep 17 00:00:00 2001 From: Maurus Cuelenaere Date: Fri, 13 Aug 2010 21:29:30 +0200 Subject: USB: s3c-hsotg: Remove DEBUG define DEBUG is defined unconditionally, remove it as this clutters the message log. Signed-off-by: Maurus Cuelenaere Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/gadget/s3c-hsotg.c b/drivers/usb/gadget/s3c-hsotg.c index 521ebed..a229744 100644 --- a/drivers/usb/gadget/s3c-hsotg.c +++ b/drivers/usb/gadget/s3c-hsotg.c @@ -12,8 +12,6 @@ * published by the Free Software Foundation. */ -#define DEBUG - #include #include #include -- cgit v0.10.2 From 08a3b3b1c2e622e378d9086aee9e2e42ce37591d Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Sat, 14 Aug 2010 11:06:19 +0200 Subject: USB: ehci-ppc-of: problems in unwind The iounmap(ehci->ohci_hcctrl_reg); should be the first thing we do because the ioremap() was the last thing we did. Also if we hit any of the goto statements in the original code then it would have led to a NULL dereference of "ehci". This bug was introduced in: 796bcae7361c "USB: powerpc: Workaround for the PPC440EPX USBH_23 errata [take 3]" I modified the few lines in front a little so that my code didn't obscure the return success code path. Signed-off-by: Dan Carpenter Reviewed-by: Grant Likely Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/host/ehci-ppc-of.c b/drivers/usb/host/ehci-ppc-of.c index 335ee69..ba52be4 100644 --- a/drivers/usb/host/ehci-ppc-of.c +++ b/drivers/usb/host/ehci-ppc-of.c @@ -192,17 +192,19 @@ ehci_hcd_ppc_of_probe(struct platform_device *op, const struct of_device_id *mat } rv = usb_add_hcd(hcd, irq, 0); - if (rv == 0) - return 0; + if (rv) + goto err_ehci; + + return 0; +err_ehci: + if (ehci->has_amcc_usb23) + iounmap(ehci->ohci_hcctrl_reg); iounmap(hcd->regs); err_ioremap: irq_dispose_mapping(irq); err_irq: release_mem_region(hcd->rsrc_start, hcd->rsrc_len); - - if (ehci->has_amcc_usb23) - iounmap(ehci->ohci_hcctrl_reg); err_rmr: usb_put_hcd(hcd); -- cgit v0.10.2 From 793f03aa7bda8f492e12ada3de711b4ad7f4d8d0 Mon Sep 17 00:00:00 2001 From: Henrik Kretzschmar Date: Fri, 20 Aug 2010 19:57:50 +0200 Subject: USB: rndis: section mismatch fix This patch removes the following section mismatch warning, by moving the function rndis_init() from .init.text to .text. WARNING: vmlinux.o(.text+0x1aeca5a): Section mismatch in reference from the function rndis_bind_config() to the function .init.text:rndis_init() The function rndis_bind_config() references the function __init rndis_init(). This is often because rndis_bind_config lacks a __init annotation or the annotation of rndis_init is wrong. Signed-off-by: Henrik Kretzschmar Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/gadget/rndis.c b/drivers/usb/gadget/rndis.c index 020fa5a..ee337f7 100644 --- a/drivers/usb/gadget/rndis.c +++ b/drivers/usb/gadget/rndis.c @@ -1148,7 +1148,7 @@ static struct proc_dir_entry *rndis_connect_state [RNDIS_MAX_CONFIGS]; #endif /* CONFIG_USB_GADGET_DEBUG_FILES */ -int __init rndis_init (void) +int rndis_init(void) { u8 i; diff --git a/drivers/usb/gadget/rndis.h b/drivers/usb/gadget/rndis.h index c236aaa..907c330 100644 --- a/drivers/usb/gadget/rndis.h +++ b/drivers/usb/gadget/rndis.h @@ -262,7 +262,7 @@ int rndis_signal_disconnect (int configNr); int rndis_state (int configNr); extern void rndis_set_host_mac (int configNr, const u8 *addr); -int __devinit rndis_init (void); +int rndis_init(void); void rndis_exit (void); #endif /* _LINUX_RNDIS_H */ -- cgit v0.10.2 From 037d3656adbd7e8cb848f01cf5dec423ed76bbe7 Mon Sep 17 00:00:00 2001 From: Maxim Osipov Date: Sat, 21 Aug 2010 14:54:06 +0400 Subject: USB: Fix kernel oops with g_ether and Windows Please find attached patch for https://bugzilla.kernel.org/show_bug.cgi?id=16023 problem. Signed-off-by: Maxim Osipov Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/gadget/rndis.c b/drivers/usb/gadget/rndis.c index ee337f7..972d5dd 100644 --- a/drivers/usb/gadget/rndis.c +++ b/drivers/usb/gadget/rndis.c @@ -293,9 +293,13 @@ gen_ndis_query_resp (int configNr, u32 OID, u8 *buf, unsigned buf_len, /* mandatory */ case OID_GEN_VENDOR_DESCRIPTION: pr_debug("%s: OID_GEN_VENDOR_DESCRIPTION\n", __func__); - length = strlen (rndis_per_dev_params [configNr].vendorDescr); - memcpy (outbuf, - rndis_per_dev_params [configNr].vendorDescr, length); + if ( rndis_per_dev_params [configNr].vendorDescr ) { + length = strlen (rndis_per_dev_params [configNr].vendorDescr); + memcpy (outbuf, + rndis_per_dev_params [configNr].vendorDescr, length); + } else { + outbuf[0] = 0; + } retval = 0; break; -- cgit v0.10.2 From c7aa8f44b4d1dc73591894a2dd6909213612d299 Mon Sep 17 00:00:00 2001 From: Dirk De Schepper Date: Tue, 24 Aug 2010 20:38:35 +0100 Subject: USB: option: fix incorrect novatel entries Unfortunately some of the hardware PID belonging to auto-install CDROM (AICD) of Novatel modems found their way into the option module. This causes the AICD to be treated as a modem in stead of a disk. Since the modem ports do not appear until after the AICD is ejected, this essentially disables the modem. After a couple of minutes the AICD should auto-eject, but it is just too long a wait. The frequency of the failure seems to depend on both the hardware and the linux distribution. Here is a patch that fixes this up, and also adds a couple of new PID, offering some explanations and removing some incomplete and unnecessary comments. Signed-off-by: Dirk De Schepper Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c index adcbdb9..c46911a 100644 --- a/drivers/usb/serial/option.c +++ b/drivers/usb/serial/option.c @@ -164,6 +164,14 @@ static void option_instat_callback(struct urb *urb); #define YISO_VENDOR_ID 0x0EAB #define YISO_PRODUCT_U893 0xC893 +/* + * NOVATEL WIRELESS PRODUCTS + * + * Note from Novatel Wireless: + * If your Novatel modem does not work on linux, don't + * change the option module, but check our website. If + * that does not help, contact ddeschepper@nvtl.com +*/ /* MERLIN EVDO PRODUCTS */ #define NOVATELWIRELESS_PRODUCT_V640 0x1100 #define NOVATELWIRELESS_PRODUCT_V620 0x1110 @@ -185,24 +193,39 @@ static void option_instat_callback(struct urb *urb); #define NOVATELWIRELESS_PRODUCT_EU730 0x2400 #define NOVATELWIRELESS_PRODUCT_EU740 0x2410 #define NOVATELWIRELESS_PRODUCT_EU870D 0x2420 - /* OVATION PRODUCTS */ #define NOVATELWIRELESS_PRODUCT_MC727 0x4100 #define NOVATELWIRELESS_PRODUCT_MC950D 0x4400 -#define NOVATELWIRELESS_PRODUCT_U727 0x5010 -#define NOVATELWIRELESS_PRODUCT_MC727_NEW 0x5100 -#define NOVATELWIRELESS_PRODUCT_MC760 0x6000 +/* + * Note from Novatel Wireless: + * All PID in the 5xxx range are currently reserved for + * auto-install CDROMs, and should not be added to this + * module. + * + * #define NOVATELWIRELESS_PRODUCT_U727 0x5010 + * #define NOVATELWIRELESS_PRODUCT_MC727_NEW 0x5100 +*/ #define NOVATELWIRELESS_PRODUCT_OVMC760 0x6002 - -/* FUTURE NOVATEL PRODUCTS */ -#define NOVATELWIRELESS_PRODUCT_EVDO_HIGHSPEED 0X6001 -#define NOVATELWIRELESS_PRODUCT_HSPA_FULLSPEED 0X7000 -#define NOVATELWIRELESS_PRODUCT_HSPA_HIGHSPEED 0X7001 -#define NOVATELWIRELESS_PRODUCT_EVDO_EMBEDDED_FULLSPEED 0X8000 -#define NOVATELWIRELESS_PRODUCT_EVDO_EMBEDDED_HIGHSPEED 0X8001 -#define NOVATELWIRELESS_PRODUCT_HSPA_EMBEDDED_FULLSPEED 0X9000 -#define NOVATELWIRELESS_PRODUCT_HSPA_EMBEDDED_HIGHSPEED 0X9001 -#define NOVATELWIRELESS_PRODUCT_GLOBAL 0XA001 +#define NOVATELWIRELESS_PRODUCT_MC780 0x6010 +#define NOVATELWIRELESS_PRODUCT_EVDO_FULLSPEED 0x6000 +#define NOVATELWIRELESS_PRODUCT_EVDO_HIGHSPEED 0x6001 +#define NOVATELWIRELESS_PRODUCT_HSPA_FULLSPEED 0x7000 +#define NOVATELWIRELESS_PRODUCT_HSPA_HIGHSPEED 0x7001 +#define NOVATELWIRELESS_PRODUCT_HSPA_HIGHSPEED3 0x7003 +#define NOVATELWIRELESS_PRODUCT_HSPA_HIGHSPEED4 0x7004 +#define NOVATELWIRELESS_PRODUCT_HSPA_HIGHSPEED5 0x7005 +#define NOVATELWIRELESS_PRODUCT_HSPA_HIGHSPEED6 0x7006 +#define NOVATELWIRELESS_PRODUCT_HSPA_HIGHSPEED7 0x7007 +#define NOVATELWIRELESS_PRODUCT_MC996D 0x7030 +#define NOVATELWIRELESS_PRODUCT_MF3470 0x7041 +#define NOVATELWIRELESS_PRODUCT_MC547 0x7042 +#define NOVATELWIRELESS_PRODUCT_EVDO_EMBEDDED_FULLSPEED 0x8000 +#define NOVATELWIRELESS_PRODUCT_EVDO_EMBEDDED_HIGHSPEED 0x8001 +#define NOVATELWIRELESS_PRODUCT_HSPA_EMBEDDED_FULLSPEED 0x9000 +#define NOVATELWIRELESS_PRODUCT_HSPA_EMBEDDED_HIGHSPEED 0x9001 +#define NOVATELWIRELESS_PRODUCT_G1 0xA001 +#define NOVATELWIRELESS_PRODUCT_G1_M 0xA002 +#define NOVATELWIRELESS_PRODUCT_G2 0xA010 /* AMOI PRODUCTS */ #define AMOI_VENDOR_ID 0x1614 @@ -490,36 +513,44 @@ static const struct usb_device_id option_ids[] = { { USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_K3765, 0xff, 0xff, 0xff) }, { USB_DEVICE_AND_INTERFACE_INFO(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_ETS1220, 0xff, 0xff, 0xff) }, { USB_DEVICE(HUAWEI_VENDOR_ID, HUAWEI_PRODUCT_E14AC) }, - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_V640) }, /* Novatel Merlin V640/XV620 */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_V620) }, /* Novatel Merlin V620/S620 */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_V740) }, /* Novatel Merlin EX720/V740/X720 */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_V720) }, /* Novatel Merlin V720/S720/PC720 */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_U730) }, /* Novatel U730/U740 (VF version) */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_U740) }, /* Novatel U740 */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_U870) }, /* Novatel U870 */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_XU870) }, /* Novatel Merlin XU870 HSDPA/3G */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_X950D) }, /* Novatel X950D */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_EV620) }, /* Novatel EV620/ES620 CDMA/EV-DO */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_ES720) }, /* Novatel ES620/ES720/U720/USB720 */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_E725) }, /* Novatel E725/E726 */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_ES620) }, /* Novatel Merlin ES620 SM Bus */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_EU730) }, /* Novatel EU730 and Vodafone EU740 */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_EU740) }, /* Novatel non-Vodafone EU740 */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_EU870D) }, /* Novatel EU850D/EU860D/EU870D */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_MC950D) }, /* Novatel MC930D/MC950D */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_MC727) }, /* Novatel MC727/U727/USB727 */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_MC727_NEW) }, /* Novatel MC727/U727/USB727 refresh */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_U727) }, /* Novatel MC727/U727/USB727 */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_MC760) }, /* Novatel MC760/U760/USB760 */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_OVMC760) }, /* Novatel Ovation MC760 */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_HSPA_FULLSPEED) }, /* Novatel HSPA product */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_EVDO_EMBEDDED_FULLSPEED) }, /* Novatel EVDO Embedded product */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_HSPA_EMBEDDED_FULLSPEED) }, /* Novatel HSPA Embedded product */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_EVDO_HIGHSPEED) }, /* Novatel EVDO product */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_HSPA_HIGHSPEED) }, /* Novatel HSPA product */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_EVDO_EMBEDDED_HIGHSPEED) }, /* Novatel EVDO Embedded product */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_HSPA_EMBEDDED_HIGHSPEED) }, /* Novatel HSPA Embedded product */ - { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_GLOBAL) }, /* Novatel Global product */ + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_V640) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_V620) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_V740) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_V720) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_U730) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_U740) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_U870) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_XU870) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_X950D) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_EV620) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_ES720) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_E725) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_ES620) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_EU730) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_EU740) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_EU870D) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_MC950D) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_MC727) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_OVMC760) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_MC780) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_EVDO_FULLSPEED) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_HSPA_FULLSPEED) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_EVDO_EMBEDDED_FULLSPEED) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_HSPA_EMBEDDED_FULLSPEED) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_EVDO_HIGHSPEED) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_HSPA_HIGHSPEED3) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_HSPA_HIGHSPEED4) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_HSPA_HIGHSPEED5) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_HSPA_HIGHSPEED6) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_HSPA_HIGHSPEED7) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_MC996D) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_MF3470) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_MC547) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_EVDO_EMBEDDED_HIGHSPEED) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_HSPA_EMBEDDED_HIGHSPEED) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_G1) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_G1_M) }, + { USB_DEVICE(NOVATELWIRELESS_VENDOR_ID, NOVATELWIRELESS_PRODUCT_G2) }, { USB_DEVICE(AMOI_VENDOR_ID, AMOI_PRODUCT_H01) }, { USB_DEVICE(AMOI_VENDOR_ID, AMOI_PRODUCT_H01A) }, -- cgit v0.10.2 From 541e05ec3add5ab5bcf238d60161b53480280b20 Mon Sep 17 00:00:00 2001 From: Craig Shelley Date: Mon, 23 Aug 2010 20:50:57 +0100 Subject: USB: CP210x Add new device ID New device ID added for Balluff RFID reader. Signed-off-by: Craig Shelley Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/cp210x.c b/drivers/usb/serial/cp210x.c index 80bf833..103f9d2 100644 --- a/drivers/usb/serial/cp210x.c +++ b/drivers/usb/serial/cp210x.c @@ -109,6 +109,7 @@ static const struct usb_device_id id_table[] = { { USB_DEVICE(0x10C4, 0x83A8) }, /* Amber Wireless AMB2560 */ { USB_DEVICE(0x10C4, 0x8411) }, /* Kyocera GPS Module */ { USB_DEVICE(0x10C4, 0x846E) }, /* BEI USB Sensor Interface (VCP) */ + { USB_DEVICE(0x10C4, 0x8477) }, /* Balluff RFID */ { USB_DEVICE(0x10C4, 0xEA60) }, /* Silicon Labs factory default */ { USB_DEVICE(0x10C4, 0xEA61) }, /* Silicon Labs factory default */ { USB_DEVICE(0x10C4, 0xEA71) }, /* Infinity GPS-MIC-1 Radio Monophone */ @@ -122,14 +123,14 @@ static const struct usb_device_id id_table[] = { { USB_DEVICE(0x1555, 0x0004) }, /* Owen AC4 USB-RS485 Converter */ { USB_DEVICE(0x166A, 0x0303) }, /* Clipsal 5500PCU C-Bus USB interface */ { USB_DEVICE(0x16D6, 0x0001) }, /* Jablotron serial interface */ - { USB_DEVICE(0x17F4, 0xAAAA) }, /* Wavesense Jazz blood glucose meter */ - { USB_DEVICE(0x1843, 0x0200) }, /* Vaisala USB Instrument Cable */ - { USB_DEVICE(0x18EF, 0xE00F) }, /* ELV USB-I2C-Interface */ - { USB_DEVICE(0x413C, 0x9500) }, /* DW700 GPS USB interface */ { USB_DEVICE(0x16DC, 0x0010) }, /* W-IE-NE-R Plein & Baus GmbH PL512 Power Supply */ { USB_DEVICE(0x16DC, 0x0011) }, /* W-IE-NE-R Plein & Baus GmbH RCM Remote Control for MARATON Power Supply */ { USB_DEVICE(0x16DC, 0x0012) }, /* W-IE-NE-R Plein & Baus GmbH MPOD Multi Channel Power Supply */ { USB_DEVICE(0x16DC, 0x0015) }, /* W-IE-NE-R Plein & Baus GmbH CML Control, Monitoring and Data Logger */ + { USB_DEVICE(0x17F4, 0xAAAA) }, /* Wavesense Jazz blood glucose meter */ + { USB_DEVICE(0x1843, 0x0200) }, /* Vaisala USB Instrument Cable */ + { USB_DEVICE(0x18EF, 0xE00F) }, /* ELV USB-I2C-Interface */ + { USB_DEVICE(0x413C, 0x9500) }, /* DW700 GPS USB interface */ { } /* Terminating Entry */ }; -- cgit v0.10.2 From 0bf7a81c5d447c21db434be35363c44c0a30f598 Mon Sep 17 00:00:00 2001 From: Jason Detring Date: Thu, 26 Aug 2010 15:08:54 -0500 Subject: USB: cp210x: Add B&G H3000 link cable ID This is the cable between an H3000 navigation unit and a multi-function display. http://www.bandg.com/en/Products/H3000/Spares-and-Accessories/Cables/H3000-CPU-USB-Cable-Pack/ Signed-off-by: Jason Detring Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/cp210x.c b/drivers/usb/serial/cp210x.c index 103f9d2..8ed6ff6 100644 --- a/drivers/usb/serial/cp210x.c +++ b/drivers/usb/serial/cp210x.c @@ -88,6 +88,7 @@ static const struct usb_device_id id_table[] = { { USB_DEVICE(0x10C4, 0x8149) }, /* West Mountain Radio Computerized Battery Analyzer */ { USB_DEVICE(0x10C4, 0x814A) }, /* West Mountain Radio RIGblaster P&P */ { USB_DEVICE(0x10C4, 0x814B) }, /* West Mountain Radio RIGtalk */ + { USB_DEVICE(0x10C4, 0x8156) }, /* B&G H3000 link cable */ { USB_DEVICE(0x10C4, 0x815E) }, /* Helicomm IP-Link 1220-DVM */ { USB_DEVICE(0x10C4, 0x818B) }, /* AVIT Research USB to TTL */ { USB_DEVICE(0x10C4, 0x819F) }, /* MJS USB Toslink Switcher */ -- cgit v0.10.2 From 5b22a32e76defeb573991b301a27d299472c5714 Mon Sep 17 00:00:00 2001 From: A E Lawrence Date: Sun, 29 Aug 2010 21:51:52 +0100 Subject: USB: cp210x usb driver: add USB_DEVICE for Pirelli DP-L10 mobile. The Pirelli DP-L10 mobile is sold under various brand names. One, already supported by cp210x, is the T-COM TC300. Here is the lsusb for that version: ------------------------------------------------------------------- Bus 001 Device 002: ID 0489:e000 Foxconn / Hon Hai T-Com TC 300 Device Descriptor: bLength 18 bDescriptorType 1 bcdUSB 1.10 bDeviceClass 0 (Defined at Interface level) bDeviceSubClass 0 bDeviceProtocol 0 bMaxPacketSize0 64 idVendor 0x0489 Foxconn / Hon Hai idProduct 0xe000 T-Com TC 300 bcdDevice 1.00 iManufacturer 1 Silicon Labs iProduct 2 TC 300 iSerial 3 0001 [snip] --------------------------------------------------------------------------- However the native Pirelli DP-L10 is not supported: ------------------------------------------------------------------ Bus 001 Device 003: ID 0489:e003 Foxconn / Hon Hai Pirelli DP-L10 Device Descriptor: bLength 18 bDescriptorType 1 bcdUSB 1.10 bDeviceClass 0 (Defined at Interface level) bDeviceSubClass 0 bDeviceProtocol 0 bMaxPacketSize0 64 idVendor 0x0489 Foxconn / Hon Hai idProduct 0xe003 Pirelli DP-L10 bcdDevice 1.00 iManufacturer 1 Silicon Labs iProduct 2 DP-L10 iSerial 3 0001 [snip] ------------------------------------------------------------------------- All that is required is an extra USB_DEVICE entry: { USB_DEVICE(0x0489, 0xE003) }, /* Pirelli Broadband S.p.A, DP-L10 SIP/GSM +Mobile */ The patch adds that entry. Tested under 2.6.36-rc2 from git. Signed-off-by: A E Lawrence Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/cp210x.c b/drivers/usb/serial/cp210x.c index 8ed6ff6..4f1744c 100644 --- a/drivers/usb/serial/cp210x.c +++ b/drivers/usb/serial/cp210x.c @@ -56,6 +56,7 @@ static int debug; static const struct usb_device_id id_table[] = { { USB_DEVICE(0x0471, 0x066A) }, /* AKTAKOM ACE-1001 cable */ { USB_DEVICE(0x0489, 0xE000) }, /* Pirelli Broadband S.p.A, DP-L10 SIP/GSM Mobile */ + { USB_DEVICE(0x0489, 0xE003) }, /* Pirelli Broadband S.p.A, DP-L10 SIP/GSM Mobile */ { USB_DEVICE(0x0745, 0x1000) }, /* CipherLab USB CCD Barcode Scanner 1000 */ { USB_DEVICE(0x08e6, 0x5501) }, /* Gemalto Prox-PU/CU contactless smartcard reader */ { USB_DEVICE(0x08FD, 0x000A) }, /* Digianswer A/S , ZigBee/802.15.4 MAC Device */ -- cgit v0.10.2 From 0791971ba8fbc44e4f476079f856335ed45e6324 Mon Sep 17 00:00:00 2001 From: Thadeu Lima de Souza Cascardo Date: Sat, 28 Aug 2010 03:06:29 -0300 Subject: usb: allow drivers to use allocated bandwidth until unbound When using the remove sysfs file, the device configuration is set to -1 (unconfigured). This eventually unbind drivers with the bandwidth_mutex held. Some drivers may call functions that hold said mutex, like usb_reset_device. This is the case for rtl8187, for example. This will lead to the same process holding the mutex twice, which deadlocks. Besides, according to Alan Stern: "The deadlock problem probably could be handled somehow, but there's a separate issue: Until the usb_disable_device call finishes unbinding the drivers, the drivers are free to continue using their allocated bandwidth. We musn't change the bandwidth allocations until after the unbinding is done. So this patch is indeed necessary." Unbinding the driver before holding the bandwidth_mutex solves the problem. If any operation after that fails, drivers are not bound again. But that would be a problem anyway that the user may solve resetting the device configuration to one that works, just like he would need to do in most other failure cases. Signed-off-by: Thadeu Lima de Souza Cascardo Cc: Alan Stern Cc: Sarah Sharp Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/core/message.c b/drivers/usb/core/message.c index fd4c36e..844683e 100644 --- a/drivers/usb/core/message.c +++ b/drivers/usb/core/message.c @@ -1724,6 +1724,15 @@ free_interfaces: if (ret) goto free_interfaces; + /* if it's already configured, clear out old state first. + * getting rid of old interfaces means unbinding their drivers. + */ + if (dev->state != USB_STATE_ADDRESS) + usb_disable_device(dev, 1); /* Skip ep0 */ + + /* Get rid of pending async Set-Config requests for this device */ + cancel_async_set_config(dev); + /* Make sure we have bandwidth (and available HCD resources) for this * configuration. Remove endpoints from the schedule if we're dropping * this configuration to set configuration 0. After this point, the @@ -1733,20 +1742,11 @@ free_interfaces: mutex_lock(&hcd->bandwidth_mutex); ret = usb_hcd_alloc_bandwidth(dev, cp, NULL, NULL); if (ret < 0) { - usb_autosuspend_device(dev); mutex_unlock(&hcd->bandwidth_mutex); + usb_autosuspend_device(dev); goto free_interfaces; } - /* if it's already configured, clear out old state first. - * getting rid of old interfaces means unbinding their drivers. - */ - if (dev->state != USB_STATE_ADDRESS) - usb_disable_device(dev, 1); /* Skip ep0 */ - - /* Get rid of pending async Set-Config requests for this device */ - cancel_async_set_config(dev); - ret = usb_control_msg(dev, usb_sndctrlpipe(dev, 0), USB_REQ_SET_CONFIGURATION, 0, configuration, 0, NULL, 0, USB_CTRL_SET_TIMEOUT); @@ -1761,8 +1761,8 @@ free_interfaces: if (!cp) { usb_set_device_state(dev, USB_STATE_ADDRESS); usb_hcd_alloc_bandwidth(dev, NULL, NULL, NULL); - usb_autosuspend_device(dev); mutex_unlock(&hcd->bandwidth_mutex); + usb_autosuspend_device(dev); goto free_interfaces; } mutex_unlock(&hcd->bandwidth_mutex); -- cgit v0.10.2 From 3c35b002da0c749ec15cf25cfe58f06aa230ae9c Mon Sep 17 00:00:00 2001 From: Bill Pemberton Date: Wed, 25 Aug 2010 18:21:23 -0400 Subject: USB: ssu100: turn off debug flag Remove the hard coding of the debug flag to 1. Signed-off-by: Bill Pemberton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/ssu100.c b/drivers/usb/serial/ssu100.c index 68c18fd..e986002 100644 --- a/drivers/usb/serial/ssu100.c +++ b/drivers/usb/serial/ssu100.c @@ -46,7 +46,7 @@ #define FULLPWRBIT 0x00000080 #define NEXT_BOARD_POWER_BIT 0x00000004 -static int debug = 1; +static int debug; /* Version Information */ #define DRIVER_VERSION "v0.1" -- cgit v0.10.2 From caf3a636a9f809fdca5fa746e6687096457accb1 Mon Sep 17 00:00:00 2001 From: Dave Ludlow Date: Tue, 31 Aug 2010 14:26:17 -0400 Subject: usb: serial: mos7840: Add USB ID to support the B&B Electronics USOPTL4-2P. Add the USB ID needed to support B&B Electronic's 2-port, optically-isolated, powered, USB to RS485 converter. Signed-off-by: Dave Ludlow Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/mos7840.c b/drivers/usb/serial/mos7840.c index 585b7e6..9a5ea9d 100644 --- a/drivers/usb/serial/mos7840.c +++ b/drivers/usb/serial/mos7840.c @@ -129,6 +129,7 @@ #define BANDB_DEVICE_ID_USOPTL4_2 0xAC42 #define BANDB_DEVICE_ID_USOPTL4_4 0xAC44 #define BANDB_DEVICE_ID_USOPTL2_4 0xAC24 +#define BANDB_DEVICE_ID_USOPTL4_2P 0xBC02 /* This driver also supports * ATEN UC2324 device using Moschip MCS7840 @@ -192,6 +193,7 @@ static const struct usb_device_id moschip_port_id_table[] = { {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL4_2)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL4_4)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL2_4)}, + {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL2_4P)}, {USB_DEVICE(USB_VENDOR_ID_ATENINTL, ATENINTL_DEVICE_ID_UC2324)}, {USB_DEVICE(USB_VENDOR_ID_ATENINTL, ATENINTL_DEVICE_ID_UC2322)}, {} /* terminating entry */ @@ -209,6 +211,7 @@ static const struct usb_device_id moschip_id_table_combined[] __devinitconst = { {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL4_2)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL4_4)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL2_4)}, + {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL2_4P)}, {USB_DEVICE(USB_VENDOR_ID_ATENINTL, ATENINTL_DEVICE_ID_UC2324)}, {USB_DEVICE(USB_VENDOR_ID_ATENINTL, ATENINTL_DEVICE_ID_UC2322)}, {} /* terminating entry */ -- cgit v0.10.2 From 4035e45632c2a8bb4edae83c20447051bd9a9604 Mon Sep 17 00:00:00 2001 From: Toby Gray Date: Wed, 1 Sep 2010 16:01:19 +0100 Subject: USB: cdc-acm: Adding second ACM channel support for various Nokia and one Samsung phones S60 phones from Nokia and Samsung expose two ACM channels. The first is a modem with a standard AT-command interface, which is picked up correctly by CDC-ACM. The second ACM port is marked as having a vendor-specific protocol. This means that the ACM driver will not claim the second channel by default. This adds support for the second ACM channel for the following devices: Nokia E63 Nokia E75 Nokia 6760 Slide Nokia E52 Nokia E55 Nokia E72 Nokia X6 Nokia N97 Mini Nokia 5800 Xpressmusic Nokia E90 Samsung GTi8510 (INNOV8) Signed-off-by: Toby Gray Cc: Oliver Neukum Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c index 1833b3a..a3d6c4d 100644 --- a/drivers/usb/class/cdc-acm.c +++ b/drivers/usb/class/cdc-acm.c @@ -1481,6 +1481,11 @@ static int acm_reset_resume(struct usb_interface *intf) USB_CLASS_COMM, USB_CDC_SUBCLASS_ACM, \ USB_CDC_ACM_PROTO_VENDOR) +#define SAMSUNG_PCSUITE_ACM_INFO(x) \ + USB_DEVICE_AND_INTERFACE_INFO(0x04e7, x, \ + USB_CLASS_COMM, USB_CDC_SUBCLASS_ACM, \ + USB_CDC_ACM_PROTO_VENDOR) + /* * USB driver structure. */ @@ -1591,6 +1596,17 @@ static const struct usb_device_id acm_ids[] = { { NOKIA_PCSUITE_ACM_INFO(0x0108), }, /* Nokia 5320 XpressMusic 2G */ { NOKIA_PCSUITE_ACM_INFO(0x01f5), }, /* Nokia N97, RM-505 */ { NOKIA_PCSUITE_ACM_INFO(0x02e3), }, /* Nokia 5230, RM-588 */ + { NOKIA_PCSUITE_ACM_INFO(0x0178), }, /* Nokia E63 */ + { NOKIA_PCSUITE_ACM_INFO(0x010e), }, /* Nokia E75 */ + { NOKIA_PCSUITE_ACM_INFO(0x02d9), }, /* Nokia 6760 Slide */ + { NOKIA_PCSUITE_ACM_INFO(0x01d0), }, /* Nokia E52 */ + { NOKIA_PCSUITE_ACM_INFO(0x0223), }, /* Nokia E72 */ + { NOKIA_PCSUITE_ACM_INFO(0x0275), }, /* Nokia X6 */ + { NOKIA_PCSUITE_ACM_INFO(0x026c), }, /* Nokia N97 Mini */ + { NOKIA_PCSUITE_ACM_INFO(0x0154), }, /* Nokia 5800 XpressMusic */ + { NOKIA_PCSUITE_ACM_INFO(0x04ce), }, /* Nokia E90 */ + { NOKIA_PCSUITE_ACM_INFO(0x01d4), }, /* Nokia E55 */ + { SAMSUNG_PCSUITE_ACM_INFO(0x6651), }, /* Samsung GTi8510 (INNOV8) */ /* NOTE: non-Nokia COMM/ACM/0xff is likely MSFT RNDIS... NOT a modem! */ -- cgit v0.10.2 From 870408c8291015872a7a0b583673a9e56b3e73f4 Mon Sep 17 00:00:00 2001 From: Dave Ludlow Date: Wed, 1 Sep 2010 12:33:30 -0400 Subject: usb: serial: mos7840: Add USB IDs to support more B&B USB/RS485 converters. Add the USB IDs needed to support the B&B USOPTL4-4P, USO9ML2-2P, and USO9ML2-4P. This patch expands and corrects a typo in the patch sent on 08-31-2010. Signed-off-by: Dave Ludlow Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/mos7840.c b/drivers/usb/serial/mos7840.c index 9a5ea9d..1c9b6e9 100644 --- a/drivers/usb/serial/mos7840.c +++ b/drivers/usb/serial/mos7840.c @@ -119,17 +119,20 @@ * by making a change here, in moschip_port_id_table, and in * moschip_id_table_combined */ -#define USB_VENDOR_ID_BANDB 0x0856 -#define BANDB_DEVICE_ID_USO9ML2_2 0xAC22 -#define BANDB_DEVICE_ID_USO9ML2_4 0xAC24 -#define BANDB_DEVICE_ID_US9ML2_2 0xAC29 -#define BANDB_DEVICE_ID_US9ML2_4 0xAC30 -#define BANDB_DEVICE_ID_USPTL4_2 0xAC31 -#define BANDB_DEVICE_ID_USPTL4_4 0xAC32 -#define BANDB_DEVICE_ID_USOPTL4_2 0xAC42 -#define BANDB_DEVICE_ID_USOPTL4_4 0xAC44 -#define BANDB_DEVICE_ID_USOPTL2_4 0xAC24 -#define BANDB_DEVICE_ID_USOPTL4_2P 0xBC02 +#define USB_VENDOR_ID_BANDB 0x0856 +#define BANDB_DEVICE_ID_USO9ML2_2 0xAC22 +#define BANDB_DEVICE_ID_USO9ML2_2P 0xBC00 +#define BANDB_DEVICE_ID_USO9ML2_4 0xAC24 +#define BANDB_DEVICE_ID_USO9ML2_4P 0xBC01 +#define BANDB_DEVICE_ID_US9ML2_2 0xAC29 +#define BANDB_DEVICE_ID_US9ML2_4 0xAC30 +#define BANDB_DEVICE_ID_USPTL4_2 0xAC31 +#define BANDB_DEVICE_ID_USPTL4_4 0xAC32 +#define BANDB_DEVICE_ID_USOPTL4_2 0xAC42 +#define BANDB_DEVICE_ID_USOPTL4_2P 0xBC02 +#define BANDB_DEVICE_ID_USOPTL4_4 0xAC44 +#define BANDB_DEVICE_ID_USOPTL4_4P 0xBC03 +#define BANDB_DEVICE_ID_USOPTL2_4 0xAC24 /* This driver also supports * ATEN UC2324 device using Moschip MCS7840 @@ -185,15 +188,18 @@ static const struct usb_device_id moschip_port_id_table[] = { {USB_DEVICE(USB_VENDOR_ID_MOSCHIP, MOSCHIP_DEVICE_ID_7840)}, {USB_DEVICE(USB_VENDOR_ID_MOSCHIP, MOSCHIP_DEVICE_ID_7820)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USO9ML2_2)}, + {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USO9ML2_2P)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USO9ML2_4)}, + {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USO9ML2_4P)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_US9ML2_2)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_US9ML2_4)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USPTL4_2)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USPTL4_4)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL4_2)}, + {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL4_2P)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL4_4)}, + {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL4_4P)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL2_4)}, - {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL2_4P)}, {USB_DEVICE(USB_VENDOR_ID_ATENINTL, ATENINTL_DEVICE_ID_UC2324)}, {USB_DEVICE(USB_VENDOR_ID_ATENINTL, ATENINTL_DEVICE_ID_UC2322)}, {} /* terminating entry */ @@ -203,15 +209,18 @@ static const struct usb_device_id moschip_id_table_combined[] __devinitconst = { {USB_DEVICE(USB_VENDOR_ID_MOSCHIP, MOSCHIP_DEVICE_ID_7840)}, {USB_DEVICE(USB_VENDOR_ID_MOSCHIP, MOSCHIP_DEVICE_ID_7820)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USO9ML2_2)}, + {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USO9ML2_2P)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USO9ML2_4)}, + {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USO9ML2_4P)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_US9ML2_2)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_US9ML2_4)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USPTL4_2)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USPTL4_4)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL4_2)}, + {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL4_2P)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL4_4)}, + {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL4_4P)}, {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL2_4)}, - {USB_DEVICE(USB_VENDOR_ID_BANDB, BANDB_DEVICE_ID_USOPTL2_4P)}, {USB_DEVICE(USB_VENDOR_ID_ATENINTL, ATENINTL_DEVICE_ID_UC2324)}, {USB_DEVICE(USB_VENDOR_ID_ATENINTL, ATENINTL_DEVICE_ID_UC2322)}, {} /* terminating entry */ -- cgit v0.10.2 From 902ffc3c707c1d459ea57428a619a807cbe412f9 Mon Sep 17 00:00:00 2001 From: Simon Arlott Date: Wed, 1 Sep 2010 18:37:12 +0100 Subject: USB: cxacru: Use a bulk/int URB to access the command endpoint The command endpoint is either a bulk or interrupt endpoint, but using the wrong type of transfer causes an error if CONFIG_USB_DEBUG is enabled after commit f661c6f8c67bd55e93348f160d590ff9edf08904, which checks for this mismatch. Detect which type of endpoint it is and use a bulk/int URB as appropriate. There are other function calls specifying a bulk pipe, but usb_clear_halt doesn't use the pipe type (only the endpoint) and usb_bulk_msg auto-detects interrupt transfers. Signed-off-by: Simon Arlott Cc: stable [.34 and newer] Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/atm/cxacru.c b/drivers/usb/atm/cxacru.c index 593fc5e..5af23cc 100644 --- a/drivers/usb/atm/cxacru.c +++ b/drivers/usb/atm/cxacru.c @@ -1127,6 +1127,7 @@ static int cxacru_bind(struct usbatm_data *usbatm_instance, { struct cxacru_data *instance; struct usb_device *usb_dev = interface_to_usbdev(intf); + struct usb_host_endpoint *cmd_ep = usb_dev->ep_in[CXACRU_EP_CMD]; int ret; /* instance init */ @@ -1171,15 +1172,34 @@ static int cxacru_bind(struct usbatm_data *usbatm_instance, goto fail; } - usb_fill_int_urb(instance->rcv_urb, + if (!cmd_ep) { + dbg("cxacru_bind: no command endpoint"); + ret = -ENODEV; + goto fail; + } + + if ((cmd_ep->desc.bmAttributes & USB_ENDPOINT_XFERTYPE_MASK) + == USB_ENDPOINT_XFER_INT) { + usb_fill_int_urb(instance->rcv_urb, usb_dev, usb_rcvintpipe(usb_dev, CXACRU_EP_CMD), instance->rcv_buf, PAGE_SIZE, cxacru_blocking_completion, &instance->rcv_done, 1); - usb_fill_int_urb(instance->snd_urb, + usb_fill_int_urb(instance->snd_urb, usb_dev, usb_sndintpipe(usb_dev, CXACRU_EP_CMD), instance->snd_buf, PAGE_SIZE, cxacru_blocking_completion, &instance->snd_done, 4); + } else { + usb_fill_bulk_urb(instance->rcv_urb, + usb_dev, usb_rcvbulkpipe(usb_dev, CXACRU_EP_CMD), + instance->rcv_buf, PAGE_SIZE, + cxacru_blocking_completion, &instance->rcv_done); + + usb_fill_bulk_urb(instance->snd_urb, + usb_dev, usb_sndbulkpipe(usb_dev, CXACRU_EP_CMD), + instance->snd_buf, PAGE_SIZE, + cxacru_blocking_completion, &instance->snd_done); + } mutex_init(&instance->cm_serialize); -- cgit v0.10.2 From 5b239f0aebd4dd6f85b13decf5e18e86e35d57f0 Mon Sep 17 00:00:00 2001 From: Philippe Corbes Date: Tue, 31 Aug 2010 19:31:32 +0200 Subject: USB: cdc-acm: Add pseudo modem without AT command capabilities cdc-acm.c : Manage pseudo-modem without AT commands capabilities Enable to drive electronic simple gadgets based on microcontrolers. The Interface descriptor is like this: bInterfaceClass 2 Communications bInterfaceSubClass 2 Abstract (modem) bInterfaceProtocol 0 None Signed-off-by: Philippe Corbes Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c index a3d6c4d..597defc 100644 --- a/drivers/usb/class/cdc-acm.c +++ b/drivers/usb/class/cdc-acm.c @@ -1615,6 +1615,10 @@ static const struct usb_device_id acm_ids[] = { .driver_info = NOT_A_MODEM, }, + /* control interfaces without any protocol set */ + { USB_INTERFACE_INFO(USB_CLASS_COMM, USB_CDC_SUBCLASS_ACM, + USB_CDC_PROTO_NONE) }, + /* control interfaces with various AT-command sets */ { USB_INTERFACE_INFO(USB_CLASS_COMM, USB_CDC_SUBCLASS_ACM, USB_CDC_ACM_PROTO_AT_V25TER) }, -- cgit v0.10.2 From 577045c0a76e34294f902a7d5d60e90b04d094d0 Mon Sep 17 00:00:00 2001 From: Toby Gray Date: Thu, 2 Sep 2010 10:46:20 +0100 Subject: USB: cdc-acm: Fixing crash when ACM probing interfaces with no endpoint descriptors. Certain USB devices, such as the Nokia X6 mobile phone, don't expose any endpoint descriptors on some of their interfaces. If the ACM driver is forced to probe all interfaces on a device the a NULL pointer dereference will occur when the ACM driver attempts to use the endpoint of the alternative settings. One way to get the ACM driver to probe all the interfaces is by using the /sys/bus/usb/drivers/cdc_acm/new_id interface. This patch checks that the endpoint pointer for the current alternate settings is non-NULL before using it. Signed-off-by: Toby Gray Cc: Oliver Neukum Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c index 597defc..bc62fae 100644 --- a/drivers/usb/class/cdc-acm.c +++ b/drivers/usb/class/cdc-acm.c @@ -965,7 +965,8 @@ static int acm_probe(struct usb_interface *intf, } if (!buflen) { - if (intf->cur_altsetting->endpoint->extralen && + if (intf->cur_altsetting->endpoint && + intf->cur_altsetting->endpoint->extralen && intf->cur_altsetting->endpoint->extra) { dev_dbg(&intf->dev, "Seeking extra descriptors on endpoint\n"); -- cgit v0.10.2 From 657373883417b2618023fd4135d251ba06a2c30a Mon Sep 17 00:00:00 2001 From: Luke Lowrey Date: Thu, 2 Sep 2010 11:39:49 +0100 Subject: USB: ftdi_sio: Added custom PIDs for ChamSys products Added the 0xDAF8 to 0xDAFF PID range for ChamSys limited USB interface/wing products Signed-off-by: Luke Lowrey Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c index c792c96..97cc87d 100644 --- a/drivers/usb/serial/ftdi_sio.c +++ b/drivers/usb/serial/ftdi_sio.c @@ -753,6 +753,14 @@ static struct usb_device_id id_table_combined [] = { { USB_DEVICE(FTDI_VID, SEGWAY_RMP200_PID) }, { USB_DEVICE(IONICS_VID, IONICS_PLUGCOMPUTER_PID), .driver_info = (kernel_ulong_t)&ftdi_jtag_quirk }, + { USB_DEVICE(FTDI_VID, FTDI_CHAMSYS_24_MASTER_WING_PID) }, + { USB_DEVICE(FTDI_VID, FTDI_CHAMSYS_PC_WING_PID) }, + { USB_DEVICE(FTDI_VID, FTDI_CHAMSYS_USB_DMX_PID) }, + { USB_DEVICE(FTDI_VID, FTDI_CHAMSYS_MIDI_TIMECODE_PID) }, + { USB_DEVICE(FTDI_VID, FTDI_CHAMSYS_MINI_WING_PID) }, + { USB_DEVICE(FTDI_VID, FTDI_CHAMSYS_MAXI_WING_PID) }, + { USB_DEVICE(FTDI_VID, FTDI_CHAMSYS_MEDIA_WING_PID) }, + { USB_DEVICE(FTDI_VID, FTDI_CHAMSYS_WING_PID) }, { }, /* Optional parameter entry */ { } /* Terminating entry */ }; diff --git a/drivers/usb/serial/ftdi_sio_ids.h b/drivers/usb/serial/ftdi_sio_ids.h index 2e95857..15a4583 100644 --- a/drivers/usb/serial/ftdi_sio_ids.h +++ b/drivers/usb/serial/ftdi_sio_ids.h @@ -135,6 +135,18 @@ #define FTDI_NDI_AURORA_SCU_PID 0xDA74 /* NDI Aurora SCU */ /* + * ChamSys Limited (www.chamsys.co.uk) USB wing/interface product IDs + */ +#define FTDI_CHAMSYS_24_MASTER_WING_PID 0xDAF8 +#define FTDI_CHAMSYS_PC_WING_PID 0xDAF9 +#define FTDI_CHAMSYS_USB_DMX_PID 0xDAFA +#define FTDI_CHAMSYS_MIDI_TIMECODE_PID 0xDAFB +#define FTDI_CHAMSYS_MINI_WING_PID 0xDAFC +#define FTDI_CHAMSYS_MAXI_WING_PID 0xDAFD +#define FTDI_CHAMSYS_MEDIA_WING_PID 0xDAFE +#define FTDI_CHAMSYS_WING_PID 0xDAFF + +/* * Westrex International devices submitted by Cory Lee */ #define FTDI_WESTREX_MODEL_777_PID 0xDC00 /* Model 777 */ -- cgit v0.10.2 From b681b5886bb5d1f5b6750a0ed7c62846da7ccea4 Mon Sep 17 00:00:00 2001 From: Haiyang Zhang Date: Tue, 3 Aug 2010 19:15:31 +0000 Subject: staging: hv: Fix missing functions for net_device_ops Fix missing functions for net_device_ops. It's a bug when porting the drivers from 2.6.27 to 2.6.32. In 2.6.27, the default functions for Ethernet, like eth_change_mtu(), were assigned by ether_setup(). But in 2.6.32, these function pointers moved to net_device_ops structure and no longer be assigned in ether_setup(). So we need to set these functions in our driver code. It will ensure the MTU won't be set beyond 1500. Otherwise, this can cause an error on the server side, because the HyperV linux driver doesn't support jumbo frame yet. Signed-off-by: Haiyang Zhang Signed-off-by: Hank Janssen Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/hv/netvsc_drv.c b/drivers/staging/hv/netvsc_drv.c index 56e1157..64a0114 100644 --- a/drivers/staging/hv/netvsc_drv.c +++ b/drivers/staging/hv/netvsc_drv.c @@ -327,6 +327,9 @@ static const struct net_device_ops device_ops = { .ndo_stop = netvsc_close, .ndo_start_xmit = netvsc_start_xmit, .ndo_set_multicast_list = netvsc_set_multicast_list, + .ndo_change_mtu = eth_change_mtu, + .ndo_validate_addr = eth_validate_addr, + .ndo_set_mac_address = eth_mac_addr, }; static int netvsc_probe(struct device *device) -- cgit v0.10.2 From 0c47a70a9a8a6d1ec37a53d2f9cb82f8b8ef8aa2 Mon Sep 17 00:00:00 2001 From: Hank Janssen Date: Thu, 5 Aug 2010 19:29:44 +0000 Subject: staging: hv: Fixed bounce kmap problem by using correct index Fixed bounce offset kmap problem by using correct index. The symptom of the problem is that in some NAS appliances this problem represents Itself by a unresponsive VM under a load with many clients writing small files. Signed-off-by:Hank Janssen Signed-off-by:Haiyang Zhang Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/hv/storvsc_drv.c b/drivers/staging/hv/storvsc_drv.c index 075b61b..3b9ccb0 100644 --- a/drivers/staging/hv/storvsc_drv.c +++ b/drivers/staging/hv/storvsc_drv.c @@ -495,7 +495,7 @@ static unsigned int copy_to_bounce_buffer(struct scatterlist *orig_sgl, /* ASSERT(orig_sgl[i].offset + orig_sgl[i].length <= PAGE_SIZE); */ - if (j == 0) + if (bounce_addr == 0) bounce_addr = (unsigned long)kmap_atomic(sg_page((&bounce_sgl[j])), KM_IRQ0); while (srclen) { @@ -556,7 +556,7 @@ static unsigned int copy_from_bounce_buffer(struct scatterlist *orig_sgl, destlen = orig_sgl[i].length; /* ASSERT(orig_sgl[i].offset + orig_sgl[i].length <= PAGE_SIZE); */ - if (j == 0) + if (bounce_addr == 0) bounce_addr = (unsigned long)kmap_atomic(sg_page((&bounce_sgl[j])), KM_IRQ0); while (destlen) { -- cgit v0.10.2 From e5fa721d1c2a54261a37eb59686e18dee34b6af6 Mon Sep 17 00:00:00 2001 From: Haiyang Zhang Date: Thu, 5 Aug 2010 19:30:01 +0000 Subject: staging: hv: Fixed the value of the 64bit-hole inside ring buffer Fixed the value of the 64bit-hole inside ring buffer, this caused a problem on Hyper-V when running checked Windows builds. Checked builds of Windows are used internally and given to external system integrators at times. They are builds that for example that all elements in a structure follow the definition of that Structure. The bug this fixed was for a field that we did not fill in at all (Because we do Not use it on the Linux side), and the checked build of windows gives errors on it internally to the Windows logs. This fixes that error. Signed-off-by:Hank Janssen Signed-off-by:Haiyang Zhang Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/hv/ring_buffer.c b/drivers/staging/hv/ring_buffer.c index 17bc762..d78c569 100644 --- a/drivers/staging/hv/ring_buffer.c +++ b/drivers/staging/hv/ring_buffer.c @@ -193,8 +193,7 @@ Description: static inline u64 GetRingBufferIndices(struct hv_ring_buffer_info *RingInfo) { - return ((u64)RingInfo->RingBuffer->WriteIndex << 32) - || RingInfo->RingBuffer->ReadIndex; + return (u64)RingInfo->RingBuffer->WriteIndex << 32; } -- cgit v0.10.2 From 15dd1c9f53b31cdc84b8072a88c23fa09527c596 Mon Sep 17 00:00:00 2001 From: Hank Janssen Date: Thu, 5 Aug 2010 19:30:31 +0000 Subject: staging: hv: Increased storvsc ringbuffer and max_io_requests Increased storvsc ringbuffer and max_io_requests. This now more closely mimics the numbers on Hyper-V. And will allow more IO requests to take place for the SCSI driver. Max_IO is set to double from what it was before, Hyper-V allows it and we have had appliance builder requests to see if it was a problem to increase the number. Ringbuffer size for storvsc is now increased because I have seen A few buffer problems on extremely busy systems. They were Set pretty low before. And since max_io_requests is increased I Really needed to increase the buffer as well. Signed-off-by:Hank Janssen Signed-off-by:Haiyang Zhang Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/hv/storvsc_api.h b/drivers/staging/hv/storvsc_api.h index 0063bde..8505a1c 100644 --- a/drivers/staging/hv/storvsc_api.h +++ b/drivers/staging/hv/storvsc_api.h @@ -28,10 +28,10 @@ #include "vmbus_api.h" /* Defines */ -#define STORVSC_RING_BUFFER_SIZE (10*PAGE_SIZE) +#define STORVSC_RING_BUFFER_SIZE (20*PAGE_SIZE) #define BLKVSC_RING_BUFFER_SIZE (20*PAGE_SIZE) -#define STORVSC_MAX_IO_REQUESTS 64 +#define STORVSC_MAX_IO_REQUESTS 128 /* * In Hyper-V, each port/path/target maps to 1 scsi host adapter. In -- cgit v0.10.2 From 77c5ceaff31645ea049c6706b99e699eae81fb88 Mon Sep 17 00:00:00 2001 From: Hank Janssen Date: Wed, 1 Sep 2010 11:10:41 -0700 Subject: staging: hv: Fixed lockup problem with bounce_buffer scatter list Fixed lockup problem with bounce_buffer scatter list which caused crashes in heavy loads. And minor code indentation cleanup in effected area. Removed whitespace and noted minor indentation changes in description as pointed out by Joe Perches. (Thanks for reviewing Joe) Signed-off-by: Hank Janssen Signed-off-by: Haiyang Zhang Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/hv/storvsc_drv.c b/drivers/staging/hv/storvsc_drv.c index 3b9ccb0..62882a4 100644 --- a/drivers/staging/hv/storvsc_drv.c +++ b/drivers/staging/hv/storvsc_drv.c @@ -615,6 +615,7 @@ static int storvsc_queuecommand(struct scsi_cmnd *scmnd, unsigned int request_size = 0; int i; struct scatterlist *sgl; + unsigned int sg_count = 0; DPRINT_DBG(STORVSC_DRV, "scmnd %p dir %d, use_sg %d buf %p len %d " "queue depth %d tagged %d", scmnd, scmnd->sc_data_direction, @@ -697,6 +698,7 @@ static int storvsc_queuecommand(struct scsi_cmnd *scmnd, request->DataBuffer.Length = scsi_bufflen(scmnd); if (scsi_sg_count(scmnd)) { sgl = (struct scatterlist *)scsi_sglist(scmnd); + sg_count = scsi_sg_count(scmnd); /* check if we need to bounce the sgl */ if (do_bounce_buffer(sgl, scsi_sg_count(scmnd)) != -1) { @@ -731,15 +733,16 @@ static int storvsc_queuecommand(struct scsi_cmnd *scmnd, scsi_sg_count(scmnd)); sgl = cmd_request->bounce_sgl; + sg_count = cmd_request->bounce_sgl_count; } request->DataBuffer.Offset = sgl[0].offset; - for (i = 0; i < scsi_sg_count(scmnd); i++) { + for (i = 0; i < sg_count; i++) { DPRINT_DBG(STORVSC_DRV, "sgl[%d] len %d offset %d\n", i, sgl[i].length, sgl[i].offset); request->DataBuffer.PfnArray[i] = - page_to_pfn(sg_page((&sgl[i]))); + page_to_pfn(sg_page((&sgl[i]))); } } else if (scsi_sglist(scmnd)) { /* ASSERT(scsi_bufflen(scmnd) <= PAGE_SIZE); */ -- cgit v0.10.2 From f8d261d39a24f9f612cb8b2d5ad68654727543cd Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Mon, 30 Aug 2010 08:57:49 +0200 Subject: staging: spectra needs MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit On one of my m68k test builds I get: drivers/staging/spectra/ffsport.c: In function ‘ioctl_read_page_data’: drivers/staging/spectra/ffsport.c:196: error: implicit declaration of function ‘kmalloc’ drivers/staging/spectra/ffsport.c:196: warning: assignment makes pointer from integer without a cast drivers/staging/spectra/ffsport.c:212: error: implicit declaration of function ‘kfree’ drivers/staging/spectra/ffsport.c: In function ‘ioctl_write_page_data’: drivers/staging/spectra/ffsport.c:229: warning: assignment makes pointer from integer without a cast drivers/staging/spectra/ffsport.c: In function ‘SBD_setup_device’: drivers/staging/spectra/ffsport.c:637: warning: assignment makes pointer from integer without a cast Signed-off-by: Geert Uytterhoeven Acked-by: Pekka Enberg Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/spectra/ffsport.c b/drivers/staging/spectra/ffsport.c index 44a7fbe..fa21a0f 100644 --- a/drivers/staging/spectra/ffsport.c +++ b/drivers/staging/spectra/ffsport.c @@ -28,6 +28,7 @@ #include #include #include +#include /**** Helper functions used for Div, Remainder operation on u64 ****/ -- cgit v0.10.2 From 10022a0675a8f8722068d956a798f1fddb02e71c Mon Sep 17 00:00:00 2001 From: Andreas Bombe Date: Sat, 14 Aug 2010 03:24:22 +0200 Subject: staging: comedi das08_cs.c: Fix io_req_t conversion Commit 90abdc3b9 converted all PCMCIA users away from io_req_t. In das08_cs.c the converted IO lines mask setting was added but the old line using the now inexistent p_dev->io was not removed. Signed-off-by: Andreas Bombe Signed-off-by: Dominik Brodowski Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/comedi/drivers/das08_cs.c b/drivers/staging/comedi/drivers/das08_cs.c index c6aa52f..48d9fb1 100644 --- a/drivers/staging/comedi/drivers/das08_cs.c +++ b/drivers/staging/comedi/drivers/das08_cs.c @@ -222,7 +222,6 @@ static int das08_pcmcia_config_loop(struct pcmcia_device *p_dev, p_dev->resource[0]->flags &= ~IO_DATA_PATH_WIDTH; p_dev->resource[0]->flags |= pcmcia_io_cfg_data_width(io->flags); - p_dev->io.IOAddrLines = io->flags & CISTPL_IO_LINES_MASK; p_dev->resource[0]->start = io->win[0].base; p_dev->resource[0]->end = io->win[0].len; if (io->nwin > 1) { -- cgit v0.10.2 From 9e693e4375689cb1cd1529aba011de0044f74ef5 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Sun, 29 Aug 2010 02:13:11 +0100 Subject: Staging: rt2870sta: Add more device IDs from vendor drivers Taken from DPO_RT3070_LinuxSTA_V2.3.0.4_20100604.tar.bz2 and 2010_0709_RT2870_Linux_STA_v2.4.0.1.tar.bz2, with duplicates removed. Signed-off-by: Ben Hutchings Cc: stable Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/rt2860/usb_main_dev.c b/drivers/staging/rt2860/usb_main_dev.c index a0fe31d..ebf9074 100644 --- a/drivers/staging/rt2860/usb_main_dev.c +++ b/drivers/staging/rt2860/usb_main_dev.c @@ -44,6 +44,7 @@ struct usb_device_id rtusb_usb_id[] = { {USB_DEVICE(0x07B8, 0x2870)}, /* AboCom */ {USB_DEVICE(0x07B8, 0x2770)}, /* AboCom */ {USB_DEVICE(0x0DF6, 0x0039)}, /* Sitecom 2770 */ + {USB_DEVICE(0x0DF6, 0x003F)}, /* Sitecom 2770 */ {USB_DEVICE(0x083A, 0x7512)}, /* Arcadyan 2770 */ {USB_DEVICE(0x0789, 0x0162)}, /* Logitec 2870 */ {USB_DEVICE(0x0789, 0x0163)}, /* Logitec 2870 */ @@ -95,7 +96,8 @@ struct usb_device_id rtusb_usb_id[] = { {USB_DEVICE(0x050d, 0x815c)}, {USB_DEVICE(0x1482, 0x3C09)}, /* Abocom */ {USB_DEVICE(0x14B2, 0x3C09)}, /* Alpha */ - {USB_DEVICE(0x04E8, 0x2018)}, /* samsung */ + {USB_DEVICE(0x04E8, 0x2018)}, /* samsung linkstick2 */ + {USB_DEVICE(0x1690, 0x0740)}, /* Askey */ {USB_DEVICE(0x5A57, 0x0280)}, /* Zinwell */ {USB_DEVICE(0x5A57, 0x0282)}, /* Zinwell */ {USB_DEVICE(0x7392, 0x7718)}, @@ -105,21 +107,34 @@ struct usb_device_id rtusb_usb_id[] = { {USB_DEVICE(0x1737, 0x0071)}, /* Linksys WUSB600N */ {USB_DEVICE(0x0411, 0x00e8)}, /* Buffalo WLI-UC-G300N */ {USB_DEVICE(0x050d, 0x815c)}, /* Belkin F5D8053 */ + {USB_DEVICE(0x100D, 0x9031)}, /* Motorola 2770 */ #endif /* RT2870 // */ #ifdef RT3070 {USB_DEVICE(0x148F, 0x3070)}, /* Ralink 3070 */ {USB_DEVICE(0x148F, 0x3071)}, /* Ralink 3071 */ {USB_DEVICE(0x148F, 0x3072)}, /* Ralink 3072 */ {USB_DEVICE(0x0DB0, 0x3820)}, /* Ralink 3070 */ + {USB_DEVICE(0x0DB0, 0x871C)}, /* Ralink 3070 */ + {USB_DEVICE(0x0DB0, 0x822C)}, /* Ralink 3070 */ + {USB_DEVICE(0x0DB0, 0x871B)}, /* Ralink 3070 */ + {USB_DEVICE(0x0DB0, 0x822B)}, /* Ralink 3070 */ {USB_DEVICE(0x0DF6, 0x003E)}, /* Sitecom 3070 */ {USB_DEVICE(0x0DF6, 0x0042)}, /* Sitecom 3072 */ + {USB_DEVICE(0x0DF6, 0x0048)}, /* Sitecom 3070 */ + {USB_DEVICE(0x0DF6, 0x0047)}, /* Sitecom 3071 */ {USB_DEVICE(0x14B2, 0x3C12)}, /* AL 3070 */ {USB_DEVICE(0x18C5, 0x0012)}, /* Corega 3070 */ {USB_DEVICE(0x083A, 0x7511)}, /* Arcadyan 3070 */ + {USB_DEVICE(0x083A, 0xA701)}, /* SMC 3070 */ + {USB_DEVICE(0x083A, 0xA702)}, /* SMC 3072 */ {USB_DEVICE(0x1740, 0x9703)}, /* EnGenius 3070 */ {USB_DEVICE(0x1740, 0x9705)}, /* EnGenius 3071 */ {USB_DEVICE(0x1740, 0x9706)}, /* EnGenius 3072 */ + {USB_DEVICE(0x1740, 0x9707)}, /* EnGenius 3070 */ + {USB_DEVICE(0x1740, 0x9708)}, /* EnGenius 3071 */ + {USB_DEVICE(0x1740, 0x9709)}, /* EnGenius 3072 */ {USB_DEVICE(0x13D3, 0x3273)}, /* AzureWave 3070 */ + {USB_DEVICE(0x13D3, 0x3305)}, /* AzureWave 3070*/ {USB_DEVICE(0x1044, 0x800D)}, /* Gigabyte GN-WB32L 3070 */ {USB_DEVICE(0x2019, 0xAB25)}, /* Planex Communications, Inc. RT3070 */ {USB_DEVICE(0x07B8, 0x3070)}, /* AboCom 3070 */ @@ -132,14 +147,36 @@ struct usb_device_id rtusb_usb_id[] = { {USB_DEVICE(0x07D1, 0x3C0D)}, /* D-Link 3070 */ {USB_DEVICE(0x07D1, 0x3C0E)}, /* D-Link 3070 */ {USB_DEVICE(0x07D1, 0x3C0F)}, /* D-Link 3070 */ + {USB_DEVICE(0x07D1, 0x3C16)}, /* D-Link 3070 */ + {USB_DEVICE(0x07D1, 0x3C17)}, /* D-Link 8070 */ {USB_DEVICE(0x1D4D, 0x000C)}, /* Pegatron Corporation 3070 */ {USB_DEVICE(0x1D4D, 0x000E)}, /* Pegatron Corporation 3070 */ {USB_DEVICE(0x5A57, 0x5257)}, /* Zinwell 3070 */ {USB_DEVICE(0x5A57, 0x0283)}, /* Zinwell 3072 */ {USB_DEVICE(0x04BB, 0x0945)}, /* I-O DATA 3072 */ + {USB_DEVICE(0x04BB, 0x0947)}, /* I-O DATA 3070 */ + {USB_DEVICE(0x04BB, 0x0948)}, /* I-O DATA 3072 */ {USB_DEVICE(0x203D, 0x1480)}, /* Encore 3070 */ + {USB_DEVICE(0x20B8, 0x8888)}, /* PARA INDUSTRIAL 3070 */ + {USB_DEVICE(0x0B05, 0x1784)}, /* Asus 3072 */ + {USB_DEVICE(0x203D, 0x14A9)}, /* Encore 3070*/ + {USB_DEVICE(0x0DB0, 0x899A)}, /* MSI 3070*/ + {USB_DEVICE(0x0DB0, 0x3870)}, /* MSI 3070*/ + {USB_DEVICE(0x0DB0, 0x870A)}, /* MSI 3070*/ + {USB_DEVICE(0x0DB0, 0x6899)}, /* MSI 3070 */ + {USB_DEVICE(0x0DB0, 0x3822)}, /* MSI 3070 */ + {USB_DEVICE(0x0DB0, 0x3871)}, /* MSI 3070 */ + {USB_DEVICE(0x0DB0, 0x871A)}, /* MSI 3070 */ + {USB_DEVICE(0x0DB0, 0x822A)}, /* MSI 3070 */ + {USB_DEVICE(0x0DB0, 0x3821)}, /* Ralink 3070 */ + {USB_DEVICE(0x0DB0, 0x821A)}, /* Ralink 3070 */ + {USB_DEVICE(0x083A, 0xA703)}, /* IO-MAGIC */ + {USB_DEVICE(0x13D3, 0x3307)}, /* Azurewave */ + {USB_DEVICE(0x13D3, 0x3321)}, /* Azurewave */ + {USB_DEVICE(0x07FA, 0x7712)}, /* Edimax */ + {USB_DEVICE(0x0789, 0x0166)}, /* Edimax */ + {USB_DEVICE(0x148F, 0x2070)}, /* Edimax */ #endif /* RT3070 // */ - {USB_DEVICE(0x0DF6, 0x003F)}, /* Sitecom WL-608 */ {USB_DEVICE(0x1737, 0x0077)}, /* Linksys WUSB54GC-EU v3 */ {USB_DEVICE(0x2001, 0x3C09)}, /* D-Link */ {USB_DEVICE(0x2001, 0x3C0A)}, /* D-Link 3072 */ -- cgit v0.10.2 From 273ad8dcef345cac55a5db910137c10953f81480 Mon Sep 17 00:00:00 2001 From: Shahar Havivi Date: Sat, 28 Aug 2010 10:09:05 +0300 Subject: Staging: zram: free device memory when init fails Signed-off-by: Shahar Havivi Cc: Nitin Gupta Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/zram/zram_drv.c b/drivers/staging/zram/zram_drv.c index 77d4d71..722c840 100644 --- a/drivers/staging/zram/zram_drv.c +++ b/drivers/staging/zram/zram_drv.c @@ -769,6 +769,7 @@ static int __init zram_init(void) free_devices: while (dev_id) destroy_device(&devices[--dev_id]); + kfree(devices); unregister: unregister_blkdev(zram_major, "zram"); out: -- cgit v0.10.2 From 11ac33a5f29a8e96ec76097ec1b3de3a4045f4b6 Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Tue, 24 Aug 2010 12:12:26 -0400 Subject: Staging: spectra: depend on X86_MRST lld_nand fails to build on arches without virt_to_bus. Since this driver is specifically for hardware enablment on Moorestown, this patch adds Moorestown MID support as a dependency. Signed-off-by: Jeff Mahoney Cc: David Woodhouse Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/spectra/Kconfig b/drivers/staging/spectra/Kconfig index 5e2ffef..d231ae2 100644 --- a/drivers/staging/spectra/Kconfig +++ b/drivers/staging/spectra/Kconfig @@ -2,6 +2,7 @@ menuconfig SPECTRA tristate "Denali Spectra Flash Translation Layer" depends on BLOCK + depends on X86_MRST default n ---help--- Enable the FTL pseudo-filesystem used with the NAND Flash -- cgit v0.10.2 From 95c46ed9c99f9a3fc4954fc1b7cc1509a359efbb Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Tue, 24 Aug 2010 14:09:40 -0700 Subject: Staging: octeon: depends on NETDEVICES OCTEON_ETHERNET should depend on NETDEVICES. Fixes this kconfig warning: warning: (NET_DSA && NET && EXPERIMENTAL && NETDEVICES && !S390 || ... || OCTEON_ETHERNET && STAGING && !STAGING_EXCLUDE_BUILD && CPU_CAVIUM_OCTEON) selects PHYLIB which has unmet direct dependencies (!S390 && NETDEVICES) Reported-by: Arnaud Lacombe Signed-off-by: Randy Dunlap Cc: support@caviumnetworks.com Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/octeon/Kconfig b/drivers/staging/octeon/Kconfig index 638ad6b..9493128 100644 --- a/drivers/staging/octeon/Kconfig +++ b/drivers/staging/octeon/Kconfig @@ -1,6 +1,6 @@ config OCTEON_ETHERNET tristate "Cavium Networks Octeon Ethernet support" - depends on CPU_CAVIUM_OCTEON + depends on CPU_CAVIUM_OCTEON && NETDEVICES select PHYLIB select MDIO_OCTEON help -- cgit v0.10.2 From aff3ea4e5d4b0280d1c631fcce048e7f009bc3e5 Mon Sep 17 00:00:00 2001 From: Karl Relton Date: Wed, 11 Aug 2010 18:16:08 +0100 Subject: Staging: wlan-ng: Explicitly set some fields in cfg80211 interface The cfg80211 api has introduced a few new fields. Rather than assume what cfg80211 api does by default, set these explicitly. Signed-off-by: Karl Relton Signed-off-by: Greg Kroah-Hartman diff --git a/drivers/staging/wlan-ng/cfg80211.c b/drivers/staging/wlan-ng/cfg80211.c index 368c30a..4af83d5 100644 --- a/drivers/staging/wlan-ng/cfg80211.c +++ b/drivers/staging/wlan-ng/cfg80211.c @@ -219,6 +219,7 @@ int prism2_get_key(struct wiphy *wiphy, struct net_device *dev, return -ENOENT; params.key_len = len; params.key = wlandev->wep_keys[key_index]; + params.seq_len = 0; callback(cookie, ¶ms); @@ -735,6 +736,8 @@ struct wiphy *wlan_create_wiphy(struct device *dev, wlandevice_t *wlandev) priv->band.n_channels = ARRAY_SIZE(prism2_channels); priv->band.bitrates = priv->rates; priv->band.n_bitrates = ARRAY_SIZE(prism2_rates); + priv->band.band = IEEE80211_BAND_2GHZ; + priv->band.ht_cap.ht_supported = false; wiphy->bands[IEEE80211_BAND_2GHZ] = &priv->band; set_wiphy_dev(wiphy, dev); -- cgit v0.10.2 From cc1a8e523333bc3d990102140f4d12908fa83caf Mon Sep 17 00:00:00 2001 From: Francisco Jerez Date: Sat, 4 Sep 2010 22:56:43 +0200 Subject: x86: Fix the address space annotations of iomap_atomic_prot_pfn() This patch fixes the sparse warnings when the return pointer of iomap_atomic_prot_pfn() is used as an argument of iowrite32() and friends. Signed-off-by: Francisco Jerez LKML-Reference: <1283633804-11749-1-git-send-email-currojerez@riseup.net> Cc: Andrew Morton Signed-off-by: Ingo Molnar diff --git a/arch/x86/include/asm/iomap.h b/arch/x86/include/asm/iomap.h index f35eb45..c4191b3 100644 --- a/arch/x86/include/asm/iomap.h +++ b/arch/x86/include/asm/iomap.h @@ -26,11 +26,11 @@ #include #include -void * +void __iomem * iomap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot); void -iounmap_atomic(void *kvaddr, enum km_type type); +iounmap_atomic(void __iomem *kvaddr, enum km_type type); int iomap_create_wc(resource_size_t base, unsigned long size, pgprot_t *prot); diff --git a/arch/x86/mm/iomap_32.c b/arch/x86/mm/iomap_32.c index 84e236c..72fc70c 100644 --- a/arch/x86/mm/iomap_32.c +++ b/arch/x86/mm/iomap_32.c @@ -74,7 +74,7 @@ void *kmap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot) /* * Map 'pfn' using fixed map 'type' and protections 'prot' */ -void * +void __iomem * iomap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot) { /* @@ -86,12 +86,12 @@ iomap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot) if (!pat_enabled && pgprot_val(prot) == pgprot_val(PAGE_KERNEL_WC)) prot = PAGE_KERNEL_UC_MINUS; - return kmap_atomic_prot_pfn(pfn, type, prot); + return (void __force __iomem *) kmap_atomic_prot_pfn(pfn, type, prot); } EXPORT_SYMBOL_GPL(iomap_atomic_prot_pfn); void -iounmap_atomic(void *kvaddr, enum km_type type) +iounmap_atomic(void __iomem *kvaddr, enum km_type type) { unsigned long vaddr = (unsigned long) kvaddr & PAGE_MASK; enum fixed_addresses idx = type + KM_TYPE_NR*smp_processor_id(); -- cgit v0.10.2 From 29bc17ecb856ffb2b47c7009a71971c6f9334205 Mon Sep 17 00:00:00 2001 From: Francisco Jerez Date: Sat, 4 Sep 2010 22:56:44 +0200 Subject: io-mapping: Fix the address space annotations Fixes a bunch of sparse warnings in io-mapping.h because of the inconsistent __iomem usage. Signed-off-by: Francisco Jerez LKML-Reference: <1283633804-11749-2-git-send-email-currojerez@riseup.net> Cc: Andrew Morton Signed-off-by: Ingo Molnar diff --git a/include/linux/io-mapping.h b/include/linux/io-mapping.h index 0a6b3d5..7fb5927 100644 --- a/include/linux/io-mapping.h +++ b/include/linux/io-mapping.h @@ -79,7 +79,7 @@ io_mapping_free(struct io_mapping *mapping) } /* Atomic map/unmap */ -static inline void * +static inline void __iomem * io_mapping_map_atomic_wc(struct io_mapping *mapping, unsigned long offset, int slot) @@ -94,12 +94,12 @@ io_mapping_map_atomic_wc(struct io_mapping *mapping, } static inline void -io_mapping_unmap_atomic(void *vaddr, int slot) +io_mapping_unmap_atomic(void __iomem *vaddr, int slot) { iounmap_atomic(vaddr, slot); } -static inline void * +static inline void __iomem * io_mapping_map_wc(struct io_mapping *mapping, unsigned long offset) { resource_size_t phys_addr; @@ -111,7 +111,7 @@ io_mapping_map_wc(struct io_mapping *mapping, unsigned long offset) } static inline void -io_mapping_unmap(void *vaddr) +io_mapping_unmap(void __iomem *vaddr) { iounmap(vaddr); } @@ -125,38 +125,38 @@ struct io_mapping; static inline struct io_mapping * io_mapping_create_wc(resource_size_t base, unsigned long size) { - return (struct io_mapping *) ioremap_wc(base, size); + return (struct io_mapping __force *) ioremap_wc(base, size); } static inline void io_mapping_free(struct io_mapping *mapping) { - iounmap(mapping); + iounmap((void __force __iomem *) mapping); } /* Atomic map/unmap */ -static inline void * +static inline void __iomem * io_mapping_map_atomic_wc(struct io_mapping *mapping, unsigned long offset, int slot) { - return ((char *) mapping) + offset; + return ((char __force __iomem *) mapping) + offset; } static inline void -io_mapping_unmap_atomic(void *vaddr, int slot) +io_mapping_unmap_atomic(void __iomem *vaddr, int slot) { } /* Non-atomic map/unmap */ -static inline void * +static inline void __iomem * io_mapping_map_wc(struct io_mapping *mapping, unsigned long offset) { - return ((char *) mapping) + offset; + return ((char __force __iomem *) mapping) + offset; } static inline void -io_mapping_unmap(void *vaddr) +io_mapping_unmap(void __iomem *vaddr) { } -- cgit v0.10.2 From 1389298f7d14f179c2fa4b4b343079bd56082c70 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Fri, 27 Aug 2010 11:20:07 +0200 Subject: x86, mcheck: Avoid duplicate sysfs links/files for thresholding banks kobject_add_internal failed for threshold_bank2 with -EEXIST, don't try to register things with the same name in the same directory: Pid: 1, comm: swapper Tainted: G W 2.6.31 #1 Call Trace: [] ? kobject_add_internal+0x156/0x180 [] ? kobject_add+0x66/0x6b [] ? kobject_init+0x42/0x82 [] ? kobject_create_and_add+0x34/0x63 [] ? threshold_create_bank+0x14f/0x259 [] ? mce_create_device+0x8d/0x1b8 [] ? threshold_init_device+0x3f/0x80 [] ? threshold_init_device+0x0/0x80 [] ? do_one_initcall+0x4f/0x143 [] ? kernel_init+0x14c/0x1a2 [] ? child_rip+0xa/0x20 [] ? kernel_init+0x0/0x1a2 [] ? child_rip+0x0/0x20 kobject_create_and_add: kobject_add error: -17 (Probably the for_each_cpu loop should be entirely removed.) Signed-off-by: Andreas Herrmann LKML-Reference: <20100827092006.GB5348@loge.amd.com> Signed-off-by: Ingo Molnar diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd.c b/arch/x86/kernel/cpu/mcheck/mce_amd.c index 224392d..5e97529 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_amd.c +++ b/arch/x86/kernel/cpu/mcheck/mce_amd.c @@ -530,7 +530,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank) err = -ENOMEM; goto out; } - if (!alloc_cpumask_var(&b->cpus, GFP_KERNEL)) { + if (!zalloc_cpumask_var(&b->cpus, GFP_KERNEL)) { kfree(b); err = -ENOMEM; goto out; @@ -543,7 +543,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank) #ifndef CONFIG_SMP cpumask_setall(b->cpus); #else - cpumask_copy(b->cpus, c->llc_shared_map); + cpumask_set_cpu(cpu, b->cpus); #endif per_cpu(threshold_banks, cpu)[bank] = b; -- cgit v0.10.2 From b3bd3de66f60df4c9a2076e2886a622458929056 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 10 Aug 2010 14:17:51 -0700 Subject: gcc-4.6: kernel/*: Fix unused but set warnings No real bugs I believe, just some dead code. Signed-off-by: Andi Kleen Cc: Peter Zijlstra Cc: andi@firstfloor.org Signed-off-by: Andrew Morton Signed-off-by: Ingo Molnar diff --git a/kernel/debug/kdb/kdb_bp.c b/kernel/debug/kdb/kdb_bp.c index 75bd9b3..20059ef 100644 --- a/kernel/debug/kdb/kdb_bp.c +++ b/kernel/debug/kdb/kdb_bp.c @@ -274,7 +274,6 @@ static int kdb_bp(int argc, const char **argv) int i, bpno; kdb_bp_t *bp, *bp_check; int diag; - int free; char *symname = NULL; long offset = 0ul; int nextarg; @@ -305,7 +304,6 @@ static int kdb_bp(int argc, const char **argv) /* * Find an empty bp structure to allocate */ - free = KDB_MAXBPT; for (bpno = 0, bp = kdb_breakpoints; bpno < KDB_MAXBPT; bpno++, bp++) { if (bp->bp_free) break; diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index ce66917..1decafb 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -1091,11 +1091,10 @@ EXPORT_SYMBOL_GPL(hrtimer_cancel); */ ktime_t hrtimer_get_remaining(const struct hrtimer *timer) { - struct hrtimer_clock_base *base; unsigned long flags; ktime_t rem; - base = lock_hrtimer_base(timer, &flags); + lock_hrtimer_base(timer, &flags); rem = hrtimer_expires_remaining(timer); unlock_hrtimer_base(timer, &flags); diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index ab661eb..134f7ed 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1313,7 +1313,7 @@ static struct sched_group * find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu, int load_idx) { - struct sched_group *idlest = NULL, *this = NULL, *group = sd->groups; + struct sched_group *idlest = NULL, *group = sd->groups; unsigned long min_load = ULONG_MAX, this_load = 0; int imbalance = 100 + (sd->imbalance_pct-100)/2; @@ -1348,7 +1348,6 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, if (local_group) { this_load = avg_load; - this = group; } else if (avg_load < min_load) { min_load = avg_load; idlest = group; diff --git a/kernel/sysctl.c b/kernel/sysctl.c index ca38e8e..f88552c 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -1713,10 +1713,7 @@ static __init int sysctl_init(void) { sysctl_set_parent(NULL, root_table); #ifdef CONFIG_SYSCTL_SYSCALL_CHECK - { - int err; - err = sysctl_check_table(current->nsproxy, root_table); - } + sysctl_check_table(current->nsproxy, root_table); #endif return 0; } diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 19cccc3..492197e 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -2985,13 +2985,11 @@ static void rb_advance_reader(struct ring_buffer_per_cpu *cpu_buffer) static void rb_advance_iter(struct ring_buffer_iter *iter) { - struct ring_buffer *buffer; struct ring_buffer_per_cpu *cpu_buffer; struct ring_buffer_event *event; unsigned length; cpu_buffer = iter->cpu_buffer; - buffer = cpu_buffer->buffer; /* * Check if we are at the end of the buffer. -- cgit v0.10.2 From 73457f0f836956747e0394320be2163c050e96ef Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Fri, 6 Aug 2010 01:59:14 +0300 Subject: cgroups: fix API thinko cgroup_attach_task_current_cg API that have upstream is backwards: we really need an API to attach to the cgroups from another process A to the current one. In our case (vhost), a priveledged user wants to attach it's task to cgroups from a less priveledged one, the API makes us run it in the other task's context, and this fails. So let's make the API generic and just pass in 'from' and 'to' tasks. Add an inline wrapper for cgroup_attach_task_current_cg to avoid breaking bisect. Signed-off-by: Michael S. Tsirkin Acked-by: Li Zefan Acked-by: Paul Menage diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index ed3e92e..5a53d8f 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -578,7 +578,11 @@ struct task_struct *cgroup_iter_next(struct cgroup *cgrp, void cgroup_iter_end(struct cgroup *cgrp, struct cgroup_iter *it); int cgroup_scan_tasks(struct cgroup_scanner *scan); int cgroup_attach_task(struct cgroup *, struct task_struct *); -int cgroup_attach_task_current_cg(struct task_struct *); +int cgroup_attach_task_all(struct task_struct *from, struct task_struct *); +static inline int cgroup_attach_task_current_cg(struct task_struct *tsk) +{ + return cgroup_attach_task_all(current, tsk); +} /* * CSS ID is ID for cgroup_subsys_state structs under subsys. This only works @@ -636,6 +640,11 @@ static inline int cgroupstats_build(struct cgroupstats *stats, } /* No cgroups - nothing to do */ +static inline int cgroup_attach_task_all(struct task_struct *from, + struct task_struct *t) +{ + return 0; +} static inline int cgroup_attach_task_current_cg(struct task_struct *t) { return 0; diff --git a/kernel/cgroup.c b/kernel/cgroup.c index a8ce099..9d90c08 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -1789,10 +1789,11 @@ out: } /** - * cgroup_attach_task_current_cg - attach task 'tsk' to current task's cgroup + * cgroup_attach_task_all - attach task 'tsk' to all cgroups of task 'from' + * @from: attach to all cgroups of a given task * @tsk: the task to be attached */ -int cgroup_attach_task_current_cg(struct task_struct *tsk) +int cgroup_attach_task_all(struct task_struct *from, struct task_struct *tsk) { struct cgroupfs_root *root; struct cgroup *cur_cg; @@ -1800,7 +1801,7 @@ int cgroup_attach_task_current_cg(struct task_struct *tsk) cgroup_lock(); for_each_active_root(root) { - cur_cg = task_cgroup_from_root(current, root); + cur_cg = task_cgroup_from_root(from, root); retval = cgroup_attach_task(cur_cg, tsk); if (retval) break; @@ -1809,7 +1810,7 @@ int cgroup_attach_task_current_cg(struct task_struct *tsk) return retval; } -EXPORT_SYMBOL_GPL(cgroup_attach_task_current_cg); +EXPORT_SYMBOL_GPL(cgroup_attach_task_all); /* * Attach task with pid 'pid' to cgroup 'cgrp'. Call with cgroup_mutex -- cgit v0.10.2 From feafb7b1714cf599a6d0fed45801ab3f66046cbd Mon Sep 17 00:00:00 2001 From: Arun Easi Date: Fri, 3 Sep 2010 14:57:00 -0700 Subject: [SCSI] qla2xxx: Fix vport delete issues Signed-off-by: Giridhar Malavali Signed-off-by: James Bottomley diff --git a/drivers/scsi/qla2xxx/qla_attr.c b/drivers/scsi/qla2xxx/qla_attr.c index 420238c..114bc5a 100644 --- a/drivers/scsi/qla2xxx/qla_attr.c +++ b/drivers/scsi/qla2xxx/qla_attr.c @@ -1838,26 +1838,33 @@ qla24xx_vport_delete(struct fc_vport *fc_vport) qla24xx_disable_vp(vha); + vha->flags.delete_progress = 1; + fc_remove_host(vha->host); scsi_remove_host(vha->host); - qla2x00_free_fcports(vha); + if (vha->timer_active) { + qla2x00_vp_stop_timer(vha); + DEBUG15(printk(KERN_INFO "scsi(%ld): timer for the vport[%d]" + " = %p has stopped\n", vha->host_no, vha->vp_idx, vha)); + } qla24xx_deallocate_vp_id(vha); + /* No pending activities shall be there on the vha now */ + DEBUG(msleep(random32()%10)); /* Just to see if something falls on + * the net we have placed below */ + + BUG_ON(atomic_read(&vha->vref_count)); + + qla2x00_free_fcports(vha); + mutex_lock(&ha->vport_lock); ha->cur_vport_count--; clear_bit(vha->vp_idx, ha->vp_idx_map); mutex_unlock(&ha->vport_lock); - if (vha->timer_active) { - qla2x00_vp_stop_timer(vha); - DEBUG15(printk ("scsi(%ld): timer for the vport[%d] = %p " - "has stopped\n", - vha->host_no, vha->vp_idx, vha)); - } - if (vha->req->id && !ha->flags.cpu_affinity_enabled) { if (qla25xx_delete_req_que(vha, vha->req) != QLA_SUCCESS) qla_printk(KERN_WARNING, ha, diff --git a/drivers/scsi/qla2xxx/qla_dbg.h b/drivers/scsi/qla2xxx/qla_dbg.h index 6cfc28a..b74e6b5 100644 --- a/drivers/scsi/qla2xxx/qla_dbg.h +++ b/drivers/scsi/qla2xxx/qla_dbg.h @@ -29,8 +29,6 @@ /* #define QL_DEBUG_LEVEL_17 */ /* Output EEH trace messages */ /* #define QL_DEBUG_LEVEL_18 */ /* Output T10 CRC trace messages */ -/* #define QL_PRINTK_BUF */ /* Captures printk to buffer */ - /* * Macros use for debugging the driver. */ diff --git a/drivers/scsi/qla2xxx/qla_def.h b/drivers/scsi/qla2xxx/qla_def.h index 3a432ea..d2a4e15 100644 --- a/drivers/scsi/qla2xxx/qla_def.h +++ b/drivers/scsi/qla2xxx/qla_def.h @@ -2641,6 +2641,7 @@ struct qla_hw_data { #define MBX_UPDATE_FLASH_ACTIVE 3 struct mutex vport_lock; /* Virtual port synchronization */ + spinlock_t vport_slock; /* order is hardware_lock, then vport_slock */ struct completion mbx_cmd_comp; /* Serialize mbx access */ struct completion mbx_intr_comp; /* Used for completion notification */ struct completion dcbx_comp; /* For set port config notification */ @@ -2828,6 +2829,7 @@ typedef struct scsi_qla_host { uint32_t management_server_logged_in :1; uint32_t process_response_queue :1; uint32_t difdix_supported:1; + uint32_t delete_progress:1; } flags; atomic_t loop_state; @@ -2922,6 +2924,8 @@ typedef struct scsi_qla_host { struct req_que *req; int fw_heartbeat_counter; int seconds_since_last_heartbeat; + + atomic_t vref_count; } scsi_qla_host_t; /* @@ -2932,6 +2936,22 @@ typedef struct scsi_qla_host { test_bit(LOOP_RESYNC_NEEDED, &ha->dpc_flags) || \ atomic_read(&ha->loop_state) == LOOP_DOWN) +#define QLA_VHA_MARK_BUSY(__vha, __bail) do { \ + atomic_inc(&__vha->vref_count); \ + mb(); \ + if (__vha->flags.delete_progress) { \ + atomic_dec(&__vha->vref_count); \ + __bail = 1; \ + } else { \ + __bail = 0; \ + } \ +} while (0) + +#define QLA_VHA_MARK_NOT_BUSY(__vha) do { \ + atomic_dec(&__vha->vref_count); \ +} while (0) + + #define qla_printk(level, ha, format, arg...) \ dev_printk(level , &((ha)->pdev->dev) , format , ## arg) diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c index d863ed2..9c383ba 100644 --- a/drivers/scsi/qla2xxx/qla_init.c +++ b/drivers/scsi/qla2xxx/qla_init.c @@ -69,21 +69,29 @@ qla2x00_ctx_sp_free(srb_t *sp) { struct srb_ctx *ctx = sp->ctx; struct srb_iocb *iocb = ctx->u.iocb_cmd; + struct scsi_qla_host *vha = sp->fcport->vha; del_timer_sync(&iocb->timer); kfree(iocb); kfree(ctx); mempool_free(sp, sp->fcport->vha->hw->srb_mempool); + + QLA_VHA_MARK_NOT_BUSY(vha); } inline srb_t * qla2x00_get_ctx_sp(scsi_qla_host_t *vha, fc_port_t *fcport, size_t size, unsigned long tmo) { - srb_t *sp; + srb_t *sp = NULL; struct qla_hw_data *ha = vha->hw; struct srb_ctx *ctx; struct srb_iocb *iocb; + uint8_t bail; + + QLA_VHA_MARK_BUSY(vha, bail); + if (bail) + return NULL; sp = mempool_alloc(ha->srb_mempool, GFP_KERNEL); if (!sp) @@ -116,6 +124,8 @@ qla2x00_get_ctx_sp(scsi_qla_host_t *vha, fc_port_t *fcport, size_t size, iocb->timer.function = qla2x00_ctx_sp_timeout; add_timer(&iocb->timer); done: + if (!sp) + QLA_VHA_MARK_NOT_BUSY(vha); return sp; } @@ -1777,11 +1787,15 @@ qla2x00_init_rings(scsi_qla_host_t *vha) qla2x00_init_response_q_entries(rsp); } + spin_lock_irqsave(&ha->vport_slock, flags); /* Clear RSCN queue. */ list_for_each_entry(vp, &ha->vp_list, list) { vp->rscn_in_ptr = 0; vp->rscn_out_ptr = 0; } + + spin_unlock_irqrestore(&ha->vport_slock, flags); + ha->isp_ops->config_rings(vha); spin_unlock_irqrestore(&ha->hardware_lock, flags); @@ -3218,12 +3232,17 @@ qla2x00_find_all_fabric_devs(scsi_qla_host_t *vha, /* Bypass virtual ports of the same host. */ found = 0; if (ha->num_vhosts) { + unsigned long flags; + + spin_lock_irqsave(&ha->vport_slock, flags); list_for_each_entry_safe(vp, tvp, &ha->vp_list, list) { if (new_fcport->d_id.b24 == vp->d_id.b24) { found = 1; break; } } + spin_unlock_irqrestore(&ha->vport_slock, flags); + if (found) continue; } @@ -3343,6 +3362,7 @@ qla2x00_find_new_loop_id(scsi_qla_host_t *vha, fc_port_t *dev) struct qla_hw_data *ha = vha->hw; struct scsi_qla_host *vp; struct scsi_qla_host *tvp; + unsigned long flags = 0; rval = QLA_SUCCESS; @@ -3367,6 +3387,8 @@ qla2x00_find_new_loop_id(scsi_qla_host_t *vha, fc_port_t *dev) /* Check for loop ID being already in use. */ found = 0; fcport = NULL; + + spin_lock_irqsave(&ha->vport_slock, flags); list_for_each_entry_safe(vp, tvp, &ha->vp_list, list) { list_for_each_entry(fcport, &vp->vp_fcports, list) { if (fcport->loop_id == dev->loop_id && @@ -3379,6 +3401,7 @@ qla2x00_find_new_loop_id(scsi_qla_host_t *vha, fc_port_t *dev) if (found) break; } + spin_unlock_irqrestore(&ha->vport_slock, flags); /* If not in use then it is free to use. */ if (!found) { @@ -3791,14 +3814,27 @@ void qla2x00_update_fcports(scsi_qla_host_t *base_vha) { fc_port_t *fcport; - struct scsi_qla_host *tvp, *vha; + struct scsi_qla_host *vha; + struct qla_hw_data *ha = base_vha->hw; + unsigned long flags; + spin_lock_irqsave(&ha->vport_slock, flags); /* Go with deferred removal of rport references. */ - list_for_each_entry_safe(vha, tvp, &base_vha->hw->vp_list, list) - list_for_each_entry(fcport, &vha->vp_fcports, list) + list_for_each_entry(vha, &base_vha->hw->vp_list, list) { + atomic_inc(&vha->vref_count); + list_for_each_entry(fcport, &vha->vp_fcports, list) { if (fcport && fcport->drport && - atomic_read(&fcport->state) != FCS_UNCONFIGURED) + atomic_read(&fcport->state) != FCS_UNCONFIGURED) { + spin_unlock_irqrestore(&ha->vport_slock, flags); + qla2x00_rport_del(fcport); + + spin_lock_irqsave(&ha->vport_slock, flags); + } + } + atomic_dec(&vha->vref_count); + } + spin_unlock_irqrestore(&ha->vport_slock, flags); } void @@ -3806,7 +3842,7 @@ qla2x00_abort_isp_cleanup(scsi_qla_host_t *vha) { struct qla_hw_data *ha = vha->hw; struct scsi_qla_host *vp, *base_vha = pci_get_drvdata(ha->pdev); - struct scsi_qla_host *tvp; + unsigned long flags; vha->flags.online = 0; ha->flags.chip_reset_done = 0; @@ -3824,8 +3860,18 @@ qla2x00_abort_isp_cleanup(scsi_qla_host_t *vha) if (atomic_read(&vha->loop_state) != LOOP_DOWN) { atomic_set(&vha->loop_state, LOOP_DOWN); qla2x00_mark_all_devices_lost(vha, 0); - list_for_each_entry_safe(vp, tvp, &base_vha->hw->vp_list, list) + + spin_lock_irqsave(&ha->vport_slock, flags); + list_for_each_entry(vp, &base_vha->hw->vp_list, list) { + atomic_inc(&vp->vref_count); + spin_unlock_irqrestore(&ha->vport_slock, flags); + qla2x00_mark_all_devices_lost(vp, 0); + + spin_lock_irqsave(&ha->vport_slock, flags); + atomic_dec(&vp->vref_count); + } + spin_unlock_irqrestore(&ha->vport_slock, flags); } else { if (!atomic_read(&vha->loop_down_timer)) atomic_set(&vha->loop_down_timer, @@ -3862,8 +3908,8 @@ qla2x00_abort_isp(scsi_qla_host_t *vha) uint8_t status = 0; struct qla_hw_data *ha = vha->hw; struct scsi_qla_host *vp; - struct scsi_qla_host *tvp; struct req_que *req = ha->req_q_map[0]; + unsigned long flags; if (vha->flags.online) { qla2x00_abort_isp_cleanup(vha); @@ -3970,10 +4016,21 @@ qla2x00_abort_isp(scsi_qla_host_t *vha) DEBUG(printk(KERN_INFO "qla2x00_abort_isp(%ld): succeeded.\n", vha->host_no)); - list_for_each_entry_safe(vp, tvp, &ha->vp_list, list) { - if (vp->vp_idx) + + spin_lock_irqsave(&ha->vport_slock, flags); + list_for_each_entry(vp, &ha->vp_list, list) { + if (vp->vp_idx) { + atomic_inc(&vp->vref_count); + spin_unlock_irqrestore(&ha->vport_slock, flags); + qla2x00_vp_abort_isp(vp); + + spin_lock_irqsave(&ha->vport_slock, flags); + atomic_dec(&vp->vref_count); + } } + spin_unlock_irqrestore(&ha->vport_slock, flags); + } else { qla_printk(KERN_INFO, ha, "qla2x00_abort_isp: **** FAILED ****\n"); @@ -5185,7 +5242,7 @@ qla82xx_restart_isp(scsi_qla_host_t *vha) struct req_que *req = ha->req_q_map[0]; struct rsp_que *rsp = ha->rsp_q_map[0]; struct scsi_qla_host *vp; - struct scsi_qla_host *tvp; + unsigned long flags; status = qla2x00_init_rings(vha); if (!status) { @@ -5272,10 +5329,21 @@ qla82xx_restart_isp(scsi_qla_host_t *vha) DEBUG(printk(KERN_INFO "qla82xx_restart_isp(%ld): succeeded.\n", vha->host_no)); - list_for_each_entry_safe(vp, tvp, &ha->vp_list, list) { - if (vp->vp_idx) + + spin_lock_irqsave(&ha->vport_slock, flags); + list_for_each_entry(vp, &ha->vp_list, list) { + if (vp->vp_idx) { + atomic_inc(&vp->vref_count); + spin_unlock_irqrestore(&ha->vport_slock, flags); + qla2x00_vp_abort_isp(vp); + + spin_lock_irqsave(&ha->vport_slock, flags); + atomic_dec(&vp->vref_count); + } } + spin_unlock_irqrestore(&ha->vport_slock, flags); + } else { qla_printk(KERN_INFO, ha, "qla82xx_restart_isp: **** FAILED ****\n"); diff --git a/drivers/scsi/qla2xxx/qla_mbx.c b/drivers/scsi/qla2xxx/qla_mbx.c index 6009b0c..a595ec8 100644 --- a/drivers/scsi/qla2xxx/qla_mbx.c +++ b/drivers/scsi/qla2xxx/qla_mbx.c @@ -2913,7 +2913,7 @@ qla24xx_report_id_acquisition(scsi_qla_host_t *vha, uint16_t stat = le16_to_cpu(rptid_entry->vp_idx); struct qla_hw_data *ha = vha->hw; scsi_qla_host_t *vp; - scsi_qla_host_t *tvp; + unsigned long flags; if (rptid_entry->entry_status != 0) return; @@ -2945,9 +2945,12 @@ qla24xx_report_id_acquisition(scsi_qla_host_t *vha, return; } - list_for_each_entry_safe(vp, tvp, &ha->vp_list, list) + spin_lock_irqsave(&ha->vport_slock, flags); + list_for_each_entry(vp, &ha->vp_list, list) if (vp_idx == vp->vp_idx) break; + spin_unlock_irqrestore(&ha->vport_slock, flags); + if (!vp) return; diff --git a/drivers/scsi/qla2xxx/qla_mid.c b/drivers/scsi/qla2xxx/qla_mid.c index 987c5b0..bc1a745 100644 --- a/drivers/scsi/qla2xxx/qla_mid.c +++ b/drivers/scsi/qla2xxx/qla_mid.c @@ -30,6 +30,7 @@ qla24xx_allocate_vp_id(scsi_qla_host_t *vha) { uint32_t vp_id; struct qla_hw_data *ha = vha->hw; + unsigned long flags; /* Find an empty slot and assign an vp_id */ mutex_lock(&ha->vport_lock); @@ -44,7 +45,11 @@ qla24xx_allocate_vp_id(scsi_qla_host_t *vha) set_bit(vp_id, ha->vp_idx_map); ha->num_vhosts++; vha->vp_idx = vp_id; + + spin_lock_irqsave(&ha->vport_slock, flags); list_add_tail(&vha->list, &ha->vp_list); + spin_unlock_irqrestore(&ha->vport_slock, flags); + mutex_unlock(&ha->vport_lock); return vp_id; } @@ -54,12 +59,31 @@ qla24xx_deallocate_vp_id(scsi_qla_host_t *vha) { uint16_t vp_id; struct qla_hw_data *ha = vha->hw; + unsigned long flags = 0; mutex_lock(&ha->vport_lock); + /* + * Wait for all pending activities to finish before removing vport from + * the list. + * Lock needs to be held for safe removal from the list (it + * ensures no active vp_list traversal while the vport is removed + * from the queue) + */ + spin_lock_irqsave(&ha->vport_slock, flags); + while (atomic_read(&vha->vref_count)) { + spin_unlock_irqrestore(&ha->vport_slock, flags); + + msleep(500); + + spin_lock_irqsave(&ha->vport_slock, flags); + } + list_del(&vha->list); + spin_unlock_irqrestore(&ha->vport_slock, flags); + vp_id = vha->vp_idx; ha->num_vhosts--; clear_bit(vp_id, ha->vp_idx_map); - list_del(&vha->list); + mutex_unlock(&ha->vport_lock); } @@ -68,12 +92,17 @@ qla24xx_find_vhost_by_name(struct qla_hw_data *ha, uint8_t *port_name) { scsi_qla_host_t *vha; struct scsi_qla_host *tvha; + unsigned long flags; + spin_lock_irqsave(&ha->vport_slock, flags); /* Locate matching device in database. */ list_for_each_entry_safe(vha, tvha, &ha->vp_list, list) { - if (!memcmp(port_name, vha->port_name, WWN_SIZE)) + if (!memcmp(port_name, vha->port_name, WWN_SIZE)) { + spin_unlock_irqrestore(&ha->vport_slock, flags); return vha; + } } + spin_unlock_irqrestore(&ha->vport_slock, flags); return NULL; } @@ -93,6 +122,12 @@ qla24xx_find_vhost_by_name(struct qla_hw_data *ha, uint8_t *port_name) static void qla2x00_mark_vp_devices_dead(scsi_qla_host_t *vha) { + /* + * !!! NOTE !!! + * This function, if called in contexts other than vp create, disable + * or delete, please make sure this is synchronized with the + * delete thread. + */ fc_port_t *fcport; list_for_each_entry(fcport, &vha->vp_fcports, list) { @@ -194,12 +229,17 @@ qla24xx_configure_vp(scsi_qla_host_t *vha) void qla2x00_alert_all_vps(struct rsp_que *rsp, uint16_t *mb) { - scsi_qla_host_t *vha, *tvha; + scsi_qla_host_t *vha; struct qla_hw_data *ha = rsp->hw; int i = 0; + unsigned long flags; - list_for_each_entry_safe(vha, tvha, &ha->vp_list, list) { + spin_lock_irqsave(&ha->vport_slock, flags); + list_for_each_entry(vha, &ha->vp_list, list) { if (vha->vp_idx) { + atomic_inc(&vha->vref_count); + spin_unlock_irqrestore(&ha->vport_slock, flags); + switch (mb[0]) { case MBA_LIP_OCCURRED: case MBA_LOOP_UP: @@ -215,9 +255,13 @@ qla2x00_alert_all_vps(struct rsp_que *rsp, uint16_t *mb) qla2x00_async_event(vha, rsp, mb); break; } + + spin_lock_irqsave(&ha->vport_slock, flags); + atomic_dec(&vha->vref_count); } i++; } + spin_unlock_irqrestore(&ha->vport_slock, flags); } int @@ -297,7 +341,7 @@ qla2x00_do_dpc_all_vps(scsi_qla_host_t *vha) int ret; struct qla_hw_data *ha = vha->hw; scsi_qla_host_t *vp; - struct scsi_qla_host *tvp; + unsigned long flags = 0; if (vha->vp_idx) return; @@ -309,10 +353,19 @@ qla2x00_do_dpc_all_vps(scsi_qla_host_t *vha) if (!(ha->current_topology & ISP_CFG_F)) return; - list_for_each_entry_safe(vp, tvp, &ha->vp_list, list) { - if (vp->vp_idx) + spin_lock_irqsave(&ha->vport_slock, flags); + list_for_each_entry(vp, &ha->vp_list, list) { + if (vp->vp_idx) { + atomic_inc(&vp->vref_count); + spin_unlock_irqrestore(&ha->vport_slock, flags); + ret = qla2x00_do_dpc_vp(vp); + + spin_lock_irqsave(&ha->vport_slock, flags); + atomic_dec(&vp->vref_count); + } } + spin_unlock_irqrestore(&ha->vport_slock, flags); } int diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index 8c80b49..1e4bff6 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -2341,16 +2341,28 @@ probe_out: static void qla2x00_remove_one(struct pci_dev *pdev) { - scsi_qla_host_t *base_vha, *vha, *temp; + scsi_qla_host_t *base_vha, *vha; struct qla_hw_data *ha; + unsigned long flags; base_vha = pci_get_drvdata(pdev); ha = base_vha->hw; - list_for_each_entry_safe(vha, temp, &ha->vp_list, list) { - if (vha && vha->fc_vport) + spin_lock_irqsave(&ha->vport_slock, flags); + list_for_each_entry(vha, &ha->vp_list, list) { + atomic_inc(&vha->vref_count); + + if (vha && vha->fc_vport) { + spin_unlock_irqrestore(&ha->vport_slock, flags); + fc_vport_terminate(vha->fc_vport); + + spin_lock_irqsave(&ha->vport_slock, flags); + } + + atomic_dec(&vha->vref_count); } + spin_unlock_irqrestore(&ha->vport_slock, flags); set_bit(UNLOADING, &base_vha->dpc_flags); @@ -2975,10 +2987,17 @@ static struct qla_work_evt * qla2x00_alloc_work(struct scsi_qla_host *vha, enum qla_work_type type) { struct qla_work_evt *e; + uint8_t bail; + + QLA_VHA_MARK_BUSY(vha, bail); + if (bail) + return NULL; e = kzalloc(sizeof(struct qla_work_evt), GFP_ATOMIC); - if (!e) + if (!e) { + QLA_VHA_MARK_NOT_BUSY(vha); return NULL; + } INIT_LIST_HEAD(&e->list); e->type = type; @@ -3135,6 +3154,9 @@ qla2x00_do_work(struct scsi_qla_host *vha) } if (e->flags & QLA_EVT_FLAG_FREE) kfree(e); + + /* For each work completed decrement vha ref count */ + QLA_VHA_MARK_NOT_BUSY(vha); } } -- cgit v0.10.2 From 970ee0c52a41cf27c1b5c346dd9475e9c236f3c5 Mon Sep 17 00:00:00 2001 From: Arun Easi Date: Fri, 3 Sep 2010 14:57:01 -0700 Subject: [SCSI] qla2xxx: make rport deletions explicit during vport removal Signed-off-by: Giridhar Malavali Signed-off-by: James Bottomley diff --git a/drivers/scsi/qla2xxx/qla_mid.c b/drivers/scsi/qla2xxx/qla_mid.c index bc1a745..2b69392 100644 --- a/drivers/scsi/qla2xxx/qla_mid.c +++ b/drivers/scsi/qla2xxx/qla_mid.c @@ -135,7 +135,6 @@ qla2x00_mark_vp_devices_dead(scsi_qla_host_t *vha) "loop_id=0x%04x :%x\n", vha->host_no, fcport->loop_id, fcport->vp_idx)); - atomic_set(&fcport->state, FCS_DEVICE_DEAD); qla2x00_mark_device_lost(vha, fcport, 0, 0); atomic_set(&fcport->state, FCS_UNCONFIGURED); } -- cgit v0.10.2 From efa786cc43a114d0bf2e4b95e856ea6911404d58 Mon Sep 17 00:00:00 2001 From: Lalit Chandivade Date: Fri, 3 Sep 2010 14:57:02 -0700 Subject: [SCSI] qla2xxx: Reset seconds_since_last_heartbeat correctly. The seconds_since_last_heartbeat should be checked for consecutive heartbeat checks. Currently it could happen that seconds_since_last_heartbeat gets set to max (2 seconds) for non-consecutive heartbeat checks. Signed-off-by: Giridhar Malavali Signed-off-by: James Bottomley diff --git a/drivers/scsi/qla2xxx/qla_nx.c b/drivers/scsi/qla2xxx/qla_nx.c index 915b77a..e54d2fa 100644 --- a/drivers/scsi/qla2xxx/qla_nx.c +++ b/drivers/scsi/qla2xxx/qla_nx.c @@ -3316,7 +3316,8 @@ qla82xx_check_fw_alive(scsi_qla_host_t *vha) complete(&ha->mbx_intr_comp); } } - } + } else + vha->seconds_since_last_heartbeat = 0; vha->fw_heartbeat_counter = fw_heartbeat_counter; } -- cgit v0.10.2 From 4142b1987f1f8ba90589642cb74566eaff3dc2e9 Mon Sep 17 00:00:00 2001 From: Giridhar Malavali Date: Fri, 3 Sep 2010 14:57:03 -0700 Subject: [SCSI] qla2xxx: Correctly set fw hung and complete only waiting mbx. The fw_hung flag should be set ir-respective of if there is a mbx command pending or not. Also the complete should be called if there is a mbx waiting. Signed-off-by: Giridhar Malavali Signed-off-by: James Bottomley diff --git a/drivers/scsi/qla2xxx/qla_nx.c b/drivers/scsi/qla2xxx/qla_nx.c index e54d2fa..ad290dc 100644 --- a/drivers/scsi/qla2xxx/qla_nx.c +++ b/drivers/scsi/qla2xxx/qla_nx.c @@ -3307,13 +3307,15 @@ qla82xx_check_fw_alive(scsi_qla_host_t *vha) set_bit(ISP_ABORT_NEEDED, &vha->dpc_flags); } qla2xxx_wake_dpc(vha); + ha->flags.fw_hung = 1; if (ha->flags.mbox_busy) { - ha->flags.fw_hung = 1; ha->flags.mbox_int = 1; DEBUG2(qla_printk(KERN_ERR, ha, - "Due to fw hung, doing premature " - "completion of mbx command\n")); - complete(&ha->mbx_intr_comp); + "Due to fw hung, doing premature " + "completion of mbx command\n")); + if (test_bit(MBX_INTR_WAIT, + &ha->mbx_cmd_flags)) + complete(&ha->mbx_intr_comp); } } } else @@ -3419,13 +3421,15 @@ void qla82xx_watchdog(scsi_qla_host_t *vha) "%s(): Adapter reset needed!\n", __func__); set_bit(ISP_ABORT_NEEDED, &vha->dpc_flags); qla2xxx_wake_dpc(vha); + ha->flags.fw_hung = 1; if (ha->flags.mbox_busy) { - ha->flags.fw_hung = 1; ha->flags.mbox_int = 1; DEBUG2(qla_printk(KERN_ERR, ha, - "Need reset, doing premature " - "completion of mbx command\n")); - complete(&ha->mbx_intr_comp); + "Need reset, doing premature " + "completion of mbx command\n")); + if (test_bit(MBX_INTR_WAIT, + &ha->mbx_cmd_flags)) + complete(&ha->mbx_intr_comp); } } else { qla82xx_check_fw_alive(vha); -- cgit v0.10.2 From 0374f55ed882a46cd4825dde16ca2392d4c367f6 Mon Sep 17 00:00:00 2001 From: Lalit Chandivade Date: Fri, 3 Sep 2010 14:57:04 -0700 Subject: [SCSI] qla2xxx: Cover UNDERRUN case where SCSI status is set. Currently, if target sets the SCSI Status (with Check condition) and there is no FCP residual bit set then driver does not check for dropped frame. This could lead to data corruption. Signed-off-by: Giridhar Malavali Signed-off-by: James Bottomley diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c index 6982ba7..28f65be 100644 --- a/drivers/scsi/qla2xxx/qla_isr.c +++ b/drivers/scsi/qla2xxx/qla_isr.c @@ -1706,19 +1706,20 @@ qla2x00_status_entry(scsi_qla_host_t *vha, struct rsp_que *rsp, void *pkt) cp->result = DID_ERROR << 16; break; } - } else if (!lscsi_status) { + } else { DEBUG2(qla_printk(KERN_INFO, ha, "scsi(%ld:%d:%d) Dropped frame(s) detected (0x%x " "of 0x%x bytes).\n", vha->host_no, cp->device->id, cp->device->lun, resid, scsi_bufflen(cp))); - cp->result = DID_ERROR << 16; - break; + cp->result = DID_ERROR << 16 | lscsi_status; + goto check_scsi_status; } cp->result = DID_OK << 16 | lscsi_status; logit = 0; +check_scsi_status: /* * Check to see if SCSI Status is non zero. If so report SCSI * Status. -- cgit v0.10.2 From 1bd58b89e84b15283aaa3148fee4969abe19af8d Mon Sep 17 00:00:00 2001 From: Giridhar Malavali Date: Fri, 3 Sep 2010 14:57:05 -0700 Subject: [SCSI] qla2xxx: Check for empty slot in request queue before posting Command type 6 request. For ISP82xx, the check for empty slot in request queue before posting command type 6 request was missing. This could lead to request queue entry corruptions causing IO timeouts. Signed-off-by: Giridhar Malavali Signed-off-by: James Bottomley diff --git a/drivers/scsi/qla2xxx/qla_nx.c b/drivers/scsi/qla2xxx/qla_nx.c index ad290dc..0a71cc7 100644 --- a/drivers/scsi/qla2xxx/qla_nx.c +++ b/drivers/scsi/qla2xxx/qla_nx.c @@ -2672,6 +2672,19 @@ qla82xx_start_scsi(srb_t *sp) sufficient_dsds: req_cnt = 1; + if (req->cnt < (req_cnt + 2)) { + cnt = (uint16_t)RD_REG_DWORD_RELAXED( + ®->req_q_out[0]); + if (req->ring_index < cnt) + req->cnt = cnt - req->ring_index; + else + req->cnt = req->length - + (req->ring_index - cnt); + } + + if (req->cnt < (req_cnt + 2)) + goto queuing_error; + ctx = sp->ctx = mempool_alloc(ha->ctx_mempool, GFP_ATOMIC); if (!sp->ctx) { DEBUG(printk(KERN_INFO -- cgit v0.10.2 From 0fb576d8251c10f498ed4c6938aeeed8d0c93cfe Mon Sep 17 00:00:00 2001 From: Madhuranath Iyengar Date: Fri, 3 Sep 2010 14:57:06 -0700 Subject: [SCSI] qla2xxx: Update version number to 8.03.04-k0. Signed-off-by: Giridhar Malavali Signed-off-by: James Bottomley diff --git a/drivers/scsi/qla2xxx/qla_version.h b/drivers/scsi/qla2xxx/qla_version.h index e75ccb9..8edbccb 100644 --- a/drivers/scsi/qla2xxx/qla_version.h +++ b/drivers/scsi/qla2xxx/qla_version.h @@ -7,9 +7,9 @@ /* * Driver version */ -#define QLA2XXX_VERSION "8.03.03-k0" +#define QLA2XXX_VERSION "8.03.04-k0" #define QLA_DRIVER_MAJOR_VER 8 #define QLA_DRIVER_MINOR_VER 3 -#define QLA_DRIVER_PATCH_VER 3 +#define QLA_DRIVER_PATCH_VER 4 #define QLA_DRIVER_BETA_VER 0 -- cgit v0.10.2 From 87d6a412bd1ed82c14cabd4b408003b23bbd2880 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Thu, 2 Sep 2010 14:05:30 +0300 Subject: vhost: fix attach to cgroups regression Since 2.6.36-rc1, non-root users of vhost-net fail to attach if they are in any cgroups. The reason is that when qemu uses vhost, vhost wants to attach its thread to all cgroups that qemu has. But we got the API backwards, so a non-priveledged process (Qemu) tried to control the priveledged one (vhost), which fails. Fix this by switching to the new cgroup_attach_task_all, and running it from the vhost thread. Signed-off-by: Michael S. Tsirkin diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index 4b99117..1afa085 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -60,22 +60,25 @@ static int vhost_poll_wakeup(wait_queue_t *wait, unsigned mode, int sync, return 0; } +static void vhost_work_init(struct vhost_work *work, vhost_work_fn_t fn) +{ + INIT_LIST_HEAD(&work->node); + work->fn = fn; + init_waitqueue_head(&work->done); + work->flushing = 0; + work->queue_seq = work->done_seq = 0; +} + /* Init poll structure */ void vhost_poll_init(struct vhost_poll *poll, vhost_work_fn_t fn, unsigned long mask, struct vhost_dev *dev) { - struct vhost_work *work = &poll->work; - init_waitqueue_func_entry(&poll->wait, vhost_poll_wakeup); init_poll_funcptr(&poll->table, vhost_poll_func); poll->mask = mask; poll->dev = dev; - INIT_LIST_HEAD(&work->node); - work->fn = fn; - init_waitqueue_head(&work->done); - work->flushing = 0; - work->queue_seq = work->done_seq = 0; + vhost_work_init(&poll->work, fn); } /* Start polling a file. We add ourselves to file's wait queue. The caller must @@ -95,35 +98,38 @@ void vhost_poll_stop(struct vhost_poll *poll) remove_wait_queue(poll->wqh, &poll->wait); } -/* Flush any work that has been scheduled. When calling this, don't hold any - * locks that are also used by the callback. */ -void vhost_poll_flush(struct vhost_poll *poll) +static void vhost_work_flush(struct vhost_dev *dev, struct vhost_work *work) { - struct vhost_work *work = &poll->work; unsigned seq; int left; int flushing; - spin_lock_irq(&poll->dev->work_lock); + spin_lock_irq(&dev->work_lock); seq = work->queue_seq; work->flushing++; - spin_unlock_irq(&poll->dev->work_lock); + spin_unlock_irq(&dev->work_lock); wait_event(work->done, ({ - spin_lock_irq(&poll->dev->work_lock); + spin_lock_irq(&dev->work_lock); left = seq - work->done_seq <= 0; - spin_unlock_irq(&poll->dev->work_lock); + spin_unlock_irq(&dev->work_lock); left; })); - spin_lock_irq(&poll->dev->work_lock); + spin_lock_irq(&dev->work_lock); flushing = --work->flushing; - spin_unlock_irq(&poll->dev->work_lock); + spin_unlock_irq(&dev->work_lock); BUG_ON(flushing < 0); } -void vhost_poll_queue(struct vhost_poll *poll) +/* Flush any work that has been scheduled. When calling this, don't hold any + * locks that are also used by the callback. */ +void vhost_poll_flush(struct vhost_poll *poll) +{ + vhost_work_flush(poll->dev, &poll->work); +} + +static inline void vhost_work_queue(struct vhost_dev *dev, + struct vhost_work *work) { - struct vhost_dev *dev = poll->dev; - struct vhost_work *work = &poll->work; unsigned long flags; spin_lock_irqsave(&dev->work_lock, flags); @@ -135,6 +141,11 @@ void vhost_poll_queue(struct vhost_poll *poll) spin_unlock_irqrestore(&dev->work_lock, flags); } +void vhost_poll_queue(struct vhost_poll *poll) +{ + vhost_work_queue(poll->dev, &poll->work); +} + static void vhost_vq_reset(struct vhost_dev *dev, struct vhost_virtqueue *vq) { @@ -236,6 +247,29 @@ long vhost_dev_check_owner(struct vhost_dev *dev) return dev->mm == current->mm ? 0 : -EPERM; } +struct vhost_attach_cgroups_struct { + struct vhost_work work; + struct task_struct *owner; + int ret; +}; + +static void vhost_attach_cgroups_work(struct vhost_work *work) +{ + struct vhost_attach_cgroups_struct *s; + s = container_of(work, struct vhost_attach_cgroups_struct, work); + s->ret = cgroup_attach_task_all(s->owner, current); +} + +static int vhost_attach_cgroups(struct vhost_dev *dev) +{ + struct vhost_attach_cgroups_struct attach; + attach.owner = current; + vhost_work_init(&attach.work, vhost_attach_cgroups_work); + vhost_work_queue(dev, &attach.work); + vhost_work_flush(dev, &attach.work); + return attach.ret; +} + /* Caller should have device mutex */ static long vhost_dev_set_owner(struct vhost_dev *dev) { @@ -255,10 +289,11 @@ static long vhost_dev_set_owner(struct vhost_dev *dev) } dev->worker = worker; - err = cgroup_attach_task_current_cg(worker); + wake_up_process(worker); /* avoid contributing to loadavg */ + + err = vhost_attach_cgroups(dev); if (err) goto err_cgroup; - wake_up_process(worker); /* avoid contributing to loadavg */ return 0; err_cgroup: -- cgit v0.10.2 From 615cc2211c17ed05a2a5d94abdac6c340a8ea508 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Thu, 2 Sep 2010 14:16:36 +0300 Subject: vhost: error handling fix vhost should set worker to NULL on cgroups attach failure, so that we won't try to destroy the worker again on close. Signed-off-by: Michael S. Tsirkin diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index 1afa085..c579dcc 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -298,6 +298,7 @@ static long vhost_dev_set_owner(struct vhost_dev *dev) return 0; err_cgroup: kthread_stop(worker); + dev->worker = NULL; err_worker: if (dev->mm) mmput(dev->mm); -- cgit v0.10.2 From 7e7b41d2ff30ed7ad4bf401d18566e6f38e42e4f Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Thu, 2 Sep 2010 21:32:32 -0400 Subject: drm/radeon/kms/evergreen: fix gpu hangs in userspace accel code These VGT regs need to be programmed via the ring rather than MMIO as on previous asics (r6xx/r7xx). Signed-off-by: Alex Deucher Cc: stable@kernel.org Signed-off-by: Dave Airlie diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index fe31b0d..b8b7f01 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -675,6 +675,43 @@ static int evergreen_cp_load_microcode(struct radeon_device *rdev) return 0; } +static int evergreen_cp_start(struct radeon_device *rdev) +{ + int r; + uint32_t cp_me; + + r = radeon_ring_lock(rdev, 7); + if (r) { + DRM_ERROR("radeon: cp failed to lock ring (%d).\n", r); + return r; + } + radeon_ring_write(rdev, PACKET3(PACKET3_ME_INITIALIZE, 5)); + radeon_ring_write(rdev, 0x1); + radeon_ring_write(rdev, 0x0); + radeon_ring_write(rdev, rdev->config.evergreen.max_hw_contexts - 1); + radeon_ring_write(rdev, PACKET3_ME_INITIALIZE_DEVICE_ID(1)); + radeon_ring_write(rdev, 0); + radeon_ring_write(rdev, 0); + radeon_ring_unlock_commit(rdev); + + cp_me = 0xff; + WREG32(CP_ME_CNTL, cp_me); + + r = radeon_ring_lock(rdev, 4); + if (r) { + DRM_ERROR("radeon: cp failed to lock ring (%d).\n", r); + return r; + } + /* init some VGT regs */ + radeon_ring_write(rdev, PACKET3(PACKET3_SET_CONTEXT_REG, 2)); + radeon_ring_write(rdev, (VGT_VERTEX_REUSE_BLOCK_CNTL - PACKET3_SET_CONTEXT_REG_START) >> 2); + radeon_ring_write(rdev, 0xe); + radeon_ring_write(rdev, 0x10); + radeon_ring_unlock_commit(rdev); + + return 0; +} + int evergreen_cp_resume(struct radeon_device *rdev) { u32 tmp; @@ -719,7 +756,7 @@ int evergreen_cp_resume(struct radeon_device *rdev) rdev->cp.rptr = RREG32(CP_RB_RPTR); rdev->cp.wptr = RREG32(CP_RB_WPTR); - r600_cp_start(rdev); + evergreen_cp_start(rdev); rdev->cp.ready = true; r = radeon_ring_test(rdev); if (r) { diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index 04f134d..afc18d8 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -2119,10 +2119,7 @@ int r600_cp_start(struct radeon_device *rdev) } radeon_ring_write(rdev, PACKET3(PACKET3_ME_INITIALIZE, 5)); radeon_ring_write(rdev, 0x1); - if (rdev->family >= CHIP_CEDAR) { - radeon_ring_write(rdev, 0x0); - radeon_ring_write(rdev, rdev->config.evergreen.max_hw_contexts - 1); - } else if (rdev->family >= CHIP_RV770) { + if (rdev->family >= CHIP_RV770) { radeon_ring_write(rdev, 0x0); radeon_ring_write(rdev, rdev->config.rv770.max_hw_contexts - 1); } else { -- cgit v0.10.2 From 54bfe496cec7586f76f713a277435dd3ac6fd4c4 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Fri, 3 Sep 2010 15:52:53 -0400 Subject: drm/radeon/kms: fix tv-out on avivo asics digital underscan support regressed tv-out. fixes: https://bugs.freedesktop.org/show_bug.cgi?id=29985 Signed-off-by: Alex Deucher Signed-off-by: Dave Airlie diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c index 71b3150..464a81a 100644 --- a/drivers/gpu/drm/radeon/atombios_crtc.c +++ b/drivers/gpu/drm/radeon/atombios_crtc.c @@ -332,6 +332,11 @@ static void atombios_crtc_set_timing(struct drm_crtc *crtc, args.usV_SyncWidth = cpu_to_le16(mode->crtc_vsync_end - mode->crtc_vsync_start); + args.ucOverscanRight = radeon_crtc->h_border; + args.ucOverscanLeft = radeon_crtc->h_border; + args.ucOverscanBottom = radeon_crtc->v_border; + args.ucOverscanTop = radeon_crtc->v_border; + if (mode->flags & DRM_MODE_FLAG_NVSYNC) misc |= ATOM_VSYNC_POLARITY; if (mode->flags & DRM_MODE_FLAG_NHSYNC) @@ -1211,8 +1216,18 @@ int atombios_crtc_mode_set(struct drm_crtc *crtc, struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); struct drm_device *dev = crtc->dev; struct radeon_device *rdev = dev->dev_private; + struct drm_encoder *encoder; + bool is_tvcv = false; - /* TODO color tiling */ + list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { + /* find tv std */ + if (encoder->crtc == crtc) { + struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); + if (radeon_encoder->active_device & + (ATOM_DEVICE_TV_SUPPORT | ATOM_DEVICE_CV_SUPPORT)) + is_tvcv = true; + } + } atombios_disable_ss(crtc); /* always set DCPLL */ @@ -1221,9 +1236,14 @@ int atombios_crtc_mode_set(struct drm_crtc *crtc, atombios_crtc_set_pll(crtc, adjusted_mode); atombios_enable_ss(crtc); - if (ASIC_IS_AVIVO(rdev)) + if (ASIC_IS_DCE4(rdev)) atombios_set_crtc_dtd_timing(crtc, adjusted_mode); - else { + else if (ASIC_IS_AVIVO(rdev)) { + if (is_tvcv) + atombios_crtc_set_timing(crtc, adjusted_mode); + else + atombios_set_crtc_dtd_timing(crtc, adjusted_mode); + } else { atombios_crtc_set_timing(crtc, adjusted_mode); if (radeon_crtc->crtc_id == 0) atombios_set_crtc_dtd_timing(crtc, adjusted_mode); -- cgit v0.10.2 From e58f637bb96d5a0ae0919b9998b891d1ba7e47c9 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Fri, 20 Aug 2010 09:13:36 +0100 Subject: drm/kms: Add a module parameter to disable polling MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Polling for a VGA device on an old system can be quite expensive, causing latencies on the order of 600ms. As we hold the mode mutex for this time and also need the same mutex to move the cursor, we trigger a user-visible stall. The real solution would involve improving the granulatity of the locking and so perhaps performing some of the probing not under the lock or some other updates can be done under different locks. Also reducing the cost of probing for a non-existent monitor would be worthwhile. However, exposing a parameter to disable polling is a simple workaround in the meantime. In order to accommodate users turning polling on and off at runtime, the polling is potentially re-enabled on every probe. This is coupled to the user calling xrandr, which seems to be a vaild time to reset the polling timeout since the information on the connection has just been updated. (The presumption being that all connections are probed in a single xrandr pass, which is currently valid.) References: Bug 29536 - 2.6.35 causes ~600ms latency every 10s https://bugs.freedesktop.org/show_bug.cgi?id=29536 Bug 16265 - Why is kslowd accumulating so much CPU time? https://bugzilla.kernel.org/show_bug.cgi?id=16265 Signed-off-by: Chris Wilson Reported-and-tested-by: Bruno Prémont Signed-off-by: Dave Airlie diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c index 7e31d43..06fc0bc 100644 --- a/drivers/gpu/drm/drm_crtc_helper.c +++ b/drivers/gpu/drm/drm_crtc_helper.c @@ -34,6 +34,9 @@ #include "drm_crtc_helper.h" #include "drm_fb_helper.h" +static bool drm_kms_helper_poll = true; +module_param_named(poll, drm_kms_helper_poll, bool, 0600); + static void drm_mode_validate_flag(struct drm_connector *connector, int flags) { @@ -99,8 +102,10 @@ int drm_helper_probe_single_connector_modes(struct drm_connector *connector, connector->status = connector_status_disconnected; if (connector->funcs->force) connector->funcs->force(connector); - } else + } else { connector->status = connector->funcs->detect(connector); + drm_helper_hpd_irq_event(dev); + } if (connector->status == connector_status_disconnected) { DRM_DEBUG_KMS("[CONNECTOR:%d:%s] disconnected\n", @@ -840,6 +845,9 @@ static void output_poll_execute(struct work_struct *work) enum drm_connector_status old_status, status; bool repoll = false, changed = false; + if (!drm_kms_helper_poll) + return; + mutex_lock(&dev->mode_config.mutex); list_for_each_entry(connector, &dev->mode_config.connector_list, head) { @@ -890,6 +898,9 @@ void drm_kms_helper_poll_enable(struct drm_device *dev) bool poll = false; struct drm_connector *connector; + if (!dev->mode_config.poll_enabled || !drm_kms_helper_poll) + return; + list_for_each_entry(connector, &dev->mode_config.connector_list, head) { if (connector->polled) poll = true; @@ -919,8 +930,10 @@ void drm_helper_hpd_irq_event(struct drm_device *dev) { if (!dev->mode_config.poll_enabled) return; + /* kill timer and schedule immediate execution, this doesn't block */ cancel_delayed_work(&dev->mode_config.output_poll_work); - queue_delayed_work(system_nrt_wq, &dev->mode_config.output_poll_work, 0); + if (drm_kms_helper_poll) + queue_delayed_work(system_nrt_wq, &dev->mode_config.output_poll_work, 0); } EXPORT_SYMBOL(drm_helper_hpd_irq_event); -- cgit v0.10.2 From c7ef35a960369bcad733b92868e4befe03ba9234 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Sun, 5 Sep 2010 16:55:25 +0100 Subject: drm: Do not force 1024x768 modes on unknown connectors Only fallback to a set of default modes on a connector iff that connector is known to be connected. The issue occurs that with limited hardware which cannot probe a connector and so reports the connector status as unknown will then attempt to retrieve the modes for it during drm_helper_probe_single_connector_modes(). Should that fail, the helper then generates a default set which fools the fb_helper and causes havoc with the console and beyond. Signed-off-by: Chris Wilson Signed-off-by: Dave Airlie diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c index 06fc0bc..d2ab01e 100644 --- a/drivers/gpu/drm/drm_crtc_helper.c +++ b/drivers/gpu/drm/drm_crtc_helper.c @@ -115,11 +115,10 @@ int drm_helper_probe_single_connector_modes(struct drm_connector *connector, } count = (*connector_funcs->get_modes)(connector); - if (!count) { + if (count == 0 && connector->status == connector_status_connected) count = drm_add_modes_noedid(connector, 1024, 768); - if (!count) - return 0; - } + if (count == 0) + goto prune; drm_mode_connector_list_update(connector); -- cgit v0.10.2 From e167976ee7f5fe4b80f7e8f55e087f6c67cf9562 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 24 Aug 2010 16:35:52 -0700 Subject: drivers/gpu/drm/i915/intel_overlay.c needs seq_file.h drivers/gpu/drm/i915/intel_overlay.c: In function 'intel_overlay_print_error_state': drivers/gpu/drm/i915/intel_overlay.c:1467: error: implicit declaration of function 'seq_printf' Addresses https://bugzilla.kernel.org/show_bug.cgi?id=16811 Reported-by: Martin Ziegler Cc: Chris Wilson Cc: Daniel Vetter Cc: Eric Anholt Cc: Dave Airlie Cc: Andre Muller Signed-off-by: Andrew Morton Signed-off-by: Chris Wilson diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c index 4f00390..1d306a4 100644 --- a/drivers/gpu/drm/i915/intel_overlay.c +++ b/drivers/gpu/drm/i915/intel_overlay.c @@ -25,6 +25,8 @@ * * Derived from Xorg ddx, xf86-video-intel, src/i830_video.c */ + +#include #include "drmP.h" #include "drm.h" #include "i915_drm.h" -- cgit v0.10.2 From 1dfd9754cd55e424f247d9a2e855ad384e3e90ef Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Mon, 6 Sep 2010 14:44:14 +0100 Subject: Revert "drm/i915: Unreference object not handle on creation" This reverts commit 86f100b136626e91f4f66f3776303475e2e58998. The kref API requires the handlecount to be initialised to one on object creation (so that kref_get() doesn't complain upon first use) so the dalliance in the drivers is required in order to sink the initial floating reference. Signed-off-by: Chris Wilson Cc: stable@kernel.org diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index df5a713..68526f4 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -135,12 +135,15 @@ i915_gem_create_ioctl(struct drm_device *dev, void *data, return -ENOMEM; ret = drm_gem_handle_create(file_priv, obj, &handle); - drm_gem_object_unreference_unlocked(obj); - if (ret) + if (ret) { + drm_gem_object_unreference_unlocked(obj); return ret; + } - args->handle = handle; + /* Sink the floating reference from kref_init(handlecount) */ + drm_gem_object_handle_unreference_unlocked(obj); + args->handle = handle; return 0; } -- cgit v0.10.2 From c74696b9c890074c1e1ee3d7496fc71eb3680ced Mon Sep 17 00:00:00 2001 From: Pavel Roskin Date: Thu, 2 Sep 2010 14:46:34 -0400 Subject: i915: revert some checks added by commit 32aad86f This fixes blur-like screen corruption on the following card: VGA compatible controller [0300]: Intel Corporation 82G33/G31 Express Integrated Graphics Controller [8086:29c2] (rev 10) intel_sdvo_mode_set() should not return prematurely just because some features are not supported. https://bugzilla.kernel.org/show_bug.cgi?id=17151 Signed-off-by: Pavel Roskin Reported-by: Jonathan Corbet Reviewed-by: Chris Wilson [ickle: Relax a couple more checks for failing LVDS modesetting] Signed-off-by: Chris Wilson diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c index 093e914..62d22ae 100644 --- a/drivers/gpu/drm/i915/intel_sdvo.c +++ b/drivers/gpu/drm/i915/intel_sdvo.c @@ -1061,8 +1061,9 @@ static bool intel_sdvo_mode_fixup(struct drm_encoder *encoder, if (!intel_sdvo_set_output_timings_from_mode(intel_sdvo, mode)) return false; - if (!intel_sdvo_set_input_timings_for_mode(intel_sdvo, mode, adjusted_mode)) - return false; + (void) intel_sdvo_set_input_timings_for_mode(intel_sdvo, + mode, + adjusted_mode); } else if (intel_sdvo->is_lvds) { drm_mode_set_crtcinfo(intel_sdvo->sdvo_lvds_fixed_mode, 0); @@ -1070,8 +1071,9 @@ static bool intel_sdvo_mode_fixup(struct drm_encoder *encoder, intel_sdvo->sdvo_lvds_fixed_mode)) return false; - if (!intel_sdvo_set_input_timings_for_mode(intel_sdvo, mode, adjusted_mode)) - return false; + (void) intel_sdvo_set_input_timings_for_mode(intel_sdvo, + mode, + adjusted_mode); } /* Make the CRTC code factor in the SDVO pixel multiplier. The @@ -1108,10 +1110,9 @@ static void intel_sdvo_mode_set(struct drm_encoder *encoder, in_out.in0 = intel_sdvo->attached_output; in_out.in1 = 0; - if (!intel_sdvo_set_value(intel_sdvo, - SDVO_CMD_SET_IN_OUT_MAP, - &in_out, sizeof(in_out))) - return; + intel_sdvo_set_value(intel_sdvo, + SDVO_CMD_SET_IN_OUT_MAP, + &in_out, sizeof(in_out)); if (intel_sdvo->is_hdmi) { if (!intel_sdvo_set_avi_infoframe(intel_sdvo, mode)) @@ -1122,11 +1123,9 @@ static void intel_sdvo_mode_set(struct drm_encoder *encoder, /* We have tried to get input timing in mode_fixup, and filled into adjusted_mode */ - if (intel_sdvo->is_tv || intel_sdvo->is_lvds) { - intel_sdvo_get_dtd_from_mode(&input_dtd, adjusted_mode); + intel_sdvo_get_dtd_from_mode(&input_dtd, adjusted_mode); + if (intel_sdvo->is_tv || intel_sdvo->is_lvds) input_dtd.part2.sdvo_flags = intel_sdvo->sdvo_flags; - } else - intel_sdvo_get_dtd_from_mode(&input_dtd, mode); /* If it's a TV, we already set the output timing in mode_fixup. * Otherwise, the output timing is equal to the input timing. @@ -1137,8 +1136,7 @@ static void intel_sdvo_mode_set(struct drm_encoder *encoder, intel_sdvo->attached_output)) return; - if (!intel_sdvo_set_output_timing(intel_sdvo, &input_dtd)) - return; + (void) intel_sdvo_set_output_timing(intel_sdvo, &input_dtd); } /* Set the input timing to the screen. Assume always input 0. */ @@ -1165,8 +1163,7 @@ static void intel_sdvo_mode_set(struct drm_encoder *encoder, intel_sdvo_set_input_timing(encoder, &input_dtd); } #else - if (!intel_sdvo_set_input_timing(intel_sdvo, &input_dtd)) - return; + (void) intel_sdvo_set_input_timing(intel_sdvo, &input_dtd); #endif sdvo_pixel_multiply = intel_sdvo_get_pixel_multiplier(mode); -- cgit v0.10.2 From 4f233eff6f32745f8894eb513bc59851213c7833 Mon Sep 17 00:00:00 2001 From: Pekka Enberg Date: Sat, 4 Sep 2010 19:24:04 +0300 Subject: i915: Fix spurious TV detection after 9d0498a2bf + 9559fcdbff Partial revert of 9d0498a2bf. Signed-off-by: Pekka Enberg Tested-by: Hugh Dickins Tested-by: Sven Joachim Signed-off-by: Chris Wilson diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c index d2029ef..c671f60 100644 --- a/drivers/gpu/drm/i915/intel_tv.c +++ b/drivers/gpu/drm/i915/intel_tv.c @@ -1231,7 +1231,6 @@ intel_tv_detect_type (struct intel_tv *intel_tv) struct drm_encoder *encoder = &intel_tv->base.enc; struct drm_device *dev = encoder->dev; struct drm_i915_private *dev_priv = dev->dev_private; - struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc); unsigned long irqflags; u32 tv_ctl, save_tv_ctl; u32 tv_dac, save_tv_dac; @@ -1268,11 +1267,15 @@ intel_tv_detect_type (struct intel_tv *intel_tv) DAC_C_0_7_V); I915_WRITE(TV_CTL, tv_ctl); I915_WRITE(TV_DAC, tv_dac); - intel_wait_for_vblank(dev, intel_crtc->pipe); + POSTING_READ(TV_DAC); + msleep(20); + tv_dac = I915_READ(TV_DAC); I915_WRITE(TV_DAC, save_tv_dac); I915_WRITE(TV_CTL, save_tv_ctl); - intel_wait_for_vblank(dev, intel_crtc->pipe); + POSTING_READ(TV_CTL); + msleep(20); + /* * A B C * 0 1 1 Composite -- cgit v0.10.2 From 300387c0b57d75e5218e2881d6ad2720657a8bcf Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Sun, 5 Sep 2010 20:25:43 +0100 Subject: drm/i915: Clear the vblank status bit before polling for the next vblank The vblank status bit is a sticky bit that must be cleared with a write of '1' prior to polling for the next vblank. Signed-off-by: Chris Wilson Tested-by: Sitsofe Wheeler jbarnes: I'd still rather see a lock, but I think you're right that we don't generally wait in code that needs not to miss an interrupt. Reviewed-by: Jesse Barnes diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 11a3394..3fc767b 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -990,6 +990,22 @@ void intel_wait_for_vblank(struct drm_device *dev, int pipe) struct drm_i915_private *dev_priv = dev->dev_private; int pipestat_reg = (pipe == 0 ? PIPEASTAT : PIPEBSTAT); + /* Clear existing vblank status. Note this will clear any other + * sticky status fields as well. + * + * This races with i915_driver_irq_handler() with the result + * that either function could miss a vblank event. Here it is not + * fatal, as we will either wait upon the next vblank interrupt or + * timeout. Generally speaking intel_wait_for_vblank() is only + * called during modeset at which time the GPU should be idle and + * should *not* be performing page flips and thus not waiting on + * vblanks... + * Currently, the result of us stealing a vblank from the irq + * handler is that a single frame will be skipped during swapbuffers. + */ + I915_WRITE(pipestat_reg, + I915_READ(pipestat_reg) | PIPE_VBLANK_INTERRUPT_STATUS); + /* Wait for vblank interrupt bit to set */ if (wait_for((I915_READ(pipestat_reg) & PIPE_VBLANK_INTERRUPT_STATUS), -- cgit v0.10.2 From 9f82d23846146990d475f6753be733e55788d88d Mon Sep 17 00:00:00 2001 From: Daniel Vetter Date: Mon, 30 Aug 2010 21:25:23 +0200 Subject: drm/i915: overlay on gen2 can't address above 1G So set the coherent dma mask accordingly. This dma mask is only used for physical objects, so it won't really matter allocation-wise. Now this never really surfaced because sane 32bit kernels only have 1G of lowmem. But some eager testers (distros?) still carry around the patch to adjust lowmem via a kconfig option. And the kernel seems to favour high allocations on boot-up, hence the overlay blowing up reliably. Because the patch is tiny and nicely shows how broken gen2 is it's imho worth to merge despite the fact that mucking around with the lowmem/ highmem division is (no longer) supported. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=28318 Cc: stable@kernel.org Signed-off-by: Daniel Vetter Signed-off-by: Chris Wilson diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c index a7ec93e..7796f45 100644 --- a/drivers/gpu/drm/i915/i915_dma.c +++ b/drivers/gpu/drm/i915/i915_dma.c @@ -2082,6 +2082,10 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags) goto free_priv; } + /* overlay on gen2 is broken and can't address above 1G */ + if (IS_GEN2(dev)) + dma_set_coherent_mask(&dev->pdev->dev, DMA_BIT_MASK(30)); + dev_priv->regs = ioremap(base, size); if (!dev_priv->regs) { DRM_ERROR("failed to map registers\n"); -- cgit v0.10.2 From df51e7aa2cf204e3a65657a1d60b96cfda133e9b Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Sat, 4 Sep 2010 14:57:27 +0100 Subject: agp/intel: Promote warning about failure to setup flush to error. Make sure we always detect when we fail to correctly allocate the Isoch Flush Page and print an error to warn the user about the likely memory corruption that will result in invalid rendering or worse. Signed-off-by: Chris Wilson Cc: stable@kernel.org diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/intel-gtt.c index d22ffb8..ce536e6 100644 --- a/drivers/char/agp/intel-gtt.c +++ b/drivers/char/agp/intel-gtt.c @@ -1068,11 +1068,11 @@ static void intel_i9xx_setup_flush(void) intel_i915_setup_chipset_flush(); } - if (intel_private.ifp_resource.start) { + if (intel_private.ifp_resource.start) intel_private.i9xx_flush_page = ioremap_nocache(intel_private.ifp_resource.start, PAGE_SIZE); - if (!intel_private.i9xx_flush_page) - dev_info(&intel_private.pcidev->dev, "can't ioremap flush page - no chipset flushing"); - } + if (!intel_private.i9xx_flush_page) + dev_err(&intel_private.pcidev->dev, + "can't ioremap flush page - no chipset flushing\n"); } static int intel_i9xx_configure(void) -- cgit v0.10.2 From 9927a403ca8c97798129953fa9cbb5dc259c7cb9 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Sat, 19 Jun 2010 15:12:51 +0200 Subject: i915: return -EFAULT if copy_to_user fails copy_to_user returns the number of bytes remaining to be copied, but we want to return a negative error code here. These are returned to userspace. Signed-off-by: Dan Carpenter Signed-off-by: Chris Wilson Cc: stable@kernel.org diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c index 7796f45..051c4db 100644 --- a/drivers/gpu/drm/i915/i915_dma.c +++ b/drivers/gpu/drm/i915/i915_dma.c @@ -620,8 +620,10 @@ static int i915_batchbuffer(struct drm_device *dev, void *data, ret = copy_from_user(cliprects, batch->cliprects, batch->num_cliprects * sizeof(struct drm_clip_rect)); - if (ret != 0) + if (ret != 0) { + ret = -EFAULT; goto fail_free; + } } mutex_lock(&dev->struct_mutex); @@ -662,8 +664,10 @@ static int i915_cmdbuffer(struct drm_device *dev, void *data, return -ENOMEM; ret = copy_from_user(batch_data, cmdbuf->buf, cmdbuf->sz); - if (ret != 0) + if (ret != 0) { + ret = -EFAULT; goto fail_batch_free; + } if (cmdbuf->num_cliprects) { cliprects = kcalloc(cmdbuf->num_cliprects, @@ -676,8 +680,10 @@ static int i915_cmdbuffer(struct drm_device *dev, void *data, ret = copy_from_user(cliprects, cmdbuf->cliprects, cmdbuf->num_cliprects * sizeof(struct drm_clip_rect)); - if (ret != 0) + if (ret != 0) { + ret = -EFAULT; goto fail_clip_free; + } } mutex_lock(&dev->struct_mutex); -- cgit v0.10.2 From c877cdce93a44eea96f6cf7fc04be7d0372db2be Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 23 Jun 2010 19:03:01 +0200 Subject: i915: return -EFAULT if copy_to_user fails copy_to_user() returns the number of bytes remaining to be copied and I'm pretty sure we want to return a negative error code here. Signed-off-by: Dan Carpenter Signed-off-by: Chris Wilson Cc: stable@kernel.org diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 68526f4..748c263 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -3588,6 +3588,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data, if (ret != 0) { DRM_ERROR("copy %d cliprects failed: %d\n", args->num_cliprects, ret); + ret = -EFAULT; goto pre_mutex_err; } } -- cgit v0.10.2 From 4d155641c81203440da64c4633b4efaab75f63b3 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Tue, 7 Sep 2010 11:58:30 +0200 Subject: ALSA: hda - Add quirk for Lenovo T400s Lenovo T400s requires the quirk to make automatic HP/mic switching working. Reported-by: Frank Becker Cc: Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_conexant.c b/sound/pci/hda/patch_conexant.c index 4f06199..71f9d64 100644 --- a/sound/pci/hda/patch_conexant.c +++ b/sound/pci/hda/patch_conexant.c @@ -3094,6 +3094,7 @@ static struct snd_pci_quirk cxt5066_cfg_tbl[] = { SND_PCI_QUIRK(0x103c, 0x360b, "HP G60", CXT5066_HP_LAPTOP), SND_PCI_QUIRK(0x1179, 0xff50, "Toshiba Satellite P500-PSPGSC-01800T", CXT5066_OLPC_XO_1_5), SND_PCI_QUIRK(0x1179, 0xffe0, "Toshiba Satellite Pro T130-15F", CXT5066_OLPC_XO_1_5), + SND_PCI_QUIRK(0x17aa, 0x20f2, "Lenovo T400s", CXT5066_THINKPAD), SND_PCI_QUIRK(0x17aa, 0x21b2, "Thinkpad X100e", CXT5066_IDEAPAD), SND_PCI_QUIRK(0x17aa, 0x21b3, "Thinkpad Edge 13 (197)", CXT5066_IDEAPAD), SND_PCI_QUIRK(0x17aa, 0x21b4, "Thinkpad Edge", CXT5066_IDEAPAD), -- cgit v0.10.2 From c96c3a8cb7fadcb33d9a5ebe35fcee8b7d0a7946 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Wed, 11 Aug 2010 09:59:24 +0100 Subject: drm/i915: Include a generation number in the device info To simplify the IS_GEN[234] macros and to enable switching. Signed-off-by: Chris Wilson diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 00befce..5363985 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -61,91 +61,86 @@ extern int intel_agp_enabled; .driver_data = (unsigned long) info } static const struct intel_device_info intel_i830_info = { - .is_i8xx = 1, .is_mobile = 1, .cursor_needs_physical = 1, + .gen = 2, .is_i8xx = 1, .is_mobile = 1, .cursor_needs_physical = 1, }; static const struct intel_device_info intel_845g_info = { - .is_i8xx = 1, + .gen = 2, .is_i8xx = 1, }; static const struct intel_device_info intel_i85x_info = { - .is_i8xx = 1, .is_i85x = 1, .is_mobile = 1, + .gen = 2, .is_i8xx = 1, .is_i85x = 1, .is_mobile = 1, .cursor_needs_physical = 1, }; static const struct intel_device_info intel_i865g_info = { - .is_i8xx = 1, + .gen = 2, .is_i8xx = 1, }; static const struct intel_device_info intel_i915g_info = { - .is_i915g = 1, .is_i9xx = 1, .cursor_needs_physical = 1, + .gen = 3, .is_i915g = 1, .is_i9xx = 1, .cursor_needs_physical = 1, }; static const struct intel_device_info intel_i915gm_info = { - .is_i9xx = 1, .is_mobile = 1, + .gen = 3, .is_i9xx = 1, .is_mobile = 1, .cursor_needs_physical = 1, }; static const struct intel_device_info intel_i945g_info = { - .is_i9xx = 1, .has_hotplug = 1, .cursor_needs_physical = 1, + .gen = 3, .is_i9xx = 1, .has_hotplug = 1, .cursor_needs_physical = 1, }; static const struct intel_device_info intel_i945gm_info = { - .is_i945gm = 1, .is_i9xx = 1, .is_mobile = 1, + .gen = 3, .is_i945gm = 1, .is_i9xx = 1, .is_mobile = 1, .has_hotplug = 1, .cursor_needs_physical = 1, }; static const struct intel_device_info intel_i965g_info = { - .is_broadwater = 1, .is_i965g = 1, .is_i9xx = 1, .has_hotplug = 1, + .gen = 4, .is_broadwater = 1, .is_i965g = 1, .is_i9xx = 1, + .has_hotplug = 1, }; static const struct intel_device_info intel_i965gm_info = { - .is_crestline = 1, .is_i965g = 1, .is_i965gm = 1, .is_i9xx = 1, - .is_mobile = 1, .has_fbc = 1, .has_rc6 = 1, - .has_hotplug = 1, + .gen = 4, .is_crestline = 1, .is_i965g = 1, .is_i965gm = 1, .is_i9xx = 1, + .is_mobile = 1, .has_fbc = 1, .has_rc6 = 1, .has_hotplug = 1, }; static const struct intel_device_info intel_g33_info = { - .is_g33 = 1, .is_i9xx = 1, .need_gfx_hws = 1, - .has_hotplug = 1, + .gen = 3, .is_g33 = 1, .is_i9xx = 1, + .need_gfx_hws = 1, .has_hotplug = 1, }; static const struct intel_device_info intel_g45_info = { - .is_i965g = 1, .is_g4x = 1, .is_i9xx = 1, .need_gfx_hws = 1, - .has_pipe_cxsr = 1, - .has_hotplug = 1, + .gen = 4, .is_i965g = 1, .is_g4x = 1, .is_i9xx = 1, .need_gfx_hws = 1, + .has_pipe_cxsr = 1, .has_hotplug = 1, }; static const struct intel_device_info intel_gm45_info = { - .is_i965g = 1, .is_g4x = 1, .is_i9xx = 1, + .gen = 4, .is_i965g = 1, .is_g4x = 1, .is_i9xx = 1, .is_mobile = 1, .need_gfx_hws = 1, .has_fbc = 1, .has_rc6 = 1, - .has_pipe_cxsr = 1, - .has_hotplug = 1, + .has_pipe_cxsr = 1, .has_hotplug = 1, }; static const struct intel_device_info intel_pineview_info = { - .is_g33 = 1, .is_pineview = 1, .is_mobile = 1, .is_i9xx = 1, - .need_gfx_hws = 1, - .has_hotplug = 1, + .gen = 3, .is_g33 = 1, .is_pineview = 1, .is_mobile = 1, .is_i9xx = 1, + .need_gfx_hws = 1, .has_hotplug = 1, }; static const struct intel_device_info intel_ironlake_d_info = { - .is_ironlake = 1, .is_i965g = 1, .is_i9xx = 1, .need_gfx_hws = 1, - .has_pipe_cxsr = 1, - .has_hotplug = 1, + .gen = 5, .is_ironlake = 1, .is_i965g = 1, .is_i9xx = 1, + .need_gfx_hws = 1, .has_pipe_cxsr = 1, .has_hotplug = 1, }; static const struct intel_device_info intel_ironlake_m_info = { - .is_ironlake = 1, .is_mobile = 1, .is_i965g = 1, .is_i9xx = 1, - .need_gfx_hws = 1, .has_fbc = 1, .has_rc6 = 1, - .has_hotplug = 1, + .gen = 5, .is_ironlake = 1, .is_mobile = 1, .is_i965g = 1, .is_i9xx = 1, + .need_gfx_hws = 1, .has_fbc = 1, .has_rc6 = 1, .has_hotplug = 1, }; static const struct intel_device_info intel_sandybridge_d_info = { - .is_i965g = 1, .is_i9xx = 1, .need_gfx_hws = 1, - .has_hotplug = 1, .is_gen6 = 1, + .gen = 6, .is_i965g = 1, .is_i9xx = 1, + .need_gfx_hws = 1, .has_hotplug = 1, }; static const struct intel_device_info intel_sandybridge_m_info = { - .is_i965g = 1, .is_mobile = 1, .is_i9xx = 1, .need_gfx_hws = 1, - .has_hotplug = 1, .is_gen6 = 1, + .gen = 6, .is_i965g = 1, .is_mobile = 1, .is_i9xx = 1, + .need_gfx_hws = 1, .has_hotplug = 1, }; static const struct pci_device_id pciidlist[] = { /* aka */ diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 047cd7c..af4a263 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -191,6 +191,7 @@ struct drm_i915_display_funcs { }; struct intel_device_info { + u8 gen; u8 is_mobile : 1; u8 is_i8xx : 1; u8 is_i85x : 1; @@ -206,7 +207,6 @@ struct intel_device_info { u8 is_broadwater : 1; u8 is_crestline : 1; u8 is_ironlake : 1; - u8 is_gen6 : 1; u8 has_fbc : 1; u8 has_rc6 : 1; u8 has_pipe_cxsr : 1; @@ -1162,7 +1162,6 @@ extern void intel_overlay_print_error_state(struct seq_file *m, struct intel_ove #define IS_845G(dev) ((dev)->pci_device == 0x2562) #define IS_I85X(dev) (INTEL_INFO(dev)->is_i85x) #define IS_I865G(dev) ((dev)->pci_device == 0x2572) -#define IS_GEN2(dev) (INTEL_INFO(dev)->is_i8xx) #define IS_I915G(dev) (INTEL_INFO(dev)->is_i915g) #define IS_I915GM(dev) ((dev)->pci_device == 0x2592) #define IS_I945G(dev) ((dev)->pci_device == 0x2772) @@ -1181,27 +1180,13 @@ extern void intel_overlay_print_error_state(struct seq_file *m, struct intel_ove #define IS_IRONLAKE_M(dev) ((dev)->pci_device == 0x0046) #define IS_IRONLAKE(dev) (INTEL_INFO(dev)->is_ironlake) #define IS_I9XX(dev) (INTEL_INFO(dev)->is_i9xx) -#define IS_GEN6(dev) (INTEL_INFO(dev)->is_gen6) #define IS_MOBILE(dev) (INTEL_INFO(dev)->is_mobile) -#define IS_GEN3(dev) (IS_I915G(dev) || \ - IS_I915GM(dev) || \ - IS_I945G(dev) || \ - IS_I945GM(dev) || \ - IS_G33(dev) || \ - IS_PINEVIEW(dev)) -#define IS_GEN4(dev) ((dev)->pci_device == 0x2972 || \ - (dev)->pci_device == 0x2982 || \ - (dev)->pci_device == 0x2992 || \ - (dev)->pci_device == 0x29A2 || \ - (dev)->pci_device == 0x2A02 || \ - (dev)->pci_device == 0x2A12 || \ - (dev)->pci_device == 0x2E02 || \ - (dev)->pci_device == 0x2E12 || \ - (dev)->pci_device == 0x2E22 || \ - (dev)->pci_device == 0x2E32 || \ - (dev)->pci_device == 0x2A42 || \ - (dev)->pci_device == 0x2E42) +#define IS_GEN2(dev) (INTEL_INFO(dev)->gen == 2) +#define IS_GEN3(dev) (INTEL_INFO(dev)->gen == 3) +#define IS_GEN4(dev) (INTEL_INFO(dev)->gen == 4) +#define IS_GEN5(dev) (INTEL_INFO(dev)->gen == 5) +#define IS_GEN6(dev) (INTEL_INFO(dev)->gen == 6) #define HAS_BSD(dev) (IS_IRONLAKE(dev) || IS_G4X(dev)) #define I915_NEED_GFX_HWS(dev) (INTEL_INFO(dev)->need_gfx_hws) -- cgit v0.10.2 From 52e68630d13f9668f8f4dd6978fa41039bacfaf6 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Sun, 8 Aug 2010 10:15:59 +0100 Subject: drm/i915: Fix offset page-flips on i965+ i965 uses the Display Registers to compute the offset from the display base so the new base does not need adjusting when flipping. The older chipsets use a fence to access the display and so do perceive the surface as linear and have a single base register which is reprogrammed using the flip. Signed-off-by: Chris Wilson Cc: Jesse Barnes Reported-by: Marty Jack Reviewed-by: Jesse Barnes diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 3fc767b..334665c 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -5042,9 +5042,9 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc, struct intel_crtc *intel_crtc = to_intel_crtc(crtc); struct intel_unpin_work *work; unsigned long flags, offset; - int pipesrc_reg = (intel_crtc->pipe == 0) ? PIPEASRC : PIPEBSRC; - int ret, pipesrc; - u32 flip_mask; + int pipe = intel_crtc->pipe; + u32 pf, pipesrc; + int ret; work = kzalloc(sizeof *work, GFP_KERNEL); if (work == NULL) @@ -5093,12 +5093,14 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc, atomic_inc(&obj_priv->pending_flip); work->pending_flip_obj = obj; - if (intel_crtc->plane) - flip_mask = MI_WAIT_FOR_PLANE_B_FLIP; - else - flip_mask = MI_WAIT_FOR_PLANE_A_FLIP; - if (IS_GEN3(dev) || IS_GEN2(dev)) { + u32 flip_mask; + + if (intel_crtc->plane) + flip_mask = MI_WAIT_FOR_PLANE_B_FLIP; + else + flip_mask = MI_WAIT_FOR_PLANE_A_FLIP; + BEGIN_LP_RING(2); OUT_RING(MI_WAIT_FOR_EVENT | flip_mask); OUT_RING(0); @@ -5106,29 +5108,56 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc, } /* Offset into the new buffer for cases of shared fbs between CRTCs */ - offset = obj_priv->gtt_offset; - offset += (crtc->y * fb->pitch) + (crtc->x * (fb->bits_per_pixel) / 8); + offset = crtc->y * fb->pitch + crtc->x * fb->bits_per_pixel/8; BEGIN_LP_RING(4); - if (IS_I965G(dev)) { + switch(INTEL_INFO(dev)->gen) { + case 2: OUT_RING(MI_DISPLAY_FLIP | MI_DISPLAY_FLIP_PLANE(intel_crtc->plane)); OUT_RING(fb->pitch); - OUT_RING(offset | obj_priv->tiling_mode); - pipesrc = I915_READ(pipesrc_reg); - OUT_RING(pipesrc & 0x0fff0fff); - } else if (IS_GEN3(dev)) { + OUT_RING(obj_priv->gtt_offset + offset); + OUT_RING(MI_NOOP); + break; + + case 3: OUT_RING(MI_DISPLAY_FLIP_I915 | MI_DISPLAY_FLIP_PLANE(intel_crtc->plane)); OUT_RING(fb->pitch); - OUT_RING(offset); + OUT_RING(obj_priv->gtt_offset + offset); OUT_RING(MI_NOOP); - } else { + break; + + case 4: + case 5: + /* i965+ uses the linear or tiled offsets from the + * Display Registers (which do not change across a page-flip) + * so we need only reprogram the base address. + */ OUT_RING(MI_DISPLAY_FLIP | MI_DISPLAY_FLIP_PLANE(intel_crtc->plane)); OUT_RING(fb->pitch); - OUT_RING(offset); - OUT_RING(MI_NOOP); + OUT_RING(obj_priv->gtt_offset | obj_priv->tiling_mode); + + /* XXX Enabling the panel-fitter across page-flip is so far + * untested on non-native modes, so ignore it for now. + * pf = I915_READ(pipe == 0 ? PFA_CTL_1 : PFB_CTL_1) & PF_ENABLE; + */ + pf = 0; + pipesrc = I915_READ(pipe == 0 ? PIPEASRC : PIPEBSRC) & 0x0fff0fff; + OUT_RING(pf | pipesrc); + break; + + case 6: + OUT_RING(MI_DISPLAY_FLIP | + MI_DISPLAY_FLIP_PLANE(intel_crtc->plane)); + OUT_RING(fb->pitch | obj_priv->tiling_mode); + OUT_RING(obj_priv->gtt_offset); + + pf = I915_READ(pipe == 0 ? PFA_CTL_1 : PFB_CTL_1) & PF_ENABLE; + pipesrc = I915_READ(pipe == 0 ? PIPEASRC : PIPEBSRC) & 0x0fff0fff; + OUT_RING(pf | pipesrc); + break; } ADVANCE_LP_RING(); -- cgit v0.10.2 From 4e6cfefc729be2aa20647415317577ed98d4f7bf Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Sun, 8 Aug 2010 13:20:19 +0100 Subject: drm/i915: Re-use set_base_atomic to share setting of the display registers Lets try to avoid repeating old bugs. Signed-off-by: Chris Wilson Reviewed-by: Jesse Barnes diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 334665c..cbb5093 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -1502,7 +1502,7 @@ intel_pipe_set_base_atomic(struct drm_crtc *crtc, struct drm_framebuffer *fb, dspcntr &= ~DISPPLANE_TILED; } - if (IS_IRONLAKE(dev)) + if (HAS_PCH_SPLIT(dev)) /* must disable */ dspcntr |= DISPPLANE_TRICKLE_FEED_DISABLE; @@ -1511,20 +1511,19 @@ intel_pipe_set_base_atomic(struct drm_crtc *crtc, struct drm_framebuffer *fb, Start = obj_priv->gtt_offset; Offset = y * fb->pitch + x * (fb->bits_per_pixel / 8); - DRM_DEBUG("Writing base %08lX %08lX %d %d\n", Start, Offset, x, y); + DRM_DEBUG_KMS("Writing base %08lX %08lX %d %d %d\n", + Start, Offset, x, y, fb->pitch); I915_WRITE(dspstride, fb->pitch); if (IS_I965G(dev)) { - I915_WRITE(dspbase, Offset); - I915_READ(dspbase); I915_WRITE(dspsurf, Start); - I915_READ(dspsurf); I915_WRITE(dsptileoff, (y << 16) | x); + I915_WRITE(dspbase, Offset); } else { I915_WRITE(dspbase, Start + Offset); - I915_READ(dspbase); } + POSTING_READ(dspbase); - if ((IS_I965G(dev) || plane == 0)) + if (IS_I965G(dev) || plane == 0) intel_update_fbc(crtc, &crtc->mode); intel_wait_for_vblank(dev, intel_crtc->pipe); @@ -1538,7 +1537,6 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, struct drm_framebuffer *old_fb) { struct drm_device *dev = crtc->dev; - struct drm_i915_private *dev_priv = dev->dev_private; struct drm_i915_master_private *master_priv; struct intel_crtc *intel_crtc = to_intel_crtc(crtc); struct intel_framebuffer *intel_fb; @@ -1546,13 +1544,6 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, struct drm_gem_object *obj; int pipe = intel_crtc->pipe; int plane = intel_crtc->plane; - unsigned long Start, Offset; - int dspbase = (plane == 0 ? DSPAADDR : DSPBADDR); - int dspsurf = (plane == 0 ? DSPASURF : DSPBSURF); - int dspstride = (plane == 0) ? DSPASTRIDE : DSPBSTRIDE; - int dsptileoff = (plane == 0 ? DSPATILEOFF : DSPBTILEOFF); - int dspcntr_reg = (plane == 0) ? DSPACNTR : DSPBCNTR; - u32 dspcntr; int ret; /* no fb bound */ @@ -1588,71 +1579,18 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, return ret; } - dspcntr = I915_READ(dspcntr_reg); - /* Mask out pixel format bits in case we change it */ - dspcntr &= ~DISPPLANE_PIXFORMAT_MASK; - switch (crtc->fb->bits_per_pixel) { - case 8: - dspcntr |= DISPPLANE_8BPP; - break; - case 16: - if (crtc->fb->depth == 15) - dspcntr |= DISPPLANE_15_16BPP; - else - dspcntr |= DISPPLANE_16BPP; - break; - case 24: - case 32: - if (crtc->fb->depth == 30) - dspcntr |= DISPPLANE_32BPP_30BIT_NO_ALPHA; - else - dspcntr |= DISPPLANE_32BPP_NO_ALPHA; - break; - default: - DRM_ERROR("Unknown color depth\n"); + ret = intel_pipe_set_base_atomic(crtc, crtc->fb, x, y); + if (ret) { i915_gem_object_unpin(obj); mutex_unlock(&dev->struct_mutex); - return -EINVAL; - } - if (IS_I965G(dev)) { - if (obj_priv->tiling_mode != I915_TILING_NONE) - dspcntr |= DISPPLANE_TILED; - else - dspcntr &= ~DISPPLANE_TILED; - } - - if (HAS_PCH_SPLIT(dev)) - /* must disable */ - dspcntr |= DISPPLANE_TRICKLE_FEED_DISABLE; - - I915_WRITE(dspcntr_reg, dspcntr); - - Start = obj_priv->gtt_offset; - Offset = y * crtc->fb->pitch + x * (crtc->fb->bits_per_pixel / 8); - - DRM_DEBUG_KMS("Writing base %08lX %08lX %d %d %d\n", - Start, Offset, x, y, crtc->fb->pitch); - I915_WRITE(dspstride, crtc->fb->pitch); - if (IS_I965G(dev)) { - I915_WRITE(dspsurf, Start); - I915_WRITE(dsptileoff, (y << 16) | x); - I915_WRITE(dspbase, Offset); - } else { - I915_WRITE(dspbase, Start + Offset); + return ret; } - POSTING_READ(dspbase); - - if ((IS_I965G(dev) || plane == 0)) - intel_update_fbc(crtc, &crtc->mode); - - intel_wait_for_vblank(dev, pipe); if (old_fb) { intel_fb = to_intel_framebuffer(old_fb); obj_priv = to_intel_bo(intel_fb->obj); i915_gem_object_unpin(intel_fb->obj); } - intel_increase_pllclock(crtc, true); mutex_unlock(&dev->struct_mutex); -- cgit v0.10.2 From 0ad6ef2c587dea59212c4e2ab3ec3b0067500a2a Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Mon, 9 Aug 2010 17:21:44 +0100 Subject: drm/i915/dp: Boost timeout for enabling transcoder to 100ms Adam Hill reported that his Arrandale system required a much longer, up to 200x500us, wait for the panel to initialise or else modesetting would fail. References: https://bugs.freedesktop.org/show_bug.cgi?id=29141 Signed-off-by: Chris Wilson Reported-and-tested-by: Adam Hill diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index cbb5093..83c8545 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -2069,7 +2069,7 @@ static void ironlake_crtc_dpms(struct drm_crtc *crtc, int mode) I915_WRITE(transconf_reg, temp | TRANS_ENABLE); I915_READ(transconf_reg); - if (wait_for(I915_READ(transconf_reg) & TRANS_STATE_ENABLE, 10, 0)) + if (wait_for(I915_READ(transconf_reg) & TRANS_STATE_ENABLE, 100, 1)) DRM_ERROR("failed to enable transcoder\n"); } -- cgit v0.10.2 From b66d842467311ac3434aa19c5c41deaab8295bd0 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Thu, 12 Aug 2010 15:26:41 +0100 Subject: drm/i915/sdvo: Restore guess of the DDC bus in absence of VBIOS If the VBIOS tells us the mapping of the SDVO device onto the DDC bus, use it. However, if there is no VBIOS available that mapping is uninitialised and we should fallback to our earlier guess. Fix regression introduced in b1083333 (which in turn is a fix for the regression caused by the introduction of this guess, 14571b4). References: Bug 29499 - [945GM] Screen disconnected because of missing VBIOS https://bugs.freedesktop.org/show_bug.cgi?id=29499 Bug 15109 - i945GM fails to detect EDID on DVI port https://bugzilla.kernel.org/show_bug.cgi?id=15109 Signed-off-by: Chris Wilson Reported-and-tested-by: Paul Neumann Cc: Adam Jackson Cc: Zhenyu Wang Cc: stable@kernel.org diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c index 62d22ae..e3b7a7e 100644 --- a/drivers/gpu/drm/i915/intel_sdvo.c +++ b/drivers/gpu/drm/i915/intel_sdvo.c @@ -1929,6 +1929,41 @@ static const struct drm_encoder_funcs intel_sdvo_enc_funcs = { .destroy = intel_sdvo_enc_destroy, }; +static void +intel_sdvo_guess_ddc_bus(struct intel_sdvo *sdvo) +{ + uint16_t mask = 0; + unsigned int num_bits; + + /* Make a mask of outputs less than or equal to our own priority in the + * list. + */ + switch (sdvo->controlled_output) { + case SDVO_OUTPUT_LVDS1: + mask |= SDVO_OUTPUT_LVDS1; + case SDVO_OUTPUT_LVDS0: + mask |= SDVO_OUTPUT_LVDS0; + case SDVO_OUTPUT_TMDS1: + mask |= SDVO_OUTPUT_TMDS1; + case SDVO_OUTPUT_TMDS0: + mask |= SDVO_OUTPUT_TMDS0; + case SDVO_OUTPUT_RGB1: + mask |= SDVO_OUTPUT_RGB1; + case SDVO_OUTPUT_RGB0: + mask |= SDVO_OUTPUT_RGB0; + break; + } + + /* Count bits to find what number we are in the priority list. */ + mask &= sdvo->caps.output_flags; + num_bits = hweight16(mask); + /* If more than 3 outputs, default to DDC bus 3 for now. */ + if (num_bits > 3) + num_bits = 3; + + /* Corresponds to SDVO_CONTROL_BUS_DDCx */ + sdvo->ddc_bus = 1 << num_bits; +} /** * Choose the appropriate DDC bus for control bus switch command for this @@ -1948,7 +1983,10 @@ intel_sdvo_select_ddc_bus(struct drm_i915_private *dev_priv, else mapping = &(dev_priv->sdvo_mappings[1]); - sdvo->ddc_bus = 1 << ((mapping->ddc_pin & 0xf0) >> 4); + if (mapping->initialized) + sdvo->ddc_bus = 1 << ((mapping->ddc_pin & 0xf0) >> 4); + else + intel_sdvo_guess_ddc_bus(sdvo); } static bool -- cgit v0.10.2 From 4f7f7b7eb94bd37c449f06932459bbed78826f8d Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Wed, 18 Aug 2010 18:12:56 +0100 Subject: drm/i915/dp: Really try 5 times before giving up. Only stop trying if the aux channel sucessfully reports that the transmission was completed, otherwise try again. On the 5th failure, bail and report that something is amiss. This fixes a sporadic failure in reading the EDID for my external panel over DP. Signed-off-by: Chris Wilson Cc: stable@kernel.org diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 9caccd0..51d1429 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -239,7 +239,6 @@ intel_dp_aux_ch(struct intel_dp *intel_dp, uint32_t ch_data = ch_ctl + 4; int i; int recv_bytes; - uint32_t ctl; uint32_t status; uint32_t aux_clock_divider; int try, precharge; @@ -263,41 +262,43 @@ intel_dp_aux_ch(struct intel_dp *intel_dp, else precharge = 5; + if (I915_READ(ch_ctl) & DP_AUX_CH_CTL_SEND_BUSY) { + DRM_ERROR("dp_aux_ch not started status 0x%08x\n", + I915_READ(ch_ctl)); + return -EBUSY; + } + /* Must try at least 3 times according to DP spec */ for (try = 0; try < 5; try++) { /* Load the send data into the aux channel data registers */ - for (i = 0; i < send_bytes; i += 4) { - uint32_t d = pack_aux(send + i, send_bytes - i); - - I915_WRITE(ch_data + i, d); - } - - ctl = (DP_AUX_CH_CTL_SEND_BUSY | - DP_AUX_CH_CTL_TIME_OUT_400us | - (send_bytes << DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT) | - (precharge << DP_AUX_CH_CTL_PRECHARGE_2US_SHIFT) | - (aux_clock_divider << DP_AUX_CH_CTL_BIT_CLOCK_2X_SHIFT) | - DP_AUX_CH_CTL_DONE | - DP_AUX_CH_CTL_TIME_OUT_ERROR | - DP_AUX_CH_CTL_RECEIVE_ERROR); + for (i = 0; i < send_bytes; i += 4) + I915_WRITE(ch_data + i, + pack_aux(send + i, send_bytes - i)); /* Send the command and wait for it to complete */ - I915_WRITE(ch_ctl, ctl); - (void) I915_READ(ch_ctl); + I915_WRITE(ch_ctl, + DP_AUX_CH_CTL_SEND_BUSY | + DP_AUX_CH_CTL_TIME_OUT_400us | + (send_bytes << DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT) | + (precharge << DP_AUX_CH_CTL_PRECHARGE_2US_SHIFT) | + (aux_clock_divider << DP_AUX_CH_CTL_BIT_CLOCK_2X_SHIFT) | + DP_AUX_CH_CTL_DONE | + DP_AUX_CH_CTL_TIME_OUT_ERROR | + DP_AUX_CH_CTL_RECEIVE_ERROR); for (;;) { - udelay(100); status = I915_READ(ch_ctl); if ((status & DP_AUX_CH_CTL_SEND_BUSY) == 0) break; + udelay(100); } /* Clear done status and any errors */ - I915_WRITE(ch_ctl, (status | - DP_AUX_CH_CTL_DONE | - DP_AUX_CH_CTL_TIME_OUT_ERROR | - DP_AUX_CH_CTL_RECEIVE_ERROR)); - (void) I915_READ(ch_ctl); - if ((status & DP_AUX_CH_CTL_TIME_OUT_ERROR) == 0) + I915_WRITE(ch_ctl, + status | + DP_AUX_CH_CTL_DONE | + DP_AUX_CH_CTL_TIME_OUT_ERROR | + DP_AUX_CH_CTL_RECEIVE_ERROR); + if (status & DP_AUX_CH_CTL_DONE) break; } @@ -324,15 +325,12 @@ intel_dp_aux_ch(struct intel_dp *intel_dp, /* Unload any bytes sent back from the other side */ recv_bytes = ((status & DP_AUX_CH_CTL_MESSAGE_SIZE_MASK) >> DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT); - if (recv_bytes > recv_size) recv_bytes = recv_size; - for (i = 0; i < recv_bytes; i += 4) { - uint32_t d = I915_READ(ch_data + i); - - unpack_aux(d, recv + i, recv_bytes - i); - } + for (i = 0; i < recv_bytes; i += 4) + unpack_aux(I915_READ(ch_data + i), + recv + i, recv_bytes - i); return recv_bytes; } -- cgit v0.10.2 From a25c25c2a2aa55e609099a9f74453c518aec29a6 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Fri, 20 Aug 2010 14:36:45 +0100 Subject: drm/i915: Allocate the PCI resource for the MCHBAR We were failing when trying to allocate the resource for MMIO of the MCHBAR because we forgot to specify what type of resource we wanted. Signed-off-by: Chris Wilson Cc: Jesse Barnes Cc: stable@kernel.org Reviewed-by: Jesse Barnes diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c index 051c4db..9d67b48 100644 --- a/drivers/gpu/drm/i915/i915_dma.c +++ b/drivers/gpu/drm/i915/i915_dma.c @@ -891,7 +891,7 @@ intel_alloc_mchbar_resource(struct drm_device *dev) int reg = IS_I965G(dev) ? MCHBAR_I965 : MCHBAR_I915; u32 temp_lo, temp_hi = 0; u64 mchbar_addr; - int ret = 0; + int ret; if (IS_I965G(dev)) pci_read_config_dword(dev_priv->bridge_dev, reg + 4, &temp_hi); @@ -901,22 +901,23 @@ intel_alloc_mchbar_resource(struct drm_device *dev) /* If ACPI doesn't have it, assume we need to allocate it ourselves */ #ifdef CONFIG_PNP if (mchbar_addr && - pnp_range_reserved(mchbar_addr, mchbar_addr + MCHBAR_SIZE)) { - ret = 0; - goto out; - } + pnp_range_reserved(mchbar_addr, mchbar_addr + MCHBAR_SIZE)) + return 0; #endif /* Get some space for it */ - ret = pci_bus_alloc_resource(dev_priv->bridge_dev->bus, &dev_priv->mch_res, + dev_priv->mch_res.name = "i915 MCHBAR"; + dev_priv->mch_res.flags = IORESOURCE_MEM; + ret = pci_bus_alloc_resource(dev_priv->bridge_dev->bus, + &dev_priv->mch_res, MCHBAR_SIZE, MCHBAR_SIZE, PCIBIOS_MIN_MEM, - 0, pcibios_align_resource, + 0, pcibios_align_resource, dev_priv->bridge_dev); if (ret) { DRM_DEBUG_DRIVER("failed bus alloc: %d\n", ret); dev_priv->mch_res.start = 0; - goto out; + return ret; } if (IS_I965G(dev)) @@ -925,8 +926,7 @@ intel_alloc_mchbar_resource(struct drm_device *dev) pci_write_config_dword(dev_priv->bridge_dev, reg, lower_32_bits(dev_priv->mch_res.start)); -out: - return ret; + return 0; } /* Setup MCHBAR if possible, return true if we should disable it again */ -- cgit v0.10.2 From 8e647a279ca30029f19eca646de08a6338eab924 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Sun, 22 Aug 2010 10:54:23 +0100 Subject: drm/i915: Tightly scope intel_encoder to prevent invalid use We reset intel_encoder for every matching encoder whilst iterating over the encoders attached to this crtc when changing mode. As such in a cloned configuration intel_encoder may not correspond to the correct is_edp encoder. By scoping intel_encoder to the loop, not only is the compiler able to spot this mistake, we also improve readiability for ourselves. [It might not be a mistake, within this function it is unclear as to whether it is permissable for eDP to be cloned...] Signed-off-by: Chris Wilson diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 83c8545..0b90443 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -3508,10 +3508,9 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc, u32 dpll = 0, fp = 0, fp2 = 0, dspcntr, pipeconf; bool ok, has_reduced_clock = false, is_sdvo = false, is_dvo = false; bool is_crt = false, is_lvds = false, is_tv = false, is_dp = false; - bool is_edp = false; + struct intel_encoder *has_edp_encoder = NULL; struct drm_mode_config *mode_config = &dev->mode_config; struct drm_encoder *encoder; - struct intel_encoder *intel_encoder = NULL; const intel_limit_t *limit; int ret; struct fdi_m_n m_n = {0}; @@ -3532,12 +3531,12 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc, drm_vblank_pre_modeset(dev, pipe); list_for_each_entry(encoder, &mode_config->encoder_list, head) { + struct intel_encoder *intel_encoder; - if (!encoder || encoder->crtc != crtc) + if (encoder->crtc != crtc) continue; intel_encoder = enc_to_intel_encoder(encoder); - switch (intel_encoder->type) { case INTEL_OUTPUT_LVDS: is_lvds = true; @@ -3561,7 +3560,7 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc, is_dp = true; break; case INTEL_OUTPUT_EDP: - is_edp = true; + has_edp_encoder = intel_encoder; break; } @@ -3639,10 +3638,10 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc, int lane = 0, link_bw, bpp; /* eDP doesn't require FDI link, so just set DP M/N according to current link config */ - if (is_edp) { + if (has_edp_encoder) { target_clock = mode->clock; - intel_edp_link_config(intel_encoder, - &lane, &link_bw); + intel_edp_link_config(has_edp_encoder, + &lane, &link_bw); } else { /* DP over FDI requires target mode clock instead of link clock */ @@ -3663,7 +3662,7 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc, temp |= PIPE_8BPC; else temp |= PIPE_6BPC; - } else if (is_edp || (is_dp && intel_pch_has_edp(crtc))) { + } else if (has_edp_encoder || (is_dp && intel_pch_has_edp(crtc))) { switch (dev_priv->edp_bpp/3) { case 8: temp |= PIPE_8BPC; @@ -3736,7 +3735,7 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc, udelay(200); - if (is_edp) { + if (has_edp_encoder) { if (dev_priv->lvds_use_ssc) { temp |= DREF_SSC1_ENABLE; I915_WRITE(PCH_DREF_CONTROL, temp); @@ -3885,7 +3884,7 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc, dpll_reg = pch_dpll_reg; } - if (!is_edp) { + if (!has_edp_encoder) { I915_WRITE(fp_reg, fp); I915_WRITE(dpll_reg, dpll & ~DPLL_VCO_ENABLE); I915_READ(dpll_reg); @@ -3980,7 +3979,7 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc, } } - if (!is_edp) { + if (!has_edp_encoder) { I915_WRITE(fp_reg, fp); I915_WRITE(dpll_reg, dpll); I915_READ(dpll_reg); @@ -4059,7 +4058,7 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc, I915_WRITE(link_m1_reg, m_n.link_m); I915_WRITE(link_n1_reg, m_n.link_n); - if (is_edp) { + if (has_edp_encoder) { ironlake_set_pll_edp(crtc, adjusted_mode->clock); } else { /* enable FDI RX PLL too */ -- cgit v0.10.2 From 4e5359cd053bfb7d8dabe4a63624a5726848ffbc Mon Sep 17 00:00:00 2001 From: Simon Farnsworth Date: Wed, 1 Sep 2010 17:47:52 +0100 Subject: drm/i915: Avoid pageflipping freeze when we miss the flip prepare interrupt When we miss the flip prepare interrupt, we never get into the software state needed to restart userspace, resulting in a freeze of a full-screen OpenGL application (such as a compositor). Work around this by checking DSPxSURF/DSPxBASE to see if the page flip has actually happened. If it has, do the work we would have done when the flip prepare interrupt comes in. Also, add debugfs information to tell us what's going on (based on the patch from Chris Wilson attached to bugs.fdo bug #29798). Signed-off-by: Simon Farnsworth Signed-off-by: Chris Wilson diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 92d5605..5e43d70 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -31,6 +31,7 @@ #include #include "drmP.h" #include "drm.h" +#include "intel_drv.h" #include "i915_drm.h" #include "i915_drv.h" @@ -121,6 +122,54 @@ static int i915_gem_object_list_info(struct seq_file *m, void *data) return 0; } +static int i915_gem_pageflip_info(struct seq_file *m, void *data) +{ + struct drm_info_node *node = (struct drm_info_node *) m->private; + struct drm_device *dev = node->minor->dev; + unsigned long flags; + struct intel_crtc *crtc; + + list_for_each_entry(crtc, &dev->mode_config.crtc_list, base.head) { + const char *pipe = crtc->pipe ? "B" : "A"; + const char *plane = crtc->plane ? "B" : "A"; + struct intel_unpin_work *work; + + spin_lock_irqsave(&dev->event_lock, flags); + work = crtc->unpin_work; + if (work == NULL) { + seq_printf(m, "No flip due on pipe %s (plane %s)\n", + pipe, plane); + } else { + if (!work->pending) { + seq_printf(m, "Flip queued on pipe %s (plane %s)\n", + pipe, plane); + } else { + seq_printf(m, "Flip pending (waiting for vsync) on pipe %s (plane %s)\n", + pipe, plane); + } + if (work->enable_stall_check) + seq_printf(m, "Stall check enabled, "); + else + seq_printf(m, "Stall check waiting for page flip ioctl, "); + seq_printf(m, "%d prepares\n", work->pending); + + if (work->old_fb_obj) { + struct drm_i915_gem_object *obj_priv = to_intel_bo(work->old_fb_obj); + if(obj_priv) + seq_printf(m, "Old framebuffer gtt_offset 0x%08x\n", obj_priv->gtt_offset ); + } + if (work->pending_flip_obj) { + struct drm_i915_gem_object *obj_priv = to_intel_bo(work->pending_flip_obj); + if(obj_priv) + seq_printf(m, "New framebuffer gtt_offset 0x%08x\n", obj_priv->gtt_offset ); + } + } + spin_unlock_irqrestore(&dev->event_lock, flags); + } + + return 0; +} + static int i915_gem_request_info(struct seq_file *m, void *data) { struct drm_info_node *node = (struct drm_info_node *) m->private; @@ -777,6 +826,7 @@ static struct drm_info_list i915_debugfs_list[] = { {"i915_gem_active", i915_gem_object_list_info, 0, (void *) ACTIVE_LIST}, {"i915_gem_flushing", i915_gem_object_list_info, 0, (void *) FLUSHING_LIST}, {"i915_gem_inactive", i915_gem_object_list_info, 0, (void *) INACTIVE_LIST}, + {"i915_gem_pageflip", i915_gem_pageflip_info, 0}, {"i915_gem_request", i915_gem_request_info, 0}, {"i915_gem_seqno", i915_gem_seqno_info, 0}, {"i915_gem_fence_regs", i915_gem_fence_regs_info, 0}, diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 16861b8..59457e8 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -887,6 +887,49 @@ static void i915_handle_error(struct drm_device *dev, bool wedged) queue_work(dev_priv->wq, &dev_priv->error_work); } +static void i915_pageflip_stall_check(struct drm_device *dev, int pipe) +{ + drm_i915_private_t *dev_priv = dev->dev_private; + struct drm_crtc *crtc = dev_priv->pipe_to_crtc_mapping[pipe]; + struct intel_crtc *intel_crtc = to_intel_crtc(crtc); + struct drm_i915_gem_object *obj_priv; + struct intel_unpin_work *work; + unsigned long flags; + bool stall_detected; + + /* Ignore early vblank irqs */ + if (intel_crtc == NULL) + return; + + spin_lock_irqsave(&dev->event_lock, flags); + work = intel_crtc->unpin_work; + + if (work == NULL || work->pending || !work->enable_stall_check) { + /* Either the pending flip IRQ arrived, or we're too early. Don't check */ + spin_unlock_irqrestore(&dev->event_lock, flags); + return; + } + + /* Potential stall - if we see that the flip has happened, assume a missed interrupt */ + obj_priv = to_intel_bo(work->pending_flip_obj); + if(IS_I965G(dev)) { + int dspsurf = intel_crtc->plane == 0 ? DSPASURF : DSPBSURF; + stall_detected = I915_READ(dspsurf) == obj_priv->gtt_offset; + } else { + int dspaddr = intel_crtc->plane == 0 ? DSPAADDR : DSPBADDR; + stall_detected = I915_READ(dspaddr) == (obj_priv->gtt_offset + + crtc->y * crtc->fb->pitch + + crtc->x * crtc->fb->bits_per_pixel/8); + } + + spin_unlock_irqrestore(&dev->event_lock, flags); + + if (stall_detected) { + DRM_DEBUG_DRIVER("Pageflip stall detected\n"); + intel_prepare_page_flip(dev, intel_crtc->plane); + } +} + irqreturn_t i915_driver_irq_handler(DRM_IRQ_ARGS) { struct drm_device *dev = (struct drm_device *) arg; @@ -1004,15 +1047,19 @@ irqreturn_t i915_driver_irq_handler(DRM_IRQ_ARGS) if (pipea_stats & vblank_status) { vblank++; drm_handle_vblank(dev, 0); - if (!dev_priv->flip_pending_is_done) + if (!dev_priv->flip_pending_is_done) { + i915_pageflip_stall_check(dev, 0); intel_finish_page_flip(dev, 0); + } } if (pipeb_stats & vblank_status) { vblank++; drm_handle_vblank(dev, 1); - if (!dev_priv->flip_pending_is_done) + if (!dev_priv->flip_pending_is_done) { + i915_pageflip_stall_check(dev, 1); intel_finish_page_flip(dev, 1); + } } if ((pipea_stats & PIPE_LEGACY_BLC_EVENT_STATUS) || diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 0b90443..1bd0c67 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -4864,15 +4864,6 @@ static void intel_crtc_destroy(struct drm_crtc *crtc) kfree(intel_crtc); } -struct intel_unpin_work { - struct work_struct work; - struct drm_device *dev; - struct drm_gem_object *old_fb_obj; - struct drm_gem_object *pending_flip_obj; - struct drm_pending_vblank_event *event; - int pending; -}; - static void intel_unpin_work_fn(struct work_struct *__work) { struct intel_unpin_work *work = @@ -4960,7 +4951,8 @@ void intel_prepare_page_flip(struct drm_device *dev, int plane) spin_lock_irqsave(&dev->event_lock, flags); if (intel_crtc->unpin_work) { - intel_crtc->unpin_work->pending = 1; + if ((++intel_crtc->unpin_work->pending) > 1) + DRM_ERROR("Prepared flip multiple times\n"); } else { DRM_DEBUG_DRIVER("preparing flip with no unpin work?\n"); } @@ -5044,6 +5036,8 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc, ADVANCE_LP_RING(); } + work->enable_stall_check = true; + /* Offset into the new buffer for cases of shared fbs between CRTCs */ offset = crtc->y * fb->pitch + crtc->x * fb->bits_per_pixel/8; diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index 0e92aa0..ad312ca 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -176,6 +176,16 @@ struct intel_crtc { #define enc_to_intel_encoder(x) container_of(x, struct intel_encoder, enc) #define to_intel_framebuffer(x) container_of(x, struct intel_framebuffer, base) +struct intel_unpin_work { + struct work_struct work; + struct drm_device *dev; + struct drm_gem_object *old_fb_obj; + struct drm_gem_object *pending_flip_obj; + struct drm_pending_vblank_event *event; + int pending; + bool enable_stall_check; +}; + struct i2c_adapter *intel_i2c_create(struct drm_device *dev, const u32 reg, const char *name); void intel_i2c_destroy(struct i2c_adapter *adapter); -- cgit v0.10.2 From 52be11964869c948fbbb9ec7845f9c52b0d3dc09 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Sun, 5 Sep 2010 10:01:13 +0100 Subject: drm/i915: Avoid use of uninitialised values when disabling panel-fitter We were passing garbage values into the panel-fitter control register when disabling it on Ironlake - those values (filter modes and reserved MBZ bits) would have then be re-used the next time panel-fitting was enabled. Signed-off-by: Chris Wilson diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 1bd0c67..cf8d5e5 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -1865,9 +1865,6 @@ static void ironlake_crtc_dpms(struct drm_crtc *crtc, int mode) int fdi_tx_reg = (pipe == 0) ? FDI_TXA_CTL : FDI_TXB_CTL; int fdi_rx_reg = (pipe == 0) ? FDI_RXA_CTL : FDI_RXB_CTL; int transconf_reg = (pipe == 0) ? TRANSACONF : TRANSBCONF; - int pf_ctl_reg = (pipe == 0) ? PFA_CTL_1 : PFB_CTL_1; - int pf_win_size = (pipe == 0) ? PFA_WIN_SZ : PFB_WIN_SZ; - int pf_win_pos = (pipe == 0) ? PFA_WIN_POS : PFB_WIN_POS; int cpu_htot_reg = (pipe == 0) ? HTOTAL_A : HTOTAL_B; int cpu_hblank_reg = (pipe == 0) ? HBLANK_A : HBLANK_B; int cpu_hsync_reg = (pipe == 0) ? HSYNC_A : HSYNC_B; @@ -1936,15 +1933,19 @@ static void ironlake_crtc_dpms(struct drm_crtc *crtc, int mode) } /* Enable panel fitting for LVDS */ - if (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS) - || HAS_eDP || intel_pch_has_edp(crtc)) { - if (dev_priv->pch_pf_size) { - temp = I915_READ(pf_ctl_reg); - I915_WRITE(pf_ctl_reg, temp | PF_ENABLE | PF_FILTER_MED_3x3); - I915_WRITE(pf_win_pos, dev_priv->pch_pf_pos); - I915_WRITE(pf_win_size, dev_priv->pch_pf_size); - } else - I915_WRITE(pf_ctl_reg, temp & ~PF_ENABLE); + if (dev_priv->pch_pf_size && + (intel_pipe_has_type(crtc, INTEL_OUTPUT_LVDS) + || HAS_eDP || intel_pch_has_edp(crtc))) { + /* Force use of hard-coded filter coefficients + * as some pre-programmed values are broken, + * e.g. x201. + */ + I915_WRITE(pipe ? PFB_CTL_1 : PFA_CTL_1, + PF_ENABLE | PF_FILTER_MED_3x3); + I915_WRITE(pipe ? PFB_WIN_POS : PFA_WIN_POS, + dev_priv->pch_pf_pos); + I915_WRITE(pipe ? PFB_WIN_SZ : PFA_WIN_SZ, + dev_priv->pch_pf_size); } /* Enable CPU pipe */ @@ -2109,14 +2110,8 @@ static void ironlake_crtc_dpms(struct drm_crtc *crtc, int mode) udelay(100); /* Disable PF */ - temp = I915_READ(pf_ctl_reg); - if ((temp & PF_ENABLE) != 0) { - I915_WRITE(pf_ctl_reg, temp & ~PF_ENABLE); - I915_READ(pf_ctl_reg); - } - I915_WRITE(pf_win_size, 0); - POSTING_READ(pf_win_size); - + I915_WRITE(pipe ? PFB_CTL_1 : PFA_CTL_1, 0); + I915_WRITE(pipe ? PFB_WIN_SZ : PFA_WIN_SZ, 0); /* disable CPU FDI tx and PCH FDI rx */ temp = I915_READ(fdi_tx_reg); -- cgit v0.10.2 From 032d2a0d068b0368296a56469761394ef03207c3 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Mon, 6 Sep 2010 16:17:22 +0100 Subject: drm/i915: Prevent double dpms on Arguably this is a bug in drm-core in that we should not be called twice in succession with DPMS_ON, however this is still occuring and we see FDI link training failures on the second call leading to the occassional blank display. For the time being ignore the repeated call. Original patch by Dave Airlie Signed-off-by: Chris Wilson Cc: stable@kernel.org diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index cf8d5e5..40cc5da 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -2370,6 +2370,9 @@ static void intel_crtc_dpms(struct drm_crtc *crtc, int mode) int pipe = intel_crtc->pipe; bool enabled; + if (intel_crtc->dpms_mode == mode) + return; + intel_crtc->dpms_mode = mode; intel_crtc->cursor_on = mode == DRM_MODE_DPMS_ON; @@ -5164,7 +5167,7 @@ static void intel_crtc_init(struct drm_device *dev, int pipe) dev_priv->pipe_to_crtc_mapping[intel_crtc->pipe] = &intel_crtc->base; intel_crtc->cursor_addr = 0; - intel_crtc->dpms_mode = DRM_MODE_DPMS_OFF; + intel_crtc->dpms_mode = -1; drm_crtc_helper_add(&intel_crtc->base, &intel_helper_funcs); intel_crtc->busy = false; -- cgit v0.10.2 From 8dfc2b14ebf538f28a05565f34913ecffedf5024 Mon Sep 17 00:00:00 2001 From: Zhenyu Wang Date: Mon, 23 Aug 2010 14:37:52 +0800 Subject: agp/intel: fix physical address mask bits for sandybridge It should shift bit 39-32 into pte's bit 11-4. Reported-by:Takashi Iwai Signed-off-by: Zhenyu Wang Cc: stable@kernel.org Signed-off-by: Chris Wilson diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/intel-gtt.c index ce536e6..7f35854 100644 --- a/drivers/char/agp/intel-gtt.c +++ b/drivers/char/agp/intel-gtt.c @@ -1333,8 +1333,8 @@ static unsigned long intel_i965_mask_memory(struct agp_bridge_data *bridge, static unsigned long intel_gen6_mask_memory(struct agp_bridge_data *bridge, dma_addr_t addr, int type) { - /* Shift high bits down */ - addr |= (addr >> 28) & 0xff; + /* gen6 has bit11-4 for physical addr bit39-32 */ + addr |= (addr >> 28) & 0xff0; /* Type checking must be done elsewhere */ return addr | bridge->driver->masks[type].mask; -- cgit v0.10.2 From 93f5f7f1249e76a5e8afbdab53f90b10c41fdb61 Mon Sep 17 00:00:00 2001 From: Zhenyu Wang Date: Fri, 27 Aug 2010 11:06:48 +0800 Subject: agp/intel: use #ifdef idiom for intel-agp.h Signed-off-by: Zhenyu Wang Cc: stable@kernel.org Signed-off-by: Chris Wilson diff --git a/drivers/char/agp/intel-agp.h b/drivers/char/agp/intel-agp.h index 08d4753..78124a8 100644 --- a/drivers/char/agp/intel-agp.h +++ b/drivers/char/agp/intel-agp.h @@ -1,6 +1,8 @@ /* * Common Intel AGPGART and GTT definitions. */ +#ifndef _INTEL_AGP_H +#define _INTEL_AGP_H /* Intel registers */ #define INTEL_APSIZE 0xb4 @@ -244,3 +246,5 @@ agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_IRONLAKE_MA_HB || \ agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_IRONLAKE_MC2_HB || \ IS_SNB) + +#endif -- cgit v0.10.2 From f8f235e5bbf4e61f3e0886a44afb1dc4cfe8f337 Mon Sep 17 00:00:00 2001 From: Zhenyu Wang Date: Fri, 27 Aug 2010 11:08:57 +0800 Subject: agp/intel: Fix cache control for Sandybridge Sandybridge GTT has new cache control bits in PTE, which controls graphics page cache in LLC or LLC/MLC, so we need to extend the mask function to respect the new bits. And set cache control to always LLC only by default on Gen6. Signed-off-by: Zhenyu Wang Cc: stable@kernel.org Signed-off-by: Chris Wilson diff --git a/drivers/char/agp/intel-agp.c b/drivers/char/agp/intel-agp.c index 710af89..74461d1 100644 --- a/drivers/char/agp/intel-agp.c +++ b/drivers/char/agp/intel-agp.c @@ -12,6 +12,7 @@ #include #include "agp.h" #include "intel-agp.h" +#include #include "intel-gtt.c" diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/intel-gtt.c index 7f35854..64b1055 100644 --- a/drivers/char/agp/intel-gtt.c +++ b/drivers/char/agp/intel-gtt.c @@ -49,6 +49,26 @@ static struct gatt_mask intel_i810_masks[] = .type = INTEL_AGP_CACHED_MEMORY} }; +#define INTEL_AGP_UNCACHED_MEMORY 0 +#define INTEL_AGP_CACHED_MEMORY_LLC 1 +#define INTEL_AGP_CACHED_MEMORY_LLC_GFDT 2 +#define INTEL_AGP_CACHED_MEMORY_LLC_MLC 3 +#define INTEL_AGP_CACHED_MEMORY_LLC_MLC_GFDT 4 + +static struct gatt_mask intel_gen6_masks[] = +{ + {.mask = I810_PTE_VALID | GEN6_PTE_UNCACHED, + .type = INTEL_AGP_UNCACHED_MEMORY }, + {.mask = I810_PTE_VALID | GEN6_PTE_LLC, + .type = INTEL_AGP_CACHED_MEMORY_LLC }, + {.mask = I810_PTE_VALID | GEN6_PTE_LLC | GEN6_PTE_GFDT, + .type = INTEL_AGP_CACHED_MEMORY_LLC_GFDT }, + {.mask = I810_PTE_VALID | GEN6_PTE_LLC_MLC, + .type = INTEL_AGP_CACHED_MEMORY_LLC_MLC }, + {.mask = I810_PTE_VALID | GEN6_PTE_LLC_MLC | GEN6_PTE_GFDT, + .type = INTEL_AGP_CACHED_MEMORY_LLC_MLC_GFDT }, +}; + static struct _intel_private { struct pci_dev *pcidev; /* device one */ u8 __iomem *registers; @@ -178,13 +198,6 @@ static void intel_agp_insert_sg_entries(struct agp_memory *mem, off_t pg_start, int mask_type) { int i, j; - u32 cache_bits = 0; - - if (agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_SANDYBRIDGE_HB || - agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_HB) - { - cache_bits = GEN6_PTE_LLC_MLC; - } for (i = 0, j = pg_start; i < mem->page_count; i++, j++) { writel(agp_bridge->driver->mask_memory(agp_bridge, @@ -317,6 +330,23 @@ static int intel_i830_type_to_mask_type(struct agp_bridge_data *bridge, return 0; } +static int intel_gen6_type_to_mask_type(struct agp_bridge_data *bridge, + int type) +{ + unsigned int type_mask = type & ~AGP_USER_CACHED_MEMORY_GFDT; + unsigned int gfdt = type & AGP_USER_CACHED_MEMORY_GFDT; + + if (type_mask == AGP_USER_UNCACHED_MEMORY) + return INTEL_AGP_UNCACHED_MEMORY; + else if (type_mask == AGP_USER_CACHED_MEMORY_LLC_MLC) + return gfdt ? INTEL_AGP_CACHED_MEMORY_LLC_MLC_GFDT : + INTEL_AGP_CACHED_MEMORY_LLC_MLC; + else /* set 'normal'/'cached' to LLC by default */ + return gfdt ? INTEL_AGP_CACHED_MEMORY_LLC_GFDT : + INTEL_AGP_CACHED_MEMORY_LLC; +} + + static int intel_i810_insert_entries(struct agp_memory *mem, off_t pg_start, int type) { @@ -1163,7 +1193,7 @@ static int intel_i915_insert_entries(struct agp_memory *mem, off_t pg_start, mask_type = agp_bridge->driver->agp_type_to_mask_type(agp_bridge, type); - if (mask_type != 0 && mask_type != AGP_PHYS_MEMORY && + if (!IS_SNB && mask_type != 0 && mask_type != AGP_PHYS_MEMORY && mask_type != INTEL_AGP_CACHED_MEMORY) goto out_err; @@ -1563,7 +1593,7 @@ static const struct agp_bridge_driver intel_gen6_driver = { .fetch_size = intel_i9xx_fetch_size, .cleanup = intel_i915_cleanup, .mask_memory = intel_gen6_mask_memory, - .masks = intel_i810_masks, + .masks = intel_gen6_masks, .agp_enable = intel_i810_agp_enable, .cache_flush = global_cache_flush, .create_gatt_table = intel_i965_create_gatt_table, @@ -1576,7 +1606,7 @@ static const struct agp_bridge_driver intel_gen6_driver = { .agp_alloc_pages = agp_generic_alloc_pages, .agp_destroy_page = agp_generic_destroy_page, .agp_destroy_pages = agp_generic_destroy_pages, - .agp_type_to_mask_type = intel_i830_type_to_mask_type, + .agp_type_to_mask_type = intel_gen6_type_to_mask_type, .chipset_flush = intel_i915_chipset_flush, #ifdef USE_PCI_DMA_API .agp_map_page = intel_agp_map_page, diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 748c263..16fca1d 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -34,6 +34,7 @@ #include #include #include +#include static uint32_t i915_gem_get_gtt_alignment(struct drm_gem_object *obj); static int i915_gem_object_flush_gpu_write_domain(struct drm_gem_object *obj); diff --git a/include/linux/intel-gtt.h b/include/linux/intel-gtt.h new file mode 100644 index 0000000..1d19ab2 --- /dev/null +++ b/include/linux/intel-gtt.h @@ -0,0 +1,20 @@ +/* + * Common Intel AGPGART and GTT definitions. + */ +#ifndef _INTEL_GTT_H +#define _INTEL_GTT_H + +#include + +/* This is for Intel only GTT controls. + * + * Sandybridge: AGP_USER_CACHED_MEMORY default to LLC only + */ + +#define AGP_USER_CACHED_MEMORY_LLC_MLC (AGP_USER_TYPES + 2) +#define AGP_USER_UNCACHED_MEMORY (AGP_USER_TYPES + 4) + +/* flag for GFDT type */ +#define AGP_USER_CACHED_MEMORY_GFDT (1 << 3) + +#endif -- cgit v0.10.2 From a69ffdbfcba8eabf2ca9d384b578e6f28b339c61 Mon Sep 17 00:00:00 2001 From: Zhenyu Wang Date: Mon, 30 Aug 2010 16:12:42 +0800 Subject: drm/i915: Enable MI_FLUSH on Sandybridge MI_FLUSH is being deprecated, but still available on Sandybridge. Make sure it's enabled as userspace still uses MI_FLUSH. Signed-off-by: Zhenyu Wang Cc: stable@kernel.org Signed-off-by: Chris Wilson diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 67e3ec1..d094e91 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -319,6 +319,7 @@ #define MI_MODE 0x0209c # define VS_TIMER_DISPATCH (1 << 6) +# define MI_FLUSH_ENABLE (1 << 11) #define SCPD0 0x0209c /* 915+ only */ #define IER 0x020a0 diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 51e9c9e7..cb3508f 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -220,9 +220,13 @@ static int init_render_ring(struct drm_device *dev, { drm_i915_private_t *dev_priv = dev->dev_private; int ret = init_ring_common(dev, ring); + int mode; + if (IS_I9XX(dev) && !IS_GEN3(dev)) { - I915_WRITE(MI_MODE, - (VS_TIMER_DISPATCH) << 16 | VS_TIMER_DISPATCH); + mode = VS_TIMER_DISPATCH << 16 | VS_TIMER_DISPATCH; + if (IS_GEN6(dev)) + mode |= MI_FLUSH_ENABLE << 16 | MI_FLUSH_ENABLE; + I915_WRITE(MI_MODE, mode); } return ret; } -- cgit v0.10.2 From 8554048070906579ec9fa19ac381deddd2d7b155 Mon Sep 17 00:00:00 2001 From: Zhenyu Wang Date: Tue, 7 Sep 2010 13:45:32 +0800 Subject: intel_agp,i915: Add more sandybridge graphics device ids New pci ids for GT2 and GT2+ on desktop and mobile sandybridge, and graphics device ids for server sandybridge. Also rename original ids string to reflect GT1 version. Signed-off-by: Zhenyu Wang Cc: stable@kernel.org Signed-off-by: Chris Wilson diff --git a/drivers/char/agp/intel-agp.c b/drivers/char/agp/intel-agp.c index 74461d1..eab58db 100644 --- a/drivers/char/agp/intel-agp.c +++ b/drivers/char/agp/intel-agp.c @@ -816,11 +816,19 @@ static const struct intel_driver_description { "HD Graphics", NULL, &intel_i965_driver }, { PCI_DEVICE_ID_INTEL_IRONLAKE_MC2_HB, PCI_DEVICE_ID_INTEL_IRONLAKE_M_IG, "HD Graphics", NULL, &intel_i965_driver }, - { PCI_DEVICE_ID_INTEL_SANDYBRIDGE_HB, PCI_DEVICE_ID_INTEL_SANDYBRIDGE_IG, + { PCI_DEVICE_ID_INTEL_SANDYBRIDGE_HB, PCI_DEVICE_ID_INTEL_SANDYBRIDGE_GT1_IG, "Sandybridge", NULL, &intel_gen6_driver }, - { PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_HB, PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_IG, + { PCI_DEVICE_ID_INTEL_SANDYBRIDGE_HB, PCI_DEVICE_ID_INTEL_SANDYBRIDGE_GT2_IG, "Sandybridge", NULL, &intel_gen6_driver }, - { PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_HB, PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_D0_IG, + { PCI_DEVICE_ID_INTEL_SANDYBRIDGE_HB, PCI_DEVICE_ID_INTEL_SANDYBRIDGE_GT2_PLUS_IG, + "Sandybridge", NULL, &intel_gen6_driver }, + { PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_HB, PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_GT1_IG, + "Sandybridge", NULL, &intel_gen6_driver }, + { PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_HB, PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_GT2_IG, + "Sandybridge", NULL, &intel_gen6_driver }, + { PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_HB, PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_GT2_PLUS_IG, + "Sandybridge", NULL, &intel_gen6_driver }, + { PCI_DEVICE_ID_INTEL_SANDYBRIDGE_S_HB, PCI_DEVICE_ID_INTEL_SANDYBRIDGE_S_IG, "Sandybridge", NULL, &intel_gen6_driver }, { 0, 0, NULL, NULL, NULL } }; @@ -1045,6 +1053,7 @@ static struct pci_device_id agp_intel_pci_table[] = { ID(PCI_DEVICE_ID_INTEL_IRONLAKE_MC2_HB), ID(PCI_DEVICE_ID_INTEL_SANDYBRIDGE_HB), ID(PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_HB), + ID(PCI_DEVICE_ID_INTEL_SANDYBRIDGE_S_HB), { } }; diff --git a/drivers/char/agp/intel-agp.h b/drivers/char/agp/intel-agp.h index 78124a8..ee189c7 100644 --- a/drivers/char/agp/intel-agp.h +++ b/drivers/char/agp/intel-agp.h @@ -202,11 +202,16 @@ #define PCI_DEVICE_ID_INTEL_IRONLAKE_MA_HB 0x0062 #define PCI_DEVICE_ID_INTEL_IRONLAKE_MC2_HB 0x006a #define PCI_DEVICE_ID_INTEL_IRONLAKE_M_IG 0x0046 -#define PCI_DEVICE_ID_INTEL_SANDYBRIDGE_HB 0x0100 -#define PCI_DEVICE_ID_INTEL_SANDYBRIDGE_IG 0x0102 -#define PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_HB 0x0104 -#define PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_IG 0x0106 -#define PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_D0_IG 0x0126 +#define PCI_DEVICE_ID_INTEL_SANDYBRIDGE_HB 0x0100 /* Desktop */ +#define PCI_DEVICE_ID_INTEL_SANDYBRIDGE_GT1_IG 0x0102 +#define PCI_DEVICE_ID_INTEL_SANDYBRIDGE_GT2_IG 0x0112 +#define PCI_DEVICE_ID_INTEL_SANDYBRIDGE_GT2_PLUS_IG 0x0122 +#define PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_HB 0x0104 /* Mobile */ +#define PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_GT1_IG 0x0106 +#define PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_GT2_IG 0x0116 +#define PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_GT2_PLUS_IG 0x0126 +#define PCI_DEVICE_ID_INTEL_SANDYBRIDGE_S_HB 0x0108 /* Server */ +#define PCI_DEVICE_ID_INTEL_SANDYBRIDGE_S_IG 0x010A /* cover 915 and 945 variants */ #define IS_I915 (agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_E7221_HB || \ @@ -233,7 +238,8 @@ agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_PINEVIEW_HB) #define IS_SNB (agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_SANDYBRIDGE_HB || \ - agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_HB) + agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_HB || \ + agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_SANDYBRIDGE_S_HB) #define IS_G4X (agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_EAGLELAKE_HB || \ agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_Q45_HB || \ diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/intel-gtt.c index 64b1055..75e0a34 100644 --- a/drivers/char/agp/intel-gtt.c +++ b/drivers/char/agp/intel-gtt.c @@ -618,8 +618,7 @@ static void intel_i830_init_gtt_entries(void) gtt_entries = 0; break; } - } else if (agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_SANDYBRIDGE_HB || - agp_bridge->dev->device == PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_HB) { + } else if (IS_SNB) { /* * SandyBridge has new memory control reg at 0x50.w */ @@ -1389,6 +1388,7 @@ static void intel_i965_get_gtt_range(int *gtt_offset, int *gtt_size) break; case PCI_DEVICE_ID_INTEL_SANDYBRIDGE_HB: case PCI_DEVICE_ID_INTEL_SANDYBRIDGE_M_HB: + case PCI_DEVICE_ID_INTEL_SANDYBRIDGE_S_HB: *gtt_offset = MB(2); pci_read_config_word(intel_private.pcidev, SNB_GMCH_CTRL, &snb_gmch_ctl); diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 5363985..216deb5 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -175,8 +175,12 @@ static const struct pci_device_id pciidlist[] = { /* aka */ INTEL_VGA_DEVICE(0x0042, &intel_ironlake_d_info), INTEL_VGA_DEVICE(0x0046, &intel_ironlake_m_info), INTEL_VGA_DEVICE(0x0102, &intel_sandybridge_d_info), + INTEL_VGA_DEVICE(0x0112, &intel_sandybridge_d_info), + INTEL_VGA_DEVICE(0x0122, &intel_sandybridge_d_info), INTEL_VGA_DEVICE(0x0106, &intel_sandybridge_m_info), + INTEL_VGA_DEVICE(0x0116, &intel_sandybridge_m_info), INTEL_VGA_DEVICE(0x0126, &intel_sandybridge_m_info), + INTEL_VGA_DEVICE(0x010A, &intel_sandybridge_d_info), {0, 0, 0} }; -- cgit v0.10.2 From 595afaf9e6ee1b48e13ec4b8bcc8c7dee888161a Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Tue, 7 Sep 2010 13:42:41 +0200 Subject: fuse: flush background queue on connection close David Bartly reported that fuse can hang in fuse_get_req_nofail() when the connection to the filesystem server is no longer active. If bg_queue is not empty then flush_bg_queue() called from request_end() can put more requests on to the pending queue. If this happens while ending requests on the processing queue then those background requests will be queued to the pending list and never ended. Another problem is that fuse_dev_release() didn't wake up processes sleeping on blocked_waitq. Solve this by: a) flushing the background queue before calling end_requests() on the pending and processing queues b) setting blocked = 0 and waking up processes waiting on blocked_waitq() Thanks to David for an excellent bug report. Reported-by: David Bartley Signed-off-by: Miklos Szeredi CC: stable@kernel.org diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 69ad053..b4fc47f 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -1769,6 +1769,14 @@ __acquires(&fc->lock) } } +static void end_queued_requests(struct fuse_conn *fc) +{ + fc->max_background = UINT_MAX; + flush_bg_queue(fc); + end_requests(fc, &fc->pending); + end_requests(fc, &fc->processing); +} + /* * Abort all requests. * @@ -1795,8 +1803,7 @@ void fuse_abort_conn(struct fuse_conn *fc) fc->connected = 0; fc->blocked = 0; end_io_requests(fc); - end_requests(fc, &fc->pending); - end_requests(fc, &fc->processing); + end_queued_requests(fc); wake_up_all(&fc->waitq); wake_up_all(&fc->blocked_waitq); kill_fasync(&fc->fasync, SIGIO, POLL_IN); @@ -1811,8 +1818,9 @@ int fuse_dev_release(struct inode *inode, struct file *file) if (fc) { spin_lock(&fc->lock); fc->connected = 0; - end_requests(fc, &fc->pending); - end_requests(fc, &fc->processing); + fc->blocked = 0; + end_queued_requests(fc); + wake_up_all(&fc->blocked_waitq); spin_unlock(&fc->lock); fuse_conn_put(fc); } -- cgit v0.10.2 From b9ca67b2ddf021491a3741d9555e8ff59b2175ba Mon Sep 17 00:00:00 2001 From: Miklos Szeredi Date: Tue, 7 Sep 2010 13:42:41 +0200 Subject: fuse: fix lock annotations Sparse doesn't understand lock annotations of the form __releases(&foo->lock). Change them to __releases(foo->lock). Same for __acquires(). Signed-off-by: Miklos Szeredi diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index b4fc47f..d367af1 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -276,7 +276,7 @@ static void flush_bg_queue(struct fuse_conn *fc) * Called with fc->lock, unlocks it */ static void request_end(struct fuse_conn *fc, struct fuse_req *req) -__releases(&fc->lock) +__releases(fc->lock) { void (*end) (struct fuse_conn *, struct fuse_req *) = req->end; req->end = NULL; @@ -306,8 +306,8 @@ __releases(&fc->lock) static void wait_answer_interruptible(struct fuse_conn *fc, struct fuse_req *req) -__releases(&fc->lock) -__acquires(&fc->lock) +__releases(fc->lock) +__acquires(fc->lock) { if (signal_pending(current)) return; @@ -325,8 +325,8 @@ static void queue_interrupt(struct fuse_conn *fc, struct fuse_req *req) } static void request_wait_answer(struct fuse_conn *fc, struct fuse_req *req) -__releases(&fc->lock) -__acquires(&fc->lock) +__releases(fc->lock) +__acquires(fc->lock) { if (!fc->no_interrupt) { /* Any signal may interrupt this */ @@ -905,8 +905,8 @@ static int request_pending(struct fuse_conn *fc) /* Wait until a request is available on the pending list */ static void request_wait(struct fuse_conn *fc) -__releases(&fc->lock) -__acquires(&fc->lock) +__releases(fc->lock) +__acquires(fc->lock) { DECLARE_WAITQUEUE(wait, current); @@ -934,7 +934,7 @@ __acquires(&fc->lock) */ static int fuse_read_interrupt(struct fuse_conn *fc, struct fuse_copy_state *cs, size_t nbytes, struct fuse_req *req) -__releases(&fc->lock) +__releases(fc->lock) { struct fuse_in_header ih; struct fuse_interrupt_in arg; @@ -1720,8 +1720,8 @@ static unsigned fuse_dev_poll(struct file *file, poll_table *wait) * This function releases and reacquires fc->lock */ static void end_requests(struct fuse_conn *fc, struct list_head *head) -__releases(&fc->lock) -__acquires(&fc->lock) +__releases(fc->lock) +__acquires(fc->lock) { while (!list_empty(head)) { struct fuse_req *req; @@ -1744,8 +1744,8 @@ __acquires(&fc->lock) * locked). */ static void end_io_requests(struct fuse_conn *fc) -__releases(&fc->lock) -__acquires(&fc->lock) +__releases(fc->lock) +__acquires(fc->lock) { while (!list_empty(&fc->io)) { struct fuse_req *req = @@ -1770,6 +1770,8 @@ __acquires(&fc->lock) } static void end_queued_requests(struct fuse_conn *fc) +__releases(fc->lock) +__acquires(fc->lock) { fc->max_background = UINT_MAX; flush_bg_queue(fc); diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 147c1f7..c822458 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -1144,8 +1144,8 @@ static void fuse_writepage_finish(struct fuse_conn *fc, struct fuse_req *req) /* Called under fc->lock, may release and reacquire it */ static void fuse_send_writepage(struct fuse_conn *fc, struct fuse_req *req) -__releases(&fc->lock) -__acquires(&fc->lock) +__releases(fc->lock) +__acquires(fc->lock) { struct fuse_inode *fi = get_fuse_inode(req->inode); loff_t size = i_size_read(req->inode); @@ -1183,8 +1183,8 @@ __acquires(&fc->lock) * Called with fc->lock */ void fuse_flush_writepages(struct inode *inode) -__releases(&fc->lock) -__acquires(&fc->lock) +__releases(fc->lock) +__acquires(fc->lock) { struct fuse_conn *fc = get_fuse_conn(inode); struct fuse_inode *fi = get_fuse_inode(inode); -- cgit v0.10.2 From 7e443312403ad1ff40ef3177590e96d1fe747c79 Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Tue, 7 Sep 2010 11:27:52 -0400 Subject: [SCSI] sd: fix medium-removal bug Commit 409f3499a2cfcd1e9c2857c53af7fcce069f027f (scsi/sd: remove big kernel lock) introduced a bug in the sd_release routine. Medium removal should be allowed when the number of open file references drops to 0, not when it becomes non-zero. This patch (as1414) adjusts the test to fix the bug. Signed-off-by: Alan Stern Signed-off-by: James Bottomley diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index cd71f46..ffa0689 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -870,7 +870,7 @@ static int sd_release(struct gendisk *disk, fmode_t mode) SCSI_LOG_HLQUEUE(3, sd_printk(KERN_INFO, sdkp, "sd_release\n")); - if (atomic_dec_return(&sdkp->openers) && sdev->removable) { + if (atomic_dec_return(&sdkp->openers) == 0 && sdev->removable) { if (scsi_block_when_processing_errors(sdev)) scsi_set_medium_removal(sdev, SCSI_REMOVAL_ALLOW); } -- cgit v0.10.2 From 8d2602e0778299e2d6084f03086b716d6e7a1e1e Mon Sep 17 00:00:00 2001 From: Nicolas Ferre Date: Fri, 20 Aug 2010 16:44:33 +0200 Subject: AT91: change dma resource index Reported-by: Dan Liang Signed-off-by: Nicolas Ferre diff --git a/arch/arm/mach-at91/at91sam9g45_devices.c b/arch/arm/mach-at91/at91sam9g45_devices.c index 809114d..501ac6f 100644 --- a/arch/arm/mach-at91/at91sam9g45_devices.c +++ b/arch/arm/mach-at91/at91sam9g45_devices.c @@ -46,7 +46,7 @@ static struct resource hdmac_resources[] = { .end = AT91_BASE_SYS + AT91_DMA + SZ_512 - 1, .flags = IORESOURCE_MEM, }, - [2] = { + [1] = { .start = AT91SAM9G45_ID_DMA, .end = AT91SAM9G45_ID_DMA, .flags = IORESOURCE_IRQ, -- cgit v0.10.2 From 7a2e8a8faab76386d8eaae9ded739ee5615be174 Mon Sep 17 00:00:00 2001 From: Valerie Aurora Date: Thu, 26 Aug 2010 11:07:22 -0700 Subject: VFS: Sanity check mount flags passed to change_mnt_propagation() Sanity check the flags passed to change_mnt_propagation(). Exactly one flag should be set. Return EINVAL otherwise. Userspace can pass in arbitrary combinations of MS_* flags to mount(). do_change_type() is called if any of MS_SHARED, MS_PRIVATE, MS_SLAVE, or MS_UNBINDABLE is set. do_change_type() clears MS_REC and then calls change_mnt_propagation() with the rest of the user-supplied flags. change_mnt_propagation() clearly assumes only one flag is set but do_change_type() does not check that this is true. For example, mount() with flags MS_SHARED | MS_RDONLY does not actually make the mount shared or read-only but does clear MNT_UNBINDABLE. Signed-off-by: Valerie Aurora Signed-off-by: Linus Torvalds diff --git a/fs/namespace.c b/fs/namespace.c index de402eb..a72eaab 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1484,13 +1484,30 @@ out_unlock: } /* + * Sanity check the flags to change_mnt_propagation. + */ + +static int flags_to_propagation_type(int flags) +{ + int type = flags & ~MS_REC; + + /* Fail if any non-propagation flags are set */ + if (type & ~(MS_SHARED | MS_PRIVATE | MS_SLAVE | MS_UNBINDABLE)) + return 0; + /* Only one propagation flag should be set */ + if (!is_power_of_2(type)) + return 0; + return type; +} + +/* * recursively change the type of the mountpoint. */ static int do_change_type(struct path *path, int flag) { struct vfsmount *m, *mnt = path->mnt; int recurse = flag & MS_REC; - int type = flag & ~MS_REC; + int type; int err = 0; if (!capable(CAP_SYS_ADMIN)) @@ -1499,6 +1516,10 @@ static int do_change_type(struct path *path, int flag) if (path->dentry != path->mnt->mnt_root) return -EINVAL; + type = flags_to_propagation_type(flag); + if (!type) + return -EINVAL; + down_write(&namespace_sem); if (type == MS_SHARED) { err = invent_group_ids(mnt, recurse); -- cgit v0.10.2 From 3ab61eb9fd751336d1b3ad50ec7704bd3b62213f Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 31 Aug 2010 16:52:11 +0100 Subject: h8300: IRQ flags should be stored in an unsigned long Fix h8300's asm/atomic.h to store the IRQ flags in an unsigned long to deal with warnings of the following type: arch/h8300/include/asm/atomic.h: In function 'atomic_add_return': arch/h8300/include/asm/atomic.h:22: warning: comparison of distinct pointer types lacks a cast arch/h8300/include/asm/atomic.h:24: warning: comparison of distinct pointer types lacks a cast Signed-off-by: David Howells Signed-off-by: Linus Torvalds diff --git a/arch/h8300/include/asm/atomic.h b/arch/h8300/include/asm/atomic.h index e936804..984221a 100644 --- a/arch/h8300/include/asm/atomic.h +++ b/arch/h8300/include/asm/atomic.h @@ -18,7 +18,8 @@ static __inline__ int atomic_add_return(int i, atomic_t *v) { - int ret,flags; + unsigned long flags; + int ret; local_irq_save(flags); ret = v->counter += i; local_irq_restore(flags); @@ -30,7 +31,8 @@ static __inline__ int atomic_add_return(int i, atomic_t *v) static __inline__ int atomic_sub_return(int i, atomic_t *v) { - int ret,flags; + unsigned long flags; + int ret; local_irq_save(flags); ret = v->counter -= i; local_irq_restore(flags); @@ -42,7 +44,8 @@ static __inline__ int atomic_sub_return(int i, atomic_t *v) static __inline__ int atomic_inc_return(atomic_t *v) { - int ret,flags; + unsigned long flags; + int ret; local_irq_save(flags); v->counter++; ret = v->counter; @@ -64,7 +67,8 @@ static __inline__ int atomic_inc_return(atomic_t *v) static __inline__ int atomic_dec_return(atomic_t *v) { - int ret,flags; + unsigned long flags; + int ret; local_irq_save(flags); --v->counter; ret = v->counter; @@ -76,7 +80,8 @@ static __inline__ int atomic_dec_return(atomic_t *v) static __inline__ int atomic_dec_and_test(atomic_t *v) { - int ret,flags; + unsigned long flags; + int ret; local_irq_save(flags); --v->counter; ret = v->counter; -- cgit v0.10.2 From b857189d94e0d8a05943349618c4f84dbc57144c Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 31 Aug 2010 16:52:16 +0100 Subject: h8300: Fix die() Fix h8300's die() to take care of a number of problems: CC arch/h8300/kernel/traps.o In file included from arch/h8300/include/asm/bitops.h:10, from include/linux/bitops.h:22, from include/linux/kernel.h:17, from include/linux/sched.h:54, from arch/h8300/kernel/traps.c:18: arch/h8300/include/asm/system.h:136: warning: 'struct pt_regs' declared inside parameter list arch/h8300/include/asm/system.h:136: warning: its scope is only this definition or declaration, which is probably not what you want arch/h8300/kernel/traps.c:100: error: conflicting types for 'die' arch/h8300/include/asm/system.h:136: error: previous declaration of 'die' was here make[2]: *** [arch/h8300/kernel/traps.o] Error 1 Signed-off-by: David Howells Signed-off-by: Linus Torvalds diff --git a/arch/h8300/include/asm/system.h b/arch/h8300/include/asm/system.h index d98d976..16bf156 100644 --- a/arch/h8300/include/asm/system.h +++ b/arch/h8300/include/asm/system.h @@ -3,6 +3,8 @@ #include +struct pt_regs; + /* * switch_to(n) should switch tasks to task ptr, first checking that * ptr isn't the current task, in which case it does nothing. This @@ -155,6 +157,6 @@ static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int siz #define arch_align_stack(x) (x) -void die(char *str, struct pt_regs *fp, unsigned long err); +extern void die(const char *str, struct pt_regs *fp, unsigned long err); #endif /* _H8300_SYSTEM_H */ diff --git a/arch/h8300/kernel/traps.c b/arch/h8300/kernel/traps.c index 3c0b66b..dfa05bd 100644 --- a/arch/h8300/kernel/traps.c +++ b/arch/h8300/kernel/traps.c @@ -96,7 +96,7 @@ static void dump(struct pt_regs *fp) printk("\n\n"); } -void die(char *str, struct pt_regs *fp, unsigned long err) +void die(const char *str, struct pt_regs *fp, unsigned long err) { static int diecount; -- cgit v0.10.2 From 1e72910e248fb94839cd63c0c62107b1f15716f9 Mon Sep 17 00:00:00 2001 From: David Howells Date: Tue, 31 Aug 2010 16:52:21 +0100 Subject: h8300: Fix missing consts in kernel_execve() Fix missing consts in h8300's kernel_execve(): arch/h8300/kernel/sys_h8300.c: In function 'kernel_execve': arch/h8300/kernel/sys_h8300.c:59: warning: initialization from incompatible pointer type arch/h8300/kernel/sys_h8300.c:60: warning: initialization from incompatible pointer type Signed-off-by: David Howells Signed-off-by: Linus Torvalds diff --git a/arch/h8300/kernel/sys_h8300.c b/arch/h8300/kernel/sys_h8300.c index dc1ac02..aaf5e5a 100644 --- a/arch/h8300/kernel/sys_h8300.c +++ b/arch/h8300/kernel/sys_h8300.c @@ -56,8 +56,8 @@ int kernel_execve(const char *filename, const char *const envp[]) { register long res __asm__("er0"); - register char *const *_c __asm__("er3") = envp; - register char *const *_b __asm__("er2") = argv; + register const char *const *_c __asm__("er3") = envp; + register const char *const *_b __asm__("er2") = argv; register const char * _a __asm__("er1") = filename; __asm__ __volatile__ ("mov.l %1,er0\n\t" "trapa #0\n\t" -- cgit v0.10.2 From e6ba59bcae6968ee18ef5a237a8020a0ea331ae1 Mon Sep 17 00:00:00 2001 From: Greg Ungerer Date: Fri, 3 Sep 2010 11:15:19 +1000 Subject: m68knommu: fix missing linker segments Recent changes to linker segments that hold per-cpu data broke linking for m68knommu targets: LD vmlinux /usr/local/bin/m68k-uclinux-ld.real: error: no memory region specified for loadable section `.data..shared_aligned' Add missing segments into the m68knommu linker script. Signed-off-by: Greg Ungerer Signed-off-by: Linus Torvalds diff --git a/arch/m68knommu/kernel/vmlinux.lds.S b/arch/m68knommu/kernel/vmlinux.lds.S index a91b271..ef33213 100644 --- a/arch/m68knommu/kernel/vmlinux.lds.S +++ b/arch/m68knommu/kernel/vmlinux.lds.S @@ -150,6 +150,8 @@ SECTIONS { _sdata = . ; DATA_DATA CACHELINE_ALIGNED_DATA(32) + PAGE_ALIGNED_DATA(PAGE_SIZE) + *(.data..shared_aligned) INIT_TASK_DATA(THREAD_SIZE) _edata = . ; } > DATA -- cgit v0.10.2 From bc41606aefa8b17000619f510d5809e6c4003d65 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Tue, 7 Sep 2010 21:51:02 +0100 Subject: Revert "drm/i915: Enable RC6 on Ironlake." This reverts commit ce17178094f368d9e3f39b2cb4303da5ed633dd4. This commit has been independently bisected a few times as being the cause of a s2ram failure. Reported-and-tested-by: Kyle McMartin Reported-and-tested-by: Andy Isaacson Cc: Zou Nan hai Signed-off-by: Chris Wilson diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 40cc5da..e0d1952 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -5696,8 +5696,7 @@ void intel_init_clock_gating(struct drm_device *dev) ILK_DPFC_DIS2 | ILK_CLK_FBC); } - if (IS_GEN6(dev)) - return; + return; } else if (IS_G4X(dev)) { uint32_t dspclk_gate; I915_WRITE(RENCLK_GATE_D1, 0); @@ -5758,11 +5757,9 @@ void intel_init_clock_gating(struct drm_device *dev) OUT_RING(MI_FLUSH); ADVANCE_LP_RING(); } - } else { + } else DRM_DEBUG_KMS("Failed to allocate render context." - "Disable RC6\n"); - return; - } + "Disable RC6\n"); } if (I915_HAS_RC6(dev) && drm_core_check_feature(dev, DRIVER_MODESET)) { -- cgit v0.10.2 From c4433be6e19e3680727f3f89c938a22e7b789b43 Mon Sep 17 00:00:00 2001 From: Giuseppe Cavallaro Date: Mon, 6 Sep 2010 05:02:11 +0200 Subject: stmmac: fix sleep inside atomic We cannot use spinlock when kmalloc is invoked with GFP_KERNEL flag because it can sleep. So this patch reviews the usage of spinlock within the stmmac_resume function avoing this bug. Signed-off-by: Giuseppe Cavallaro Reported-by: Jiri Slaby Signed-off-by: David S. Miller diff --git a/drivers/net/stmmac/stmmac_main.c b/drivers/net/stmmac/stmmac_main.c index bbb7951..ea0461e 100644 --- a/drivers/net/stmmac/stmmac_main.c +++ b/drivers/net/stmmac/stmmac_main.c @@ -1865,15 +1865,15 @@ static int stmmac_resume(struct platform_device *pdev) if (!netif_running(dev)) return 0; - spin_lock(&priv->lock); - if (priv->shutdown) { /* Re-open the interface and re-init the MAC/DMA - and the rings. */ + and the rings (i.e. on hibernation stage) */ stmmac_open(dev); - goto out_resume; + return 0; } + spin_lock(&priv->lock); + /* Power Down bit, into the PM register, is cleared * automatically as soon as a magic packet or a Wake-up frame * is received. Anyway, it's better to manually clear @@ -1901,7 +1901,6 @@ static int stmmac_resume(struct platform_device *pdev) netif_start_queue(dev); -out_resume: spin_unlock(&priv->lock); return 0; } -- cgit v0.10.2 From cb32f2a0d194212e4e750a8cdedcc610c9ca4876 Mon Sep 17 00:00:00 2001 From: Jiri Bohac Date: Thu, 2 Sep 2010 05:45:54 +0000 Subject: bonding: Fix jiffies overflow problems (again) The time_before_eq()/time_after_eq() functions operate on unsigned long and only work if the difference between the two compared values is smaller than half the range of unsigned long (31 bits on i386). Some of the variables (slave->jiffies, dev->trans_start, dev->last_rx) used by bonding store a copy of jiffies and may not be updated for a long time. With HZ=1000, time_before_eq()/time_after_eq() will start giving bad results after ~25 days. jiffies will never be before slave->jiffies, dev->trans_start, dev->last_rx by more than possibly a couple ticks caused by preemption of this code. This allows us to detect/prevent these overflows by replacing time_before_eq()/time_after_eq() with time_in_range(). Signed-off-by: Jiri Bohac Signed-off-by: Jean Delvare Signed-off-by: David S. Miller diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 2cc4cfc..3b16f62 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -2797,9 +2797,15 @@ void bond_loadbalance_arp_mon(struct work_struct *work) * so it can wait */ bond_for_each_slave(bond, slave, i) { + unsigned long trans_start = dev_trans_start(slave->dev); + if (slave->link != BOND_LINK_UP) { - if (time_before_eq(jiffies, dev_trans_start(slave->dev) + delta_in_ticks) && - time_before_eq(jiffies, slave->dev->last_rx + delta_in_ticks)) { + if (time_in_range(jiffies, + trans_start - delta_in_ticks, + trans_start + delta_in_ticks) && + time_in_range(jiffies, + slave->dev->last_rx - delta_in_ticks, + slave->dev->last_rx + delta_in_ticks)) { slave->link = BOND_LINK_UP; slave->state = BOND_STATE_ACTIVE; @@ -2827,8 +2833,12 @@ void bond_loadbalance_arp_mon(struct work_struct *work) * when the source ip is 0, so don't take the link down * if we don't know our ip yet */ - if (time_after_eq(jiffies, dev_trans_start(slave->dev) + 2*delta_in_ticks) || - (time_after_eq(jiffies, slave->dev->last_rx + 2*delta_in_ticks))) { + if (!time_in_range(jiffies, + trans_start - delta_in_ticks, + trans_start + 2 * delta_in_ticks) || + !time_in_range(jiffies, + slave->dev->last_rx - delta_in_ticks, + slave->dev->last_rx + 2 * delta_in_ticks)) { slave->link = BOND_LINK_DOWN; slave->state = BOND_STATE_BACKUP; @@ -2883,13 +2893,16 @@ static int bond_ab_arp_inspect(struct bonding *bond, int delta_in_ticks) { struct slave *slave; int i, commit = 0; + unsigned long trans_start; bond_for_each_slave(bond, slave, i) { slave->new_link = BOND_LINK_NOCHANGE; if (slave->link != BOND_LINK_UP) { - if (time_before_eq(jiffies, slave_last_rx(bond, slave) + - delta_in_ticks)) { + if (time_in_range(jiffies, + slave_last_rx(bond, slave) - delta_in_ticks, + slave_last_rx(bond, slave) + delta_in_ticks)) { + slave->new_link = BOND_LINK_UP; commit++; } @@ -2902,8 +2915,9 @@ static int bond_ab_arp_inspect(struct bonding *bond, int delta_in_ticks) * active. This avoids bouncing, as the last receive * times need a full ARP monitor cycle to be updated. */ - if (!time_after_eq(jiffies, slave->jiffies + - 2 * delta_in_ticks)) + if (time_in_range(jiffies, + slave->jiffies - delta_in_ticks, + slave->jiffies + 2 * delta_in_ticks)) continue; /* @@ -2921,8 +2935,10 @@ static int bond_ab_arp_inspect(struct bonding *bond, int delta_in_ticks) */ if (slave->state == BOND_STATE_BACKUP && !bond->current_arp_slave && - time_after(jiffies, slave_last_rx(bond, slave) + - 3 * delta_in_ticks)) { + !time_in_range(jiffies, + slave_last_rx(bond, slave) - delta_in_ticks, + slave_last_rx(bond, slave) + 3 * delta_in_ticks)) { + slave->new_link = BOND_LINK_DOWN; commit++; } @@ -2933,11 +2949,15 @@ static int bond_ab_arp_inspect(struct bonding *bond, int delta_in_ticks) * - (more than 2*delta since receive AND * the bond has an IP address) */ + trans_start = dev_trans_start(slave->dev); if ((slave->state == BOND_STATE_ACTIVE) && - (time_after_eq(jiffies, dev_trans_start(slave->dev) + - 2 * delta_in_ticks) || - (time_after_eq(jiffies, slave_last_rx(bond, slave) - + 2 * delta_in_ticks)))) { + (!time_in_range(jiffies, + trans_start - delta_in_ticks, + trans_start + 2 * delta_in_ticks) || + !time_in_range(jiffies, + slave_last_rx(bond, slave) - delta_in_ticks, + slave_last_rx(bond, slave) + 2 * delta_in_ticks))) { + slave->new_link = BOND_LINK_DOWN; commit++; } @@ -2956,6 +2976,7 @@ static void bond_ab_arp_commit(struct bonding *bond, int delta_in_ticks) { struct slave *slave; int i; + unsigned long trans_start; bond_for_each_slave(bond, slave, i) { switch (slave->new_link) { @@ -2963,10 +2984,11 @@ static void bond_ab_arp_commit(struct bonding *bond, int delta_in_ticks) continue; case BOND_LINK_UP: + trans_start = dev_trans_start(slave->dev); if ((!bond->curr_active_slave && - time_before_eq(jiffies, - dev_trans_start(slave->dev) + - delta_in_ticks)) || + time_in_range(jiffies, + trans_start - delta_in_ticks, + trans_start + delta_in_ticks)) || bond->curr_active_slave != slave) { slave->link = BOND_LINK_UP; bond->current_arp_slave = NULL; -- cgit v0.10.2 From deabc772f39405054a438d711f408d2d94d26d96 Mon Sep 17 00:00:00 2001 From: Helmut Schaa Date: Fri, 3 Sep 2010 02:39:56 +0000 Subject: net: fix tx queue selection for bridged devices implementing select_queue When a net device is implementing the select_queue callback and is part of a bridge, frames coming from the bridge already have a tx queue associated to the socket (introduced in commit a4ee3ce3293dc931fab19beb472a8bde1295aebe, "net: Use sk_tx_queue_mapping for connected sockets"). The call to sk_tx_queue_get will then return the tx queue used by the bridge instead of calling the select_queue callback. In case of mac80211 this broke QoS which is implemented by using the select_queue callback. Furthermore it introduced problems with rt2x00 because frames with the same TID and RA sometimes appeared on different tx queues which the hw cannot handle correctly. Fix this by always calling select_queue first if it is available and only afterwards use the socket tx queue mapping. Signed-off-by: Helmut Schaa Signed-off-by: David S. Miller diff --git a/net/core/dev.c b/net/core/dev.c index 3721fbb..b9b22a3 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -2058,16 +2058,16 @@ static struct netdev_queue *dev_pick_tx(struct net_device *dev, struct sk_buff *skb) { int queue_index; - struct sock *sk = skb->sk; + const struct net_device_ops *ops = dev->netdev_ops; - queue_index = sk_tx_queue_get(sk); - if (queue_index < 0) { - const struct net_device_ops *ops = dev->netdev_ops; + if (ops->ndo_select_queue) { + queue_index = ops->ndo_select_queue(dev, skb); + queue_index = dev_cap_txqueue(dev, queue_index); + } else { + struct sock *sk = skb->sk; + queue_index = sk_tx_queue_get(sk); + if (queue_index < 0) { - if (ops->ndo_select_queue) { - queue_index = ops->ndo_select_queue(dev, skb); - queue_index = dev_cap_txqueue(dev, queue_index); - } else { queue_index = 0; if (dev->real_num_tx_queues > 1) queue_index = skb_tx_hash(dev, skb); -- cgit v0.10.2 From 70789d7052239992824628db8133de08dc78e593 Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Fri, 3 Sep 2010 05:13:05 +0000 Subject: ipv6: discard overlapping fragment RFC5722 prohibits reassembling fragments when some data overlaps. Bug spotted by Zhang Zuotao . Signed-off-by: Nicolas Dichtel Signed-off-by: David S. Miller diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c index 545c414..64cfef1 100644 --- a/net/ipv6/reassembly.c +++ b/net/ipv6/reassembly.c @@ -149,13 +149,6 @@ int ip6_frag_match(struct inet_frag_queue *q, void *a) } EXPORT_SYMBOL(ip6_frag_match); -/* Memory Tracking Functions. */ -static void frag_kfree_skb(struct netns_frags *nf, struct sk_buff *skb) -{ - atomic_sub(skb->truesize, &nf->mem); - kfree_skb(skb); -} - void ip6_frag_init(struct inet_frag_queue *q, void *a) { struct frag_queue *fq = container_of(q, struct frag_queue, q); @@ -346,58 +339,22 @@ static int ip6_frag_queue(struct frag_queue *fq, struct sk_buff *skb, } found: - /* We found where to put this one. Check for overlap with - * preceding fragment, and, if needed, align things so that - * any overlaps are eliminated. + /* RFC5722, Section 4: + * When reassembling an IPv6 datagram, if + * one or more its constituent fragments is determined to be an + * overlapping fragment, the entire datagram (and any constituent + * fragments, including those not yet received) MUST be silently + * discarded. */ - if (prev) { - int i = (FRAG6_CB(prev)->offset + prev->len) - offset; - if (i > 0) { - offset += i; - if (end <= offset) - goto err; - if (!pskb_pull(skb, i)) - goto err; - if (skb->ip_summed != CHECKSUM_UNNECESSARY) - skb->ip_summed = CHECKSUM_NONE; - } - } + /* Check for overlap with preceding fragment. */ + if (prev && + (FRAG6_CB(prev)->offset + prev->len) - offset > 0) + goto discard_fq; - /* Look for overlap with succeeding segments. - * If we can merge fragments, do it. - */ - while (next && FRAG6_CB(next)->offset < end) { - int i = end - FRAG6_CB(next)->offset; /* overlap is 'i' bytes */ - - if (i < next->len) { - /* Eat head of the next overlapped fragment - * and leave the loop. The next ones cannot overlap. - */ - if (!pskb_pull(next, i)) - goto err; - FRAG6_CB(next)->offset += i; /* next fragment */ - fq->q.meat -= i; - if (next->ip_summed != CHECKSUM_UNNECESSARY) - next->ip_summed = CHECKSUM_NONE; - break; - } else { - struct sk_buff *free_it = next; - - /* Old fragment is completely overridden with - * new one drop it. - */ - next = next->next; - - if (prev) - prev->next = next; - else - fq->q.fragments = next; - - fq->q.meat -= free_it->len; - frag_kfree_skb(fq->q.net, free_it); - } - } + /* Look for overlap with succeeding segment. */ + if (next && FRAG6_CB(next)->offset < end) + goto discard_fq; FRAG6_CB(skb)->offset = offset; @@ -436,6 +393,8 @@ found: write_unlock(&ip6_frags.lock); return -1; +discard_fq: + fq_kill(fq); err: IP6_INC_STATS(net, ip6_dst_idev(skb_dst(skb)), IPSTATS_MIB_REASMFAILS); -- cgit v0.10.2 From 1ee89bd0fe72e381c8e4ef887d53c19c8adc6c93 Mon Sep 17 00:00:00 2001 From: Nicolas Dichtel Date: Fri, 3 Sep 2010 05:13:07 +0000 Subject: netfilter: discard overlapping IPv6 fragment RFC5722 prohibits reassembling IPv6 fragments when some data overlaps. Bug spotted by Zhang Zuotao . Signed-off-by: Nicolas Dichtel Signed-off-by: David S. Miller diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c index 13ef5bc..578f3c1 100644 --- a/net/ipv6/netfilter/nf_conntrack_reasm.c +++ b/net/ipv6/netfilter/nf_conntrack_reasm.c @@ -113,14 +113,6 @@ static void nf_skb_free(struct sk_buff *skb) kfree_skb(NFCT_FRAG6_CB(skb)->orig); } -/* Memory Tracking Functions. */ -static void frag_kfree_skb(struct sk_buff *skb) -{ - atomic_sub(skb->truesize, &nf_init_frags.mem); - nf_skb_free(skb); - kfree_skb(skb); -} - /* Destruction primitives. */ static __inline__ void fq_put(struct nf_ct_frag6_queue *fq) @@ -282,66 +274,22 @@ static int nf_ct_frag6_queue(struct nf_ct_frag6_queue *fq, struct sk_buff *skb, } found: - /* We found where to put this one. Check for overlap with - * preceding fragment, and, if needed, align things so that - * any overlaps are eliminated. - */ - if (prev) { - int i = (NFCT_FRAG6_CB(prev)->offset + prev->len) - offset; - - if (i > 0) { - offset += i; - if (end <= offset) { - pr_debug("overlap\n"); - goto err; - } - if (!pskb_pull(skb, i)) { - pr_debug("Can't pull\n"); - goto err; - } - if (skb->ip_summed != CHECKSUM_UNNECESSARY) - skb->ip_summed = CHECKSUM_NONE; - } - } - - /* Look for overlap with succeeding segments. - * If we can merge fragments, do it. + /* RFC5722, Section 4: + * When reassembling an IPv6 datagram, if + * one or more its constituent fragments is determined to be an + * overlapping fragment, the entire datagram (and any constituent + * fragments, including those not yet received) MUST be silently + * discarded. */ - while (next && NFCT_FRAG6_CB(next)->offset < end) { - /* overlap is 'i' bytes */ - int i = end - NFCT_FRAG6_CB(next)->offset; - - if (i < next->len) { - /* Eat head of the next overlapped fragment - * and leave the loop. The next ones cannot overlap. - */ - pr_debug("Eat head of the overlapped parts.: %d", i); - if (!pskb_pull(next, i)) - goto err; - /* next fragment */ - NFCT_FRAG6_CB(next)->offset += i; - fq->q.meat -= i; - if (next->ip_summed != CHECKSUM_UNNECESSARY) - next->ip_summed = CHECKSUM_NONE; - break; - } else { - struct sk_buff *free_it = next; - - /* Old fragmnet is completely overridden with - * new one drop it. - */ - next = next->next; + /* Check for overlap with preceding fragment. */ + if (prev && + (NFCT_FRAG6_CB(prev)->offset + prev->len) - offset > 0) + goto discard_fq; - if (prev) - prev->next = next; - else - fq->q.fragments = next; - - fq->q.meat -= free_it->len; - frag_kfree_skb(free_it); - } - } + /* Look for overlap with succeeding segment. */ + if (next && NFCT_FRAG6_CB(next)->offset < end) + goto discard_fq; NFCT_FRAG6_CB(skb)->offset = offset; @@ -371,6 +319,8 @@ found: write_unlock(&nf_frags.lock); return 0; +discard_fq: + fq_kill(fq); err: return -1; } -- cgit v0.10.2 From 89b12faba4f3441c9457c5278851e8a93ffd008d Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Mon, 6 Sep 2010 18:28:56 -0700 Subject: 3c59x: Fix deadlock in vortex_error() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This fixes a bug introduced in commit de847272149365363a6043a963a6f42fb91566e2 "3c59x: Use fine-grained locks for MII and windowed register access". vortex_interrupt() holds vp->window_lock over multiple register accesses to reduce locking overhead. However it also needs to call vortex_error() sometimes, and that uses the regular functions for access to windowed registers, which will try to acquire window_lock again. Therefore, drop window_lock around the call to vortex_error() and set the window afterward reacquiring the lock. Since vortex_error() may call vortex_rx(), which *does* require its caller to hold window_lock, lift that call up into vortex_interrupt(). This also removes the potential for calling vortex_rx() on a later-generation NIC. Reported-and-tested-by: Jens Schüßler [in Debian's 2.6.32] Signed-off-by: Ben Hutchings Signed-off-by: David S. Miller diff --git a/drivers/net/3c59x.c b/drivers/net/3c59x.c index a045559..85671ad 100644 --- a/drivers/net/3c59x.c +++ b/drivers/net/3c59x.c @@ -1994,10 +1994,9 @@ vortex_error(struct net_device *dev, int status) } } - if (status & RxEarly) { /* Rx early is unused. */ - vortex_rx(dev); + if (status & RxEarly) /* Rx early is unused. */ iowrite16(AckIntr | RxEarly, ioaddr + EL3_CMD); - } + if (status & StatsFull) { /* Empty statistics. */ static int DoneDidThat; if (vortex_debug > 4) @@ -2298,7 +2297,12 @@ vortex_interrupt(int irq, void *dev_id) if (status & (HostError | RxEarly | StatsFull | TxComplete | IntReq)) { if (status == 0xffff) break; + if (status & RxEarly) + vortex_rx(dev); + spin_unlock(&vp->window_lock); vortex_error(dev, status); + spin_lock(&vp->window_lock); + window_set(vp, 7); } if (--work_done < 0) { -- cgit v0.10.2 From cf9b94f88bdbe8a02015fc30d7c232b2d262d4ad Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Sat, 4 Sep 2010 03:14:35 +0000 Subject: irda: off by one This is an off by one. We would go past the end when we NUL terminate the "value" string at end of the function. The "value" buffer is allocated in irlan_client_parse_response() or irlan_provider_parse_command(). CC: stable@kernel.org Signed-off-by: Dan Carpenter Signed-off-by: David S. Miller diff --git a/net/irda/irlan/irlan_common.c b/net/irda/irlan/irlan_common.c index a788f9e..6130f9d 100644 --- a/net/irda/irlan/irlan_common.c +++ b/net/irda/irlan/irlan_common.c @@ -1102,7 +1102,7 @@ int irlan_extract_param(__u8 *buf, char *name, char *value, __u16 *len) memcpy(&val_len, buf+n, 2); /* To avoid alignment problems */ le16_to_cpus(&val_len); n+=2; - if (val_len > 1016) { + if (val_len >= 1016) { IRDA_DEBUG(2, "%s(), parameter length to long\n", __func__ ); return -RSP_INVALID_COMMAND_FORMAT; } -- cgit v0.10.2 From 32737e934a952c1b0c744f2a78d80089d15c7ee3 Mon Sep 17 00:00:00 2001 From: Mark Lord Date: Sat, 4 Sep 2010 14:17:59 +0000 Subject: PATCH: b44 Handle RX FIFO overflow better (simplified) This patch is a simplified version of the original patch from James Courtier-Dutton. >From: James Courtier-Dutton >Subject: [PATCH] Fix b44 RX FIFO overflow recovery. >Date: Wednesday, June 30, 2010 - 1:11 pm > >This patch improves the recovery after a RX FIFO overflow on the b44 >Ethernet NIC. >Before it would do a complete chip reset, resulting is loss of link >for a few seconds. >This patch improves this to do recovery in about 20ms without loss of link. > >Signed off by: James@superbug.co.uk Signed-off-by: Mark Lord Signed-off-by: David S. Miller diff --git a/drivers/net/b44.c b/drivers/net/b44.c index 37617ab..1e620e2 100644 --- a/drivers/net/b44.c +++ b/drivers/net/b44.c @@ -848,6 +848,15 @@ static int b44_poll(struct napi_struct *napi, int budget) b44_tx(bp); /* spin_unlock(&bp->tx_lock); */ } + if (bp->istat & ISTAT_RFO) { /* fast recovery, in ~20msec */ + bp->istat &= ~ISTAT_RFO; + b44_disable_ints(bp); + ssb_device_enable(bp->sdev, 0); /* resets ISTAT_RFO */ + b44_init_rings(bp); + b44_init_hw(bp, B44_FULL_RESET_SKIP_PHY); + netif_wake_queue(bp->dev); + } + spin_unlock_irqrestore(&bp->lock, flags); work_done = 0; -- cgit v0.10.2 From 8df73ff90f00f14d2c7ff7156f7ef153f7e9d3b7 Mon Sep 17 00:00:00 2001 From: Tetsuo Handa Date: Sat, 4 Sep 2010 01:34:28 +0000 Subject: UNIX: Do not loop forever at unix_autobind(). We assumed that unix_autobind() never fails if kzalloc() succeeded. But unix_autobind() allows only 1048576 names. If /proc/sys/fs/file-max is larger than 1048576 (e.g. systems with more than 10GB of RAM), a local user can consume all names using fork()/socket()/bind(). If all names are in use, those who call bind() with addr_len == sizeof(short) or connect()/sendmsg() with setsockopt(SO_PASSCRED) will continue while (1) yield(); loop at unix_autobind() till a name becomes available. This patch adds a loop counter in order to give up after 1048576 attempts. Calling yield() for once per 256 attempts may not be sufficient when many names are already in use, for __unix_find_socket_byname() can take long time under such circumstance. Therefore, this patch also adds cond_resched() call. Note that currently a local user can consume 2GB of kernel memory if the user is allowed to create and autobind 1048576 UNIX domain sockets. We should consider adding some restriction for autobind operation. Signed-off-by: Tetsuo Handa Signed-off-by: David S. Miller diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 4414a18..0b39b24 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -692,6 +692,7 @@ static int unix_autobind(struct socket *sock) static u32 ordernum = 1; struct unix_address *addr; int err; + unsigned int retries = 0; mutex_lock(&u->readlock); @@ -717,9 +718,17 @@ retry: if (__unix_find_socket_byname(net, addr->name, addr->len, sock->type, addr->hash)) { spin_unlock(&unix_table_lock); - /* Sanity yield. It is unusual case, but yet... */ - if (!(ordernum&0xFF)) - yield(); + /* + * __unix_find_socket_byname() may take long time if many names + * are already in use. + */ + cond_resched(); + /* Give up if all names seems to be in use. */ + if (retries++ == 0xFFFFF) { + err = -ENOSPC; + kfree(addr); + goto out; + } goto retry; } addr->hash ^= sk->sk_type; -- cgit v0.10.2 From 6f86b325189e0a53c97bf86cff0c8b02ff624934 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 6 Sep 2010 22:36:19 -0700 Subject: ipv4: Fix reverse path filtering with multipath routing. Actually iterate over the next-hops to make sure we have a device match. Otherwise RP filtering is always elided when the route matched has multiple next-hops. Reported-by: Igor M Podlesny Signed-off-by: David S. Miller diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index a439689..7d02a9f 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -246,6 +246,7 @@ int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif, struct fib_result res; int no_addr, rpf, accept_local; + bool dev_match; int ret; struct net *net; @@ -273,12 +274,22 @@ int fib_validate_source(__be32 src, __be32 dst, u8 tos, int oif, } *spec_dst = FIB_RES_PREFSRC(res); fib_combine_itag(itag, &res); + dev_match = false; + #ifdef CONFIG_IP_ROUTE_MULTIPATH - if (FIB_RES_DEV(res) == dev || res.fi->fib_nhs > 1) + for (ret = 0; ret < res.fi->fib_nhs; ret++) { + struct fib_nh *nh = &res.fi->fib_nh[ret]; + + if (nh->nh_dev == dev) { + dev_match = true; + break; + } + } #else if (FIB_RES_DEV(res) == dev) + dev_match = true; #endif - { + if (dev_match) { ret = FIB_RES_NH(res).nh_scope >= RT_SCOPE_HOST; fib_res_put(&res); return ret; -- cgit v0.10.2 From de2b96f1212722eb0af80bf9a029d03d8fc673a9 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 7 Sep 2010 13:49:44 -0700 Subject: via-velocity: Turn scatter-gather support back off. It causes all kinds of DMA API debugging assertions and all straight-forward attempts to fix it have failed. So turn off SG, and we'll tackle making this work properly in net-next-2.6 Reported-by: Dave Jones Tested-by: Dave Jones Signed-off-by: David S. Miller diff --git a/drivers/net/via-velocity.c b/drivers/net/via-velocity.c index fd69095..f534123 100644 --- a/drivers/net/via-velocity.c +++ b/drivers/net/via-velocity.c @@ -2824,7 +2824,7 @@ static int __devinit velocity_found1(struct pci_dev *pdev, const struct pci_devi netif_napi_add(dev, &vptr->napi, velocity_poll, VELOCITY_NAPI_WEIGHT); dev->features |= NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_FILTER | - NETIF_F_HW_VLAN_RX | NETIF_F_IP_CSUM | NETIF_F_SG; + NETIF_F_HW_VLAN_RX | NETIF_F_IP_CSUM; ret = register_netdev(dev); if (ret < 0) -- cgit v0.10.2 From e819ff519b2d74373eca4a9a2b417ebf4c1e1b29 Mon Sep 17 00:00:00 2001 From: John Johansen Date: Fri, 27 Aug 2010 18:33:26 -0700 Subject: AppArmor: Drop hack to remove appended " (deleted)" string The 2.6.36 kernel has refactored __d_path() so that it no longer appends " (deleted)" to unlinked paths. So drop the hack that was used to detect and remove the appended string. Signed-off-by: John Johansen Signed-off-by: James Morris diff --git a/security/apparmor/path.c b/security/apparmor/path.c index 19358dc..8239605 100644 --- a/security/apparmor/path.c +++ b/security/apparmor/path.c @@ -59,8 +59,7 @@ static int d_namespace_path(struct path *path, char *buf, int buflen, { struct path root, tmp; char *res; - int deleted, connected; - int error = 0; + int connected, error = 0; /* Get the root we want to resolve too, released below */ if (flags & PATH_CHROOT_REL) { @@ -74,19 +73,8 @@ static int d_namespace_path(struct path *path, char *buf, int buflen, } spin_lock(&dcache_lock); - /* There is a race window between path lookup here and the - * need to strip the " (deleted) string that __d_path applies - * Detect the race and relookup the path - * - * The stripping of (deleted) is a hack that could be removed - * with an updated __d_path - */ - do { - tmp = root; - deleted = d_unlinked(path->dentry); - res = __d_path(path, &tmp, buf, buflen); - - } while (deleted != d_unlinked(path->dentry)); + tmp = root; + res = __d_path(path, &tmp, buf, buflen); spin_unlock(&dcache_lock); *name = res; @@ -98,21 +86,17 @@ static int d_namespace_path(struct path *path, char *buf, int buflen, *name = buf; goto out; } - if (deleted) { - /* On some filesystems, newly allocated dentries appear to the - * security_path hooks as a deleted dentry except without an - * inode allocated. - * - * Remove the appended deleted text and return as string for - * normal mediation, or auditing. The (deleted) string is - * guaranteed to be added in this case, so just strip it. - */ - buf[buflen - 11] = 0; /* - (len(" (deleted)") +\0) */ - if (path->dentry->d_inode && !(flags & PATH_MEDIATE_DELETED)) { + /* Handle two cases: + * 1. A deleted dentry && profile is not allowing mediation of deleted + * 2. On some filesystems, newly allocated dentries appear to the + * security_path hooks as a deleted dentry except without an inode + * allocated. + */ + if (d_unlinked(path->dentry) && path->dentry->d_inode && + !(flags & PATH_MEDIATE_DELETED)) { error = -ENOENT; goto out; - } } /* Determine if the path is connected to the expected root */ -- cgit v0.10.2 From 3a2dc8382a3e85a51ed9c6f57ea80665ea7a0c95 Mon Sep 17 00:00:00 2001 From: John Johansen Date: Mon, 6 Sep 2010 10:10:20 -0700 Subject: AppArmor: Fix security_task_setrlimit logic for 2.6.36 changes 2.6.36 introduced the abilitiy to specify the task that is having its rlimits set. Update mediation to ensure that confined tasks can only set their own group_leader as expected by current policy. Add TODO note about extending policy to support setting other tasks rlimits. Signed-off-by: John Johansen Signed-off-by: James Morris diff --git a/security/apparmor/include/resource.h b/security/apparmor/include/resource.h index 3c88be9..02baec7 100644 --- a/security/apparmor/include/resource.h +++ b/security/apparmor/include/resource.h @@ -33,8 +33,8 @@ struct aa_rlimit { }; int aa_map_resource(int resource); -int aa_task_setrlimit(struct aa_profile *profile, unsigned int resource, - struct rlimit *new_rlim); +int aa_task_setrlimit(struct aa_profile *profile, struct task_struct *, + unsigned int resource, struct rlimit *new_rlim); void __aa_transition_rlimits(struct aa_profile *old, struct aa_profile *new); diff --git a/security/apparmor/lsm.c b/security/apparmor/lsm.c index f73e2c2..cf1de44 100644 --- a/security/apparmor/lsm.c +++ b/security/apparmor/lsm.c @@ -614,7 +614,7 @@ static int apparmor_task_setrlimit(struct task_struct *task, int error = 0; if (!unconfined(profile)) - error = aa_task_setrlimit(profile, resource, new_rlim); + error = aa_task_setrlimit(profile, task, resource, new_rlim); return error; } diff --git a/security/apparmor/resource.c b/security/apparmor/resource.c index 4a368f1..a4136c1 100644 --- a/security/apparmor/resource.c +++ b/security/apparmor/resource.c @@ -72,6 +72,7 @@ int aa_map_resource(int resource) /** * aa_task_setrlimit - test permission to set an rlimit * @profile - profile confining the task (NOT NULL) + * @task - task the resource is being set on * @resource - the resource being set * @new_rlim - the new resource limit (NOT NULL) * @@ -79,18 +80,21 @@ int aa_map_resource(int resource) * * Returns: 0 or error code if setting resource failed */ -int aa_task_setrlimit(struct aa_profile *profile, unsigned int resource, - struct rlimit *new_rlim) +int aa_task_setrlimit(struct aa_profile *profile, struct task_struct *task, + unsigned int resource, struct rlimit *new_rlim) { int error = 0; - if (profile->rlimits.mask & (1 << resource) && - new_rlim->rlim_max > profile->rlimits.limits[resource].rlim_max) - - error = audit_resource(profile, resource, new_rlim->rlim_max, - -EACCES); + /* TODO: extend resource control to handle other (non current) + * processes. AppArmor rules currently have the implicit assumption + * that the task is setting the resource of the current process + */ + if ((task != current->group_leader) || + (profile->rlimits.mask & (1 << resource) && + new_rlim->rlim_max > profile->rlimits.limits[resource].rlim_max)) + error = -EACCES; - return error; + return audit_resource(profile, resource, new_rlim->rlim_max, error); } /** -- cgit v0.10.2 From 04ccd53f09741c4bc54ab36db000bc1383e4812e Mon Sep 17 00:00:00 2001 From: John Johansen Date: Fri, 27 Aug 2010 18:33:28 -0700 Subject: AppArmor: Fix splitting an fqname into separate namespace and profile names As per Dan Carpenter If we have a ns name without a following profile then in the original code it did "*ns_name = &name[1];". "name" is NULL so "*ns_name" is 0x1. That isn't useful and could cause an oops when this function is called from aa_remove_profiles(). Beyond this the assignment of the namespace name was wrong in the case where the profile name was provided as it was being set to &name[1] after name = skip_spaces(split + 1); Move the ns_name assignment before updating name for the split and also add skip_spaces, making the interface more robust. Signed-off-by: John Johansen Signed-off-by: James Morris diff --git a/security/apparmor/lib.c b/security/apparmor/lib.c index 6e85cdb..506d2ba 100644 --- a/security/apparmor/lib.c +++ b/security/apparmor/lib.c @@ -40,6 +40,7 @@ char *aa_split_fqname(char *fqname, char **ns_name) *ns_name = NULL; if (name[0] == ':') { char *split = strchr(&name[1], ':'); + *ns_name = skip_spaces(&name[1]); if (split) { /* overwrite ':' with \0 */ *split = 0; @@ -47,7 +48,6 @@ char *aa_split_fqname(char *fqname, char **ns_name) } else /* a ns name without a following profile is allowed */ name = NULL; - *ns_name = &name[1]; } if (name && *name == 0) name = NULL; -- cgit v0.10.2 From 999b4f0aa2314b76857775334cb94bafa053db64 Mon Sep 17 00:00:00 2001 From: John Johansen Date: Fri, 27 Aug 2010 18:33:29 -0700 Subject: AppArmor: Fix locking from removal of profile namespace The locking for profile namespace removal is wrong, when removing a profile namespace, it needs to be removed from its parent's list. Lock the parent of namespace list instead of the namespace being removed. Signed-off-by: John Johansen Signed-off-by: James Morris diff --git a/security/apparmor/policy.c b/security/apparmor/policy.c index 3cdc1ad..52cc865 100644 --- a/security/apparmor/policy.c +++ b/security/apparmor/policy.c @@ -1151,12 +1151,14 @@ ssize_t aa_remove_profiles(char *fqname, size_t size) /* released below */ ns = aa_get_namespace(root); - write_lock(&ns->lock); if (!name) { /* remove namespace - can only happen if fqname[0] == ':' */ + write_lock(&ns->parent->lock); __remove_namespace(ns); + write_unlock(&ns->parent->lock); } else { /* remove profile */ + write_lock(&ns->lock); profile = aa_get_profile(__lookup_profile(&ns->base, name)); if (!profile) { error = -ENOENT; @@ -1165,8 +1167,8 @@ ssize_t aa_remove_profiles(char *fqname, size_t size) } name = profile->base.hname; __remove_profile(profile); + write_unlock(&ns->lock); } - write_unlock(&ns->lock); /* don't fail removal if audit fails */ (void) audit_policy(OP_PROF_RM, GFP_KERNEL, name, info, error); -- cgit v0.10.2 From e950598d43dce8d97e7d5270808393425d1e5cbd Mon Sep 17 00:00:00 2001 From: Mimi Zohar Date: Tue, 31 Aug 2010 09:38:51 -0400 Subject: ima: always maintain counters commit 8262bb85da allocated the inode integrity struct (iint) before any inodes were created. Only after IMA was initialized in late_initcall were the counters updated. This patch updates the counters, whether or not IMA has been initialized, to resolve 'imbalance' messages. This patch fixes the bug as reported in bugzilla: 15673. When the i915 is builtin, the ring_buffer is initialized before IMA, causing the imbalance message on suspend. Reported-by: Thomas Meyer Signed-off-by: Mimi Zohar Tested-by: Thomas Meyer Tested-by: David Safford Cc: Stable Kernel Signed-off-by: James Morris diff --git a/security/integrity/ima/ima.h b/security/integrity/ima/ima.h index 16d100d..3fbcd1d 100644 --- a/security/integrity/ima/ima.h +++ b/security/integrity/ima/ima.h @@ -35,6 +35,7 @@ enum tpm_pcrs { TPM_PCR0 = 0, TPM_PCR8 = 8 }; #define IMA_MEASURE_HTABLE_SIZE (1 << IMA_HASH_BITS) /* set during initialization */ +extern int iint_initialized; extern int ima_initialized; extern int ima_used_chip; extern char *ima_hash; diff --git a/security/integrity/ima/ima_iint.c b/security/integrity/ima/ima_iint.c index 7625b85..afba4ae 100644 --- a/security/integrity/ima/ima_iint.c +++ b/security/integrity/ima/ima_iint.c @@ -22,9 +22,10 @@ RADIX_TREE(ima_iint_store, GFP_ATOMIC); DEFINE_SPINLOCK(ima_iint_lock); - static struct kmem_cache *iint_cache __read_mostly; +int iint_initialized = 0; + /* ima_iint_find_get - return the iint associated with an inode * * ima_iint_find_get gets a reference to the iint. Caller must @@ -141,6 +142,7 @@ static int __init ima_iintcache_init(void) iint_cache = kmem_cache_create("iint_cache", sizeof(struct ima_iint_cache), 0, SLAB_PANIC, init_once); + iint_initialized = 1; return 0; } security_initcall(ima_iintcache_init); diff --git a/security/integrity/ima/ima_main.c b/security/integrity/ima/ima_main.c index f936413..e662b89 100644 --- a/security/integrity/ima/ima_main.c +++ b/security/integrity/ima/ima_main.c @@ -148,12 +148,14 @@ void ima_counts_get(struct file *file) struct ima_iint_cache *iint; int rc; - if (!ima_initialized || !S_ISREG(inode->i_mode)) + if (!iint_initialized || !S_ISREG(inode->i_mode)) return; iint = ima_iint_find_get(inode); if (!iint) return; mutex_lock(&iint->mutex); + if (!ima_initialized) + goto out; rc = ima_must_measure(iint, inode, MAY_READ, FILE_CHECK); if (rc < 0) goto out; @@ -213,7 +215,7 @@ void ima_file_free(struct file *file) struct inode *inode = file->f_dentry->d_inode; struct ima_iint_cache *iint; - if (!ima_initialized || !S_ISREG(inode->i_mode)) + if (!iint_initialized || !S_ISREG(inode->i_mode)) return; iint = ima_iint_find_get(inode); if (!iint) @@ -230,7 +232,7 @@ static int process_measurement(struct file *file, const unsigned char *filename, { struct inode *inode = file->f_dentry->d_inode; struct ima_iint_cache *iint; - int rc; + int rc = 0; if (!ima_initialized || !S_ISREG(inode->i_mode)) return 0; -- cgit v0.10.2 From dc696aced9f09f05b1f927b93f5a7918017a3e49 Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Thu, 12 Aug 2010 16:24:25 -0700 Subject: ocfs2: Fix metaecc error messages Like tools, the checksum validate function now prints the values in hex. Signed-off-by: Sunil Mushran Singed-off-by: Tao Ma diff --git a/fs/ocfs2/blockcheck.c b/fs/ocfs2/blockcheck.c index ec6d123..c7ee03c 100644 --- a/fs/ocfs2/blockcheck.c +++ b/fs/ocfs2/blockcheck.c @@ -439,7 +439,7 @@ int ocfs2_block_check_validate(void *data, size_t blocksize, ocfs2_blockcheck_inc_failure(stats); mlog(ML_ERROR, - "CRC32 failed: stored: %u, computed %u. Applying ECC.\n", + "CRC32 failed: stored: 0x%x, computed 0x%x. Applying ECC.\n", (unsigned int)check.bc_crc32e, (unsigned int)crc); /* Ok, try ECC fixups */ @@ -453,7 +453,7 @@ int ocfs2_block_check_validate(void *data, size_t blocksize, goto out; } - mlog(ML_ERROR, "Fixed CRC32 failed: stored: %u, computed %u\n", + mlog(ML_ERROR, "Fixed CRC32 failed: stored: 0x%x, computed 0x%x\n", (unsigned int)check.bc_crc32e, (unsigned int)crc); rc = -EIO; -- cgit v0.10.2 From f5ce5a08a40f2086435858ddc80cb40394b082eb Mon Sep 17 00:00:00 2001 From: Sunil Mushran Date: Thu, 12 Aug 2010 16:24:26 -0700 Subject: ocfs2: Fix incorrect checksum validation error For local mounts, ocfs2_read_locked_inode() calls ocfs2_read_blocks_sync() to read the inode off the disk. The latter first checks to see if that block is cached in the journal, and, if so, returns that block. That is ok. But ocfs2_read_locked_inode() goes wrong when it tries to validate the checksum of such blocks. Blocks that are cached in the journal may not have had their checksum computed as yet. We should not validate the checksums of such blocks. Fixes ossbz#1282 http://oss.oracle.com/bugzilla/show_bug.cgi?id=1282 Signed-off-by: Sunil Mushran Cc: stable@kernel.org Singed-off-by: Tao Ma diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c index 0492464..eece3e0 100644 --- a/fs/ocfs2/inode.c +++ b/fs/ocfs2/inode.c @@ -488,7 +488,11 @@ static int ocfs2_read_locked_inode(struct inode *inode, OCFS2_BH_IGNORE_CACHE); } else { status = ocfs2_read_blocks_sync(osb, args->fi_blkno, 1, &bh); - if (!status) + /* + * If buffer is in jbd, then its checksum may not have been + * computed as yet. + */ + if (!status && !buffer_jbd(bh)) status = ocfs2_validate_inode_block(osb->sb, bh); } if (status < 0) { -- cgit v0.10.2 From f63afdb2c32db850fa1bfccf84643a8885cbeb61 Mon Sep 17 00:00:00 2001 From: Tao Ma Date: Sat, 17 Jul 2010 21:45:49 +0800 Subject: ocfs2: make __ocfs2_page_mkwrite handle file end properly. __ocfs2_page_mkwrite now is broken in handling file end. 1. the last page should be the page contains i_size - 1. 2. the len in the last page is also calculated wrong. So change them accordingly. Acked-by: Mark Fasheh Signed-off-by: Tao Ma diff --git a/fs/ocfs2/mmap.c b/fs/ocfs2/mmap.c index af2b8fe..4c18f4a 100644 --- a/fs/ocfs2/mmap.c +++ b/fs/ocfs2/mmap.c @@ -74,9 +74,11 @@ static int __ocfs2_page_mkwrite(struct inode *inode, struct buffer_head *di_bh, /* * Another node might have truncated while we were waiting on * cluster locks. + * We don't check size == 0 before the shift. This is borrowed + * from do_generic_file_read. */ - last_index = size >> PAGE_CACHE_SHIFT; - if (page->index > last_index) { + last_index = (size - 1) >> PAGE_CACHE_SHIFT; + if (unlikely(!size || page->index > last_index)) { ret = -EINVAL; goto out; } @@ -107,7 +109,7 @@ static int __ocfs2_page_mkwrite(struct inode *inode, struct buffer_head *di_bh, * because the "write" would invalidate their data. */ if (page->index == last_index) - len = size & ~PAGE_CACHE_MASK; + len = ((size - 1) & ~PAGE_CACHE_MASK) + 1; ret = ocfs2_write_begin_nolock(mapping, pos, len, 0, &locked_page, &fsdata, di_bh, page); -- cgit v0.10.2 From 04eda1a18019bb387dc7e97ee99979dd88dc608a Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Thu, 5 Aug 2010 20:32:45 +0200 Subject: ocfs2: Flush drive's caches on fdatasync When 'barrier' mount option is specified, we have to issue a cache flush during fdatasync(2). We have to do this even if inode doesn't have I_DIRTY_DATASYNC set because we still have to get written *data* to disk so that they are not lost in case of crash. Acked-by: Tao Ma Signed-off-by: Jan Kara Singed-off-by: Tao Ma diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 81296b4..6b2be0f 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -36,6 +36,7 @@ #include #include #include +#include #define MLOG_MASK_PREFIX ML_INODE #include @@ -190,8 +191,16 @@ static int ocfs2_sync_file(struct file *file, int datasync) if (err) goto bail; - if (datasync && !(inode->i_state & I_DIRTY_DATASYNC)) + if (datasync && !(inode->i_state & I_DIRTY_DATASYNC)) { + /* + * We still have to flush drive's caches to get data to the + * platter + */ + if (osb->s_mount_opt & OCFS2_MOUNT_BARRIER) + blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, + NULL, BLKDEV_IFL_WAIT); goto bail; + } journal = osb->journal->j_journal; err = jbd2_journal_force_commit(journal); -- cgit v0.10.2 From 889f004a8c83d515f275078687f859bc0d5ede9d Mon Sep 17 00:00:00 2001 From: Tao Ma Date: Thu, 2 Sep 2010 13:10:10 +0800 Subject: ocfs2: Use the right group in nfs sync check. We have added discontig block group now, and now an inode can be allocated in an discontig block group. So get it in ocfs2_get_suballoc_slot_bit. The old ocfs2_test_suballoc_bit gets group block no from the allocation inode which is wrong. Fix it by passing the right group. Acked-by: Mark Fasheh Signed-off-by: Tao Ma diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c index a8e6a95..8a009ee 100644 --- a/fs/ocfs2/suballoc.c +++ b/fs/ocfs2/suballoc.c @@ -2567,7 +2567,8 @@ out: * suballoc_bit. */ static int ocfs2_get_suballoc_slot_bit(struct ocfs2_super *osb, u64 blkno, - u16 *suballoc_slot, u16 *suballoc_bit) + u16 *suballoc_slot, u64 *group_blkno, + u16 *suballoc_bit) { int status; struct buffer_head *inode_bh = NULL; @@ -2604,6 +2605,8 @@ static int ocfs2_get_suballoc_slot_bit(struct ocfs2_super *osb, u64 blkno, *suballoc_slot = le16_to_cpu(inode_fe->i_suballoc_slot); if (suballoc_bit) *suballoc_bit = le16_to_cpu(inode_fe->i_suballoc_bit); + if (group_blkno) + *group_blkno = le64_to_cpu(inode_fe->i_suballoc_loc); bail: brelse(inode_bh); @@ -2621,7 +2624,8 @@ bail: */ static int ocfs2_test_suballoc_bit(struct ocfs2_super *osb, struct inode *suballoc, - struct buffer_head *alloc_bh, u64 blkno, + struct buffer_head *alloc_bh, + u64 group_blkno, u64 blkno, u16 bit, int *res) { struct ocfs2_dinode *alloc_di; @@ -2642,10 +2646,8 @@ static int ocfs2_test_suballoc_bit(struct ocfs2_super *osb, goto bail; } - if (alloc_di->i_suballoc_loc) - bg_blkno = le64_to_cpu(alloc_di->i_suballoc_loc); - else - bg_blkno = ocfs2_which_suballoc_group(blkno, bit); + bg_blkno = group_blkno ? group_blkno : + ocfs2_which_suballoc_group(blkno, bit); status = ocfs2_read_group_descriptor(suballoc, alloc_di, bg_blkno, &group_bh); if (status < 0) { @@ -2680,6 +2682,7 @@ bail: int ocfs2_test_inode_bit(struct ocfs2_super *osb, u64 blkno, int *res) { int status; + u64 group_blkno = 0; u16 suballoc_bit = 0, suballoc_slot = 0; struct inode *inode_alloc_inode; struct buffer_head *alloc_bh = NULL; @@ -2687,7 +2690,7 @@ int ocfs2_test_inode_bit(struct ocfs2_super *osb, u64 blkno, int *res) mlog_entry("blkno: %llu", (unsigned long long)blkno); status = ocfs2_get_suballoc_slot_bit(osb, blkno, &suballoc_slot, - &suballoc_bit); + &group_blkno, &suballoc_bit); if (status < 0) { mlog(ML_ERROR, "get alloc slot and bit failed %d\n", status); goto bail; @@ -2715,7 +2718,7 @@ int ocfs2_test_inode_bit(struct ocfs2_super *osb, u64 blkno, int *res) } status = ocfs2_test_suballoc_bit(osb, inode_alloc_inode, alloc_bh, - blkno, suballoc_bit, res); + group_blkno, blkno, suballoc_bit, res); if (status < 0) mlog(ML_ERROR, "test suballoc bit failed %d\n", status); -- cgit v0.10.2 From b2b6ebf5f740e015b2155343958f067e594323ea Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Thu, 26 Aug 2010 13:06:50 -0700 Subject: ocfs2: properly set and use inode group alloc hint We were setting ac->ac_last_group in ocfs2_claim_suballoc_bits from res->sr_bg_blkno. Unfortunately, res->sr_bg_blkno is going to be zero under normal (non-fragmented) circumstances. The discontig block group patches effectively turned off that feature. Fix this by correctly calculating what the next group hint should be. Acked-by: Tao Ma Signed-off-by: Mark Fasheh Tested-by: Goldwyn Rodrigues Signed-off-by: Tao Ma diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c index 8a009ee..b93d7e7 100644 --- a/fs/ocfs2/suballoc.c +++ b/fs/ocfs2/suballoc.c @@ -62,6 +62,17 @@ struct ocfs2_suballoc_result { unsigned int sr_bits; /* How many bits we claimed */ }; +static u64 ocfs2_group_from_res(struct ocfs2_suballoc_result *res) +{ + if (res->sr_blkno == 0) + return 0; + + if (res->sr_bg_blkno) + return res->sr_bg_blkno; + + return ocfs2_which_suballoc_group(res->sr_blkno, res->sr_bit_offset); +} + static inline void ocfs2_debug_bg(struct ocfs2_group_desc *bg); static inline void ocfs2_debug_suballoc_inode(struct ocfs2_dinode *fe); static inline u16 ocfs2_find_victim_chain(struct ocfs2_chain_list *cl); @@ -1845,6 +1856,7 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_alloc_context *ac, int status; u16 victim, i; u16 bits_left = 0; + u64 hint = ac->ac_last_group; struct ocfs2_chain_list *cl; struct ocfs2_dinode *fe; @@ -1872,7 +1884,7 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_alloc_context *ac, goto bail; } - res->sr_bg_blkno = ac->ac_last_group; + res->sr_bg_blkno = hint; if (res->sr_bg_blkno) { /* Attempt to short-circuit the usual search mechanism * by jumping straight to the most recently used @@ -1896,8 +1908,10 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_alloc_context *ac, status = ocfs2_search_chain(ac, handle, bits_wanted, min_bits, res, &bits_left); - if (!status) + if (!status) { + hint = ocfs2_group_from_res(res); goto set_hint; + } if (status < 0 && status != -ENOSPC) { mlog_errno(status); goto bail; @@ -1920,8 +1934,10 @@ static int ocfs2_claim_suballoc_bits(struct ocfs2_alloc_context *ac, ac->ac_chain = i; status = ocfs2_search_chain(ac, handle, bits_wanted, min_bits, res, &bits_left); - if (!status) + if (!status) { + hint = ocfs2_group_from_res(res); break; + } if (status < 0 && status != -ENOSPC) { mlog_errno(status); goto bail; @@ -1936,7 +1952,7 @@ set_hint: if (bits_left < min_bits) ac->ac_last_group = 0; else - ac->ac_last_group = res->sr_bg_blkno; + ac->ac_last_group = hint; } bail: -- cgit v0.10.2 From 9b4c0ff32ccd87ab52d4c5bd0a0536febce11370 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Tue, 24 Aug 2010 14:28:03 +0200 Subject: ocfs2: Fix deadlock when allocating page We cannot call grab_cache_page() when holding filesystem locks or with a transaction started as grab_cache_page() calls page allocation with GFP_KERNEL flag and thus page reclaim can recurse back into the filesystem causing deadlocks or various assertion failures. We have to use find_or_create_page() instead and pass it GFP_NOFS as we do with other allocations. Acked-by: Mark Fasheh Signed-off-by: Jan Kara Signed-off-by: Tao Ma diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c index 215e12c..592fae5 100644 --- a/fs/ocfs2/alloc.c +++ b/fs/ocfs2/alloc.c @@ -6672,7 +6672,7 @@ int ocfs2_grab_pages(struct inode *inode, loff_t start, loff_t end, last_page_bytes = PAGE_ALIGN(end); index = start >> PAGE_CACHE_SHIFT; do { - pages[numpages] = grab_cache_page(mapping, index); + pages[numpages] = find_or_create_page(mapping, index, GFP_NOFS); if (!pages[numpages]) { ret = -ENOMEM; mlog_errno(ret); diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 6b2be0f..2caa3a7 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -783,7 +783,7 @@ static int ocfs2_write_zero_page(struct inode *inode, u64 abs_from, BUG_ON(abs_to > (((u64)index + 1) << PAGE_CACHE_SHIFT)); BUG_ON(abs_from & (inode->i_blkbits - 1)); - page = grab_cache_page(mapping, index); + page = find_or_create_page(mapping, index, GFP_NOFS); if (!page) { ret = -ENOMEM; mlog_errno(ret); diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c index 73a11cc..0afeda831 100644 --- a/fs/ocfs2/refcounttree.c +++ b/fs/ocfs2/refcounttree.c @@ -2960,7 +2960,7 @@ static int ocfs2_duplicate_clusters_by_page(handle_t *handle, if (map_end & (PAGE_CACHE_SIZE - 1)) to = map_end & (PAGE_CACHE_SIZE - 1); - page = grab_cache_page(mapping, page_index); + page = find_or_create_page(mapping, page_index, GFP_NOFS); /* * In case PAGE_CACHE_SIZE <= CLUSTER_SIZE, This page @@ -3179,7 +3179,8 @@ static int ocfs2_cow_sync_writeback(struct super_block *sb, if (map_end > end) map_end = end; - page = grab_cache_page(context->inode->i_mapping, page_index); + page = find_or_create_page(context->inode->i_mapping, + page_index, GFP_NOFS); BUG_ON(!page); wait_on_page_writeback(page); -- cgit v0.10.2 From 81c8c82b5a39f9127e8b239e9b406a6c3a41b228 Mon Sep 17 00:00:00 2001 From: Tristan Ye Date: Thu, 19 Aug 2010 15:15:00 +0800 Subject: Ocfs2: Fix a regression bug from mainline commit(6b933c8e6f1a2f3118082c455eef25f9b1ac7b45). The patch is to fix the regression bug brought from commit 6b933c8...( 'ocfs2: Avoid direct write if we fall back to buffered I/O'): http://oss.oracle.com/bugzilla/show_bug.cgi?id=1285 The commit 6b933c8e6f1a2f3118082c455eef25f9b1ac7b45 changed __generic_file_aio_write to generic_file_buffered_write, which didn't call filemap_{write,wait}_range to flush the pagecaches when we were falling O_DIRECT writes back to buffered ones. it did hurt the O_DIRECT semantics somehow in extented odirect writes. This patch tries to guarantee O_DIRECT writes of 'fall back to buffered' to be correctly flushed. Signed-off-by: Tristan Ye Signed-off-by: Tao Ma diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 2caa3a7..9a03c15 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -2338,7 +2338,7 @@ out_dio: BUG_ON(ret == -EIOCBQUEUED && !(file->f_flags & O_DIRECT)); if (((file->f_flags & O_DSYNC) && !direct_io) || IS_SYNC(inode) || - ((file->f_flags & O_DIRECT) && has_refcount)) { + ((file->f_flags & O_DIRECT) && !direct_io)) { ret = filemap_fdatawrite_range(file->f_mapping, pos, pos + count - 1); if (ret < 0) -- cgit v0.10.2 From 021960cab320ae3cc4e9aba9cca42f9f5ce785f3 Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Fri, 13 Aug 2010 15:15:15 -0700 Subject: ocfs2: split out inode alloc code from ocfs2_mknod_locked Do this by splitting the bulk of the function away from the inode allocation code at the very tom of ocfs2_mknod_locked(). Existing callers don't need to change and won't see any difference. The new function created, __ocfs2_mknod_locked() will be used shortly. Signed-off-by: Mark Fasheh Signed-off-by: Tao Ma diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index f171b51..2aa66b6 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -472,32 +472,23 @@ leave: return status; } -static int ocfs2_mknod_locked(struct ocfs2_super *osb, - struct inode *dir, - struct inode *inode, - dev_t dev, - struct buffer_head **new_fe_bh, - struct buffer_head *parent_fe_bh, - handle_t *handle, - struct ocfs2_alloc_context *inode_ac) +static int __ocfs2_mknod_locked(struct inode *dir, + struct inode *inode, + dev_t dev, + struct buffer_head **new_fe_bh, + struct buffer_head *parent_fe_bh, + handle_t *handle, + struct ocfs2_alloc_context *inode_ac, + u64 fe_blkno, u64 suballoc_loc, u16 suballoc_bit) { int status = 0; + struct ocfs2_super *osb = OCFS2_SB(dir->i_sb); struct ocfs2_dinode *fe = NULL; struct ocfs2_extent_list *fel; - u64 suballoc_loc, fe_blkno = 0; - u16 suballoc_bit; u16 feat; *new_fe_bh = NULL; - status = ocfs2_claim_new_inode(handle, dir, parent_fe_bh, - inode_ac, &suballoc_loc, - &suballoc_bit, &fe_blkno); - if (status < 0) { - mlog_errno(status); - goto leave; - } - /* populate as many fields early on as possible - many of * these are used by the support functions here and in * callers. */ @@ -591,6 +582,34 @@ leave: return status; } +static int ocfs2_mknod_locked(struct ocfs2_super *osb, + struct inode *dir, + struct inode *inode, + dev_t dev, + struct buffer_head **new_fe_bh, + struct buffer_head *parent_fe_bh, + handle_t *handle, + struct ocfs2_alloc_context *inode_ac) +{ + int status = 0; + u64 suballoc_loc, fe_blkno = 0; + u16 suballoc_bit; + + *new_fe_bh = NULL; + + status = ocfs2_claim_new_inode(handle, dir, parent_fe_bh, + inode_ac, &suballoc_loc, + &suballoc_bit, &fe_blkno); + if (status < 0) { + mlog_errno(status); + return status; + } + + return __ocfs2_mknod_locked(dir, inode, dev, new_fe_bh, + parent_fe_bh, handle, inode_ac, + fe_blkno, suballoc_loc, suballoc_bit); +} + static int ocfs2_mkdir(struct inode *dir, struct dentry *dentry, int mode) -- cgit v0.10.2 From d51349829c378c06ba4aa7d4b16ca23739858608 Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Fri, 13 Aug 2010 15:15:16 -0700 Subject: ocfs2: use ocfs2_alloc_dinode_update_counts() instead of open coding ocfs2_search_chain() makes the same updates as ocfs2_alloc_dinode_update_counts to the alloc inode. Instead of open coding the bitmap update, use our helper function. Signed-off-by: Mark Fasheh Signed-off-by: Tao Ma diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c index b93d7e7..e7edda8 100644 --- a/fs/ocfs2/suballoc.c +++ b/fs/ocfs2/suballoc.c @@ -1719,7 +1719,6 @@ static int ocfs2_search_chain(struct ocfs2_alloc_context *ac, { int status; u16 chain; - u32 tmp_used; u64 next_group; struct inode *alloc_inode = ac->ac_inode; struct buffer_head *group_bh = NULL; @@ -1807,22 +1806,14 @@ static int ocfs2_search_chain(struct ocfs2_alloc_context *ac, } } - /* Ok, claim our bits now: set the info on dinode, chainlist - * and then the group */ - status = ocfs2_journal_access_di(handle, - INODE_CACHE(alloc_inode), - ac->ac_bh, - OCFS2_JOURNAL_ACCESS_WRITE); - if (status < 0) { + status = ocfs2_alloc_dinode_update_counts(alloc_inode, handle, + ac->ac_bh, res->sr_bits, + chain); + if (status) { mlog_errno(status); goto bail; } - tmp_used = le32_to_cpu(fe->id1.bitmap1.i_used); - fe->id1.bitmap1.i_used = cpu_to_le32(res->sr_bits + tmp_used); - le32_add_cpu(&cl->cl_recs[chain].c_free, -res->sr_bits); - ocfs2_journal_dirty(handle, ac->ac_bh); - status = ocfs2_block_group_set_bits(handle, alloc_inode, bg, -- cgit v0.10.2 From e49e27674d1dd2717ad90b21ece8f83102153315 Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Fri, 13 Aug 2010 15:15:17 -0700 Subject: ocfs2: allow return of new inode block location before allocation of the inode This allows code which needs to know the eventual block number of an inode but can't allocate it yet due to transaction or lock ordering. For example, ocfs2_create_inode_in_orphan() currently gives a junk blkno for preparation of the orphan dir because it can't yet know where the actual inode is placed - that code is actually in ocfs2_mknod_locked. This is a problem when the orphan dirs are indexed as the junk inode number will create an index entry which goes unused (and fails the later removal from the orphan dir). Now with these interfaces, ocfs2_create_inode_in_orphan() can run the block group search (and get back the inode block number) *before* any actual allocation occurs. Signed-off-by: Mark Fasheh Signed-off-by: Tao Ma diff --git a/fs/ocfs2/suballoc.c b/fs/ocfs2/suballoc.c index e7edda8..8a286f5 100644 --- a/fs/ocfs2/suballoc.c +++ b/fs/ocfs2/suballoc.c @@ -57,6 +57,12 @@ struct ocfs2_suballoc_result { u64 sr_bg_blkno; /* The bg we allocated from. Set to 0 when a block group is contiguous. */ + u64 sr_bg_stable_blkno; /* + * Doesn't change, always + * set to target block + * group descriptor + * block. + */ u64 sr_blkno; /* The first allocated block */ unsigned int sr_bit_offset; /* The bit in the bg */ unsigned int sr_bits; /* How many bits we claimed */ @@ -149,6 +155,10 @@ void ocfs2_free_ac_resource(struct ocfs2_alloc_context *ac) brelse(ac->ac_bh); ac->ac_bh = NULL; ac->ac_resv = NULL; + if (ac->ac_find_loc_priv) { + kfree(ac->ac_find_loc_priv); + ac->ac_find_loc_priv = NULL; + } } void ocfs2_free_alloc_context(struct ocfs2_alloc_context *ac) @@ -1689,6 +1699,15 @@ static int ocfs2_search_one_group(struct ocfs2_alloc_context *ac, if (!ret) ocfs2_bg_discontig_fix_result(ac, gd, res); + /* + * sr_bg_blkno might have been changed by + * ocfs2_bg_discontig_fix_result + */ + res->sr_bg_stable_blkno = group_bh->b_blocknr; + + if (ac->ac_find_loc_only) + goto out_loc_only; + ret = ocfs2_alloc_dinode_update_counts(alloc_inode, handle, ac->ac_bh, res->sr_bits, le16_to_cpu(gd->bg_chain)); @@ -1702,6 +1721,7 @@ static int ocfs2_search_one_group(struct ocfs2_alloc_context *ac, if (ret < 0) mlog_errno(ret); +out_loc_only: *bits_left = le16_to_cpu(gd->bg_free_bits_count); out: @@ -1780,6 +1800,11 @@ static int ocfs2_search_chain(struct ocfs2_alloc_context *ac, if (!status) ocfs2_bg_discontig_fix_result(ac, bg, res); + /* + * sr_bg_blkno might have been changed by + * ocfs2_bg_discontig_fix_result + */ + res->sr_bg_stable_blkno = group_bh->b_blocknr; /* * Keep track of previous block descriptor read. When @@ -1806,6 +1831,9 @@ static int ocfs2_search_chain(struct ocfs2_alloc_context *ac, } } + if (ac->ac_find_loc_only) + goto out_loc_only; + status = ocfs2_alloc_dinode_update_counts(alloc_inode, handle, ac->ac_bh, res->sr_bits, chain); @@ -1828,6 +1856,7 @@ static int ocfs2_search_chain(struct ocfs2_alloc_context *ac, mlog(0, "Allocated %u bits from suballocator %llu\n", res->sr_bits, (unsigned long long)le64_to_cpu(fe->i_blkno)); +out_loc_only: *bits_left = le16_to_cpu(bg->bg_free_bits_count); bail: brelse(group_bh); @@ -2023,6 +2052,136 @@ static inline void ocfs2_save_inode_ac_group(struct inode *dir, OCFS2_I(dir)->ip_last_used_slot = ac->ac_alloc_slot; } +int ocfs2_find_new_inode_loc(struct inode *dir, + struct buffer_head *parent_fe_bh, + struct ocfs2_alloc_context *ac, + u64 *fe_blkno) +{ + int ret; + handle_t *handle = NULL; + struct ocfs2_suballoc_result *res; + + BUG_ON(!ac); + BUG_ON(ac->ac_bits_given != 0); + BUG_ON(ac->ac_bits_wanted != 1); + BUG_ON(ac->ac_which != OCFS2_AC_USE_INODE); + + res = kzalloc(sizeof(*res), GFP_NOFS); + if (res == NULL) { + ret = -ENOMEM; + mlog_errno(ret); + goto out; + } + + ocfs2_init_inode_ac_group(dir, parent_fe_bh, ac); + + /* + * The handle started here is for chain relink. Alternatively, + * we could just disable relink for these calls. + */ + handle = ocfs2_start_trans(OCFS2_SB(dir->i_sb), OCFS2_SUBALLOC_ALLOC); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + handle = NULL; + mlog_errno(ret); + goto out; + } + + /* + * This will instruct ocfs2_claim_suballoc_bits and + * ocfs2_search_one_group to search but save actual allocation + * for later. + */ + ac->ac_find_loc_only = 1; + + ret = ocfs2_claim_suballoc_bits(ac, handle, 1, 1, res); + if (ret < 0) { + mlog_errno(ret); + goto out; + } + + ac->ac_find_loc_priv = res; + *fe_blkno = res->sr_blkno; + +out: + if (handle) + ocfs2_commit_trans(OCFS2_SB(dir->i_sb), handle); + + if (ret) + kfree(res); + + return ret; +} + +int ocfs2_claim_new_inode_at_loc(handle_t *handle, + struct inode *dir, + struct ocfs2_alloc_context *ac, + u64 *suballoc_loc, + u16 *suballoc_bit, + u64 di_blkno) +{ + int ret; + u16 chain; + struct ocfs2_suballoc_result *res = ac->ac_find_loc_priv; + struct buffer_head *bg_bh = NULL; + struct ocfs2_group_desc *bg; + struct ocfs2_dinode *di = (struct ocfs2_dinode *) ac->ac_bh->b_data; + + /* + * Since di_blkno is being passed back in, we check for any + * inconsistencies which may have happened between + * calls. These are code bugs as di_blkno is not expected to + * change once returned from ocfs2_find_new_inode_loc() + */ + BUG_ON(res->sr_blkno != di_blkno); + + ret = ocfs2_read_group_descriptor(ac->ac_inode, di, + res->sr_bg_stable_blkno, &bg_bh); + if (ret) { + mlog_errno(ret); + goto out; + } + + bg = (struct ocfs2_group_desc *) bg_bh->b_data; + chain = le16_to_cpu(bg->bg_chain); + + ret = ocfs2_alloc_dinode_update_counts(ac->ac_inode, handle, + ac->ac_bh, res->sr_bits, + chain); + if (ret) { + mlog_errno(ret); + goto out; + } + + ret = ocfs2_block_group_set_bits(handle, + ac->ac_inode, + bg, + bg_bh, + res->sr_bit_offset, + res->sr_bits); + if (ret < 0) { + mlog_errno(ret); + goto out; + } + + mlog(0, "Allocated %u bits from suballocator %llu\n", res->sr_bits, + (unsigned long long)di_blkno); + + atomic_inc(&OCFS2_SB(ac->ac_inode->i_sb)->alloc_stats.bg_allocs); + + BUG_ON(res->sr_bits != 1); + + *suballoc_loc = res->sr_bg_blkno; + *suballoc_bit = res->sr_bit_offset; + ac->ac_bits_given++; + ocfs2_save_inode_ac_group(dir, ac); + +out: + brelse(bg_bh); + + return ret; +} + int ocfs2_claim_new_inode(handle_t *handle, struct inode *dir, struct buffer_head *parent_fe_bh, diff --git a/fs/ocfs2/suballoc.h b/fs/ocfs2/suballoc.h index a017dd3..b8afabf 100644 --- a/fs/ocfs2/suballoc.h +++ b/fs/ocfs2/suballoc.h @@ -56,6 +56,9 @@ struct ocfs2_alloc_context { u64 ac_max_block; /* Highest block number to allocate. 0 is is the same as ~0 - unlimited */ + int ac_find_loc_only; /* hack for reflink operation ordering */ + struct ocfs2_suballoc_result *ac_find_loc_priv; /* */ + struct ocfs2_alloc_reservation *ac_resv; }; @@ -197,4 +200,22 @@ int ocfs2_lock_allocators(struct inode *inode, struct ocfs2_extent_tree *et, struct ocfs2_alloc_context **meta_ac); int ocfs2_test_inode_bit(struct ocfs2_super *osb, u64 blkno, int *res); + + + +/* + * The following two interfaces are for ocfs2_create_inode_in_orphan(). + */ +int ocfs2_find_new_inode_loc(struct inode *dir, + struct buffer_head *parent_fe_bh, + struct ocfs2_alloc_context *ac, + u64 *fe_blkno); + +int ocfs2_claim_new_inode_at_loc(handle_t *handle, + struct inode *dir, + struct ocfs2_alloc_context *ac, + u64 *suballoc_loc, + u16 *suballoc_bit, + u64 di_blkno); + #endif /* _CHAINALLOC_H_ */ -- cgit v0.10.2 From dd43bcde23c527f64897eef41aa1fed2c9905ea9 Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Fri, 13 Aug 2010 15:15:18 -0700 Subject: ocfs2: split out ocfs2_prepare_orphan_dir() into locking and prep functions We do this because ocfs2_create_inode_in_orphan() wants to order locking of the orphan dir with respect to locking of the inode allocator *before* making any changes to the directory. Signed-off-by: Mark Fasheh Signed-off-by: Tao Ma diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index 2aa66b6..54c6298 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -1871,61 +1871,117 @@ bail: return status; } -static int ocfs2_prepare_orphan_dir(struct ocfs2_super *osb, - struct inode **ret_orphan_dir, - u64 blkno, - char *name, - struct ocfs2_dir_lookup_result *lookup) +static int ocfs2_lookup_lock_orphan_dir(struct ocfs2_super *osb, + struct inode **ret_orphan_dir, + struct buffer_head **ret_orphan_dir_bh) { struct inode *orphan_dir_inode; struct buffer_head *orphan_dir_bh = NULL; - int status = 0; - - status = ocfs2_blkno_stringify(blkno, name); - if (status < 0) { - mlog_errno(status); - return status; - } + int ret = 0; orphan_dir_inode = ocfs2_get_system_file_inode(osb, ORPHAN_DIR_SYSTEM_INODE, osb->slot_num); if (!orphan_dir_inode) { - status = -ENOENT; - mlog_errno(status); - return status; + ret = -ENOENT; + mlog_errno(ret); + return ret; } mutex_lock(&orphan_dir_inode->i_mutex); - status = ocfs2_inode_lock(orphan_dir_inode, &orphan_dir_bh, 1); - if (status < 0) { - mlog_errno(status); - goto leave; + ret = ocfs2_inode_lock(orphan_dir_inode, &orphan_dir_bh, 1); + if (ret < 0) { + mutex_unlock(&orphan_dir_inode->i_mutex); + iput(orphan_dir_inode); + + mlog_errno(ret); + return ret; } - status = ocfs2_prepare_dir_for_insert(osb, orphan_dir_inode, - orphan_dir_bh, name, - OCFS2_ORPHAN_NAMELEN, lookup); - if (status < 0) { - ocfs2_inode_unlock(orphan_dir_inode, 1); + *ret_orphan_dir = orphan_dir_inode; + *ret_orphan_dir_bh = orphan_dir_bh; - mlog_errno(status); - goto leave; + return 0; +} + +static int __ocfs2_prepare_orphan_dir(struct inode *orphan_dir_inode, + struct buffer_head *orphan_dir_bh, + u64 blkno, + char *name, + struct ocfs2_dir_lookup_result *lookup) +{ + int ret; + struct ocfs2_super *osb = OCFS2_SB(orphan_dir_inode->i_sb); + + ret = ocfs2_blkno_stringify(blkno, name); + if (ret < 0) { + mlog_errno(ret); + return ret; + } + + ret = ocfs2_prepare_dir_for_insert(osb, orphan_dir_inode, + orphan_dir_bh, name, + OCFS2_ORPHAN_NAMELEN, lookup); + if (ret < 0) { + mlog_errno(ret); + return ret; + } + + return 0; +} + +/** + * ocfs2_prepare_orphan_dir() - Prepare an orphan directory for + * insertion of an orphan. + * @osb: ocfs2 file system + * @ret_orphan_dir: Orphan dir inode - returned locked! + * @blkno: Actual block number of the inode to be inserted into orphan dir. + * @lookup: dir lookup result, to be passed back into functions like + * ocfs2_orphan_add + * + * Returns zero on success and the ret_orphan_dir, name and lookup + * fields will be populated. + * + * Returns non-zero on failure. + */ +static int ocfs2_prepare_orphan_dir(struct ocfs2_super *osb, + struct inode **ret_orphan_dir, + u64 blkno, + char *name, + struct ocfs2_dir_lookup_result *lookup) +{ + struct inode *orphan_dir_inode = NULL; + struct buffer_head *orphan_dir_bh = NULL; + int ret = 0; + + ret = ocfs2_lookup_lock_orphan_dir(osb, &orphan_dir_inode, + &orphan_dir_bh); + if (ret < 0) { + mlog_errno(ret); + return ret; + } + + ret = __ocfs2_prepare_orphan_dir(orphan_dir_inode, orphan_dir_bh, + blkno, name, lookup); + if (ret < 0) { + mlog_errno(ret); + goto out; } *ret_orphan_dir = orphan_dir_inode; -leave: - if (status) { +out: + brelse(orphan_dir_bh); + + if (ret) { + ocfs2_inode_unlock(orphan_dir_inode, 1); mutex_unlock(&orphan_dir_inode->i_mutex); iput(orphan_dir_inode); } - brelse(orphan_dir_bh); - - mlog_exit(status); - return status; + mlog_exit(ret); + return ret; } static int ocfs2_orphan_add(struct ocfs2_super *osb, -- cgit v0.10.2 From 97b8f4a9dfd932997677136e11980eb2fafea91d Mon Sep 17 00:00:00 2001 From: Mark Fasheh Date: Fri, 13 Aug 2010 15:15:19 -0700 Subject: ocfs2: Fix orphan add in ocfs2_create_inode_in_orphan ocfs2_create_inode_in_orphan() is used by reflink to create the newly reflinked inode simultaneously in the orphan dir. This allows us to easily handle partially-reflinked files during recovery cleanup. We have a problem though - the orphan dir stringifies inode # to determine a unique name under which the orphan entry dirent can be created. Since ocfs2_create_inode_in_orphan() needs the space allocated in the orphan dir before it can allocate the inode, we currently call into the orphan code: /* * We give the orphan dir the root blkno to fake an orphan name, * and allocate enough space for our insertion. */ status = ocfs2_prepare_orphan_dir(osb, &orphan_dir, osb->root_blkno, orphan_name, &orphan_insert); Using osb->root_blkno might work fine on unindexed directories, but the orphan dir can have an index. When it has that index, the above code fails to allocate the proper index entry. Later, when we try to remove the file from the orphan dir (using the actual inode #), the reflink operation will fail. To fix this, I created a function ocfs2_alloc_orphaned_file() which uses the newly split out orphan and inode alloc code to figure out what the inode block number will be (once allocated) and then prepare the orphan dir from that data. Signed-off-by: Mark Fasheh Signed-off-by: Tao Ma diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index 54c6298..a00dda2 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -2128,6 +2128,99 @@ leave: return status; } +/** + * ocfs2_prep_new_orphaned_file() - Prepare the orphan dir to recieve a newly + * allocated file. This is different from the typical 'add to orphan dir' + * operation in that the inode does not yet exist. This is a problem because + * the orphan dir stringifies the inode block number to come up with it's + * dirent. Obviously if the inode does not yet exist we have a chicken and egg + * problem. This function works around it by calling deeper into the orphan + * and suballoc code than other callers. Use this only by necessity. + * @dir: The directory which this inode will ultimately wind up under - not the + * orphan dir! + * @dir_bh: buffer_head the @dir inode block + * @orphan_name: string of length (CFS2_ORPHAN_NAMELEN + 1). Will be filled + * with the string to be used for orphan dirent. Pass back to the orphan dir + * code. + * @ret_orphan_dir: orphan dir inode returned to be passed back into orphan + * dir code. + * @ret_di_blkno: block number where the new inode will be allocated. + * @orphan_insert: Dir insert context to be passed back into orphan dir code. + * @ret_inode_ac: Inode alloc context to be passed back to the allocator. + * + * Returns zero on success and the ret_orphan_dir, name and lookup + * fields will be populated. + * + * Returns non-zero on failure. + */ +static int ocfs2_prep_new_orphaned_file(struct inode *dir, + struct buffer_head *dir_bh, + char *orphan_name, + struct inode **ret_orphan_dir, + u64 *ret_di_blkno, + struct ocfs2_dir_lookup_result *orphan_insert, + struct ocfs2_alloc_context **ret_inode_ac) +{ + int ret; + u64 di_blkno; + struct ocfs2_super *osb = OCFS2_SB(dir->i_sb); + struct inode *orphan_dir = NULL; + struct buffer_head *orphan_dir_bh = NULL; + struct ocfs2_alloc_context *inode_ac = NULL; + + ret = ocfs2_lookup_lock_orphan_dir(osb, &orphan_dir, &orphan_dir_bh); + if (ret < 0) { + mlog_errno(ret); + return ret; + } + + /* reserve an inode spot */ + ret = ocfs2_reserve_new_inode(osb, &inode_ac); + if (ret < 0) { + if (ret != -ENOSPC) + mlog_errno(ret); + goto out; + } + + ret = ocfs2_find_new_inode_loc(dir, dir_bh, inode_ac, + &di_blkno); + if (ret) { + mlog_errno(ret); + goto out; + } + + ret = __ocfs2_prepare_orphan_dir(orphan_dir, orphan_dir_bh, + di_blkno, orphan_name, orphan_insert); + if (ret < 0) { + mlog_errno(ret); + goto out; + } + +out: + if (ret == 0) { + *ret_orphan_dir = orphan_dir; + *ret_di_blkno = di_blkno; + *ret_inode_ac = inode_ac; + /* + * orphan_name and orphan_insert are already up to + * date via prepare_orphan_dir + */ + } else { + /* Unroll reserve_new_inode* */ + if (inode_ac) + ocfs2_free_alloc_context(inode_ac); + + /* Unroll orphan dir locking */ + mutex_unlock(&orphan_dir->i_mutex); + ocfs2_inode_unlock(orphan_dir, 1); + iput(orphan_dir); + } + + brelse(orphan_dir_bh); + + return 0; +} + int ocfs2_create_inode_in_orphan(struct inode *dir, int mode, struct inode **new_inode) @@ -2143,6 +2236,8 @@ int ocfs2_create_inode_in_orphan(struct inode *dir, struct buffer_head *new_di_bh = NULL; struct ocfs2_alloc_context *inode_ac = NULL; struct ocfs2_dir_lookup_result orphan_insert = { NULL, }; + u64 uninitialized_var(di_blkno), suballoc_loc; + u16 suballoc_bit; status = ocfs2_inode_lock(dir, &parent_di_bh, 1); if (status < 0) { @@ -2151,20 +2246,9 @@ int ocfs2_create_inode_in_orphan(struct inode *dir, return status; } - /* - * We give the orphan dir the root blkno to fake an orphan name, - * and allocate enough space for our insertion. - */ - status = ocfs2_prepare_orphan_dir(osb, &orphan_dir, - osb->root_blkno, - orphan_name, &orphan_insert); - if (status < 0) { - mlog_errno(status); - goto leave; - } - - /* reserve an inode spot */ - status = ocfs2_reserve_new_inode(osb, &inode_ac); + status = ocfs2_prep_new_orphaned_file(dir, parent_di_bh, + orphan_name, &orphan_dir, + &di_blkno, &orphan_insert, &inode_ac); if (status < 0) { if (status != -ENOSPC) mlog_errno(status); @@ -2191,17 +2275,20 @@ int ocfs2_create_inode_in_orphan(struct inode *dir, goto leave; did_quota_inode = 1; - inode->i_nlink = 0; - /* do the real work now. */ - status = ocfs2_mknod_locked(osb, dir, inode, - 0, &new_di_bh, parent_di_bh, handle, - inode_ac); + status = ocfs2_claim_new_inode_at_loc(handle, dir, inode_ac, + &suballoc_loc, + &suballoc_bit, di_blkno); if (status < 0) { mlog_errno(status); goto leave; } - status = ocfs2_blkno_stringify(OCFS2_I(inode)->ip_blkno, orphan_name); + inode->i_nlink = 0; + /* do the real work now. */ + status = __ocfs2_mknod_locked(dir, inode, + 0, &new_di_bh, parent_di_bh, handle, + inode_ac, di_blkno, suballoc_loc, + suballoc_bit); if (status < 0) { mlog_errno(status); goto leave; -- cgit v0.10.2 From a769cbcf60cee51f4431c0938acd39e7e5b76b8d Mon Sep 17 00:00:00 2001 From: Brian Austin Date: Tue, 7 Sep 2010 14:36:22 -0500 Subject: ALSA: hda - Add errata initverb sequence for CS42xx codecs Add init verb sequence for errata ER880C3 http://www.cirrus.com/en/pubs/errata/ER880C3.pdf Signed-off-by: Brian Austin Cc: Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_cirrus.c b/sound/pci/hda/patch_cirrus.c index 4ef5efa..488fd9a 100644 --- a/sound/pci/hda/patch_cirrus.c +++ b/sound/pci/hda/patch_cirrus.c @@ -972,6 +972,53 @@ static struct hda_verb cs_coef_init_verbs[] = { {} /* terminator */ }; +/* Errata: CS4207 rev C0/C1/C2 Silicon + * + * http://www.cirrus.com/en/pubs/errata/ER880C3.pdf + * + * 6. At high temperature (TA > +85°C), the digital supply current (IVD) + * may be excessive (up to an additional 200 μA), which is most easily + * observed while the part is being held in reset (RESET# active low). + * + * Root Cause: At initial powerup of the device, the logic that drives + * the clock and write enable to the S/PDIF SRC RAMs is not properly + * initialized. + * Certain random patterns will cause a steady leakage current in those + * RAM cells. The issue will resolve once the SRCs are used (turned on). + * + * Workaround: The following verb sequence briefly turns on the S/PDIF SRC + * blocks, which will alleviate the issue. + */ + +static struct hda_verb cs_errata_init_verbs[] = { + {0x01, AC_VERB_SET_POWER_STATE, 0x00}, /* AFG: D0 */ + {0x11, AC_VERB_SET_PROC_STATE, 0x01}, /* VPW: processing on */ + + {0x11, AC_VERB_SET_COEF_INDEX, 0x0008}, + {0x11, AC_VERB_SET_PROC_COEF, 0x9999}, + {0x11, AC_VERB_SET_COEF_INDEX, 0x0017}, + {0x11, AC_VERB_SET_PROC_COEF, 0xa412}, + {0x11, AC_VERB_SET_COEF_INDEX, 0x0001}, + {0x11, AC_VERB_SET_PROC_COEF, 0x0009}, + + {0x07, AC_VERB_SET_POWER_STATE, 0x00}, /* S/PDIF Rx: D0 */ + {0x08, AC_VERB_SET_POWER_STATE, 0x00}, /* S/PDIF Tx: D0 */ + + {0x11, AC_VERB_SET_COEF_INDEX, 0x0017}, + {0x11, AC_VERB_SET_PROC_COEF, 0x2412}, + {0x11, AC_VERB_SET_COEF_INDEX, 0x0008}, + {0x11, AC_VERB_SET_PROC_COEF, 0x0000}, + {0x11, AC_VERB_SET_COEF_INDEX, 0x0001}, + {0x11, AC_VERB_SET_PROC_COEF, 0x0008}, + {0x11, AC_VERB_SET_PROC_STATE, 0x00}, + + {0x07, AC_VERB_SET_POWER_STATE, 0x03}, /* S/PDIF Rx: D3 */ + {0x08, AC_VERB_SET_POWER_STATE, 0x03}, /* S/PDIF Tx: D3 */ + /*{0x01, AC_VERB_SET_POWER_STATE, 0x03},*/ /* AFG: D3 This is already handled */ + + {} /* terminator */ +}; + /* SPDIF setup */ static void init_digital(struct hda_codec *codec) { @@ -991,6 +1038,9 @@ static int cs_init(struct hda_codec *codec) { struct cs_spec *spec = codec->spec; + /* init_verb sequence for C0/C1/C2 errata*/ + snd_hda_sequence_write(codec, cs_errata_init_verbs); + snd_hda_sequence_write(codec, cs_coef_init_verbs); if (spec->gpio_mask) { -- cgit v0.10.2 From 080dc7bc2562615a5be0a705a9d1a8c24eb198d4 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 8 Sep 2010 08:38:41 +0200 Subject: ALSA: hda - Enable PC-beep for EeePC with ALC269 codec EeePC 1001HAG has a similar problem like other ASUS machine, which doesn't set the codec SSID properly for indicating the beep capability. To enable PC-beep again, put this to the whitelist. Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 627bf99..bcbf916 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -5334,6 +5334,7 @@ static void fillup_priv_adc_nids(struct hda_codec *codec, hda_nid_t *nids, static struct snd_pci_quirk beep_white_list[] = { SND_PCI_QUIRK(0x1043, 0x829f, "ASUS", 1), + SND_PCI_QUIRK(0x1043, 0x83ce, "EeePC", 1), SND_PCI_QUIRK(0x8086, 0xd613, "Intel", 1), {} }; -- cgit v0.10.2 From 12e8ba25ef52f19e7a42e61aecb3c1fef83b2a82 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Tue, 7 Sep 2010 23:39:28 +0100 Subject: Revert "drm/i915: Allow LVDS on pipe A on gen4+" This reverts commit 0f3ee801b332d6ff22285386675fe5aaedf035c3. Enabling LVDS on pipe A was causing excessive wakeups on otherwise idle systems due to i915 interrupts. So restrict the LVDS to pipe B once more, whilst the issue is properly diagnosed. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=16307 Reported-and-tested-by: Enrico Bandiello Poked-by: Florian Mickler Signed-off-by: Chris Wilson Cc: Adam Jackson Cc: stable@kernel.org diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c index b819c10..4fbb016 100644 --- a/drivers/gpu/drm/i915/intel_lvds.c +++ b/drivers/gpu/drm/i915/intel_lvds.c @@ -875,8 +875,6 @@ void intel_lvds_init(struct drm_device *dev) intel_encoder->clone_mask = (1 << INTEL_LVDS_CLONE_BIT); intel_encoder->crtc_mask = (1 << 1); - if (IS_I965G(dev)) - intel_encoder->crtc_mask |= (1 << 0); drm_encoder_helper_add(encoder, &intel_lvds_helper_funcs); drm_connector_helper_add(connector, &intel_lvds_connector_helper_funcs); connector->display_info.subpixel_order = SubPixelHorizontalRGB; -- cgit v0.10.2 From c3add4b63438555d5e88c5893d238ab80d1f5959 Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Wed, 8 Sep 2010 09:14:08 +0100 Subject: Revert "drm/i915: Warn if we run out of FIFO space for a mode" This reverts commit b9421ae8f30958deea98d71477b4a77a066856b4. This warning was so prelevant, even for apparently working machines, that it was just causing fear, anxiety and panic. The root cause still remains, so we will add some better debugging when we focus on fixing it. Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=17021 Reported-by: Maciej Rutecki Signed-off-by: Chris Wilson diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index e0d1952..7c91030 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -2767,14 +2767,8 @@ static unsigned long intel_calculate_wm(unsigned long clock_in_khz, /* Don't promote wm_size to unsigned... */ if (wm_size > (long)wm->max_wm) wm_size = wm->max_wm; - if (wm_size <= 0) { + if (wm_size <= 0) wm_size = wm->default_wm; - DRM_ERROR("Insufficient FIFO for plane, expect flickering:" - " entries required = %ld, available = %lu.\n", - entries_required + wm->guard_size, - wm->fifo_size); - } - return wm_size; } -- cgit v0.10.2 From 0485e18bc4112a3b548baa314c24bfbece4d156b Mon Sep 17 00:00:00 2001 From: Russell King Date: Sun, 5 Sep 2010 12:19:50 +0100 Subject: Revert "[ARM] pxa: remove now unnecessary dma_needs_bounce()" This reverts commit 4fa5518, which causes a compilation regression for IXP4xx platforms. Reported-by: Richard Cochran Acked-by: Eric Miao Signed-off-by: Russell King diff --git a/arch/arm/common/it8152.c b/arch/arm/common/it8152.c index 6c09135..7974baa 100644 --- a/arch/arm/common/it8152.c +++ b/arch/arm/common/it8152.c @@ -263,6 +263,14 @@ static int it8152_pci_platform_notify_remove(struct device *dev) return 0; } +int dma_needs_bounce(struct device *dev, dma_addr_t dma_addr, size_t size) +{ + dev_dbg(dev, "%s: dma_addr %08x, size %08x\n", + __func__, dma_addr, size); + return (dev->bus == &pci_bus_type) && + ((dma_addr + size - PHYS_OFFSET) >= SZ_64M); +} + int __init it8152_pci_setup(int nr, struct pci_sys_data *sys) { it8152_io.start = IT8152_IO_BASE + 0x12000; diff --git a/arch/arm/include/asm/dma-mapping.h b/arch/arm/include/asm/dma-mapping.h index c226fe1..c568da7 100644 --- a/arch/arm/include/asm/dma-mapping.h +++ b/arch/arm/include/asm/dma-mapping.h @@ -288,15 +288,7 @@ extern void dmabounce_unregister_dev(struct device *); * DMA access and 1 if the buffer needs to be bounced. * */ -#ifdef CONFIG_SA1111 extern int dma_needs_bounce(struct device*, dma_addr_t, size_t); -#else -static inline int dma_needs_bounce(struct device *dev, dma_addr_t addr, - size_t size) -{ - return 0; -} -#endif /* * The DMA API, implemented by dmabounce.c. See below for descriptions. -- cgit v0.10.2 From a387f0f5409276c5cf75eecb61ef6e6896f851b4 Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Fri, 3 Sep 2010 17:14:54 +0100 Subject: ARM: 6359/1: ep93xx: move clock initialization earlier Commit 7cfe24947 ("ARM: AMBA: Add pclk support to AMBA bus infrastructure") changed AMBA bus to handle the PCLK automatically. However, in EP93xx clock initialization is arch_initcall which is done later than AMBA device identification. This causes amba_get_enable_pclk() to fail resulting device where UARTs are not functional. So change ep93xx_clock_init() to be postcore_initcall. Signed-off-by: Mika Westerberg Acked-by: H Hartley Sweeten Signed-off-by: Russell King diff --git a/arch/arm/mach-ep93xx/clock.c b/arch/arm/mach-ep93xx/clock.c index 8bf3cec..4566bd1 100644 --- a/arch/arm/mach-ep93xx/clock.c +++ b/arch/arm/mach-ep93xx/clock.c @@ -560,4 +560,4 @@ static int __init ep93xx_clock_init(void) clkdev_add_table(clocks, ARRAY_SIZE(clocks)); return 0; } -arch_initcall(ep93xx_clock_init); +postcore_initcall(ep93xx_clock_init); -- cgit v0.10.2 From 122661b67899980f1372812d907e73ebcfb3d037 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 8 Sep 2010 14:57:04 +0200 Subject: ALSA: hda - Fix wrong HP pin detection in snd_hda_parse_pin_def_config() snd_hda_parse_pin_def_config() has some workaround for re-assigning some pins declared as headphones to line-outs. This didn't work properly for some cases because it used memmove() stupidly wrongly. Reference: Novell bnc#637263 https://bugzilla.novell.com/show_bug.cgi?id=637263 Cc: Signed-off-by: Takashi Iwai diff --git a/sound/pci/hda/hda_codec.c b/sound/pci/hda/hda_codec.c index 3827092..1482921 100644 --- a/sound/pci/hda/hda_codec.c +++ b/sound/pci/hda/hda_codec.c @@ -4536,7 +4536,7 @@ int snd_hda_parse_pin_def_config(struct hda_codec *codec, cfg->hp_outs--; memmove(cfg->hp_pins + i, cfg->hp_pins + i + 1, sizeof(cfg->hp_pins[0]) * (cfg->hp_outs - i)); - memmove(sequences_hp + i - 1, sequences_hp + i, + memmove(sequences_hp + i, sequences_hp + i + 1, sizeof(sequences_hp[0]) * (cfg->hp_outs - i)); } } -- cgit v0.10.2 From febc88c5948f81114f64c3412011d695aecae233 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Tue, 7 Sep 2010 14:46:37 +0200 Subject: semaphore: Add DEFINE_SEMAPHORE The full cleanup of init_MUTEX[_LOCKED] and DECLARE_MUTEX has not been done. Some of the users are real semaphores and we should name them as such instead of confusing everyone with "MUTEX". Provide the infrastructure to get finally rid of init_MUTEX[_LOCKED] and DECLARE_MUTEX. Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: Christoph Hellwig LKML-Reference: <20100907125054.795929962@linutronix.de> diff --git a/include/linux/semaphore.h b/include/linux/semaphore.h index 7415839..5310d27 100644 --- a/include/linux/semaphore.h +++ b/include/linux/semaphore.h @@ -26,6 +26,9 @@ struct semaphore { .wait_list = LIST_HEAD_INIT((name).wait_list), \ } +#define DEFINE_SEMAPHORE(name) \ + struct semaphore name = __SEMAPHORE_INITIALIZER(name, 1) + #define DECLARE_MUTEX(name) \ struct semaphore name = __SEMAPHORE_INITIALIZER(name, 1) -- cgit v0.10.2 From 9c55cb12c1c172e2d51e85fbb5a4796ca86b77e7 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Wed, 8 Sep 2010 11:20:37 -0400 Subject: tracing: Do not allow llseek to set_ftrace_filter Reading the file set_ftrace_filter does three things. 1) shows whether or not filters are set for the function tracer 2) shows what functions are set for the function tracer 3) shows what triggers are set on any functions 3 is independent from 1 and 2. The way this file currently works is that it is a state machine, and as you read it, it may change state. But this assumption breaks when you use lseek() on the file. The state machine gets out of sync and the t_show() may use the wrong pointer and cause a kernel oops. Luckily, this will only kill the app that does the lseek, but the app dies while holding a mutex. This prevents anyone else from using the set_ftrace_filter file (or any other function tracing file for that matter). A real fix for this is to rewrite the code, but that is too much for a -rc release or stable. This patch simply disables llseek on the set_ftrace_filter() file for now, and we can do the proper fix for the next major release. Reported-by: Robert Swiecki Cc: Chris Wright Cc: Tavis Ormandy Cc: Eugene Teo Cc: vendor-sec@lst.de Cc: Signed-off-by: Steven Rostedt diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 7cb1f45..83a16e9 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -2416,7 +2416,7 @@ static const struct file_operations ftrace_filter_fops = { .open = ftrace_filter_open, .read = seq_read, .write = ftrace_filter_write, - .llseek = ftrace_regex_lseek, + .llseek = no_llseek, .release = ftrace_filter_release, }; -- cgit v0.10.2 From 61a527362234ac3352a91ac67c50c6f7cd248eb1 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Fri, 27 Aug 2010 20:38:46 +0900 Subject: tracing/kprobe: Fix a memory leak in error case Fix a memory leak which happens when a field name conflicts with others. In error case, free_trace_probe() will free all arguments until nr_args, so this increments nr_args the begining of the loop instead of the end. Cc: Steven Rostedt Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Mathieu Desnoyers LKML-Reference: <20100827113846.22882.12670.stgit@ltc236.sdl.hitachi.co.jp> Signed-off-by: Masami Hiramatsu Signed-off-by: Arnaldo Carvalho de Melo diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 8b27c98..0116c03 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -992,6 +992,9 @@ static int create_trace_probe(int argc, char **argv) /* parse arguments */ ret = 0; for (i = 0; i < argc && i < MAX_TRACE_ARGS; i++) { + /* Increment count for freeing args in error case */ + tp->nr_args++; + /* Parse argument name */ arg = strchr(argv[i], '='); if (arg) @@ -1021,11 +1024,8 @@ static int create_trace_probe(int argc, char **argv) ret = parse_probe_arg(arg, tp, &tp->args[i], is_return); if (ret) { pr_info("Parse error at argument%d. (%d)\n", i, ret); - kfree(tp->args[i].name); goto error; } - - tp->nr_args++; } ret = register_trace_probe(tp); -- cgit v0.10.2 From 04ddd04b044d8896a4f8a921b23ba09d365df196 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Fri, 27 Aug 2010 20:38:53 +0900 Subject: perf probe: Fix return probe support Fix a bug to support %return probe syntax again. Previous commit 4235b04 has a bug which disables the %return syntax on perf probe. Cc: Peter Zijlstra Cc: Paul Mackerras Cc: Ingo Molnar Cc: Arnaldo Carvalho de Melo Cc: Frederic Weisbecker LKML-Reference: <20100827113852.22882.87447.stgit@ltc236.sdl.hitachi.co.jp> Signed-off-by: Masami Hiramatsu Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c index e72f05c..fcc16e4 100644 --- a/tools/perf/util/probe-event.c +++ b/tools/perf/util/probe-event.c @@ -1539,6 +1539,7 @@ static int convert_to_probe_trace_events(struct perf_probe_event *pev, goto error; } tev->point.offset = pev->point.offset; + tev->point.retprobe = pev->point.retprobe; tev->nargs = pev->nargs; if (tev->nargs) { tev->args = zalloc(sizeof(struct probe_trace_arg) diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c index 5251366..9b0e1b1 100644 --- a/tools/perf/util/probe-finder.c +++ b/tools/perf/util/probe-finder.c @@ -783,6 +783,16 @@ static int convert_probe_point(Dwarf_Die *sp_die, struct probe_finder *pf) /* This function has no name. */ tev->point.offset = (unsigned long)pf->addr; + /* Return probe must be on the head of a subprogram */ + if (pf->pev->point.retprobe) { + if (tev->point.offset != 0) { + pr_warning("Return probe must be on the head of" + " a real function\n"); + return -EINVAL; + } + tev->point.retprobe = true; + } + pr_debug("Probe point found: %s+%lu\n", tev->point.symbol, tev->point.offset); -- cgit v0.10.2 From 367e94c10092469c896a226a77ef13cf6da757e4 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Fri, 27 Aug 2010 20:38:59 +0900 Subject: perf probe: Fix handling of arguments names Don't make argument names from raw parameters (means the parameters are written in kprobe-tracer syntax), because the argument syntax may include special characters. Just leave it, then kprobe-tracer gives a new name. Reported-by: Srikar Dronamraju Cc: Peter Zijlstra Cc: Paul Mackerras Cc: Ingo Molnar Cc: Arnaldo Carvalho de Melo Cc: Frederic Weisbecker LKML-Reference: <20100827113859.22882.75598.stgit@ltc236.sdl.hitachi.co.jp> Signed-off-by: Masami Hiramatsu Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c index 9b0e1b1..32b81f7 100644 --- a/tools/perf/util/probe-finder.c +++ b/tools/perf/util/probe-finder.c @@ -686,6 +686,25 @@ static int find_variable(Dwarf_Die *sp_die, struct probe_finder *pf) char buf[32], *ptr; int ret, nscopes; + if (!is_c_varname(pf->pvar->var)) { + /* Copy raw parameters */ + pf->tvar->value = strdup(pf->pvar->var); + if (pf->tvar->value == NULL) + return -ENOMEM; + if (pf->pvar->type) { + pf->tvar->type = strdup(pf->pvar->type); + if (pf->tvar->type == NULL) + return -ENOMEM; + } + if (pf->pvar->name) { + pf->tvar->name = strdup(pf->pvar->name); + if (pf->tvar->name == NULL) + return -ENOMEM; + } else + pf->tvar->name = NULL; + return 0; + } + if (pf->pvar->name) pf->tvar->name = strdup(pf->pvar->name); else { @@ -700,19 +719,6 @@ static int find_variable(Dwarf_Die *sp_die, struct probe_finder *pf) if (pf->tvar->name == NULL) return -ENOMEM; - if (!is_c_varname(pf->pvar->var)) { - /* Copy raw parameters */ - pf->tvar->value = strdup(pf->pvar->var); - if (pf->tvar->value == NULL) - return -ENOMEM; - if (pf->pvar->type) { - pf->tvar->type = strdup(pf->pvar->type); - if (pf->tvar->type == NULL) - return -ENOMEM; - } - return 0; - } - pr_debug("Searching '%s' variable in context.\n", pf->pvar->var); /* Search child die for local variables and parameters. */ -- cgit v0.10.2 From aba91595cfcebd193425e20aabc407531526a1c5 Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Fri, 27 Aug 2010 20:39:06 +0900 Subject: tracing/kprobes: Fix handling of argument names Set "argN" name for each argument automatically if it has no specified name. Since dynamic trace event(kprobe_events) accepts special characters for its argument, its format can show those special characters (e.g. '$', '%', '+'). However, perf can't parse those format because of the character (especially '%') mess up the format. This sets "argX" name for those arguments if user omitted the argument names. E.g. # echo 'p do_fork %ax IP=%ip $stack' > tracing/kprobe_events # cat tracing/kprobe_events p:kprobes/p_do_fork_0 do_fork arg1=%ax IP=%ip arg3=$stack Reported-by: Srikar Dronamraju Cc: Steven Rostedt Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Mathieu Desnoyers LKML-Reference: <20100827113906.22882.59312.stgit@ltc236.sdl.hitachi.co.jp> Signed-off-by: Masami Hiramatsu Signed-off-by: Arnaldo Carvalho de Melo diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 0116c03..a39251e 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -997,15 +997,18 @@ static int create_trace_probe(int argc, char **argv) /* Parse argument name */ arg = strchr(argv[i], '='); - if (arg) + if (arg) { *arg++ = '\0'; - else + tp->args[i].name = kstrdup(argv[i], GFP_KERNEL); + } else { arg = argv[i]; + /* If argument name is omitted, set "argN" */ + snprintf(buf, MAX_EVENT_NAME_LEN, "arg%d", i + 1); + tp->args[i].name = kstrdup(buf, GFP_KERNEL); + } - tp->args[i].name = kstrdup(argv[i], GFP_KERNEL); if (!tp->args[i].name) { - pr_info("Failed to allocate argument%d name '%s'.\n", - i, argv[i]); + pr_info("Failed to allocate argument[%d] name.\n", i); ret = -ENOMEM; goto error; } @@ -1014,7 +1017,7 @@ static int create_trace_probe(int argc, char **argv) *tmp = '_'; /* convert : to _ */ if (conflict_field_name(tp->args[i].name, tp->args, i)) { - pr_info("Argument%d name '%s' conflicts with " + pr_info("Argument[%d] name '%s' conflicts with " "another field.\n", i, argv[i]); ret = -EINVAL; goto error; @@ -1023,7 +1026,7 @@ static int create_trace_probe(int argc, char **argv) /* Parse fetch argument */ ret = parse_probe_arg(arg, tp, &tp->args[i], is_return); if (ret) { - pr_info("Parse error at argument%d. (%d)\n", i, ret); + pr_info("Parse error at argument[%d]. (%d)\n", i, ret); goto error; } } -- cgit v0.10.2 From da34634fd39958725310d2c30c9b4543945f968b Mon Sep 17 00:00:00 2001 From: Masami Hiramatsu Date: Fri, 27 Aug 2010 20:39:12 +0900 Subject: tracing/kprobe: Fix handling of C-unlike argument names Check the argument name whether it is invalid (not C-like symbol name). This makes event format simple. Reported-by: Srikar Dronamraju Cc: Steven Rostedt Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Mathieu Desnoyers LKML-Reference: <20100827113912.22882.62313.stgit@ltc236.sdl.hitachi.co.jp> Signed-off-by: Masami Hiramatsu Signed-off-by: Arnaldo Carvalho de Melo diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index a39251e..544301d 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -514,8 +514,8 @@ static int kprobe_dispatcher(struct kprobe *kp, struct pt_regs *regs); static int kretprobe_dispatcher(struct kretprobe_instance *ri, struct pt_regs *regs); -/* Check the name is good for event/group */ -static int check_event_name(const char *name) +/* Check the name is good for event/group/fields */ +static int is_good_name(const char *name) { if (!isalpha(*name) && *name != '_') return 0; @@ -557,7 +557,7 @@ static struct trace_probe *alloc_trace_probe(const char *group, else tp->rp.kp.pre_handler = kprobe_dispatcher; - if (!event || !check_event_name(event)) { + if (!event || !is_good_name(event)) { ret = -EINVAL; goto error; } @@ -567,7 +567,7 @@ static struct trace_probe *alloc_trace_probe(const char *group, if (!tp->call.name) goto error; - if (!group || !check_event_name(group)) { + if (!group || !is_good_name(group)) { ret = -EINVAL; goto error; } @@ -883,7 +883,7 @@ static int create_trace_probe(int argc, char **argv) int i, ret = 0; int is_return = 0, is_delete = 0; char *symbol = NULL, *event = NULL, *group = NULL; - char *arg, *tmp; + char *arg; unsigned long offset = 0; void *addr = NULL; char buf[MAX_EVENT_NAME_LEN]; @@ -1012,9 +1012,13 @@ static int create_trace_probe(int argc, char **argv) ret = -ENOMEM; goto error; } - tmp = strchr(tp->args[i].name, ':'); - if (tmp) - *tmp = '_'; /* convert : to _ */ + + if (!is_good_name(tp->args[i].name)) { + pr_info("Invalid argument[%d] name: %s\n", + i, tp->args[i].name); + ret = -EINVAL; + goto error; + } if (conflict_field_name(tp->args[i].name, tp->args, i)) { pr_info("Argument[%d] name '%s' conflicts with " -- cgit v0.10.2 From 2be23c475af8ae4e25f8bab08d815b17593bd547 Mon Sep 17 00:00:00 2001 From: Russell King Date: Wed, 8 Sep 2010 16:27:56 +0100 Subject: ARM: Ensure PTE modifications via dma_alloc_coherent are visible Dave Hylands reports: | We've observed a problem with dma_alloc_writecombine when the system | is under heavy load (heavy bus traffic). We've managed to reduce the | problem to the following snippet, which is run from a kthread in a | continuous loop: | | void *virtAddr; | dma_addr_t physAddr; | unsigned int numBytes = 256; | | for (;;) { | virtAddr = dma_alloc_writecombine(NULL, | numBytes, &physAddr, GFP_KERNEL); | if (virtAddr == NULL) { | printk(KERN_ERR "Running out of memory\n"); | break; | } | | /* access DMA memory allocated */ | tmp = virtAddr; | *tmp = 0x77; | | /* free DMA memory */ | dma_free_writecombine(NULL, | numBytes, virtAddr, physAddr); | | ...sleep here... | } | | By itself, the code will run forever with no issues. However, as we | increase our bus traffic (typically using DMA) then the *tmp = 0x77 | line will eventually cause a page fault. If we add a small delay (a | few microseconds) before the *tmp = 0x77, then we don't see a page | fault, even under heavy load. A dsb() is required after modifying the PTE entries to ensure that they will always be visible. Add this dsb(). Reported-by: Dave Hylands Tested-by: Dave Hylands Signed-off-by: Russell King diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c index c704eed..4bc43e5 100644 --- a/arch/arm/mm/dma-mapping.c +++ b/arch/arm/mm/dma-mapping.c @@ -229,6 +229,8 @@ __dma_alloc_remap(struct page *page, size_t size, gfp_t gfp, pgprot_t prot) } } while (size -= PAGE_SIZE); + dsb(); + return (void *)c->vm_start; } return NULL; -- cgit v0.10.2 From 64289c8e6851bca0e589e064c9a5c9fbd6ae5dd4 Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Sat, 4 Sep 2010 10:34:29 +0000 Subject: gro: Re-fix different skb headrooms The patch: "gro: fix different skb headrooms" in its part: "2) allocate a minimal skb for head of frag_list" is buggy. The copied skb has p->data set at the ip header at the moment, and skb_gro_offset is the length of ip + tcp headers. So, after the change the length of mac header is skipped. Later skb_set_mac_header() sets it into the NET_SKB_PAD area (if it's long enough) and ip header is misaligned at NET_SKB_PAD + NET_IP_ALIGN offset. There is no reason to assume the original skb was wrongly allocated, so let's copy it as it was. bugzilla : https://bugzilla.kernel.org/show_bug.cgi?id=16626 fixes commit: 3d3be4333fdf6faa080947b331a6a19bce1a4f57 Reported-by: Plamen Petrov Signed-off-by: Jarek Poplawski CC: Eric Dumazet Acked-by: Eric Dumazet Tested-by: Plamen Petrov Signed-off-by: David S. Miller diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 26396ff..c83b421 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -2706,7 +2706,7 @@ int skb_gro_receive(struct sk_buff **head, struct sk_buff *skb) } else if (skb_gro_len(p) != pinfo->gso_size) return -E2BIG; - headroom = NET_SKB_PAD + NET_IP_ALIGN; + headroom = skb_headroom(p); nskb = alloc_skb(headroom + skb_gro_offset(p), GFP_ATOMIC); if (unlikely(!nskb)) return -ENOMEM; -- cgit v0.10.2 From 6523ce1525e88c598c75a1a6b8c4edddfa9defe8 Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Sun, 5 Sep 2010 18:02:29 +0000 Subject: ipvs: fix active FTP - Do not create expectation when forwarding the PORT command to avoid blocking the connection. The problem is that nf_conntrack_ftp.c:help() tries to create the same expectation later in POST_ROUTING and drops the packet with "dropping packet" message after failure in nf_ct_expect_related. - Change ip_vs_update_conntrack to alter the conntrack for related connections from real server. If we do not alter the reply in this direction the next packet from client sent to vport 20 comes as NEW connection. We alter it but may be some collision happens for both conntracks and the second conntrack gets destroyed immediately. The connection stucks too. Signed-off-by: Julian Anastasov Signed-off-by: Simon Horman Signed-off-by: David S. Miller diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h index a4747a0..f976885 100644 --- a/include/net/ip_vs.h +++ b/include/net/ip_vs.h @@ -955,6 +955,9 @@ static inline __wsum ip_vs_check_diff2(__be16 old, __be16 new, __wsum oldsum) return csum_partial(diff, sizeof(diff), oldsum); } +extern void ip_vs_update_conntrack(struct sk_buff *skb, struct ip_vs_conn *cp, + int outin); + #endif /* __KERNEL__ */ #endif /* _NET_IP_VS_H */ diff --git a/net/netfilter/ipvs/ip_vs_core.c b/net/netfilter/ipvs/ip_vs_core.c index 4f8ddba4..4c2f89d 100644 --- a/net/netfilter/ipvs/ip_vs_core.c +++ b/net/netfilter/ipvs/ip_vs_core.c @@ -924,6 +924,7 @@ handle_response(int af, struct sk_buff *skb, struct ip_vs_protocol *pp, ip_vs_out_stats(cp, skb); ip_vs_set_state(cp, IP_VS_DIR_OUTPUT, skb, pp); + ip_vs_update_conntrack(skb, cp, 0); ip_vs_conn_put(cp); skb->ipvs_property = 1; diff --git a/net/netfilter/ipvs/ip_vs_ftp.c b/net/netfilter/ipvs/ip_vs_ftp.c index 33b329b..7e9af5b 100644 --- a/net/netfilter/ipvs/ip_vs_ftp.c +++ b/net/netfilter/ipvs/ip_vs_ftp.c @@ -410,7 +410,6 @@ static int ip_vs_ftp_in(struct ip_vs_app *app, struct ip_vs_conn *cp, union nf_inet_addr to; __be16 port; struct ip_vs_conn *n_cp; - struct nf_conn *ct; #ifdef CONFIG_IP_VS_IPV6 /* This application helper doesn't work with IPv6 yet, @@ -497,11 +496,6 @@ static int ip_vs_ftp_in(struct ip_vs_app *app, struct ip_vs_conn *cp, ip_vs_control_add(n_cp, cp); } - ct = (struct nf_conn *)skb->nfct; - if (ct && ct != &nf_conntrack_untracked) - ip_vs_expect_related(skb, ct, n_cp, - IPPROTO_TCP, &n_cp->dport, 1); - /* * Move tunnel to listen state */ diff --git a/net/netfilter/ipvs/ip_vs_xmit.c b/net/netfilter/ipvs/ip_vs_xmit.c index 21e1a5e..49df6be 100644 --- a/net/netfilter/ipvs/ip_vs_xmit.c +++ b/net/netfilter/ipvs/ip_vs_xmit.c @@ -349,8 +349,8 @@ ip_vs_bypass_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, } #endif -static void -ip_vs_update_conntrack(struct sk_buff *skb, struct ip_vs_conn *cp) +void +ip_vs_update_conntrack(struct sk_buff *skb, struct ip_vs_conn *cp, int outin) { struct nf_conn *ct = (struct nf_conn *)skb->nfct; struct nf_conntrack_tuple new_tuple; @@ -365,11 +365,17 @@ ip_vs_update_conntrack(struct sk_buff *skb, struct ip_vs_conn *cp) * real-server we will see RIP->DIP. */ new_tuple = ct->tuplehash[IP_CT_DIR_REPLY].tuple; - new_tuple.src.u3 = cp->daddr; + if (outin) + new_tuple.src.u3 = cp->daddr; + else + new_tuple.dst.u3 = cp->vaddr; /* * This will also take care of UDP and other protocols. */ - new_tuple.src.u.tcp.port = cp->dport; + if (outin) + new_tuple.src.u.tcp.port = cp->dport; + else + new_tuple.dst.u.tcp.port = cp->vport; nf_conntrack_alter_reply(ct, &new_tuple); } @@ -428,7 +434,7 @@ ip_vs_nat_xmit(struct sk_buff *skb, struct ip_vs_conn *cp, IP_VS_DBG_PKT(10, pp, skb, 0, "After DNAT"); - ip_vs_update_conntrack(skb, cp); + ip_vs_update_conntrack(skb, cp, 1); /* FIXME: when application helper enlarges the packet and the length is larger than the MTU of outgoing device, there will be still @@ -506,7 +512,7 @@ ip_vs_nat_xmit_v6(struct sk_buff *skb, struct ip_vs_conn *cp, IP_VS_DBG_PKT(10, pp, skb, 0, "After DNAT"); - ip_vs_update_conntrack(skb, cp); + ip_vs_update_conntrack(skb, cp, 1); /* FIXME: when application helper enlarges the packet and the length is larger than the MTU of outgoing device, there will be still -- cgit v0.10.2 From 16518d5ada690643453eb0aef3cc7841d3623c2d Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Thu, 26 Aug 2010 14:31:30 +0300 Subject: KVM: x86 emulator: fix regression with cmpxchg8b on i386 hosts operand::val and operand::orig_val are 32-bit on i386, whereas cmpxchg8b operands are 64-bit. Fix by adding val64 and orig_val64 union members to struct operand, and using them where needed. Signed-off-by: Avi Kivity Signed-off-by: Marcelo Tosatti diff --git a/arch/x86/include/asm/kvm_emulate.h b/arch/x86/include/asm/kvm_emulate.h index 51cfd73..1f99ecf 100644 --- a/arch/x86/include/asm/kvm_emulate.h +++ b/arch/x86/include/asm/kvm_emulate.h @@ -152,9 +152,14 @@ struct x86_emulate_ops { struct operand { enum { OP_REG, OP_MEM, OP_IMM, OP_NONE } type; unsigned int bytes; - unsigned long orig_val, *ptr; + union { + unsigned long orig_val; + u64 orig_val64; + }; + unsigned long *ptr; union { unsigned long val; + u64 val64; char valptr[sizeof(unsigned long) + 2]; }; }; diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index b38bd8b..66ca98a 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -1870,17 +1870,16 @@ static inline int emulate_grp9(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) { struct decode_cache *c = &ctxt->decode; - u64 old = c->dst.orig_val; + u64 old = c->dst.orig_val64; if (((u32) (old >> 0) != (u32) c->regs[VCPU_REGS_RAX]) || ((u32) (old >> 32) != (u32) c->regs[VCPU_REGS_RDX])) { - c->regs[VCPU_REGS_RAX] = (u32) (old >> 0); c->regs[VCPU_REGS_RDX] = (u32) (old >> 32); ctxt->eflags &= ~EFLG_ZF; } else { - c->dst.val = ((u64)c->regs[VCPU_REGS_RCX] << 32) | - (u32) c->regs[VCPU_REGS_RBX]; + c->dst.val64 = ((u64)c->regs[VCPU_REGS_RCX] << 32) | + (u32) c->regs[VCPU_REGS_RBX]; ctxt->eflags |= EFLG_ZF; } @@ -2616,7 +2615,7 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) c->src.valptr, c->src.bytes); if (rc != X86EMUL_CONTINUE) goto done; - c->src.orig_val = c->src.val; + c->src.orig_val64 = c->src.val64; } if (c->src2.type == OP_MEM) { -- cgit v0.10.2 From ae0635b3580844e320a253a16985f4ce8b5bd0b6 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Tue, 27 Jul 2010 14:51:44 +0300 Subject: KVM: fix i8259 oops when no vcpus are online If there are no vcpus, found will be NULL. Check before doing anything with it. Signed-off-by: Avi Kivity diff --git a/arch/x86/kvm/i8259.c b/arch/x86/kvm/i8259.c index 8d10c06..4b7b73c 100644 --- a/arch/x86/kvm/i8259.c +++ b/arch/x86/kvm/i8259.c @@ -64,6 +64,9 @@ static void pic_unlock(struct kvm_pic *s) if (!found) found = s->kvm->bsp_vcpu; + if (!found) + return; + kvm_vcpu_kick(found); } } -- cgit v0.10.2 From eebb5f31b8d9a2620dcf32297096f8ce1240b179 Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Mon, 30 Aug 2010 12:49:13 +0300 Subject: KVM: i8259: fix migration Top of kvm_kpic_state structure should have the same memory layout as kvm_pic_state since it is copied by memcpy. Signed-off-by: Gleb Natapov Signed-off-by: Avi Kivity diff --git a/arch/x86/kvm/irq.h b/arch/x86/kvm/irq.h index ffed068..63c3145 100644 --- a/arch/x86/kvm/irq.h +++ b/arch/x86/kvm/irq.h @@ -43,7 +43,6 @@ struct kvm_kpic_state { u8 irr; /* interrupt request register */ u8 imr; /* interrupt mask register */ u8 isr; /* interrupt service register */ - u8 isr_ack; /* interrupt ack detection */ u8 priority_add; /* highest irq priority */ u8 irq_base; u8 read_reg_select; @@ -56,6 +55,7 @@ struct kvm_kpic_state { u8 init4; /* true if 4 byte init */ u8 elcr; /* PIIX edge/trigger selection */ u8 elcr_mask; + u8 isr_ack; /* interrupt ack detection */ struct kvm_pic *pics_state; }; -- cgit v0.10.2 From ee9c5cfad29c8a13199962614b9b16f1c4137ac9 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Tue, 7 Sep 2010 04:35:19 +0000 Subject: niu: Fix kernel buffer overflow for ETHTOOL_GRXCLSRLALL niu_get_ethtool_tcam_all() assumes that its output buffer is the right size, and warns before returning if it is not. However, the output buffer size is under user control and ETHTOOL_GRXCLSRLALL is an unprivileged ethtool command. Therefore this is at least a local denial-of-service vulnerability. Change it to check before writing each entry and to return an error if the buffer is already full. Compile-tested only. Signed-off-by: Ben Hutchings Signed-off-by: David S. Miller diff --git a/drivers/net/niu.c b/drivers/net/niu.c index b9b9508..340cbec 100644 --- a/drivers/net/niu.c +++ b/drivers/net/niu.c @@ -7272,32 +7272,28 @@ static int niu_get_ethtool_tcam_all(struct niu *np, struct niu_parent *parent = np->parent; struct niu_tcam_entry *tp; int i, idx, cnt; - u16 n_entries; unsigned long flags; - + int ret = 0; /* put the tcam size here */ nfc->data = tcam_get_size(np); niu_lock_parent(np, flags); - n_entries = nfc->rule_cnt; for (cnt = 0, i = 0; i < nfc->data; i++) { idx = tcam_get_index(np, i); tp = &parent->tcam[idx]; if (!tp->valid) continue; + if (cnt == nfc->rule_cnt) { + ret = -EMSGSIZE; + break; + } rule_locs[cnt] = i; cnt++; } niu_unlock_parent(np, flags); - if (n_entries != cnt) { - /* print warning, this should not happen */ - netdev_info(np->dev, "niu%d: In %s(): n_entries[%d] != cnt[%d]!!!\n", - np->parent->index, __func__, n_entries, cnt); - } - - return 0; + return ret; } static int niu_get_nfc(struct net_device *dev, struct ethtool_rxnfc *cmd, -- cgit v0.10.2 From f6b085b69d1cbbd62f49f34e71a3d58cb6d34b7e Mon Sep 17 00:00:00 2001 From: Jarek Poplawski Date: Tue, 7 Sep 2010 07:51:17 +0000 Subject: ipv4: Suppress lockdep-RCU false positive in FIB trie (3) Hi, Here is one more of these warnings and a patch below: Sep 5 23:52:33 del kernel: [46044.244833] =================================================== Sep 5 23:52:33 del kernel: [46044.269681] [ INFO: suspicious rcu_dereference_check() usage. ] Sep 5 23:52:33 del kernel: [46044.277000] --------------------------------------------------- Sep 5 23:52:33 del kernel: [46044.285185] net/ipv4/fib_trie.c:1756 invoked rcu_dereference_check() without protection! Sep 5 23:52:33 del kernel: [46044.293627] Sep 5 23:52:33 del kernel: [46044.293632] other info that might help us debug this: Sep 5 23:52:33 del kernel: [46044.293634] Sep 5 23:52:33 del kernel: [46044.325333] Sep 5 23:52:33 del kernel: [46044.325335] rcu_scheduler_active = 1, debug_locks = 0 Sep 5 23:52:33 del kernel: [46044.348013] 1 lock held by pppd/1717: Sep 5 23:52:33 del kernel: [46044.357548] #0: (rtnl_mutex){+.+.+.}, at: [] rtnl_lock+0xf/0x20 Sep 5 23:52:33 del kernel: [46044.367647] Sep 5 23:52:33 del kernel: [46044.367652] stack backtrace: Sep 5 23:52:33 del kernel: [46044.387429] Pid: 1717, comm: pppd Not tainted 2.6.35.4.4a #3 Sep 5 23:52:33 del kernel: [46044.398764] Call Trace: Sep 5 23:52:33 del kernel: [46044.409596] [] ? printk+0x18/0x1e Sep 5 23:52:33 del kernel: [46044.420761] [] lockdep_rcu_dereference+0xa9/0xb0 Sep 5 23:52:33 del kernel: [46044.432229] [] trie_firstleaf+0x65/0x70 Sep 5 23:52:33 del kernel: [46044.443941] [] fib_table_flush+0x14/0x170 Sep 5 23:52:33 del kernel: [46044.455823] [] ? local_bh_enable_ip+0x62/0xd0 Sep 5 23:52:33 del kernel: [46044.467995] [] ? _raw_spin_unlock_bh+0x2f/0x40 Sep 5 23:52:33 del kernel: [46044.480404] [] ? fib_sync_down_dev+0x120/0x180 Sep 5 23:52:33 del kernel: [46044.493025] [] fib_flush+0x2d/0x60 Sep 5 23:52:33 del kernel: [46044.505796] [] fib_disable_ip+0x25/0x50 Sep 5 23:52:33 del kernel: [46044.518772] [] fib_netdev_event+0x73/0xd0 Sep 5 23:52:33 del kernel: [46044.531918] [] notifier_call_chain+0x2d/0x70 Sep 5 23:52:33 del kernel: [46044.545358] [] raw_notifier_call_chain+0x1a/0x20 Sep 5 23:52:33 del kernel: [46044.559092] [] call_netdevice_notifiers+0x27/0x60 Sep 5 23:52:33 del kernel: [46044.573037] [] __dev_notify_flags+0x5c/0x80 Sep 5 23:52:33 del kernel: [46044.586489] [] dev_change_flags+0x37/0x60 Sep 5 23:52:33 del kernel: [46044.599394] [] devinet_ioctl+0x54d/0x630 Sep 5 23:52:33 del kernel: [46044.612277] [] inet_ioctl+0x97/0xc0 Sep 5 23:52:34 del kernel: [46044.625208] [] sock_ioctl+0x6f/0x270 Sep 5 23:52:34 del kernel: [46044.638046] [] ? handle_mm_fault+0x420/0x6c0 Sep 5 23:52:34 del kernel: [46044.650968] [] ? sock_ioctl+0x0/0x270 Sep 5 23:52:34 del kernel: [46044.663865] [] vfs_ioctl+0x28/0xa0 Sep 5 23:52:34 del kernel: [46044.676556] [] do_vfs_ioctl+0x6a/0x5c0 Sep 5 23:52:34 del kernel: [46044.688989] [] ? up_read+0x16/0x30 Sep 5 23:52:34 del kernel: [46044.701411] [] ? do_page_fault+0x1d6/0x3a0 Sep 5 23:52:34 del kernel: [46044.714223] [] ? fget_light+0xf8/0x2f0 Sep 5 23:52:34 del kernel: [46044.726601] [] ? sys_socketcall+0x208/0x2c0 Sep 5 23:52:34 del kernel: [46044.739140] [] sys_ioctl+0x63/0x70 Sep 5 23:52:34 del kernel: [46044.751967] [] syscall_call+0x7/0xb Sep 5 23:52:34 del kernel: [46044.764734] [] ? cookie_v6_check+0x3d0/0x630 --------------> This patch fixes the warning: =================================================== [ INFO: suspicious rcu_dereference_check() usage. ] --------------------------------------------------- net/ipv4/fib_trie.c:1756 invoked rcu_dereference_check() without protection! other info that might help us debug this: rcu_scheduler_active = 1, debug_locks = 0 1 lock held by pppd/1717: #0: (rtnl_mutex){+.+.+.}, at: [] rtnl_lock+0xf/0x20 stack backtrace: Pid: 1717, comm: pppd Not tainted 2.6.35.4a #3 Call Trace: [] ? printk+0x18/0x1e [] lockdep_rcu_dereference+0xa9/0xb0 [] trie_firstleaf+0x65/0x70 [] fib_table_flush+0x14/0x170 ... Allow trie_firstleaf() to be called either under rcu_read_lock() protection or with RTNL held. The same annotation is added to node_parent_rcu() to prevent a similar warning a bit later. Followup of commits 634a4b20 and 4eaa0e3c. Signed-off-by: Jarek Poplawski Signed-off-by: David S. Miller diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c index 79d057a..4a8e370 100644 --- a/net/ipv4/fib_trie.c +++ b/net/ipv4/fib_trie.c @@ -186,7 +186,9 @@ static inline struct tnode *node_parent_rcu(struct node *node) { struct tnode *ret = node_parent(node); - return rcu_dereference(ret); + return rcu_dereference_check(ret, + rcu_read_lock_held() || + lockdep_rtnl_is_held()); } /* Same as rcu_assign_pointer @@ -1753,7 +1755,9 @@ static struct leaf *leaf_walk_rcu(struct tnode *p, struct node *c) static struct leaf *trie_firstleaf(struct trie *t) { - struct tnode *n = (struct tnode *) rcu_dereference(t->trie); + struct tnode *n = (struct tnode *) rcu_dereference_check(t->trie, + rcu_read_lock_held() || + lockdep_rtnl_is_held()); if (!n) return NULL; -- cgit v0.10.2 From 70c9db0fdfd703781c3b8c2caf9287806f642e02 Mon Sep 17 00:00:00 2001 From: Chien Tung Date: Tue, 7 Sep 2010 16:31:20 +0000 Subject: RDMA/nes: Write correct register write to set TX pause param Setting TX pause param writes to the wrong register location causing the adapter to hang. Correct the define used to write the reigster. Addresses: https://bugs.openfabrics.org/show_bug.cgi?id=2116 Reported-by: Shiri Franchi Signed-off-by: Chien Tung Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/nes/nes_nic.c b/drivers/infiniband/hw/nes/nes_nic.c index 6dfdd49..10560c7 100644 --- a/drivers/infiniband/hw/nes/nes_nic.c +++ b/drivers/infiniband/hw/nes/nes_nic.c @@ -1446,14 +1446,14 @@ static int nes_netdev_set_pauseparam(struct net_device *netdev, NES_IDX_MAC_TX_CONFIG + (nesdev->mac_index*0x200)); u32temp |= NES_IDX_MAC_TX_CONFIG_ENABLE_PAUSE; nes_write_indexed(nesdev, - NES_IDX_MAC_TX_CONFIG_ENABLE_PAUSE + (nesdev->mac_index*0x200), u32temp); + NES_IDX_MAC_TX_CONFIG + (nesdev->mac_index*0x200), u32temp); nesdev->disable_tx_flow_control = 0; } else if ((et_pauseparam->tx_pause == 0) && (nesdev->disable_tx_flow_control == 0)) { u32temp = nes_read_indexed(nesdev, NES_IDX_MAC_TX_CONFIG + (nesdev->mac_index*0x200)); u32temp &= ~NES_IDX_MAC_TX_CONFIG_ENABLE_PAUSE; nes_write_indexed(nesdev, - NES_IDX_MAC_TX_CONFIG_ENABLE_PAUSE + (nesdev->mac_index*0x200), u32temp); + NES_IDX_MAC_TX_CONFIG + (nesdev->mac_index*0x200), u32temp); nesdev->disable_tx_flow_control = 1; } if ((et_pauseparam->rx_pause == 1) && (nesdev->disable_rx_flow_control == 1)) { -- cgit v0.10.2 From ae2688d59b5f861dc70a091d003773975d2ae7fb Mon Sep 17 00:00:00 2001 From: Jianzhao Wang Date: Wed, 8 Sep 2010 14:35:43 -0700 Subject: net: blackhole route should always be recalculated Blackhole routes are used when xfrm_lookup() returns -EREMOTE (error triggered by IKE for example), hence this kind of route is always temporary and so we should check if a better route exists for next packets. Bug has been introduced by commit d11a4dc18bf41719c9f0d7ed494d295dd2973b92. Signed-off-by: Jianzhao Wang Signed-off-by: Nicolas Dichtel Signed-off-by: David S. Miller diff --git a/net/ipv4/route.c b/net/ipv4/route.c index 3f56b6e..6298f75 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -2738,6 +2738,11 @@ slow_output: } EXPORT_SYMBOL_GPL(__ip_route_output_key); +static struct dst_entry *ipv4_blackhole_dst_check(struct dst_entry *dst, u32 cookie) +{ + return NULL; +} + static void ipv4_rt_blackhole_update_pmtu(struct dst_entry *dst, u32 mtu) { } @@ -2746,7 +2751,7 @@ static struct dst_ops ipv4_dst_blackhole_ops = { .family = AF_INET, .protocol = cpu_to_be16(ETH_P_IP), .destroy = ipv4_dst_destroy, - .check = ipv4_dst_check, + .check = ipv4_blackhole_dst_check, .update_pmtu = ipv4_rt_blackhole_update_pmtu, .entries = ATOMIC_INIT(0), }; -- cgit v0.10.2 From dae58728dc64e9ad71c40ac90b463bff6ecce271 Mon Sep 17 00:00:00 2001 From: Faisal Latif Date: Sat, 14 Aug 2010 21:04:56 +0000 Subject: RDMA/nes: Fix double CLOSE event indication crash During a stress testing in a large cluster, multiple close event are detected and BUG() is hit in the iWARP core. The cause is that the active node gave up while waiting for an MPA response from the peer and tried to close the connection by sending RST. The passive node driver receives the RST but is waiting for MPA response from the user. When the MPA accept is received, the driver offloads the connection and sends a CLOSE event. The driver gets an AE indicating RESET received and also sends a CLOSE event, hitting a BUG(). Fix this by correcting RESET handling and sending CLOSE events. Signed-off-by: Faisal Latif Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/nes/nes_cm.c b/drivers/infiniband/hw/nes/nes_cm.c index 443cea5..61e0efd 100644 --- a/drivers/infiniband/hw/nes/nes_cm.c +++ b/drivers/infiniband/hw/nes/nes_cm.c @@ -502,7 +502,9 @@ int schedule_nes_timer(struct nes_cm_node *cm_node, struct sk_buff *skb, static void nes_retrans_expired(struct nes_cm_node *cm_node) { struct iw_cm_id *cm_id = cm_node->cm_id; - switch (cm_node->state) { + enum nes_cm_node_state state = cm_node->state; + cm_node->state = NES_CM_STATE_CLOSED; + switch (state) { case NES_CM_STATE_SYN_RCVD: case NES_CM_STATE_CLOSING: rem_ref_cm_node(cm_node->cm_core, cm_node); @@ -511,7 +513,6 @@ static void nes_retrans_expired(struct nes_cm_node *cm_node) case NES_CM_STATE_FIN_WAIT1: if (cm_node->cm_id) cm_id->rem_ref(cm_id); - cm_node->state = NES_CM_STATE_CLOSED; send_reset(cm_node, NULL); break; default: @@ -1439,9 +1440,6 @@ static void handle_rst_pkt(struct nes_cm_node *cm_node, struct sk_buff *skb, break; case NES_CM_STATE_MPAREQ_RCVD: passive_state = atomic_add_return(1, &cm_node->passive_state); - if (passive_state == NES_SEND_RESET_EVENT) - create_event(cm_node, NES_CM_EVENT_RESET); - cm_node->state = NES_CM_STATE_CLOSED; dev_kfree_skb_any(skb); break; case NES_CM_STATE_ESTABLISHED: @@ -1456,6 +1454,7 @@ static void handle_rst_pkt(struct nes_cm_node *cm_node, struct sk_buff *skb, case NES_CM_STATE_CLOSED: drop_packet(skb); break; + case NES_CM_STATE_FIN_WAIT2: case NES_CM_STATE_FIN_WAIT1: case NES_CM_STATE_LAST_ACK: cm_node->cm_id->rem_ref(cm_node->cm_id); @@ -2777,6 +2776,12 @@ int nes_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) return -EINVAL; } + passive_state = atomic_add_return(1, &cm_node->passive_state); + if (passive_state == NES_SEND_RESET_EVENT) { + rem_ref_cm_node(cm_node->cm_core, cm_node); + return -ECONNRESET; + } + /* associate the node with the QP */ nesqp->cm_node = (void *)cm_node; cm_node->nesqp = nesqp; @@ -2979,9 +2984,6 @@ int nes_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param) printk(KERN_ERR "%s[%u] OFA CM event_handler returned, " "ret=%d\n", __func__, __LINE__, ret); - passive_state = atomic_add_return(1, &cm_node->passive_state); - if (passive_state == NES_SEND_RESET_EVENT) - create_event(cm_node, NES_CM_EVENT_RESET); return 0; } -- cgit v0.10.2 From 67d70721151726286763209ecadc3fce86abfdce Mon Sep 17 00:00:00 2001 From: Faisal Latif Date: Sat, 14 Aug 2010 21:05:04 +0000 Subject: RDMA/nes: Change state to closing after FIN When the driver receives an AE for FIN received, it closes the connection without changing the state of the connection in the hardware to closing. By changing the state to closing, hardware will do a normal close sequence. Signed-off-by: Faisal Latif Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/nes/nes_hw.c b/drivers/infiniband/hw/nes/nes_hw.c index f8233c8..ba93a8b 100644 --- a/drivers/infiniband/hw/nes/nes_hw.c +++ b/drivers/infiniband/hw/nes/nes_hw.c @@ -3468,6 +3468,18 @@ static void nes_process_iwarp_aeqe(struct nes_device *nesdev, return; /* Ignore it, wait for close complete */ if (atomic_inc_return(&nesqp->close_timer_started) == 1) { + if ((tcp_state == NES_AEQE_TCP_STATE_CLOSE_WAIT) && + (nesqp->ibqp_state == IB_QPS_RTS)) { + spin_lock_irqsave(&nesqp->lock, flags); + nesqp->hw_iwarp_state = iwarp_state; + nesqp->hw_tcp_state = tcp_state; + nesqp->last_aeq = async_event_id; + next_iwarp_state = NES_CQP_QP_IWARP_STATE_CLOSING; + nesqp->hw_iwarp_state = NES_AEQE_IWARP_STATE_CLOSING; + spin_unlock_irqrestore(&nesqp->lock, flags); + nes_hw_modify_qp(nesdev, nesqp, next_iwarp_state, 0, 0); + nes_cm_disconn(nesqp); + } nesqp->cm_id->add_ref(nesqp->cm_id); schedule_nes_timer(nesqp->cm_node, (struct sk_buff *)nesqp, NES_TIMER_TYPE_CLOSE, 1, 0); @@ -3477,7 +3489,6 @@ static void nes_process_iwarp_aeqe(struct nes_device *nesdev, nesqp->hwqp.qp_id, atomic_read(&nesqp->refcount), async_event_id, nesqp->last_aeq, tcp_state); } - break; case NES_AEQE_AEID_LLP_CLOSE_COMPLETE: if (nesqp->term_flags) { -- cgit v0.10.2 From 29da03b9d1c6f24548d98cebda1e15a25d80ee1b Mon Sep 17 00:00:00 2001 From: Faisal Latif Date: Wed, 1 Sep 2010 15:43:11 +0000 Subject: RDMA/nes: Fix hang with modified FIN handling on A0 cards Changing state to CLOSING when FIN is received causes A0 cards to hang. Fix this by checking for A0 cards in FIN handling. Signed-off-by: Faisal Latif Signed-off-by: Roland Dreier diff --git a/drivers/infiniband/hw/nes/nes_hw.c b/drivers/infiniband/hw/nes/nes_hw.c index ba93a8b..1980a46 100644 --- a/drivers/infiniband/hw/nes/nes_hw.c +++ b/drivers/infiniband/hw/nes/nes_hw.c @@ -3469,7 +3469,8 @@ static void nes_process_iwarp_aeqe(struct nes_device *nesdev, if (atomic_inc_return(&nesqp->close_timer_started) == 1) { if ((tcp_state == NES_AEQE_TCP_STATE_CLOSE_WAIT) && - (nesqp->ibqp_state == IB_QPS_RTS)) { + (nesqp->ibqp_state == IB_QPS_RTS) && + ((nesadapter->eeprom_version >> 16) != NES_A0)) { spin_lock_irqsave(&nesqp->lock, flags); nesqp->hw_iwarp_state = iwarp_state; nesqp->hw_tcp_state = tcp_state; diff --git a/drivers/infiniband/hw/nes/nes_hw.h b/drivers/infiniband/hw/nes/nes_hw.h index aa9183d..1204c34 100644 --- a/drivers/infiniband/hw/nes/nes_hw.h +++ b/drivers/infiniband/hw/nes/nes_hw.h @@ -45,6 +45,7 @@ #define NES_PHY_TYPE_KR 9 #define NES_MULTICAST_PF_MAX 8 +#define NES_A0 3 enum pci_regs { NES_INT_STAT = 0x0000, -- cgit v0.10.2 From 152e1d592071c8b312bb898bc1118b64e4aea535 Mon Sep 17 00:00:00 2001 From: Colin Cross Date: Fri, 3 Sep 2010 01:24:07 +0200 Subject: PM: Prevent waiting forever on asynchronous resume after failing suspend During suspend, the power.completion is expected to be set when a device has not yet started suspending. Set it on init to fix a corner case where a device is resumed when its parent has never suspended. Consider three drivers, A, B, and C. The parent of A is C, and C has async_suspend set. On boot, C->power.completion is initialized to 0. During the first suspend: suspend_devices_and_enter(...) dpm_resume(...) device_suspend(A) device_suspend(B) returns error, aborts suspend dpm_resume_end(...) dpm_resume(...) device_resume(A) dpm_wait(A->parent == C) wait_for_completion(C->power.completion) The wait_for_completion will never complete, because complete_all(C->power.completion) will only be called from device_suspend(C) or device_resume(C), neither of which is called if suspend is aborted before C. After a successful suspend->resume cycle, where B doesn't abort suspend, C->power.completion is left in the completed state by the call to device_resume(C), and the same call path will work if B aborts suspend. Signed-off-by: Colin Cross Signed-off-by: Rafael J. Wysocki diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 5419a49..276d5a7 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -59,6 +59,7 @@ void device_pm_init(struct device *dev) { dev->power.status = DPM_ON; init_completion(&dev->power.completion); + complete_all(&dev->power.completion); dev->power.wakeup_count = 0; pm_runtime_init(dev); } -- cgit v0.10.2 From b19f8200392db88d529576a4f8178d08b70047be Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 8 Sep 2010 20:57:59 -0700 Subject: sparc: Kill all BKL usage. They were all bogus artifacts and completely unnecessary. Signed-off-by: David S. Miller diff --git a/arch/sparc/kernel/sys_sparc_32.c b/arch/sparc/kernel/sys_sparc_32.c index ee995b7..de45de4 100644 --- a/arch/sparc/kernel/sys_sparc_32.c +++ b/arch/sparc/kernel/sys_sparc_32.c @@ -166,7 +166,6 @@ sparc_breakpoint (struct pt_regs *regs) { siginfo_t info; - lock_kernel(); #ifdef DEBUG_SPARC_BREAKPOINT printk ("TRAP: Entering kernel PC=%x, nPC=%x\n", regs->pc, regs->npc); #endif @@ -180,7 +179,6 @@ sparc_breakpoint (struct pt_regs *regs) #ifdef DEBUG_SPARC_BREAKPOINT printk ("TRAP: Returning to space: PC=%x nPC=%x\n", regs->pc, regs->npc); #endif - unlock_kernel(); } asmlinkage int diff --git a/arch/sparc/kernel/unaligned_32.c b/arch/sparc/kernel/unaligned_32.c index f8514e2..12b9f35 100644 --- a/arch/sparc/kernel/unaligned_32.c +++ b/arch/sparc/kernel/unaligned_32.c @@ -323,7 +323,6 @@ asmlinkage void user_unaligned_trap(struct pt_regs *regs, unsigned int insn) { enum direction dir; - lock_kernel(); if(!(current->thread.flags & SPARC_FLAG_UNALIGNED) || (((insn >> 30) & 3) != 3)) goto kill_user; @@ -377,5 +376,5 @@ asmlinkage void user_unaligned_trap(struct pt_regs *regs, unsigned int insn) kill_user: user_mna_trap_fault(regs, insn); out: - unlock_kernel(); + ; } diff --git a/arch/sparc/kernel/windows.c b/arch/sparc/kernel/windows.c index f24d298..b351770 100644 --- a/arch/sparc/kernel/windows.c +++ b/arch/sparc/kernel/windows.c @@ -112,7 +112,6 @@ void try_to_clear_window_buffer(struct pt_regs *regs, int who) struct thread_info *tp = current_thread_info(); int window; - lock_kernel(); flush_user_windows(); for(window = 0; window < tp->w_saved; window++) { unsigned long sp = tp->rwbuf_stkptrs[window]; @@ -123,5 +122,4 @@ void try_to_clear_window_buffer(struct pt_regs *regs, int who) do_exit(SIGILL); } tp->w_saved = 0; - unlock_kernel(); } -- cgit v0.10.2 From 719f835853a92f6090258114a72ffe41f09155cd Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 8 Sep 2010 05:08:44 +0000 Subject: udp: add rehash on connect() commit 30fff923 introduced in linux-2.6.33 (udp: bind() optimisation) added a secondary hash on UDP, hashed on (local addr, local port). Problem is that following sequence : fd = socket(...) connect(fd, &remote, ...) not only selects remote end point (address and port), but also sets local address, while UDP stack stored in secondary hash table the socket while its local address was INADDR_ANY (or ipv6 equivalent) Sequence is : - autobind() : choose a random local port, insert socket in hash tables [while local address is INADDR_ANY] - connect() : set remote address and port, change local address to IP given by a route lookup. When an incoming UDP frame comes, if more than 10 sockets are found in primary hash table, we switch to secondary table, and fail to find socket because its local address changed. One solution to this problem is to rehash datagram socket if needed. We add a new rehash(struct socket *) method in "struct proto", and implement this method for UDP v4 & v6, using a common helper. This rehashing only takes care of secondary hash table, since primary hash (based on local port only) is not changed. Reported-by: Krzysztof Piotr Oledzki Signed-off-by: Eric Dumazet Tested-by: Krzysztof Piotr Oledzki Signed-off-by: David S. Miller diff --git a/include/net/sock.h b/include/net/sock.h index ac53bfb..adab9dc 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -752,6 +752,7 @@ struct proto { /* Keeping track of sk's, looking them up, and port selection methods. */ void (*hash)(struct sock *sk); void (*unhash)(struct sock *sk); + void (*rehash)(struct sock *sk); int (*get_port)(struct sock *sk, unsigned short snum); /* Keeping track of sockets in use */ diff --git a/include/net/udp.h b/include/net/udp.h index 7abdf30..a184d34 100644 --- a/include/net/udp.h +++ b/include/net/udp.h @@ -151,6 +151,7 @@ static inline void udp_lib_hash(struct sock *sk) } extern void udp_lib_unhash(struct sock *sk); +extern void udp_lib_rehash(struct sock *sk, u16 new_hash); static inline void udp_lib_close(struct sock *sk, long timeout) { diff --git a/net/ipv4/datagram.c b/net/ipv4/datagram.c index f055094..721a8a3 100644 --- a/net/ipv4/datagram.c +++ b/net/ipv4/datagram.c @@ -62,8 +62,11 @@ int ip4_datagram_connect(struct sock *sk, struct sockaddr *uaddr, int addr_len) } if (!inet->inet_saddr) inet->inet_saddr = rt->rt_src; /* Update source address */ - if (!inet->inet_rcv_saddr) + if (!inet->inet_rcv_saddr) { inet->inet_rcv_saddr = rt->rt_src; + if (sk->sk_prot->rehash) + sk->sk_prot->rehash(sk); + } inet->inet_daddr = rt->rt_dst; inet->inet_dport = usin->sin_port; sk->sk_state = TCP_ESTABLISHED; diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 32e0bef..fb23c2e 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -1260,6 +1260,49 @@ void udp_lib_unhash(struct sock *sk) } EXPORT_SYMBOL(udp_lib_unhash); +/* + * inet_rcv_saddr was changed, we must rehash secondary hash + */ +void udp_lib_rehash(struct sock *sk, u16 newhash) +{ + if (sk_hashed(sk)) { + struct udp_table *udptable = sk->sk_prot->h.udp_table; + struct udp_hslot *hslot, *hslot2, *nhslot2; + + hslot2 = udp_hashslot2(udptable, udp_sk(sk)->udp_portaddr_hash); + nhslot2 = udp_hashslot2(udptable, newhash); + udp_sk(sk)->udp_portaddr_hash = newhash; + if (hslot2 != nhslot2) { + hslot = udp_hashslot(udptable, sock_net(sk), + udp_sk(sk)->udp_port_hash); + /* we must lock primary chain too */ + spin_lock_bh(&hslot->lock); + + spin_lock(&hslot2->lock); + hlist_nulls_del_init_rcu(&udp_sk(sk)->udp_portaddr_node); + hslot2->count--; + spin_unlock(&hslot2->lock); + + spin_lock(&nhslot2->lock); + hlist_nulls_add_head_rcu(&udp_sk(sk)->udp_portaddr_node, + &nhslot2->head); + nhslot2->count++; + spin_unlock(&nhslot2->lock); + + spin_unlock_bh(&hslot->lock); + } + } +} +EXPORT_SYMBOL(udp_lib_rehash); + +static void udp_v4_rehash(struct sock *sk) +{ + u16 new_hash = udp4_portaddr_hash(sock_net(sk), + inet_sk(sk)->inet_rcv_saddr, + inet_sk(sk)->inet_num); + udp_lib_rehash(sk, new_hash); +} + static int __udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) { int rc; @@ -1843,6 +1886,7 @@ struct proto udp_prot = { .backlog_rcv = __udp_queue_rcv_skb, .hash = udp_lib_hash, .unhash = udp_lib_unhash, + .rehash = udp_v4_rehash, .get_port = udp_v4_get_port, .memory_allocated = &udp_memory_allocated, .sysctl_mem = sysctl_udp_mem, diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c index 7d929a2..ef371aa 100644 --- a/net/ipv6/datagram.c +++ b/net/ipv6/datagram.c @@ -105,9 +105,12 @@ ipv4_connected: if (ipv6_addr_any(&np->saddr)) ipv6_addr_set_v4mapped(inet->inet_saddr, &np->saddr); - if (ipv6_addr_any(&np->rcv_saddr)) + if (ipv6_addr_any(&np->rcv_saddr)) { ipv6_addr_set_v4mapped(inet->inet_rcv_saddr, &np->rcv_saddr); + if (sk->sk_prot->rehash) + sk->sk_prot->rehash(sk); + } goto out; } @@ -181,6 +184,8 @@ ipv4_connected: if (ipv6_addr_any(&np->rcv_saddr)) { ipv6_addr_copy(&np->rcv_saddr, &fl.fl6_src); inet->inet_rcv_saddr = LOOPBACK4_IPV6; + if (sk->sk_prot->rehash) + sk->sk_prot->rehash(sk); } ip6_dst_store(sk, dst, diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 1dd1aff..5acb356 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -111,6 +111,15 @@ int udp_v6_get_port(struct sock *sk, unsigned short snum) return udp_lib_get_port(sk, snum, ipv6_rcv_saddr_equal, hash2_nulladdr); } +static void udp_v6_rehash(struct sock *sk) +{ + u16 new_hash = udp6_portaddr_hash(sock_net(sk), + &inet6_sk(sk)->rcv_saddr, + inet_sk(sk)->inet_num); + + udp_lib_rehash(sk, new_hash); +} + static inline int compute_score(struct sock *sk, struct net *net, unsigned short hnum, struct in6_addr *saddr, __be16 sport, @@ -1447,6 +1456,7 @@ struct proto udpv6_prot = { .backlog_rcv = udpv6_queue_rcv_skb, .hash = udp_lib_hash, .unhash = udp_lib_unhash, + .rehash = udp_v6_rehash, .get_port = udp_v6_get_port, .memory_allocated = &udp_memory_allocated, .sysctl_mem = sysctl_udp_mem, -- cgit v0.10.2 From 972c40b5bee429c84ba727f8ac0a08292bc5dc3d Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 8 Sep 2010 13:26:55 +0000 Subject: KS8851: Correct RX packet allocation Use netdev_alloc_skb_ip_align() helper and do correct allocation Tested-by: Abraham Arce Signed-off-by: Abraham Arce Signed-off-by: David S. Miller diff --git a/drivers/net/ks8851.c b/drivers/net/ks8851.c index b4fb07a..51919fc 100644 --- a/drivers/net/ks8851.c +++ b/drivers/net/ks8851.c @@ -503,30 +503,33 @@ static void ks8851_rx_pkts(struct ks8851_net *ks) ks8851_wrreg16(ks, KS_RXQCR, ks->rc_rxqcr | RXQCR_SDA | RXQCR_ADRFE); - if (rxlen > 0) { - skb = netdev_alloc_skb(ks->netdev, rxlen + 2 + 8); - if (!skb) { - /* todo - dump frame and move on */ - } + if (rxlen > 4) { + unsigned int rxalign; + + rxlen -= 4; + rxalign = ALIGN(rxlen, 4); + skb = netdev_alloc_skb_ip_align(ks->netdev, rxalign); + if (skb) { - /* two bytes to ensure ip is aligned, and four bytes - * for the status header and 4 bytes of garbage */ - skb_reserve(skb, 2 + 4 + 4); + /* 4 bytes of status header + 4 bytes of + * garbage: we put them before ethernet + * header, so that they are copied, + * but ignored. + */ - rxpkt = skb_put(skb, rxlen - 4) - 8; + rxpkt = skb_put(skb, rxlen) - 8; - /* align the packet length to 4 bytes, and add 4 bytes - * as we're getting the rx status header as well */ - ks8851_rdfifo(ks, rxpkt, ALIGN(rxlen, 4) + 8); + ks8851_rdfifo(ks, rxpkt, rxalign + 8); - if (netif_msg_pktdata(ks)) - ks8851_dbg_dumpkkt(ks, rxpkt); + if (netif_msg_pktdata(ks)) + ks8851_dbg_dumpkkt(ks, rxpkt); - skb->protocol = eth_type_trans(skb, ks->netdev); - netif_rx(skb); + skb->protocol = eth_type_trans(skb, ks->netdev); + netif_rx(skb); - ks->netdev->stats.rx_packets++; - ks->netdev->stats.rx_bytes += rxlen - 4; + ks->netdev->stats.rx_packets++; + ks->netdev->stats.rx_bytes += rxlen; + } } ks8851_wrreg16(ks, KS_RXQCR, ks->rc_rxqcr); -- cgit v0.10.2 From 7839d956fc6aecbb66d645b4050e8e88e2e821cd Mon Sep 17 00:00:00 2001 From: Chris Wilson Date: Thu, 9 Sep 2010 00:02:03 +0100 Subject: drm/i915: Double check that the wait_request is not pending before warning If we are busy, then we may have woken up the wait_request handler but not yet serviced it before the hang check fires. So in hang check, double check that the i915_gem_do_wait_request() is still pending the wake-up before declaring all hope lost. Fixes regression with e78d73b16bcde921c9cf458d2e4de8e4fc2518f3. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=30073 Reported-and-tested-by: Sitsofe Wheeler Signed-off-by: Chris Wilson diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 59457e8..744225e 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -1350,17 +1350,25 @@ void i915_hangcheck_elapsed(unsigned long data) i915_seqno_passed(i915_get_gem_seqno(dev, &dev_priv->render_ring), i915_get_tail_request(dev)->seqno)) { + bool missed_wakeup = false; + dev_priv->hangcheck_count = 0; /* Issue a wake-up to catch stuck h/w. */ - if (dev_priv->render_ring.waiting_gem_seqno | - dev_priv->bsd_ring.waiting_gem_seqno) { - DRM_ERROR("Hangcheck timer elapsed... GPU idle, missed IRQ.\n"); - if (dev_priv->render_ring.waiting_gem_seqno) - DRM_WAKEUP(&dev_priv->render_ring.irq_queue); - if (dev_priv->bsd_ring.waiting_gem_seqno) - DRM_WAKEUP(&dev_priv->bsd_ring.irq_queue); + if (dev_priv->render_ring.waiting_gem_seqno && + waitqueue_active(&dev_priv->render_ring.irq_queue)) { + DRM_WAKEUP(&dev_priv->render_ring.irq_queue); + missed_wakeup = true; + } + + if (dev_priv->bsd_ring.waiting_gem_seqno && + waitqueue_active(&dev_priv->bsd_ring.irq_queue)) { + DRM_WAKEUP(&dev_priv->bsd_ring.irq_queue); + missed_wakeup = true; } + + if (missed_wakeup) + DRM_ERROR("Hangcheck timer elapsed... GPU idle, missed IRQ.\n"); return; } -- cgit v0.10.2 From 3a5c19c23db65a554f2e4f5df5f307c668277056 Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Mon, 16 Aug 2010 10:06:26 -0500 Subject: [SCSI] fix use-after-free in scsi_init_io() we're using a pointer through a freed command to reset the request, which has shown up as an oops with slab poisoning: Reported-by: Tejun Heo Reported-by: Alexey Dobriyan Signed-off-by: James Bottomley diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index 9ade720..ee02d38 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -1011,8 +1011,8 @@ int scsi_init_io(struct scsi_cmnd *cmd, gfp_t gfp_mask) err_exit: scsi_release_buffers(cmd); - scsi_put_command(cmd); cmd->request->special = NULL; + scsi_put_command(cmd); return error; } EXPORT_SYMBOL(scsi_init_io); -- cgit v0.10.2 From 39aa3cb3e8250db9188a6f1e3fb62ffa1a717678 Mon Sep 17 00:00:00 2001 From: Stefan Bader Date: Tue, 31 Aug 2010 15:52:27 +0200 Subject: mm: Move vma_stack_continue into mm.h So it can be used by all that need to check for that. Signed-off-by: Stefan Bader Signed-off-by: Linus Torvalds diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 439fc1f..271afc4 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -224,7 +224,8 @@ static void show_map_vma(struct seq_file *m, struct vm_area_struct *vma) /* We don't show the stack guard page in /proc/maps */ start = vma->vm_start; if (vma->vm_flags & VM_GROWSDOWN) - start += PAGE_SIZE; + if (!vma_stack_continue(vma->vm_prev, vma->vm_start)) + start += PAGE_SIZE; seq_printf(m, "%08lx-%08lx %c%c%c%c %08llx %02x:%02x %lu %n", start, diff --git a/include/linux/mm.h b/include/linux/mm.h index e6b1210..74949fb 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -864,6 +864,12 @@ int set_page_dirty(struct page *page); int set_page_dirty_lock(struct page *page); int clear_page_dirty_for_io(struct page *page); +/* Is the vma a continuation of the stack vma above it? */ +static inline int vma_stack_continue(struct vm_area_struct *vma, unsigned long addr) +{ + return vma && (vma->vm_end == addr) && (vma->vm_flags & VM_GROWSDOWN); +} + extern unsigned long move_page_tables(struct vm_area_struct *vma, unsigned long old_addr, struct vm_area_struct *new_vma, unsigned long new_addr, unsigned long len); diff --git a/mm/mlock.c b/mm/mlock.c index cbae7c5..b70919c 100644 --- a/mm/mlock.c +++ b/mm/mlock.c @@ -135,12 +135,6 @@ void munlock_vma_page(struct page *page) } } -/* Is the vma a continuation of the stack vma above it? */ -static inline int vma_stack_continue(struct vm_area_struct *vma, unsigned long addr) -{ - return vma && (vma->vm_end == addr) && (vma->vm_flags & VM_GROWSDOWN); -} - static inline int stack_guard_page(struct vm_area_struct *vma, unsigned long addr) { return (vma->vm_flags & VM_GROWSDOWN) && -- cgit v0.10.2 From a73f8844e1fc54c3762555c1cf1f71774142ca91 Mon Sep 17 00:00:00 2001 From: Jonathan Corbet Date: Wed, 8 Sep 2010 16:54:54 -0600 Subject: lglock: make lg_lock_global() actually lock globally lg_lock_global() currently only acquires spinlocks for online CPUs, but it's meant to lock all possible CPUs. Lglock-protected resources may be associated with removed CPUs - and, indeed, that could happen with the per-superblock open files lists. At Nick's suggestion, change for_each_online_cpu() to for_each_possible_cpu() to protect accesses to those resources. Cc: Al Viro Acked-by: Nick Piggin Signed-off-by: Jonathan Corbet Signed-off-by: Linus Torvalds diff --git a/include/linux/lglock.h b/include/linux/lglock.h index b288cb7..f549056 100644 --- a/include/linux/lglock.h +++ b/include/linux/lglock.h @@ -150,7 +150,7 @@ int i; \ preempt_disable(); \ rwlock_acquire(&name##_lock_dep_map, 0, 0, _RET_IP_); \ - for_each_online_cpu(i) { \ + for_each_possible_cpu(i) { \ arch_spinlock_t *lock; \ lock = &per_cpu(name##_lock, i); \ arch_spin_lock(lock); \ @@ -161,7 +161,7 @@ void name##_global_unlock(void) { \ int i; \ rwlock_release(&name##_lock_dep_map, 1, _RET_IP_); \ - for_each_online_cpu(i) { \ + for_each_possible_cpu(i) { \ arch_spinlock_t *lock; \ lock = &per_cpu(name##_lock, i); \ arch_spin_unlock(lock); \ -- cgit v0.10.2 From da908f2fb4e783c2a4de751fb90f11a0dd041161 Mon Sep 17 00:00:00 2001 From: Zachary Amsden Date: Thu, 19 Aug 2010 22:07:27 -1000 Subject: KVM: x86: Perform hardware_enable in CPU_STARTING callback The CPU_STARTING callback was added upstream with the intention of being used for KVM, specifically for the hardware enablement that must be done before we can run in hardware virt. It had bugs on the x86_64 architecture at the time, where it was called after CPU_ONLINE. The arches have since merged and the bug is gone. It might be noted other features should probably start making use of this callback; microcode updates in particular which might be fixing important erratums would be best applied before beginning to run user tasks. Signed-off-by: Zachary Amsden Signed-off-by: Marcelo Tosatti diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index b78b794..d4853a5 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -1958,10 +1958,10 @@ static int kvm_cpu_hotplug(struct notifier_block *notifier, unsigned long val, cpu); hardware_disable(NULL); break; - case CPU_ONLINE: + case CPU_STARTING: printk(KERN_INFO "kvm: enabling virtualization on CPU%d\n", cpu); - smp_call_function_single(cpu, hardware_enable, NULL, 1); + hardware_enable(NULL); break; } return NOTIFY_OK; @@ -2096,7 +2096,6 @@ int kvm_io_bus_unregister_dev(struct kvm *kvm, enum kvm_bus bus_idx, static struct notifier_block kvm_cpu_notifier = { .notifier_call = kvm_cpu_hotplug, - .priority = 20, /* must be > scheduler priority */ }; static int vm_stat_get(void *_offset, u64 *val) -- cgit v0.10.2 From 5afddee415c829704d3a91031ae634008bc332e9 Mon Sep 17 00:00:00 2001 From: Nicolas Ferre Date: Thu, 9 Sep 2010 19:58:23 +0200 Subject: AT91: clock: peripheral clocks can have other parent than mck While registering clock allow to set parent clock other than mck. It is useful for clocks than can be seen as child clock of a peripheral. Signed-off-by: Nicolas Ferre Acked-by: Jean-Christophe PLAGNIOL-VILLARD Acked-by: Andrew Victor diff --git a/arch/arm/mach-at91/clock.c b/arch/arm/mach-at91/clock.c index 7f7da43..7525cee 100644 --- a/arch/arm/mach-at91/clock.c +++ b/arch/arm/mach-at91/clock.c @@ -501,7 +501,8 @@ postcore_initcall(at91_clk_debugfs_init); int __init clk_register(struct clk *clk) { if (clk_is_peripheral(clk)) { - clk->parent = &mck; + if (!clk->parent) + clk->parent = &mck; clk->mode = pmc_periph_mode; list_add_tail(&clk->node, &clocks); } -- cgit v0.10.2 From ab64511cbbd03196d84bcc32c6e7b9d46543df7b Mon Sep 17 00:00:00 2001 From: Fabian Godehardt Date: Fri, 3 Sep 2010 13:31:33 +0100 Subject: AT91: SAM9G45 - add a separate clock entry for every single TC block Without this patch you will not be able to register the first block because of the second association call on at91_add_device_tc(). Signed-off-by: Fabian Godehardt [nicolas.ferre@atmel.com: change tcb1_clk to fake child clock of tcb0_clk] Acked-by: Jean-Christophe PLAGNIOL-VILLARD Acked-by: Andrew Victor Signed-off-by: Nicolas Ferre diff --git a/arch/arm/mach-at91/at91sam9g45.c b/arch/arm/mach-at91/at91sam9g45.c index 753c0d3..c67b47f 100644 --- a/arch/arm/mach-at91/at91sam9g45.c +++ b/arch/arm/mach-at91/at91sam9g45.c @@ -121,8 +121,8 @@ static struct clk ssc1_clk = { .pmc_mask = 1 << AT91SAM9G45_ID_SSC1, .type = CLK_TYPE_PERIPHERAL, }; -static struct clk tcb_clk = { - .name = "tcb_clk", +static struct clk tcb0_clk = { + .name = "tcb0_clk", .pmc_mask = 1 << AT91SAM9G45_ID_TCB, .type = CLK_TYPE_PERIPHERAL, }; @@ -192,6 +192,14 @@ static struct clk ohci_clk = { .parent = &uhphs_clk, }; +/* One additional fake clock for second TC block */ +static struct clk tcb1_clk = { + .name = "tcb1_clk", + .pmc_mask = 0, + .type = CLK_TYPE_PERIPHERAL, + .parent = &tcb0_clk, +}; + static struct clk *periph_clocks[] __initdata = { &pioA_clk, &pioB_clk, @@ -208,7 +216,7 @@ static struct clk *periph_clocks[] __initdata = { &spi1_clk, &ssc0_clk, &ssc1_clk, - &tcb_clk, + &tcb0_clk, &pwm_clk, &tsc_clk, &dma_clk, @@ -221,6 +229,7 @@ static struct clk *periph_clocks[] __initdata = { &mmc1_clk, // irq0 &ohci_clk, + &tcb1_clk, }; /* diff --git a/arch/arm/mach-at91/at91sam9g45_devices.c b/arch/arm/mach-at91/at91sam9g45_devices.c index 501ac6f..5e71ccd 100644 --- a/arch/arm/mach-at91/at91sam9g45_devices.c +++ b/arch/arm/mach-at91/at91sam9g45_devices.c @@ -835,9 +835,9 @@ static struct platform_device at91sam9g45_tcb1_device = { static void __init at91_add_device_tc(void) { /* this chip has one clock and irq for all six TC channels */ - at91_clock_associate("tcb_clk", &at91sam9g45_tcb0_device.dev, "t0_clk"); + at91_clock_associate("tcb0_clk", &at91sam9g45_tcb0_device.dev, "t0_clk"); platform_device_register(&at91sam9g45_tcb0_device); - at91_clock_associate("tcb_clk", &at91sam9g45_tcb1_device.dev, "t0_clk"); + at91_clock_associate("tcb1_clk", &at91sam9g45_tcb1_device.dev, "t0_clk"); platform_device_register(&at91sam9g45_tcb1_device); } #else -- cgit v0.10.2 From 673424c0890a00e22398017c9adf999577526220 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Mon, 30 Aug 2010 17:37:05 +0200 Subject: pata_artop: Fix device ID parity check x % 1 always evaluates to 0, which clearly isn't the intent. The author probably had "% 2" or "& 1" in mind, and mispelled it. Signed-off-by: Jean Delvare Cc: Jeff Garzik Cc: Alan Cox Signed-off-by: Jeff Garzik diff --git a/drivers/ata/pata_artop.c b/drivers/ata/pata_artop.c index ba43f0f..2215632 100644 --- a/drivers/ata/pata_artop.c +++ b/drivers/ata/pata_artop.c @@ -74,7 +74,8 @@ static int artop6260_pre_reset(struct ata_link *link, unsigned long deadline) struct pci_dev *pdev = to_pci_dev(ap->host->dev); /* Odd numbered device ids are the units with enable bits (the -R cards) */ - if (pdev->device % 1 && !pci_test_config_bits(pdev, &artop_enable_bits[ap->port_no])) + if ((pdev->device & 1) && + !pci_test_config_bits(pdev, &artop_enable_bits[ap->port_no])) return -ENOENT; return ata_sff_prereset(link, deadline); -- cgit v0.10.2 From f1f5a807b051eddd3f302e503d39214e5bde0ef2 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Fri, 27 Aug 2010 11:09:15 +0200 Subject: ahci: fix hang on failed softreset ahci_do_softreset() compared the current time and deadline in reverse when calculating timeout for SRST issue. The result is that if @deadline is in future, SRST is issued with 0 timeout, which hasn't caused any problem because it later waits for DRDY with the correct timeout. If deadline is already exceeded by the time SRST is about to be issued, the timeout calculation underflows and if the device doesn't respond, timeout doesn't trigger for a _very_ long time. Reverse the incorrect comparison order. Signed-off-by: Tejun Heo Reported-by: Anssi Hannula Tested-by: Gwendal Grignou Cc: stable@kernel.org Signed-off-by: Jeff Garzik diff --git a/drivers/ata/libahci.c b/drivers/ata/libahci.c index 666850d..68dc678 100644 --- a/drivers/ata/libahci.c +++ b/drivers/ata/libahci.c @@ -1326,7 +1326,7 @@ int ahci_do_softreset(struct ata_link *link, unsigned int *class, /* issue the first D2H Register FIS */ msecs = 0; now = jiffies; - if (time_after(now, deadline)) + if (time_after(deadline, now)) msecs = jiffies_to_msecs(deadline - now); tf.ctl |= ATA_SRST; -- cgit v0.10.2 From 9cb627d5f38830ca19aa0dca52d1d3a633018bf7 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Wed, 1 Sep 2010 12:58:43 +0200 Subject: perf, trace: Fix module leak Commit 1c024eca (perf, trace: Optimize tracepoints by using per-tracepoint-per-cpu hlist to track events) caused a module refcount leak. Reported-And-Tested-by: Avi Kivity Signed-off-by: Peter Zijlstra LKML-Reference: <4C7E1F12.8030304@cn.fujitsu.com> Signed-off-by: Ingo Molnar diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c index 000e6e8..31cc4cb 100644 --- a/kernel/trace/trace_event_perf.c +++ b/kernel/trace/trace_event_perf.c @@ -91,6 +91,8 @@ int perf_trace_init(struct perf_event *p_event) tp_event->class && tp_event->class->reg && try_module_get(tp_event->mod)) { ret = perf_trace_event_init(tp_event, p_event); + if (ret) + module_put(tp_event->mod); break; } } @@ -146,6 +148,7 @@ void perf_trace_destroy(struct perf_event *p_event) } } out: + module_put(tp_event->mod); mutex_unlock(&event_mutex); } -- cgit v0.10.2 From 5e11637e2c929e34dcc0fbbfb48bdb638937701a Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 11 Jun 2010 13:35:08 +0200 Subject: perf: Fix CPU hotplug Since we have UP_PREPARE, we should also have UP_CANCELED. Signed-off-by: Peter Zijlstra Cc: paulus LKML-Reference: Signed-off-by: Ingo Molnar diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 657555a..db5b560 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -5761,15 +5761,15 @@ perf_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu) { unsigned int cpu = (long)hcpu; - switch (action) { + switch (action & ~CPU_TASKS_FROZEN) { case CPU_UP_PREPARE: - case CPU_UP_PREPARE_FROZEN: + case CPU_DOWN_FAILED: perf_event_init_cpu(cpu); break; + case CPU_UP_CANCELED: case CPU_DOWN_PREPARE: - case CPU_DOWN_PREPARE_FROZEN: perf_event_exit_cpu(cpu); break; -- cgit v0.10.2 From da2b71edd8a7db44fe1746261410a981f3e03632 Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Mon, 23 Aug 2010 13:42:51 -0700 Subject: sched: Move sched_avg_update() to update_cpu_load() Currently sched_avg_update() (which updates rt_avg stats in the rq) is getting called from scale_rt_power() (in the load balance context) which doesn't take rq->lock. Fix it by moving the sched_avg_update() to more appropriate update_cpu_load() where the CFS load gets updated as well. Signed-off-by: Suresh Siddha Signed-off-by: Peter Zijlstra LKML-Reference: <1282596171.2694.3.camel@sbsiddha-MOBL3> Signed-off-by: Ingo Molnar diff --git a/kernel/sched.c b/kernel/sched.c index 09b574e..ed09d4f 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -1294,6 +1294,10 @@ static void resched_task(struct task_struct *p) static void sched_rt_avg_update(struct rq *rq, u64 rt_delta) { } + +static void sched_avg_update(struct rq *rq) +{ +} #endif /* CONFIG_SMP */ #if BITS_PER_LONG == 32 @@ -3182,6 +3186,8 @@ static void update_cpu_load(struct rq *this_rq) this_rq->cpu_load[i] = (old_load * (scale - 1) + new_load) >> i; } + + sched_avg_update(this_rq); } static void update_cpu_load_active(struct rq *this_rq) diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index ab661eb..f53ec75 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -2268,8 +2268,6 @@ unsigned long scale_rt_power(int cpu) struct rq *rq = cpu_rq(cpu); u64 total, available; - sched_avg_update(rq); - total = sched_avg_period() + (rq->clock - rq->age_stamp); available = total - rq->rt_avg; -- cgit v0.10.2 From 85e00b55517a5696282b80836655e6b6f5702df0 Mon Sep 17 00:00:00 2001 From: Jovi Zhang Date: Thu, 9 Sep 2010 13:30:59 -0300 Subject: perf symbols: Fix multiple initialization of symbol system By returning immediately if it was already initialized, do it as well at symbol__exit, refusing multiple deinitializations. This fixes problems in the kmem, sched and timechart commands. Reported-by: Davidlohr Bueso Cc: Davidlohr Bueso Signed-off-by: Jovi Zhang LKML-Reference: AANLkTi=9Cn=R8SPMCRp5z+gEjXbaBHeb-AaOtRbuwwcn@mail.gmail.com Signed-off-by: Arnaldo Carvalho de Melo diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c index 1a36773..b2f5ae9 100644 --- a/tools/perf/util/symbol.c +++ b/tools/perf/util/symbol.c @@ -2268,6 +2268,9 @@ static int setup_list(struct strlist **list, const char *list_str, int symbol__init(void) { + if (symbol_conf.initialized) + return 0; + elf_version(EV_CURRENT); if (symbol_conf.sort_by_name) symbol_conf.priv_size += (sizeof(struct symbol_name_rb_node) - @@ -2293,6 +2296,7 @@ int symbol__init(void) symbol_conf.sym_list_str, "symbol") < 0) goto out_free_comm_list; + symbol_conf.initialized = true; return 0; out_free_dso_list: @@ -2304,11 +2308,14 @@ out_free_comm_list: void symbol__exit(void) { + if (!symbol_conf.initialized) + return; strlist__delete(symbol_conf.sym_list); strlist__delete(symbol_conf.dso_list); strlist__delete(symbol_conf.comm_list); vmlinux_path__exit(); symbol_conf.sym_list = symbol_conf.dso_list = symbol_conf.comm_list = NULL; + symbol_conf.initialized = false; } int machines__create_kernel_maps(struct rb_root *self, pid_t pid) diff --git a/tools/perf/util/symbol.h b/tools/perf/util/symbol.h index b7a8da4..ea95c27 100644 --- a/tools/perf/util/symbol.h +++ b/tools/perf/util/symbol.h @@ -69,7 +69,8 @@ struct symbol_conf { show_nr_samples, use_callchain, exclude_other, - show_cpu_utilization; + show_cpu_utilization, + initialized; const char *vmlinux_name, *source_prefix, *field_sep; -- cgit v0.10.2 From 9e84ed63dc71e13b62cea5ec6b0049260cca0b7a Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 9 Sep 2010 22:39:41 +0100 Subject: ARM: Partially revert "Auto calculate ZRELADDR and provide option for exceptions" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Partially revert e69edc7, which introduced automatic zreladdr support. The change in the way the manual definition is defined seems to be error and conflict prone. Go back to the original way we were handling this for the time being, while keeping the automatic zreladdr facility. Acked-by: Uwe Kleine-König diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 16bc8eb..553b7cf 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -1576,97 +1576,6 @@ config AUTO_ZRELADDR 0xf8000000. This assumes the zImage being placed in the first 128MB from start of memory. -config ZRELADDR - hex "Physical address of the decompressed kernel image" - depends on !AUTO_ZRELADDR - default 0x00008000 if ARCH_BCMRING ||\ - ARCH_CNS3XXX ||\ - ARCH_DOVE ||\ - ARCH_EBSA110 ||\ - ARCH_FOOTBRIDGE ||\ - ARCH_INTEGRATOR ||\ - ARCH_IOP13XX ||\ - ARCH_IOP33X ||\ - ARCH_IXP2000 ||\ - ARCH_IXP23XX ||\ - ARCH_IXP4XX ||\ - ARCH_KIRKWOOD ||\ - ARCH_KS8695 ||\ - ARCH_LOKI ||\ - ARCH_MMP ||\ - ARCH_MV78XX0 ||\ - ARCH_NOMADIK ||\ - ARCH_NUC93X ||\ - ARCH_NS9XXX ||\ - ARCH_ORION5X ||\ - ARCH_SPEAR3XX ||\ - ARCH_SPEAR6XX ||\ - ARCH_TEGRA ||\ - ARCH_U8500 ||\ - ARCH_VERSATILE ||\ - ARCH_W90X900 - default 0x08008000 if ARCH_MX1 ||\ - ARCH_SHARK - default 0x10008000 if ARCH_MSM ||\ - ARCH_OMAP1 ||\ - ARCH_RPC - default 0x20008000 if ARCH_S5P6440 ||\ - ARCH_S5P6442 ||\ - ARCH_S5PC100 ||\ - ARCH_S5PV210 - default 0x30008000 if ARCH_S3C2410 ||\ - ARCH_S3C2400 ||\ - ARCH_S3C2412 ||\ - ARCH_S3C2416 ||\ - ARCH_S3C2440 ||\ - ARCH_S3C2443 - default 0x40008000 if ARCH_STMP378X ||\ - ARCH_STMP37XX ||\ - ARCH_SH7372 ||\ - ARCH_SH7377 ||\ - ARCH_S5PV310 - default 0x50008000 if ARCH_S3C64XX ||\ - ARCH_SH7367 - default 0x60008000 if ARCH_VEXPRESS - default 0x80008000 if ARCH_MX25 ||\ - ARCH_MX3 ||\ - ARCH_NETX ||\ - ARCH_OMAP2PLUS ||\ - ARCH_PNX4008 - default 0x90008000 if ARCH_MX5 ||\ - ARCH_MX91231 - default 0xa0008000 if ARCH_IOP32X ||\ - ARCH_PXA ||\ - MACH_MX27 - default 0xc0008000 if ARCH_LH7A40X ||\ - MACH_MX21 - default 0xf0008000 if ARCH_AAEC2000 ||\ - ARCH_L7200 - default 0xc0028000 if ARCH_CLPS711X - default 0x70008000 if ARCH_AT91 && (ARCH_AT91CAP9 || ARCH_AT91SAM9G45) - default 0x20008000 if ARCH_AT91 && !(ARCH_AT91CAP9 || ARCH_AT91SAM9G45) - default 0xc0008000 if ARCH_DAVINCI && ARCH_DAVINCI_DA8XX - default 0x80008000 if ARCH_DAVINCI && !ARCH_DAVINCI_DA8XX - default 0x00008000 if ARCH_EP93XX && EP93XX_SDCE3_SYNC_PHYS_OFFSET - default 0xc0008000 if ARCH_EP93XX && EP93XX_SDCE0_PHYS_OFFSET - default 0xd0008000 if ARCH_EP93XX && EP93XX_SDCE1_PHYS_OFFSET - default 0xe0008000 if ARCH_EP93XX && EP93XX_SDCE2_PHYS_OFFSET - default 0xf0008000 if ARCH_EP93XX && EP93XX_SDCE3_ASYNC_PHYS_OFFSET - default 0x00008000 if ARCH_GEMINI && GEMINI_MEM_SWAP - default 0x10008000 if ARCH_GEMINI && !GEMINI_MEM_SWAP - default 0x70008000 if ARCH_REALVIEW && REALVIEW_HIGH_PHYS_OFFSET - default 0x00008000 if ARCH_REALVIEW && !REALVIEW_HIGH_PHYS_OFFSET - default 0xc0208000 if ARCH_SA1100 && SA1111 - default 0xc0008000 if ARCH_SA1100 && !SA1111 - default 0x30108000 if ARCH_S3C2410 && PM_H1940 - default 0x28E08000 if ARCH_U300 && MACH_U300_SINGLE_RAM - default 0x48008000 if ARCH_U300 && !MACH_U300_SINGLE_RAM - help - ZRELADDR is the physical address where the decompressed kernel - image will be placed. ZRELADDR has to be specified when the - assumption of AUTO_ZRELADDR is not valid, or when ZBOOT_ROM is - selected. - endmenu menu "CPU Power Management" diff --git a/arch/arm/boot/Makefile b/arch/arm/boot/Makefile index f705213..4a590f4 100644 --- a/arch/arm/boot/Makefile +++ b/arch/arm/boot/Makefile @@ -14,16 +14,18 @@ MKIMAGE := $(srctree)/scripts/mkuboot.sh ifneq ($(MACHINE),) --include $(srctree)/$(MACHINE)/Makefile.boot +include $(srctree)/$(MACHINE)/Makefile.boot endif # Note: the following conditions must always be true: +# ZRELADDR == virt_to_phys(PAGE_OFFSET + TEXT_OFFSET) # PARAMS_PHYS must be within 4MB of ZRELADDR # INITRD_PHYS must be in RAM +ZRELADDR := $(zreladdr-y) PARAMS_PHYS := $(params_phys-y) INITRD_PHYS := $(initrd_phys-y) -export INITRD_PHYS PARAMS_PHYS +export ZRELADDR INITRD_PHYS PARAMS_PHYS targets := Image zImage xipImage bootpImage uImage @@ -65,7 +67,7 @@ quiet_cmd_uimage = UIMAGE $@ ifeq ($(CONFIG_ZBOOT_ROM),y) $(obj)/uImage: LOADADDR=$(CONFIG_ZBOOT_ROM_TEXT) else -$(obj)/uImage: LOADADDR=$(CONFIG_ZRELADDR) +$(obj)/uImage: LOADADDR=$(ZRELADDR) endif ifeq ($(CONFIG_THUMB2_KERNEL),y) diff --git a/arch/arm/boot/compressed/Makefile b/arch/arm/boot/compressed/Makefile index 68775e3..b23f6bc 100644 --- a/arch/arm/boot/compressed/Makefile +++ b/arch/arm/boot/compressed/Makefile @@ -79,6 +79,10 @@ endif EXTRA_CFLAGS := -fpic -fno-builtin EXTRA_AFLAGS := -Wa,-march=all +# Supply ZRELADDR to the decompressor via a linker symbol. +ifneq ($(CONFIG_AUTO_ZRELADDR),y) +LDFLAGS_vmlinux := --defsym zreladdr=$(ZRELADDR) +endif ifeq ($(CONFIG_CPU_ENDIAN_BE8),y) LDFLAGS_vmlinux += --be8 endif diff --git a/arch/arm/boot/compressed/head.S b/arch/arm/boot/compressed/head.S index 6af9907..6825c34 100644 --- a/arch/arm/boot/compressed/head.S +++ b/arch/arm/boot/compressed/head.S @@ -177,7 +177,7 @@ not_angel: and r4, pc, #0xf8000000 add r4, r4, #TEXT_OFFSET #else - ldr r4, =CONFIG_ZRELADDR + ldr r4, =zreladdr #endif subs r0, r0, r1 @ calculate the delta offset -- cgit v0.10.2 From a14d0404088a7b0d51bb5219baf33d77e5592007 Mon Sep 17 00:00:00 2001 From: Russell King Date: Thu, 9 Sep 2010 22:43:02 +0100 Subject: ARM: Update mach-types Signed-off-by: Russell King diff --git a/arch/arm/tools/mach-types b/arch/arm/tools/mach-types index 48cbdcb..55590a4 100644 --- a/arch/arm/tools/mach-types +++ b/arch/arm/tools/mach-types @@ -12,7 +12,7 @@ # # http://www.arm.linux.org.uk/developer/machines/?action=new # -# Last update: Mon Jul 12 21:10:14 2010 +# Last update: Thu Sep 9 22:43:01 2010 # # machine_is_xxx CONFIG_xxxx MACH_TYPE_xxx number # @@ -2622,7 +2622,7 @@ kraken MACH_KRAKEN KRAKEN 2634 gw2388 MACH_GW2388 GW2388 2635 jadecpu MACH_JADECPU JADECPU 2636 carlisle MACH_CARLISLE CARLISLE 2637 -lux_sf9 MACH_LUX_SFT9 LUX_SFT9 2638 +lux_sf9 MACH_LUX_SF9 LUX_SF9 2638 nemid_tb MACH_NEMID_TB NEMID_TB 2639 terrier MACH_TERRIER TERRIER 2640 turbot MACH_TURBOT TURBOT 2641 @@ -2950,3 +2950,97 @@ davinci_dm365_dvr MACH_DAVINCI_DM365_DVR DAVINCI_DM365_DVR 2963 netviz MACH_NETVIZ NETVIZ 2964 flexibity MACH_FLEXIBITY FLEXIBITY 2965 wlan_computer MACH_WLAN_COMPUTER WLAN_COMPUTER 2966 +lpc24xx MACH_LPC24XX LPC24XX 2967 +spica MACH_SPICA SPICA 2968 +gpsdisplay MACH_GPSDISPLAY GPSDISPLAY 2969 +bipnet MACH_BIPNET BIPNET 2970 +overo_ctu_inertial MACH_OVERO_CTU_INERTIAL OVERO_CTU_INERTIAL 2971 +davinci_dm355_mmm MACH_DAVINCI_DM355_MMM DAVINCI_DM355_MMM 2972 +pc9260_v2 MACH_PC9260_V2 PC9260_V2 2973 +ptx7545 MACH_PTX7545 PTX7545 2974 +tm_efdc MACH_TM_EFDC TM_EFDC 2975 +omap3_waldo1 MACH_OMAP3_WALDO1 OMAP3_WALDO1 2977 +flyer MACH_FLYER FLYER 2978 +tornado3240 MACH_TORNADO3240 TORNADO3240 2979 +soli_01 MACH_SOLI_01 SOLI_01 2980 +omapl138_europalc MACH_OMAPL138_EUROPALC OMAPL138_EUROPALC 2981 +helios_v1 MACH_HELIOS_V1 HELIOS_V1 2982 +netspace_lite_v2 MACH_NETSPACE_LITE_V2 NETSPACE_LITE_V2 2983 +ssc MACH_SSC SSC 2984 +premierwave_en MACH_PREMIERWAVE_EN PREMIERWAVE_EN 2985 +wasabi MACH_WASABI WASABI 2986 +vivow MACH_VIVOW VIVOW 2987 +mx50_rdp MACH_MX50_RDP MX50_RDP 2988 +universal MACH_UNIVERSAL UNIVERSAL 2989 +real6410 MACH_REAL6410 REAL6410 2990 +spx_sakura MACH_SPX_SAKURA SPX_SAKURA 2991 +ij3k_2440 MACH_IJ3K_2440 IJ3K_2440 2992 +omap3_bc10 MACH_OMAP3_BC10 OMAP3_BC10 2993 +thebe MACH_THEBE THEBE 2994 +rv082 MACH_RV082 RV082 2995 +armlguest MACH_ARMLGUEST ARMLGUEST 2996 +tjinc1000 MACH_TJINC1000 TJINC1000 2997 +dockstar MACH_DOCKSTAR DOCKSTAR 2998 +ax8008 MACH_AX8008 AX8008 2999 +gnet_sgce MACH_GNET_SGCE GNET_SGCE 3000 +pxwnas_500_1000 MACH_PXWNAS_500_1000 PXWNAS_500_1000 3001 +ea20 MACH_EA20 EA20 3002 +awm2 MACH_AWM2 AWM2 3003 +ti8148evm MACH_TI8148EVM TI8148EVM 3004 +tegra_seaboard MACH_TEGRA_SEABOARD TEGRA_SEABOARD 3005 +linkstation_chlv2 MACH_LINKSTATION_CHLV2 LINKSTATION_CHLV2 3006 +tera_pro2_rack MACH_TERA_PRO2_RACK TERA_PRO2_RACK 3007 +rubys MACH_RUBYS RUBYS 3008 +aquarius MACH_AQUARIUS AQUARIUS 3009 +mx53_ard MACH_MX53_ARD MX53_ARD 3010 +mx53_smd MACH_MX53_SMD MX53_SMD 3011 +lswxl MACH_LSWXL LSWXL 3012 +dove_avng_v3 MACH_DOVE_AVNG_V3 DOVE_AVNG_V3 3013 +sdi_ess_9263 MACH_SDI_ESS_9263 SDI_ESS_9263 3014 +jocpu550 MACH_JOCPU550 JOCPU550 3015 +msm8x60_rumi3 MACH_MSM8X60_RUMI3 MSM8X60_RUMI3 3016 +msm8x60_ffa MACH_MSM8X60_FFA MSM8X60_FFA 3017 +yanomami MACH_YANOMAMI YANOMAMI 3018 +gta04 MACH_GTA04 GTA04 3019 +cm_a510 MACH_CM_A510 CM_A510 3020 +omap3_rfs200 MACH_OMAP3_RFS200 OMAP3_RFS200 3021 +kx33xx MACH_KX33XX KX33XX 3022 +ptx7510 MACH_PTX7510 PTX7510 3023 +top9000 MACH_TOP9000 TOP9000 3024 +teenote MACH_TEENOTE TEENOTE 3025 +ts3 MACH_TS3 TS3 3026 +a0 MACH_A0 A0 3027 +fsm9xxx_surf MACH_FSM9XXX_SURF FSM9XXX_SURF 3028 +fsm9xxx_ffa MACH_FSM9XXX_FFA FSM9XXX_FFA 3029 +frrhwcdma60w MACH_FRRHWCDMA60W FRRHWCDMA60W 3030 +remus MACH_REMUS REMUS 3031 +at91cap7xdk MACH_AT91CAP7XDK AT91CAP7XDK 3032 +at91cap7stk MACH_AT91CAP7STK AT91CAP7STK 3033 +kt_sbc_sam9_1 MACH_KT_SBC_SAM9_1 KT_SBC_SAM9_1 3034 +oratisrouter MACH_ORATISROUTER ORATISROUTER 3035 +armada_xp_db MACH_ARMADA_XP_DB ARMADA_XP_DB 3036 +spdm MACH_SPDM SPDM 3037 +gtib MACH_GTIB GTIB 3038 +dgm3240 MACH_DGM3240 DGM3240 3039 +atlas_i_lpe MACH_ATLAS_I_LPE ATLAS_I_LPE 3040 +htcmega MACH_HTCMEGA HTCMEGA 3041 +tricorder MACH_TRICORDER TRICORDER 3042 +tx28 MACH_TX28 TX28 3043 +bstbrd MACH_BSTBRD BSTBRD 3044 +pwb3090 MACH_PWB3090 PWB3090 3045 +idea6410 MACH_IDEA6410 IDEA6410 3046 +qbc9263 MACH_QBC9263 QBC9263 3047 +borabora MACH_BORABORA BORABORA 3048 +valdez MACH_VALDEZ VALDEZ 3049 +ls9g20 MACH_LS9G20 LS9G20 3050 +mios_v1 MACH_MIOS_V1 MIOS_V1 3051 +s5pc110_crespo MACH_S5PC110_CRESPO S5PC110_CRESPO 3052 +controltek9g20 MACH_CONTROLTEK9G20 CONTROLTEK9G20 3053 +tin307 MACH_TIN307 TIN307 3054 +tin510 MACH_TIN510 TIN510 3055 +bluecheese MACH_BLUECHEESE BLUECHEESE 3057 +tem3x30 MACH_TEM3X30 TEM3X30 3058 +harvest_desoto MACH_HARVEST_DESOTO HARVEST_DESOTO 3059 +msm8x60_qrdc MACH_MSM8X60_QRDC MSM8X60_QRDC 3060 +spear900 MACH_SPEAR900 SPEAR900 3061 +pcontrol_g20 MACH_PCONTROL_G20 PCONTROL_G20 3062 -- cgit v0.10.2 From 123031c0eeda6144b4002dc3285375aa9ae9dc11 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Wed, 8 Sep 2010 11:04:21 +0000 Subject: sctp: fix test for end of loop Add a list_has_sctp_addr function to simplify loop Based on a patches by Dan Carpenter and David Miller Signed-off-by: Joe Perches Acked-by: Vlad Yasevich Signed-off-by: David S. Miller diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 24b2cd5..d344dc4 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -1232,6 +1232,18 @@ out: return 0; } +static bool list_has_sctp_addr(const struct list_head *list, + union sctp_addr *ipaddr) +{ + struct sctp_transport *addr; + + list_for_each_entry(addr, list, transports) { + if (sctp_cmp_addr_exact(ipaddr, &addr->ipaddr)) + return true; + } + + return false; +} /* A restart is occurring, check to make sure no new addresses * are being added as we may be under a takeover attack. */ @@ -1240,10 +1252,10 @@ static int sctp_sf_check_restart_addrs(const struct sctp_association *new_asoc, struct sctp_chunk *init, sctp_cmd_seq_t *commands) { - struct sctp_transport *new_addr, *addr; - int found; + struct sctp_transport *new_addr; + int ret = 1; - /* Implementor's Guide - Sectin 5.2.2 + /* Implementor's Guide - Section 5.2.2 * ... * Before responding the endpoint MUST check to see if the * unexpected INIT adds new addresses to the association. If new @@ -1254,31 +1266,19 @@ static int sctp_sf_check_restart_addrs(const struct sctp_association *new_asoc, /* Search through all current addresses and make sure * we aren't adding any new ones. */ - new_addr = NULL; - found = 0; - list_for_each_entry(new_addr, &new_asoc->peer.transport_addr_list, - transports) { - found = 0; - list_for_each_entry(addr, &asoc->peer.transport_addr_list, - transports) { - if (sctp_cmp_addr_exact(&new_addr->ipaddr, - &addr->ipaddr)) { - found = 1; - break; - } - } - if (!found) + transports) { + if (!list_has_sctp_addr(&asoc->peer.transport_addr_list, + &new_addr->ipaddr)) { + sctp_sf_send_restart_abort(&new_addr->ipaddr, init, + commands); + ret = 0; break; - } - - /* If a new address was added, ABORT the sender. */ - if (!found && new_addr) { - sctp_sf_send_restart_abort(&new_addr->ipaddr, init, commands); + } } /* Return success if all addresses were found. */ - return found; + return ret; } /* Populate the verification/tie tags based on overlapping INIT -- cgit v0.10.2 From f3c65b2870f2481f3646bc410a58a12989ecc704 Mon Sep 17 00:00:00 2001 From: David Vrabel Date: Thu, 9 Sep 2010 16:37:24 -0700 Subject: mmc: avoid getting CID on SDIO-only cards The introduction of support for SD combo cards breaks the initialization of all CSR SDIO chips. The GO_IDLE (CMD0) in mmc_sd_get_cid() causes CSR chips to be reset (this is non-standard behavior). When initializing an SDIO card check for a combo card by using the memory present bit in the R4 response to IO_SEND_OP_COND (CMD5). This avoids the call to mmc_sd_get_cid() on an SDIO-only card. Signed-off-by: David Vrabel Acked-by: Michal Mirolaw Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/mmc/core/sdio.c b/drivers/mmc/core/sdio.c index bd2755e..f332c52 100644 --- a/drivers/mmc/core/sdio.c +++ b/drivers/mmc/core/sdio.c @@ -362,9 +362,8 @@ static int mmc_sdio_init_card(struct mmc_host *host, u32 ocr, goto err; } - err = mmc_sd_get_cid(host, host->ocr & ocr, card->raw_cid); - - if (!err) { + if (ocr & R4_MEMORY_PRESENT + && mmc_sd_get_cid(host, host->ocr & ocr, card->raw_cid) == 0) { card->type = MMC_TYPE_SD_COMBO; if (oldcard && (oldcard->type != MMC_TYPE_SD_COMBO || diff --git a/include/linux/mmc/sdio.h b/include/linux/mmc/sdio.h index 329a8fa..245cdac 100644 --- a/include/linux/mmc/sdio.h +++ b/include/linux/mmc/sdio.h @@ -38,6 +38,8 @@ * [8:0] Byte/block count */ +#define R4_MEMORY_PRESENT (1 << 27) + /* SDIO status in R5 Type -- cgit v0.10.2 From e0bf1024b36be90da241af3c2767311e055b612c Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Thu, 9 Sep 2010 16:37:26 -0700 Subject: kfifo: add parenthesis for macro parameter reference Some macro parameter references inside typeof() operator are not enclosed with parenthesis. It should be safer to add them. Signed-off-by: Huang Ying Acked-by: Stefani Seibold Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/kfifo.h b/include/linux/kfifo.h index 4aa95f2..62dbee5 100644 --- a/include/linux/kfifo.h +++ b/include/linux/kfifo.h @@ -214,7 +214,7 @@ __kfifo_must_check_helper(unsigned int val) */ #define kfifo_reset(fifo) \ (void)({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ __tmp->kfifo.in = __tmp->kfifo.out = 0; \ }) @@ -228,7 +228,7 @@ __kfifo_must_check_helper(unsigned int val) */ #define kfifo_reset_out(fifo) \ (void)({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ __tmp->kfifo.out = __tmp->kfifo.in; \ }) @@ -238,7 +238,7 @@ __kfifo_must_check_helper(unsigned int val) */ #define kfifo_len(fifo) \ ({ \ - typeof(fifo + 1) __tmpl = (fifo); \ + typeof((fifo) + 1) __tmpl = (fifo); \ __tmpl->kfifo.in - __tmpl->kfifo.out; \ }) @@ -248,7 +248,7 @@ __kfifo_must_check_helper(unsigned int val) */ #define kfifo_is_empty(fifo) \ ({ \ - typeof(fifo + 1) __tmpq = (fifo); \ + typeof((fifo) + 1) __tmpq = (fifo); \ __tmpq->kfifo.in == __tmpq->kfifo.out; \ }) @@ -258,7 +258,7 @@ __kfifo_must_check_helper(unsigned int val) */ #define kfifo_is_full(fifo) \ ({ \ - typeof(fifo + 1) __tmpq = (fifo); \ + typeof((fifo) + 1) __tmpq = (fifo); \ kfifo_len(__tmpq) > __tmpq->kfifo.mask; \ }) @@ -269,7 +269,7 @@ __kfifo_must_check_helper(unsigned int val) #define kfifo_avail(fifo) \ __kfifo_must_check_helper( \ ({ \ - typeof(fifo + 1) __tmpq = (fifo); \ + typeof((fifo) + 1) __tmpq = (fifo); \ const size_t __recsize = sizeof(*__tmpq->rectype); \ unsigned int __avail = kfifo_size(__tmpq) - kfifo_len(__tmpq); \ (__recsize) ? ((__avail <= __recsize) ? 0 : \ @@ -284,7 +284,7 @@ __kfifo_must_check_helper( \ */ #define kfifo_skip(fifo) \ (void)({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ if (__recsize) \ @@ -302,7 +302,7 @@ __kfifo_must_check_helper( \ #define kfifo_peek_len(fifo) \ __kfifo_must_check_helper( \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ (!__recsize) ? kfifo_len(__tmp) * sizeof(*__tmp->type) : \ @@ -325,7 +325,7 @@ __kfifo_must_check_helper( \ #define kfifo_alloc(fifo, size, gfp_mask) \ __kfifo_must_check_helper( \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ __is_kfifo_ptr(__tmp) ? \ __kfifo_alloc(__kfifo, size, sizeof(*__tmp->type), gfp_mask) : \ @@ -339,7 +339,7 @@ __kfifo_must_check_helper( \ */ #define kfifo_free(fifo) \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ if (__is_kfifo_ptr(__tmp)) \ __kfifo_free(__kfifo); \ @@ -358,7 +358,7 @@ __kfifo_must_check_helper( \ */ #define kfifo_init(fifo, buffer, size) \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ __is_kfifo_ptr(__tmp) ? \ __kfifo_init(__kfifo, buffer, size, sizeof(*__tmp->type)) : \ @@ -379,8 +379,8 @@ __kfifo_must_check_helper( \ */ #define kfifo_put(fifo, val) \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ - typeof(val + 1) __val = (val); \ + typeof((fifo) + 1) __tmp = (fifo); \ + typeof((val) + 1) __val = (val); \ unsigned int __ret; \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ @@ -421,8 +421,8 @@ __kfifo_must_check_helper( \ #define kfifo_get(fifo, val) \ __kfifo_must_check_helper( \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ - typeof(val + 1) __val = (val); \ + typeof((fifo) + 1) __tmp = (fifo); \ + typeof((val) + 1) __val = (val); \ unsigned int __ret; \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ @@ -462,8 +462,8 @@ __kfifo_must_check_helper( \ #define kfifo_peek(fifo, val) \ __kfifo_must_check_helper( \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ - typeof(val + 1) __val = (val); \ + typeof((fifo) + 1) __tmp = (fifo); \ + typeof((val) + 1) __val = (val); \ unsigned int __ret; \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ @@ -501,8 +501,8 @@ __kfifo_must_check_helper( \ */ #define kfifo_in(fifo, buf, n) \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ - typeof(buf + 1) __buf = (buf); \ + typeof((fifo) + 1) __tmp = (fifo); \ + typeof((buf) + 1) __buf = (buf); \ unsigned long __n = (n); \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ @@ -554,8 +554,8 @@ __kfifo_must_check_helper( \ #define kfifo_out(fifo, buf, n) \ __kfifo_must_check_helper( \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ - typeof(buf + 1) __buf = (buf); \ + typeof((fifo) + 1) __tmp = (fifo); \ + typeof((buf) + 1) __buf = (buf); \ unsigned long __n = (n); \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ @@ -611,7 +611,7 @@ __kfifo_must_check_helper( \ #define kfifo_from_user(fifo, from, len, copied) \ __kfifo_must_check_helper( \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ const void __user *__from = (from); \ unsigned int __len = (len); \ unsigned int *__copied = (copied); \ @@ -639,7 +639,7 @@ __kfifo_must_check_helper( \ #define kfifo_to_user(fifo, to, len, copied) \ __kfifo_must_check_helper( \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ void __user *__to = (to); \ unsigned int __len = (len); \ unsigned int *__copied = (copied); \ @@ -666,7 +666,7 @@ __kfifo_must_check_helper( \ */ #define kfifo_dma_in_prepare(fifo, sgl, nents, len) \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ struct scatterlist *__sgl = (sgl); \ int __nents = (nents); \ unsigned int __len = (len); \ @@ -690,7 +690,7 @@ __kfifo_must_check_helper( \ */ #define kfifo_dma_in_finish(fifo, len) \ (void)({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ unsigned int __len = (len); \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ @@ -717,7 +717,7 @@ __kfifo_must_check_helper( \ */ #define kfifo_dma_out_prepare(fifo, sgl, nents, len) \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ struct scatterlist *__sgl = (sgl); \ int __nents = (nents); \ unsigned int __len = (len); \ @@ -741,7 +741,7 @@ __kfifo_must_check_helper( \ */ #define kfifo_dma_out_finish(fifo, len) \ (void)({ \ - typeof(fifo + 1) __tmp = (fifo); \ + typeof((fifo) + 1) __tmp = (fifo); \ unsigned int __len = (len); \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ @@ -766,8 +766,8 @@ __kfifo_must_check_helper( \ #define kfifo_out_peek(fifo, buf, n) \ __kfifo_must_check_helper( \ ({ \ - typeof(fifo + 1) __tmp = (fifo); \ - typeof(buf + 1) __buf = (buf); \ + typeof((fifo) + 1) __tmp = (fifo); \ + typeof((buf) + 1) __buf = (buf); \ unsigned long __n = (n); \ const size_t __recsize = sizeof(*__tmp->rectype); \ struct __kfifo *__kfifo = &__tmp->kfifo; \ -- cgit v0.10.2 From 110b7e9698601b28f313c2c560d51a8b1c742002 Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Thu, 9 Sep 2010 16:37:27 -0700 Subject: rtc-bfin: fix inverted logic in suspend path The int_clear helper takes a bitmask of interrupts to keep, not to disable. When suspending without wakeup enabled, we want to disable all interrupts, so use 0 (keep none) instead of -1 (keep all). Signed-off-by: Mike Frysinger Acked-by: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-bfin.c b/drivers/rtc/rtc-bfin.c index 72b2bcc..65facfb 100644 --- a/drivers/rtc/rtc-bfin.c +++ b/drivers/rtc/rtc-bfin.c @@ -426,7 +426,7 @@ static int bfin_rtc_suspend(struct platform_device *pdev, pm_message_t state) enable_irq_wake(IRQ_RTC); bfin_rtc_sync_pending(&pdev->dev); } else - bfin_rtc_int_clear(-1); + bfin_rtc_int_clear(0); return 0; } -- cgit v0.10.2 From b6de860651d5a9e56ba4f4e3edc1aa52ac2ac849 Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Thu, 9 Sep 2010 16:37:29 -0700 Subject: rtc-bfin: fix state restoration when resuming Much (but not all) of the RTC state is kept in the RTC peripheral which has its own power domain. Periodically (1 HZ), that state is synced from one power domain to the other (peripheral->core). When we are resuming, we need to wait for the sync to occur so that we don't get a mismatch of reading undefined state in the rest of the driver. Further, once the externally maintained bits have been synced back into the core, we then need to restore the bits maintained in the core. In our particular case, that is just the write completion interrupt bit. If we don't do any of this, working with the RTC causes ~5 second delays from time to time after waking up due to the write completion interrupt never firing. Reported-by: Michael Dean Reported-by: Michael Hennerich Signed-off-by: Mike Frysinger Acked-by: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-bfin.c b/drivers/rtc/rtc-bfin.c index 65facfb..d4fb82d 100644 --- a/drivers/rtc/rtc-bfin.c +++ b/drivers/rtc/rtc-bfin.c @@ -435,8 +435,17 @@ static int bfin_rtc_resume(struct platform_device *pdev) { if (device_may_wakeup(&pdev->dev)) disable_irq_wake(IRQ_RTC); - else - bfin_write_RTC_ISTAT(-1); + + /* + * Since only some of the RTC bits are maintained externally in the + * Vbat domain, we need to wait for the RTC MMRs to be synced into + * the core after waking up. This happens every RTC 1HZ. Once that + * has happened, we can go ahead and re-enable the important write + * complete interrupt event. + */ + while (!(bfin_read_RTC_ISTAT() & RTC_ISTAT_SEC)) + continue; + bfin_rtc_int_set(RTC_ISTAT_WRITE_COMPLETE); return 0; } -- cgit v0.10.2 From 841a23ae1e7a8f8758ac915ceaf2e11a1cc235c5 Mon Sep 17 00:00:00 2001 From: Jiri Slaby Date: Thu, 9 Sep 2010 16:37:30 -0700 Subject: MAINTAINERS: remove drivers/char/hpet.c bob.picco@hp.com doesn't work any more and Bob says that he's unlikely to work on hpet.c in the future. Signed-off-by: Jiri Slaby Acked-by: Bob Picco Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/MAINTAINERS b/MAINTAINERS index 7e189b5..05e5537 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2787,11 +2787,6 @@ S: Maintained F: arch/x86/kernel/hpet.c F: arch/x86/include/asm/hpet.h -HPET: ACPI -M: Bob Picco -S: Maintained -F: drivers/char/hpet.c - HPFS FILESYSTEM M: Mikulas Patocka W: http://artax.karlin.mff.cuni.cz/~mikulas/vyplody/hpfs/index-e.cgi -- cgit v0.10.2 From 4e70598c3b56e6fec551454c495d4d4025834749 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Thu, 9 Sep 2010 16:37:31 -0700 Subject: hp_accel: add quirks for HP ProBook 532x and HP Mini 5102 Added missing axis-mapping for HP ProBook 532x and HP Mini 5102. Signed-off-by: Takashi Iwai Cc: Eric Piel Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/hwmon/hp_accel.c b/drivers/hwmon/hp_accel.c index 7580f55..36e9575 100644 --- a/drivers/hwmon/hp_accel.c +++ b/drivers/hwmon/hp_accel.c @@ -221,6 +221,8 @@ static struct dmi_system_id lis3lv02d_dmi_ids[] = { AXIS_DMI_MATCH("HPB442x", "HP ProBook 442", xy_rotated_left), AXIS_DMI_MATCH("HPB452x", "HP ProBook 452", y_inverted), AXIS_DMI_MATCH("HPB522x", "HP ProBook 522", xy_swap), + AXIS_DMI_MATCH("HPB532x", "HP ProBook 532", y_inverted), + AXIS_DMI_MATCH("Mini5102", "HP Mini 5102", xy_rotated_left_usd), { NULL, } /* Laptop models without axis info (yet): * "NC6910" "HP Compaq 6910" -- cgit v0.10.2 From 7a801ac6f5067539ceb5fad0fe90ec49fc156e47 Mon Sep 17 00:00:00 2001 From: Jeff Moyer Date: Thu, 9 Sep 2010 16:37:33 -0700 Subject: O_DIRECT: fix the splitting up of contiguous I/O commit c2c6ca4 (direct-io: do not merge logically non-contiguous requests) introduced a bug whereby all O_DIRECT I/Os were submitted a page at a time to the block layer. The problem is that the code expected dio->block_in_file to correspond to the current page in the dio. In fact, it corresponds to the previous page submitted via submit_page_section. This was purely an oversight, as the dio->cur_page_fs_offset field was introduced for just this purpose. This patch simply uses the correct variable when calculating whether there is a mismatch between contiguous logical blocks and contiguous physical blocks (as described in the comments). I also switched the if conditional following this check to an else if, to ensure that we never call dio_bio_submit twice for the same dio (in theory, this should not happen, anyway). I've tested this by running blktrace and verifying that a 64KB I/O was submitted as a single I/O. I also ran the patched kernel through xfstests' aio tests using xfs, ext4 (with 1k and 4k block sizes) and btrfs and verified that there were no regressions as compared to an unpatched kernel. Signed-off-by: Jeff Moyer Acked-by: Josef Bacik Cc: Christoph Hellwig Cc: Chris Mason Cc: [2.6.35.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/direct-io.c b/fs/direct-io.c index 51f270b..48d74c7 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -634,7 +634,7 @@ static int dio_send_cur_page(struct dio *dio) int ret = 0; if (dio->bio) { - loff_t cur_offset = dio->block_in_file << dio->blkbits; + loff_t cur_offset = dio->cur_page_fs_offset; loff_t bio_next_offset = dio->logical_offset_in_bio + dio->bio->bi_size; @@ -659,7 +659,7 @@ static int dio_send_cur_page(struct dio *dio) * Submit now if the underlying fs is about to perform a * metadata read */ - if (dio->boundary) + else if (dio->boundary) dio_bio_submit(dio); } -- cgit v0.10.2 From 2f327dad14aa8bc939a4f0a2d3fdcf64a2d8c09e Mon Sep 17 00:00:00 2001 From: Simon Horman Date: Thu, 9 Sep 2010 16:37:33 -0700 Subject: MAINTAINERS: kexec-tools has moved Kexec tools has been moved to http://kernel.org/pub/linux/utils/kernel/kexec/ as user-space code shouldn't be in /pub/linux/kernel Signed-off-by: Simon Horman Acked-by: "Eric W. Biederman" Cc: "H. Peter Anvin" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/MAINTAINERS b/MAINTAINERS index 05e5537..a3da87f 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3399,7 +3399,7 @@ F: drivers/s390/kvm/ KEXEC M: Eric Biederman -W: http://ftp.kernel.org/pub/linux/kernel/people/horms/kexec-tools/ +W: http://kernel.org/pub/linux/utils/kernel/kexec/ L: kexec@lists.infradead.org S: Maintained F: include/linux/kexec.h -- cgit v0.10.2 From 85a0fdfd0f967507f3903e8419bc7e408f5a59de Mon Sep 17 00:00:00 2001 From: Peter Oberparleiter Date: Thu, 9 Sep 2010 16:37:35 -0700 Subject: gcov: fix null-pointer dereference for certain module types The gcov-kernel infrastructure expects that each object file is loaded only once. This may not be true, e.g. when loading multiple kernel modules which are linked to the same object file. As a result, loading such kernel modules will result in incorrect gcov results while unloading will cause a null-pointer dereference. This patch fixes these problems by changing the gcov-kernel infrastructure so that multiple profiling data sets can be associated with one debugfs entry. It applies to 2.6.36-rc1. Signed-off-by: Peter Oberparleiter Reported-by: Werner Spies Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/gcov/fs.c b/kernel/gcov/fs.c index ef3c3f8..f83972b 100644 --- a/kernel/gcov/fs.c +++ b/kernel/gcov/fs.c @@ -33,10 +33,11 @@ * @children: child nodes * @all: list head for list of all nodes * @parent: parent node - * @info: associated profiling data structure if not a directory - * @ghost: when an object file containing profiling data is unloaded we keep a - * copy of the profiling data here to allow collecting coverage data - * for cleanup code. Such a node is called a "ghost". + * @loaded_info: array of pointers to profiling data sets for loaded object + * files. + * @num_loaded: number of profiling data sets for loaded object files. + * @unloaded_info: accumulated copy of profiling data sets for unloaded + * object files. Used only when gcov_persist=1. * @dentry: main debugfs entry, either a directory or data file * @links: associated symbolic links * @name: data file basename @@ -51,10 +52,11 @@ struct gcov_node { struct list_head children; struct list_head all; struct gcov_node *parent; - struct gcov_info *info; - struct gcov_info *ghost; + struct gcov_info **loaded_info; + struct gcov_info *unloaded_info; struct dentry *dentry; struct dentry **links; + int num_loaded; char name[0]; }; @@ -136,16 +138,37 @@ static const struct seq_operations gcov_seq_ops = { }; /* - * Return the profiling data set for a given node. This can either be the - * original profiling data structure or a duplicate (also called "ghost") - * in case the associated object file has been unloaded. + * Return a profiling data set associated with the given node. This is + * either a data set for a loaded object file or a data set copy in case + * all associated object files have been unloaded. */ static struct gcov_info *get_node_info(struct gcov_node *node) { - if (node->info) - return node->info; + if (node->num_loaded > 0) + return node->loaded_info[0]; - return node->ghost; + return node->unloaded_info; +} + +/* + * Return a newly allocated profiling data set which contains the sum of + * all profiling data associated with the given node. + */ +static struct gcov_info *get_accumulated_info(struct gcov_node *node) +{ + struct gcov_info *info; + int i = 0; + + if (node->unloaded_info) + info = gcov_info_dup(node->unloaded_info); + else + info = gcov_info_dup(node->loaded_info[i++]); + if (!info) + return NULL; + for (; i < node->num_loaded; i++) + gcov_info_add(info, node->loaded_info[i]); + + return info; } /* @@ -163,9 +186,10 @@ static int gcov_seq_open(struct inode *inode, struct file *file) mutex_lock(&node_lock); /* * Read from a profiling data copy to minimize reference tracking - * complexity and concurrent access. + * complexity and concurrent access and to keep accumulating multiple + * profiling data sets associated with one node simple. */ - info = gcov_info_dup(get_node_info(node)); + info = get_accumulated_info(node); if (!info) goto out_unlock; iter = gcov_iter_new(info); @@ -225,12 +249,25 @@ static struct gcov_node *get_node_by_name(const char *name) return NULL; } +/* + * Reset all profiling data associated with the specified node. + */ +static void reset_node(struct gcov_node *node) +{ + int i; + + if (node->unloaded_info) + gcov_info_reset(node->unloaded_info); + for (i = 0; i < node->num_loaded; i++) + gcov_info_reset(node->loaded_info[i]); +} + static void remove_node(struct gcov_node *node); /* * write() implementation for gcov data files. Reset profiling data for the - * associated file. If the object file has been unloaded (i.e. this is - * a "ghost" node), remove the debug fs node as well. + * corresponding file. If all associated object files have been unloaded, + * remove the debug fs node as well. */ static ssize_t gcov_seq_write(struct file *file, const char __user *addr, size_t len, loff_t *pos) @@ -245,10 +282,10 @@ static ssize_t gcov_seq_write(struct file *file, const char __user *addr, node = get_node_by_name(info->filename); if (node) { /* Reset counts or remove node for unloaded modules. */ - if (node->ghost) + if (node->num_loaded == 0) remove_node(node); else - gcov_info_reset(node->info); + reset_node(node); } /* Reset counts for open file. */ gcov_info_reset(info); @@ -378,7 +415,10 @@ static void init_node(struct gcov_node *node, struct gcov_info *info, INIT_LIST_HEAD(&node->list); INIT_LIST_HEAD(&node->children); INIT_LIST_HEAD(&node->all); - node->info = info; + if (node->loaded_info) { + node->loaded_info[0] = info; + node->num_loaded = 1; + } node->parent = parent; if (name) strcpy(node->name, name); @@ -394,9 +434,13 @@ static struct gcov_node *new_node(struct gcov_node *parent, struct gcov_node *node; node = kzalloc(sizeof(struct gcov_node) + strlen(name) + 1, GFP_KERNEL); - if (!node) { - pr_warning("out of memory\n"); - return NULL; + if (!node) + goto err_nomem; + if (info) { + node->loaded_info = kcalloc(1, sizeof(struct gcov_info *), + GFP_KERNEL); + if (!node->loaded_info) + goto err_nomem; } init_node(node, info, name, parent); /* Differentiate between gcov data file nodes and directory nodes. */ @@ -416,6 +460,11 @@ static struct gcov_node *new_node(struct gcov_node *parent, list_add(&node->all, &all_head); return node; + +err_nomem: + kfree(node); + pr_warning("out of memory\n"); + return NULL; } /* Remove symbolic links associated with node. */ @@ -441,8 +490,9 @@ static void release_node(struct gcov_node *node) list_del(&node->all); debugfs_remove(node->dentry); remove_links(node); - if (node->ghost) - gcov_info_free(node->ghost); + kfree(node->loaded_info); + if (node->unloaded_info) + gcov_info_free(node->unloaded_info); kfree(node); } @@ -477,7 +527,7 @@ static struct gcov_node *get_child_by_name(struct gcov_node *parent, /* * write() implementation for reset file. Reset all profiling data to zero - * and remove ghost nodes. + * and remove nodes for which all associated object files are unloaded. */ static ssize_t reset_write(struct file *file, const char __user *addr, size_t len, loff_t *pos) @@ -487,8 +537,8 @@ static ssize_t reset_write(struct file *file, const char __user *addr, mutex_lock(&node_lock); restart: list_for_each_entry(node, &all_head, all) { - if (node->info) - gcov_info_reset(node->info); + if (node->num_loaded > 0) + reset_node(node); else if (list_empty(&node->children)) { remove_node(node); /* Several nodes may have gone - restart loop. */ @@ -564,37 +614,115 @@ err_remove: } /* - * The profiling data set associated with this node is being unloaded. Store a - * copy of the profiling data and turn this node into a "ghost". + * Associate a profiling data set with an existing node. Needs to be called + * with node_lock held. */ -static int ghost_node(struct gcov_node *node) +static void add_info(struct gcov_node *node, struct gcov_info *info) { - node->ghost = gcov_info_dup(node->info); - if (!node->ghost) { - pr_warning("could not save data for '%s' (out of memory)\n", - node->info->filename); - return -ENOMEM; + struct gcov_info **loaded_info; + int num = node->num_loaded; + + /* + * Prepare new array. This is done first to simplify cleanup in + * case the new data set is incompatible, the node only contains + * unloaded data sets and there's not enough memory for the array. + */ + loaded_info = kcalloc(num + 1, sizeof(struct gcov_info *), GFP_KERNEL); + if (!loaded_info) { + pr_warning("could not add '%s' (out of memory)\n", + info->filename); + return; + } + memcpy(loaded_info, node->loaded_info, + num * sizeof(struct gcov_info *)); + loaded_info[num] = info; + /* Check if the new data set is compatible. */ + if (num == 0) { + /* + * A module was unloaded, modified and reloaded. The new + * data set replaces the copy of the last one. + */ + if (!gcov_info_is_compatible(node->unloaded_info, info)) { + pr_warning("discarding saved data for %s " + "(incompatible version)\n", info->filename); + gcov_info_free(node->unloaded_info); + node->unloaded_info = NULL; + } + } else { + /* + * Two different versions of the same object file are loaded. + * The initial one takes precedence. + */ + if (!gcov_info_is_compatible(node->loaded_info[0], info)) { + pr_warning("could not add '%s' (incompatible " + "version)\n", info->filename); + kfree(loaded_info); + return; + } } - node->info = NULL; + /* Overwrite previous array. */ + kfree(node->loaded_info); + node->loaded_info = loaded_info; + node->num_loaded = num + 1; +} - return 0; +/* + * Return the index of a profiling data set associated with a node. + */ +static int get_info_index(struct gcov_node *node, struct gcov_info *info) +{ + int i; + + for (i = 0; i < node->num_loaded; i++) { + if (node->loaded_info[i] == info) + return i; + } + return -ENOENT; } /* - * Profiling data for this node has been loaded again. Add profiling data - * from previous instantiation and turn this node into a regular node. + * Save the data of a profiling data set which is being unloaded. */ -static void revive_node(struct gcov_node *node, struct gcov_info *info) +static void save_info(struct gcov_node *node, struct gcov_info *info) { - if (gcov_info_is_compatible(node->ghost, info)) - gcov_info_add(info, node->ghost); + if (node->unloaded_info) + gcov_info_add(node->unloaded_info, info); else { - pr_warning("discarding saved data for '%s' (version changed)\n", + node->unloaded_info = gcov_info_dup(info); + if (!node->unloaded_info) { + pr_warning("could not save data for '%s' " + "(out of memory)\n", info->filename); + } + } +} + +/* + * Disassociate a profiling data set from a node. Needs to be called with + * node_lock held. + */ +static void remove_info(struct gcov_node *node, struct gcov_info *info) +{ + int i; + + i = get_info_index(node, info); + if (i < 0) { + pr_warning("could not remove '%s' (not found)\n", info->filename); + return; } - gcov_info_free(node->ghost); - node->ghost = NULL; - node->info = info; + if (gcov_persist) + save_info(node, info); + /* Shrink array. */ + node->loaded_info[i] = node->loaded_info[node->num_loaded - 1]; + node->num_loaded--; + if (node->num_loaded > 0) + return; + /* Last loaded data set was removed. */ + kfree(node->loaded_info); + node->loaded_info = NULL; + node->num_loaded = 0; + if (!node->unloaded_info) + remove_node(node); } /* @@ -609,30 +737,18 @@ void gcov_event(enum gcov_action action, struct gcov_info *info) node = get_node_by_name(info->filename); switch (action) { case GCOV_ADD: - /* Add new node or revive ghost. */ - if (!node) { + if (node) + add_info(node, info); + else add_node(info); - break; - } - if (gcov_persist) - revive_node(node, info); - else { - pr_warning("could not add '%s' (already exists)\n", - info->filename); - } break; case GCOV_REMOVE: - /* Remove node or turn into ghost. */ - if (!node) { + if (node) + remove_info(node, info); + else { pr_warning("could not remove '%s' (not found)\n", info->filename); - break; } - if (gcov_persist) { - if (!ghost_node(node)) - break; - } - remove_node(node); break; } mutex_unlock(&node_lock); -- cgit v0.10.2 From ed430fec756ad65f7cfba24f8ad17c3d5a403290 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Thu, 9 Sep 2010 16:37:36 -0700 Subject: proc: export uncached bit properly in /proc/kpageflags Fix the left-over old ifdef for PG_uncached in /proc/kpageflags. Now it's used by x86, too. Signed-off-by: Takashi Iwai Cc: Wu Fengguang Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/proc/page.c b/fs/proc/page.c index 180cf5a..3b8b456 100644 --- a/fs/proc/page.c +++ b/fs/proc/page.c @@ -146,7 +146,7 @@ u64 stable_page_flags(struct page *page) u |= kpf_copy_bit(k, KPF_HWPOISON, PG_hwpoison); #endif -#ifdef CONFIG_IA64_UNCACHED_ALLOCATOR +#ifdef CONFIG_ARCH_USES_PG_UNCACHED u |= kpf_copy_bit(k, KPF_UNCACHED, PG_uncached); #endif -- cgit v0.10.2 From 31583bb0cf6cc40f2a468a4d2f3b9cbefd24f891 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Thu, 9 Sep 2010 16:37:37 -0700 Subject: cgroups: fix API thinko Add cgroup_attach_task_all() The existing cgroup_attach_task_current_cg() API is called by a thread to attach another thread to all of its cgroups; this is unsuitable for cases where a privileged task wants to attach itself to the cgroups of a less privileged one, since the call must be made from the context of the target task. This patch adds a more generic cgroup_attach_task_all() API that allows both the source task and to-be-moved task to be specified. cgroup_attach_task_current_cg() becomes a specialization of the more generic new function. [menage@google.com: rewrote changelog] [akpm@linux-foundation.org: address reviewer comments] Signed-off-by: Michael S. Tsirkin Tested-by: Alex Williamson Acked-by: Paul Menage Cc: Li Zefan Cc: Ben Blum Cc: Sridhar Samudrala Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index ed3e92e..0c99102 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -578,7 +578,12 @@ struct task_struct *cgroup_iter_next(struct cgroup *cgrp, void cgroup_iter_end(struct cgroup *cgrp, struct cgroup_iter *it); int cgroup_scan_tasks(struct cgroup_scanner *scan); int cgroup_attach_task(struct cgroup *, struct task_struct *); -int cgroup_attach_task_current_cg(struct task_struct *); +int cgroup_attach_task_all(struct task_struct *from, struct task_struct *); + +static inline int cgroup_attach_task_current_cg(struct task_struct *tsk) +{ + return cgroup_attach_task_all(current, tsk); +} /* * CSS ID is ID for cgroup_subsys_state structs under subsys. This only works @@ -636,6 +641,11 @@ static inline int cgroupstats_build(struct cgroupstats *stats, } /* No cgroups - nothing to do */ +static inline int cgroup_attach_task_all(struct task_struct *from, + struct task_struct *t) +{ + return 0; +} static inline int cgroup_attach_task_current_cg(struct task_struct *t) { return 0; diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 192f88c..c9483d8 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -1791,19 +1791,20 @@ out: } /** - * cgroup_attach_task_current_cg - attach task 'tsk' to current task's cgroup + * cgroup_attach_task_all - attach task 'tsk' to all cgroups of task 'from' + * @from: attach to all cgroups of a given task * @tsk: the task to be attached */ -int cgroup_attach_task_current_cg(struct task_struct *tsk) +int cgroup_attach_task_all(struct task_struct *from, struct task_struct *tsk) { struct cgroupfs_root *root; - struct cgroup *cur_cg; int retval = 0; cgroup_lock(); for_each_active_root(root) { - cur_cg = task_cgroup_from_root(current, root); - retval = cgroup_attach_task(cur_cg, tsk); + struct cgroup *from_cg = task_cgroup_from_root(from, root); + + retval = cgroup_attach_task(from_cg, tsk); if (retval) break; } @@ -1811,7 +1812,7 @@ int cgroup_attach_task_current_cg(struct task_struct *tsk) return retval; } -EXPORT_SYMBOL_GPL(cgroup_attach_task_current_cg); +EXPORT_SYMBOL_GPL(cgroup_attach_task_all); /* * Attach task with pid 'pid' to cgroup 'cgrp'. Call with cgroup_mutex -- cgit v0.10.2 From b78d6c5f51935ba89df8db33a57bacb547aa7325 Mon Sep 17 00:00:00 2001 From: Yusuke Goda Date: Thu, 9 Sep 2010 16:37:39 -0700 Subject: tmio_mmc: don't clear unhandled pending interrupts Previously, it was possible for ack_mmc_irqs() to clear pending interrupt bits in the CTL_STATUS register, even though the interrupt handler had not been called. This was because of a race that existed when doing a read-modify-write sequence on CTL_STATUS. After the read step in this sequence, if an interrupt occurred (causing one of the bits in CTL_STATUS to be set) the write step would inadvertently clear it. Observed with the TMIO_STAT_RXRDY bit together with CMD53 on AR6002 and BCM4318 SDIO cards in polled mode. This patch eliminates this race by only writing to CTL_STATUS and clearing the interrupts that were passed as an argument to ack_mmc_irqs()." [matt@console-pimps.org: rewrote changelog] Signed-off-by: Yusuke Goda Acked-by: Magnus Damm " Tested-by: Arnd Hannemann " Acked-by: Ian Molton Cc: Matt Fleming Cc: Samuel Ortiz Cc: Paul Mundt Cc: Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/mmc/host/tmio_mmc.h b/drivers/mmc/host/tmio_mmc.h index 64f7d5d..7944604 100644 --- a/drivers/mmc/host/tmio_mmc.h +++ b/drivers/mmc/host/tmio_mmc.h @@ -82,10 +82,7 @@ #define ack_mmc_irqs(host, i) \ do { \ - u32 mask;\ - mask = sd_ctrl_read32((host), CTL_STATUS); \ - mask &= ~((i) & TMIO_MASK_IRQ); \ - sd_ctrl_write32((host), CTL_STATUS, mask); \ + sd_ctrl_write32((host), CTL_STATUS, ~(i)); \ } while (0) -- cgit v0.10.2 From 5600efb1bc2745d93ae0bc08130117a84f2b9d69 Mon Sep 17 00:00:00 2001 From: Guennadi Liakhovetski Date: Thu, 9 Sep 2010 16:37:43 -0700 Subject: mmc: fix the use of kunmap_atomic() in tmio_mmc.h kunmap_atomic() takes the cookie, returned by the kmap_atomic() as its argument and not the page address, used as an argument to kmap_atomic(). This patch fixes the compile error: In file included from drivers/mmc/host/tmio_mmc.c:37: drivers/mmc/host/tmio_mmc.h: In function 'tmio_mmc_kunmap_atomic': drivers/mmc/host/tmio_mmc.h:192: error: negative width in bit-field '' Signed-off-by: Guennadi Liakhovetski Acked-by: Eric Miao Tested-by: Magnus Damm Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/mmc/host/tmio_mmc.c b/drivers/mmc/host/tmio_mmc.c index ee7d0a5..69d98e3 100644 --- a/drivers/mmc/host/tmio_mmc.c +++ b/drivers/mmc/host/tmio_mmc.c @@ -164,6 +164,7 @@ tmio_mmc_start_command(struct tmio_mmc_host *host, struct mmc_command *cmd) static void tmio_mmc_pio_irq(struct tmio_mmc_host *host) { struct mmc_data *data = host->data; + void *sg_virt; unsigned short *buf; unsigned int count; unsigned long flags; @@ -173,8 +174,8 @@ static void tmio_mmc_pio_irq(struct tmio_mmc_host *host) return; } - buf = (unsigned short *)(tmio_mmc_kmap_atomic(host, &flags) + - host->sg_off); + sg_virt = tmio_mmc_kmap_atomic(host->sg_ptr, &flags); + buf = (unsigned short *)(sg_virt + host->sg_off); count = host->sg_ptr->length - host->sg_off; if (count > data->blksz) @@ -191,7 +192,7 @@ static void tmio_mmc_pio_irq(struct tmio_mmc_host *host) host->sg_off += count; - tmio_mmc_kunmap_atomic(host, &flags); + tmio_mmc_kunmap_atomic(sg_virt, &flags); if (host->sg_off == host->sg_ptr->length) tmio_mmc_next_sg(host); diff --git a/drivers/mmc/host/tmio_mmc.h b/drivers/mmc/host/tmio_mmc.h index 7944604..0fedc78 100644 --- a/drivers/mmc/host/tmio_mmc.h +++ b/drivers/mmc/host/tmio_mmc.h @@ -174,19 +174,17 @@ static inline int tmio_mmc_next_sg(struct tmio_mmc_host *host) return --host->sg_len; } -static inline char *tmio_mmc_kmap_atomic(struct tmio_mmc_host *host, +static inline char *tmio_mmc_kmap_atomic(struct scatterlist *sg, unsigned long *flags) { - struct scatterlist *sg = host->sg_ptr; - local_irq_save(*flags); return kmap_atomic(sg_page(sg), KM_BIO_SRC_IRQ) + sg->offset; } -static inline void tmio_mmc_kunmap_atomic(struct tmio_mmc_host *host, +static inline void tmio_mmc_kunmap_atomic(void *virt, unsigned long *flags) { - kunmap_atomic(sg_page(host->sg_ptr), KM_BIO_SRC_IRQ); + kunmap_atomic(virt, KM_BIO_SRC_IRQ); local_irq_restore(*flags); } -- cgit v0.10.2 From 60c2c0d5658082468b569d039f4d0dc24f92c66b Mon Sep 17 00:00:00 2001 From: Jiri Pinkava Date: Tue, 25 May 2010 09:48:58 +0200 Subject: ARM: SAMSUNG: MMC: fix build error when both DMA and PIO mode selected [cjb: fix line-wrapped patch] Signed-off-by: Jiri Pinkava Signed-off-by: Chris Ball Cc: Matt Fleming Cc: Russell King Cc: Ben Dooks Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/mmc/host/s3cmci.c b/drivers/mmc/host/s3cmci.c index 2e16e0a..976330d 100644 --- a/drivers/mmc/host/s3cmci.c +++ b/drivers/mmc/host/s3cmci.c @@ -1600,7 +1600,7 @@ static int __devinit s3cmci_probe(struct platform_device *pdev) host->pio_active = XFER_NONE; #ifdef CONFIG_MMC_S3C_PIODMA - host->dodma = host->pdata->dma; + host->dodma = host->pdata->use_dma; #endif host->mem = platform_get_resource(pdev, IORESOURCE_MEM, 0); -- cgit v0.10.2 From 16d9b130783c54c30cab80e24810ab1ab9596e11 Mon Sep 17 00:00:00 2001 From: Sergio Aguirre Date: Thu, 9 Sep 2010 16:37:46 -0700 Subject: omap_hsmmc: remove unused local `state' This fixes the following warning: drivers/mmc/host/omap_hsmmc.c: In function 'omap_hsmmc_suspend': drivers/mmc/host/omap_hsmmc.c:2275: warning: unused variable 'state' Introduced by commit ID: commit 1a13f8fa76c880be41d6b1e6a2b44404bcbfdf9e Author: Matt Fleming Date: Wed May 26 14:42:08 2010 -0700 mmc: remove the "state" argument to mmc_suspend_host() The unique usage of this var was removed there, and missed removing the respective declaration aswell. Signed-off-by: Sergio Aguirre Signed-off-by: Chris Ball Acked-by: Matt Fleming Cc: Madhusudhan Chikkature Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c index 4a8776f..fb6446a 100644 --- a/drivers/mmc/host/omap_hsmmc.c +++ b/drivers/mmc/host/omap_hsmmc.c @@ -2305,7 +2305,6 @@ static int omap_hsmmc_suspend(struct device *dev) int ret = 0; struct platform_device *pdev = to_platform_device(dev); struct omap_hsmmc_host *host = platform_get_drvdata(pdev); - pm_message_t state = PMSG_SUSPEND; /* unused by MMC core */ if (host && host->suspended) return 0; -- cgit v0.10.2 From 23ef309a6e070490da0a37b9b6383819f8170ea3 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Thu, 9 Sep 2010 16:37:48 -0700 Subject: mmc: at91_mci: add missing linux/highmem.h include Fix the following error: at91_mci.c: In function 'at91_mci_sg_to_dma': at91_mci.c:236: error: implicit declaration of function 'kmap_atomic' at91_mci.c:236: error: 'KM_BIO_SRC_IRQ' undeclared (first use in this function) at91_mci.c:236: error: (Each undeclared identifier is reported only once at91_mci.c:236: error: for each function it appears in.) at91_mci.c:236: warning: assignment makes pointer from integer without a cast at91_mci.c:252: error: implicit declaration of function 'kunmap_atomic' at91_mci.c: In function 'at91_mci_post_dma_read': at91_mci.c:302: error: 'KM_BIO_SRC_IRQ' undeclared (first use in this function) at91_mci.c:302: warning: assignment makes pointer from integer without a cast at91_mci.c:317: error: implicit declaration of function 'flush_kernel_dcache_page' Signed-off-by: Marc Kleine-Budde Signed-off-by: Chris Ball Cc: Nicolas Ferre Cc: Andrew Victor Cc: Wolfgang Muees Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/mmc/host/at91_mci.c b/drivers/mmc/host/at91_mci.c index 5f3a599..87226cd 100644 --- a/drivers/mmc/host/at91_mci.c +++ b/drivers/mmc/host/at91_mci.c @@ -66,6 +66,7 @@ #include #include #include +#include #include -- cgit v0.10.2 From e7cb756fc3c7c32040283963572258381b342dff Mon Sep 17 00:00:00 2001 From: Ethan Du Date: Thu, 9 Sep 2010 16:37:49 -0700 Subject: omap hsmmc: fix a racing case between kmmcd and omap_hsmmc_suspend If suspend called when kmmcd is doing host->ops->disable, as kmmcd already increased host->en_dis_recurs to 1, the mmc_host_enable in suspend function will return directly without increase the nesting_cnt, which will cause the followed register access carried out to the disabled host. mmc_suspend_host will enable host itself. No need to enable host before it. Also works on kmmcd will get flushed in mmc_suspend_host, enable host after it will be safe. So make the mmc_host_enable after it. [cjb: rebase against current Linus] Signed-off-by: Ethan Signed-off-by: Chris Ball Acked-by: Adrian Hunter Acked-by: Madhusudhan Chikkature Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c index fb6446a..4526d27 100644 --- a/drivers/mmc/host/omap_hsmmc.c +++ b/drivers/mmc/host/omap_hsmmc.c @@ -2323,8 +2323,8 @@ static int omap_hsmmc_suspend(struct device *dev) } } cancel_work_sync(&host->mmc_carddetect_work); - mmc_host_enable(host->mmc); ret = mmc_suspend_host(host->mmc); + mmc_host_enable(host->mmc); if (ret == 0) { omap_hsmmc_disable_irq(host); OMAP_HSMMC_WRITE(host->base, HCTL, -- cgit v0.10.2 From 7c5367f205f7d53659fb19b9fdf65b7bc1a592c6 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Thu, 9 Sep 2010 16:37:50 -0700 Subject: drivers/mmc/host/imxmmc.c: adjust confusing if indentation Move the second if (reg & ...) test into the branch indicated by its indentation. The test was previously always executed after the if containing that branch, but it was always false unless the if branch was taken. The semantic match that finds this problem is as follows: (http://coccinelle.lip6.fr/) // @r disable braces4@ position p1,p2; statement S1,S2; @@ ( if (...) { ... } | if (...) S1@p1 S2@p2 ) @script:python@ p1 << r.p1; p2 << r.p2; @@ if (p1[0].column == p2[0].column): cocci.print_main("branch",p1) cocci.print_secs("after",p2) // Signed-off-by: Julia Lawall Signed-off-by: Chris Ball Cc: Pavel Pisa Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/mmc/host/imxmmc.c b/drivers/mmc/host/imxmmc.c index 9a68ff4..5a950b1 100644 --- a/drivers/mmc/host/imxmmc.c +++ b/drivers/mmc/host/imxmmc.c @@ -148,11 +148,12 @@ static int imxmci_start_clock(struct imxmci_host *host) while (delay--) { reg = readw(host->base + MMC_REG_STATUS); - if (reg & STATUS_CARD_BUS_CLK_RUN) + if (reg & STATUS_CARD_BUS_CLK_RUN) { /* Check twice before cut */ reg = readw(host->base + MMC_REG_STATUS); if (reg & STATUS_CARD_BUS_CLK_RUN) return 0; + } if (test_bit(IMXMCI_PEND_STARTED_b, &host->pending_events)) return 0; -- cgit v0.10.2 From 4969c1192d15afa3389e7ae3302096ff684ba655 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 9 Sep 2010 16:37:52 -0700 Subject: mm: fix swapin race condition The pte_same check is reliable only if the swap entry remains pinned (by the page lock on swapcache). We've also to ensure the swapcache isn't removed before we take the lock as try_to_free_swap won't care about the page pin. One of the possible impacts of this patch is that a KSM-shared page can point to the anon_vma of another process, which could exit before the page is freed. This can leave a page with a pointer to a recycled anon_vma object, or worse, a pointer to something that is no longer an anon_vma. [riel@redhat.com: changelog help] Signed-off-by: Andrea Arcangeli Acked-by: Hugh Dickins Reviewed-by: Rik van Riel Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/ksm.h b/include/linux/ksm.h index 74d691e..3319a69 100644 --- a/include/linux/ksm.h +++ b/include/linux/ksm.h @@ -16,6 +16,9 @@ struct stable_node; struct mem_cgroup; +struct page *ksm_does_need_to_copy(struct page *page, + struct vm_area_struct *vma, unsigned long address); + #ifdef CONFIG_KSM int ksm_madvise(struct vm_area_struct *vma, unsigned long start, unsigned long end, int advice, unsigned long *vm_flags); @@ -70,19 +73,14 @@ static inline void set_page_stable_node(struct page *page, * We'd like to make this conditional on vma->vm_flags & VM_MERGEABLE, * but what if the vma was unmerged while the page was swapped out? */ -struct page *ksm_does_need_to_copy(struct page *page, - struct vm_area_struct *vma, unsigned long address); -static inline struct page *ksm_might_need_to_copy(struct page *page, +static inline int ksm_might_need_to_copy(struct page *page, struct vm_area_struct *vma, unsigned long address) { struct anon_vma *anon_vma = page_anon_vma(page); - if (!anon_vma || - (anon_vma->root == vma->anon_vma->root && - page->index == linear_page_index(vma, address))) - return page; - - return ksm_does_need_to_copy(page, vma, address); + return anon_vma && + (anon_vma->root != vma->anon_vma->root || + page->index != linear_page_index(vma, address)); } int page_referenced_ksm(struct page *page, @@ -115,10 +113,10 @@ static inline int ksm_madvise(struct vm_area_struct *vma, unsigned long start, return 0; } -static inline struct page *ksm_might_need_to_copy(struct page *page, +static inline int ksm_might_need_to_copy(struct page *page, struct vm_area_struct *vma, unsigned long address) { - return page; + return 0; } static inline int page_referenced_ksm(struct page *page, diff --git a/mm/ksm.c b/mm/ksm.c index e2ae004..b1873cf 100644 --- a/mm/ksm.c +++ b/mm/ksm.c @@ -1504,8 +1504,6 @@ struct page *ksm_does_need_to_copy(struct page *page, { struct page *new_page; - unlock_page(page); /* any racers will COW it, not modify it */ - new_page = alloc_page_vma(GFP_HIGHUSER_MOVABLE, vma, address); if (new_page) { copy_user_highpage(new_page, page, address, vma); @@ -1521,7 +1519,6 @@ struct page *ksm_does_need_to_copy(struct page *page, add_page_to_unevictable_list(new_page); } - page_cache_release(page); return new_page; } diff --git a/mm/memory.c b/mm/memory.c index 6b2ab10..71b161b 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -2623,7 +2623,7 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma, unsigned int flags, pte_t orig_pte) { spinlock_t *ptl; - struct page *page; + struct page *page, *swapcache = NULL; swp_entry_t entry; pte_t pte; struct mem_cgroup *ptr = NULL; @@ -2679,10 +2679,23 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma, lock_page(page); delayacct_clear_flag(DELAYACCT_PF_SWAPIN); - page = ksm_might_need_to_copy(page, vma, address); - if (!page) { - ret = VM_FAULT_OOM; - goto out; + /* + * Make sure try_to_free_swap didn't release the swapcache + * from under us. The page pin isn't enough to prevent that. + */ + if (unlikely(!PageSwapCache(page))) + goto out_page; + + if (ksm_might_need_to_copy(page, vma, address)) { + swapcache = page; + page = ksm_does_need_to_copy(page, vma, address); + + if (unlikely(!page)) { + ret = VM_FAULT_OOM; + page = swapcache; + swapcache = NULL; + goto out_page; + } } if (mem_cgroup_try_charge_swapin(mm, page, GFP_KERNEL, &ptr)) { @@ -2735,6 +2748,18 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma, if (vm_swap_full() || (vma->vm_flags & VM_LOCKED) || PageMlocked(page)) try_to_free_swap(page); unlock_page(page); + if (swapcache) { + /* + * Hold the lock to avoid the swap entry to be reused + * until we take the PT lock for the pte_same() check + * (to avoid false positives from pte_same). For + * further safety release the lock after the swap_free + * so that the swap count won't change under a + * parallel locked swapcache. + */ + unlock_page(swapcache); + page_cache_release(swapcache); + } if (flags & FAULT_FLAG_WRITE) { ret |= do_wp_page(mm, vma, address, page_table, pmd, ptl, pte); @@ -2756,6 +2781,10 @@ out_page: unlock_page(page); out_release: page_cache_release(page); + if (swapcache) { + unlock_page(swapcache); + page_cache_release(swapcache); + } return ret; } -- cgit v0.10.2 From 152e0659fc001029c70fa4373af1792b1ae0d01c Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Thu, 9 Sep 2010 16:37:53 -0700 Subject: mm: avoid warning when COMPACTION is selected COMPACTION enables MIGRATION, but MIGRATION spawns a warning if numa or memhotplug aren't selected. However MIGRATION doesn't depend on them. I guess it's just trying to be strict doing a double check on who's enabling it, but it doesn't know that compaction also enables MIGRATION. Signed-off-by: Andrea Arcangeli Acked-by: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/Kconfig b/mm/Kconfig index f4e516e..f0fb912 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -189,7 +189,7 @@ config COMPACTION config MIGRATION bool "Page migration" def_bool y - depends on NUMA || ARCH_ENABLE_MEMORY_HOTREMOVE + depends on NUMA || ARCH_ENABLE_MEMORY_HOTREMOVE || COMPACTION help Allows the migration of the physical location of pages of processes while the virtual addresses are not changed. This is useful in -- cgit v0.10.2 From ee3aebdd8f5f8eac41c25c80ceee3d728f920f3b Mon Sep 17 00:00:00 2001 From: Jan Sembera Date: Thu, 9 Sep 2010 16:37:54 -0700 Subject: binfmt_misc: fix binfmt_misc priority Commit 74641f584da ("alpha: binfmt_aout fix") (May 2009) introduced a regression - binfmt_misc is now consulted after binfmt_elf, which will unfortunately break ia32el. ia32 ELF binaries on ia64 used to be matched using binfmt_misc and executed using wrapper. As 32bit binaries are now matched by binfmt_elf before bindmt_misc kicks in, the wrapper is ignored. The fix increases precedence of binfmt_misc to the original state. Signed-off-by: Jan Sembera Cc: Ivan Kokshaysky Cc: Al Viro Cc: Richard Henderson [2.6.everything.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index a7528b9..fd0cc0b 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -724,7 +724,7 @@ static int __init init_misc_binfmt(void) { int err = register_filesystem(&bm_fs_type); if (!err) { - err = register_binfmt(&misc_format); + err = insert_binfmt(&misc_format); if (err) unregister_filesystem(&bm_fs_type); } -- cgit v0.10.2 From 408929bed7841686ce5fdd06366fb652cb653d6c Mon Sep 17 00:00:00 2001 From: Atsushi Nemoto Date: Thu, 9 Sep 2010 16:37:56 -0700 Subject: rtc: m41t80: do not use rtc_valid_tm in m41t80_rtc_read_alarm Commit b485fe5ea ("rtc/m41t80: use rtc_valid_tm() to check returned tm") added rtc_valid_tm to m41t80_rtc_read_alarm() but it was wrong while the t->time does not contain complete date/time. This patch also fixes a warning: warning: passing argument 1 of 'rtc_valid_tm' from incompatible pointer type Signed-off-by: Atsushi Nemoto Cc: Wan ZongShun Cc: Alessandro Zummo Cc: Geert Uytterhoeven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-m41t80.c b/drivers/rtc/rtc-m41t80.c index 66377f3..d60557c 100644 --- a/drivers/rtc/rtc-m41t80.c +++ b/drivers/rtc/rtc-m41t80.c @@ -364,7 +364,7 @@ static int m41t80_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *t) t->time.tm_isdst = -1; t->enabled = !!(reg[M41T80_REG_ALARM_MON] & M41T80_ALMON_AFE); t->pending = !!(reg[M41T80_REG_FLAGS] & M41T80_FLAGS_AF); - return rtc_valid_tm(t); + return 0; } static struct rtc_class_ops m41t80_rtc_ops = { -- cgit v0.10.2 From 94131e174fedd9f3f9bb148cee4be12f2d46d68e Mon Sep 17 00:00:00 2001 From: "Ira W. Snyder" Date: Thu, 9 Sep 2010 16:37:58 -0700 Subject: arch/powerpc/include/asm/fsldma.h needs slab.h The slab.h header is required to use the kmalloc() family of functions. Due to recent kernel changes, this header must be directly included by code that calls into the memory allocator. Without this patch, any code which includes this header fails to build. Signed-off-by: Ira W. Snyder Cc: Dan Williams Cc: Benjamin Herrenschmidt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/arch/powerpc/include/asm/fsldma.h b/arch/powerpc/include/asm/fsldma.h index a67aeed..debc5ed 100644 --- a/arch/powerpc/include/asm/fsldma.h +++ b/arch/powerpc/include/asm/fsldma.h @@ -11,6 +11,7 @@ #ifndef __ARCH_POWERPC_ASM_FSLDMA_H__ #define __ARCH_POWERPC_ASM_FSLDMA_H__ +#include #include /* -- cgit v0.10.2 From 1c24de60e50fb19b94d94225458da17c720f0729 Mon Sep 17 00:00:00 2001 From: Jerome Marchand Date: Thu, 9 Sep 2010 16:37:59 -0700 Subject: kernel/groups.c: fix integer overflow in groups_search gid_t is a unsigned int. If group_info contains a gid greater than MAX_INT, groups_search() function may look on the wrong side of the search tree. This solves some unfair "permission denied" problems. Signed-off-by: Jerome Marchand Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/kernel/groups.c b/kernel/groups.c index 53b1916..253dc0f 100644 --- a/kernel/groups.c +++ b/kernel/groups.c @@ -143,10 +143,9 @@ int groups_search(const struct group_info *group_info, gid_t grp) right = group_info->ngroups; while (left < right) { unsigned int mid = (left+right)/2; - int cmp = grp - GROUP_AT(group_info, mid); - if (cmp > 0) + if (grp > GROUP_AT(group_info, mid)) left = mid + 1; - else if (cmp < 0) + else if (grp < GROUP_AT(group_info, mid)) right = mid; else return 1; -- cgit v0.10.2 From bc6930457460788e14b2c0808ed4632a1592bd61 Mon Sep 17 00:00:00 2001 From: Minchan Kim Date: Thu, 9 Sep 2010 16:38:00 -0700 Subject: mm: compaction: handle active and inactive fairly in too_many_isolated Iram reported that compaction's too_many_isolated() loops forever. (http://www.spinics.net/lists/linux-mm/msg08123.html) The meminfo when the situation happened was inactive anon is zero. That's because the system has no memory pressure until then. While all anon pages were in the active lru, compaction could select active lru as well as inactive lru. That's a different thing from vmscan's isolated. So we has been two too_many_isolated. While compaction can isolate pages in both active and inactive, current implementation of too_many_isolated only considers inactive. It made Iram's problem. This patch handles active and inactive fairly. That's because we can't expect where from and how many compaction would isolated pages. This patch changes (nr_isolated > nr_inactive) with nr_isolated > (nr_active + nr_inactive) / 2. Signed-off-by: Minchan Kim Reported-by: Iram Shahzad Acked-by: Mel Gorman Acked-by: Wu Fengguang Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/compaction.c b/mm/compaction.c index 94cce51..4d709ee 100644 --- a/mm/compaction.c +++ b/mm/compaction.c @@ -214,15 +214,16 @@ static void acct_isolated(struct zone *zone, struct compact_control *cc) /* Similar to reclaim, but different enough that they don't share logic */ static bool too_many_isolated(struct zone *zone) { - - unsigned long inactive, isolated; + unsigned long active, inactive, isolated; inactive = zone_page_state(zone, NR_INACTIVE_FILE) + zone_page_state(zone, NR_INACTIVE_ANON); + active = zone_page_state(zone, NR_ACTIVE_FILE) + + zone_page_state(zone, NR_ACTIVE_ANON); isolated = zone_page_state(zone, NR_ISOLATED_FILE) + zone_page_state(zone, NR_ISOLATED_ANON); - return isolated > inactive; + return isolated > (inactive + active) / 2; } /* -- cgit v0.10.2 From 0dcc48c15f63ee86c2fcd33968b08d651f0360a5 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Thu, 9 Sep 2010 16:38:01 -0700 Subject: memory hotplug: fix next block calculation in is_removable next_active_pageblock() is for finding next _used_ freeblock. It skips several blocks when it finds there are a chunk of free pages lager than pageblock. But it has 2 bugs. 1. We have no lock. page_order(page) - pageblock_order can be minus. 2. pageblocks_stride += is wrong. it should skip page_order(p) of pages. Signed-off-by: KAMEZAWA Hiroyuki Cc: Michal Hocko Cc: Wu Fengguang Cc: Mel Gorman Cc: KAMEZAWA Hiroyuki Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index a4cfcdc..dd186c1 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -584,19 +584,19 @@ static inline int pageblock_free(struct page *page) /* Return the start of the next active pageblock after a given page */ static struct page *next_active_pageblock(struct page *page) { - int pageblocks_stride; - /* Ensure the starting page is pageblock-aligned */ BUG_ON(page_to_pfn(page) & (pageblock_nr_pages - 1)); - /* Move forward by at least 1 * pageblock_nr_pages */ - pageblocks_stride = 1; - /* If the entire pageblock is free, move to the end of free page */ - if (pageblock_free(page)) - pageblocks_stride += page_order(page) - pageblock_order; + if (pageblock_free(page)) { + int order; + /* be careful. we don't have locks, page_order can be changed.*/ + order = page_order(page); + if ((order < MAX_ORDER) && (order >= pageblock_order)) + return page + (1 << order); + } - return page + (pageblocks_stride * pageblock_nr_pages); + return page + pageblock_nr_pages; } /* Checks if this range of memory is likely to be hot-removable. */ -- cgit v0.10.2 From 5affb607720d734ca572b8a77c5c7d62d3042b6f Mon Sep 17 00:00:00 2001 From: Gregory Bean Date: Thu, 9 Sep 2010 16:38:02 -0700 Subject: gpio: sx150x: correct and refine reset-on-probe behavior Replace the arbitrary software-reset call from the device-probe method, because: - It is defective. To work correctly, it should be two byte writes, not a single word write. As it stands, it does nothing. - Some devices with sx150x expanders installed have their NRESET pins ganged on the same line, so resetting one causes the others to reset - not a nice thing to do arbitrarily! - The probe, usually taking place at boot, implies a recent hard-reset, so a software reset at this point is just a waste of energy anyway. Therefore, make it optional, defaulting to off, as this will match the common case of probing at powerup and also matches the current broken no-op behavior. Signed-off-by: Gregory Bean Reviewed-by: Jean Delvare Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/gpio/sx150x.c b/drivers/gpio/sx150x.c index b42f42c..823559a 100644 --- a/drivers/gpio/sx150x.c +++ b/drivers/gpio/sx150x.c @@ -459,17 +459,33 @@ static int sx150x_init_io(struct sx150x_chip *chip, u8 base, u16 cfg) return err; } -static int sx150x_init_hw(struct sx150x_chip *chip, - struct sx150x_platform_data *pdata) +static int sx150x_reset(struct sx150x_chip *chip) { - int err = 0; + int err; - err = i2c_smbus_write_word_data(chip->client, + err = i2c_smbus_write_byte_data(chip->client, chip->dev_cfg->reg_reset, - 0x3412); + 0x12); if (err < 0) return err; + err = i2c_smbus_write_byte_data(chip->client, + chip->dev_cfg->reg_reset, + 0x34); + return err; +} + +static int sx150x_init_hw(struct sx150x_chip *chip, + struct sx150x_platform_data *pdata) +{ + int err = 0; + + if (pdata->reset_during_probe) { + err = sx150x_reset(chip); + if (err < 0) + return err; + } + err = sx150x_i2c_write(chip->client, chip->dev_cfg->reg_misc, 0x01); diff --git a/include/linux/i2c/sx150x.h b/include/linux/i2c/sx150x.h index ee3049c..52baa79 100644 --- a/include/linux/i2c/sx150x.h +++ b/include/linux/i2c/sx150x.h @@ -63,6 +63,9 @@ * IRQ lines will appear. Similarly to gpio_base, the expander * will create a block of irqs beginning at this number. * This value is ignored if irq_summary is < 0. + * @reset_during_probe: If set to true, the driver will trigger a full + * reset of the chip at the beginning of the probe + * in order to place it in a known state. */ struct sx150x_platform_data { unsigned gpio_base; @@ -73,6 +76,7 @@ struct sx150x_platform_data { u16 io_polarity; int irq_summary; unsigned irq_base; + bool reset_during_probe; }; #endif /* __LINUX_I2C_SX150X_H */ -- cgit v0.10.2 From c956126c137d97acb6f4d56fa9572d0bcc84e4ed Mon Sep 17 00:00:00 2001 From: David Brownell Date: Thu, 9 Sep 2010 16:38:03 -0700 Subject: gpio: doc updates There's been some recent confusion about error checking GPIO numbers. briefly, it should be handled mostly during setup, when gpio_request() is called, and NEVER by expectig gpio_is_valid to report more than never-usable GPIO numbers. [akpm@linux-foundation.org: terminate unterminated comment] Signed-off-by: David Brownell Cc: Eric Miao" Cc: "Ryan Mallon" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/Documentation/gpio.txt b/Documentation/gpio.txt index d96a6db..9633da0 100644 --- a/Documentation/gpio.txt +++ b/Documentation/gpio.txt @@ -109,17 +109,19 @@ use numbers 2000-2063 to identify GPIOs in a bank of I2C GPIO expanders. If you want to initialize a structure with an invalid GPIO number, use some negative number (perhaps "-EINVAL"); that will never be valid. To -test if a number could reference a GPIO, you may use this predicate: +test if such number from such a structure could reference a GPIO, you +may use this predicate: int gpio_is_valid(int number); A number that's not valid will be rejected by calls which may request or free GPIOs (see below). Other numbers may also be rejected; for -example, a number might be valid but unused on a given board. - -Whether a platform supports multiple GPIO controllers is currently a -platform-specific implementation issue. +example, a number might be valid but temporarily unused on a given board. +Whether a platform supports multiple GPIO controllers is a platform-specific +implementation issue, as are whether that support can leave "holes" in the space +of GPIO numbers, and whether new controllers can be added at runtime. Such issues +can affect things including whether adjacent GPIO numbers are both valid. Using GPIOs ----------- @@ -480,12 +482,16 @@ To support this framework, a platform's Kconfig will "select" either ARCH_REQUIRE_GPIOLIB or ARCH_WANT_OPTIONAL_GPIOLIB and arrange that its includes and defines three functions: gpio_get_value(), gpio_set_value(), and gpio_cansleep(). -They may also want to provide a custom value for ARCH_NR_GPIOS. -ARCH_REQUIRE_GPIOLIB means that the gpio-lib code will always get compiled +It may also provide a custom value for ARCH_NR_GPIOS, so that it better +reflects the number of GPIOs in actual use on that platform, without +wasting static table space. (It should count both built-in/SoC GPIOs and +also ones on GPIO expanders. + +ARCH_REQUIRE_GPIOLIB means that the gpiolib code will always get compiled into the kernel on that architecture. -ARCH_WANT_OPTIONAL_GPIOLIB means the gpio-lib code defaults to off and the user +ARCH_WANT_OPTIONAL_GPIOLIB means the gpiolib code defaults to off and the user can enable it and build it into the kernel optionally. If neither of these options are selected, the platform does not support diff --git a/include/asm-generic/gpio.h b/include/asm-generic/gpio.h index c7376bf..8ca18e2 100644 --- a/include/asm-generic/gpio.h +++ b/include/asm-generic/gpio.h @@ -16,15 +16,27 @@ * While the GPIO programming interface defines valid GPIO numbers * to be in the range 0..MAX_INT, this library restricts them to the * smaller range 0..ARCH_NR_GPIOS-1. + * + * ARCH_NR_GPIOS is somewhat arbitrary; it usually reflects the sum of + * builtin/SoC GPIOs plus a number of GPIOs on expanders; the latter is + * actually an estimate of a board-specific value. */ #ifndef ARCH_NR_GPIOS #define ARCH_NR_GPIOS 256 #endif +/* + * "valid" GPIO numbers are nonnegative and may be passed to + * setup routines like gpio_request(). only some valid numbers + * can successfully be requested and used. + * + * Invalid GPIO numbers are useful for indicating no-such-GPIO in + * platform data and other tables. + */ + static inline int gpio_is_valid(int number) { - /* only some non-negative numbers are valid */ return ((unsigned)number) < ARCH_NR_GPIOS; } -- cgit v0.10.2 From b7bbbf354ed0edbbffdc70c9c17f1a5d4d78204c Mon Sep 17 00:00:00 2001 From: Wolfram Sang Date: Thu, 9 Sep 2010 16:38:04 -0700 Subject: MAINTAINERS: correct entry for legacy RTC-driver Because no one dared to remove it so far, let's keep the entry correct, at least. Signed-off-by: Wolfram Sang Acked-by: Mike Frysinger Cc: Paul Gortmaker Cc: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/MAINTAINERS b/MAINTAINERS index a3da87f..9800de5 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4838,12 +4838,10 @@ F: kernel/rcu* F: kernel/srcu* X: kernel/rcutorture.c -REAL TIME CLOCK DRIVER +REAL TIME CLOCK DRIVER (LEGACY) M: Paul Gortmaker S: Maintained -F: Documentation/rtc.txt -F: drivers/rtc/ -F: include/linux/rtc.h +F: drivers/char/rtc.c REAL TIME CLOCK (RTC) SUBSYSTEM M: Alessandro Zummo -- cgit v0.10.2 From 47016434257b90445113eed1c5b8b57eb2d35330 Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Thu, 9 Sep 2010 16:38:04 -0700 Subject: drivers/rtc/rtc-pl031.c: do not mark PL031 IRQ as shared It was a mistake to mark the PL031 IRQ as shared (for the U8500), we misread the datasheet. Get rid of this. Signed-off-by: Linus Walleij Cc: Jonas Aberg Cc: Mian Yousaf Kaukab Acked-by: Alessandro Zummo Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/drivers/rtc/rtc-pl031.c b/drivers/rtc/rtc-pl031.c index 6c418fe..b7a6690 100644 --- a/drivers/rtc/rtc-pl031.c +++ b/drivers/rtc/rtc-pl031.c @@ -403,7 +403,7 @@ static int pl031_probe(struct amba_device *adev, struct amba_id *id) } if (request_irq(adev->irq[0], pl031_interrupt, - IRQF_DISABLED | IRQF_SHARED, "rtc-pl031", ldata)) { + IRQF_DISABLED, "rtc-pl031", ldata)) { ret = -EIO; goto out_no_irq; } -- cgit v0.10.2 From ac8456d6f9a3011c824176bd6084d39e5f70a382 Mon Sep 17 00:00:00 2001 From: Gary King Date: Thu, 9 Sep 2010 16:38:05 -0700 Subject: bounce: call flush_dcache_page() after bounce_copy_vec() I have been seeing problems on Tegra 2 (ARMv7 SMP) systems with HIGHMEM enabled on 2.6.35 (plus some patches targetted at 2.6.36 to perform cache maintenance lazily), and the root cause appears to be that the mm bouncing code is calling flush_dcache_page before it copies the bounce buffer into the bio. The bounced page needs to be flushed after data is copied into it, to ensure that architecture implementations can synchronize instruction and data caches if necessary. Signed-off-by: Gary King Cc: Tejun Heo Cc: Russell King Acked-by: Jens Axboe Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/bounce.c b/mm/bounce.c index 13b6dad..1481de6 100644 --- a/mm/bounce.c +++ b/mm/bounce.c @@ -116,8 +116,8 @@ static void copy_to_high_bio_irq(struct bio *to, struct bio *from) */ vfrom = page_address(fromvec->bv_page) + tovec->bv_offset; - flush_dcache_page(tovec->bv_page); bounce_copy_vec(tovec, vfrom); + flush_dcache_page(tovec->bv_page); } } -- cgit v0.10.2 From 910321ea817a202ff70fac666e37e2c8e2f88823 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Thu, 9 Sep 2010 16:38:07 -0700 Subject: swap: revert special hibernation allocation Please revert 2.6.36-rc commit d2997b1042ec150616c1963b5e5e919ffd0b0ebf "hibernation: freeze swap at hibernation". It complicated matters by adding a second swap allocation path, just for hibernation; without in any way fixing the issue that it was intended to address - page reclaim after fixing the hibernation image might free swap from a page already imaged as swapcache, letting its swap be reallocated to store a different page of the image: resulting in data corruption if the imaged page were freed as clean then swapped back in. Pages freed to si->swap_map were still in danger of being reallocated by the alternative allocation path. I guess it inadvertently fixed slow SSD swap allocation for hibernation, as reported by Nigel Cunningham: by missing out the discards that occur on the usual swap allocation path; but that was unintentional, and needs a separate fix. Signed-off-by: Hugh Dickins Cc: KAMEZAWA Hiroyuki Cc: KOSAKI Motohiro Cc: "Rafael J. Wysocki" Cc: Ondrej Zary Cc: Andrea Gelmini Cc: Balbir Singh Cc: Andrea Arcangeli Cc: Nigel Cunningham Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/swap.h b/include/linux/swap.h index 2fee51a..bf4eb62 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -315,6 +315,7 @@ extern long nr_swap_pages; extern long total_swap_pages; extern void si_swapinfo(struct sysinfo *); extern swp_entry_t get_swap_page(void); +extern swp_entry_t get_swap_page_of_type(int); extern int valid_swaphandles(swp_entry_t, unsigned long *); extern int add_swap_count_continuation(swp_entry_t, gfp_t); extern void swap_shmem_alloc(swp_entry_t); @@ -331,13 +332,6 @@ extern int reuse_swap_page(struct page *); extern int try_to_free_swap(struct page *); struct backing_dev_info; -#ifdef CONFIG_HIBERNATION -void hibernation_freeze_swap(void); -void hibernation_thaw_swap(void); -swp_entry_t get_swap_for_hibernation(int type); -void swap_free_for_hibernation(swp_entry_t val); -#endif - /* linux/mm/thrash.c */ extern struct mm_struct *swap_token_mm; extern void grab_swap_token(struct mm_struct *); diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c index c779639..8dc31e0 100644 --- a/kernel/power/hibernate.c +++ b/kernel/power/hibernate.c @@ -338,7 +338,6 @@ int hibernation_snapshot(int platform_mode) goto Close; suspend_console(); - hibernation_freeze_swap(); saved_mask = clear_gfp_allowed_mask(GFP_IOFS); error = dpm_suspend_start(PMSG_FREEZE); if (error) diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 5e7edfb..f6cd6fa 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -1086,7 +1086,6 @@ void swsusp_free(void) buffer = NULL; alloc_normal = 0; alloc_highmem = 0; - hibernation_thaw_swap(); } /* Helper functions used for the shrinking of memory. */ diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 5d0059e..e6a5bdf 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -136,10 +136,10 @@ sector_t alloc_swapdev_block(int swap) { unsigned long offset; - offset = swp_offset(get_swap_for_hibernation(swap)); + offset = swp_offset(get_swap_page_of_type(swap)); if (offset) { if (swsusp_extents_insert(offset)) - swap_free_for_hibernation(swp_entry(swap, offset)); + swap_free(swp_entry(swap, offset)); else return swapdev_block(swap, offset); } @@ -163,7 +163,7 @@ void free_all_swap_pages(int swap) ext = container_of(node, struct swsusp_extent, node); rb_erase(node, &swsusp_extents); for (offset = ext->start; offset <= ext->end; offset++) - swap_free_for_hibernation(swp_entry(swap, offset)); + swap_free(swp_entry(swap, offset)); kfree(ext); } diff --git a/mm/swapfile.c b/mm/swapfile.c index 1f3f9c5..f08d165 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -47,8 +47,6 @@ long nr_swap_pages; long total_swap_pages; static int least_priority; -static bool swap_for_hibernation; - static const char Bad_file[] = "Bad swap file entry "; static const char Unused_file[] = "Unused swap file entry "; static const char Bad_offset[] = "Bad swap offset entry "; @@ -453,8 +451,6 @@ swp_entry_t get_swap_page(void) spin_lock(&swap_lock); if (nr_swap_pages <= 0) goto noswap; - if (swap_for_hibernation) - goto noswap; nr_swap_pages--; for (type = swap_list.next; type >= 0 && wrapped < 2; type = next) { @@ -487,6 +483,28 @@ noswap: return (swp_entry_t) {0}; } +/* The only caller of this function is now susupend routine */ +swp_entry_t get_swap_page_of_type(int type) +{ + struct swap_info_struct *si; + pgoff_t offset; + + spin_lock(&swap_lock); + si = swap_info[type]; + if (si && (si->flags & SWP_WRITEOK)) { + nr_swap_pages--; + /* This is called for allocating swap entry, not cache */ + offset = scan_swap_map(si, 1); + if (offset) { + spin_unlock(&swap_lock); + return swp_entry(type, offset); + } + nr_swap_pages++; + } + spin_unlock(&swap_lock); + return (swp_entry_t) {0}; +} + static struct swap_info_struct *swap_info_get(swp_entry_t entry) { struct swap_info_struct *p; @@ -746,74 +764,6 @@ int mem_cgroup_count_swap_user(swp_entry_t ent, struct page **pagep) #endif #ifdef CONFIG_HIBERNATION - -static pgoff_t hibernation_offset[MAX_SWAPFILES]; -/* - * Once hibernation starts to use swap, we freeze swap_map[]. Otherwise, - * saved swap_map[] image to the disk will be an incomplete because it's - * changing without synchronization with hibernation snap shot. - * At resume, we just make swap_for_hibernation=false. We can forget - * used maps easily. - */ -void hibernation_freeze_swap(void) -{ - int i; - - spin_lock(&swap_lock); - - printk(KERN_INFO "PM: Freeze Swap\n"); - swap_for_hibernation = true; - for (i = 0; i < MAX_SWAPFILES; i++) - hibernation_offset[i] = 1; - spin_unlock(&swap_lock); -} - -void hibernation_thaw_swap(void) -{ - spin_lock(&swap_lock); - if (swap_for_hibernation) { - printk(KERN_INFO "PM: Thaw Swap\n"); - swap_for_hibernation = false; - } - spin_unlock(&swap_lock); -} - -/* - * Because updateing swap_map[] can make not-saved-status-change, - * we use our own easy allocator. - * Please see kernel/power/swap.c, Used swaps are recorded into - * RB-tree. - */ -swp_entry_t get_swap_for_hibernation(int type) -{ - pgoff_t off; - swp_entry_t val = {0}; - struct swap_info_struct *si; - - spin_lock(&swap_lock); - - si = swap_info[type]; - if (!si || !(si->flags & SWP_WRITEOK)) - goto done; - - for (off = hibernation_offset[type]; off < si->max; ++off) { - if (!si->swap_map[off]) - break; - } - if (off < si->max) { - val = swp_entry(type, off); - hibernation_offset[type] = off + 1; - } -done: - spin_unlock(&swap_lock); - return val; -} - -void swap_free_for_hibernation(swp_entry_t ent) -{ - /* Nothing to do */ -} - /* * Find the swap type that corresponds to given device (if any). * -- cgit v0.10.2 From b73d7fcecd93dc15eaa3c45c8c587b613f6673c4 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Thu, 9 Sep 2010 16:38:09 -0700 Subject: swap: prevent reuse during hibernation Move the hibernation check from scan_swap_map() into try_to_free_swap(): to catch not only the common case when hibernation's allocation itself triggers swap reuse, but also the less likely case when concurrent page reclaim (shrink_page_list) might happen to try_to_free_swap from a page. Hibernation already clears __GFP_IO from the gfp_allowed_mask, to stop reclaim from going to swap: check that to prevent swap reuse too. Signed-off-by: Hugh Dickins Cc: KAMEZAWA Hiroyuki Cc: KOSAKI Motohiro Cc: "Rafael J. Wysocki" Cc: Ondrej Zary Cc: Andrea Gelmini Cc: Balbir Singh Cc: Andrea Arcangeli Cc: Nigel Cunningham Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/swapfile.c b/mm/swapfile.c index f08d165..ed51510 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -318,10 +318,8 @@ checks: if (offset > si->highest_bit) scan_base = offset = si->lowest_bit; - /* reuse swap entry of cache-only swap if not hibernation. */ - if (vm_swap_full() - && usage == SWAP_HAS_CACHE - && si->swap_map[offset] == SWAP_HAS_CACHE) { + /* reuse swap entry of cache-only swap if not busy. */ + if (vm_swap_full() && si->swap_map[offset] == SWAP_HAS_CACHE) { int swap_was_freed; spin_unlock(&swap_lock); swap_was_freed = __try_to_reclaim_swap(si, offset); @@ -688,6 +686,24 @@ int try_to_free_swap(struct page *page) if (page_swapcount(page)) return 0; + /* + * Once hibernation has begun to create its image of memory, + * there's a danger that one of the calls to try_to_free_swap() + * - most probably a call from __try_to_reclaim_swap() while + * hibernation is allocating its own swap pages for the image, + * but conceivably even a call from memory reclaim - will free + * the swap from a page which has already been recorded in the + * image as a clean swapcache page, and then reuse its swap for + * another page of the image. On waking from hibernation, the + * original page might be freed under memory pressure, then + * later read back in from swap, now with the wrong data. + * + * Hibernation clears bits from gfp_allowed_mask to prevent + * memory reclaim from writing to disk, so check that here. + */ + if (!(gfp_allowed_mask & __GFP_IO)) + return 0; + delete_from_swap_cache(page); SetPageDirty(page); return 1; -- cgit v0.10.2 From 8f2ae0faa3a119158c4dcfe89926d6fad5f5332c Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Thu, 9 Sep 2010 16:38:10 -0700 Subject: swap: do not send discards as barriers The swap code already uses synchronous discards, no need to add I/O barriers. This fixes the worst of the terrible slowdown in swap allocation for hibernation, reported on 2.6.35 by Nigel Cunningham; but does not entirely eliminate that regression. [tj@kernel.org: superflous newlines removed] Signed-off-by: Christoph Hellwig Tested-by: Nigel Cunningham Signed-off-by: Tejun Heo Signed-off-by: Hugh Dickins Cc: Jens Axboe Cc: James Bottomley Cc: "Martin K. Petersen" Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/swapfile.c b/mm/swapfile.c index ed51510..1894dea 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -139,8 +139,7 @@ static int discard_swap(struct swap_info_struct *si) nr_blocks = ((sector_t)se->nr_pages - 1) << (PAGE_SHIFT - 9); if (nr_blocks) { err = blkdev_issue_discard(si->bdev, start_block, - nr_blocks, GFP_KERNEL, - BLKDEV_IFL_WAIT | BLKDEV_IFL_BARRIER); + nr_blocks, GFP_KERNEL, BLKDEV_IFL_WAIT); if (err) return err; cond_resched(); @@ -151,8 +150,7 @@ static int discard_swap(struct swap_info_struct *si) nr_blocks = (sector_t)se->nr_pages << (PAGE_SHIFT - 9); err = blkdev_issue_discard(si->bdev, start_block, - nr_blocks, GFP_KERNEL, - BLKDEV_IFL_WAIT | BLKDEV_IFL_BARRIER); + nr_blocks, GFP_KERNEL, BLKDEV_IFL_WAIT); if (err) break; @@ -191,8 +189,7 @@ static void discard_swap_cluster(struct swap_info_struct *si, start_block <<= PAGE_SHIFT - 9; nr_blocks <<= PAGE_SHIFT - 9; if (blkdev_issue_discard(si->bdev, start_block, - nr_blocks, GFP_NOIO, BLKDEV_IFL_WAIT | - BLKDEV_IFL_BARRIER)) + nr_blocks, GFP_NOIO, BLKDEV_IFL_WAIT)) break; } -- cgit v0.10.2 From 3399446632739fcd05fd8b272b476a69c6e6d14a Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Thu, 9 Sep 2010 16:38:11 -0700 Subject: swap: discard while swapping only if SWAP_FLAG_DISCARD Tests with recent firmware on Intel X25-M 80GB and OCZ Vertex 60GB SSDs show a shift since I last tested in December: in part because of firmware updates, in part because of the necessary move from barriers to awaiting completion at the block layer. While discard at swapon still shows as slightly beneficial on both, discarding 1MB swap cluster when allocating is now disadvanteous: adds 25% overhead on Intel, adds 230% on OCZ (YMMV). Surrender: discard as presently implemented is more hindrance than help for swap; but might prove useful on other devices, or with improvements. So continue to do the discard at swapon, but make discard while swapping conditional on a SWAP_FLAG_DISCARD to sys_swapon() (which has been using only the lower 16 bits of int flags). We can add a --discard or -d to swapon(8), and a "discard" to swap in /etc/fstab: matching the mount option for btrfs, ext4, fat, gfs2, nilfs2. Signed-off-by: Hugh Dickins Cc: Christoph Hellwig Cc: Nigel Cunningham Cc: Tejun Heo Cc: Jens Axboe Cc: James Bottomley Cc: "Martin K. Petersen" Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/swap.h b/include/linux/swap.h index bf4eb62..7cdd633 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -19,6 +19,7 @@ struct bio; #define SWAP_FLAG_PREFER 0x8000 /* set if swap priority specified */ #define SWAP_FLAG_PRIO_MASK 0x7fff #define SWAP_FLAG_PRIO_SHIFT 0 +#define SWAP_FLAG_DISCARD 0x10000 /* discard swap cluster after use */ static inline int current_is_kswapd(void) { @@ -142,7 +143,7 @@ struct swap_extent { enum { SWP_USED = (1 << 0), /* is slot in swap_info[] used? */ SWP_WRITEOK = (1 << 1), /* ok to write to this swap? */ - SWP_DISCARDABLE = (1 << 2), /* blkdev supports discard */ + SWP_DISCARDABLE = (1 << 2), /* swapon+blkdev support discard */ SWP_DISCARDING = (1 << 3), /* now discarding a free cluster */ SWP_SOLIDSTATE = (1 << 4), /* blkdev seeks are cheap */ SWP_CONTINUED = (1 << 5), /* swap_map has count continuation */ diff --git a/mm/swapfile.c b/mm/swapfile.c index 1894dea..7c703ff 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -2047,7 +2047,7 @@ SYSCALL_DEFINE2(swapon, const char __user *, specialfile, int, swap_flags) p->flags |= SWP_SOLIDSTATE; p->cluster_next = 1 + (random32() % p->highest_bit); } - if (discard_swap(p) == 0) + if (discard_swap(p) == 0 && (swap_flags & SWAP_FLAG_DISCARD)) p->flags |= SWP_DISCARDABLE; } -- cgit v0.10.2 From 3ab04d5cf9736b7a4e9dfcf28285d8663b01aa0e Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Thu, 9 Sep 2010 16:38:12 -0700 Subject: vfs: take O_NONBLOCK out of the O_* uniqueness test O_NONBLOCK on parisc has a dual value: #define O_NONBLOCK 000200004 /* HPUX has separate NDELAY & NONBLOCK */ It is caught by the O_* bits uniqueness check and leads to a parisc compile error. The fix would be to take O_NONBLOCK out. Signed-off-by: Wu Fengguang Signed-off-by: James Bottomley Cc: Jamie Lokier Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/fcntl.c b/fs/fcntl.c index 6769fd0..f8cc34f 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -769,11 +769,15 @@ EXPORT_SYMBOL(kill_fasync); static int __init fcntl_init(void) { - /* please add new bits here to ensure allocation uniqueness */ - BUILD_BUG_ON(19 - 1 /* for O_RDONLY being 0 */ != HWEIGHT32( + /* + * Please add new bits here to ensure allocation uniqueness. + * Exceptions: O_NONBLOCK is a two bit define on parisc; O_NDELAY + * is defined as O_NONBLOCK on some platforms and not on others. + */ + BUILD_BUG_ON(18 - 1 /* for O_RDONLY being 0 */ != HWEIGHT32( O_RDONLY | O_WRONLY | O_RDWR | O_CREAT | O_EXCL | O_NOCTTY | - O_TRUNC | O_APPEND | O_NONBLOCK | + O_TRUNC | O_APPEND | /* O_NONBLOCK | */ __O_SYNC | O_DSYNC | FASYNC | O_DIRECT | O_LARGEFILE | O_DIRECTORY | O_NOFOLLOW | O_NOATIME | O_CLOEXEC | -- cgit v0.10.2 From 5ee28a447625b9fe64fbf7cff026561084fc5f16 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Thu, 9 Sep 2010 16:38:14 -0700 Subject: vmstat: update zone stat threshold when onlining a cpu refresh_zone_stat_thresholds() calculates parameter based on the number of online cpus. It's called at cpu offlining but needs to be called at onlining, too. Signed-off-by: KAMEZAWA Hiroyuki Cc: Christoph Lameter Acked-by: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/vmstat.c b/mm/vmstat.c index f389168..a8d6b59 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -998,6 +998,7 @@ static int __cpuinit vmstat_cpuup_callback(struct notifier_block *nfb, switch (action) { case CPU_ONLINE: case CPU_ONLINE_FROZEN: + refresh_zone_stat_thresholds(); start_cpu_timer(cpu); node_set_state(cpu_to_node(cpu), N_CPU); break; -- cgit v0.10.2 From 72853e2991a2702ae93aaf889ac7db743a415dd3 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Thu, 9 Sep 2010 16:38:16 -0700 Subject: mm: page allocator: update free page counters after pages are placed on the free list When allocating a page, the system uses NR_FREE_PAGES counters to determine if watermarks would remain intact after the allocation was made. This check is made without interrupts disabled or the zone lock held and so is race-prone by nature. Unfortunately, when pages are being freed in batch, the counters are updated before the pages are added on the list. During this window, the counters are misleading as the pages do not exist yet. When under significant pressure on systems with large numbers of CPUs, it's possible for processes to make progress even though they should have been stalled. This is particularly problematic if a number of the processes are using GFP_ATOMIC as the min watermark can be accidentally breached and in extreme cases, the system can livelock. This patch updates the counters after the pages have been added to the list. This makes the allocator more cautious with respect to preserving the watermarks and mitigates livelock possibilities. [akpm@linux-foundation.org: avoid modifying incoming args] Signed-off-by: Mel Gorman Reviewed-by: Rik van Riel Reviewed-by: Minchan Kim Reviewed-by: KAMEZAWA Hiroyuki Reviewed-by: Christoph Lameter Reviewed-by: KOSAKI Motohiro Acked-by: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/page_alloc.c b/mm/page_alloc.c index a9649f4..452e2ba0 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -588,13 +588,13 @@ static void free_pcppages_bulk(struct zone *zone, int count, { int migratetype = 0; int batch_free = 0; + int to_free = count; spin_lock(&zone->lock); zone->all_unreclaimable = 0; zone->pages_scanned = 0; - __mod_zone_page_state(zone, NR_FREE_PAGES, count); - while (count) { + while (to_free) { struct page *page; struct list_head *list; @@ -619,8 +619,9 @@ static void free_pcppages_bulk(struct zone *zone, int count, /* MIGRATE_MOVABLE list may include MIGRATE_RESERVEs */ __free_one_page(page, zone, 0, page_private(page)); trace_mm_page_pcpu_drain(page, 0, page_private(page)); - } while (--count && --batch_free && !list_empty(list)); + } while (--to_free && --batch_free && !list_empty(list)); } + __mod_zone_page_state(zone, NR_FREE_PAGES, count); spin_unlock(&zone->lock); } @@ -631,8 +632,8 @@ static void free_one_page(struct zone *zone, struct page *page, int order, zone->all_unreclaimable = 0; zone->pages_scanned = 0; - __mod_zone_page_state(zone, NR_FREE_PAGES, 1 << order); __free_one_page(page, zone, order, migratetype); + __mod_zone_page_state(zone, NR_FREE_PAGES, 1 << order); spin_unlock(&zone->lock); } -- cgit v0.10.2 From aa45484031ddee09b06350ab8528bfe5b2c76d1c Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Thu, 9 Sep 2010 16:38:17 -0700 Subject: mm: page allocator: calculate a better estimate of NR_FREE_PAGES when memory is low and kswapd is awake Ordinarily watermark checks are based on the vmstat NR_FREE_PAGES as it is cheaper than scanning a number of lists. To avoid synchronization overhead, counter deltas are maintained on a per-cpu basis and drained both periodically and when the delta is above a threshold. On large CPU systems, the difference between the estimated and real value of NR_FREE_PAGES can be very high. If NR_FREE_PAGES is much higher than number of real free page in buddy, the VM can allocate pages below min watermark, at worst reducing the real number of pages to zero. Even if the OOM killer kills some victim for freeing memory, it may not free memory if the exit path requires a new page resulting in livelock. This patch introduces a zone_page_state_snapshot() function (courtesy of Christoph) that takes a slightly more accurate view of an arbitrary vmstat counter. It is used to read NR_FREE_PAGES while kswapd is awake to avoid the watermark being accidentally broken. The estimate is not perfect and may result in cache line bounces but is expected to be lighter than the IPI calls necessary to continually drain the per-cpu counters while kswapd is awake. Signed-off-by: Christoph Lameter Signed-off-by: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 6e6e626..3984c4e 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -284,6 +284,13 @@ struct zone { unsigned long watermark[NR_WMARK]; /* + * When free pages are below this point, additional steps are taken + * when reading the number of free pages to avoid per-cpu counter + * drift allowing watermarks to be breached + */ + unsigned long percpu_drift_mark; + + /* * We don't know if the memory that we're going to allocate will be freeable * or/and it will be released eventually, so to avoid totally wasting several * GB of ram we must reserve some of the lower zone memory (otherwise we risk @@ -441,6 +448,12 @@ static inline int zone_is_oom_locked(const struct zone *zone) return test_bit(ZONE_OOM_LOCKED, &zone->flags); } +#ifdef CONFIG_SMP +unsigned long zone_nr_free_pages(struct zone *zone); +#else +#define zone_nr_free_pages(zone) zone_page_state(zone, NR_FREE_PAGES) +#endif /* CONFIG_SMP */ + /* * The "priority" of VM scanning is how much of the queues we will scan in one * go. A value of 12 for DEF_PRIORITY implies that we will scan 1/4096th of the diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h index 7f43ccd..eaaea37 100644 --- a/include/linux/vmstat.h +++ b/include/linux/vmstat.h @@ -170,6 +170,28 @@ static inline unsigned long zone_page_state(struct zone *zone, return x; } +/* + * More accurate version that also considers the currently pending + * deltas. For that we need to loop over all cpus to find the current + * deltas. There is no synchronization so the result cannot be + * exactly accurate either. + */ +static inline unsigned long zone_page_state_snapshot(struct zone *zone, + enum zone_stat_item item) +{ + long x = atomic_long_read(&zone->vm_stat[item]); + +#ifdef CONFIG_SMP + int cpu; + for_each_online_cpu(cpu) + x += per_cpu_ptr(zone->pageset, cpu)->vm_stat_diff[item]; + + if (x < 0) + x = 0; +#endif + return x; +} + extern unsigned long global_reclaimable_pages(void); extern unsigned long zone_reclaimable_pages(struct zone *zone); diff --git a/mm/mmzone.c b/mm/mmzone.c index f5b7d17..e35bfb8 100644 --- a/mm/mmzone.c +++ b/mm/mmzone.c @@ -87,3 +87,24 @@ int memmap_valid_within(unsigned long pfn, return 1; } #endif /* CONFIG_ARCH_HAS_HOLES_MEMORYMODEL */ + +#ifdef CONFIG_SMP +/* Called when a more accurate view of NR_FREE_PAGES is needed */ +unsigned long zone_nr_free_pages(struct zone *zone) +{ + unsigned long nr_free_pages = zone_page_state(zone, NR_FREE_PAGES); + + /* + * While kswapd is awake, it is considered the zone is under some + * memory pressure. Under pressure, there is a risk that + * per-cpu-counter-drift will allow the min watermark to be breached + * potentially causing a live-lock. While kswapd is awake and + * free pages are low, get a better estimate for free pages + */ + if (nr_free_pages < zone->percpu_drift_mark && + !waitqueue_active(&zone->zone_pgdat->kswapd_wait)) + return zone_page_state_snapshot(zone, NR_FREE_PAGES); + + return nr_free_pages; +} +#endif /* CONFIG_SMP */ diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 452e2ba0..b2d21e0 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1462,7 +1462,7 @@ int zone_watermark_ok(struct zone *z, int order, unsigned long mark, { /* free_pages my go negative - that's OK */ long min = mark; - long free_pages = zone_page_state(z, NR_FREE_PAGES) - (1 << order) + 1; + long free_pages = zone_nr_free_pages(z) - (1 << order) + 1; int o; if (alloc_flags & ALLOC_HIGH) @@ -2424,7 +2424,7 @@ void show_free_areas(void) " all_unreclaimable? %s" "\n", zone->name, - K(zone_page_state(zone, NR_FREE_PAGES)), + K(zone_nr_free_pages(zone)), K(min_wmark_pages(zone)), K(low_wmark_pages(zone)), K(high_wmark_pages(zone)), diff --git a/mm/vmstat.c b/mm/vmstat.c index a8d6b59..355a9e6 100644 --- a/mm/vmstat.c +++ b/mm/vmstat.c @@ -138,11 +138,24 @@ static void refresh_zone_stat_thresholds(void) int threshold; for_each_populated_zone(zone) { + unsigned long max_drift, tolerate_drift; + threshold = calculate_threshold(zone); for_each_online_cpu(cpu) per_cpu_ptr(zone->pageset, cpu)->stat_threshold = threshold; + + /* + * Only set percpu_drift_mark if there is a danger that + * NR_FREE_PAGES reports the low watermark is ok when in fact + * the min watermark could be breached by an allocation + */ + tolerate_drift = low_wmark_pages(zone) - min_wmark_pages(zone); + max_drift = num_online_cpus() * threshold; + if (max_drift > tolerate_drift) + zone->percpu_drift_mark = high_wmark_pages(zone) + + max_drift; } } @@ -813,7 +826,7 @@ static void zoneinfo_show_print(struct seq_file *m, pg_data_t *pgdat, "\n scanned %lu" "\n spanned %lu" "\n present %lu", - zone_page_state(zone, NR_FREE_PAGES), + zone_nr_free_pages(zone), min_wmark_pages(zone), low_wmark_pages(zone), high_wmark_pages(zone), -- cgit v0.10.2 From 9ee493ce0a60bf42c0f8fd0b0fe91df5704a1cbf Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Thu, 9 Sep 2010 16:38:18 -0700 Subject: mm: page allocator: drain per-cpu lists after direct reclaim allocation fails When under significant memory pressure, a process enters direct reclaim and immediately afterwards tries to allocate a page. If it fails and no further progress is made, it's possible the system will go OOM. However, on systems with large amounts of memory, it's possible that a significant number of pages are on per-cpu lists and inaccessible to the calling process. This leads to a process entering direct reclaim more often than it should increasing the pressure on the system and compounding the problem. This patch notes that if direct reclaim is making progress but allocations are still failing that the system is already under heavy pressure. In this case, it drains the per-cpu lists and tries the allocation a second time before continuing. Signed-off-by: Mel Gorman Reviewed-by: Minchan Kim Reviewed-by: KAMEZAWA Hiroyuki Reviewed-by: KOSAKI Motohiro Reviewed-by: Christoph Lameter Cc: Dave Chinner Cc: Wu Fengguang Cc: David Rientjes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/mm/page_alloc.c b/mm/page_alloc.c index b2d21e0..a8cfa9c 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1847,6 +1847,7 @@ __alloc_pages_direct_reclaim(gfp_t gfp_mask, unsigned int order, struct page *page = NULL; struct reclaim_state reclaim_state; struct task_struct *p = current; + bool drained = false; cond_resched(); @@ -1865,14 +1866,25 @@ __alloc_pages_direct_reclaim(gfp_t gfp_mask, unsigned int order, cond_resched(); - if (order != 0) - drain_all_pages(); + if (unlikely(!(*did_some_progress))) + return NULL; - if (likely(*did_some_progress)) - page = get_page_from_freelist(gfp_mask, nodemask, order, +retry: + page = get_page_from_freelist(gfp_mask, nodemask, order, zonelist, high_zoneidx, alloc_flags, preferred_zone, migratetype); + + /* + * If an allocation failed after direct reclaim, it could be because + * pages are pinned on the per-cpu lists. Drain them and try again + */ + if (!page && !drained) { + drain_all_pages(); + drained = true; + goto retry; + } + return page; } -- cgit v0.10.2 From eee743fd7eac9f2ea69ad06d093dfb5a12538fe5 Mon Sep 17 00:00:00 2001 From: "Jorge Boncompte [DTI2]" Date: Thu, 9 Sep 2010 16:38:19 -0700 Subject: minix: fix regression in minix_mkdir() Commit 9eed1fb721c ("minix: replace inode uid,gid,mode init with helper") broke directory creation on minix filesystems. Fix it by passing the needed mode flag to inode init helper. Signed-off-by: Jorge Boncompte [DTI2] Cc: Dmitry Monakhov Cc: Al Viro Cc: [2.6.35.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds diff --git a/fs/minix/namei.c b/fs/minix/namei.c index e20ee85..f3f3578 100644 --- a/fs/minix/namei.c +++ b/fs/minix/namei.c @@ -115,7 +115,7 @@ static int minix_mkdir(struct inode * dir, struct dentry *dentry, int mode) inode_inc_link_count(dir); - inode = minix_new_inode(dir, mode, &err); + inode = minix_new_inode(dir, S_IFDIR | mode, &err); if (!inode) goto out_dir; -- cgit v0.10.2 From 40c6023031369ae5573e622ca54fa3ffe89fb865 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Thu, 9 Sep 2010 17:13:31 +0200 Subject: libata,pata_via: revert ata_wait_idle() removal from ata_sff/via_tf_load() Commit 978c0666 (libata: Remove excess delay in the tf_load path) removed ata_wait_idle() from ata_sff_tf_load() and via_tf_load(). This caused obscure detection problems in sata_sil. https://bugzilla.kernel.org/show_bug.cgi?id=16606 The commit was pure performance optimization. Revert it for now. Reported-by: Dieter Plaetinck Reported-by: Jan Beulich Bisected-by: gianluca Cc: stable@kernel.org Signed-off-by: Jeff Garzik diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c index 3b82d8e..dee3c2c 100644 --- a/drivers/ata/libata-sff.c +++ b/drivers/ata/libata-sff.c @@ -418,6 +418,7 @@ void ata_sff_tf_load(struct ata_port *ap, const struct ata_taskfile *tf) if (ioaddr->ctl_addr) iowrite8(tf->ctl, ioaddr->ctl_addr); ap->last_ctl = tf->ctl; + ata_wait_idle(ap); } if (is_addr && (tf->flags & ATA_TFLAG_LBA48)) { @@ -453,6 +454,8 @@ void ata_sff_tf_load(struct ata_port *ap, const struct ata_taskfile *tf) iowrite8(tf->device, ioaddr->device_addr); VPRINTK("device 0x%X\n", tf->device); } + + ata_wait_idle(ap); } EXPORT_SYMBOL_GPL(ata_sff_tf_load); diff --git a/drivers/ata/pata_via.c b/drivers/ata/pata_via.c index 5e65988..ac8d7d9 100644 --- a/drivers/ata/pata_via.c +++ b/drivers/ata/pata_via.c @@ -417,6 +417,8 @@ static void via_tf_load(struct ata_port *ap, const struct ata_taskfile *tf) tf->lbam, tf->lbah); } + + ata_wait_idle(ap); } static int via_port_start(struct ata_port *ap) -- cgit v0.10.2 From 238e149c7a92eb79ab9f48c171e907a5bde18333 Mon Sep 17 00:00:00 2001 From: Seth Heasley Date: Thu, 9 Sep 2010 09:42:40 -0700 Subject: ata_piix: IDE Mode SATA patch for Intel Patsburg DeviceIDs This patch adds the Intel Patsburg (PCH) IDE mode SATA Controller DeviceIDs. Signed-off-by: Seth Heasley Signed-off-by: Jeff Garzik diff --git a/drivers/ata/ata_piix.c b/drivers/ata/ata_piix.c index 3971bc0..d712675 100644 --- a/drivers/ata/ata_piix.c +++ b/drivers/ata/ata_piix.c @@ -302,6 +302,10 @@ static const struct pci_device_id piix_pci_tbl[] = { { 0x8086, 0x1c08, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich8_2port_sata }, /* SATA Controller IDE (CPT) */ { 0x8086, 0x1c09, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich8_2port_sata }, + /* SATA Controller IDE (PBG) */ + { 0x8086, 0x1d00, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich8_sata }, + /* SATA Controller IDE (PBG) */ + { 0x8086, 0x1d08, PCI_ANY_ID, PCI_ANY_ID, 0, 0, ich8_2port_sata }, { } /* terminate list */ }; -- cgit v0.10.2 From 992b3fb9b5391bc4de5b42bb810dc6dd583a6c4a Mon Sep 17 00:00:00 2001 From: Seth Heasley Date: Thu, 9 Sep 2010 09:44:56 -0700 Subject: ahci: AHCI and RAID mode SATA patch for Intel Patsburg DeviceIDs This patch adds the Intel Patsburg (PCH) SATA AHCI and RAID Controller DeviceIDs. Signed-off-by: Seth Heasley Signed-off-by: Jeff Garzik diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c index 013727b..ff1c945 100644 --- a/drivers/ata/ahci.c +++ b/drivers/ata/ahci.c @@ -253,6 +253,9 @@ static const struct pci_device_id ahci_pci_tbl[] = { { PCI_VDEVICE(INTEL, 0x1c05), board_ahci }, /* CPT RAID */ { PCI_VDEVICE(INTEL, 0x1c06), board_ahci }, /* CPT RAID */ { PCI_VDEVICE(INTEL, 0x1c07), board_ahci }, /* CPT RAID */ + { PCI_VDEVICE(INTEL, 0x1d02), board_ahci }, /* PBG AHCI */ + { PCI_VDEVICE(INTEL, 0x1d04), board_ahci }, /* PBG RAID */ + { PCI_VDEVICE(INTEL, 0x1d06), board_ahci }, /* PBG RAID */ /* JMicron 360/1/3/5/6, match class to avoid IDE function */ { PCI_VENDOR_ID_JMICRON, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, -- cgit v0.10.2 From e2f3d75fc0e4a0d03c61872bad39ffa2e74a04ff Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Tue, 7 Sep 2010 14:05:31 +0200 Subject: libata: skip EH autopsy and recovery during suspend For some mysterious reason, certain hardware reacts badly to usual EH actions while the system is going for suspend. As the devices won't be needed until the system is resumed, ask EH to skip usual autopsy and recovery and proceed directly to suspend. Signed-off-by: Tejun Heo Tested-by: Stephan Diestelhorst Cc: stable@kernel.org Signed-off-by: Jeff Garzik diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index c035b3d..932eaee 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -5418,6 +5418,7 @@ static int ata_host_request_pm(struct ata_host *host, pm_message_t mesg, */ int ata_host_suspend(struct ata_host *host, pm_message_t mesg) { + unsigned int ehi_flags = ATA_EHI_QUIET; int rc; /* @@ -5426,7 +5427,18 @@ int ata_host_suspend(struct ata_host *host, pm_message_t mesg) */ ata_lpm_enable(host); - rc = ata_host_request_pm(host, mesg, 0, ATA_EHI_QUIET, 1); + /* + * On some hardware, device fails to respond after spun down + * for suspend. As the device won't be used before being + * resumed, we don't need to touch the device. Ask EH to skip + * the usual stuff and proceed directly to suspend. + * + * http://thread.gmane.org/gmane.linux.ide/46764 + */ + if (mesg.event == PM_EVENT_SUSPEND) + ehi_flags |= ATA_EHI_NO_AUTOPSY | ATA_EHI_NO_RECOVERY; + + rc = ata_host_request_pm(host, mesg, 0, ehi_flags, 1); if (rc == 0) host->dev->power.power_state = mesg; return rc; diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c index c9ae299..e48302e 100644 --- a/drivers/ata/libata-eh.c +++ b/drivers/ata/libata-eh.c @@ -3235,6 +3235,10 @@ static int ata_eh_skip_recovery(struct ata_link *link) if (link->flags & ATA_LFLAG_DISABLED) return 1; + /* skip if explicitly requested */ + if (ehc->i.flags & ATA_EHI_NO_RECOVERY) + return 1; + /* thaw frozen port and recover failed devices */ if ((ap->pflags & ATA_PFLAG_FROZEN) || ata_link_nr_enabled(link)) return 0; diff --git a/include/linux/libata.h b/include/linux/libata.h index f010f18..7de282d 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -335,6 +335,7 @@ enum { ATA_EHI_HOTPLUGGED = (1 << 0), /* could have been hotplugged */ ATA_EHI_NO_AUTOPSY = (1 << 2), /* no autopsy */ ATA_EHI_QUIET = (1 << 3), /* be quiet */ + ATA_EHI_NO_RECOVERY = (1 << 4), /* no recovery */ ATA_EHI_DID_SOFTRESET = (1 << 16), /* already soft-reset this port */ ATA_EHI_DID_HARDRESET = (1 << 17), /* already soft-reset this port */ -- cgit v0.10.2 From ea3c64506ea7965f86f030155e6fdef381de10e2 Mon Sep 17 00:00:00 2001 From: Gwendal Grignou Date: Tue, 31 Aug 2010 16:20:36 -0700 Subject: libata-sff: Reenable Port Multiplier after libata-sff remodeling. Keep track of the link on the which the current request is in progress. It allows support of links behind port multiplier. Not all libata-sff is PMP compliant. Code for native BMDMA controller does not take in accound PMP. Tested on Marvell 7042 and Sil7526. Signed-off-by: Gwendal Grignou Signed-off-by: Jeff Garzik diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c index dee3c2c..e30c537 100644 --- a/drivers/ata/libata-sff.c +++ b/drivers/ata/libata-sff.c @@ -1045,7 +1045,8 @@ static void ata_hsm_qc_complete(struct ata_queued_cmd *qc, int in_wq) int ata_sff_hsm_move(struct ata_port *ap, struct ata_queued_cmd *qc, u8 status, int in_wq) { - struct ata_eh_info *ehi = &ap->link.eh_info; + struct ata_link *link = qc->dev->link; + struct ata_eh_info *ehi = &link->eh_info; unsigned long flags = 0; int poll_next; @@ -1301,8 +1302,14 @@ fsm_start: } EXPORT_SYMBOL_GPL(ata_sff_hsm_move); -void ata_sff_queue_pio_task(struct ata_port *ap, unsigned long delay) +void ata_sff_queue_pio_task(struct ata_link *link, unsigned long delay) { + struct ata_port *ap = link->ap; + + WARN_ON((ap->sff_pio_task_link != NULL) && + (ap->sff_pio_task_link != link)); + ap->sff_pio_task_link = link; + /* may fail if ata_sff_flush_pio_task() in progress */ queue_delayed_work(ata_sff_wq, &ap->sff_pio_task, msecs_to_jiffies(delay)); @@ -1324,14 +1331,18 @@ static void ata_sff_pio_task(struct work_struct *work) { struct ata_port *ap = container_of(work, struct ata_port, sff_pio_task.work); + struct ata_link *link = ap->sff_pio_task_link; struct ata_queued_cmd *qc; u8 status; int poll_next; + BUG_ON(ap->sff_pio_task_link == NULL); /* qc can be NULL if timeout occurred */ - qc = ata_qc_from_tag(ap, ap->link.active_tag); - if (!qc) + qc = ata_qc_from_tag(ap, link->active_tag); + if (!qc) { + ap->sff_pio_task_link = NULL; return; + } fsm_start: WARN_ON_ONCE(ap->hsm_task_state == HSM_ST_IDLE); @@ -1348,11 +1359,16 @@ fsm_start: msleep(2); status = ata_sff_busy_wait(ap, ATA_BUSY, 10); if (status & ATA_BUSY) { - ata_sff_queue_pio_task(ap, ATA_SHORT_PAUSE); + ata_sff_queue_pio_task(link, ATA_SHORT_PAUSE); return; } } + /* + * hsm_move() may trigger another command to be processed. + * clean the link beforehand. + */ + ap->sff_pio_task_link = NULL; /* move the HSM */ poll_next = ata_sff_hsm_move(ap, qc, status, 1); @@ -1379,6 +1395,7 @@ fsm_start: unsigned int ata_sff_qc_issue(struct ata_queued_cmd *qc) { struct ata_port *ap = qc->ap; + struct ata_link *link = qc->dev->link; /* Use polling pio if the LLD doesn't handle * interrupt driven pio and atapi CDB interrupt. @@ -1399,7 +1416,7 @@ unsigned int ata_sff_qc_issue(struct ata_queued_cmd *qc) ap->hsm_task_state = HSM_ST_LAST; if (qc->tf.flags & ATA_TFLAG_POLLING) - ata_sff_queue_pio_task(ap, 0); + ata_sff_queue_pio_task(link, 0); break; @@ -1412,7 +1429,7 @@ unsigned int ata_sff_qc_issue(struct ata_queued_cmd *qc) if (qc->tf.flags & ATA_TFLAG_WRITE) { /* PIO data out protocol */ ap->hsm_task_state = HSM_ST_FIRST; - ata_sff_queue_pio_task(ap, 0); + ata_sff_queue_pio_task(link, 0); /* always send first data block using the * ata_sff_pio_task() codepath. @@ -1422,7 +1439,7 @@ unsigned int ata_sff_qc_issue(struct ata_queued_cmd *qc) ap->hsm_task_state = HSM_ST; if (qc->tf.flags & ATA_TFLAG_POLLING) - ata_sff_queue_pio_task(ap, 0); + ata_sff_queue_pio_task(link, 0); /* if polling, ata_sff_pio_task() handles the * rest. otherwise, interrupt handler takes @@ -1444,7 +1461,7 @@ unsigned int ata_sff_qc_issue(struct ata_queued_cmd *qc) /* send cdb by polling if no cdb interrupt */ if ((!(qc->dev->flags & ATA_DFLAG_CDB_INTR)) || (qc->tf.flags & ATA_TFLAG_POLLING)) - ata_sff_queue_pio_task(ap, 0); + ata_sff_queue_pio_task(link, 0); break; default: @@ -2737,6 +2754,7 @@ EXPORT_SYMBOL_GPL(ata_bmdma_dumb_qc_prep); unsigned int ata_bmdma_qc_issue(struct ata_queued_cmd *qc) { struct ata_port *ap = qc->ap; + struct ata_link *link = qc->dev->link; /* defer PIO handling to sff_qc_issue */ if (!ata_is_dma(qc->tf.protocol)) @@ -2765,7 +2783,7 @@ unsigned int ata_bmdma_qc_issue(struct ata_queued_cmd *qc) /* send cdb by polling if no cdb interrupt */ if (!(qc->dev->flags & ATA_DFLAG_CDB_INTR)) - ata_sff_queue_pio_task(ap, 0); + ata_sff_queue_pio_task(link, 0); break; default: diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c index 8198259..a9fd970 100644 --- a/drivers/ata/sata_mv.c +++ b/drivers/ata/sata_mv.c @@ -2284,7 +2284,7 @@ static unsigned int mv_qc_issue_fis(struct ata_queued_cmd *qc) } if (qc->tf.flags & ATA_TFLAG_POLLING) - ata_sff_queue_pio_task(ap, 0); + ata_sff_queue_pio_task(link, 0); return 0; } diff --git a/include/linux/libata.h b/include/linux/libata.h index 7de282d..45fb296 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -724,6 +724,7 @@ struct ata_port { struct ata_ioports ioaddr; /* ATA cmd/ctl/dma register blocks */ u8 ctl; /* cache of ATA control register */ u8 last_ctl; /* Cache last written value */ + struct ata_link* sff_pio_task_link; /* link currently used */ struct delayed_work sff_pio_task; #ifdef CONFIG_ATA_BMDMA struct ata_bmdma_prd *bmdma_prd; /* BMDMA SG list */ @@ -1595,7 +1596,7 @@ extern void ata_sff_irq_on(struct ata_port *ap); extern void ata_sff_irq_clear(struct ata_port *ap); extern int ata_sff_hsm_move(struct ata_port *ap, struct ata_queued_cmd *qc, u8 status, int in_wq); -extern void ata_sff_queue_pio_task(struct ata_port *ap, unsigned long delay); +extern void ata_sff_queue_pio_task(struct ata_link *link, unsigned long delay); extern unsigned int ata_sff_qc_issue(struct ata_queued_cmd *qc); extern bool ata_sff_qc_fill_rtf(struct ata_queued_cmd *qc); extern unsigned int ata_sff_port_intr(struct ata_port *ap, -- cgit v0.10.2 From df09162550fbb53354f0c88e85b5d0e6129ee9cc Mon Sep 17 00:00:00 2001 From: Chris Wright Date: Thu, 9 Sep 2010 16:34:59 -0700 Subject: tracing: t_start: reset FTRACE_ITER_HASH in case of seek/pread Be sure to avoid entering t_show() with FTRACE_ITER_HASH set without having properly started the iterator to iterate the hash. This case is degenerate and, as discovered by Robert Swiecki, can cause t_hash_show() to misuse a pointer. This causes a NULL ptr deref with possible security implications. Tracked as CVE-2010-3079. Cc: Robert Swiecki Cc: Eugene Teo Cc: Signed-off-by: Chris Wright Signed-off-by: Steven Rostedt diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 83a16e9..fa7ece6 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -1510,6 +1510,8 @@ static void *t_start(struct seq_file *m, loff_t *pos) if (*pos > 0) return t_hash_start(m, pos); iter->flags |= FTRACE_ITER_PRINTALL; + /* reset in case of seek/pread */ + iter->flags &= ~FTRACE_ITER_HASH; return iter; } -- cgit v0.10.2 From 201b6bab6798fcd8ec1cb30165b91fdb89a91775 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Tue, 7 Sep 2010 20:33:24 +0000 Subject: MAINTAINERS: Add CAIF Signed-off-by: Joe Perches Signed-off-by: David S. Miller diff --git a/MAINTAINERS b/MAINTAINERS index 087912a..b5fb9bd 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1445,6 +1445,16 @@ S: Maintained F: Documentation/video4linux/cafe_ccic F: drivers/media/video/cafe_ccic* +CAIF NETWORK LAYER +M: Sjur Braendeland +L: netdev@vger.kernel.org +S: Supported +F: Documentation/networking/caif/ +F: drivers/net/caif/ +F: include/linux/caif/ +F: include/net/caif/ +F: net/caif/ + CALGARY x86-64 IOMMU M: Muli Ben-Yehuda M: "Jon D. Mason" -- cgit v0.10.2 From c9cedbba0fc591e1c0587f838932ca3f3c6fec57 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Wed, 8 Sep 2010 07:50:47 +0000 Subject: ipheth: remove incorrect devtype to WWAN The 'wwan' devtype is meant for devices that require preconfiguration and *every* time setup before the ethernet interface can be used, like cellular modems which require a series of setup commands on serial ports or other mechanisms before the ethernet interface will handle packets. As ipheth only requires one-per-hotplug pairing setup with no preconfiguration (like APN, phone #, etc) and the network interface is usable at any time after that initial setup, remove the incorrect devtype wwan. Signed-off-by: Dan Williams Signed-off-by: David S. Miller diff --git a/drivers/net/usb/ipheth.c b/drivers/net/usb/ipheth.c index 8ed30fa..b2bcf99 100644 --- a/drivers/net/usb/ipheth.c +++ b/drivers/net/usb/ipheth.c @@ -429,10 +429,6 @@ static const struct net_device_ops ipheth_netdev_ops = { .ndo_get_stats = &ipheth_stats, }; -static struct device_type ipheth_type = { - .name = "wwan", -}; - static int ipheth_probe(struct usb_interface *intf, const struct usb_device_id *id) { @@ -450,7 +446,7 @@ static int ipheth_probe(struct usb_interface *intf, netdev->netdev_ops = &ipheth_netdev_ops; netdev->watchdog_timeo = IPHETH_TX_TIMEOUT; - strcpy(netdev->name, "wwan%d"); + strcpy(netdev->name, "eth%d"); dev = netdev_priv(netdev); dev->udev = udev; @@ -500,7 +496,6 @@ static int ipheth_probe(struct usb_interface *intf, SET_NETDEV_DEV(netdev, &intf->dev); SET_ETHTOOL_OPS(netdev, &ops); - SET_NETDEV_DEVTYPE(netdev, &ipheth_type); retval = register_netdev(netdev); if (retval) { -- cgit v0.10.2 From be14eb619108fa8b7120eb2c42d66d5f623ae10e Mon Sep 17 00:00:00 2001 From: Brian King Date: Fri, 10 Sep 2010 09:03:21 +0200 Subject: block: Range check cpu in blk_cpu_to_group While testing CPU DLPAR, the following problem was discovered. We were DLPAR removing the first CPU, which in this case was logical CPUs 0-3. CPUs 0-2 were already marked offline and we were in the process of offlining CPU 3. After marking the CPU inactive and offline in cpu_disable, but before the cpu was completely idle (cpu_die), we ended up in __make_request on CPU 3. There we looked at the topology map to see which CPU to complete the I/O on and found no CPUs in the cpu_sibling_map. This resulted in the block layer setting the completion cpu to be NR_CPUS, which then caused an oops when we tried to complete the I/O. Fix this by sanity checking the value we return from blk_cpu_to_group to be a valid cpu value. Signed-off-by: Brian King Signed-off-by: Jens Axboe diff --git a/block/blk.h b/block/blk.h index 6e7dc87..d6b911a 100644 --- a/block/blk.h +++ b/block/blk.h @@ -142,14 +142,18 @@ static inline int queue_congestion_off_threshold(struct request_queue *q) static inline int blk_cpu_to_group(int cpu) { + int group = NR_CPUS; #ifdef CONFIG_SCHED_MC const struct cpumask *mask = cpu_coregroup_mask(cpu); - return cpumask_first(mask); + group = cpumask_first(mask); #elif defined(CONFIG_SCHED_SMT) - return cpumask_first(topology_thread_cpumask(cpu)); + group = cpumask_first(topology_thread_cpumask(cpu)); #else return cpu; #endif + if (likely(group < NR_CPUS)) + return group; + return cpu; } /* -- cgit v0.10.2 From 1879c45cce6411fe97b0b65ccad38f5206700dda Mon Sep 17 00:00:00 2001 From: Nicolas Ferre Date: Fri, 10 Sep 2010 11:38:43 +0200 Subject: AT91: dm9000 initialization update Add information in dm9000 mac/phy chip initialization: - irq resource details - platform data details Signed-off-by: Nicolas Ferre diff --git a/arch/arm/mach-at91/board-sam9261ek.c b/arch/arm/mach-at91/board-sam9261ek.c index c4c8865..b4307ab 100644 --- a/arch/arm/mach-at91/board-sam9261ek.c +++ b/arch/arm/mach-at91/board-sam9261ek.c @@ -93,11 +93,12 @@ static struct resource dm9000_resource[] = { .start = AT91_PIN_PC11, .end = AT91_PIN_PC11, .flags = IORESOURCE_IRQ + | IORESOURCE_IRQ_LOWEDGE | IORESOURCE_IRQ_HIGHEDGE, } }; static struct dm9000_plat_data dm9000_platdata = { - .flags = DM9000_PLATF_16BITONLY, + .flags = DM9000_PLATF_16BITONLY | DM9000_PLATF_NO_EEPROM, }; static struct platform_device dm9000_device = { -- cgit v0.10.2 From 64d72bbeeb68b536a27319471f212d413f58bf6f Mon Sep 17 00:00:00 2001 From: Nicolas Ferre Date: Fri, 10 Sep 2010 11:26:42 +0200 Subject: AT91: at91sam9261ek board: remove warnings related to use of SPI or SD/MMC The sd/mmc data structure is not used if SPI is selected. The configuration of PIO on the board prevent from using both interfaces at the same time (board dependent). Remove the warnings at compilation time adding a preprocessor condition. Signed-off-by: Nicolas Ferre diff --git a/arch/arm/mach-at91/board-sam9261ek.c b/arch/arm/mach-at91/board-sam9261ek.c index b4307ab..19d4d2c 100644 --- a/arch/arm/mach-at91/board-sam9261ek.c +++ b/arch/arm/mach-at91/board-sam9261ek.c @@ -169,17 +169,6 @@ static struct at91_udc_data __initdata ek_udc_data = { /* - * MCI (SD/MMC) - */ -static struct at91_mmc_data __initdata ek_mmc_data = { - .wire4 = 1, -// .det_pin = ... not connected -// .wp_pin = ... not connected -// .vcc_pin = ... not connected -}; - - -/* * NAND flash */ static struct mtd_partition __initdata ek_nand_partition[] = { @@ -247,6 +236,10 @@ static void __init ek_add_device_nand(void) at91_add_device_nand(&ek_nand_data); } +/* + * SPI related devices + */ +#if defined(CONFIG_SPI_ATMEL) || defined(CONFIG_SPI_ATMEL_MODULE) /* * ADS7846 Touchscreen @@ -357,6 +350,21 @@ static struct spi_board_info ek_spi_devices[] = { #endif }; +#else /* CONFIG_SPI_ATMEL_* */ + +/* + * MCI (SD/MMC) + * spi0 and mmc/sd share the same PIO pins: cannot be used at the same time + */ +static struct at91_mmc_data __initdata ek_mmc_data = { + .wire4 = 1, +// .det_pin = ... not connected +// .wp_pin = ... not connected +// .vcc_pin = ... not connected +}; + +#endif /* CONFIG_SPI_ATMEL_* */ + /* * LCD Controller -- cgit v0.10.2 From 4deb22a6001eb0d4058295c3645c36190a0ec603 Mon Sep 17 00:00:00 2001 From: Nicolas Ferre Date: Fri, 10 Sep 2010 14:36:06 +0200 Subject: AT91: at91sam9261ek: remove C99 comments but keep information Signed-off-by: Nicolas Ferre diff --git a/arch/arm/mach-at91/board-sam9261ek.c b/arch/arm/mach-at91/board-sam9261ek.c index 19d4d2c..65eb094 100644 --- a/arch/arm/mach-at91/board-sam9261ek.c +++ b/arch/arm/mach-at91/board-sam9261ek.c @@ -351,16 +351,14 @@ static struct spi_board_info ek_spi_devices[] = { }; #else /* CONFIG_SPI_ATMEL_* */ +/* spi0 and mmc/sd share the same PIO pins: cannot be used at the same time */ /* * MCI (SD/MMC) - * spi0 and mmc/sd share the same PIO pins: cannot be used at the same time + * det_pin, wp_pin and vcc_pin are not connected */ static struct at91_mmc_data __initdata ek_mmc_data = { .wire4 = 1, -// .det_pin = ... not connected -// .wp_pin = ... not connected -// .vcc_pin = ... not connected }; #endif /* CONFIG_SPI_ATMEL_* */ -- cgit v0.10.2 From a122eb2fdfd78b58c6dd992d6f4b1aaef667eef9 Mon Sep 17 00:00:00 2001 From: Dan Rosenberg Date: Mon, 6 Sep 2010 18:24:57 -0400 Subject: xfs: prevent reading uninitialized stack memory The XFS_IOC_FSGETXATTR ioctl allows unprivileged users to read 12 bytes of uninitialized stack memory, because the fsxattr struct declared on the stack in xfs_ioc_fsgetxattr() does not alter (or zero) the 12-byte fsx_pad member before copying it back to the user. This patch takes care of it. Signed-off-by: Dan Rosenberg Reviewed-by: Eric Sandeen Signed-off-by: Alex Elder diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c index 4fec427..3b9e626 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl.c +++ b/fs/xfs/linux-2.6/xfs_ioctl.c @@ -785,6 +785,8 @@ xfs_ioc_fsgetxattr( { struct fsxattr fa; + memset(&fa, 0, sizeof(struct fsxattr)); + xfs_ilock(ip, XFS_ILOCK_SHARED); fa.fsx_xflags = xfs_ip2xflags(ip); fa.fsx_extsize = ip->i_d.di_extsize << ip->i_mount->m_sb.sb_blocklog; -- cgit v0.10.2 From dd8849c8f59ec1cee4809a0c5e603e045abe860e Mon Sep 17 00:00:00 2001 From: Jesse Barnes Date: Thu, 9 Sep 2010 11:58:02 -0700 Subject: drm/i915: don't enable self-refresh on Ironlake We don't know how to enable it safely, especially as outputs turn on and off. When disabling LP1 we also need to make sure LP2 and 3 are already disabled. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=29173 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=29082 Reported-by: Chris Lord Signed-off-by: Jesse Barnes Tested-by: Daniel Vetter Cc: stable@kernel.org Signed-off-by: Chris Wilson diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index d094e91..4f5e155 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -2206,9 +2206,17 @@ #define WM1_LP_SR_EN (1<<31) #define WM1_LP_LATENCY_SHIFT 24 #define WM1_LP_LATENCY_MASK (0x7f<<24) +#define WM1_LP_FBC_LP1_MASK (0xf<<20) +#define WM1_LP_FBC_LP1_SHIFT 20 #define WM1_LP_SR_MASK (0x1ff<<8) #define WM1_LP_SR_SHIFT 8 #define WM1_LP_CURSOR_MASK (0x3f) +#define WM2_LP_ILK 0x4510c +#define WM2_LP_EN (1<<31) +#define WM3_LP_ILK 0x45110 +#define WM3_LP_EN (1<<31) +#define WM1S_LP_ILK 0x45120 +#define WM1S_LP_EN (1<<31) /* Memory latency timer register */ #define MLTR_ILK 0x11222 diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 7c91030..19daead 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -3382,8 +3382,7 @@ static void ironlake_update_wm(struct drm_device *dev, int planea_clock, reg_value = I915_READ(WM1_LP_ILK); reg_value &= ~(WM1_LP_LATENCY_MASK | WM1_LP_SR_MASK | WM1_LP_CURSOR_MASK); - reg_value |= WM1_LP_SR_EN | - (ilk_sr_latency << WM1_LP_LATENCY_SHIFT) | + reg_value |= (ilk_sr_latency << WM1_LP_LATENCY_SHIFT) | (sr_wm << WM1_LP_SR_SHIFT) | cursor_wm; I915_WRITE(WM1_LP_ILK, reg_value); @@ -5669,6 +5668,9 @@ void intel_init_clock_gating(struct drm_device *dev) I915_WRITE(DISP_ARB_CTL, (I915_READ(DISP_ARB_CTL) | DISP_FBC_WM_DIS)); + I915_WRITE(WM3_LP_ILK, 0); + I915_WRITE(WM2_LP_ILK, 0); + I915_WRITE(WM1_LP_ILK, 0); } /* * Based on the document from hardware guys the following bits -- cgit v0.10.2 From 9d1ac65a9698513d00e5608d93fca0c53f536c14 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 10 Sep 2010 09:59:46 +0100 Subject: KEYS: Fix RCU no-lock warning in keyctl_session_to_parent() There's an protected access to the parent process's credentials in the middle of keyctl_session_to_parent(). This results in the following RCU warning: =================================================== [ INFO: suspicious rcu_dereference_check() usage. ] --------------------------------------------------- security/keys/keyctl.c:1291 invoked rcu_dereference_check() without protection! other info that might help us debug this: rcu_scheduler_active = 1, debug_locks = 0 1 lock held by keyctl-session-/2137: #0: (tasklist_lock){.+.+..}, at: [] keyctl_session_to_parent+0x60/0x236 stack backtrace: Pid: 2137, comm: keyctl-session- Not tainted 2.6.36-rc2-cachefs+ #1 Call Trace: [] lockdep_rcu_dereference+0xaa/0xb3 [] keyctl_session_to_parent+0xed/0x236 [] sys_keyctl+0xb4/0xb6 [] system_call_fastpath+0x16/0x1b The code should take the RCU read lock to make sure the parents credentials don't go away, even though it's holding a spinlock and has IRQ disabled. Signed-off-by: David Howells Signed-off-by: Linus Torvalds diff --git a/security/keys/keyctl.c b/security/keys/keyctl.c index b2b0998..3868c67 100644 --- a/security/keys/keyctl.c +++ b/security/keys/keyctl.c @@ -1272,6 +1272,7 @@ long keyctl_session_to_parent(void) keyring_r = NULL; me = current; + rcu_read_lock(); write_lock_irq(&tasklist_lock); parent = me->real_parent; @@ -1319,6 +1320,7 @@ long keyctl_session_to_parent(void) set_ti_thread_flag(task_thread_info(parent), TIF_NOTIFY_RESUME); write_unlock_irq(&tasklist_lock); + rcu_read_unlock(); if (oldcred) put_cred(oldcred); return 0; @@ -1327,6 +1329,7 @@ already_same: ret = 0; not_permitted: write_unlock_irq(&tasklist_lock); + rcu_read_unlock(); put_cred(cred); return ret; -- cgit v0.10.2 From 3d96406c7da1ed5811ea52a3b0905f4f0e295376 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 10 Sep 2010 09:59:51 +0100 Subject: KEYS: Fix bug in keyctl_session_to_parent() if parent has no session keyring Fix a bug in keyctl_session_to_parent() whereby it tries to check the ownership of the parent process's session keyring whether or not the parent has a session keyring [CVE-2010-2960]. This results in the following oops: BUG: unable to handle kernel NULL pointer dereference at 00000000000000a0 IP: [] keyctl_session_to_parent+0x251/0x443 ... Call Trace: [] ? keyctl_session_to_parent+0x67/0x443 [] ? __do_fault+0x24b/0x3d0 [] sys_keyctl+0xb4/0xb8 [] system_call_fastpath+0x16/0x1b if the parent process has no session keyring. If the system is using pam_keyinit then it mostly protected against this as all processes derived from a login will have inherited the session keyring created by pam_keyinit during the log in procedure. To test this, pam_keyinit calls need to be commented out in /etc/pam.d/. Reported-by: Tavis Ormandy Signed-off-by: David Howells Acked-by: Tavis Ormandy Signed-off-by: Linus Torvalds diff --git a/security/keys/keyctl.c b/security/keys/keyctl.c index 3868c67..60924f6 100644 --- a/security/keys/keyctl.c +++ b/security/keys/keyctl.c @@ -1305,7 +1305,8 @@ long keyctl_session_to_parent(void) goto not_permitted; /* the keyrings must have the same UID */ - if (pcred->tgcred->session_keyring->uid != mycred->euid || + if ((pcred->tgcred->session_keyring && + pcred->tgcred->session_keyring->uid != mycred->euid) || mycred->tgcred->session_keyring->uid != mycred->euid) goto not_permitted; -- cgit v0.10.2 From 1b528181b2ffa14721fb28ad1bd539fe1732c583 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Tue, 7 Sep 2010 19:35:49 -0700 Subject: setup_arg_pages: diagnose excessive argument size The CONFIG_STACK_GROWSDOWN variant of setup_arg_pages() does not check the size of the argument/environment area on the stack. When it is unworkably large, shift_arg_pages() hits its BUG_ON. This is exploitable with a very large RLIMIT_STACK limit, to create a crash pretty easily. Check that the initial stack is not too large to make it possible to map in any executable. We're not checking that the actual executable (or intepreter, for binfmt_elf) will fit. So those mappings might clobber part of the initial stack mapping. But that is just userland lossage that userland made happen, not a kernel problem. Signed-off-by: Roland McGrath Reviewed-by: KOSAKI Motohiro Signed-off-by: Linus Torvalds diff --git a/fs/exec.c b/fs/exec.c index 2d94552..1b63237 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -594,6 +594,11 @@ int setup_arg_pages(struct linux_binprm *bprm, #else stack_top = arch_align_stack(stack_top); stack_top = PAGE_ALIGN(stack_top); + + if (unlikely(stack_top < mmap_min_addr) || + unlikely(vma->vm_end - vma->vm_start >= stack_top - mmap_min_addr)) + return -ENOMEM; + stack_shift = vma->vm_end - stack_top; bprm->p -= stack_shift; -- cgit v0.10.2 From 7993bc1f4663c0db67bb8f0d98e6678145b387cd Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Tue, 7 Sep 2010 19:36:28 -0700 Subject: execve: improve interactivity with large arguments This adds a preemption point during the copying of the argument and environment strings for execve, in copy_strings(). There is already a preemption point in the count() loop, so this doesn't add any new points in the abstract sense. When the total argument+environment strings are very large, the time spent copying them can be much more than a normal user time slice. So this change improves the interactivity of the rest of the system when one process is doing an execve with very large arguments. Signed-off-by: Roland McGrath Reviewed-by: KOSAKI Motohiro Signed-off-by: Linus Torvalds diff --git a/fs/exec.c b/fs/exec.c index 1b63237..6f2d777 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -419,6 +419,8 @@ static int copy_strings(int argc, const char __user *const __user *argv, while (len > 0) { int offset, bytes_to_copy; + cond_resched(); + offset = pos % PAGE_SIZE; if (offset == 0) offset = PAGE_SIZE; -- cgit v0.10.2 From 9aea5a65aa7a1af9a4236dfaeb0088f1624f9919 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Tue, 7 Sep 2010 19:37:06 -0700 Subject: execve: make responsive to SIGKILL with large arguments An execve with a very large total of argument/environment strings can take a really long time in the execve system call. It runs uninterruptibly to count and copy all the strings. This change makes it abort the exec quickly if sent a SIGKILL. Note that this is the conservative change, to interrupt only for SIGKILL, by using fatal_signal_pending(). It would be perfectly correct semantics to let any signal interrupt the string-copying in execve, i.e. use signal_pending() instead of fatal_signal_pending(). We'll save that change for later, since it could have user-visible consequences, such as having a timer set too quickly make it so that an execve can never complete, though it always happened to work before. Signed-off-by: Roland McGrath Reviewed-by: KOSAKI Motohiro Signed-off-by: Linus Torvalds diff --git a/fs/exec.c b/fs/exec.c index 6f2d777..828dd24 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -376,6 +376,9 @@ static int count(const char __user * const __user * argv, int max) argv++; if (i++ >= max) return -E2BIG; + + if (fatal_signal_pending(current)) + return -ERESTARTNOHAND; cond_resched(); } } @@ -419,6 +422,10 @@ static int copy_strings(int argc, const char __user *const __user *argv, while (len > 0) { int offset, bytes_to_copy; + if (fatal_signal_pending(current)) { + ret = -ERESTARTNOHAND; + goto out; + } cond_resched(); offset = pos % PAGE_SIZE; -- cgit v0.10.2 From 51749e47e191db8e588ad5cebea731caf7b705d7 Mon Sep 17 00:00:00 2001 From: Dave Chinner Date: Wed, 8 Sep 2010 09:00:22 +0000 Subject: xfs: log IO completion workqueue is a high priority queue The workqueue implementation in 2.6.36-rcX has changed, resulting in the workqueues no longer having dedicated threads for work processing. This has caused severe livelocks under heavy parallel create workloads because the log IO completions have been getting held up behind metadata IO completions. Hence log commits would stall, memory allocation would stall because pages could not be cleaned, and lock contention on the AIL during inode IO completion processing was being seen to slow everything down even further. By making the log Io completion workqueue a high priority workqueue, they are queued ahead of all data/metadata IO completions and processed before the data/metadata completions. Hence the log never gets stalled, and operations needed to clean memory can continue as quickly as possible. This avoids the livelock conditions and allos the system to keep running under heavy load as per normal. Signed-off-by: Dave Chinner Reviewed-by: Christoph Hellwig Signed-off-by: Alex Elder diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index d72cf2b..286e36e 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -1932,7 +1932,8 @@ xfs_buf_init(void) if (!xfs_buf_zone) goto out; - xfslogd_workqueue = create_workqueue("xfslogd"); + xfslogd_workqueue = alloc_workqueue("xfslogd", + WQ_RESCUER | WQ_HIGHPRI, 1); if (!xfslogd_workqueue) goto out_free_buf_zone; -- cgit v0.10.2 From 0109c2c48d062a04685638926a35ed20153fedc8 Mon Sep 17 00:00:00 2001 From: mark gross Date: Thu, 9 Sep 2010 23:20:09 +0200 Subject: PM QoS: Correct pr_debug() misuse and improve parameter checks Correct some pr_debug() misuse and add a stronger parameter check to pm_qos_write() for the ASCII hex value case. Thanks to Dan Carpenter for pointing out the problem! Signed-off-by: mark gross Signed-off-by: Rafael J. Wysocki diff --git a/kernel/pm_qos_params.c b/kernel/pm_qos_params.c index b7e4c36..645e541 100644 --- a/kernel/pm_qos_params.c +++ b/kernel/pm_qos_params.c @@ -389,10 +389,12 @@ static ssize_t pm_qos_power_write(struct file *filp, const char __user *buf, } else if (count == 11) { /* len('0x12345678/0') */ if (copy_from_user(ascii_value, buf, 11)) return -EFAULT; + if (strlen(ascii_value) != 10) + return -EINVAL; x = sscanf(ascii_value, "%x", &value); if (x != 1) return -EINVAL; - pr_debug(KERN_ERR "%s, %d, 0x%x\n", ascii_value, x, value); + pr_debug("%s, %d, 0x%x\n", ascii_value, x, value); } else return -EINVAL; -- cgit v0.10.2 From 5ee5e97ee9bca919af11c562beeaf61741ad33f1 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 10 Sep 2010 22:32:53 +0200 Subject: x86, tsc: Fix a preemption leak in restore_sched_clock_state() A real life genuine preemption leak.. Reported-and-tested-by: Jeff Chua Signed-off-by: Peter Zijlstra Acked-by: Suresh Siddha Signed-off-by: Linus Torvalds diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index d632934..26a863a 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -655,7 +655,7 @@ void restore_sched_clock_state(void) local_irq_save(flags); - get_cpu_var(cyc2ns_offset) = 0; + __get_cpu_var(cyc2ns_offset) = 0; offset = cyc2ns_suspend - sched_clock(); for_each_possible_cpu(cpu) -- cgit v0.10.2 From 55496c896b8a695140045099d4e0175cf09d4eae Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 10 Sep 2010 22:32:53 +0200 Subject: x86, tsc: Fix a preemption leak in restore_sched_clock_state() Doh, a real life genuine preemption leak.. This caused a suspend failure. Reported-bisected-and-tested-by-the-invaluable: Jeff Chua Acked-by: Suresh Siddha Signed-off-by: Peter Zijlstra Cc: Rafael J. Wysocki Cc: Nico Schottelius Cc: Jesse Barnes Cc: Linus Torvalds Cc: Florian Pritz Cc: Suresh Siddha Cc: Len Brown Cc: # Greg, please apply after: cd7240c ("x86, tsc, sched: Recompute cyc2ns_offset's during resume from") sleep states LKML-Reference: <1284150773.402.122.camel@laptop> Signed-off-by: Ingo Molnar diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index d632934..26a863a 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -655,7 +655,7 @@ void restore_sched_clock_state(void) local_irq_save(flags); - get_cpu_var(cyc2ns_offset) = 0; + __get_cpu_var(cyc2ns_offset) = 0; offset = cyc2ns_suspend - sched_clock(); for_each_possible_cpu(cpu) -- cgit v0.10.2 From 6715045ddc7472a22be5e49d4047d2d89b391f45 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Sat, 11 Sep 2010 20:58:27 +0200 Subject: PM / Hibernate: Avoid hitting OOM during preallocation of memory There is a problem in hibernate_preallocate_memory() that it calls preallocate_image_memory() with an argument that may be greater than the total number of available non-highmem memory pages. If that's the case, the OOM condition is guaranteed to trigger, which in turn can cause significant slowdown to occur during hibernation. To avoid that, make preallocate_image_memory() adjust its argument before calling preallocate_image_pages(), so that the total number of saveable non-highem pages left is not less than the minimum size of a hibernation image. Change hibernate_preallocate_memory() to try to allocate from highmem if the number of pages allocated by preallocate_image_memory() is too low. Modify free_unnecessary_pages() to take all possible memory allocation patterns into account. Reported-by: KOSAKI Motohiro Signed-off-by: Rafael J. Wysocki Tested-by: M. Vefa Bicakci diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 5e7edfb..5209b39 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -1122,9 +1122,19 @@ static unsigned long preallocate_image_pages(unsigned long nr_pages, gfp_t mask) return nr_alloc; } -static unsigned long preallocate_image_memory(unsigned long nr_pages) +static unsigned long preallocate_image_memory(unsigned long nr_pages, + unsigned long avail_normal) { - return preallocate_image_pages(nr_pages, GFP_IMAGE); + unsigned long alloc; + + if (avail_normal <= alloc_normal) + return 0; + + alloc = avail_normal - alloc_normal; + if (nr_pages < alloc) + alloc = nr_pages; + + return preallocate_image_pages(alloc, GFP_IMAGE); } #ifdef CONFIG_HIGHMEM @@ -1170,15 +1180,22 @@ static inline unsigned long preallocate_highmem_fraction(unsigned long nr_pages, */ static void free_unnecessary_pages(void) { - unsigned long save_highmem, to_free_normal, to_free_highmem; + unsigned long save, to_free_normal, to_free_highmem; - to_free_normal = alloc_normal - count_data_pages(); - save_highmem = count_highmem_pages(); - if (alloc_highmem > save_highmem) { - to_free_highmem = alloc_highmem - save_highmem; + save = count_data_pages(); + if (alloc_normal >= save) { + to_free_normal = alloc_normal - save; + save = 0; + } else { + to_free_normal = 0; + save -= alloc_normal; + } + save += count_highmem_pages(); + if (alloc_highmem >= save) { + to_free_highmem = alloc_highmem - save; } else { to_free_highmem = 0; - to_free_normal -= save_highmem - alloc_highmem; + to_free_normal -= save - alloc_highmem; } memory_bm_position_reset(©_bm); @@ -1259,7 +1276,7 @@ int hibernate_preallocate_memory(void) { struct zone *zone; unsigned long saveable, size, max_size, count, highmem, pages = 0; - unsigned long alloc, save_highmem, pages_highmem; + unsigned long alloc, save_highmem, pages_highmem, avail_normal; struct timeval start, stop; int error; @@ -1296,6 +1313,7 @@ int hibernate_preallocate_memory(void) else count += zone_page_state(zone, NR_FREE_PAGES); } + avail_normal = count; count += highmem; count -= totalreserve_pages; @@ -1310,12 +1328,21 @@ int hibernate_preallocate_memory(void) */ if (size >= saveable) { pages = preallocate_image_highmem(save_highmem); - pages += preallocate_image_memory(saveable - pages); + pages += preallocate_image_memory(saveable - pages, avail_normal); goto out; } /* Estimate the minimum size of the image. */ pages = minimum_image_size(saveable); + /* + * To avoid excessive pressure on the normal zone, leave room in it to + * accommodate an image of the minimum size (unless it's already too + * small, in which case don't preallocate pages from it at all). + */ + if (avail_normal > pages) + avail_normal -= pages; + else + avail_normal = 0; if (size < pages) size = min_t(unsigned long, pages, max_size); @@ -1336,16 +1363,34 @@ int hibernate_preallocate_memory(void) */ pages_highmem = preallocate_image_highmem(highmem / 2); alloc = (count - max_size) - pages_highmem; - pages = preallocate_image_memory(alloc); - if (pages < alloc) - goto err_out; - size = max_size - size; - alloc = size; - size = preallocate_highmem_fraction(size, highmem, count); - pages_highmem += size; - alloc -= size; - pages += preallocate_image_memory(alloc); - pages += pages_highmem; + pages = preallocate_image_memory(alloc, avail_normal); + if (pages < alloc) { + /* We have exhausted non-highmem pages, try highmem. */ + alloc -= pages; + pages += pages_highmem; + pages_highmem = preallocate_image_highmem(alloc); + if (pages_highmem < alloc) + goto err_out; + pages += pages_highmem; + /* + * size is the desired number of saveable pages to leave in + * memory, so try to preallocate (all memory - size) pages. + */ + alloc = (count - pages) - size; + pages += preallocate_image_highmem(alloc); + } else { + /* + * There are approximately max_size saveable pages at this point + * and we want to reduce this number down to size. + */ + alloc = max_size - size; + size = preallocate_highmem_fraction(alloc, highmem, count); + pages_highmem += size; + alloc -= size; + size = preallocate_image_memory(alloc, avail_normal); + pages_highmem += preallocate_image_highmem(alloc - size); + pages += pages_highmem + size; + } /* * We only need as many page frames for the image as there are saveable -- cgit v0.10.2 From 1f3a66889c4c80c821f3eadf899c140e91452f8e Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sat, 11 Sep 2010 15:55:12 -0700 Subject: kernel-doc: ignore case when stripping attributes There are valid attributes that could have upper case letters, but we still want to remove, like for example __attribute__((aligned(NETDEV_ALIGN))) as encountered in the wireless code. Signed-off-by: Johannes Berg Signed-off-by: Randy Dunlap Signed-off-by: Linus Torvalds diff --git a/scripts/kernel-doc b/scripts/kernel-doc index 102e123..a68240c 100755 --- a/scripts/kernel-doc +++ b/scripts/kernel-doc @@ -1679,7 +1679,7 @@ sub check_sections($$$$$$) { foreach $px (0 .. $#prms) { $prm_clean = $prms[$px]; $prm_clean =~ s/\[.*\]//; - $prm_clean =~ s/__attribute__\s*\(\([a-z,_\*\s\(\)]*\)\)//; + $prm_clean =~ s/__attribute__\s*\(\([a-z,_\*\s\(\)]*\)\)//i; # ignore array size in a parameter string; # however, the original param string may contain # spaces, e.g.: addr[6 + 2] -- cgit v0.10.2 From eda603f6cdba4b14dbf80531fab2fe545232e7a0 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Sat, 11 Sep 2010 15:55:22 -0700 Subject: docbook: warn on unused doc entries When you don't use !E or !I but only !F, then it's very easy to miss including some functions, structs etc. in documentation. To help finding which ones were missed, allow printing out the unused ones as warnings. For example, using this on mac80211 yields a lot of warnings like this: Warning: didn't use docs for DOC: mac80211 workqueue Warning: didn't use docs for ieee80211_max_queues Warning: didn't use docs for ieee80211_bss_change Warning: didn't use docs for ieee80211_bss_conf when generating the documentation for it. Signed-off-by: Johannes Berg Signed-off-by: Randy Dunlap Signed-off-by: Linus Torvalds diff --git a/Documentation/kernel-doc-nano-HOWTO.txt b/Documentation/kernel-doc-nano-HOWTO.txt index 27a52b3..3d8a977 100644 --- a/Documentation/kernel-doc-nano-HOWTO.txt +++ b/Documentation/kernel-doc-nano-HOWTO.txt @@ -345,5 +345,10 @@ documentation, in , for the functions listed. section titled
from . Spaces are allowed in
; do not quote the
. +!C is replaced by nothing, but makes the tools check that +all DOC: sections and documented functions, symbols, etc. are used. +This makes sense to use when you use !F/!P only and want to verify +that all documentation is included. + Tim. */ diff --git a/scripts/basic/docproc.c b/scripts/basic/docproc.c index 79ab973..fc3b18d 100644 --- a/scripts/basic/docproc.c +++ b/scripts/basic/docproc.c @@ -34,12 +34,14 @@ * */ +#define _GNU_SOURCE #include #include #include #include #include #include +#include #include #include @@ -54,6 +56,7 @@ typedef void FILEONLY(char * file); FILEONLY *internalfunctions; FILEONLY *externalfunctions; FILEONLY *symbolsonly; +FILEONLY *findall; typedef void FILELINE(char * file, char * line); FILELINE * singlefunctions; @@ -65,12 +68,30 @@ FILELINE * docsection; #define KERNELDOCPATH "scripts/" #define KERNELDOC "kernel-doc" #define DOCBOOK "-docbook" +#define LIST "-list" #define FUNCTION "-function" #define NOFUNCTION "-nofunction" #define NODOCSECTIONS "-no-doc-sections" static char *srctree, *kernsrctree; +static char **all_list = NULL; +static int all_list_len = 0; + +static void consume_symbol(const char *sym) +{ + int i; + + for (i = 0; i < all_list_len; i++) { + if (!all_list[i]) + continue; + if (strcmp(sym, all_list[i])) + continue; + all_list[i] = NULL; + break; + } +} + static void usage (void) { fprintf(stderr, "Usage: docproc {doc|depend} file\n"); @@ -248,6 +269,7 @@ static void docfunctions(char * filename, char * type) struct symfile * sym = &symfilelist[i]; for (j=0; j < sym->symbolcnt; j++) { vec[idx++] = type; + consume_symbol(sym->symbollist[j].name); vec[idx++] = sym->symbollist[j].name; } } @@ -287,6 +309,11 @@ static void singfunc(char * filename, char * line) vec[idx++] = &line[i]; } } + for (i = 0; i < idx; i++) { + if (strcmp(vec[i], FUNCTION)) + continue; + consume_symbol(vec[i + 1]); + } vec[idx++] = filename; vec[idx] = NULL; exec_kernel_doc(vec); @@ -306,6 +333,10 @@ static void docsect(char *filename, char *line) if (*s == '\n') *s = '\0'; + asprintf(&s, "DOC: %s", line); + consume_symbol(s); + free(s); + vec[0] = KERNELDOC; vec[1] = DOCBOOK; vec[2] = FUNCTION; @@ -315,6 +346,84 @@ static void docsect(char *filename, char *line) exec_kernel_doc(vec); } +static void find_all_symbols(char *filename) +{ + char *vec[4]; /* kerneldoc -list file NULL */ + pid_t pid; + int ret, i, count, start; + char real_filename[PATH_MAX + 1]; + int pipefd[2]; + char *data, *str; + size_t data_len = 0; + + vec[0] = KERNELDOC; + vec[1] = LIST; + vec[2] = filename; + vec[3] = NULL; + + if (pipe(pipefd)) { + perror("pipe"); + exit(1); + } + + switch (pid=fork()) { + case -1: + perror("fork"); + exit(1); + case 0: + close(pipefd[0]); + dup2(pipefd[1], 1); + memset(real_filename, 0, sizeof(real_filename)); + strncat(real_filename, kernsrctree, PATH_MAX); + strncat(real_filename, "/" KERNELDOCPATH KERNELDOC, + PATH_MAX - strlen(real_filename)); + execvp(real_filename, vec); + fprintf(stderr, "exec "); + perror(real_filename); + exit(1); + default: + close(pipefd[1]); + data = malloc(4096); + do { + while ((ret = read(pipefd[0], + data + data_len, + 4096)) > 0) { + data_len += ret; + data = realloc(data, data_len + 4096); + } + } while (ret == -EAGAIN); + if (ret != 0) { + perror("read"); + exit(1); + } + waitpid(pid, &ret ,0); + } + if (WIFEXITED(ret)) + exitstatus |= WEXITSTATUS(ret); + else + exitstatus = 0xff; + + count = 0; + /* poor man's strtok, but with counting */ + for (i = 0; i < data_len; i++) { + if (data[i] == '\n') { + count++; + data[i] = '\0'; + } + } + start = all_list_len; + all_list_len += count; + all_list = realloc(all_list, sizeof(char *) * all_list_len); + str = data; + for (i = 0; i < data_len && start != all_list_len; i++) { + if (data[i] == '\0') { + all_list[start] = str; + str = data + i + 1; + start++; + } + } +} + /* * Parse file, calling action specific functions for: * 1) Lines containing !E @@ -322,7 +431,8 @@ static void docsect(char *filename, char *line) * 3) Lines containing !D * 4) Lines containing !F * 5) Lines containing !P - * 6) Default lines - lines not matching the above + * 6) Lines containing !C + * 7) Default lines - lines not matching the above */ static void parse_file(FILE *infile) { @@ -365,6 +475,12 @@ static void parse_file(FILE *infile) s++; docsection(line + 2, s); break; + case 'C': + while (*s && !isspace(*s)) s++; + *s = '\0'; + if (findall) + findall(line+2); + break; default: defaultline(line); } @@ -380,6 +496,7 @@ static void parse_file(FILE *infile) int main(int argc, char *argv[]) { FILE * infile; + int i; srctree = getenv("SRCTREE"); if (!srctree) @@ -415,6 +532,7 @@ int main(int argc, char *argv[]) symbolsonly = find_export_symbols; singlefunctions = noaction2; docsection = noaction2; + findall = find_all_symbols; parse_file(infile); /* Rewind to start from beginning of file again */ @@ -425,8 +543,16 @@ int main(int argc, char *argv[]) symbolsonly = printline; singlefunctions = singfunc; docsection = docsect; + findall = NULL; parse_file(infile); + + for (i = 0; i < all_list_len; i++) { + if (!all_list[i]) + continue; + fprintf(stderr, "Warning: didn't use docs for %s\n", + all_list[i]); + } } else if (strcmp("depend", argv[1]) == 0) { @@ -439,6 +565,7 @@ int main(int argc, char *argv[]) symbolsonly = adddep; singlefunctions = adddep2; docsection = adddep2; + findall = adddep; parse_file(infile); printf("\n"); } diff --git a/scripts/kernel-doc b/scripts/kernel-doc index a68240c..cdb6dc1 100755 --- a/scripts/kernel-doc +++ b/scripts/kernel-doc @@ -44,12 +44,13 @@ use strict; # Note: This only supports 'c'. # usage: -# kernel-doc [ -docbook | -html | -text | -man ] [ -no-doc-sections ] +# kernel-doc [ -docbook | -html | -text | -man | -list ] [ -no-doc-sections ] # [ -function funcname [ -function funcname ...] ] c file(s)s > outputfile # or # [ -nofunction funcname [ -function funcname ...] ] c file(s)s > outputfile # # Set output format using one of -docbook -html -text or -man. Default is man. +# The -list format is for internal use by docproc. # # -no-doc-sections # Do not output DOC: sections @@ -210,9 +211,16 @@ my %highlights_text = ( $type_constant, "\$1", $type_param, "\$1" ); my $blankline_text = ""; +# list mode +my %highlights_list = ( $type_constant, "\$1", + $type_func, "\$1", + $type_struct, "\$1", + $type_param, "\$1" ); +my $blankline_list = ""; sub usage { - print "Usage: $0 [ -v ] [ -docbook | -html | -text | -man ] [ -no-doc-sections ]\n"; + print "Usage: $0 [ -v ] [ -docbook | -html | -text | -man | -list ]\n"; + print " [ -no-doc-sections ]\n"; print " [ -function funcname [ -function funcname ...] ]\n"; print " [ -nofunction funcname [ -nofunction funcname ...] ]\n"; print " c source file(s) > outputfile\n"; @@ -318,6 +326,10 @@ while ($ARGV[0] =~ m/^-(.*)/) { $output_mode = "xml"; %highlights = %highlights_xml; $blankline = $blankline_xml; + } elsif ($cmd eq "-list") { + $output_mode = "list"; + %highlights = %highlights_list; + $blankline = $blankline_list; } elsif ($cmd eq "-gnome") { $output_mode = "gnome"; %highlights = %highlights_gnome; @@ -1361,6 +1373,42 @@ sub output_blockhead_text(%) { } } +## list mode output functions + +sub output_function_list(%) { + my %args = %{$_[0]}; + + print $args{'function'} . "\n"; +} + +# output enum in list +sub output_enum_list(%) { + my %args = %{$_[0]}; + print $args{'enum'} . "\n"; +} + +# output typedef in list +sub output_typedef_list(%) { + my %args = %{$_[0]}; + print $args{'typedef'} . "\n"; +} + +# output struct as list +sub output_struct_list(%) { + my %args = %{$_[0]}; + + print $args{'struct'} . "\n"; +} + +sub output_blockhead_list(%) { + my %args = %{$_[0]}; + my ($parameter, $section); + + foreach $section (@{$args{'sectionlist'}}) { + print "DOC: $section\n"; + } +} + ## # generic output function for all types (function, struct/union, typedef, enum); # calls the generated, variable output_ function name based on -- cgit v0.10.2 From 39d709392ffc1d05f825191b07b7d865d434f282 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sat, 11 Sep 2010 15:55:26 -0700 Subject: docbook: skip files with no docs since they generate scary warnings Fix docbook templates that reference files that do not contain the expected kernel-doc notation. Fixes these warnings: Warning(arch/x86/include/asm/unaligned.h): no structured comments found Warning(lib/vsprintf.c): no structured comments found These cause errors in the generated html output, like below, so drop these lines. Name arch/x86/include/asm/unaligned.h - Document generation inconsistency Oops Warning The template for this document tried to insert the structured comment from the file arch/x86/include/asm/unaligned.h at this point, but none was found. This dummy section is inserted to allow generation to continue. Signed-off-by: Randy Dunlap Signed-off-by: Linus Torvalds diff --git a/Documentation/DocBook/device-drivers.tmpl b/Documentation/DocBook/device-drivers.tmpl index ecd35e9..feca075 100644 --- a/Documentation/DocBook/device-drivers.tmpl +++ b/Documentation/DocBook/device-drivers.tmpl @@ -46,7 +46,6 @@ Atomic and pointer manipulation !Iarch/x86/include/asm/atomic.h -!Iarch/x86/include/asm/unaligned.h Delaying, scheduling, and timer routines diff --git a/Documentation/DocBook/kernel-api.tmpl b/Documentation/DocBook/kernel-api.tmpl index a20c6f6..6899f47 100644 --- a/Documentation/DocBook/kernel-api.tmpl +++ b/Documentation/DocBook/kernel-api.tmpl @@ -57,7 +57,6 @@ String Conversions -!Ilib/vsprintf.c !Elib/vsprintf.c String Manipulation -- cgit v0.10.2 From 49553c2ef88749dd502687f4eb9c258bb10a4f44 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 12 Sep 2010 16:07:37 -0700 Subject: Linux 2.6.36-rc4 diff --git a/Makefile b/Makefile index 4df9873..92ab33f 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 36 -EXTRAVERSION = -rc3 +EXTRAVERSION = -rc4 NAME = Sheep on Meth # *DOCUMENTATION* -- cgit v0.10.2 From 147fcf1c211f1a87bf4d0711b7e9637f3d6ce080 Mon Sep 17 00:00:00 2001 From: Joe Perches Date: Sat, 11 Sep 2010 22:10:59 -0700 Subject: sound: Remove pr_ uses of KERN_ Signed-off-by: Joe Perches Acked-by: Mark Brown Acked-by: Geoff Levand Signed-off-by: Takashi Iwai diff --git a/sound/ppc/snd_ps3.c b/sound/ppc/snd_ps3.c index 2f12da4..581a670 100644 --- a/sound/ppc/snd_ps3.c +++ b/sound/ppc/snd_ps3.c @@ -579,7 +579,7 @@ static int snd_ps3_delay_to_bytes(struct snd_pcm_substream *substream, rate * delay_ms / 1000) * substream->runtime->channels; - pr_debug(KERN_ERR "%s: time=%d rate=%d bytes=%ld, frames=%d, ret=%d\n", + pr_debug("%s: time=%d rate=%d bytes=%ld, frames=%d, ret=%d\n", __func__, delay_ms, rate, diff --git a/sound/soc/s3c24xx/s3c-dma.c b/sound/soc/s3c24xx/s3c-dma.c index 1b61c23..f1b1bc4 100644 --- a/sound/soc/s3c24xx/s3c-dma.c +++ b/sound/soc/s3c24xx/s3c-dma.c @@ -94,8 +94,7 @@ static void s3c_dma_enqueue(struct snd_pcm_substream *substream) if ((pos + len) > prtd->dma_end) { len = prtd->dma_end - pos; - pr_debug(KERN_DEBUG "%s: corrected dma len %ld\n", - __func__, len); + pr_debug("%s: corrected dma len %ld\n", __func__, len); } ret = s3c2410_dma_enqueue(prtd->params->channel, -- cgit v0.10.2