From 9fbcbd7e1fa9acde67f3516f7aceef2c0d968a7b Mon Sep 17 00:00:00 2001 From: Hillf Danton Date: Thu, 4 Aug 2011 22:38:31 +0800 Subject: [PATCH 01/75] MIPS: VPE: Select correct tc If we could find tc on the tc list for @index, the found tc should be returned. Signed-off-by: Hillf Danton To: LKML Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2692/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/vpe.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/kernel/vpe.c b/arch/mips/kernel/vpe.c index 2cd50ad0d5c6..3efcb065f78a 100644 --- a/arch/mips/kernel/vpe.c +++ b/arch/mips/kernel/vpe.c @@ -192,7 +192,7 @@ static struct tc *get_tc(int index) } spin_unlock(&vpecontrol.tc_list_lock); - return NULL; + return res; } /* allocate a vpe and associate it with this minor (or index) */ From fe0b030cf016ee9a4b0ae4adb0095c46d0e461cc Mon Sep 17 00:00:00 2001 From: Hillf Danton Date: Thu, 4 Aug 2011 22:46:41 +0800 Subject: [PATCH 02/75] MIPS: i8259: Correct comment for i8259 The comment for the slave PIC is changed from 8259A-1 to 8259A-2. Signed-off-by: Hillf Danton To: LKML Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2693/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/i8259.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/kernel/i8259.c b/arch/mips/kernel/i8259.c index 5c74eb797f08..9b22db5dcef1 100644 --- a/arch/mips/kernel/i8259.c +++ b/arch/mips/kernel/i8259.c @@ -229,7 +229,7 @@ static void i8259A_shutdown(void) */ if (i8259A_auto_eoi >= 0) { outb(0xff, PIC_MASTER_IMR); /* mask all of 8259A-1 */ - outb(0xff, PIC_SLAVE_IMR); /* mask all of 8259A-1 */ + outb(0xff, PIC_SLAVE_IMR); /* mask all of 8259A-2 */ } } From 16650107579aed81e0b0534a60400f9ba911fe9b Mon Sep 17 00:00:00 2001 From: Kevin Cernekee Date: Sat, 18 Jun 2011 11:28:48 -0700 Subject: [PATCH 03/75] MIPS: Trivial style cleanups in mmap.c Fix checkpatch warnings. Rename arch_get_unmapped_area_foo() to arch_get_unmapped_area_common(). Make indentations and spacing more consistent. Add for likely/unlikely. Signed-off-by: Kevin Cernekee Cc: Jian Peng Cc: David Daney Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2506/ Signed-off-by: Ralf Baechle --- arch/mips/mm/mmap.c | 48 +++++++++++++++++++++++---------------------- 1 file changed, 25 insertions(+), 23 deletions(-) diff --git a/arch/mips/mm/mmap.c b/arch/mips/mm/mmap.c index 9ff5d0fac556..302d779d5b0d 100644 --- a/arch/mips/mm/mmap.c +++ b/arch/mips/mm/mmap.c @@ -6,6 +6,7 @@ * Copyright (C) 2011 Wind River Systems, * written by Ralf Baechle */ +#include #include #include #include @@ -15,12 +16,11 @@ #include unsigned long shm_align_mask = PAGE_SIZE - 1; /* Sane caches */ - EXPORT_SYMBOL(shm_align_mask); /* gap between mmap and stack */ #define MIN_GAP (128*1024*1024UL) -#define MAX_GAP ((TASK_SIZE)/6*5) +#define MAX_GAP ((TASK_SIZE)/6*5) static int mmap_is_legacy(void) { @@ -57,13 +57,13 @@ static inline unsigned long COLOUR_ALIGN_DOWN(unsigned long addr, return base - off; } -#define COLOUR_ALIGN(addr,pgoff) \ +#define COLOUR_ALIGN(addr, pgoff) \ ((((addr) + shm_align_mask) & ~shm_align_mask) + \ (((pgoff) << PAGE_SHIFT) & shm_align_mask)) enum mmap_allocation_direction {UP, DOWN}; -static unsigned long arch_get_unmapped_area_foo(struct file *filp, +static unsigned long arch_get_unmapped_area_common(struct file *filp, unsigned long addr0, unsigned long len, unsigned long pgoff, unsigned long flags, enum mmap_allocation_direction dir) { @@ -103,16 +103,16 @@ static unsigned long arch_get_unmapped_area_foo(struct file *filp, vma = find_vma(mm, addr); if (TASK_SIZE - len >= addr && - (!vma || addr + len <= vma->vm_start)) + (!vma || addr + len <= vma->vm_start)) return addr; } if (dir == UP) { addr = mm->mmap_base; - if (do_color_align) - addr = COLOUR_ALIGN(addr, pgoff); - else - addr = PAGE_ALIGN(addr); + if (do_color_align) + addr = COLOUR_ALIGN(addr, pgoff); + else + addr = PAGE_ALIGN(addr); for (vma = find_vma(current->mm, addr); ; vma = vma->vm_next) { /* At this point: (!vma || addr < vma->vm_end). */ @@ -131,28 +131,30 @@ static unsigned long arch_get_unmapped_area_foo(struct file *filp, mm->free_area_cache = mm->mmap_base; } - /* either no address requested or can't fit in requested address hole */ + /* + * either no address requested, or the mapping can't fit into + * the requested address hole + */ addr = mm->free_area_cache; - if (do_color_align) { - unsigned long base = - COLOUR_ALIGN_DOWN(addr - len, pgoff); - + if (do_color_align) { + unsigned long base = + COLOUR_ALIGN_DOWN(addr - len, pgoff); addr = base + len; - } + } /* make sure it can fit in the remaining address space */ if (likely(addr > len)) { vma = find_vma(mm, addr - len); if (!vma || addr <= vma->vm_start) { - /* remember the address as a hint for next time */ - return mm->free_area_cache = addr-len; + /* cache the address as a hint for next time */ + return mm->free_area_cache = addr - len; } } if (unlikely(mm->mmap_base < len)) goto bottomup; - addr = mm->mmap_base-len; + addr = mm->mmap_base - len; if (do_color_align) addr = COLOUR_ALIGN_DOWN(addr, pgoff); @@ -163,8 +165,8 @@ static unsigned long arch_get_unmapped_area_foo(struct file *filp, * return with success: */ vma = find_vma(mm, addr); - if (likely(!vma || addr+len <= vma->vm_start)) { - /* remember the address as a hint for next time */ + if (likely(!vma || addr + len <= vma->vm_start)) { + /* cache the address as a hint for next time */ return mm->free_area_cache = addr; } @@ -173,7 +175,7 @@ static unsigned long arch_get_unmapped_area_foo(struct file *filp, mm->cached_hole_size = vma->vm_start - addr; /* try just below the current vma->vm_start */ - addr = vma->vm_start-len; + addr = vma->vm_start - len; if (do_color_align) addr = COLOUR_ALIGN_DOWN(addr, pgoff); } while (likely(len < vma->vm_start)); @@ -201,7 +203,7 @@ bottomup: unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr0, unsigned long len, unsigned long pgoff, unsigned long flags) { - return arch_get_unmapped_area_foo(filp, + return arch_get_unmapped_area_common(filp, addr0, len, pgoff, flags, UP); } @@ -213,7 +215,7 @@ unsigned long arch_get_unmapped_area_topdown(struct file *filp, unsigned long addr0, unsigned long len, unsigned long pgoff, unsigned long flags) { - return arch_get_unmapped_area_foo(filp, + return arch_get_unmapped_area_common(filp, addr0, len, pgoff, flags, DOWN); } From d4001e086babc77f26b599893d803a2b6e46d602 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Thu, 23 Jun 2011 22:07:13 +0200 Subject: [PATCH 04/75] MIPS: Remove unneeded version.h includes from arch/mips/ It was pointed out by 'make versioncheck' that some includes of linux/version.h are not needed in arch/mips/. This patch removes them. Signed-off-by: Jesper Juhl To: LKML Cc: trivial@kernel.org Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2540/ Signed-off-by: Ralf Baechle --- arch/mips/include/asm/mach-powertv/dma-coherence.h | 1 - arch/mips/lantiq/xway/ebu.c | 1 - arch/mips/lantiq/xway/pmu.c | 1 - 3 files changed, 3 deletions(-) diff --git a/arch/mips/include/asm/mach-powertv/dma-coherence.h b/arch/mips/include/asm/mach-powertv/dma-coherence.h index 62c094085947..35371641575d 100644 --- a/arch/mips/include/asm/mach-powertv/dma-coherence.h +++ b/arch/mips/include/asm/mach-powertv/dma-coherence.h @@ -13,7 +13,6 @@ #define __ASM_MACH_POWERTV_DMA_COHERENCE_H #include -#include #include #include diff --git a/arch/mips/lantiq/xway/ebu.c b/arch/mips/lantiq/xway/ebu.c index 66eb52fa50a1..033b3184c7a7 100644 --- a/arch/mips/lantiq/xway/ebu.c +++ b/arch/mips/lantiq/xway/ebu.c @@ -10,7 +10,6 @@ #include #include -#include #include #include diff --git a/arch/mips/lantiq/xway/pmu.c b/arch/mips/lantiq/xway/pmu.c index 9d69f01e352b..39f0d2641cbf 100644 --- a/arch/mips/lantiq/xway/pmu.c +++ b/arch/mips/lantiq/xway/pmu.c @@ -8,7 +8,6 @@ #include #include -#include #include #include From 01f96c0a9922cd9919baf9d16febdf7016177a12 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Wed, 21 Sep 2011 15:30:20 +1000 Subject: [PATCH 05/75] md: Avoid waking up a thread after it has been freed. Two related problems: 1/ some error paths call "md_unregister_thread(mddev->thread)" without subsequently clearing ->thread. A subsequent call to mddev_unlock will try to wake the thread, and crash. 2/ Most calls to md_wakeup_thread are protected against the thread disappeared either by: - holding the ->mutex - having an active request, so something else must be keeping the array active. However mddev_unlock calls md_wakeup_thread after dropping the mutex and without any certainty of an active request, so the ->thread could theoretically disappear. So we need a spinlock to provide some protections. So change md_unregister_thread to take a pointer to the thread pointer, and ensure that it always does the required locking, and clears the pointer properly. Reported-by: "Moshe Melnikov" Signed-off-by: NeilBrown cc: stable@kernel.org --- drivers/md/md.c | 22 +++++++++++++++++++--- drivers/md/md.h | 2 +- drivers/md/multipath.c | 3 +-- drivers/md/raid1.c | 3 +-- drivers/md/raid10.c | 5 ++--- drivers/md/raid5.c | 6 ++---- 6 files changed, 26 insertions(+), 15 deletions(-) diff --git a/drivers/md/md.c b/drivers/md/md.c index 5404b2295820..5c95ccb59500 100644 --- a/drivers/md/md.c +++ b/drivers/md/md.c @@ -61,6 +61,11 @@ static void autostart_arrays(int part); #endif +/* pers_list is a list of registered personalities protected + * by pers_lock. + * pers_lock does extra service to protect accesses to + * mddev->thread when the mutex cannot be held. + */ static LIST_HEAD(pers_list); static DEFINE_SPINLOCK(pers_lock); @@ -739,7 +744,12 @@ static void mddev_unlock(mddev_t * mddev) } else mutex_unlock(&mddev->reconfig_mutex); + /* was we've dropped the mutex we need a spinlock to + * make sur the thread doesn't disappear + */ + spin_lock(&pers_lock); md_wakeup_thread(mddev->thread); + spin_unlock(&pers_lock); } static mdk_rdev_t * find_rdev_nr(mddev_t *mddev, int nr) @@ -6429,11 +6439,18 @@ mdk_thread_t *md_register_thread(void (*run) (mddev_t *), mddev_t *mddev, return thread; } -void md_unregister_thread(mdk_thread_t *thread) +void md_unregister_thread(mdk_thread_t **threadp) { + mdk_thread_t *thread = *threadp; if (!thread) return; dprintk("interrupting MD-thread pid %d\n", task_pid_nr(thread->tsk)); + /* Locking ensures that mddev_unlock does not wake_up a + * non-existent thread + */ + spin_lock(&pers_lock); + *threadp = NULL; + spin_unlock(&pers_lock); kthread_stop(thread->tsk); kfree(thread); @@ -7340,8 +7357,7 @@ static void reap_sync_thread(mddev_t *mddev) mdk_rdev_t *rdev; /* resync has finished, collect result */ - md_unregister_thread(mddev->sync_thread); - mddev->sync_thread = NULL; + md_unregister_thread(&mddev->sync_thread); if (!test_bit(MD_RECOVERY_INTR, &mddev->recovery) && !test_bit(MD_RECOVERY_REQUESTED, &mddev->recovery)) { /* success...*/ diff --git a/drivers/md/md.h b/drivers/md/md.h index 1e586bb4452e..0a309dc29b45 100644 --- a/drivers/md/md.h +++ b/drivers/md/md.h @@ -560,7 +560,7 @@ extern int register_md_personality(struct mdk_personality *p); extern int unregister_md_personality(struct mdk_personality *p); extern mdk_thread_t * md_register_thread(void (*run) (mddev_t *mddev), mddev_t *mddev, const char *name); -extern void md_unregister_thread(mdk_thread_t *thread); +extern void md_unregister_thread(mdk_thread_t **threadp); extern void md_wakeup_thread(mdk_thread_t *thread); extern void md_check_recovery(mddev_t *mddev); extern void md_write_start(mddev_t *mddev, struct bio *bi); diff --git a/drivers/md/multipath.c b/drivers/md/multipath.c index 3535c23af288..d5b5fb300171 100644 --- a/drivers/md/multipath.c +++ b/drivers/md/multipath.c @@ -514,8 +514,7 @@ static int multipath_stop (mddev_t *mddev) { multipath_conf_t *conf = mddev->private; - md_unregister_thread(mddev->thread); - mddev->thread = NULL; + md_unregister_thread(&mddev->thread); blk_sync_queue(mddev->queue); /* the unplug fn references 'conf'*/ mempool_destroy(conf->pool); kfree(conf->multipaths); diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index f4622dd8fc59..d9587dffe533 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -2562,8 +2562,7 @@ static int stop(mddev_t *mddev) raise_barrier(conf); lower_barrier(conf); - md_unregister_thread(mddev->thread); - mddev->thread = NULL; + md_unregister_thread(&mddev->thread); if (conf->r1bio_pool) mempool_destroy(conf->r1bio_pool); kfree(conf->mirrors); diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c index d7a8468ddeab..0cd9672cf9cb 100644 --- a/drivers/md/raid10.c +++ b/drivers/md/raid10.c @@ -2955,7 +2955,7 @@ static int run(mddev_t *mddev) return 0; out_free_conf: - md_unregister_thread(mddev->thread); + md_unregister_thread(&mddev->thread); if (conf->r10bio_pool) mempool_destroy(conf->r10bio_pool); safe_put_page(conf->tmppage); @@ -2973,8 +2973,7 @@ static int stop(mddev_t *mddev) raise_barrier(conf, 0); lower_barrier(conf); - md_unregister_thread(mddev->thread); - mddev->thread = NULL; + md_unregister_thread(&mddev->thread); blk_sync_queue(mddev->queue); /* the unplug fn references 'conf'*/ if (conf->r10bio_pool) mempool_destroy(conf->r10bio_pool); diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index 43709fa6b6df..ac5e8b57e50f 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -4941,8 +4941,7 @@ static int run(mddev_t *mddev) return 0; abort: - md_unregister_thread(mddev->thread); - mddev->thread = NULL; + md_unregister_thread(&mddev->thread); if (conf) { print_raid5_conf(conf); free_conf(conf); @@ -4956,8 +4955,7 @@ static int stop(mddev_t *mddev) { raid5_conf_t *conf = mddev->private; - md_unregister_thread(mddev->thread); - mddev->thread = NULL; + md_unregister_thread(&mddev->thread); if (mddev->queue) mddev->queue->backing_dev_info.congested_fn = NULL; free_conf(conf); From 20ef5d3a23d5aacb85d0a2808d949e3daf83cd20 Mon Sep 17 00:00:00 2001 From: Yoichi Yuasa Date: Tue, 26 Jul 2011 18:19:01 +0900 Subject: [PATCH 06/75] MIPS: RC32434: Fix PCI build error arch/mips/pci/pci-rc32434.c: In function 'rc32434_pci_init': arch/mips/pci/pci-rc32434.c:217:16: error: 'rcrc32434_res_pci_io1' undeclared (first use in this function) arch/mips/pci/pci-rc32434.c:217:16: note: each undeclared identifier is reported only once for each function it appears in make[1]: *** [arch/mips/pci/pci-rc32434.o] Error 1 This problem is included in the following commit. commit 28f65c11f2ffb3957259dece647a24f8ad2e241b Author: Joe Perches Date: Thu Jun 9 09:13:32 2011 -0700 treewide: Convert uses of struct resource to resource_size(ptr) Signed-off-by: Yoichi Yuasa Cc: linux-mips Cc: Joe Perches Patchwork: https://patchwork.linux-mips.org/patch/2654/ Signed-off-by: Ralf Baechle --- arch/mips/pci/pci-rc32434.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/pci/pci-rc32434.c b/arch/mips/pci/pci-rc32434.c index 764362ce5e40..5f3a69cebad1 100644 --- a/arch/mips/pci/pci-rc32434.c +++ b/arch/mips/pci/pci-rc32434.c @@ -215,7 +215,7 @@ static int __init rc32434_pci_init(void) rc32434_pcibridge_init(); io_map_base = ioremap(rc32434_res_pci_io1.start, - resource_size(&rcrc32434_res_pci_io1)); + resource_size(&rc32434_res_pci_io1)); if (!io_map_base) return -ENOMEM; From 49de830ad7b94d05f6dbfbae97a03f80169a7462 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sat, 23 Jul 2011 12:41:23 +0000 Subject: [PATCH 07/75] MIPS: Ftrace: Fix the CONFIG_DYNAMIC_FTRACE=n compile MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit arch/mips/kernel/ftrace.c: In function ‘ftrace_get_parent_ra_addr’: arch/mips/kernel/ftrace.c:212: error: implicit declaration of function ‘in_kernel_space’ arch/mips/kernel/ftrace.c: In function ‘prepare_ftrace_return’: arch/mips/kernel/ftrace.c:314: error: ‘MCOUNT_OFFSET_INSNS’ undeclared (first use in this function) arch/mips/kernel/ftrace.c:314: error: (Each undeclared identifier is reported only once arch/mips/kernel/ftrace.c:314: error: for each function it appears in.) Signed-off-by: Thomas Gleixner To: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2634/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/ftrace.c | 39 ++++++++++++++++++++------------------- 1 file changed, 20 insertions(+), 19 deletions(-) diff --git a/arch/mips/kernel/ftrace.c b/arch/mips/kernel/ftrace.c index feb8021a305f..6a2d758dd8e9 100644 --- a/arch/mips/kernel/ftrace.c +++ b/arch/mips/kernel/ftrace.c @@ -19,6 +19,26 @@ #include +#if defined(KBUILD_MCOUNT_RA_ADDRESS) && defined(CONFIG_32BIT) +#define MCOUNT_OFFSET_INSNS 5 +#else +#define MCOUNT_OFFSET_INSNS 4 +#endif + +/* + * Check if the address is in kernel space + * + * Clone core_kernel_text() from kernel/extable.c, but doesn't call + * init_kernel_text() for Ftrace doesn't trace functions in init sections. + */ +static inline int in_kernel_space(unsigned long ip) +{ + if (ip >= (unsigned long)_stext && + ip <= (unsigned long)_etext) + return 1; + return 0; +} + #ifdef CONFIG_DYNAMIC_FTRACE #define JAL 0x0c000000 /* jump & link: ip --> ra, jump to target */ @@ -54,20 +74,6 @@ static inline void ftrace_dyn_arch_init_insns(void) #endif } -/* - * Check if the address is in kernel space - * - * Clone core_kernel_text() from kernel/extable.c, but doesn't call - * init_kernel_text() for Ftrace doesn't trace functions in init sections. - */ -static inline int in_kernel_space(unsigned long ip) -{ - if (ip >= (unsigned long)_stext && - ip <= (unsigned long)_etext) - return 1; - return 0; -} - static int ftrace_modify_code(unsigned long ip, unsigned int new_code) { int faulted; @@ -112,11 +118,6 @@ static int ftrace_modify_code(unsigned long ip, unsigned int new_code) * 1: offset = 4 instructions */ -#if defined(KBUILD_MCOUNT_RA_ADDRESS) && defined(CONFIG_32BIT) -#define MCOUNT_OFFSET_INSNS 5 -#else -#define MCOUNT_OFFSET_INSNS 4 -#endif #define INSN_B_1F (0x10000000 | MCOUNT_OFFSET_INSNS) int ftrace_make_nop(struct module *mod, From 4d85f6afa43d366868fdaa19595722309e4843f6 Mon Sep 17 00:00:00 2001 From: Wu Zhangjin Date: Sat, 23 Jul 2011 12:41:24 +0000 Subject: [PATCH 08/75] MIPS: Make the die_lock be raw On preempt-rt this lock needs to be raw, so it does not get converted to a sleeping spinlock. Trying to sleep in a panic is not really desireable. Signed-off-by: Wu Zhangjin Signed-off-by: Thomas Gleixner Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2636/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/traps.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c index b7517e3abc85..01eff7e609e8 100644 --- a/arch/mips/kernel/traps.c +++ b/arch/mips/kernel/traps.c @@ -364,7 +364,7 @@ static int regs_to_trapnr(struct pt_regs *regs) return (regs->cp0_cause >> 2) & 0x1f; } -static DEFINE_SPINLOCK(die_lock); +static DEFINE_RAW_SPINLOCK(die_lock); void __noreturn die(const char *str, struct pt_regs *regs) { @@ -378,7 +378,7 @@ void __noreturn die(const char *str, struct pt_regs *regs) sig = 0; console_verbose(); - spin_lock_irq(&die_lock); + raw_spin_lock_irq(&die_lock); bust_spinlocks(1); #ifdef CONFIG_MIPS_MT_SMTC mips_mt_regdump(dvpret); @@ -387,7 +387,7 @@ void __noreturn die(const char *str, struct pt_regs *regs) printk("%s[#%d]:\n", str, ++die_counter); show_registers(regs); add_taint(TAINT_DIE); - spin_unlock_irq(&die_lock); + raw_spin_unlock_irq(&die_lock); if (in_interrupt()) panic("Fatal exception in interrupt"); From 5a4a4ad851dd8db2d888fb86c8bd946b2ae79f60 Mon Sep 17 00:00:00 2001 From: Wu Zhangjin Date: Sat, 23 Jul 2011 12:41:24 +0000 Subject: [PATCH 09/75] MIPS: Mark cascade and low level interrupts IRQF_NO_THREAD Mark interrupts with no_action handler, cascade interrupts, low level interrupts (bus error, halt ..) with IRQF_NO_THREAD to exclude them from forced threading. Signed-off-by: Wu Zhangjin Signed-off-by: Thomas Gleixner Cc: linux-mips@linux-mips.org Cc: Wu Zhangjin Signed-off-by: Ralf Baechle --- arch/mips/ar7/irq.c | 3 ++- arch/mips/bcm63xx/irq.c | 1 + arch/mips/cobalt/irq.c | 1 + arch/mips/dec/setup.c | 4 ++++ arch/mips/emma/markeins/irq.c | 2 +- arch/mips/lasat/interrupt.c | 1 + arch/mips/mti-malta/malta-int.c | 6 ++++-- arch/mips/pmc-sierra/msp71xx/msp_irq.c | 2 ++ arch/mips/pnx8550/common/int.c | 2 +- arch/mips/sgi-ip22/ip22-int.c | 10 +++++----- arch/mips/sni/rm200.c | 1 + arch/mips/vr41xx/common/irq.c | 1 + 12 files changed, 24 insertions(+), 10 deletions(-) diff --git a/arch/mips/ar7/irq.c b/arch/mips/ar7/irq.c index 03db3daadbd8..88c4babfdb5d 100644 --- a/arch/mips/ar7/irq.c +++ b/arch/mips/ar7/irq.c @@ -98,7 +98,8 @@ static struct irq_chip ar7_sec_irq_type = { static struct irqaction ar7_cascade_action = { .handler = no_action, - .name = "AR7 cascade interrupt" + .name = "AR7 cascade interrupt", + .flags = IRQF_NO_THREAD, }; static void __init ar7_irq_init(int base) diff --git a/arch/mips/bcm63xx/irq.c b/arch/mips/bcm63xx/irq.c index cea6021cb8d7..162e11b4ed75 100644 --- a/arch/mips/bcm63xx/irq.c +++ b/arch/mips/bcm63xx/irq.c @@ -222,6 +222,7 @@ static struct irq_chip bcm63xx_external_irq_chip = { static struct irqaction cpu_ip2_cascade_action = { .handler = no_action, .name = "cascade_ip2", + .flags = IRQF_NO_THREAD, }; void __init arch_init_irq(void) diff --git a/arch/mips/cobalt/irq.c b/arch/mips/cobalt/irq.c index cb9bf820fe53..965c777d3561 100644 --- a/arch/mips/cobalt/irq.c +++ b/arch/mips/cobalt/irq.c @@ -48,6 +48,7 @@ asmlinkage void plat_irq_dispatch(void) static struct irqaction cascade = { .handler = no_action, .name = "cascade", + .flags = IRQF_NO_THREAD, }; void __init arch_init_irq(void) diff --git a/arch/mips/dec/setup.c b/arch/mips/dec/setup.c index fa45e924be05..f7b7ba6d5c45 100644 --- a/arch/mips/dec/setup.c +++ b/arch/mips/dec/setup.c @@ -101,20 +101,24 @@ int cpu_fpu_mask = DEC_CPU_IRQ_MASK(DEC_CPU_INR_FPU); static struct irqaction ioirq = { .handler = no_action, .name = "cascade", + .flags = IRQF_NO_THREAD, }; static struct irqaction fpuirq = { .handler = no_action, .name = "fpu", + .flags = IRQF_NO_THREAD, }; static struct irqaction busirq = { .flags = IRQF_DISABLED, .name = "bus error", + .flags = IRQF_NO_THREAD, }; static struct irqaction haltirq = { .handler = dec_intr_halt, .name = "halt", + .flags = IRQF_NO_THREAD, }; diff --git a/arch/mips/emma/markeins/irq.c b/arch/mips/emma/markeins/irq.c index 3dbd7a5a6ad3..7798887a1288 100644 --- a/arch/mips/emma/markeins/irq.c +++ b/arch/mips/emma/markeins/irq.c @@ -169,7 +169,7 @@ void emma2rh_gpio_irq_init(void) static struct irqaction irq_cascade = { .handler = no_action, - .flags = 0, + .flags = IRQF_NO_THREAD, .name = "cascade", .dev_id = NULL, .next = NULL, diff --git a/arch/mips/lasat/interrupt.c b/arch/mips/lasat/interrupt.c index de4c165515d7..d608b6ef0edd 100644 --- a/arch/mips/lasat/interrupt.c +++ b/arch/mips/lasat/interrupt.c @@ -105,6 +105,7 @@ asmlinkage void plat_irq_dispatch(void) static struct irqaction cascade = { .handler = no_action, .name = "cascade", + .flags = IRQF_NO_THREAD, }; void __init arch_init_irq(void) diff --git a/arch/mips/mti-malta/malta-int.c b/arch/mips/mti-malta/malta-int.c index 1d36c511a7a5..d53ff91b277c 100644 --- a/arch/mips/mti-malta/malta-int.c +++ b/arch/mips/mti-malta/malta-int.c @@ -350,12 +350,14 @@ unsigned int plat_ipi_resched_int_xlate(unsigned int cpu) static struct irqaction i8259irq = { .handler = no_action, - .name = "XT-PIC cascade" + .name = "XT-PIC cascade", + .flags = IRQF_NO_THREAD, }; static struct irqaction corehi_irqaction = { .handler = no_action, - .name = "CoreHi" + .name = "CoreHi", + .flags = IRQF_NO_THREAD, }; static msc_irqmap_t __initdata msc_irqmap[] = { diff --git a/arch/mips/pmc-sierra/msp71xx/msp_irq.c b/arch/mips/pmc-sierra/msp71xx/msp_irq.c index 4531c4a514bc..42a4c04e21d5 100644 --- a/arch/mips/pmc-sierra/msp71xx/msp_irq.c +++ b/arch/mips/pmc-sierra/msp71xx/msp_irq.c @@ -109,11 +109,13 @@ asmlinkage void plat_irq_dispatch(struct pt_regs *regs) static struct irqaction cic_cascade_msp = { .handler = no_action, .name = "MSP CIC cascade" + .flags = IRQF_NO_THREAD, }; static struct irqaction per_cascade_msp = { .handler = no_action, .name = "MSP PER cascade" + .flags = IRQF_NO_THREAD, }; void __init arch_init_irq(void) diff --git a/arch/mips/pnx8550/common/int.c b/arch/mips/pnx8550/common/int.c index 6b93c81779c1..1ebe22bdadc8 100644 --- a/arch/mips/pnx8550/common/int.c +++ b/arch/mips/pnx8550/common/int.c @@ -167,7 +167,7 @@ static struct irq_chip level_irq_type = { static struct irqaction gic_action = { .handler = no_action, - .flags = IRQF_DISABLED, + .flags = IRQF_DISABLED | IRQF_NO_THREAD, .name = "GIC", }; diff --git a/arch/mips/sgi-ip22/ip22-int.c b/arch/mips/sgi-ip22/ip22-int.c index b4d08e4d2ea9..f72c336ea27b 100644 --- a/arch/mips/sgi-ip22/ip22-int.c +++ b/arch/mips/sgi-ip22/ip22-int.c @@ -155,32 +155,32 @@ static void __irq_entry indy_buserror_irq(void) static struct irqaction local0_cascade = { .handler = no_action, - .flags = IRQF_DISABLED, + .flags = IRQF_DISABLED | IRQF_NO_THREAD, .name = "local0 cascade", }; static struct irqaction local1_cascade = { .handler = no_action, - .flags = IRQF_DISABLED, + .flags = IRQF_DISABLED | IRQF_NO_THREAD, .name = "local1 cascade", }; static struct irqaction buserr = { .handler = no_action, - .flags = IRQF_DISABLED, + .flags = IRQF_DISABLED | IRQF_NO_THREAD, .name = "Bus Error", }; static struct irqaction map0_cascade = { .handler = no_action, - .flags = IRQF_DISABLED, + .flags = IRQF_DISABLED | IRQF_NO_THREAD, .name = "mapable0 cascade", }; #ifdef USE_LIO3_IRQ static struct irqaction map1_cascade = { .handler = no_action, - .flags = IRQF_DISABLED, + .flags = IRQF_DISABLED | IRQF_NO_THREAD, .name = "mapable1 cascade", }; #define SGI_INTERRUPTS SGINT_END diff --git a/arch/mips/sni/rm200.c b/arch/mips/sni/rm200.c index a7e5a6d917b1..3ab5b5d25b0a 100644 --- a/arch/mips/sni/rm200.c +++ b/arch/mips/sni/rm200.c @@ -359,6 +359,7 @@ void sni_rm200_init_8259A(void) static struct irqaction sni_rm200_irq2 = { .handler = no_action, .name = "cascade", + .flags = IRQF_NO_THREAD, }; static struct resource sni_rm200_pic1_resource = { diff --git a/arch/mips/vr41xx/common/irq.c b/arch/mips/vr41xx/common/irq.c index 70a3b85f3757..fad2bef432cd 100644 --- a/arch/mips/vr41xx/common/irq.c +++ b/arch/mips/vr41xx/common/irq.c @@ -34,6 +34,7 @@ static irq_cascade_t irq_cascade[NR_IRQS] __cacheline_aligned; static struct irqaction cascade_irqaction = { .handler = no_action, .name = "cascade", + .flags = IRQF_NO_THREAD, }; int cascade_irq(unsigned int irq, int (*get_irq)(unsigned int)) From 77cbece76723dc9e77497c042827155388bdae6d Mon Sep 17 00:00:00 2001 From: Wu Zhangjin Date: Sat, 23 Jul 2011 12:41:24 +0000 Subject: [PATCH 10/75] MIPS: Loongson: Mark cascade interrupts IRQF_NO_THREAD There are two cascade interrupts in Loongson machines, one for bonito northbridge, another for the 8259A controller in the southbridge. Both want to be non threaded. Signed-off-by: Wu Zhangjin Signed-off-by: Thomas Gleixner Cc: linux-mips@linux-mips.org Cc: Wu Zhangjin Patchwork: https://patchwork.linux-mips.org/patch/2638/ Signed-off-by: Ralf Baechle --- arch/mips/loongson/fuloong-2e/irq.c | 1 + arch/mips/loongson/lemote-2f/irq.c | 3 ++- 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/mips/loongson/fuloong-2e/irq.c b/arch/mips/loongson/fuloong-2e/irq.c index d61a04222b87..3cf1fef29f0e 100644 --- a/arch/mips/loongson/fuloong-2e/irq.c +++ b/arch/mips/loongson/fuloong-2e/irq.c @@ -42,6 +42,7 @@ asmlinkage void mach_irq_dispatch(unsigned int pending) static struct irqaction cascade_irqaction = { .handler = no_action, .name = "cascade", + .flags = IRQF_NO_THREAD, }; void __init mach_init_irq(void) diff --git a/arch/mips/loongson/lemote-2f/irq.c b/arch/mips/loongson/lemote-2f/irq.c index 081db102bb98..14b081841b6b 100644 --- a/arch/mips/loongson/lemote-2f/irq.c +++ b/arch/mips/loongson/lemote-2f/irq.c @@ -96,12 +96,13 @@ static irqreturn_t ip6_action(int cpl, void *dev_id) struct irqaction ip6_irqaction = { .handler = ip6_action, .name = "cascade", - .flags = IRQF_SHARED, + .flags = IRQF_SHARED | IRQF_NO_THREAD, }; struct irqaction cascade_irqaction = { .handler = no_action, .name = "cascade", + .flags = IRQF_NO_THREAD, }; void __init mach_init_irq(void) From 5c22cd407574988cc1580108e0df25ee64ea3d59 Mon Sep 17 00:00:00 2001 From: Liming Wang Date: Fri, 26 Aug 2011 07:00:04 +0800 Subject: [PATCH 11/75] MIPS: i8259: Mark cascade interrupt non-threaded Cascade interrupts cannot be threaded. Signed-off-by: Liming Wang Signed-off-by: Bruce Ashfield Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1314370804-21266-1-git-send-email-liming.wang@windriver.com Signed-off-by: Thomas Gleixner Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2770/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/i8259.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/mips/kernel/i8259.c b/arch/mips/kernel/i8259.c index 9b22db5dcef1..32b397b646ee 100644 --- a/arch/mips/kernel/i8259.c +++ b/arch/mips/kernel/i8259.c @@ -295,6 +295,7 @@ static void init_8259A(int auto_eoi) static struct irqaction irq2 = { .handler = no_action, .name = "cascade", + .flags = IRQF_NO_THREAD, }; static struct resource pic1_io_resource = { From 0f462e3c1281fd679b4c626612bfa59eb964bafd Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sat, 23 Jul 2011 12:41:25 +0000 Subject: [PATCH 12/75] MIPS: Allow forced irq threading All low level interrupts have been marked NO_THREAD, so MIPS can enjoy the wonderful world of forced threaded interrupt handlers. Signed-off-by: Thomas Gleixner To: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2639/ Signed-off-by: Ralf Baechle --- arch/mips/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index 177cdaf83564..d300c2bf8365 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -24,6 +24,7 @@ config MIPS select GENERIC_IRQ_PROBE select GENERIC_IRQ_SHOW select HAVE_ARCH_JUMP_LABEL + select IRQ_FORCED_THREADING menu "Machine selection" From 1f717929e987fc55bb6be02df994945f8edbd46c Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Wed, 27 Jul 2011 11:44:47 +0100 Subject: [PATCH 13/75] MIPS: Handle __put_user() sleeping. do_signal() does __put_user() which can fault, resulting in a might_sleep() warning in down_read(&mm->mmap_sem) and a "scheduling while atomic" warning when mmap_sem is contented. On Swarm this also results in: WARNING: at kernel/smp.c:459 smp_call_function_many+0x148/0x398() Modules linked in: Call Trace: [] dump_stack+0x1c/0x50 [] warn_slowpath_common+0x8c/0xc8 [] warn_slowpath_null+0x2c/0x40 [] smp_call_function_many+0x148/0x398 [] smp_call_function+0x58/0xa8 [] r4k_flush_data_cache_page+0x54/0xd8 [] handle_pte_fault+0xa9c/0xad0 [] handle_mm_fault+0x158/0x200 [] do_page_fault+0x218/0x3b0 [] ret_from_exception+0x0/0x10 [] copy_siginfo_to_user32+0x50/0x298 [] setup_rt_frame_32+0x90/0x250 [] do_notify_resume+0x154/0x358 [] work_notifysig+0xc/0x14 Fixed by enabling interrupts in do_notify_resume before delivering signals. [ralf@linux-mips.org: Reported and original fix by tglx but I wanted to minimize the amount of code being run with interrupts disabled so I moved the local_irq_disable() call right into do_notify_resume. Which is saner than doing it in entry.S.] Reported-by: Thomas Gleixner Signed-off-by: Ralf Baechle --- arch/mips/kernel/signal.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/mips/kernel/signal.c b/arch/mips/kernel/signal.c index dbbe0ce48d89..f8524003676a 100644 --- a/arch/mips/kernel/signal.c +++ b/arch/mips/kernel/signal.c @@ -8,6 +8,7 @@ * Copyright (C) 1999, 2000 Silicon Graphics, Inc. */ #include +#include #include #include #include @@ -658,6 +659,8 @@ static void do_signal(struct pt_regs *regs) asmlinkage void do_notify_resume(struct pt_regs *regs, void *unused, __u32 thread_info_flags) { + local_irq_enable(); + /* deal with pending signal delivery */ if (thread_info_flags & (_TIF_SIGPENDING | _TIF_RESTORE_SIGMASK)) do_signal(regs); From a705dc7cf4edbf110af19ea55cb5e5d0465adcf4 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Thu, 28 Jul 2011 13:31:48 +0100 Subject: [PATCH 14/75] MIPS: MSP71xx: Fix build error. MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit CC arch/mips/pmc-sierra/msp71xx/msp_irq.o /home/ralf/src/linux/linux-mips/arch/mips/pmc-sierra/msp71xx/msp_irq.c:112:2: error: request for member ‘flags’ in something not a structure or union /home/ralf/src/linux/linux-mips/arch/mips/pmc-sierra/msp71xx/msp_irq.c:118:2: error: request for member ‘flags’ in something not a structure or union make[4]: *** [arch/mips/pmc-sierra/msp71xx/msp_irq.o] Error 1 caused by 57336bc1056798d89714b7fb1b1d197e6bda6819 [MIPS: Mark cascade and low level interrupts IRQF_NO_THREAD]. Commas to separate struct initializers generally are considered useful to enhance the compilation experience of the user. Signed-off-by: Ralf Baechle --- arch/mips/pmc-sierra/msp71xx/msp_irq.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/mips/pmc-sierra/msp71xx/msp_irq.c b/arch/mips/pmc-sierra/msp71xx/msp_irq.c index 42a4c04e21d5..d3c3d81757a5 100644 --- a/arch/mips/pmc-sierra/msp71xx/msp_irq.c +++ b/arch/mips/pmc-sierra/msp71xx/msp_irq.c @@ -108,13 +108,13 @@ asmlinkage void plat_irq_dispatch(struct pt_regs *regs) static struct irqaction cic_cascade_msp = { .handler = no_action, - .name = "MSP CIC cascade" + .name = "MSP CIC cascade", .flags = IRQF_NO_THREAD, }; static struct irqaction per_cascade_msp = { .handler = no_action, - .name = "MSP PER cascade" + .name = "MSP PER cascade", .flags = IRQF_NO_THREAD, }; From c78c48827336b8aec2a5afcf249f0d0a7fe16859 Mon Sep 17 00:00:00 2001 From: Manuel Lauss Date: Tue, 2 Aug 2011 15:50:56 +0200 Subject: [PATCH 15/75] MIPS: Alchemy: Fix typo in MAC0 registration Harmless typo which prints an error message although MAC0 was registered successfully. Signed-off-by: Manuel Lauss To: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2672/ Signed-off-by: Ralf Baechle --- arch/mips/alchemy/common/platform.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/alchemy/common/platform.c b/arch/mips/alchemy/common/platform.c index 3b2c18b14341..f72c48d4804c 100644 --- a/arch/mips/alchemy/common/platform.c +++ b/arch/mips/alchemy/common/platform.c @@ -492,7 +492,7 @@ static void __init alchemy_setup_macs(int ctype) memcpy(au1xxx_eth0_platform_data.mac, ethaddr, 6); ret = platform_device_register(&au1xxx_eth0_device); - if (!ret) + if (ret) printk(KERN_INFO "Alchemy: failed to register MAC0\n"); From 870168a031b3c5b78191d719d375221f01e98e11 Mon Sep 17 00:00:00 2001 From: Manuel Lauss Date: Tue, 2 Aug 2011 15:50:58 +0200 Subject: [PATCH 16/75] MIPS: Alchemy: Include Au1100 in PM code. The current code forgets the Au1100 when looking for the correct method to suspend the chip. Signed-off-by: Manuel Lauss To: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2675/ Signed-off-by: Ralf Baechle --- arch/mips/alchemy/common/power.c | 22 ++++++++++++++-------- 1 file changed, 14 insertions(+), 8 deletions(-) diff --git a/arch/mips/alchemy/common/power.c b/arch/mips/alchemy/common/power.c index 647e518c90bc..b86324a42601 100644 --- a/arch/mips/alchemy/common/power.c +++ b/arch/mips/alchemy/common/power.c @@ -158,15 +158,21 @@ static void restore_core_regs(void) void au_sleep(void) { - int cpuid = alchemy_get_cputype(); - if (cpuid != ALCHEMY_CPU_UNKNOWN) { - save_core_regs(); - if (cpuid <= ALCHEMY_CPU_AU1500) - alchemy_sleep_au1000(); - else if (cpuid <= ALCHEMY_CPU_AU1200) - alchemy_sleep_au1550(); - restore_core_regs(); + save_core_regs(); + + switch (alchemy_get_cputype()) { + case ALCHEMY_CPU_AU1000: + case ALCHEMY_CPU_AU1500: + case ALCHEMY_CPU_AU1100: + alchemy_sleep_au1000(); + break; + case ALCHEMY_CPU_AU1550: + case ALCHEMY_CPU_AU1200: + alchemy_sleep_au1550(); + break; } + + restore_core_regs(); } #endif /* CONFIG_PM */ From d954ffe34a606d4ec34cd3c2b0b5974779759f25 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Tue, 2 Aug 2011 22:52:48 +0100 Subject: [PATCH 17/75] MIPS: tlbex: Fix build error in R3000 code. Only some GCC versions such as gcc 4.2 notice that the variable wr in build_r3000_tlb_modify_handler is used uninitialized. When using one of those GCCs the build will fail due to -Werror. GCC 4.6 does not warn about the uninitialized use of wr. This issue was introduced by 7211f4d7a3dcbe57c5d396c334dca525315dceb2 [MIPS: Close races in TLB modify handlers.] Reported-by: Ganesan Ramalingam Signed-off-by: Ralf Baechle --- arch/mips/mm/tlbex.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/arch/mips/mm/tlbex.c b/arch/mips/mm/tlbex.c index b6e1cff50667..46f33c7c1a09 100644 --- a/arch/mips/mm/tlbex.c +++ b/arch/mips/mm/tlbex.c @@ -1759,14 +1759,13 @@ static void __cpuinit build_r3000_tlb_modify_handler(void) u32 *p = handle_tlbm; struct uasm_label *l = labels; struct uasm_reloc *r = relocs; - struct work_registers wr; memset(handle_tlbm, 0, sizeof(handle_tlbm)); memset(labels, 0, sizeof(labels)); memset(relocs, 0, sizeof(relocs)); build_r3000_tlbchange_handler_head(&p, K0, K1); - build_pte_modifiable(&p, &r, wr.r1, wr.r2, wr.r3, label_nopage_tlbm); + build_pte_modifiable(&p, &r, K0, K1, -1, label_nopage_tlbm); uasm_i_nop(&p); /* load delay */ build_make_write(&p, &r, K0, K1); build_r3000_pte_reload_tlbwi(&p, K0, K1); From 77fbdb30f045084f1c65b72bd457d7406343aad7 Mon Sep 17 00:00:00 2001 From: John Crispin Date: Mon, 18 Jul 2011 18:08:40 +0200 Subject: [PATCH 18/75] MIPS: Lantiq: Fix external interrupt sources The irq base offset needs to be ignored when matching irqs to external interrupt pins. Taking the offset into account resulted in the EIU not being brought up properly. Signed-off-by: John Crispin Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2616/ Signed-off-by: Ralf Baechle --- arch/mips/lantiq/irq.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/arch/mips/lantiq/irq.c b/arch/mips/lantiq/irq.c index fc89795cafdb..f9737bb3c5ab 100644 --- a/arch/mips/lantiq/irq.c +++ b/arch/mips/lantiq/irq.c @@ -123,11 +123,10 @@ void ltq_enable_irq(struct irq_data *d) static unsigned int ltq_startup_eiu_irq(struct irq_data *d) { int i; - int irq_nr = d->irq - INT_NUM_IRQ0; ltq_enable_irq(d); for (i = 0; i < MAX_EIU; i++) { - if (irq_nr == ltq_eiu_irq[i]) { + if (d->irq == ltq_eiu_irq[i]) { /* low level - we should really handle set_type */ ltq_eiu_w32(ltq_eiu_r32(LTQ_EIU_EXIN_C) | (0x6 << (i * 4)), LTQ_EIU_EXIN_C); @@ -147,11 +146,10 @@ static unsigned int ltq_startup_eiu_irq(struct irq_data *d) static void ltq_shutdown_eiu_irq(struct irq_data *d) { int i; - int irq_nr = d->irq - INT_NUM_IRQ0; ltq_disable_irq(d); for (i = 0; i < MAX_EIU; i++) { - if (irq_nr == ltq_eiu_irq[i]) { + if (d->irq == ltq_eiu_irq[i]) { /* disable */ ltq_eiu_w32(ltq_eiu_r32(LTQ_EIU_EXIN_INEN) & ~(1 << i), LTQ_EIU_EXIN_INEN); From 0596954681859f1b7b7adfd58fa738c09ed2affa Mon Sep 17 00:00:00 2001 From: John Crispin Date: Mon, 18 Jul 2011 18:04:12 +0200 Subject: [PATCH 19/75] MIPS: Lantiq: Fix setting the PCI bus speed on AR9 The bits used to set the PCI bus speed on AR9 are slightly different to those used on Danube. Signed-off-by: John Crispin Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2614/ Signed-off-by: Ralf Baechle --- arch/mips/pci/pci-lantiq.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/arch/mips/pci/pci-lantiq.c b/arch/mips/pci/pci-lantiq.c index 603d7493e966..8656388b34bd 100644 --- a/arch/mips/pci/pci-lantiq.c +++ b/arch/mips/pci/pci-lantiq.c @@ -171,8 +171,13 @@ static int __devinit ltq_pci_startup(struct ltq_pci_data *conf) u32 temp_buffer; /* set clock to 33Mhz */ - ltq_cgu_w32(ltq_cgu_r32(LTQ_CGU_IFCCR) & ~0xf00000, LTQ_CGU_IFCCR); - ltq_cgu_w32(ltq_cgu_r32(LTQ_CGU_IFCCR) | 0x800000, LTQ_CGU_IFCCR); + if (ltq_is_ar9()) { + ltq_cgu_w32(ltq_cgu_r32(LTQ_CGU_IFCCR) & ~0x1f00000, LTQ_CGU_IFCCR); + ltq_cgu_w32(ltq_cgu_r32(LTQ_CGU_IFCCR) | 0xe00000, LTQ_CGU_IFCCR); + } else { + ltq_cgu_w32(ltq_cgu_r32(LTQ_CGU_IFCCR) & ~0xf00000, LTQ_CGU_IFCCR); + ltq_cgu_w32(ltq_cgu_r32(LTQ_CGU_IFCCR) | 0x800000, LTQ_CGU_IFCCR); + } /* external or internal clock ? */ if (conf->clock) { From b7867f1bfcb76c75d98d35f576fcd9d7759a96fe Mon Sep 17 00:00:00 2001 From: John Crispin Date: Wed, 3 Aug 2011 13:37:24 +0100 Subject: [PATCH 20/75] SERIAL: Lantiq: Set timeout in uart_port Without this patch apps using readline hang. Signed-off-by: Thomas Langer Signed-off-by: John Crispin Cc: linux-mips@linux-mips.org Cc: linux-serial@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/2687/ Signed-off-by: Alan Cox Signed-off-by: Ralf Baechle --- drivers/tty/serial/lantiq.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/tty/serial/lantiq.c b/drivers/tty/serial/lantiq.c index 58cf279ed879..bc95f52cad8b 100644 --- a/drivers/tty/serial/lantiq.c +++ b/drivers/tty/serial/lantiq.c @@ -478,8 +478,10 @@ lqasc_set_termios(struct uart_port *port, spin_unlock_irqrestore(<q_asc_lock, flags); /* Don't rewrite B0 */ - if (tty_termios_baud_rate(new)) + if (tty_termios_baud_rate(new)) tty_termios_encode_baud_rate(new, baud, baud); + + uart_update_timeout(port, cflag, baud); } static const char* From dd0a028183369cccc0826199a7ccdc850ece289b Mon Sep 17 00:00:00 2001 From: Manuel Lauss Date: Fri, 12 Aug 2011 08:28:35 +0200 Subject: [PATCH 21/75] MIPS: Alchemy: DB1200: Disable cascade IRQ in handler Disable the cascade IRQ in the cascade handler. This is required to get the DB1300 working, and also gets rid of all spurious interrupts previously observed on the DB1200; so Config[OD] can be disabled again for better performance. Signed-off-by: Manuel Lauss To: Linux-MIPS Patchwork: https://patchwork.linux-mips.org/patch/2708/ Signed-off-by: Ralf Baechle --- arch/mips/alchemy/devboards/bcsr.c | 4 ++++ arch/mips/alchemy/devboards/db1200/setup.c | 7 ------- 2 files changed, 4 insertions(+), 7 deletions(-) diff --git a/arch/mips/alchemy/devboards/bcsr.c b/arch/mips/alchemy/devboards/bcsr.c index 596ad00e7f05..463d2c4d9441 100644 --- a/arch/mips/alchemy/devboards/bcsr.c +++ b/arch/mips/alchemy/devboards/bcsr.c @@ -89,8 +89,12 @@ static void bcsr_csc_handler(unsigned int irq, struct irq_desc *d) { unsigned short bisr = __raw_readw(bcsr_virt + BCSR_REG_INTSTAT); + disable_irq_nosync(irq); + for ( ; bisr; bisr &= bisr - 1) generic_handle_irq(bcsr_csc_base + __ffs(bisr)); + + enable_irq(irq); } /* NOTE: both the enable and mask bits must be cleared, otherwise the diff --git a/arch/mips/alchemy/devboards/db1200/setup.c b/arch/mips/alchemy/devboards/db1200/setup.c index 1dac4f27d334..4a8980027ecf 100644 --- a/arch/mips/alchemy/devboards/db1200/setup.c +++ b/arch/mips/alchemy/devboards/db1200/setup.c @@ -23,13 +23,6 @@ void __init board_setup(void) unsigned long freq0, clksrc, div, pfc; unsigned short whoami; - /* Set Config[OD] (disable overlapping bus transaction): - * This gets rid of a _lot_ of spurious interrupts (especially - * wrt. IDE); but incurs ~10% performance hit in some - * cpu-bound applications. - */ - set_c0_config(1 << 19); - bcsr_init(DB1200_BCSR_PHYS_ADDR, DB1200_BCSR_PHYS_ADDR + DB1200_BCSR_HEXLED_OFS); From b8ecf341f1b7ca7bf3cb80c48cbbae33b79947d0 Mon Sep 17 00:00:00 2001 From: Arnaud Lacombe Date: Mon, 15 Aug 2011 13:07:05 -0400 Subject: [PATCH 22/75] MIPS: Do not use EXTRA_CFLAGS Usage of these flags has been deprecated for nearly 4 years by: commit f77bf01425b11947eeb3b5b54685212c302741b8 Author: Sam Ravnborg Date: Mon Oct 15 22:25:06 2007 +0200 kbuild: introduce ccflags-y, asflags-y and ldflags-y Moreover, these flags (at least EXTRA_CFLAGS) have been documented for command line use. By default, gmake(1) do not override command line setting, so this is likely to result in build failure or unexpected behavior. Replace their usage by Kbuild's `{as,cc,ld}flags-y'. To: linux-kernel@vger.kernel.org Cc: Sam Ravnborg Patchwork: https://patchwork.linux-mips.org/patch/2710/ Cc: linux-mips@linux-mips.org Signed-off-by: Arnaud Lacombe Signed-off-by: Ralf Baechle --- arch/mips/netlogic/xlr/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/netlogic/xlr/Makefile b/arch/mips/netlogic/xlr/Makefile index 9bd3f731f62e..2dca585dd2f7 100644 --- a/arch/mips/netlogic/xlr/Makefile +++ b/arch/mips/netlogic/xlr/Makefile @@ -2,4 +2,4 @@ obj-y += setup.o platform.o irq.o setup.o time.o obj-$(CONFIG_SMP) += smp.o smpboot.o obj-$(CONFIG_EARLY_PRINTK) += xlr_console.o -EXTRA_CFLAGS += -Werror +ccflags-y += -Werror From 1eec6cd08b4feb72a73aff468ab72bdd21e1dc61 Mon Sep 17 00:00:00 2001 From: Yong Zhang Date: Tue, 16 Aug 2011 09:54:54 +0800 Subject: [PATCH 23/75] MIPS: Compat: Use 32-bit wrapper for compat_sys_futex. We can't trust userspace to pass signed-extend arguments. Not correctly sign-extended arguments to futex-wait result in architecturally undefined operation of 32-bit arithmetic instructions. For example, if 'val' is too big and bit-31 is 1, the caller may enter endless loop at: futex_wait_setup() { ... if (uval != val) { queue_unlock(q, *hb); ret = -EWOULDBLOCK; ... } Signed-off-by: Yong Zhang To: linux-mips@linux-mips.org To: linux-kernel@vger.kernel.org Patchwork: https://patchwork.linux-mips.org/patch/2714/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/linux32.c | 7 +++++++ arch/mips/kernel/scall64-n32.S | 2 +- arch/mips/kernel/scall64-o32.S | 2 +- 3 files changed, 9 insertions(+), 2 deletions(-) diff --git a/arch/mips/kernel/linux32.c b/arch/mips/kernel/linux32.c index 876a75cc376f..922a554cd108 100644 --- a/arch/mips/kernel/linux32.c +++ b/arch/mips/kernel/linux32.c @@ -349,3 +349,10 @@ SYSCALL_DEFINE6(32_fanotify_mark, int, fanotify_fd, unsigned int, flags, return sys_fanotify_mark(fanotify_fd, flags, merge_64(a3, a4), dfd, pathname); } + +SYSCALL_DEFINE6(32_futex, u32 __user *, uaddr, int, op, u32, val, + struct compat_timespec __user *, utime, u32 __user *, uaddr2, + u32, val3) +{ + return compat_sys_futex(uaddr, op, val, utime, uaddr2, val3); +} diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S index f9296e894e46..6de1f598346e 100644 --- a/arch/mips/kernel/scall64-n32.S +++ b/arch/mips/kernel/scall64-n32.S @@ -315,7 +315,7 @@ EXPORT(sysn32_call_table) PTR sys_fremovexattr PTR sys_tkill PTR sys_ni_syscall - PTR compat_sys_futex + PTR sys_32_futex PTR compat_sys_sched_setaffinity /* 6195 */ PTR compat_sys_sched_getaffinity PTR sys_cacheflush diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S index 4d7c9827706f..1d813169e453 100644 --- a/arch/mips/kernel/scall64-o32.S +++ b/arch/mips/kernel/scall64-o32.S @@ -441,7 +441,7 @@ sys_call_table: PTR sys_fremovexattr /* 4235 */ PTR sys_tkill PTR sys_sendfile64 - PTR compat_sys_futex + PTR sys_32_futex PTR compat_sys_sched_setaffinity PTR compat_sys_sched_getaffinity /* 4240 */ PTR compat_sys_io_setup From bd7100099a46b59f433dd15ad60adbb4d4f3d625 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Thu, 2 Jun 2011 21:06:48 +0200 Subject: [PATCH 24/75] MIPS: PM: Use struct syscore_ops instead of sysdevs for PM (v2) Convert some MIPS architecture's code to using struct syscore_ops objects for power management instead of sysdev classes and sysdevs. This simplifies the code and reduces the kernel's memory footprint. It also is necessary for removing sysdevs from the kernel entirely in the future. Signed-off-by: Rafael J. Wysocki Acked-by: Greg Kroah-Hartman Acked-and-tested-by: Lars-Peter Clausen Signed-off-by: Lars-Peter Clausen Cc: linux-kernel@vger.kernel.org Cc: "Rafael J. Wysocki" Patchwork: http://patchwork.linux-mips.org/patch/2431/ Signed-off-by: Ralf Baechle --- arch/mips/jz4740/gpio.c | 56 ++++++++++++++++++----------------------- 1 file changed, 24 insertions(+), 32 deletions(-) diff --git a/arch/mips/jz4740/gpio.c b/arch/mips/jz4740/gpio.c index 73031f7fc827..4397972949fa 100644 --- a/arch/mips/jz4740/gpio.c +++ b/arch/mips/jz4740/gpio.c @@ -18,7 +18,7 @@ #include #include -#include +#include #include #include #include @@ -86,7 +86,6 @@ struct jz_gpio_chip { spinlock_t lock; struct gpio_chip gpio_chip; - struct sys_device sysdev; }; static struct jz_gpio_chip jz4740_gpio_chips[]; @@ -459,49 +458,47 @@ static struct jz_gpio_chip jz4740_gpio_chips[] = { JZ4740_GPIO_CHIP(D), }; -static inline struct jz_gpio_chip *sysdev_to_chip(struct sys_device *dev) +static void jz4740_gpio_suspend_chip(struct jz_gpio_chip *chip) { - return container_of(dev, struct jz_gpio_chip, sysdev); -} - -static int jz4740_gpio_suspend(struct sys_device *dev, pm_message_t state) -{ - struct jz_gpio_chip *chip = sysdev_to_chip(dev); - chip->suspend_mask = readl(chip->base + JZ_REG_GPIO_MASK); writel(~(chip->wakeup), chip->base + JZ_REG_GPIO_MASK_SET); writel(chip->wakeup, chip->base + JZ_REG_GPIO_MASK_CLEAR); +} + +static int jz4740_gpio_suspend(void) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(jz4740_gpio_chips); i++) + jz4740_gpio_suspend_chip(&jz4740_gpio_chips[i]); return 0; } -static int jz4740_gpio_resume(struct sys_device *dev) +static void jz4740_gpio_resume_chip(struct jz_gpio_chip *chip) { - struct jz_gpio_chip *chip = sysdev_to_chip(dev); uint32_t mask = chip->suspend_mask; writel(~mask, chip->base + JZ_REG_GPIO_MASK_CLEAR); writel(mask, chip->base + JZ_REG_GPIO_MASK_SET); - - return 0; } -static struct sysdev_class jz4740_gpio_sysdev_class = { - .name = "gpio", +static void jz4740_gpio_resume(void) +{ + int i; + + for (i = ARRAY_SIZE(jz4740_gpio_chips) - 1; i >= 0 ; i--) + jz4740_gpio_resume_chip(&jz4740_gpio_chips[i]); +} + +static struct syscore_ops jz4740_gpio_syscore_ops = { .suspend = jz4740_gpio_suspend, .resume = jz4740_gpio_resume, }; -static int jz4740_gpio_chip_init(struct jz_gpio_chip *chip, unsigned int id) +static void jz4740_gpio_chip_init(struct jz_gpio_chip *chip, unsigned int id) { - int ret, irq; - - chip->sysdev.id = id; - chip->sysdev.cls = &jz4740_gpio_sysdev_class; - ret = sysdev_register(&chip->sysdev); - - if (ret) - return ret; + int irq; spin_lock_init(&chip->lock); @@ -519,22 +516,17 @@ static int jz4740_gpio_chip_init(struct jz_gpio_chip *chip, unsigned int id) irq_set_chip_and_handler(irq, &jz_gpio_irq_chip, handle_level_irq); } - - return 0; } static int __init jz4740_gpio_init(void) { unsigned int i; - int ret; - - ret = sysdev_class_register(&jz4740_gpio_sysdev_class); - if (ret) - return ret; for (i = 0; i < ARRAY_SIZE(jz4740_gpio_chips); ++i) jz4740_gpio_chip_init(&jz4740_gpio_chips[i], i); + register_syscore_ops(&jz4740_gpio_syscore_ops); + printk(KERN_INFO "JZ4740 GPIO initialized\n"); return 0; From 465aaed0030b23d905f3eca5e65481773bea9ea1 Mon Sep 17 00:00:00 2001 From: David Daney Date: Sat, 20 Aug 2011 08:44:00 -0700 Subject: [PATCH 25/75] MIPS: Octeon: Select CONFIG_HOLES_IN_ZONE Current Octeon systems do in fact have holes in their memory zones. We need to select HOLES_IN_ZONE. If we do not, some memory configurations will result in crashes at boot time like this: . . . CPU 6 Unable to handle kernel paging request at virtual address 0000000000700000, epc == ffffffff8118fe00, ra == ffffffff8118fe9c Oops[#1]: Cpu 6 . . . ... Call Trace: [] setup_per_zone_wmarks+0x1b0/0x338 [] init_per_zone_wmark_min+0x64/0xd0 [] do_one_initcall+0x38/0x160 . . . Reported-by: Jason Kwon Signed-off-by: David Daney To: linux-mips@linux-mips.org Cc: Jason Kwon Patchwork: https://patchwork.linux-mips.org/patch/2724/ Tested-by: Guenter Roeck Signed-off-by: Ralf Baechle --- arch/mips/Kconfig | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index d300c2bf8365..b122adc8bdbb 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -723,6 +723,7 @@ config CAVIUM_OCTEON_SIMULATOR select SYS_SUPPORTS_HIGHMEM select SYS_SUPPORTS_HOTPLUG_CPU select SYS_HAS_CPU_CAVIUM_OCTEON + select HOLES_IN_ZONE help The Octeon simulator is software performance model of the Cavium Octeon Processor. It supports simulating Octeon processors on x86 @@ -745,6 +746,7 @@ config CAVIUM_OCTEON_REFERENCE_BOARD select ZONE_DMA32 select USB_ARCH_HAS_OHCI select USB_ARCH_HAS_EHCI + select HOLES_IN_ZONE help This option supports all of the Octeon reference boards from Cavium Networks. It builds a kernel that dynamically determines the Octeon @@ -974,6 +976,9 @@ config ISA_DMA_API config GENERIC_GPIO bool +config HOLES_IN_ZONE + bool + # # Endianess selection. Sufficiently obscure so many users don't know what to # answer,so we try hard to limit the available choices. Also the use of a From d968275921f9fa7fbc602ac6618ffec6a062ee3c Mon Sep 17 00:00:00 2001 From: David Daney Date: Tue, 30 Aug 2011 06:45:20 -0700 Subject: [PATCH 26/75] MIPS: Don't clobber CP0_STATUS value for CONFIG_MIPS_MT_SMTC Reported-by: Edgar E. Iglesias Signed-off-by: David Daney Patchwork: https://patchwork.linux-mips.org/patch/2753/ Signed-off-by: Ralf Baechle --- arch/mips/include/asm/stackframe.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/mips/include/asm/stackframe.h b/arch/mips/include/asm/stackframe.h index b4ba2449444b..cb41af5f3406 100644 --- a/arch/mips/include/asm/stackframe.h +++ b/arch/mips/include/asm/stackframe.h @@ -195,9 +195,9 @@ * to cover the pipeline delay. */ .set mips32 - mfc0 v1, CP0_TCSTATUS + mfc0 k0, CP0_TCSTATUS .set mips0 - LONG_S v1, PT_TCSTATUS(sp) + LONG_S k0, PT_TCSTATUS(sp) #endif /* CONFIG_MIPS_MT_SMTC */ LONG_S $4, PT_R4(sp) LONG_S $5, PT_R5(sp) From 0f4ccbc835036cbcc2513585bb2e93ee62e12674 Mon Sep 17 00:00:00 2001 From: David Daney Date: Fri, 16 Sep 2011 18:06:02 -0700 Subject: [PATCH 27/75] MIPS: No branches in delay slots for huge pages in handle_tlbl For the case PM_DEFAULT_MASK == 0, we were placing a branch in the delay slot of another branch. This leads to undefined behavior. Signed-off-by: David Daney Cc: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2775/ Signed-off-by: Ralf Baechle --- arch/mips/mm/tlbex.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/mips/mm/tlbex.c b/arch/mips/mm/tlbex.c index 46f33c7c1a09..e06370f58ef3 100644 --- a/arch/mips/mm/tlbex.c +++ b/arch/mips/mm/tlbex.c @@ -1962,7 +1962,8 @@ static void __cpuinit build_r4000_tlb_load_handler(void) uasm_i_andi(&p, wr.r3, wr.r3, 2); uasm_il_beqz(&p, &r, wr.r3, label_tlbl_goaround2); } - + if (PM_DEFAULT_MASK == 0) + uasm_i_nop(&p); /* * We clobbered C0_PAGEMASK, restore it. On the other branch * it is restored in build_huge_tlb_write_entry. From 2f19d080fb14bdddf11bf54d4db6306235c46c99 Mon Sep 17 00:00:00 2001 From: David Daney Date: Tue, 5 Jul 2011 16:35:53 -0700 Subject: [PATCH 28/75] MIPS: Octeon: Enable C0_UserLocal probing. Octeon2 processor cores have a UserLocal register. Remove the hard coded negative probe and allow the standard probing to detect this feature. Signed-off-by: David Daney To: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2578/ Signed-off-by: Ralf Baechle --- arch/mips/include/asm/mach-cavium-octeon/cpu-feature-overrides.h | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/mips/include/asm/mach-cavium-octeon/cpu-feature-overrides.h b/arch/mips/include/asm/mach-cavium-octeon/cpu-feature-overrides.h index 0d5a42b5f47a..a58addb98cfd 100644 --- a/arch/mips/include/asm/mach-cavium-octeon/cpu-feature-overrides.h +++ b/arch/mips/include/asm/mach-cavium-octeon/cpu-feature-overrides.h @@ -54,7 +54,6 @@ #define cpu_has_mips_r2_exec_hazard 0 #define cpu_has_dsp 0 #define cpu_has_mipsmt 0 -#define cpu_has_userlocal 0 #define cpu_has_vint 0 #define cpu_has_veic 0 #define cpu_hwrena_impl_bits 0xc0000000 From 68e58a294fb26f692697179e3f3ecf88dd8cb97c Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Sun, 25 Sep 2011 23:26:15 +0100 Subject: [PATCH 29/75] dm: flakey fix corrupt_bio_byte error path If no arguments were provided to the corrupt_bio_byte feature an error should be returned immediately. Reported-by: Zdenek Kabelac Signed-off-by: Mike Snitzer Signed-off-by: Alasdair G Kergon --- drivers/md/dm-flakey.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/md/dm-flakey.c b/drivers/md/dm-flakey.c index 89f73ca22cfa..f84c08029b21 100644 --- a/drivers/md/dm-flakey.c +++ b/drivers/md/dm-flakey.c @@ -81,8 +81,10 @@ static int parse_features(struct dm_arg_set *as, struct flakey_c *fc, * corrupt_bio_byte */ if (!strcasecmp(arg_name, "corrupt_bio_byte")) { - if (!argc) + if (!argc) { ti->error = "Feature corrupt_bio_byte requires parameters"; + return -EINVAL; + } r = dm_read_arg(_args + 1, as, &fc->corrupt_bio_byte, &ti->error); if (r) From 876fbba1db4a377f050a2bb49b474c7527b2995d Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Sun, 25 Sep 2011 23:26:17 +0100 Subject: [PATCH 30/75] dm table: avoid crash if integrity profile changes Commit a63a5cf (dm: improve block integrity support) introduced a two-phase initialization of a DM device's integrity profile. This patch avoids dereferencing a NULL 'template_disk' pointer in blk_integrity_register() if there is an integrity profile mismatch in dm_table_set_integrity(). This can occur if the integrity profiles for stacked devices in a DM table are changed between the call to dm_table_prealloc_integrity() and dm_table_set_integrity(). Reported-by: Zdenek Kabelac Signed-off-by: Mike Snitzer Signed-off-by: Alasdair G Kergon Cc: stable@kernel.org # 2.6.39 --- drivers/md/dm-table.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index 986b8754bb08..322669807077 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -1238,14 +1238,15 @@ static void dm_table_set_integrity(struct dm_table *t) return; template_disk = dm_table_get_integrity_disk(t, true); - if (!template_disk && - blk_integrity_is_initialized(dm_disk(t->md))) { + if (template_disk) + blk_integrity_register(dm_disk(t->md), + blk_get_integrity(template_disk)); + else if (blk_integrity_is_initialized(dm_disk(t->md))) DMWARN("%s: device no longer has a valid integrity profile", dm_device_name(t->md)); - return; - } - blk_integrity_register(dm_disk(t->md), - blk_get_integrity(template_disk)); + else + DMWARN("%s: unable to establish an integrity profile", + dm_device_name(t->md)); } static int device_flush_capable(struct dm_target *ti, struct dm_dev *dev, From 8232480944d173378082ebb2cac8a3207c08cf31 Mon Sep 17 00:00:00 2001 From: Jonthan Brassow Date: Sun, 25 Sep 2011 23:26:19 +0100 Subject: [PATCH 31/75] dm: raid fix write_mostly arg validation Fix off-by-one error in validation of write_mostly. The user-supplied value given for the 'write_mostly' argument must be an index starting at 0. The validation of the supplied argument failed to check for 'N' ('>' vs '>='), which would have caused an access beyond the end of the array. Reported-by: Doug Ledford Signed-off-by: Jonathan Brassow Signed-off-by: Alasdair G Kergon --- drivers/md/dm-raid.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/md/dm-raid.c b/drivers/md/dm-raid.c index a002dd85db1e..86df8b2cf927 100644 --- a/drivers/md/dm-raid.c +++ b/drivers/md/dm-raid.c @@ -449,7 +449,7 @@ static int parse_raid_params(struct raid_set *rs, char **argv, rs->ti->error = "write_mostly option is only valid for RAID1"; return -EINVAL; } - if (value > rs->md.raid_disks) { + if (value >= rs->md.raid_disks) { rs->ti->error = "Invalid write_mostly drive index given"; return -EINVAL; } From 983c7db347db8ce2d8453fd1d89b7a4bb6920d56 Mon Sep 17 00:00:00 2001 From: Milan Broz Date: Sun, 25 Sep 2011 23:26:21 +0100 Subject: [PATCH 32/75] dm crypt: always disable discard_zeroes_data If optional discard support in dm-crypt is enabled, discards requests bypass the crypt queue and blocks of the underlying device are discarded. For the read path, discarded blocks are handled the same as normal ciphertext blocks, thus decrypted. So if the underlying device announces discarded regions return zeroes, dm-crypt must disable this flag because after decryption there is just random noise instead of zeroes. Signed-off-by: Milan Broz Signed-off-by: Alasdair G Kergon --- drivers/md/dm-crypt.c | 2 ++ drivers/md/dm-table.c | 19 +++++++++++++++++++ include/linux/device-mapper.h | 5 +++++ 3 files changed, 26 insertions(+) diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c index 49da55c1528a..8c2a000cf3f5 100644 --- a/drivers/md/dm-crypt.c +++ b/drivers/md/dm-crypt.c @@ -1698,6 +1698,8 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv) } ti->num_flush_requests = 1; + ti->discard_zeroes_data_unsupported = 1; + return 0; bad: diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c index 322669807077..bc04518e9d8b 100644 --- a/drivers/md/dm-table.c +++ b/drivers/md/dm-table.c @@ -1283,6 +1283,22 @@ static bool dm_table_supports_flush(struct dm_table *t, unsigned flush) return 0; } +static bool dm_table_discard_zeroes_data(struct dm_table *t) +{ + struct dm_target *ti; + unsigned i = 0; + + /* Ensure that all targets supports discard_zeroes_data. */ + while (i < dm_table_get_num_targets(t)) { + ti = dm_table_get_target(t, i++); + + if (ti->discard_zeroes_data_unsupported) + return 0; + } + + return 1; +} + void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q, struct queue_limits *limits) { @@ -1305,6 +1321,9 @@ void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q, } blk_queue_flush(q, flush); + if (!dm_table_discard_zeroes_data(t)) + q->limits.discard_zeroes_data = 0; + dm_table_set_integrity(t); /* diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h index 3fa1f3d90ce0..99e3e50b5c57 100644 --- a/include/linux/device-mapper.h +++ b/include/linux/device-mapper.h @@ -197,6 +197,11 @@ struct dm_target { * whether or not its underlying devices have support. */ unsigned discards_supported:1; + + /* + * Set if this target does not return zeroes on discarded blocks. + */ + unsigned discard_zeroes_data_unsupported:1; }; /* Each target can link one of these into the table */ From a4be637b34a543af5b5421a8ee0ee04d30f5f58e Mon Sep 17 00:00:00 2001 From: "Roy.Li" Date: Mon, 26 Sep 2011 09:08:00 +0800 Subject: [PATCH 33/75] staging/octeon: Software should check the checksum of no tcp/udp packets Icmp packets with wrong checksum are never dropped since skb->ip_summed is set to CHECKSUM_UNNECESSARY. When icmp packets with wrong checksum pass through the octeon net driver, the not_IP, IP_exc, L4_error hardware indicators show no error. so the driver sets CHECKSUM_UNNECESSARY on skb->ip_summed. L4_error only works for TCP/UDP, not for ICMP. Signed-off-by: Roy.Li To: linux-mips@linux-mips.org Cc: netdev@vger.kernel.org Cc: ralf@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2798/ Acked-by: David Daney Cc: Greg KH --- drivers/staging/octeon/ethernet-rx.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/staging/octeon/ethernet-rx.c b/drivers/staging/octeon/ethernet-rx.c index 1a7c19ae766f..8b307b428791 100644 --- a/drivers/staging/octeon/ethernet-rx.c +++ b/drivers/staging/octeon/ethernet-rx.c @@ -411,7 +411,8 @@ static int cvm_oct_napi_poll(struct napi_struct *napi, int budget) skb->protocol = eth_type_trans(skb, dev); skb->dev = dev; - if (unlikely(work->word2.s.not_IP || work->word2.s.IP_exc || work->word2.s.L4_error)) + if (unlikely(work->word2.s.not_IP || work->word2.s.IP_exc || + work->word2.s.L4_error || !work->word2.s.tcp_or_udp)) skb->ip_summed = CHECKSUM_NONE; else skb->ip_summed = CHECKSUM_UNNECESSARY; From 3696d303d6a39f916f9ace37d79e3db7937c68a7 Mon Sep 17 00:00:00 2001 From: Bryan Buckley Date: Fri, 30 Sep 2011 11:05:55 -0700 Subject: [PATCH 34/75] ARM: OMAP4: MMC: fix power and audio issue, decouple USBC1 from MMC1 Remove OMAP4_USBC1_ICUSB_PWRDNZ_MASK during enable/disable PWRDNZ mode for MMC1_PBIAS and associated extended-drain MMC1 I/O cell. This is in accordance with the control module programming guide. This fixes a bug where if trying to use gpio_98 or gpio_99 and MMC1 at the same time the GPIO signal will be affected by a changing SDMMC1_VDDS. Software must keep MMC1_PBIAS cell and MMC1_IO cell PWRDNZ signals low whenever SDMMC1_VDDS ramps up/down or changes for cell protection purposes. MMC1 is based on SDMMC1_VDDS whereas USBC1 is based on SIM_VDDS therefore they can operate independently. Signed-off-by: Bryan Buckley Acked-by: Kishore Kadiyala Tested-by: Balaji T K Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/hsmmc.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/arch/arm/mach-omap2/hsmmc.c b/arch/arm/mach-omap2/hsmmc.c index a9b45c76e1d3..097a42d81e59 100644 --- a/arch/arm/mach-omap2/hsmmc.c +++ b/arch/arm/mach-omap2/hsmmc.c @@ -137,8 +137,7 @@ static void omap4_hsmmc1_before_set_reg(struct device *dev, int slot, */ reg = omap4_ctrl_pad_readl(control_pbias_offset); reg &= ~(OMAP4_MMC1_PBIASLITE_PWRDNZ_MASK | - OMAP4_MMC1_PWRDNZ_MASK | - OMAP4_USBC1_ICUSB_PWRDNZ_MASK); + OMAP4_MMC1_PWRDNZ_MASK); omap4_ctrl_pad_writel(reg, control_pbias_offset); } @@ -156,8 +155,7 @@ static void omap4_hsmmc1_after_set_reg(struct device *dev, int slot, else reg |= OMAP4_MMC1_PBIASLITE_VMODE_MASK; reg |= (OMAP4_MMC1_PBIASLITE_PWRDNZ_MASK | - OMAP4_MMC1_PWRDNZ_MASK | - OMAP4_USBC1_ICUSB_PWRDNZ_MASK); + OMAP4_MMC1_PWRDNZ_MASK); omap4_ctrl_pad_writel(reg, control_pbias_offset); timeout = jiffies + msecs_to_jiffies(5); @@ -171,16 +169,14 @@ static void omap4_hsmmc1_after_set_reg(struct device *dev, int slot, if (reg & OMAP4_MMC1_PBIASLITE_VMODE_ERROR_MASK) { pr_err("Pbias Voltage is not same as LDO\n"); /* Caution : On VMODE_ERROR Power Down MMC IO */ - reg &= ~(OMAP4_MMC1_PWRDNZ_MASK | - OMAP4_USBC1_ICUSB_PWRDNZ_MASK); + reg &= ~(OMAP4_MMC1_PWRDNZ_MASK); omap4_ctrl_pad_writel(reg, control_pbias_offset); } } else { reg = omap4_ctrl_pad_readl(control_pbias_offset); reg |= (OMAP4_MMC1_PBIASLITE_PWRDNZ_MASK | OMAP4_MMC1_PWRDNZ_MASK | - OMAP4_MMC1_PBIASLITE_VMODE_MASK | - OMAP4_USBC1_ICUSB_PWRDNZ_MASK); + OMAP4_MMC1_PBIASLITE_VMODE_MASK); omap4_ctrl_pad_writel(reg, control_pbias_offset); } } From bfd46a543f8f90bb197225900e355ce3c7daeb1f Mon Sep 17 00:00:00 2001 From: Tony Lindgren Date: Fri, 30 Sep 2011 11:05:55 -0700 Subject: [PATCH 35/75] ARM: OMAP: Fix i2c init for twl4030 Looks like 2600 kHz rate does not work reliably on 2430, so just use the 100 kHz rate. Otherwise the system often fails to boot properly with: omap_i2c omap_i2c.2: timeout waiting for bus ready omap_i2c omap_i2c.2: timeout waiting for bus ready twl: i2c_write failed to transfer all messages omap_i2c omap_i2c.2: timeout waiting for bus ready twl: i2c_write failed to transfer all messages omap_i2c omap_i2c.2: timeout waiting for bus ready twl: i2c_write failed to transfer all messages twl: clock init err [-110] omap_i2c omap_i2c.2: timeout waiting for bus ready twl: i2c_write failed to transfer all messages TWL4030 Unable to unlock IDCODE registers --110 Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/board-2430sdp.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/arm/mach-omap2/board-2430sdp.c b/arch/arm/mach-omap2/board-2430sdp.c index 2028464cf5b9..f79b7d2a8ed4 100644 --- a/arch/arm/mach-omap2/board-2430sdp.c +++ b/arch/arm/mach-omap2/board-2430sdp.c @@ -193,7 +193,8 @@ static int __init omap2430_i2c_init(void) { omap_register_i2c_bus(1, 100, sdp2430_i2c1_boardinfo, ARRAY_SIZE(sdp2430_i2c1_boardinfo)); - omap2_pmic_init("twl4030", &sdp2430_twldata); + omap_pmic_init(2, 100, "twl4030", INT_24XX_SYS_NIRQ, + &sdp2430_twldata); return 0; } From b8e111a74d281b70e88bd67fec48cfda4f9bdce2 Mon Sep 17 00:00:00 2001 From: Axel Lin Date: Fri, 30 Sep 2011 11:05:55 -0700 Subject: [PATCH 36/75] ARM: OMAP: musb: Remove a redundant omap4430_phy_init call in usb_musb_init Current code calls omap4430_phy_init() twice in usb_musb_init(). Calling omap4430_phy_init() once is enough. This patch removes the first omap4430_phy_init() call, which using an uninitialized pointer as parameter. This patch elimates below build warning: arch/arm/mach-omap2/usb-musb.c: In function 'usb_musb_init': arch/arm/mach-omap2/usb-musb.c:141: warning: 'dev' may be used uninitialized in this function Signed-off-by: Axel Lin Bjarne Steinsbo Acked-by: Felipe Balbi Signed-off-by: Tony Lindgren --- arch/arm/mach-omap2/usb-musb.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/arch/arm/mach-omap2/usb-musb.c b/arch/arm/mach-omap2/usb-musb.c index a65145b02a55..19e4dac62a8c 100644 --- a/arch/arm/mach-omap2/usb-musb.c +++ b/arch/arm/mach-omap2/usb-musb.c @@ -137,9 +137,6 @@ void __init usb_musb_init(struct omap_musb_board_data *musb_board_data) musb_plat.mode = board_data->mode; musb_plat.extvbus = board_data->extvbus; - if (cpu_is_omap44xx()) - omap4430_phy_init(dev); - if (cpu_is_omap3517() || cpu_is_omap3505()) { oh_name = "am35x_otg_hs"; name = "musb-am35x"; From 002ea9eefec98dada56fd5f8e432a4e8570c2a26 Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Thu, 29 Sep 2011 09:37:23 +0100 Subject: [PATCH 37/75] ARM: 7113/1: mm: Align bank start to MAX_ORDER_NR_PAGES The VM subsystem assumes that there are valid memmap entries from the bank start aligned to MAX_ORDER_NR_PAGES. On the Ux500 we have a lot of mem=N arguments on the commandline triggering this bug several times over and causing kernel oops messages. Cc: stable@kernel.org Cc: Michael Bohan Cc: Nicolas Pitre Signed-off-by: Johan Palsson Signed-off-by: Rabin Vincent Signed-off-by: Linus Walleij Signed-off-by: Russell King --- arch/arm/mm/init.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c index cc7e2d8be9aa..f8037ba338ac 100644 --- a/arch/arm/mm/init.c +++ b/arch/arm/mm/init.c @@ -496,6 +496,13 @@ static void __init free_unused_memmap(struct meminfo *mi) */ bank_start = min(bank_start, ALIGN(prev_bank_end, PAGES_PER_SECTION)); +#else + /* + * Align down here since the VM subsystem insists that the + * memmap entries are valid from the bank start aligned to + * MAX_ORDER_NR_PAGES. + */ + bank_start = round_down(bank_start, MAX_ORDER_NR_PAGES); #endif /* * If we had a previous bank, and there is a space From 8742cd23471635f8b069bf9a6806200a77397ddb Mon Sep 17 00:00:00 2001 From: Nathan Lynch Date: Fri, 30 Sep 2011 13:49:35 -0500 Subject: [PATCH 38/75] MIPS: Call oops_enter, oops_exit in die This allows pause_on_oops and mtdoops to work. Signed-off-by: Nathan Lynch To: linux-mips@linux-mips.org Patchwork: https://patchwork.linux-mips.org/patch/2810/ Signed-off-by: Ralf Baechle --- arch/mips/kernel/traps.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c index 01eff7e609e8..cbea618af0b4 100644 --- a/arch/mips/kernel/traps.c +++ b/arch/mips/kernel/traps.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include #include @@ -371,14 +372,19 @@ void __noreturn die(const char *str, struct pt_regs *regs) static int die_counter; int sig = SIGSEGV; #ifdef CONFIG_MIPS_MT_SMTC - unsigned long dvpret = dvpe(); + unsigned long dvpret; #endif /* CONFIG_MIPS_MT_SMTC */ + oops_enter(); + if (notify_die(DIE_OOPS, str, regs, 0, regs_to_trapnr(regs), SIGSEGV) == NOTIFY_STOP) sig = 0; console_verbose(); raw_spin_lock_irq(&die_lock); +#ifdef CONFIG_MIPS_MT_SMTC + dvpret = dvpe(); +#endif /* CONFIG_MIPS_MT_SMTC */ bust_spinlocks(1); #ifdef CONFIG_MIPS_MT_SMTC mips_mt_regdump(dvpret); @@ -389,6 +395,8 @@ void __noreturn die(const char *str, struct pt_regs *regs) add_taint(TAINT_DIE); raw_spin_unlock_irq(&die_lock); + oops_exit(); + if (in_interrupt()) panic("Fatal exception in interrupt"); From 9bfacd01dc9b7519e1e6da12b01963550b9d09a2 Mon Sep 17 00:00:00 2001 From: Roland Dreier Date: Thu, 22 Sep 2011 00:06:05 -0700 Subject: [PATCH 39/75] [SCSI] qla2xxx: Fix crash in qla2x00_abort_all_cmds() on unload I hit a crash in qla2x00_abort_all_cmds() if the qla2xxx module is unloaded right after it is loaded. I debugged this down to the abort handling improperly treating a command of type SRB_ADISC_CMD as if it had a bsg_job to complete when that command actually uses the iocb_cmd part of the union. (I guess to hit this one has to unload the module while the async FC initialization is still in progress) It seems we should only look for a bsg_job if type is SRB_ELS_CMD_RPT, SRB_ELS_CMD_HST or SRB_CT_CMD, so switch the test to make that explicit. Signed-off-by: Roland Dreier Acked-by: Chad Dupuis Cc: stable@kernel.org Signed-off-by: James Bottomley --- drivers/scsi/qla2xxx/qla_os.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index 4cace3f20c04..1e69527f1e4e 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -1328,10 +1328,9 @@ qla2x00_abort_all_cmds(scsi_qla_host_t *vha, int res) qla2x00_sp_compl(ha, sp); } else { ctx = sp->ctx; - if (ctx->type == SRB_LOGIN_CMD || - ctx->type == SRB_LOGOUT_CMD) { - ctx->u.iocb_cmd->free(sp); - } else { + if (ctx->type == SRB_ELS_CMD_RPT || + ctx->type == SRB_ELS_CMD_HST || + ctx->type == SRB_CT_CMD) { struct fc_bsg_job *bsg_job = ctx->u.bsg_job; if (bsg_job->request->msgcode @@ -1343,6 +1342,8 @@ qla2x00_abort_all_cmds(scsi_qla_host_t *vha, int res) kfree(sp->ctx); mempool_free(sp, ha->srb_mempool); + } else { + ctx->u.iocb_cmd->free(sp); } } } From a73914c35b05d80f8ce78288e10056c91090b666 Mon Sep 17 00:00:00 2001 From: Mark Salyzyn Date: Thu, 22 Sep 2011 08:32:23 -0700 Subject: [PATCH 40/75] [SCSI] libsas: fix panic when single phy is disabled on a wide port When a wide port is being utilized to a target, if one disables only one of the phys, we get an OS crash: BUG: unable to handle kernel NULL pointer dereference at 0000000000000238 IP: [] mutex_lock+0x21/0x50 PGD 4103f5067 PUD 41dba9067 PMD 0 Oops: 0002 [#1] SMP last sysfs file: /sys/bus/pci/slots/5/address CPU 0 Modules linked in: pm8001(U) ses enclosure fuse nfsd exportfs autofs4 ipmi_devintf ipmi_si ipmi_msghandler nfs lockd fscache nfs_acl auth_rpcgss 8021q fcoe libfcoe garp libfc scsi_transport_fc stp scsi_tgt llc sunrpc cpufreq_ondemand acpi_cpufreq freq_table ipv6 sr_mod cdrom dm_mirror dm_region_hash dm_log uinput sg i2c_i801 i2c_core iTCO_wdt iTCO_vendor_support e1000e mlx4_ib ib_mad ib_core mlx4_en mlx4_core ext3 jbd mbcache sd_mod crc_t10dif usb_storage ata_generic pata_acpi ata_piix libsas(U) scsi_transport_sas dm_mod [last unloaded: pm8001] Modules linked in: pm8001(U) ses enclosure fuse nfsd exportfs autofs4 ipmi_devintf ipmi_si ipmi_msghandler nfs lockd fscache nfs_acl auth_rpcgss 8021q fcoe libfcoe garp libfc scsi_transport_fc stp scsi_tgt llc sunrpc cpufreq_ondemand acpi_cpufreq freq_table ipv6 sr_mod cdrom dm_mirror dm_region_hash dm_log uinput sg i2c_i801 i2c_core iTCO_wdt iTCO_vendor_support e1000e mlx4_ib ib_mad ib_core mlx4_en mlx4_core ext3 jbd mbcache sd_mod crc_t10dif usb_storage ata_generic pata_acpi ata_piix libsas(U) scsi_transport_sas dm_mod [last unloaded: pm8001] Pid: 5146, comm: scsi_wq_5 Not tainted 2.6.32-71.29.1.el6.lustre.7.x86_64 #1 Storage Server RIP: 0010:[] [] mutex_lock+0x21/0x50 RSP: 0018:ffff8803e4e33d30 EFLAGS: 00010246 RAX: 0000000000000000 RBX: 0000000000000238 RCX: 0000000000000000 RDX: 0000000000000000 RSI: ffff8803e664c800 RDI: 0000000000000238 RBP: ffff8803e4e33d40 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000001 R12: 0000000000000000 R13: 0000000000000238 R14: ffff88041acb7200 R15: ffff88041c51ada0 FS: 0000000000000000(0000) GS:ffff880028200000(0000) knlGS:0000000000000000 CS: 0010 DS: 0018 ES: 0018 CR0: 000000008005003b CR2: 0000000000000238 CR3: 0000000410143000 CR4: 00000000000006f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process scsi_wq_5 (pid: 5146, threadinfo ffff8803e4e32000, task ffff8803e4e294a0) Stack: ffff8803e664c800 0000000000000000 ffff8803e4e33d70 ffffffffa001f06e <0> ffff8803e4e33d60 ffff88041c51ada0 ffff88041acb7200 ffff88041bc0aa00 <0> ffff8803e4e33d90 ffffffffa0032b6c 0000000000000014 ffff88041acb7200 Call Trace: [] sas_port_delete_phy+0x2e/0xa0 [scsi_transport_sas] [] sas_unregister_devs_sas_addr+0xac/0xe0 [libsas] [] sas_ex_revalidate_domain+0x204/0x330 [libsas] [] ? sas_revalidate_domain+0x0/0x90 [libsas] [] sas_revalidate_domain+0x65/0x90 [libsas] [] worker_thread+0x170/0x2a0 [] ? autoremove_wake_function+0x0/0x40 [] ? worker_thread+0x0/0x2a0 [] kthread+0x96/0xa0 [] child_rip+0xa/0x20 [] ? kthread+0x0/0xa0 [] ? child_rip+0x0/0x20 Code: ff ff 85 c0 75 ed eb d6 66 90 55 48 89 e5 48 83 ec 10 48 89 1c 24 4c 89 64 24 08 0f 1f 44 00 00 48 89 fb e8 92 f4 ff ff 48 89 df ff 0f 79 05 e8 25 00 00 00 65 48 8b 04 25 08 cc 00 00 48 2d RIP [] mutex_lock+0x21/0x50 RSP CR2: 0000000000000238 The following patch is admittedly a band-aid, and does not solve the root cause, but it still is a good candidate for hardening as a pointer check before reference. Signed-off-by: Mark Salyzyn Tested-by: Jack Wang Cc: stable@kernel.org Signed-off-by: James Bottomley --- drivers/scsi/libsas/sas_expander.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/drivers/scsi/libsas/sas_expander.c b/drivers/scsi/libsas/sas_expander.c index c9e3dc024bc3..16ad97df5ba6 100644 --- a/drivers/scsi/libsas/sas_expander.c +++ b/drivers/scsi/libsas/sas_expander.c @@ -1769,10 +1769,12 @@ static void sas_unregister_devs_sas_addr(struct domain_device *parent, sas_disable_routing(parent, phy->attached_sas_addr); } memset(phy->attached_sas_addr, 0, SAS_ADDR_SIZE); - sas_port_delete_phy(phy->port, phy->phy); - if (phy->port->num_phys == 0) - sas_port_delete(phy->port); - phy->port = NULL; + if (phy->port) { + sas_port_delete_phy(phy->port, phy->phy); + if (phy->port->num_phys == 0) + sas_port_delete(phy->port); + phy->port = NULL; + } } static int sas_discover_bfs_by_root_level(struct domain_device *root, From cb2d0f3e968bff7c6d262aca3e3ab8d4184e69b2 Mon Sep 17 00:00:00 2001 From: David Ward Date: Sun, 18 Sep 2011 12:53:20 +0000 Subject: [PATCH 41/75] macvlan/macvtap: Fix unicast between macvtap interfaces in bridge mode Packets should always be forwarded to the lowerdev using dev_forward_skb. vlan->forward is for packets being forwarded directly to another macvlan/ macvtap device (used for multicast in bridge mode). Reported-and-tested-by: Shlomo Pongratz Signed-off-by: David Ward Signed-off-by: David S. Miller --- drivers/net/macvlan.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c index 05172c39a0ce..376e3e94bae0 100644 --- a/drivers/net/macvlan.c +++ b/drivers/net/macvlan.c @@ -239,7 +239,7 @@ static int macvlan_queue_xmit(struct sk_buff *skb, struct net_device *dev) dest = macvlan_hash_lookup(port, eth->h_dest); if (dest && dest->mode == MACVLAN_MODE_BRIDGE) { /* send to lowerdev first for its network taps */ - vlan->forward(vlan->lowerdev, skb); + dev_forward_skb(vlan->lowerdev, skb); return NET_XMIT_SUCCESS; } From 260fcbeb1ae9e768a44c9925338fbacb0d7e5ba9 Mon Sep 17 00:00:00 2001 From: "Yan, Zheng" Date: Thu, 29 Sep 2011 17:10:10 +0000 Subject: [PATCH 42/75] tcp: properly handle md5sig_pool references tcp_v4_clear_md5_list() assumes that multiple tcp md5sig peers only hold one reference to md5sig_pool. but tcp_v4_md5_do_add() increases use count of md5sig_pool for each peer. This patch makes tcp_v4_md5_do_add() only increases use count for the first tcp md5sig peer. Signed-off-by: Zheng Yan Signed-off-by: David S. Miller --- net/ipv4/tcp_ipv4.c | 11 +++++++---- net/ipv6/tcp_ipv6.c | 8 +++++--- 2 files changed, 12 insertions(+), 7 deletions(-) diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index c34f01513945..7963e03f1068 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -927,18 +927,21 @@ int tcp_v4_md5_do_add(struct sock *sk, __be32 addr, } sk_nocaps_add(sk, NETIF_F_GSO_MASK); } - if (tcp_alloc_md5sig_pool(sk) == NULL) { + + md5sig = tp->md5sig_info; + if (md5sig->entries4 == 0 && + tcp_alloc_md5sig_pool(sk) == NULL) { kfree(newkey); return -ENOMEM; } - md5sig = tp->md5sig_info; if (md5sig->alloced4 == md5sig->entries4) { keys = kmalloc((sizeof(*keys) * (md5sig->entries4 + 1)), GFP_ATOMIC); if (!keys) { kfree(newkey); - tcp_free_md5sig_pool(); + if (md5sig->entries4 == 0) + tcp_free_md5sig_pool(); return -ENOMEM; } @@ -982,6 +985,7 @@ int tcp_v4_md5_do_del(struct sock *sk, __be32 addr) kfree(tp->md5sig_info->keys4); tp->md5sig_info->keys4 = NULL; tp->md5sig_info->alloced4 = 0; + tcp_free_md5sig_pool(); } else if (tp->md5sig_info->entries4 != i) { /* Need to do some manipulation */ memmove(&tp->md5sig_info->keys4[i], @@ -989,7 +993,6 @@ int tcp_v4_md5_do_del(struct sock *sk, __be32 addr) (tp->md5sig_info->entries4 - i) * sizeof(struct tcp4_md5sig_key)); } - tcp_free_md5sig_pool(); return 0; } } diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 79cc6469508d..7b8fc5794352 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -591,7 +591,8 @@ static int tcp_v6_md5_do_add(struct sock *sk, const struct in6_addr *peer, } sk_nocaps_add(sk, NETIF_F_GSO_MASK); } - if (tcp_alloc_md5sig_pool(sk) == NULL) { + if (tp->md5sig_info->entries6 == 0 && + tcp_alloc_md5sig_pool(sk) == NULL) { kfree(newkey); return -ENOMEM; } @@ -600,8 +601,9 @@ static int tcp_v6_md5_do_add(struct sock *sk, const struct in6_addr *peer, (tp->md5sig_info->entries6 + 1)), GFP_ATOMIC); if (!keys) { - tcp_free_md5sig_pool(); kfree(newkey); + if (tp->md5sig_info->entries6 == 0) + tcp_free_md5sig_pool(); return -ENOMEM; } @@ -647,6 +649,7 @@ static int tcp_v6_md5_do_del(struct sock *sk, const struct in6_addr *peer) kfree(tp->md5sig_info->keys6); tp->md5sig_info->keys6 = NULL; tp->md5sig_info->alloced6 = 0; + tcp_free_md5sig_pool(); } else { /* shrink the database */ if (tp->md5sig_info->entries6 != i) @@ -655,7 +658,6 @@ static int tcp_v6_md5_do_del(struct sock *sk, const struct in6_addr *peer) (tp->md5sig_info->entries6 - i) * sizeof (tp->md5sig_info->keys6[0])); } - tcp_free_md5sig_pool(); return 0; } } From 1e5289e121372a3494402b1b131b41bfe1cf9b7f Mon Sep 17 00:00:00 2001 From: "Yan, Zheng" Date: Sun, 2 Oct 2011 04:21:50 +0000 Subject: [PATCH 43/75] tcp: properly update lost_cnt_hint during shifting lost_skb_hint is used by tcp_mark_head_lost() to mark the first unhandled skb. lost_cnt_hint is the number of packets or sacked packets before the lost_skb_hint; When shifting a skb that is before the lost_skb_hint, if tcp_is_fack() is ture, the skb has already been counted in the lost_cnt_hint; if tcp_is_fack() is false, tcp_sacktag_one() will increase the lost_cnt_hint. So tcp_shifted_skb() does not need to adjust the lost_cnt_hint by itself. When shifting a skb that is equal to lost_skb_hint, the shifted packets will not be counted by tcp_mark_head_lost(). So tcp_shifted_skb() should adjust the lost_cnt_hint even tcp_is_fack(tp) is true. Signed-off-by: Zheng Yan Signed-off-by: David S. Miller --- net/ipv4/tcp_input.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 21fab3edb92c..d73aab3fbfc0 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -1389,9 +1389,7 @@ static int tcp_shifted_skb(struct sock *sk, struct sk_buff *skb, BUG_ON(!pcount); - /* Tweak before seqno plays */ - if (!tcp_is_fack(tp) && tcp_is_sack(tp) && tp->lost_skb_hint && - !before(TCP_SKB_CB(tp->lost_skb_hint)->seq, TCP_SKB_CB(skb)->seq)) + if (skb == tp->lost_skb_hint) tp->lost_cnt_hint += pcount; TCP_SKB_CB(prev)->end_seq += shifted; From d6069dae4700cf5290a75002c46b9be8ea7eef3c Mon Sep 17 00:00:00 2001 From: Jason Gerecke Date: Tue, 4 Oct 2011 22:50:45 -0700 Subject: [PATCH 44/75] Input: wacom - revert "Cintiq 21UX2 does not have menu strips" This reverts commit 71c86ce59791bcd67af937bbea719a508079d7c2. The 21UX2 does have touchstrips, but they are in a somewhat- hidden location. Signed-off-by: Jason Gerecke Acked-by: Ping Cheng Signed-off-by: Dmitry Torokhov --- drivers/input/tablet/wacom_wac.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/drivers/input/tablet/wacom_wac.c b/drivers/input/tablet/wacom_wac.c index 0dc97ec15c28..9dea71849f40 100644 --- a/drivers/input/tablet/wacom_wac.c +++ b/drivers/input/tablet/wacom_wac.c @@ -1124,11 +1124,8 @@ void wacom_setup_input_capabilities(struct input_dev *input_dev, for (i = 0; i < 8; i++) __set_bit(BTN_0 + i, input_dev->keybit); - if (wacom_wac->features.type != WACOM_21UX2) { - input_set_abs_params(input_dev, ABS_RX, 0, 4096, 0, 0); - input_set_abs_params(input_dev, ABS_RY, 0, 4096, 0, 0); - } - + input_set_abs_params(input_dev, ABS_RX, 0, 4096, 0, 0); + input_set_abs_params(input_dev, ABS_RY, 0, 4096, 0, 0); input_set_abs_params(input_dev, ABS_Z, -900, 899, 0, 0); __set_bit(INPUT_PROP_DIRECT, input_dev->propbit); From 4f332844cc87c5f99c5300f788abbe8a8c731390 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 4 Oct 2011 17:23:15 -0400 Subject: [PATCH 45/75] drm/radeon/kms: retry aux transactions if there are status flags MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If there are error flags in the aux status, retry the transaction. This makes aux much more reliable, especially on llano systems. Signed-off-by: Alex Deucher Reviewed-by: Michel Dänzer Cc: stable@kernel.org Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/atombios_dp.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c index 4da23889fea6..79e8ebc05307 100644 --- a/drivers/gpu/drm/radeon/atombios_dp.c +++ b/drivers/gpu/drm/radeon/atombios_dp.c @@ -129,7 +129,9 @@ static int radeon_dp_aux_native_write(struct radeon_connector *radeon_connector, for (retry = 0; retry < 4; retry++) { ret = radeon_process_aux_ch(dig_connector->dp_i2c_bus, msg, msg_bytes, NULL, 0, delay, &ack); - if (ret < 0) + if (ret == -EBUSY) + continue; + else if (ret < 0) return ret; if ((ack & AUX_NATIVE_REPLY_MASK) == AUX_NATIVE_REPLY_ACK) return send_bytes; @@ -160,7 +162,9 @@ static int radeon_dp_aux_native_read(struct radeon_connector *radeon_connector, for (retry = 0; retry < 4; retry++) { ret = radeon_process_aux_ch(dig_connector->dp_i2c_bus, msg, msg_bytes, recv, recv_bytes, delay, &ack); - if (ret < 0) + if (ret == -EBUSY) + continue; + else if (ret < 0) return ret; if ((ack & AUX_NATIVE_REPLY_MASK) == AUX_NATIVE_REPLY_ACK) return ret; @@ -236,7 +240,9 @@ int radeon_dp_i2c_aux_ch(struct i2c_adapter *adapter, int mode, for (retry = 0; retry < 4; retry++) { ret = radeon_process_aux_ch(auxch, msg, msg_bytes, reply, reply_bytes, 0, &ack); - if (ret < 0) { + if (ret == -EBUSY) + continue; + else if (ret < 0) { DRM_DEBUG_KMS("aux_ch failed %d\n", ret); return ret; } From 6777a4f6898a53974ef7fe7ce09ec41fae0f32db Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Tue, 4 Oct 2011 12:23:24 -0400 Subject: [PATCH 46/75] drm/radeon/kms: fix dp_detect handling for DP bridge chips MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The HPD pin is not reliable for detecting whether a monitor is connected or not. Skip HPD and just use DDC or load detection. Fixes phantom VGA connected bugs. [Michel: fixes phantom VGA bugs on his llano system.] Signed-off-by: Alex Deucher Reviewed-by: Michel Dänzer Tested-by: Michel Dänzer Cc: stable@kernel.org Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon_connectors.c | 21 ++++++--------------- 1 file changed, 6 insertions(+), 15 deletions(-) diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index bce63fd329d4..449c3d8c6836 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -1303,23 +1303,14 @@ radeon_dp_detect(struct drm_connector *connector, bool force) /* get the DPCD from the bridge */ radeon_dp_getdpcd(radeon_connector); - if (radeon_hpd_sense(rdev, radeon_connector->hpd.hpd)) - ret = connector_status_connected; - else { - /* need to setup ddc on the bridge */ - if (encoder) - radeon_atom_ext_encoder_setup_ddc(encoder); + if (encoder) { + /* setup ddc on the bridge */ + radeon_atom_ext_encoder_setup_ddc(encoder); if (radeon_ddc_probe(radeon_connector, - radeon_connector->requires_extended_probe)) + radeon_connector->requires_extended_probe)) /* try DDC */ ret = connector_status_connected; - } - - if ((ret == connector_status_disconnected) && - radeon_connector->dac_load_detect) { - struct drm_encoder *encoder = radeon_best_single_encoder(connector); - struct drm_encoder_helper_funcs *encoder_funcs; - if (encoder) { - encoder_funcs = encoder->helper_private; + else if (radeon_connector->dac_load_detect) { /* try load detection */ + struct drm_encoder_helper_funcs *encoder_funcs = encoder->helper_private; ret = encoder_funcs->detect(encoder, connector); } } From 3458e21c0d384ca04b27a2ea24d9314c1b57530f Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 5 Oct 2011 03:24:43 +0000 Subject: [PATCH 47/75] netfilter: Use proper rwlock init function Replace the open coded initialization with the init function. Signed-off-by: Thomas Gleixner Acked-by: Hans Schillstrom Signed-off-by: David S. Miller --- net/netfilter/ipvs/ip_vs_ctl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index 2b771dc708a3..5290ac353a5e 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -3679,7 +3679,7 @@ int __net_init ip_vs_control_net_init(struct net *net) int idx; struct netns_ipvs *ipvs = net_ipvs(net); - ipvs->rs_lock = __RW_LOCK_UNLOCKED(ipvs->rs_lock); + rwlock_init(&ipvs->rs_lock); /* Initialize rs_table */ for (idx = 0; idx < IP_VS_RTAB_SIZE; idx++) From cb7cf41961fe10773c491c75ae73539ad4bbed66 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 5 Oct 2011 18:36:50 -0400 Subject: [PATCH 48/75] drm/radeon/kms: use hardcoded dig encoder to transmitter mapping for DCE4.1 The encoders are supposedly fully routeable, but changing the mapping doesn't always seem to take. Using a hardcoded mapping is much more reliable. Fixes: https://bugs.freedesktop.org/show_bug.cgi?id=41366 Signed-off-by: Alex Deucher Tested-by: Simon Farnsworth Cc: stable@kernel.org Signed-off-by: Dave Airlie --- drivers/gpu/drm/radeon/radeon_encoders.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/radeon/radeon_encoders.c b/drivers/gpu/drm/radeon/radeon_encoders.c index 13690f3eb4a4..8a171b21b453 100644 --- a/drivers/gpu/drm/radeon/radeon_encoders.c +++ b/drivers/gpu/drm/radeon/radeon_encoders.c @@ -1755,9 +1755,12 @@ static int radeon_atom_pick_dig_encoder(struct drm_encoder *encoder) /* DCE4/5 */ if (ASIC_IS_DCE4(rdev)) { dig = radeon_encoder->enc_priv; - if (ASIC_IS_DCE41(rdev)) - return radeon_crtc->crtc_id; - else { + if (ASIC_IS_DCE41(rdev)) { + if (dig->linkb) + return 1; + else + return 0; + } else { switch (radeon_encoder->encoder_id) { case ENCODER_OBJECT_ID_INTERNAL_UNIPHY: if (dig->linkb) From b64b73d7d0c480f75684519c6134e79d50c1b341 Mon Sep 17 00:00:00 2001 From: stephen hemminger Date: Mon, 3 Oct 2011 18:14:45 +0000 Subject: [PATCH 49/75] bridge: leave carrier on for empty bridge This resolves a regression seen by some users of bridging. Some users use the bridge like a dummy device. They expect to be able to put an IPv6 address on the device with no ports attached. Although there are better ways of doing this, there is no reason to not allow it. Note: the bridge still will reflect the state of ports in the bridge if there are any added. Signed-off-by: Stephen Hemminger Signed-off-by: David S. Miller --- net/bridge/br_device.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index 32b8f9f7f79e..ff3ed6086ce1 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -91,7 +91,6 @@ static int br_dev_open(struct net_device *dev) { struct net_bridge *br = netdev_priv(dev); - netif_carrier_off(dev); netdev_update_features(dev); netif_start_queue(dev); br_stp_enable_bridge(br); @@ -108,8 +107,6 @@ static int br_dev_stop(struct net_device *dev) { struct net_bridge *br = netdev_priv(dev); - netif_carrier_off(dev); - br_stp_disable_bridge(br); br_multicast_stop(br); From 186c6bbced722cfeff041d2a1264c95f5d042050 Mon Sep 17 00:00:00 2001 From: Benjamin Poirier Date: Tue, 4 Oct 2011 04:00:30 +0000 Subject: [PATCH 50/75] net: fix typos in Documentation/networking/scaling.txt The second hunk fixes rps_sock_flow_table but has to re-wrap the paragraph. Signed-off-by: Benjamin Poirier Signed-off-by: David S. Miller --- Documentation/networking/scaling.txt | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/Documentation/networking/scaling.txt b/Documentation/networking/scaling.txt index 8ce7c30e7230..fe67b5c79f0f 100644 --- a/Documentation/networking/scaling.txt +++ b/Documentation/networking/scaling.txt @@ -27,7 +27,7 @@ applying a filter to each packet that assigns it to one of a small number of logical flows. Packets for each flow are steered to a separate receive queue, which in turn can be processed by separate CPUs. This mechanism is generally known as “Receive-side Scaling” (RSS). The goal of RSS and -the other scaling techniques to increase performance uniformly. +the other scaling techniques is to increase performance uniformly. Multi-queue distribution can also be used for traffic prioritization, but that is not the focus of these techniques. @@ -186,10 +186,10 @@ are steered using plain RPS. Multiple table entries may point to the same CPU. Indeed, with many flows and few CPUs, it is very likely that a single application thread handles flows with many different flow hashes. -rps_sock_table is a global flow table that contains the *desired* CPU for -flows: the CPU that is currently processing the flow in userspace. Each -table value is a CPU index that is updated during calls to recvmsg and -sendmsg (specifically, inet_recvmsg(), inet_sendmsg(), inet_sendpage() +rps_sock_flow_table is a global flow table that contains the *desired* CPU +for flows: the CPU that is currently processing the flow in userspace. +Each table value is a CPU index that is updated during calls to recvmsg +and sendmsg (specifically, inet_recvmsg(), inet_sendmsg(), inet_sendpage() and tcp_splice_read()). When the scheduler moves a thread to a new CPU while it has outstanding From 29cf7a30f8a0ce4af2406d93d5a332099be26923 Mon Sep 17 00:00:00 2001 From: Paul Menzel Date: Wed, 31 Aug 2011 17:07:10 +0200 Subject: [PATCH 51/75] x86/PCI: use host bridge _CRS info on ASUS M2V-MX SE In summary, this DMI quirk uses the _CRS info by default for the ASUS M2V-MX SE by turning on `pci=use_crs` and is similar to the quirk added by commit 2491762cfb47 ("x86/PCI: use host bridge _CRS info on ASRock ALiveSATA2-GLAN") whose commit message should be read for further information. Since commit 3e3da00c01d0 ("x86/pci: AMD one chain system to use pci read out res") Linux gives the following oops: parport0: PC-style at 0x378, irq 7 [PCSPP,TRISTATE] HDA Intel 0000:20:01.0: PCI INT A -> GSI 17 (level, low) -> IRQ 17 HDA Intel 0000:20:01.0: setting latency timer to 64 BUG: unable to handle kernel paging request at ffffc90011c08000 IP: [] azx_probe+0x3ad/0x86b [snd_hda_intel] PGD 13781a067 PUD 13781b067 PMD 1300ba067 PTE 800000fd00000173 Oops: 0009 [#1] SMP last sysfs file: /sys/module/snd_pcm/initstate CPU 0 Modules linked in: snd_hda_intel(+) snd_hda_codec snd_hwdep snd_pcm_oss snd_mixer_oss snd_pcm snd_seq_midi snd_rawmidi snd_seq_midi_event tpm_tis tpm snd_seq tpm_bios psmouse parport_pc snd_timer snd_seq_device parport processor evdev snd i2c_viapro thermal_sys amd64_edac_mod k8temp i2c_core soundcore shpchp pcspkr serio_raw asus_atk0110 pci_hotplug edac_core button snd_page_alloc edac_mce_amd ext3 jbd mbcache sha256_generic cryptd aes_x86_64 aes_generic cbc dm_crypt dm_mod raid1 md_mod usbhid hid sg sd_mod crc_t10dif sr_mod cdrom ata_generic uhci_hcd sata_via pata_via libata ehci_hcd usbcore scsi_mod via_rhine mii nls_base [last unloaded: scsi_wait_scan] Pid: 1153, comm: work_for_cpu Not tainted 2.6.37-1-amd64 #1 M2V-MX SE/System Product Name RIP: 0010:[] [] azx_probe+0x3ad/0x86b [snd_hda_intel] RSP: 0018:ffff88013153fe50 EFLAGS: 00010286 RAX: ffffc90011c08000 RBX: ffff88013029ec00 RCX: 0000000000000006 RDX: 0000000000000000 RSI: 0000000000000246 RDI: 0000000000000246 RBP: ffff88013341d000 R08: 0000000000000000 R09: 0000000000000040 R10: 0000000000000286 R11: 0000000000003731 R12: ffff88013029c400 R13: 0000000000000000 R14: 0000000000000000 R15: ffff88013341d090 FS: 0000000000000000(0000) GS:ffff8800bfc00000(0000) knlGS:00000000f7610ab0 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: ffffc90011c08000 CR3: 0000000132f57000 CR4: 00000000000006f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process work_for_cpu (pid: 1153, threadinfo ffff88013153e000, task ffff8801303c86c0) Stack: 0000000000000005 ffffffff8123ad65 00000000000136c0 ffff88013029c400 ffff8801303c8998 ffff88013341d000 ffff88013341d090 ffff8801322d9dc8 ffff88013341d208 0000000000000000 0000000000000000 ffffffff811ad232 Call Trace: [] ? __pm_runtime_set_status+0x162/0x186 [] ? local_pci_probe+0x49/0x92 [] ? do_work_for_cpu+0x0/0x1b [] ? do_work_for_cpu+0x0/0x1b [] ? do_work_for_cpu+0xb/0x1b [] ? kthread+0x7a/0x82 [] ? kernel_thread_helper+0x4/0x10 [] ? kthread+0x0/0x82 [] ? kernel_thread_helper+0x0/0x10 Code: f4 01 00 00 ef 31 f6 48 89 df e8 29 dd ff ff 85 c0 0f 88 2b 03 00 00 48 89 ef e8 b4 39 c3 e0 8b 7b 40 e8 fc 9d b1 e0 48 8b 43 38 <66> 8b 10 66 89 14 24 8b 43 14 83 e8 03 83 f8 01 77 32 31 d2 be RIP [] azx_probe+0x3ad/0x86b [snd_hda_intel] RSP CR2: ffffc90011c08000 ---[ end trace 8d1f3ebc136437fd ]--- Trusting the ACPI _CRS information (`pci=use_crs`) fixes this problem. $ dmesg | grep -i crs # with the quirk PCI: Using host bridge windows from ACPI; if necessary, use "pci=nocrs" and report a bug The match has to be against the DMI board entries though since the vendor entries are not populated. DMI: System manufacturer System Product Name/M2V-MX SE, BIOS 0304 10/30/2007 This quirk should be removed when `pci=use_crs` is enabled for machines from 2006 or earlier or some other solution is implemented. Using coreboot [1] with this board the problem does not exist but this quirk also does not affect it either. To be safe though the check is tightened to only take effect when the BIOS from American Megatrends is used. 15:13 < ruik> but coreboot does not need that 15:13 < ruik> because i have there only one root bus 15:13 < ruik> the audio is behind a bridge $ sudo dmidecode BIOS Information Vendor: American Megatrends Inc. Version: 0304 Release Date: 10/30/2007 [1] http://www.coreboot.org/ Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=30552 Cc: stable@kernel.org (2.6.34) Cc: Thomas Gleixner Cc: Ingo Molnar Cc: H. Peter Anvin Cc: x86@kernel.org Signed-off-by: Paul Menzel Signed-off-by: Bjorn Helgaas Acked-by: Jesse Barnes Signed-off-by: Linus Torvalds --- arch/x86/pci/acpi.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c index 039d91315bc5..404f21a3ff9e 100644 --- a/arch/x86/pci/acpi.c +++ b/arch/x86/pci/acpi.c @@ -43,6 +43,17 @@ static const struct dmi_system_id pci_use_crs_table[] __initconst = { DMI_MATCH(DMI_PRODUCT_NAME, "ALiveSATA2-GLAN"), }, }, + /* https://bugzilla.kernel.org/show_bug.cgi?id=30552 */ + /* 2006 AMD HT/VIA system with two host bridges */ + { + .callback = set_use_crs, + .ident = "ASUS M2V-MX SE", + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."), + DMI_MATCH(DMI_BOARD_NAME, "M2V-MX SE"), + DMI_MATCH(DMI_BIOS_VENDOR, "American Megatrends Inc."), + }, + }, {} }; From 98e87d57aab9b1594f9cc53a386fcb6f2f2ba6e2 Mon Sep 17 00:00:00 2001 From: srinidhi kasagar Date: Tue, 20 Sep 2011 11:15:46 +0530 Subject: [PATCH 52/75] ARM: mach-ux500: enable fix for ARM errata 754322 This applies ARM errata fix 754322 for all ux500 platforms. Cc: stable@kernel.org Signed-off-by: srinidhi kasagar Signed-off-by: Linus Walleij --- arch/arm/mach-ux500/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/mach-ux500/Kconfig b/arch/arm/mach-ux500/Kconfig index 4210cb434dbc..a3e0c8692f0d 100644 --- a/arch/arm/mach-ux500/Kconfig +++ b/arch/arm/mach-ux500/Kconfig @@ -6,6 +6,7 @@ config UX500_SOC_COMMON select ARM_GIC select HAS_MTU select ARM_ERRATA_753970 + select ARM_ERRATA_754322 menu "Ux500 SoC" From 9d1e397b7bafe843564eab26ed3f682637159c6c Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 6 Oct 2011 23:14:07 -0500 Subject: [PATCH 53/75] [CIFS] Fix first time message on mount, ntlmv2 upgrade delayed to 3.2 Microsoft has a bug with ntlmv2 that requires use of ntlmssp, but we didn't get the required information on when/how to use ntlmssp to old (but once very popular) legacy servers (various NT4 fixpacks for example) until too late to merge for 3.1. Will upgrade to NTLMv2 in NTLMSSP in 3.2 Signed-off-by: Steve French Reviewed-by: Jeff Layton --- fs/cifs/connect.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index f4af4cc37500..71beb0201970 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -2018,7 +2018,7 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb_vol *volume_info) warned_on_ntlm = true; cERROR(1, "default security mechanism requested. The default " "security mechanism will be upgraded from ntlm to " - "ntlmv2 in kernel release 3.1"); + "ntlmv2 in kernel release 3.2"); } ses->overrideSecFlg = volume_info->secFlg; From 746fb63c54d2659e1da3c5e90a77ccb23a309cc5 Mon Sep 17 00:00:00 2001 From: Olof Johansson Date: Fri, 7 Oct 2011 13:27:48 -0700 Subject: [PATCH 54/75] MAINTAINERS: Update tegra maintainer information A couple of changes to the Tegra maintainership setup: I'm very glad to bring on Stephen Warren on board as a maintainer. The work he has done so far is excellent, and the fact that he works for Nvidia means he has long-term interest in the platform. Erik Gilling did an astounding amount of work on getting things up and running but has been a silent partner on the maintainership side for a while, and is stepping down. Thanks for your contributions so far, Erik. Finally, update the git URL since I'll take over running the main repo for a while. Overall maintainership model isn't changing much at this time: We'll all three review patches as appropriate, and one of us will collect the main repo (me at this time). Signed-off-by: Olof Johansson Cc: Erik Gilling Acked-by: Colin Cross Acked-by: Stephen Warren Signed-off-by: Linus Torvalds --- MAINTAINERS | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index ace8f9c81b96..5483e0c93b4b 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -6366,10 +6366,10 @@ F: net/ipv4/tcp_lp.c TEGRA SUPPORT M: Colin Cross -M: Erik Gilling M: Olof Johansson +M: Stephen Warren L: linux-tegra@vger.kernel.org -T: git git://android.git.kernel.org/kernel/tegra.git +T: git git://git.kernel.org/pub/scm/linux/kernel/git/olof/tegra.git S: Supported F: arch/arm/mach-tegra From 3faf12cd8603e3efbcc7e570e07dfd1ea2a17095 Mon Sep 17 00:00:00 2001 From: Marc Dietrich Date: Fri, 7 Oct 2011 08:31:41 -0700 Subject: [PATCH 55/75] ARM: tegra: fix compilation error due to mach/hardware.h removal This fixes a compilation error in cpu-tegra.c which was introduced in dc8d966bccde ("ARM: convert PCI defines to variables") which removed the now obsolete mach/hardware.h from the mach-tegra subtree. Signed-off-by: Marc Dietrich Signed-off-by: Olof Johansson Cc: Sergei Shtylyov Signed-off-by: Linus Torvalds --- arch/arm/mach-tegra/cpu-tegra.c | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/arm/mach-tegra/cpu-tegra.c b/arch/arm/mach-tegra/cpu-tegra.c index 0e1016a827ac..0e0fd4d889bd 100644 --- a/arch/arm/mach-tegra/cpu-tegra.c +++ b/arch/arm/mach-tegra/cpu-tegra.c @@ -32,7 +32,6 @@ #include -#include #include /* Frequency table index must be sequential starting at 0 */ From c5d6136e10d667965e46f998c01863802f7b89c8 Mon Sep 17 00:00:00 2001 From: Thadeu Lima de Souza Cascardo Date: Mon, 10 Oct 2011 06:42:23 +0000 Subject: [PATCH 56/75] mlx4_en: fix endianness with blue frame support The doorbell register was being unconditionally swapped. In x86, that meant it was being swapped to BE and written to the descriptor and to memory, depending on the case of blue frame support or writing to doorbell register. On PPC, this meant it was being swapped to LE and then swapped back to BE while writing to the register. But in the blue frame case, it was being written as LE to the descriptor. The fix is not to swap doorbell unconditionally, write it to the register as BE and convert it to BE when writing it to the descriptor. Signed-off-by: Thadeu Lima de Souza Cascardo Reported-by: Richard Hendrickson Cc: Eli Cohen Cc: Yevgeny Petrilin Cc: Benjamin Herrenschmidt Signed-off-by: David S. Miller --- drivers/net/mlx4/en_tx.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/mlx4/en_tx.c b/drivers/net/mlx4/en_tx.c index 6e03de034ac7..f76ab6bf3096 100644 --- a/drivers/net/mlx4/en_tx.c +++ b/drivers/net/mlx4/en_tx.c @@ -172,7 +172,7 @@ int mlx4_en_activate_tx_ring(struct mlx4_en_priv *priv, memset(ring->buf, 0, ring->buf_size); ring->qp_state = MLX4_QP_STATE_RST; - ring->doorbell_qpn = swab32(ring->qp.qpn << 8); + ring->doorbell_qpn = ring->qp.qpn << 8; mlx4_en_fill_qp_context(priv, ring->size, ring->stride, 1, 0, ring->qpn, ring->cqn, &ring->context); @@ -791,7 +791,7 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) skb_orphan(skb); if (ring->bf_enabled && desc_size <= MAX_BF && !bounce && !vlan_tag) { - *(u32 *) (&tx_desc->ctrl.vlan_tag) |= ring->doorbell_qpn; + *(__be32 *) (&tx_desc->ctrl.vlan_tag) |= cpu_to_be32(ring->doorbell_qpn); op_own |= htonl((bf_index & 0xffff) << 8); /* Ensure new descirptor hits memory * before setting ownership of this descriptor to HW */ @@ -812,7 +812,7 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev) wmb(); tx_desc->ctrl.owner_opcode = op_own; wmb(); - writel(ring->doorbell_qpn, ring->bf.uar->map + MLX4_SEND_DOORBELL); + iowrite32be(ring->doorbell_qpn, ring->bf.uar->map + MLX4_SEND_DOORBELL); } /* Poll CQ here */ From 134d0f974c55c942541f1663d51fa846257eff2a Mon Sep 17 00:00:00 2001 From: Dmitry Kravkov Date: Sun, 9 Oct 2011 23:57:36 +0000 Subject: [PATCH 57/75] bnx2x: fix cl_id allocation for non-eth clients for NPAR mode There are some consolidations of NPAR configuration when FCoE and iSCSI L2 clients will get the same id, in this case FCoE ring will be non-functional. Signed-off-by: Dmitry Kravkov Signed-off-by: Eilon Greenstein Signed-off-by: David S. Miller --- drivers/net/bnx2x/bnx2x.h | 18 ++++++++++++------ drivers/net/bnx2x/bnx2x_cmn.h | 2 +- 2 files changed, 13 insertions(+), 7 deletions(-) diff --git a/drivers/net/bnx2x/bnx2x.h b/drivers/net/bnx2x/bnx2x.h index e46df5331c55..9a7eb3b36cf3 100644 --- a/drivers/net/bnx2x/bnx2x.h +++ b/drivers/net/bnx2x/bnx2x.h @@ -239,13 +239,19 @@ void bnx2x_int_disable(struct bnx2x *bp); * FUNC_N_CLID_X = N * NUM_SPECIAL_CLIENTS + FUNC_0_CLID_X * */ -/* iSCSI L2 */ -#define BNX2X_ISCSI_ETH_CL_ID_IDX 1 -#define BNX2X_ISCSI_ETH_CID 49 +enum { + BNX2X_ISCSI_ETH_CL_ID_IDX, + BNX2X_FCOE_ETH_CL_ID_IDX, + BNX2X_MAX_CNIC_ETH_CL_ID_IDX, +}; -/* FCoE L2 */ -#define BNX2X_FCOE_ETH_CL_ID_IDX 2 -#define BNX2X_FCOE_ETH_CID 50 +#define BNX2X_CNIC_START_ETH_CID 48 +enum { + /* iSCSI L2 */ + BNX2X_ISCSI_ETH_CID = BNX2X_CNIC_START_ETH_CID, + /* FCoE L2 */ + BNX2X_FCOE_ETH_CID, +}; /** Additional rings budgeting */ #ifdef BCM_CNIC diff --git a/drivers/net/bnx2x/bnx2x_cmn.h b/drivers/net/bnx2x/bnx2x_cmn.h index 223bfeebc597..2dc1199239d0 100644 --- a/drivers/net/bnx2x/bnx2x_cmn.h +++ b/drivers/net/bnx2x/bnx2x_cmn.h @@ -1297,7 +1297,7 @@ static inline void bnx2x_init_txdata(struct bnx2x *bp, static inline u8 bnx2x_cnic_eth_cl_id(struct bnx2x *bp, u8 cl_idx) { return bp->cnic_base_cl_id + cl_idx + - (bp->pf_num >> 1) * NON_ETH_CONTEXT_USE; + (bp->pf_num >> 1) * BNX2X_MAX_CNIC_ETH_CL_ID_IDX; } static inline u8 bnx2x_cnic_fw_sb_id(struct bnx2x *bp) From cdaf557034bb4397b2c86bb424de1788b134f610 Mon Sep 17 00:00:00 2001 From: "Yan, Zheng" Date: Sat, 8 Oct 2011 22:34:35 +0000 Subject: [PATCH 58/75] gro: refetch inet6_protos[] after pulling ext headers ipv6_gro_receive() doesn't update the protocol ops after pulling the ext headers. It looks like a typo. Signed-off-by: Zheng Yan Acked-by: Eric Dumazet Signed-off-by: David S. Miller --- net/ipv6/af_inet6.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index 3b5669a2582d..d27c797f9f05 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -875,6 +875,7 @@ static struct sk_buff **ipv6_gro_receive(struct sk_buff **head, skb_reset_transport_header(skb); __skb_push(skb, skb_gro_offset(skb)); + ops = rcu_dereference(inet6_protos[proto]); if (!ops || !ops->gro_receive) goto out_unlock; From a3a4bfde8a271df7ee56fcf31b8d907d50e7f216 Mon Sep 17 00:00:00 2001 From: Wolfgang Grandegger Date: Fri, 7 Oct 2011 09:28:14 +0000 Subject: [PATCH 59/75] mscan: too much data copied to CAN frame due to 16 bit accesses Due to the 16 bit access to mscan registers there's too much data copied to the zero initialized CAN frame when having an odd number of bytes to copy. This patch ensures that only the requested bytes are copied by using an 8 bit access for the remaining byte. Reported-by: Andre Naujoks Signed-off-by: Oliver Hartkopp Signed-off-by: Wolfgang Grandegger Signed-off-by: David S. Miller --- drivers/net/can/mscan/mscan.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/net/can/mscan/mscan.c b/drivers/net/can/mscan/mscan.c index 92feac68b66e..4cc6f44c2ba2 100644 --- a/drivers/net/can/mscan/mscan.c +++ b/drivers/net/can/mscan/mscan.c @@ -261,11 +261,13 @@ static netdev_tx_t mscan_start_xmit(struct sk_buff *skb, struct net_device *dev) void __iomem *data = ®s->tx.dsr1_0; u16 *payload = (u16 *)frame->data; - /* It is safe to write into dsr[dlc+1] */ - for (i = 0; i < (frame->can_dlc + 1) / 2; i++) { + for (i = 0; i < frame->can_dlc / 2; i++) { out_be16(data, *payload++); data += 2 + _MSCAN_RESERVED_DSR_SIZE; } + /* write remaining byte if necessary */ + if (frame->can_dlc & 1) + out_8(data, frame->data[frame->can_dlc - 1]); } out_8(®s->tx.dlr, frame->can_dlc); @@ -330,10 +332,13 @@ static void mscan_get_rx_frame(struct net_device *dev, struct can_frame *frame) void __iomem *data = ®s->rx.dsr1_0; u16 *payload = (u16 *)frame->data; - for (i = 0; i < (frame->can_dlc + 1) / 2; i++) { + for (i = 0; i < frame->can_dlc / 2; i++) { *payload++ = in_be16(data); data += 2 + _MSCAN_RESERVED_DSR_SIZE; } + /* read remaining byte if necessary */ + if (frame->can_dlc & 1) + frame->data[frame->can_dlc - 1] = in_8(data); } out_8(®s->canrflg, MSCAN_RXF); From 2a0f7f5769992bae5b3f97157fd80b2b943be485 Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Mon, 10 Oct 2011 15:43:34 -0400 Subject: [PATCH 60/75] Btrfs: fix recursive auto-defrag Follow those steps: # mount -o autodefrag /dev/sda7 /mnt # dd if=/dev/urandom of=/mnt/tmp bs=200K count=1 # sync # dd if=/dev/urandom of=/mnt/tmp bs=8K count=1 conv=notrunc and then it'll go into a loop: writeback -> defrag -> writeback ... It's because writeback writes [8K, 200K] and then writes [0, 8K]. I tried to make writeback know if the pages are dirtied by defrag, but the patch was a bit intrusive. Here I simply set writeback_index when we defrag a file. Signed-off-by: Li Zefan Signed-off-by: Chris Mason --- fs/btrfs/ioctl.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 6f89bcc4e555..df40b7c5f06b 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -1055,6 +1055,13 @@ int btrfs_defrag_file(struct inode *inode, struct file *file, if (!max_to_defrag) max_to_defrag = last_index - 1; + /* + * make writeback starts from i, so the defrag range can be + * written sequentially. + */ + if (i < inode->i_mapping->writeback_index) + inode->i_mapping->writeback_index = i; + while (i <= last_index && defrag_count < max_to_defrag) { /* * make sure we stop running if someone unmounts From 2b666859ec323403ac9a3a441d16eab30945404b Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Thu, 6 Oct 2011 00:40:47 +0300 Subject: [PATCH 61/75] x86: Default to vsyscall=native for now This UML breakage: linux-2.6.30.1[3800] vsyscall fault (exploit attempt?) ip:ffffffffff600000 cs:33 sp:7fbfb9c498 ax:ffffffffff600000 si:0 di:606790 linux-2.6.30.1[3856] vsyscall fault (exploit attempt?) ip:ffffffffff600000 cs:33 sp:7fbfb13168 ax:ffffffffff600000 si:0 di:606790 Is caused by commit 3ae36655 ("x86-64: Rework vsyscall emulation and add vsyscall= parameter") - the vsyscall emulation code is not fully cooked yet as UML relies on some rather fragile SIGSEGV semantics. Linus suggested in https://lkml.org/lkml/2011/8/9/376 to default to vsyscall=native for now, this patch implements that. Signed-off-by: Adrian Bunk Acked-by: Andrew Lutomirski Cc: H. Peter Anvin Link: http://lkml.kernel.org/r/20111005214047.GE14406@localhost.pp.htv.fi Signed-off-by: Ingo Molnar --- Documentation/kernel-parameters.txt | 7 ++++--- arch/x86/kernel/vsyscall_64.c | 2 +- 2 files changed, 5 insertions(+), 4 deletions(-) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 854ed5ca7e3f..d6e6724446c8 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -2706,10 +2706,11 @@ bytes respectively. Such letter suffixes can also be entirely omitted. functions are at fixed addresses, they make nice targets for exploits that can control RIP. - emulate [default] Vsyscalls turn into traps and are - emulated reasonably safely. + emulate Vsyscalls turn into traps and are emulated + reasonably safely. - native Vsyscalls are native syscall instructions. + native [default] Vsyscalls are native syscall + instructions. This is a little bit faster than trapping and makes a few dynamic recompilers work better than they would in emulation mode. diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c index 18ae83dd1cd7..b56c65de384d 100644 --- a/arch/x86/kernel/vsyscall_64.c +++ b/arch/x86/kernel/vsyscall_64.c @@ -56,7 +56,7 @@ DEFINE_VVAR(struct vsyscall_gtod_data, vsyscall_gtod_data) = .lock = __SEQLOCK_UNLOCKED(__vsyscall_gtod_data.lock), }; -static enum { EMULATE, NATIVE, NONE } vsyscall_mode = EMULATE; +static enum { EMULATE, NATIVE, NONE } vsyscall_mode = NATIVE; static int __init vsyscall_setup(char *str) { From f7f43cc84152e53b5687cd0eb8823310ba065524 Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Tue, 11 Oct 2011 11:41:40 -0400 Subject: [PATCH 62/75] Btrfs: make sure not to defrag extents past i_size The btrfs file defrag code will loop through the extents and force COW on them. But there is a concurrent truncate in the middle of the defrag, it might end up defragging the same range over and over again. The problem is that writepage won't go through and do anything on pages past i_size, so the cow won't happen, so the file will appear to still be fragmented. defrag will end up hitting the same extents again and again. In the worst case, the truncate can actually live lock with the defrag because the defrag keeps creating new ordered extents which the truncate code keeps waiting on. The fix here is to make defrag check for i_size inside the main loop, instead of just once before the looping starts. Signed-off-by: Chris Mason --- fs/btrfs/ioctl.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index df40b7c5f06b..efc4e4a85a69 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -1062,7 +1062,9 @@ int btrfs_defrag_file(struct inode *inode, struct file *file, if (i < inode->i_mapping->writeback_index) inode->i_mapping->writeback_index = i; - while (i <= last_index && defrag_count < max_to_defrag) { + while (i <= last_index && defrag_count < max_to_defrag && + (i < (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> + PAGE_CACHE_SHIFT)) { /* * make sure we stop running if someone unmounts * the FS From bc6e588a8971aa74c02e42db4d6e0248679f3738 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 11 Oct 2011 15:14:08 +0000 Subject: [PATCH 63/75] xfs: do not update xa_last_pushed_lsn for locked items If an item was locked we should not update xa_last_pushed_lsn and thus skip it when restarting the AIL scan as we need to be able to lock and write it out as soon as possible. Otherwise heavy lock contention might starve AIL pushing too easily, especially given the larger backoff once we moved xa_last_pushed_lsn all the way to the target lsn. Signed-off-by: Christoph Hellwig Reported-by: Stefan Priebe Tested-by: Stefan Priebe Reviewed-by: Dave Chinner Signed-off-by: Alex Elder --- fs/xfs/xfs_trans_ail.c | 1 - 1 file changed, 1 deletion(-) diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c index c15aa29fa169..9bd1ee55ef6a 100644 --- a/fs/xfs/xfs_trans_ail.c +++ b/fs/xfs/xfs_trans_ail.c @@ -440,7 +440,6 @@ xfs_ail_worker( case XFS_ITEM_LOCKED: XFS_STATS_INC(xs_push_ail_locked); - ailp->xa_last_pushed_lsn = lsn; stuck++; break; From 17b38471c3c07a49f0bbc2ecc2e92050c164e226 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 11 Oct 2011 15:14:09 +0000 Subject: [PATCH 64/75] xfs: force the log if we encounter pinned buffers in .iop_pushbuf We need to check for pinned buffers even in .iop_pushbuf given that inode items flush into the same buffers that may be pinned directly due operations on the unlinked inode list operating directly on buffers. To do this add a return value to .iop_pushbuf that tells the AIL push about this and use the existing log force mechanisms to unpin it. Signed-off-by: Christoph Hellwig Reported-by: Stefan Priebe Tested-by: Stefan Priebe Reviewed-by: Dave Chinner Signed-off-by: Alex Elder --- fs/xfs/xfs_buf_item.c | 3 ++- fs/xfs/xfs_dquot_item.c | 10 +++++++--- fs/xfs/xfs_inode_item.c | 10 +++++++--- fs/xfs/xfs_trans.h | 2 +- fs/xfs/xfs_trans_ail.c | 9 +++++++-- 5 files changed, 24 insertions(+), 10 deletions(-) diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c index cac2ecfa6746..ef43fce519a1 100644 --- a/fs/xfs/xfs_buf_item.c +++ b/fs/xfs/xfs_buf_item.c @@ -629,7 +629,7 @@ xfs_buf_item_push( * the xfsbufd to get this buffer written. We have to unlock the buffer * to allow the xfsbufd to write it, too. */ -STATIC void +STATIC bool xfs_buf_item_pushbuf( struct xfs_log_item *lip) { @@ -643,6 +643,7 @@ xfs_buf_item_pushbuf( xfs_buf_delwri_promote(bp); xfs_buf_relse(bp); + return true; } STATIC void diff --git a/fs/xfs/xfs_dquot_item.c b/fs/xfs/xfs_dquot_item.c index 9e0e2fa3f2c8..bb3f71d236d2 100644 --- a/fs/xfs/xfs_dquot_item.c +++ b/fs/xfs/xfs_dquot_item.c @@ -183,13 +183,14 @@ xfs_qm_dqunpin_wait( * search the buffer cache can be a time consuming thing, and AIL lock is a * spinlock. */ -STATIC void +STATIC bool xfs_qm_dquot_logitem_pushbuf( struct xfs_log_item *lip) { struct xfs_dq_logitem *qlip = DQUOT_ITEM(lip); struct xfs_dquot *dqp = qlip->qli_dquot; struct xfs_buf *bp; + bool ret = true; ASSERT(XFS_DQ_IS_LOCKED(dqp)); @@ -201,17 +202,20 @@ xfs_qm_dquot_logitem_pushbuf( if (completion_done(&dqp->q_flush) || !(lip->li_flags & XFS_LI_IN_AIL)) { xfs_dqunlock(dqp); - return; + return true; } bp = xfs_incore(dqp->q_mount->m_ddev_targp, qlip->qli_format.qlf_blkno, dqp->q_mount->m_quotainfo->qi_dqchunklen, XBF_TRYLOCK); xfs_dqunlock(dqp); if (!bp) - return; + return true; if (XFS_BUF_ISDELAYWRITE(bp)) xfs_buf_delwri_promote(bp); + if (xfs_buf_ispinned(bp)) + ret = false; xfs_buf_relse(bp); + return ret; } /* diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c index 588406dc6a35..836ad80d4f2b 100644 --- a/fs/xfs/xfs_inode_item.c +++ b/fs/xfs/xfs_inode_item.c @@ -708,13 +708,14 @@ xfs_inode_item_committed( * marked delayed write. If that's the case, we'll promote it and that will * allow the caller to write the buffer by triggering the xfsbufd to run. */ -STATIC void +STATIC bool xfs_inode_item_pushbuf( struct xfs_log_item *lip) { struct xfs_inode_log_item *iip = INODE_ITEM(lip); struct xfs_inode *ip = iip->ili_inode; struct xfs_buf *bp; + bool ret = true; ASSERT(xfs_isilocked(ip, XFS_ILOCK_SHARED)); @@ -725,7 +726,7 @@ xfs_inode_item_pushbuf( if (completion_done(&ip->i_flush) || !(lip->li_flags & XFS_LI_IN_AIL)) { xfs_iunlock(ip, XFS_ILOCK_SHARED); - return; + return true; } bp = xfs_incore(ip->i_mount->m_ddev_targp, iip->ili_format.ilf_blkno, @@ -733,10 +734,13 @@ xfs_inode_item_pushbuf( xfs_iunlock(ip, XFS_ILOCK_SHARED); if (!bp) - return; + return true; if (XFS_BUF_ISDELAYWRITE(bp)) xfs_buf_delwri_promote(bp); + if (xfs_buf_ispinned(bp)) + ret = false; xfs_buf_relse(bp); + return ret; } /* diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h index 06a9759b6352..53597f4db9b5 100644 --- a/fs/xfs/xfs_trans.h +++ b/fs/xfs/xfs_trans.h @@ -350,7 +350,7 @@ typedef struct xfs_item_ops { void (*iop_unlock)(xfs_log_item_t *); xfs_lsn_t (*iop_committed)(xfs_log_item_t *, xfs_lsn_t); void (*iop_push)(xfs_log_item_t *); - void (*iop_pushbuf)(xfs_log_item_t *); + bool (*iop_pushbuf)(xfs_log_item_t *); void (*iop_committing)(xfs_log_item_t *, xfs_lsn_t); } xfs_item_ops_t; diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c index 9bd1ee55ef6a..e4cd9180d33a 100644 --- a/fs/xfs/xfs_trans_ail.c +++ b/fs/xfs/xfs_trans_ail.c @@ -427,8 +427,13 @@ xfs_ail_worker( case XFS_ITEM_PUSHBUF: XFS_STATS_INC(xs_push_ail_pushbuf); - IOP_PUSHBUF(lip); - ailp->xa_last_pushed_lsn = lsn; + + if (!IOP_PUSHBUF(lip)) { + stuck++; + flush_log = 1; + } else { + ailp->xa_last_pushed_lsn = lsn; + } push_xfsbufd = 1; break; From 0030807c66f058230bcb20d2573bcaf28852e804 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 11 Oct 2011 11:14:10 -0400 Subject: [PATCH 65/75] xfs: revert to using a kthread for AIL pushing Currently we have a few issues with the way the workqueue code is used to implement AIL pushing: - it accidentally uses the same workqueue as the syncer action, and thus can be prevented from running if there are enough sync actions active in the system. - it doesn't use the HIGHPRI flag to queue at the head of the queue of work items At this point I'm not confident enough in getting all the workqueue flags and tweaks right to provide a perfectly reliable execution context for AIL pushing, which is the most important piece in XFS to make forward progress when the log fills. Revert back to use a kthread per filesystem which fixes all the above issues at the cost of having a task struct and stack around for each mounted filesystem. In addition this also gives us much better ways to diagnose any issues involving hung AIL pushing and removes a small amount of code. Signed-off-by: Christoph Hellwig Reported-by: Stefan Priebe Tested-by: Stefan Priebe Reviewed-by: Dave Chinner Signed-off-by: Alex Elder --- fs/xfs/xfs_linux.h | 2 ++ fs/xfs/xfs_super.c | 13 +------- fs/xfs/xfs_trans_ail.c | 73 +++++++++++++++++++++++------------------ fs/xfs/xfs_trans_priv.h | 8 +---- 4 files changed, 45 insertions(+), 51 deletions(-) diff --git a/fs/xfs/xfs_linux.h b/fs/xfs/xfs_linux.h index 1e8a45e74c3e..828662f70d64 100644 --- a/fs/xfs/xfs_linux.h +++ b/fs/xfs/xfs_linux.h @@ -68,6 +68,8 @@ #include #include #include +#include +#include #include #include diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 2366c54cc4fa..5cf06b85fd9d 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -1652,24 +1652,13 @@ xfs_init_workqueues(void) */ xfs_syncd_wq = alloc_workqueue("xfssyncd", WQ_CPU_INTENSIVE, 8); if (!xfs_syncd_wq) - goto out; - - xfs_ail_wq = alloc_workqueue("xfsail", WQ_CPU_INTENSIVE, 8); - if (!xfs_ail_wq) - goto out_destroy_syncd; - + return -ENOMEM; return 0; - -out_destroy_syncd: - destroy_workqueue(xfs_syncd_wq); -out: - return -ENOMEM; } STATIC void xfs_destroy_workqueues(void) { - destroy_workqueue(xfs_ail_wq); destroy_workqueue(xfs_syncd_wq); } diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c index e4cd9180d33a..3a1e7ca54c2d 100644 --- a/fs/xfs/xfs_trans_ail.c +++ b/fs/xfs/xfs_trans_ail.c @@ -28,8 +28,6 @@ #include "xfs_trans_priv.h" #include "xfs_error.h" -struct workqueue_struct *xfs_ail_wq; /* AIL workqueue */ - #ifdef DEBUG /* * Check that the list is sorted as it should be. @@ -356,16 +354,10 @@ xfs_ail_delete( xfs_trans_ail_cursor_clear(ailp, lip); } -/* - * xfs_ail_worker does the work of pushing on the AIL. It will requeue itself - * to run at a later time if there is more work to do to complete the push. - */ -STATIC void -xfs_ail_worker( - struct work_struct *work) +static long +xfsaild_push( + struct xfs_ail *ailp) { - struct xfs_ail *ailp = container_of(to_delayed_work(work), - struct xfs_ail, xa_work); xfs_mount_t *mp = ailp->xa_mount; struct xfs_ail_cursor cur; xfs_log_item_t *lip; @@ -505,20 +497,6 @@ out_done: /* We're past our target or empty, so idle */ ailp->xa_last_pushed_lsn = 0; - /* - * We clear the XFS_AIL_PUSHING_BIT first before checking - * whether the target has changed. If the target has changed, - * this pushes the requeue race directly onto the result of the - * atomic test/set bit, so we are guaranteed that either the - * the pusher that changed the target or ourselves will requeue - * the work (but not both). - */ - clear_bit(XFS_AIL_PUSHING_BIT, &ailp->xa_flags); - smp_rmb(); - if (XFS_LSN_CMP(ailp->xa_target, target) == 0 || - test_and_set_bit(XFS_AIL_PUSHING_BIT, &ailp->xa_flags)) - return; - tout = 50; } else if (XFS_LSN_CMP(lsn, target) >= 0) { /* @@ -541,9 +519,30 @@ out_done: tout = 20; } - /* There is more to do, requeue us. */ - queue_delayed_work(xfs_syncd_wq, &ailp->xa_work, - msecs_to_jiffies(tout)); + return tout; +} + +static int +xfsaild( + void *data) +{ + struct xfs_ail *ailp = data; + long tout = 0; /* milliseconds */ + + while (!kthread_should_stop()) { + if (tout && tout <= 20) + __set_current_state(TASK_KILLABLE); + else + __set_current_state(TASK_INTERRUPTIBLE); + schedule_timeout(tout ? + msecs_to_jiffies(tout) : MAX_SCHEDULE_TIMEOUT); + + try_to_freeze(); + + tout = xfsaild_push(ailp); + } + + return 0; } /* @@ -578,8 +577,9 @@ xfs_ail_push( */ smp_wmb(); xfs_trans_ail_copy_lsn(ailp, &ailp->xa_target, &threshold_lsn); - if (!test_and_set_bit(XFS_AIL_PUSHING_BIT, &ailp->xa_flags)) - queue_delayed_work(xfs_syncd_wq, &ailp->xa_work, 0); + smp_wmb(); + + wake_up_process(ailp->xa_task); } /* @@ -817,9 +817,18 @@ xfs_trans_ail_init( INIT_LIST_HEAD(&ailp->xa_ail); INIT_LIST_HEAD(&ailp->xa_cursors); spin_lock_init(&ailp->xa_lock); - INIT_DELAYED_WORK(&ailp->xa_work, xfs_ail_worker); + + ailp->xa_task = kthread_run(xfsaild, ailp, "xfsaild/%s", + ailp->xa_mount->m_fsname); + if (IS_ERR(ailp->xa_task)) + goto out_free_ailp; + mp->m_ail = ailp; return 0; + +out_free_ailp: + kmem_free(ailp); + return ENOMEM; } void @@ -828,6 +837,6 @@ xfs_trans_ail_destroy( { struct xfs_ail *ailp = mp->m_ail; - cancel_delayed_work_sync(&ailp->xa_work); + kthread_stop(ailp->xa_task); kmem_free(ailp); } diff --git a/fs/xfs/xfs_trans_priv.h b/fs/xfs/xfs_trans_priv.h index 212946b97239..22750b5e4a8f 100644 --- a/fs/xfs/xfs_trans_priv.h +++ b/fs/xfs/xfs_trans_priv.h @@ -64,23 +64,17 @@ struct xfs_ail_cursor { */ struct xfs_ail { struct xfs_mount *xa_mount; + struct task_struct *xa_task; struct list_head xa_ail; xfs_lsn_t xa_target; struct list_head xa_cursors; spinlock_t xa_lock; - struct delayed_work xa_work; xfs_lsn_t xa_last_pushed_lsn; - unsigned long xa_flags; }; -#define XFS_AIL_PUSHING_BIT 0 - /* * From xfs_trans_ail.c */ - -extern struct workqueue_struct *xfs_ail_wq; /* AIL workqueue */ - void xfs_trans_ail_update_bulk(struct xfs_ail *ailp, struct xfs_ail_cursor *cur, struct xfs_log_item **log_items, int nr_items, From 1d1136012163ab5e33a97f9779af269cf4049cb9 Mon Sep 17 00:00:00 2001 From: Johann Felix Soden Date: Mon, 10 Oct 2011 11:37:00 -0700 Subject: [PATCH 66/75] ide: Fix file references in drivers/ide/ Fix file references in drivers/ide/ There are a lot of file references to now moved or deleted files in the whole tree, especially in documentation and Kconfig files. This patch fixes the references in drivers/ide/. Signed-off-by: Johann Felix Soden Signed-off-by: Randy Dunlap Acked-by: David S. Miller Signed-off-by: Linus Torvalds --- drivers/ide/Kconfig | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/drivers/ide/Kconfig b/drivers/ide/Kconfig index 9827c5e686cb..811dbbd9306c 100644 --- a/drivers/ide/Kconfig +++ b/drivers/ide/Kconfig @@ -327,7 +327,7 @@ config BLK_DEV_OPTI621 select BLK_DEV_IDEPCI help This is a driver for the OPTi 82C621 EIDE controller. - Please read the comments at the top of . + Please read the comments at the top of . config BLK_DEV_RZ1000 tristate "RZ1000 chipset bugfix/support" @@ -365,7 +365,7 @@ config BLK_DEV_ALI15X3 normal dual channel support. Please read the comments at the top of - . + . If unsure, say N. @@ -528,7 +528,7 @@ config BLK_DEV_NS87415 This driver adds detection and support for the NS87415 chip (used mainly on SPARC64 and PA-RISC machines). - Please read the comments at the top of . + Please read the comments at the top of . config BLK_DEV_PDC202XX_OLD tristate "PROMISE PDC202{46|62|65|67} support" @@ -547,7 +547,7 @@ config BLK_DEV_PDC202XX_OLD for more than one card. Please read the comments at the top of - . + . If unsure, say N. @@ -593,7 +593,7 @@ config BLK_DEV_SIS5513 ATA100: SiS635, SiS645, SiS650, SiS730, SiS735, SiS740, SiS745, SiS750 - Please read the comments at the top of . + Please read the comments at the top of . config BLK_DEV_SL82C105 tristate "Winbond SL82c105 support" @@ -616,7 +616,7 @@ config BLK_DEV_SLC90E66 look-a-like to the PIIX4 it should be a nice addition. Please read the comments at the top of - . + . config BLK_DEV_TRM290 tristate "Tekram TRM290 chipset support" @@ -625,7 +625,7 @@ config BLK_DEV_TRM290 This driver adds support for bus master DMA transfers using the Tekram TRM290 PCI IDE chip. Volunteers are needed for further tweaking and development. - Please read the comments at the top of . + Please read the comments at the top of . config BLK_DEV_VIA82CXXX tristate "VIA82CXXX chipset support" @@ -836,7 +836,7 @@ config BLK_DEV_ALI14XX of the ALI M1439/1443/1445/1487/1489 chipsets, and permits faster I/O speeds to be set as well. See the files and - for more info. + for more info. config BLK_DEV_DTC2278 tristate "DTC-2278 support" @@ -847,7 +847,7 @@ config BLK_DEV_DTC2278 boot parameter. It enables support for the secondary IDE interface of the DTC-2278 card, and permits faster I/O speeds to be set as well. See the and - files for more info. + files for more info. config BLK_DEV_HT6560B tristate "Holtek HT6560B support" @@ -858,7 +858,7 @@ config BLK_DEV_HT6560B boot parameter. It enables support for the secondary IDE interface of the Holtek card, and permits faster I/O speeds to be set as well. See the and - files for more info. + files for more info. config BLK_DEV_QD65XX tristate "QDI QD65xx support" @@ -867,7 +867,7 @@ config BLK_DEV_QD65XX help This driver is enabled at runtime using the "qd65xx.probe" kernel boot parameter. It permits faster I/O speeds to be set. See the - and + and for more info. config BLK_DEV_UMC8672 @@ -879,7 +879,7 @@ config BLK_DEV_UMC8672 boot parameter. It enables support for the secondary IDE interface of the UMC-8672, and permits faster I/O speeds to be set as well. See the files and - for more info. + for more info. endif From d52104b29a3735f2b4dceefcb01b19d370a783ea Mon Sep 17 00:00:00 2001 From: Chris Metcalf Date: Wed, 5 Oct 2011 17:09:29 -0400 Subject: [PATCH 67/75] tile: revert change from to in asm files The 32-bit TILEPro support uses some #defines in for atomic support routines in assembly. To make this more explicit, I've turned those includes into includes of , which should hopefully make it clear that they shouldn't be bombed into in any cleanups. Signed-off-by: Chris Metcalf --- arch/tile/kernel/intvec_32.S | 2 +- arch/tile/lib/atomic_asm_32.S | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/tile/kernel/intvec_32.S b/arch/tile/kernel/intvec_32.S index fc94607f0bd5..aecc8ed5f39b 100644 --- a/arch/tile/kernel/intvec_32.S +++ b/arch/tile/kernel/intvec_32.S @@ -21,7 +21,7 @@ #include #include #include -#include +#include #include #include #include diff --git a/arch/tile/lib/atomic_asm_32.S b/arch/tile/lib/atomic_asm_32.S index 1f75a2a56101..30638042691d 100644 --- a/arch/tile/lib/atomic_asm_32.S +++ b/arch/tile/lib/atomic_asm_32.S @@ -70,7 +70,7 @@ */ #include -#include +#include #include #include From 78a43158724793f9dc25f1e4c866393654704b87 Mon Sep 17 00:00:00 2001 From: Janusz Krzysztofik Date: Tue, 23 Aug 2011 13:42:24 +0200 Subject: [PATCH 68/75] gpio/omap: fix build error with certain OMAP1 configs With commit f64ad1a0e21a, "gpio/omap: cleanup _set_gpio_wakeup(), remove ifdefs", access to build time conditionally omitted 'suspend_wakeup' member of the 'gpio_bank' structure has been placed unconditionally in function _set_gpio_wakeup(), which is always built. This resulted in the driver compilation broken for certain OMAP1, i.e., non-OMAP16xx, configurations. Really required or not in previously excluded cases, define this structure member unconditionally as a fix. Tested with a custom OMAP1510 only configuration. Signed-off-by: Janusz Krzysztofik Acked-by: Kevin Hilman Tested-by: Aaro Koskinen Signed-off-by: Grant Likely --- drivers/gpio/gpio-omap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpio/gpio-omap.c b/drivers/gpio/gpio-omap.c index 0599854e2217..118ec12d2d5f 100644 --- a/drivers/gpio/gpio-omap.c +++ b/drivers/gpio/gpio-omap.c @@ -34,8 +34,8 @@ struct gpio_bank { u16 irq; u16 virtual_irq_start; int method; -#if defined(CONFIG_ARCH_OMAP16XX) || defined(CONFIG_ARCH_OMAP2PLUS) u32 suspend_wakeup; +#if defined(CONFIG_ARCH_OMAP16XX) || defined(CONFIG_ARCH_OMAP2PLUS) u32 saved_wakeup; #endif u32 non_wakeup_gpios; From 25fcf2b7f1f65d2cc12182ced3ccd47576970be4 Mon Sep 17 00:00:00 2001 From: Hartmut Knaack Date: Tue, 11 Oct 2011 00:22:45 +0200 Subject: [PATCH 69/75] gpio-pca953x: fix gpio_base gpio_base was set to 0 if no system platform data or open firmware platform data was provided. This led to conflicts, if any other gpiochip with a gpiobase of 0 was instantiated already. Setting it to -1 will automatically use the first one available. Signed-off-by: Hartmut Knaack Signed-off-by: Grant Likely --- drivers/gpio/gpio-pca953x.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpio/gpio-pca953x.c b/drivers/gpio/gpio-pca953x.c index c43b8ff626a7..0550dcb85814 100644 --- a/drivers/gpio/gpio-pca953x.c +++ b/drivers/gpio/gpio-pca953x.c @@ -577,6 +577,7 @@ pca953x_get_alt_pdata(struct i2c_client *client, int *gpio_base, int *invert) void pca953x_get_alt_pdata(struct i2c_client *client, int *gpio_base, int *invert) { + *gpio_base = -1; } #endif From bf164c58e58328c40ebc597a8ac00cc6840f9703 Mon Sep 17 00:00:00 2001 From: Jean Delvare Date: Thu, 13 Oct 2011 15:49:08 -0400 Subject: [PATCH 70/75] hwmon: (w83627ehf) Properly report thermal diode sensors The w83627ehf driver is improperly reporting thermal diode sensors as type 2, instead of 3. This caused "sensors" and possibly other monitoring tools to report these sensors as "transistor" instead of "thermal diode". Furthermore, diode subtype selection (CPU vs. external) is only supported by the original W83627EHF/EHG. All later models only support CPU diode type, and some (NCT6776F) don't even have the register in question so we should avoid reading from it. Signed-off-by: Jean Delvare Cc: stable@kernel.org Signed-off-by: Guenter Roeck --- drivers/hwmon/w83627ehf.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/drivers/hwmon/w83627ehf.c b/drivers/hwmon/w83627ehf.c index f2b377c56a3a..7b0260dc76fb 100644 --- a/drivers/hwmon/w83627ehf.c +++ b/drivers/hwmon/w83627ehf.c @@ -1715,7 +1715,8 @@ static void w83627ehf_device_remove_files(struct device *dev) } /* Get the monitoring functions started */ -static inline void __devinit w83627ehf_init_device(struct w83627ehf_data *data) +static inline void __devinit w83627ehf_init_device(struct w83627ehf_data *data, + enum kinds kind) { int i; u8 tmp, diode; @@ -1746,10 +1747,16 @@ static inline void __devinit w83627ehf_init_device(struct w83627ehf_data *data) w83627ehf_write_value(data, W83627EHF_REG_VBAT, tmp | 0x01); /* Get thermal sensor types */ - diode = w83627ehf_read_value(data, W83627EHF_REG_DIODE); + switch (kind) { + case w83627ehf: + diode = w83627ehf_read_value(data, W83627EHF_REG_DIODE); + break; + default: + diode = 0x70; + } for (i = 0; i < 3; i++) { if ((tmp & (0x02 << i))) - data->temp_type[i] = (diode & (0x10 << i)) ? 1 : 2; + data->temp_type[i] = (diode & (0x10 << i)) ? 1 : 3; else data->temp_type[i] = 4; /* thermistor */ } @@ -2016,7 +2023,7 @@ static int __devinit w83627ehf_probe(struct platform_device *pdev) } /* Initialize the chip */ - w83627ehf_init_device(data); + w83627ehf_init_device(data, sio_data->kind); data->vrm = vid_which_vrm(); superio_enter(sio_data->sioreg); From 153b19a3b9fd8b9478495b9ee1f93f6a77c564f9 Mon Sep 17 00:00:00 2001 From: Mika Westerberg Date: Thu, 13 Oct 2011 12:04:20 +0300 Subject: [PATCH 71/75] x86, mrst: use a temporary variable for SFI irq SFI tables reside in RAM and should not be modified once they are written. Current code went to set pentry->irq to zero which causes subsequent reads to fail with invalid SFI table checksum. This will break kexec as the second kernel fails to validate SFI tables. To fix this we use temporary variable for irq number. Signed-off-by: Mika Westerberg Reviewed-by: Kirill A. Shutemov Cc: stable@kernel.org Signed-off-by: Linus Torvalds --- arch/x86/platform/mrst/mrst.c | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-) diff --git a/arch/x86/platform/mrst/mrst.c b/arch/x86/platform/mrst/mrst.c index 58425adc22c6..fe73276e026b 100644 --- a/arch/x86/platform/mrst/mrst.c +++ b/arch/x86/platform/mrst/mrst.c @@ -678,38 +678,40 @@ static int __init sfi_parse_devs(struct sfi_table_header *table) pentry = (struct sfi_device_table_entry *)sb->pentry; for (i = 0; i < num; i++, pentry++) { - if (pentry->irq != (u8)0xff) { /* native RTE case */ + int irq = pentry->irq; + + if (irq != (u8)0xff) { /* native RTE case */ /* these SPI2 devices are not exposed to system as PCI * devices, but they have separate RTE entry in IOAPIC * so we have to enable them one by one here */ - ioapic = mp_find_ioapic(pentry->irq); + ioapic = mp_find_ioapic(irq); irq_attr.ioapic = ioapic; - irq_attr.ioapic_pin = pentry->irq; + irq_attr.ioapic_pin = irq; irq_attr.trigger = 1; irq_attr.polarity = 1; - io_apic_set_pci_routing(NULL, pentry->irq, &irq_attr); + io_apic_set_pci_routing(NULL, irq, &irq_attr); } else - pentry->irq = 0; /* No irq */ + irq = 0; /* No irq */ switch (pentry->type) { case SFI_DEV_TYPE_IPC: /* ID as IRQ is a hack that will go away */ - pdev = platform_device_alloc(pentry->name, pentry->irq); + pdev = platform_device_alloc(pentry->name, irq); if (pdev == NULL) { pr_err("out of memory for SFI platform device '%s'.\n", pentry->name); continue; } - install_irq_resource(pdev, pentry->irq); + install_irq_resource(pdev, irq); pr_debug("info[%2d]: IPC bus, name = %16.16s, " - "irq = 0x%2x\n", i, pentry->name, pentry->irq); + "irq = 0x%2x\n", i, pentry->name, irq); sfi_handle_ipc_dev(pdev); break; case SFI_DEV_TYPE_SPI: memset(&spi_info, 0, sizeof(spi_info)); strncpy(spi_info.modalias, pentry->name, SFI_NAME_LEN); - spi_info.irq = pentry->irq; + spi_info.irq = irq; spi_info.bus_num = pentry->host_num; spi_info.chip_select = pentry->addr; spi_info.max_speed_hz = pentry->max_freq; @@ -726,7 +728,7 @@ static int __init sfi_parse_devs(struct sfi_table_header *table) memset(&i2c_info, 0, sizeof(i2c_info)); bus = pentry->host_num; strncpy(i2c_info.type, pentry->name, SFI_NAME_LEN); - i2c_info.irq = pentry->irq; + i2c_info.irq = irq; i2c_info.addr = pentry->addr; pr_debug("info[%2d]: I2C bus = %d, name = %16.16s, " "irq = 0x%2x, addr = 0x%x\n", i, bus, From 29a541f6c1f6e4a85628bb86071b9e72c9f8be2c Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Mon, 3 Oct 2011 18:30:53 +0100 Subject: [PATCH 72/75] ARM: 7117/1: perf: fix HW_CACHE_* events on Cortex-A9 Using COHERENT_LINE_{MISS,HIT} for cache misses and references respectively is completely wrong. Instead, use the L1D events which are a better and more useful approximation despite ignoring instruction traffic. Reported-by: Alasdair Grant Reported-by: Matt Horsnell Reported-by: Michael Williams Cc: stable@kernel.org Cc: Jean Pihet Signed-off-by: Will Deacon Signed-off-by: Russell King --- arch/arm/kernel/perf_event_v7.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/kernel/perf_event_v7.c b/arch/arm/kernel/perf_event_v7.c index 4c851834f68e..6be3e2e4d838 100644 --- a/arch/arm/kernel/perf_event_v7.c +++ b/arch/arm/kernel/perf_event_v7.c @@ -321,8 +321,8 @@ static const unsigned armv7_a9_perf_map[PERF_COUNT_HW_MAX] = { [PERF_COUNT_HW_CPU_CYCLES] = ARMV7_PERFCTR_CPU_CYCLES, [PERF_COUNT_HW_INSTRUCTIONS] = ARMV7_PERFCTR_INST_OUT_OF_RENAME_STAGE, - [PERF_COUNT_HW_CACHE_REFERENCES] = ARMV7_PERFCTR_COHERENT_LINE_HIT, - [PERF_COUNT_HW_CACHE_MISSES] = ARMV7_PERFCTR_COHERENT_LINE_MISS, + [PERF_COUNT_HW_CACHE_REFERENCES] = ARMV7_PERFCTR_DCACHE_ACCESS, + [PERF_COUNT_HW_CACHE_MISSES] = ARMV7_PERFCTR_DCACHE_REFILL, [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = ARMV7_PERFCTR_PC_WRITE, [PERF_COUNT_HW_BRANCH_MISSES] = ARMV7_PERFCTR_PC_BRANCH_MIS_PRED, [PERF_COUNT_HW_BUS_CYCLES] = ARMV7_PERFCTR_CLOCK_CYCLES, From bb1ac3ec95cf0cce5145617b140f2090ed45d565 Mon Sep 17 00:00:00 2001 From: Shawn Guo Date: Thu, 6 Oct 2011 14:57:24 +0100 Subject: [PATCH 73/75] ARM: 7122/1: localtimer: add header linux/errno.h explicitly MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Per the text in Documentation/SubmitChecklist as below, we should explicitly have header linux/errno.h in localtimer.h for ENXIO reference. 1: If you use a facility then #include the file that defines/declares that facility. Don't depend on other header files pulling in ones that you use. Otherwise, we may run into some compiling error like the following one, if any file includes localtimer.h without CONFIG_LOCAL_TIMERS defined. arch/arm/include/asm/localtimer.h: In function ‘local_timer_setup’: arch/arm/include/asm/localtimer.h:53:10: error: ‘ENXIO’ undeclared (first use in this function) Signed-off-by: Shawn Guo Signed-off-by: Russell King --- arch/arm/include/asm/localtimer.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm/include/asm/localtimer.h b/arch/arm/include/asm/localtimer.h index 080d74f8128d..ff66638ff54d 100644 --- a/arch/arm/include/asm/localtimer.h +++ b/arch/arm/include/asm/localtimer.h @@ -10,6 +10,8 @@ #ifndef __ASM_ARM_LOCALTIMER_H #define __ASM_ARM_LOCALTIMER_H +#include + struct clock_event_device; /* From f8be12d15374413cf437741a282e97129e759209 Mon Sep 17 00:00:00 2001 From: Zoltan Devai Date: Mon, 10 Oct 2011 14:54:12 +0100 Subject: [PATCH 74/75] ARM: 7128/1: vic: Don't write to the read-only register VIC_IRQ_STATUS This is unneeded and causes an abort on the SPMP8000 platform. Acked-by: Linus Walleij Signed-off-by: Zoltan Devai Signed-off-by: Russell King --- arch/arm/common/vic.c | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/arm/common/vic.c b/arch/arm/common/vic.c index 7aa4262ada7a..197f81c77351 100644 --- a/arch/arm/common/vic.c +++ b/arch/arm/common/vic.c @@ -259,7 +259,6 @@ static void __init vic_disable(void __iomem *base) writel(0, base + VIC_INT_SELECT); writel(0, base + VIC_INT_ENABLE); writel(~0, base + VIC_INT_ENABLE_CLEAR); - writel(0, base + VIC_IRQ_STATUS); writel(0, base + VIC_ITCR); writel(~0, base + VIC_INT_SOFT_CLEAR); } From a84a79e4d369a73c0130b5858199e949432da4c6 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Mon, 17 Oct 2011 08:24:24 -0700 Subject: [PATCH 75/75] Avoid using variable-length arrays in kernel/sys.c The size is always valid, but variable-length arrays generate worse code for no good reason (unless the function happens to be inlined and the compiler sees the length for the simple constant it is). Also, there seems to be some code generation problem on POWER, where Henrik Bakken reports that register r28 can get corrupted under some subtle circumstances (interrupt happening at the wrong time?). That all indicates some seriously broken compiler issues, but since variable length arrays are bad regardless, there's little point in trying to chase it down. "Just don't do that, then". Reported-by: Henrik Grindal Bakken Cc: Benjamin Herrenschmidt Cc: stable@kernel.org Signed-off-by: Linus Torvalds --- kernel/sys.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/sys.c b/kernel/sys.c index 18ee1d2f6474..1dbbe695a5ef 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -1172,7 +1172,7 @@ DECLARE_RWSEM(uts_sem); static int override_release(char __user *release, int len) { int ret = 0; - char buf[len]; + char buf[65]; if (current->personality & UNAME26) { char *rest = UTS_RELEASE;