From 6a0bfff44e4aa4ee1721b3daa004d2039576c70d Mon Sep 17 00:00:00 2001 From: Tim Pepper Date: Mon, 27 Oct 2008 12:18:36 +0800 Subject: [PATCH 001/160] Blackfin arch: handle case of d_path() returning error in decode_address() d_path() can return an error. Most of its callers do something or other to make up something sane in that case. Do similar for blackfin's decode_address() call to d_path(). Signed-off-by: Tim Pepper Signed-off-by: Bryan Wu --- arch/blackfin/kernel/traps.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/arch/blackfin/kernel/traps.c b/arch/blackfin/kernel/traps.c index 1aa2c788e22..0003616d02a 100644 --- a/arch/blackfin/kernel/traps.c +++ b/arch/blackfin/kernel/traps.c @@ -147,9 +147,12 @@ static void decode_address(char *buf, unsigned long address) char *name = p->comm; struct file *file = vma->vm_file; - if (file) - name = d_path(&file->f_path, _tmpbuf, + if (file) { + char *d_name = d_path(&file->f_path, _tmpbuf, sizeof(_tmpbuf)); + if (!IS_ERR(d_name)) + name = d_name; + } /* FLAT does not have its text aligned to the start of * the map while FDPIC ELF does ... From 6776cf4476833df0f1e96bd9dba18c1ea4f582d5 Mon Sep 17 00:00:00 2001 From: Graf Yang Date: Mon, 27 Oct 2008 18:12:53 +0800 Subject: [PATCH 002/160] Blackfin arch: fix bug - Fail to boot jffs2 kernel for BF561 with SMP patch only if the cplb block overlapped with kernel area, this cplb need be locked Signed-off-by: Graf Yang Signed-off-by: Bryan Wu --- arch/blackfin/kernel/cplb-nompu/cplbinit.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/arch/blackfin/kernel/cplb-nompu/cplbinit.c b/arch/blackfin/kernel/cplb-nompu/cplbinit.c index 512f8c92ead..2debc900e24 100644 --- a/arch/blackfin/kernel/cplb-nompu/cplbinit.c +++ b/arch/blackfin/kernel/cplb-nompu/cplbinit.c @@ -188,10 +188,11 @@ static struct cplb_desc cplb_data[] = { static u16 __init lock_kernel_check(u32 start, u32 end) { - if ((end <= (u32) _end && end >= (u32)_stext) || - (start <= (u32) _end && start >= (u32)_stext)) - return IN_KERNEL; - return 0; + if (start >= (u32)_end || end <= (u32)_stext) + return 0; + + /* This cplb block overlapped with kernel area. */ + return IN_KERNEL; } static unsigned short __init From 3b1f26a50a2bfbd2825345b49b1d7f78432a7a4c Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Mon, 27 Oct 2008 18:21:43 +0800 Subject: [PATCH 003/160] Blackfin arch: don't copy bss when copying L1 when copying L1 regions, go to the start of bss rather than end since we have code to zero it out already Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu --- arch/blackfin/include/asm/bfin-global.h | 2 +- arch/blackfin/kernel/setup.c | 10 +++++----- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/blackfin/include/asm/bfin-global.h b/arch/blackfin/include/asm/bfin-global.h index 56dcb0a2d24..77295666c34 100644 --- a/arch/blackfin/include/asm/bfin-global.h +++ b/arch/blackfin/include/asm/bfin-global.h @@ -101,7 +101,7 @@ extern u16 _bfin_swrst; /* shadow for Software Reset Register (SWRST) */ extern unsigned long _ramstart, _ramend, _rambase; extern unsigned long memory_start, memory_end, physical_mem_end; extern char _stext_l1[], _etext_l1[], _sdata_l1[], _edata_l1[], _sbss_l1[], - _ebss_l1[], _l1_lma_start[], _sdata_b_l1[], _ebss_b_l1[], + _ebss_l1[], _l1_lma_start[], _sdata_b_l1[], _sbss_b_l1[], _ebss_b_l1[], _stext_l2[], _etext_l2[], _sdata_l2[], _edata_l2[], _sbss_l2[], _ebss_l2[], _l2_lma_start[]; diff --git a/arch/blackfin/kernel/setup.c b/arch/blackfin/kernel/setup.c index 7f35d1046cd..8337dc3a62d 100644 --- a/arch/blackfin/kernel/setup.c +++ b/arch/blackfin/kernel/setup.c @@ -119,23 +119,23 @@ void __init bfin_relocate_l1_mem(void) /* Copy _stext_l1 to _etext_l1 to L1 instruction SRAM */ dma_memcpy(_stext_l1, _l1_lma_start, l1_code_length); - l1_data_a_length = _ebss_l1 - _sdata_l1; + l1_data_a_length = _sbss_l1 - _sdata_l1; if (l1_data_a_length > L1_DATA_A_LENGTH) panic("L1 Data SRAM Bank A Overflow\n"); - /* Copy _sdata_l1 to _ebss_l1 to L1 data bank A SRAM */ + /* Copy _sdata_l1 to _sbss_l1 to L1 data bank A SRAM */ dma_memcpy(_sdata_l1, _l1_lma_start + l1_code_length, l1_data_a_length); - l1_data_b_length = _ebss_b_l1 - _sdata_b_l1; + l1_data_b_length = _sbss_b_l1 - _sdata_b_l1; if (l1_data_b_length > L1_DATA_B_LENGTH) panic("L1 Data SRAM Bank B Overflow\n"); - /* Copy _sdata_b_l1 to _ebss_b_l1 to L1 data bank B SRAM */ + /* Copy _sdata_b_l1 to _sbss_b_l1 to L1 data bank B SRAM */ dma_memcpy(_sdata_b_l1, _l1_lma_start + l1_code_length + l1_data_a_length, l1_data_b_length); if (L2_LENGTH != 0) { - l2_length = _ebss_l2 - _stext_l2; + l2_length = _sbss_l2 - _stext_l2; if (l2_length > L2_LENGTH) panic("L2 SRAM Overflow\n"); From 4ee1c45337e7b529eed644c6f62399d797dcbc10 Mon Sep 17 00:00:00 2001 From: Robin Getz Date: Tue, 28 Oct 2008 11:36:11 +0800 Subject: [PATCH 004/160] Blackfin arch: Fix typo when adding CONFIG_DEBUG_VERBOSE Signed-off-by: Robin Getz Signed-off-by: Bryan Wu --- arch/blackfin/kernel/traps.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/blackfin/kernel/traps.c b/arch/blackfin/kernel/traps.c index 0003616d02a..bef025b0744 100644 --- a/arch/blackfin/kernel/traps.c +++ b/arch/blackfin/kernel/traps.c @@ -59,7 +59,7 @@ #endif -#ifdef CONFIG_VERBOSE_DEBUG +#ifdef CONFIG_DEBUG_VERBOSE #define verbose_printk(fmt, arg...) \ printk(fmt, ##arg) #else @@ -574,7 +574,7 @@ asmlinkage void trap_c(struct pt_regs *fp) #endif panic("Kernel exception"); } else { -#ifdef CONFIG_VERBOSE_DEBUG +#ifdef CONFIG_DEBUG_VERBOSE unsigned long *stack; /* Dump the user space stack */ stack = (unsigned long *)rdusp(); From 39e96c8835c36b6867b4e18698b06746972cdfcc Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Tue, 18 Nov 2008 17:48:22 +0800 Subject: [PATCH 005/160] Blackfin arch: fix bug - dmacopy test case fail on all platform The cache code I added flushes 1 line too little if the start address is not aligned to the cache size. Cache align the start address so that when we straddle cache aligns, we get the right count. Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu --- arch/blackfin/mach-common/cache.S | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/blackfin/mach-common/cache.S b/arch/blackfin/mach-common/cache.S index db532181fbd..d6780b49524 100644 --- a/arch/blackfin/mach-common/cache.S +++ b/arch/blackfin/mach-common/cache.S @@ -25,9 +25,13 @@ */ .macro do_flush flushins:req optflushins optnopins label + R2 = -L1_CACHE_BYTES; + + /* start = (start & -L1_CACHE_BYTES) */ + R0 = R0 & R2; + /* end = ((end - 1) & -L1_CACHE_BYTES) + L1_CACHE_BYTES; */ R1 += -1; - R2 = -L1_CACHE_BYTES; R1 = R1 & R2; R1 += L1_CACHE_BYTES; From 7f6b2e7b1ff70bc60cedc9a00b01c1fad5c21371 Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Tue, 28 Oct 2008 12:29:26 +0800 Subject: [PATCH 006/160] Blackfin arch: fix bug - kernel build with write back policy fails to be booted up Make sure IFLUSH is not the last instruction in the hardware loop to avoid infinite core stall. The dcache/icache function that only gets used in writeback mode was putting IFLUSH as the last instruction in the hardware loop ... we know from design that this may often lead to inifite core stalling, so switch the FLUSH/IFLUSH order. Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu --- arch/blackfin/mach-common/cache.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/blackfin/mach-common/cache.S b/arch/blackfin/mach-common/cache.S index d6780b49524..a028e945041 100644 --- a/arch/blackfin/mach-common/cache.S +++ b/arch/blackfin/mach-common/cache.S @@ -67,7 +67,7 @@ ENDPROC(_blackfin_icache_flush_range) /* Flush all cache lines assocoiated with this area of memory. */ ENTRY(_blackfin_icache_dcache_flush_range) - do_flush IFLUSH, FLUSH + do_flush FLUSH, IFLUSH ENDPROC(_blackfin_icache_dcache_flush_range) /* Throw away all D-cached data in specified region without any obligation to From da986b9ffff79224417b69cf43506192bd9c29dc Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Tue, 28 Oct 2008 13:58:15 +0800 Subject: [PATCH 007/160] Blackfin arch: dont warn when running a kernel on the oldest supported silicon Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu --- arch/blackfin/kernel/setup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/blackfin/kernel/setup.c b/arch/blackfin/kernel/setup.c index 8337dc3a62d..71a9a8c53ce 100644 --- a/arch/blackfin/kernel/setup.c +++ b/arch/blackfin/kernel/setup.c @@ -827,7 +827,7 @@ void __init setup_arch(char **cmdline_p) printk(KERN_ERR "Warning: Compiled for Rev %d, but running on Rev %d\n", bfin_compiled_revid(), bfin_revid()); } - if (bfin_revid() <= CONFIG_BF_REV_MIN || bfin_revid() > CONFIG_BF_REV_MAX) + if (bfin_revid() < CONFIG_BF_REV_MIN || bfin_revid() > CONFIG_BF_REV_MAX) printk(KERN_ERR "Warning: Unsupported Chip Revision ADSP-%s Rev 0.%d detected\n", CPU, bfin_revid()); } From a10101d5ff9e34c0a1a526725474eef84409529a Mon Sep 17 00:00:00 2001 From: Michael Hennerich Date: Tue, 28 Oct 2008 14:18:29 +0800 Subject: [PATCH 008/160] Blackfin arch: fix bug - Cpufreq assumes clocks in kHz and not Hz. Signed-off-by: Michael Hennerich Signed-off-by: Bryan Wu --- arch/blackfin/mach-common/cpufreq.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/arch/blackfin/mach-common/cpufreq.c b/arch/blackfin/mach-common/cpufreq.c index c22c47b6012..dda5443b37e 100644 --- a/arch/blackfin/mach-common/cpufreq.c +++ b/arch/blackfin/mach-common/cpufreq.c @@ -72,13 +72,13 @@ unsigned int __bfin_cycles_mod; /**************************************************************************/ -static unsigned int bfin_getfreq(unsigned int cpu) +static unsigned int bfin_getfreq_khz(unsigned int cpu) { /* The driver only support single cpu */ if (cpu != 0) return -1; - return get_cclk(); + return get_cclk() / 1000; } @@ -96,7 +96,7 @@ static int bfin_target(struct cpufreq_policy *policy, cclk_hz = bfin_freq_table[index].frequency; - freqs.old = bfin_getfreq(0); + freqs.old = bfin_getfreq_khz(0); freqs.new = cclk_hz; freqs.cpu = 0; @@ -137,8 +137,8 @@ static int __init __bfin_cpu_init(struct cpufreq_policy *policy) if (policy->cpu != 0) return -EINVAL; - cclk = get_cclk(); - sclk = get_sclk(); + cclk = get_cclk() / 1000; + sclk = get_sclk() / 1000; #if ANOMALY_05000273 || (!defined(CONFIG_BF54x) && defined(CONFIG_BFIN_DCACHE)) min_cclk = sclk * 2; @@ -152,7 +152,7 @@ static int __init __bfin_cpu_init(struct cpufreq_policy *policy) dpm_state_table[index].csel = csel << 4; /* Shift now into PLL_DIV bitpos */ dpm_state_table[index].tscale = (TIME_SCALE / (1 << csel)) - 1; - pr_debug("cpufreq: freq:%d csel:%d tscale:%d\n", + pr_debug("cpufreq: freq:%d csel:0x%x tscale:%d\n", bfin_freq_table[index].frequency, dpm_state_table[index].csel, dpm_state_table[index].tscale); @@ -173,7 +173,7 @@ static struct freq_attr *bfin_freq_attr[] = { static struct cpufreq_driver bfin_driver = { .verify = bfin_verify_speed, .target = bfin_target, - .get = bfin_getfreq, + .get = bfin_getfreq_khz, .init = __bfin_cpu_init, .name = "bfin cpufreq", .owner = THIS_MODULE, From 72edff8dd45fdee6e1a2bc431baefd8a5372f7cb Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Tue, 28 Oct 2008 15:42:13 +0800 Subject: [PATCH 009/160] Blackfin arch: fix incorrect limit check for bf54x check_gpio Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu --- arch/blackfin/kernel/bfin_gpio.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/blackfin/kernel/bfin_gpio.c b/arch/blackfin/kernel/bfin_gpio.c index 6e08f425bb4..5c0800adb4d 100644 --- a/arch/blackfin/kernel/bfin_gpio.c +++ b/arch/blackfin/kernel/bfin_gpio.c @@ -218,7 +218,7 @@ inline int check_gpio(unsigned gpio) if (gpio == GPIO_PB15 || gpio == GPIO_PC14 || gpio == GPIO_PC15 || gpio == GPIO_PH14 || gpio == GPIO_PH15 || gpio == GPIO_PJ14 || gpio == GPIO_PJ15 - || gpio > MAX_BLACKFIN_GPIOS) + || gpio >= MAX_BLACKFIN_GPIOS) return -EINVAL; return 0; } From b2c2f30388c682520ae2d07c8852b4225dd4a4d7 Mon Sep 17 00:00:00 2001 From: Jie Zhang Date: Tue, 28 Oct 2008 15:57:49 +0800 Subject: [PATCH 010/160] Blackfin arch: fix bug - shared lib function in L2 failed be called Allow user space to access L2 SRAM. Signed-off-by: Jie Zhang Signed-off-by: Bryan Wu --- arch/blackfin/kernel/process.c | 7 ++++++- arch/blackfin/mm/sram-alloc.c | 8 ++++---- 2 files changed, 10 insertions(+), 5 deletions(-) diff --git a/arch/blackfin/kernel/process.c b/arch/blackfin/kernel/process.c index 77800dd83e5..0c3ea118b65 100644 --- a/arch/blackfin/kernel/process.c +++ b/arch/blackfin/kernel/process.c @@ -351,9 +351,14 @@ int _access_ok(unsigned long addr, unsigned long size) return 1; #endif #if L1_DATA_B_LENGTH != 0 - if (addr >= L1_DATA_B_START + if (addr >= L1_DATA_B_START + (_ebss_b_l1 - _sdata_b_l1) && addr + size <= L1_DATA_B_START + L1_DATA_B_LENGTH) return 1; +#endif +#if L2_LENGTH != 0 + if (addr >= L2_START + (_ebss_l2 - _stext_l2) + && addr + size <= L2_START + L2_LENGTH) + return 1; #endif return 0; } diff --git a/arch/blackfin/mm/sram-alloc.c b/arch/blackfin/mm/sram-alloc.c index 0f1ca6930c1..cc6f336e731 100644 --- a/arch/blackfin/mm/sram-alloc.c +++ b/arch/blackfin/mm/sram-alloc.c @@ -183,10 +183,10 @@ static void __init l2_sram_init(void) return; } - free_l2_sram_head.next->paddr = (void *)L2_START + - (_etext_l2 - _stext_l2) + (_edata_l2 - _sdata_l2); - free_l2_sram_head.next->size = L2_LENGTH - - (_etext_l2 - _stext_l2) + (_edata_l2 - _sdata_l2); + free_l2_sram_head.next->paddr = + (void *)L2_START + (_ebss_l2 - _stext_l2); + free_l2_sram_head.next->size = + L2_LENGTH - (_ebss_l2 - _stext_l2); free_l2_sram_head.next->pid = 0; free_l2_sram_head.next->next = NULL; From 3044646148cdfa83a311bf1c146a70e550280159 Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Sun, 9 Nov 2008 10:07:58 -0800 Subject: [PATCH 011/160] x86: move iomap.h to the new include location a new file was accidentally added to include/asm-x86; move it to the new arch/x86/include/asm location Signed-off-by: Arjan van de Ven --- {include/asm-x86 => arch/x86/include/asm}/iomap.h | 0 1 file changed, 0 insertions(+), 0 deletions(-) rename {include/asm-x86 => arch/x86/include/asm}/iomap.h (100%) diff --git a/include/asm-x86/iomap.h b/arch/x86/include/asm/iomap.h similarity index 100% rename from include/asm-x86/iomap.h rename to arch/x86/include/asm/iomap.h From 6cd10f8db385ba547811baa5b26f672fdff232e6 Mon Sep 17 00:00:00 2001 From: James Bottomley Date: Sun, 9 Nov 2008 11:53:14 -0600 Subject: [PATCH 012/160] x86, voyager: fix smp generic helper voyager breakage Impact: build/boot fix for x86/Voyager This change: | commit 3d4422332711ef48ef0f132f1fcbfcbd56c7f3d1 | Author: Jens Axboe | Date: Thu Jun 26 11:21:34 2008 +0200 | | Add generic helpers for arch IPI function calls didn't wire up the voyager smp call function correctly, so do that here. Also make CONFIG_USE_GENERIC_SMP_HELPERS a def_bool y again, since we now use the generic helpers for every x86 architecture. Signed-off-by: James Bottomley Cc: Jens Axboe Signed-off-by: Ingo Molnar --- arch/x86/Kconfig | 5 ++++- arch/x86/mach-voyager/voyager_smp.c | 16 ++++++++++++++-- 2 files changed, 18 insertions(+), 3 deletions(-) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 4cf0ab13d18..ac22bb7719f 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -167,9 +167,12 @@ config GENERIC_PENDING_IRQ config X86_SMP bool depends on SMP && ((X86_32 && !X86_VOYAGER) || X86_64) - select USE_GENERIC_SMP_HELPERS default y +config USE_GENERIC_SMP_HELPERS + def_bool y + depends on SMP + config X86_32_SMP def_bool y depends on X86_32 && SMP diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c index 0e331652681..52145007bd7 100644 --- a/arch/x86/mach-voyager/voyager_smp.c +++ b/arch/x86/mach-voyager/voyager_smp.c @@ -7,6 +7,7 @@ * This file provides all the same external entries as smp.c but uses * the voyager hal to provide the functionality */ +#include #include #include #include @@ -1790,6 +1791,17 @@ void __init smp_setup_processor_id(void) x86_write_percpu(cpu_number, hard_smp_processor_id()); } +static void voyager_send_call_func(cpumask_t callmask) +{ + __u32 mask = cpus_addr(callmask)[0] & ~(1 << smp_processor_id()); + send_CPI(mask, VIC_CALL_FUNCTION_CPI); +} + +static void voyager_send_call_func_single(int cpu) +{ + send_CPI(1 << cpu, VIC_CALL_FUNCTION_SINGLE_CPI); +} + struct smp_ops smp_ops = { .smp_prepare_boot_cpu = voyager_smp_prepare_boot_cpu, .smp_prepare_cpus = voyager_smp_prepare_cpus, @@ -1799,6 +1811,6 @@ struct smp_ops smp_ops = { .smp_send_stop = voyager_smp_send_stop, .smp_send_reschedule = voyager_smp_send_reschedule, - .send_call_func_ipi = native_send_call_func_ipi, - .send_call_func_single_ipi = native_send_call_func_single_ipi, + .send_call_func_ipi = voyager_send_call_func, + .send_call_func_single_ipi = voyager_send_call_func_single, }; From a358324466b171e145df20bdb74fe81759906de6 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Tue, 11 Nov 2008 15:01:42 -0500 Subject: [PATCH 013/160] ring-buffer: buffer record on/off switch Impact: enable/disable ring buffer recording API added Several kernel developers have requested that there be a way to stop recording into the ring buffers with a simple switch that can also be enabled from userspace. This patch addes a new kernel API to the ring buffers called: tracing_on() tracing_off() When tracing_off() is called, all ring buffers will not be able to record into their buffers. tracing_on() will enable the ring buffers again. These two act like an on/off switch. That is, there is no counting of the number of times tracing_off or tracing_on has been called. A new file is added to the debugfs/tracing directory called tracing_on This allows for userspace applications to also flip the switch. echo 0 > debugfs/tracing/tracing_on disables the tracing. echo 1 > /debugfs/tracing/tracing_on enables it. Note, this does not disable or enable any tracers. It only sets or clears a flag that needs to be set in order for the ring buffers to write to their buffers. It is a global flag, and affects all ring buffers. The buffers start out with tracing_on enabled. There are now three flags that control recording into the buffers: tracing_on: which affects all ring buffer tracers. buffer->record_disabled: which affects an allocated buffer, which may be set if an anomaly is detected, and tracing is disabled. cpu_buffer->record_disabled: which is set by tracing_stop() or if an anomaly is detected. tracing_start can not reenable this if an anomaly occurred. The userspace debugfs/tracing/tracing_enabled is implemented with tracing_stop() but the user space code can not enable it if the kernel called tracing_stop(). Userspace can enable the tracing_on even if the kernel disabled it. It is just a switch used to stop tracing if a condition was hit. tracing_on is not for protecting critical areas in the kernel nor is it for stopping tracing if an anomaly occurred. This is because userspace can reenable it at any time. Side effect: With this patch, I discovered a dead variable in ftrace.c called tracing_on. This patch removes it. Signed-off-by: Steven Rostedt --- include/linux/ring_buffer.h | 3 ++ kernel/trace/ftrace.c | 8 +-- kernel/trace/ring_buffer.c | 101 ++++++++++++++++++++++++++++++++++++ 3 files changed, 106 insertions(+), 6 deletions(-) diff --git a/include/linux/ring_buffer.h b/include/linux/ring_buffer.h index 536b0ca46a0..e097c2e6b6d 100644 --- a/include/linux/ring_buffer.h +++ b/include/linux/ring_buffer.h @@ -120,6 +120,9 @@ unsigned long ring_buffer_overruns(struct ring_buffer *buffer); u64 ring_buffer_time_stamp(int cpu); void ring_buffer_normalize_time_stamp(int cpu, u64 *ts); +void tracing_on(void); +void tracing_off(void); + enum ring_buffer_flags { RB_FL_OVERWRITE = 1 << 0, }; diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 4a39d24568c..14fa52297b2 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -185,7 +185,6 @@ enum { }; static int ftrace_filtered; -static int tracing_on; static LIST_HEAD(ftrace_new_addrs); @@ -506,13 +505,10 @@ static int __ftrace_modify_code(void *data) { int *command = data; - if (*command & FTRACE_ENABLE_CALLS) { + if (*command & FTRACE_ENABLE_CALLS) ftrace_replace_code(1); - tracing_on = 1; - } else if (*command & FTRACE_DISABLE_CALLS) { + else if (*command & FTRACE_DISABLE_CALLS) ftrace_replace_code(0); - tracing_on = 0; - } if (*command & FTRACE_UPDATE_TRACE_FUNC) ftrace_update_ftrace_func(ftrace_trace_function); diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 2f76193c348..b08ee9f00c8 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -16,6 +16,35 @@ #include #include +#include "trace.h" + +/* Global flag to disable all recording to ring buffers */ +static int ring_buffers_off __read_mostly; + +/** + * tracing_on - enable all tracing buffers + * + * This function enables all tracing buffers that may have been + * disabled with tracing_off. + */ +void tracing_on(void) +{ + ring_buffers_off = 0; +} + +/** + * tracing_off - turn off all tracing buffers + * + * This function stops all tracing buffers from recording data. + * It does not disable any overhead the tracers themselves may + * be causing. This function simply causes all recording to + * the ring buffers to fail. + */ +void tracing_off(void) +{ + ring_buffers_off = 1; +} + /* Up this if you want to test the TIME_EXTENTS and normalization */ #define DEBUG_SHIFT 0 @@ -1133,6 +1162,9 @@ ring_buffer_lock_reserve(struct ring_buffer *buffer, struct ring_buffer_event *event; int cpu, resched; + if (ring_buffers_off) + return NULL; + if (atomic_read(&buffer->record_disabled)) return NULL; @@ -1249,6 +1281,9 @@ int ring_buffer_write(struct ring_buffer *buffer, int ret = -EBUSY; int cpu, resched; + if (ring_buffers_off) + return -EBUSY; + if (atomic_read(&buffer->record_disabled)) return -EBUSY; @@ -2070,3 +2105,69 @@ int ring_buffer_swap_cpu(struct ring_buffer *buffer_a, return 0; } +static ssize_t +rb_simple_read(struct file *filp, char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + int *p = filp->private_data; + char buf[64]; + int r; + + /* !ring_buffers_off == tracing_on */ + r = sprintf(buf, "%d\n", !*p); + + return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); +} + +static ssize_t +rb_simple_write(struct file *filp, const char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + int *p = filp->private_data; + char buf[64]; + long val; + int ret; + + if (cnt >= sizeof(buf)) + return -EINVAL; + + if (copy_from_user(&buf, ubuf, cnt)) + return -EFAULT; + + buf[cnt] = 0; + + ret = strict_strtoul(buf, 10, &val); + if (ret < 0) + return ret; + + /* !ring_buffers_off == tracing_on */ + *p = !val; + + (*ppos)++; + + return cnt; +} + +static struct file_operations rb_simple_fops = { + .open = tracing_open_generic, + .read = rb_simple_read, + .write = rb_simple_write, +}; + + +static __init int rb_init_debugfs(void) +{ + struct dentry *d_tracer; + struct dentry *entry; + + d_tracer = tracing_init_dentry(); + + entry = debugfs_create_file("tracing_on", 0644, d_tracer, + &ring_buffers_off, &rb_simple_fops); + if (!entry) + pr_warning("Could not create debugfs 'tracing_on' entry\n"); + + return 0; +} + +fs_initcall(rb_init_debugfs); From 47e74f2ba8fbf9fb1378e2524e6cfdc2fb37f160 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Wed, 12 Nov 2008 00:01:27 -0500 Subject: [PATCH 014/160] ring-buffer: no preempt for sched_clock() Impact: disable preemption when calling sched_clock() The ring_buffer_time_stamp still uses sched_clock as its counter. But it is a bug to call it with preemption enabled. This requirement should not be pushed to the ring_buffer_time_stamp callers, so the ring_buffer_time_stamp needs to disable preemption when calling sched_clock. Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/trace/ring_buffer.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index b08ee9f00c8..231db209fa8 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -51,8 +51,14 @@ void tracing_off(void) /* FIXME!!! */ u64 ring_buffer_time_stamp(int cpu) { + u64 time; + + preempt_disable_notrace(); /* shift to debug/test normalization and TIME_EXTENTS */ - return sched_clock() << DEBUG_SHIFT; + time = sched_clock() << DEBUG_SHIFT; + preempt_enable_notrace(); + + return time; } void ring_buffer_normalize_time_stamp(int cpu, u64 *ts) From 97a70e548bd97d5a46ae9d44f24aafcc013fd701 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 12 Nov 2008 23:22:35 +0100 Subject: [PATCH 015/160] x86, hibernate: fix breakage on x86_32 with CONFIG_NUMA set Impact: fix crash during hibernation on 32-bit NUMA The NUMA code on x86_32 creates special memory mapping that allows each node's pgdat to be located in this node's memory. For this purpose it allocates a memory area at the end of each node's memory and maps this area so that it is accessible with virtual addresses belonging to low memory. As a result, if there is high memory, these NUMA-allocated areas are physically located in high memory, although they are mapped to low memory addresses. Our hibernation code does not take that into account and for this reason hibernation fails on all x86_32 systems with CONFIG_NUMA=y and with high memory present. Fix this by adding a special mapping for the NUMA-allocated memory areas to the temporary page tables created during the last phase of resume. Signed-off-by: Rafael J. Wysocki Signed-off-by: Ingo Molnar --- arch/x86/include/asm/mmzone_32.h | 4 ++++ arch/x86/mm/numa_32.c | 35 ++++++++++++++++++++++++++++++++ arch/x86/power/hibernate_32.c | 4 ++++ 3 files changed, 43 insertions(+) diff --git a/arch/x86/include/asm/mmzone_32.h b/arch/x86/include/asm/mmzone_32.h index 485bdf059ff..07f1af494ca 100644 --- a/arch/x86/include/asm/mmzone_32.h +++ b/arch/x86/include/asm/mmzone_32.h @@ -34,10 +34,14 @@ static inline void get_memcfg_numa(void) extern int early_pfn_to_nid(unsigned long pfn); +extern void resume_map_numa_kva(pgd_t *pgd); + #else /* !CONFIG_NUMA */ #define get_memcfg_numa get_memcfg_numa_flat +static inline void resume_map_numa_kva(pgd_t *pgd) {} + #endif /* CONFIG_NUMA */ #ifdef CONFIG_DISCONTIGMEM diff --git a/arch/x86/mm/numa_32.c b/arch/x86/mm/numa_32.c index 847c164725f..8518c678d83 100644 --- a/arch/x86/mm/numa_32.c +++ b/arch/x86/mm/numa_32.c @@ -222,6 +222,41 @@ static void __init remap_numa_kva(void) } } +#ifdef CONFIG_HIBERNATION +/** + * resume_map_numa_kva - add KVA mapping to the temporary page tables created + * during resume from hibernation + * @pgd_base - temporary resume page directory + */ +void resume_map_numa_kva(pgd_t *pgd_base) +{ + int node; + + for_each_online_node(node) { + unsigned long start_va, start_pfn, size, pfn; + + start_va = (unsigned long)node_remap_start_vaddr[node]; + start_pfn = node_remap_start_pfn[node]; + size = node_remap_size[node]; + + printk(KERN_DEBUG "%s: node %d\n", __FUNCTION__, node); + + for (pfn = 0; pfn < size; pfn += PTRS_PER_PTE) { + unsigned long vaddr = start_va + (pfn << PAGE_SHIFT); + pgd_t *pgd = pgd_base + pgd_index(vaddr); + pud_t *pud = pud_offset(pgd, vaddr); + pmd_t *pmd = pmd_offset(pud, vaddr); + + set_pmd(pmd, pfn_pmd(start_pfn + pfn, + PAGE_KERNEL_LARGE_EXEC)); + + printk(KERN_DEBUG "%s: %08lx -> pfn %08lx\n", + __FUNCTION__, vaddr, start_pfn + pfn); + } + } +} +#endif + static unsigned long calculate_numa_remap_pages(void) { int nid; diff --git a/arch/x86/power/hibernate_32.c b/arch/x86/power/hibernate_32.c index f2b6e3f11bf..81197c62d5b 100644 --- a/arch/x86/power/hibernate_32.c +++ b/arch/x86/power/hibernate_32.c @@ -12,6 +12,7 @@ #include #include #include +#include /* Defined in hibernate_asm_32.S */ extern int restore_image(void); @@ -127,6 +128,9 @@ static int resume_physical_mapping_init(pgd_t *pgd_base) } } } + + resume_map_numa_kva(pgd_base); + return 0; } From 604d20554883cf03f888440d58ea7c6d36899839 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 12 Nov 2008 23:26:14 +0100 Subject: [PATCH 016/160] x86: make NUMA on 32-bit depend on EXPERIMENTAL again My previous patch to make CONFIG_NUMA on x86_32 depend on BROKEN turned out to be unnecessary, after all, since the source of the hibernation vs CONFIG_NUMA problem turned out to be the fact that we didn't take the NUMA KVA remapping into account in the hibernation code. Signed-off-by: Rafael J. Wysocki Signed-off-by: Ingo Molnar --- arch/x86/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 93224b56918..4cf0ab13d18 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -957,7 +957,7 @@ config ARCH_PHYS_ADDR_T_64BIT config NUMA bool "Numa Memory Allocation and Scheduler Support (EXPERIMENTAL)" depends on SMP - depends on X86_64 || (X86_32 && HIGHMEM64G && (X86_NUMAQ || X86_BIGSMP || X86_SUMMIT && ACPI) && BROKEN) + depends on X86_64 || (X86_32 && HIGHMEM64G && (X86_NUMAQ || X86_BIGSMP || X86_SUMMIT && ACPI) && EXPERIMENTAL) default n if X86_PC default y if (X86_NUMAQ || X86_SUMMIT || X86_BIGSMP) help From 2fe401e38602e853e01376cdb670b0bc4d526a6d Mon Sep 17 00:00:00 2001 From: Adrian Knoth Date: Wed, 12 Nov 2008 16:23:55 -0800 Subject: [PATCH 017/160] sched: correct sched-rt-group.txt pathname in init/Kconfig init/Kconfig directs the user to Documentation/sched-rt-group.txt, but the file is actually in Documentation/scheduler/sched-rt-group.txt. This patch corrects the pathname mentioned in init/Kconfig. Signed-off-by: Adrian Knoth Signed-off-by: Andrew Morton Signed-off-by: Ingo Molnar --- init/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/init/Kconfig b/init/Kconfig index 86b00c53fad..2f850d800d9 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -354,7 +354,7 @@ config RT_GROUP_SCHED setting below. If enabled, it will also make it impossible to schedule realtime tasks for non-root users until you allocate realtime bandwidth for them. - See Documentation/sched-rt-group.txt for more information. + See Documentation/scheduler/sched-rt-group.txt for more information. choice depends on GROUP_SCHED From 4d4e9bb339cfcde7811af10859ba1ce2fe3d46b4 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Wed, 12 Nov 2008 16:45:04 +0100 Subject: [PATCH 018/160] ALSA: hda - Add digital beep playback switch for STAC/IDT codecs The digital beep widget may have no mute control, and always enabling the beep is ofen pretty annoying, especially on laptops. This patch adds a mixer control "PC Beep Playback Switch" when there is no mixer amp mute is found, and controls it on software. Reference: Novell bnc#444572 https://bugzilla.novell.com/show_bug.cgi?id=444572 Signed-off-by: Takashi Iwai --- sound/pci/hda/hda_beep.c | 4 ++ sound/pci/hda/hda_beep.h | 1 + sound/pci/hda/patch_sigmatel.c | 69 +++++++++++++++++++++++++++++++--- 3 files changed, 68 insertions(+), 6 deletions(-) diff --git a/sound/pci/hda/hda_beep.c b/sound/pci/hda/hda_beep.c index 9b77b3e0fa9..b1796ae1e8f 100644 --- a/sound/pci/hda/hda_beep.c +++ b/sound/pci/hda/hda_beep.c @@ -37,6 +37,9 @@ static void snd_hda_generate_beep(struct work_struct *work) container_of(work, struct hda_beep, beep_work); struct hda_codec *codec = beep->codec; + if (!beep->enabled) + return; + /* generate tone */ snd_hda_codec_write_cache(codec, beep->nid, 0, AC_VERB_SET_BEEP_CONTROL, beep->tone); @@ -115,6 +118,7 @@ int snd_hda_attach_beep_device(struct hda_codec *codec, int nid) beep->nid = nid; beep->dev = input_dev; beep->codec = codec; + beep->enabled = 1; codec->beep = beep; INIT_WORK(&beep->beep_work, &snd_hda_generate_beep); diff --git a/sound/pci/hda/hda_beep.h b/sound/pci/hda/hda_beep.h index de4036e6e71..b9679f081ca 100644 --- a/sound/pci/hda/hda_beep.h +++ b/sound/pci/hda/hda_beep.h @@ -31,6 +31,7 @@ struct hda_beep { char phys[32]; int tone; int nid; + int enabled; struct work_struct beep_work; /* scheduled task for beep event */ }; diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c index 4300a679cd8..1633ef2c654 100644 --- a/sound/pci/hda/patch_sigmatel.c +++ b/sound/pci/hda/patch_sigmatel.c @@ -2587,8 +2587,10 @@ static struct snd_kcontrol_new stac92xx_control_templates[] = { }; /* add dynamic controls */ -static int stac92xx_add_control_idx(struct sigmatel_spec *spec, int type, - int idx, const char *name, unsigned long val) +static int stac92xx_add_control_temp(struct sigmatel_spec *spec, + struct snd_kcontrol_new *ktemp, + int idx, const char *name, + unsigned long val) { struct snd_kcontrol_new *knew; @@ -2607,20 +2609,29 @@ static int stac92xx_add_control_idx(struct sigmatel_spec *spec, int type, } knew = &spec->kctl_alloc[spec->num_kctl_used]; - *knew = stac92xx_control_templates[type]; + *knew = *ktemp; knew->index = idx; knew->name = kstrdup(name, GFP_KERNEL); - if (! knew->name) + if (!knew->name) return -ENOMEM; knew->private_value = val; spec->num_kctl_used++; return 0; } +static inline int stac92xx_add_control_idx(struct sigmatel_spec *spec, + int type, int idx, const char *name, + unsigned long val) +{ + return stac92xx_add_control_temp(spec, + &stac92xx_control_templates[type], + idx, name, val); +} + /* add dynamic controls */ -static int stac92xx_add_control(struct sigmatel_spec *spec, int type, - const char *name, unsigned long val) +static inline int stac92xx_add_control(struct sigmatel_spec *spec, int type, + const char *name, unsigned long val) { return stac92xx_add_control_idx(spec, type, 0, name, val); } @@ -3062,6 +3073,43 @@ static int stac92xx_auto_create_beep_ctls(struct hda_codec *codec, return 0; } +#ifdef CONFIG_SND_HDA_INPUT_BEEP +#define stac92xx_dig_beep_switch_info snd_ctl_boolean_mono_info + +static int stac92xx_dig_beep_switch_get(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + ucontrol->value.integer.value[0] = codec->beep->enabled; + return 0; +} + +static int stac92xx_dig_beep_switch_put(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + int enabled = !!ucontrol->value.integer.value[0]; + if (codec->beep->enabled != enabled) { + codec->beep->enabled = enabled; + return 1; + } + return 0; +} + +static struct snd_kcontrol_new stac92xx_dig_beep_ctrl = { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .info = stac92xx_dig_beep_switch_info, + .get = stac92xx_dig_beep_switch_get, + .put = stac92xx_dig_beep_switch_put, +}; + +static int stac92xx_beep_switch_ctl(struct hda_codec *codec) +{ + return stac92xx_add_control_temp(codec->spec, &stac92xx_dig_beep_ctrl, + 0, "PC Beep Playback Switch", 0); +} +#endif + static int stac92xx_auto_create_mux_input_ctls(struct hda_codec *codec) { struct sigmatel_spec *spec = codec->spec; @@ -3368,6 +3416,7 @@ static int stac92xx_parse_auto_config(struct hda_codec *codec, hda_nid_t dig_out #ifdef CONFIG_SND_HDA_INPUT_BEEP if (spec->digbeep_nid > 0) { hda_nid_t nid = spec->digbeep_nid; + unsigned int caps; err = stac92xx_auto_create_beep_ctls(codec, nid); if (err < 0) @@ -3375,6 +3424,14 @@ static int stac92xx_parse_auto_config(struct hda_codec *codec, hda_nid_t dig_out err = snd_hda_attach_beep_device(codec, nid); if (err < 0) return err; + /* if no beep switch is available, make its own one */ + caps = query_amp_caps(codec, nid, HDA_OUTPUT); + if (codec->beep && + !((caps & AC_AMPCAP_MUTE) >> AC_AMPCAP_MUTE_SHIFT)) { + err = stac92xx_beep_switch_ctl(codec); + if (err < 0) + return err; + } } #endif From 6a12afb5640217454cad21e8cf9a74b038fa3717 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Thu, 13 Nov 2008 13:08:56 +0100 Subject: [PATCH 019/160] ALSA: hda - Missing NULL check in hda_beep.c Added a NULL check of input_allocate_device() in hda_beep.c. Signed-off-by: Takashi Iwai --- sound/pci/hda/hda_beep.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/sound/pci/hda/hda_beep.c b/sound/pci/hda/hda_beep.c index b1796ae1e8f..3ecd7e797de 100644 --- a/sound/pci/hda/hda_beep.c +++ b/sound/pci/hda/hda_beep.c @@ -88,6 +88,10 @@ int snd_hda_attach_beep_device(struct hda_codec *codec, int nid) snprintf(beep->phys, sizeof(beep->phys), "card%d/codec#%d/beep0", codec->bus->card->number, codec->addr); input_dev = input_allocate_device(); + if (!input_dev) { + kfree(beep); + return -ENOMEM; + } /* setup digital beep device */ input_dev->name = "HDA Digital PCBeep"; From ee51a1de7e3837577412be269e0100038068e691 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 13 Nov 2008 14:58:31 +0100 Subject: [PATCH 020/160] tracing: fix mmiotrace resizing crash Pekka reported a crash when resizing the mmiotrace tracer (if only mmiotrace is enabled). This happens because in that case we do not allocate the max buffer, but we try to use it. Make ring_buffer_resize() idempotent against NULL buffers. Reported-by: Pekka Paalanen Signed-off-by: Ingo Molnar --- kernel/trace/ring_buffer.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 231db209fa8..036456cbb4f 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -538,6 +538,12 @@ int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size) LIST_HEAD(pages); int i, cpu; + /* + * Always succeed at resizing a non-existent buffer: + */ + if (!buffer) + return size; + size = DIV_ROUND_UP(size, BUF_PAGE_SIZE); size *= BUF_PAGE_SIZE; buffer_size = buffer->pages * BUF_PAGE_SIZE; From c39555d6edd0142972f577405d5259879ecc1635 Mon Sep 17 00:00:00 2001 From: Matthew Ranostay Date: Fri, 14 Nov 2008 17:46:22 -0500 Subject: [PATCH 021/160] ALSA: hda: STAC_VREF_EVENT value change Changed value for STAC_VREF_EVENT from 0x40 to 0x00 because the unsol response value is only 6-bits width and the former value was 1<<6 which is an overrun. Signed-off-by: Matthew Ranostay Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_sigmatel.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c index 1633ef2c654..f205570def1 100644 --- a/sound/pci/hda/patch_sigmatel.c +++ b/sound/pci/hda/patch_sigmatel.c @@ -36,9 +36,11 @@ #include "hda_beep.h" #define NUM_CONTROL_ALLOC 32 + +#define STAC_VREF_EVENT 0x00 +#define STAC_INSERT_EVENT 0x10 #define STAC_PWR_EVENT 0x20 #define STAC_HP_EVENT 0x30 -#define STAC_VREF_EVENT 0x40 enum { STAC_REF, From d6f35e3f0da14a5eb02114708d88ae2f59dc8283 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Fri, 14 Nov 2008 19:08:18 +0100 Subject: [PATCH 022/160] ALSA: sound/pci/pcxhr/pcxhr.c: introduce missing kfree and pci_disable_device Error handling code following a kzalloc should free the allocated data. The error handling code is adjusted to call pci_disable_device(pci); as well, as done later in the function The semantic match that finds the problem is as follows: (http://www.emn.fr/x-info/coccinelle/) // @r exists@ local idexpression x; statement S; expression E; identifier f,l; position p1,p2; expression *ptr != NULL; @@ ( if ((x@p1 = \(kmalloc\|kzalloc\|kcalloc\)(...)) == NULL) S | x@p1 = \(kmalloc\|kzalloc\|kcalloc\)(...); ... if (x == NULL) S ) <... when != x when != if (...) { <+...x...+> } x->f = E ...> ( return \(0\|<+...x...+>\|ptr\); | return@p2 ...; ) @script:python@ p1 << r.p1; p2 << r.p2; @@ print "* file: %s kmalloc %s return %s" % (p1[0].file,p1[0].line,p2[0].line) // Signed-off-by: Julia Lawall Signed-off-by: Takashi Iwai --- sound/pci/pcxhr/pcxhr.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/sound/pci/pcxhr/pcxhr.c b/sound/pci/pcxhr/pcxhr.c index 0e06c6c9fcc..73de6e989b3 100644 --- a/sound/pci/pcxhr/pcxhr.c +++ b/sound/pci/pcxhr/pcxhr.c @@ -1229,8 +1229,11 @@ static int __devinit pcxhr_probe(struct pci_dev *pci, const struct pci_device_id return -ENOMEM; } - if (snd_BUG_ON(pci_id->driver_data >= PCI_ID_LAST)) + if (snd_BUG_ON(pci_id->driver_data >= PCI_ID_LAST)) { + kfree(mgr); + pci_disable_device(pci); return -ENODEV; + } card_name = pcxhr_board_params[pci_id->driver_data].board_name; mgr->playback_chips = pcxhr_board_params[pci_id->driver_data].playback_chips; mgr->capture_chips = pcxhr_board_params[pci_id->driver_data].capture_chips; From 8e5f262bfcd90c041160a491a238661ebbb584a4 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Sat, 15 Nov 2008 19:28:54 +0100 Subject: [PATCH 023/160] ALSA: hda - Check model type instead of SSID in patch_92hd71bxx() Check board preset model instead of codec->subsystem_id in patch_92hd71bxx() so that other hardwares configured via the model option work like the given model. Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_sigmatel.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c index f205570def1..0df6f979f2a 100644 --- a/sound/pci/hda/patch_sigmatel.c +++ b/sound/pci/hda/patch_sigmatel.c @@ -4488,8 +4488,8 @@ again: codec->slave_dig_outs = stac92hd71bxx_slave_dig_outs; break; case 0x111d7608: /* 5 Port with Analog Mixer */ - switch (codec->subsystem_id) { - case 0x103c361a: + switch (spec->board_config) { + case STAC_HP_M4: /* Enable VREF power saving on GPIO1 detect */ snd_hda_codec_write(codec, codec->afg, 0, AC_VERB_SET_GPIO_UNSOLICITED_RSP_MASK, 0x02); From 29d7b90c15035741d15421b36000509212b3e135 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Sun, 16 Nov 2008 08:07:15 +0100 Subject: [PATCH 024/160] sched: fix kernel warning on /proc/sched_debug access Luis Henriques reported that with CONFIG_PREEMPT=y + CONFIG_PREEMPT_DEBUG=y + CONFIG_SCHED_DEBUG=y + CONFIG_LATENCYTOP=y enabled, the following warning triggers when using latencytop: > [ 775.663239] BUG: using smp_processor_id() in preemptible [00000000] code: latencytop/6585 > [ 775.663303] caller is native_sched_clock+0x3a/0x80 > [ 775.663314] Pid: 6585, comm: latencytop Tainted: G W 2.6.28-rc4-00355-g9c7c354 #1 > [ 775.663322] Call Trace: > [ 775.663343] [] debug_smp_processor_id+0xe4/0xf0 > [ 775.663356] [] native_sched_clock+0x3a/0x80 > [ 775.663368] [] sched_clock+0x9/0x10 > [ 775.663381] [] proc_sched_show_task+0x8bd/0x10e0 > [ 775.663395] [] sched_show+0x3e/0x80 > [ 775.663408] [] seq_read+0xdb/0x350 > [ 775.663421] [] ? security_file_permission+0x16/0x20 > [ 775.663435] [] vfs_read+0xc8/0x170 > [ 775.663447] [] sys_read+0x55/0x90 > [ 775.663460] [] system_call_fastpath+0x16/0x1b > ... This breakage was caused by me via: 7cbaef9: sched: optimize sched_clock() a bit Change the calls to cpu_clock(). Reported-by: Luis Henriques --- kernel/sched_debug.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c index 48ecc51e770..26ed8e3d1c1 100644 --- a/kernel/sched_debug.c +++ b/kernel/sched_debug.c @@ -423,10 +423,11 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m) #undef __P { + unsigned int this_cpu = raw_smp_processor_id(); u64 t0, t1; - t0 = sched_clock(); - t1 = sched_clock(); + t0 = cpu_clock(this_cpu); + t1 = cpu_clock(this_cpu); SEQ_printf(m, "%-35s:%21Ld\n", "clock-delta", (long long)(t1-t0)); } From d1f1e9c01006b4b050e090055c75278f80c2a5c5 Mon Sep 17 00:00:00 2001 From: Markus Metzger Date: Sat, 15 Nov 2008 11:00:17 +0100 Subject: [PATCH 025/160] x86, bts: fix unlock problem in ds.c Fix a problem where ds_request() returned an error without releasing the ds lock. Reported-by: Stephane Eranian Signed-off-by: Markus Metzger Cc: Signed-off-by: Ingo Molnar --- arch/x86/kernel/ds.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c index 2b69994fd3a..ac1d5b0586b 100644 --- a/arch/x86/kernel/ds.c +++ b/arch/x86/kernel/ds.c @@ -384,8 +384,9 @@ static int ds_request(struct task_struct *task, void *base, size_t size, spin_lock(&ds_lock); + error = -EPERM; if (!check_tracer(task)) - return -EPERM; + goto out_unlock; error = -ENOMEM; context = ds_alloc_context(task); From e47411b1f4456480d6c60ebdc7a733e81ccd5d66 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Sun, 16 Nov 2008 08:27:53 +0100 Subject: [PATCH 026/160] MAINTAINERS: remove me as RAID maintainer Neil has been the maintainer of the RAID/MD code for a long time, remove me as a co-maintainer. Signed-off-by: Ingo Molnar --- MAINTAINERS | 2 -- 1 file changed, 2 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 8e0777fae3c..627e4c89328 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3928,8 +3928,6 @@ M: bootc@bootc.net S: Maintained SOFTWARE RAID (Multiple Disks) SUPPORT -P: Ingo Molnar -M: mingo@redhat.com P: Neil Brown M: neilb@suse.de L: linux-raid@vger.kernel.org From 5821e1b74f0d08952cb5da4bfd2d9a388d8df58e Mon Sep 17 00:00:00 2001 From: walimis Date: Sat, 15 Nov 2008 15:19:06 +0800 Subject: [PATCH 027/160] function tracing: fix wrong pos computing when read buffer has been fulfilled Impact: make output of available_filter_functions complete phenomenon: The first value of dyn_ftrace_total_info is not equal with `cat available_filter_functions | wc -l`, but they should be equal. root cause: When printing functions with seq_printf in t_show, if the read buffer is just overflowed by current function record, then this function won't be printed to user space through read buffer, it will just be dropped. So we can't see this function printing. So, every time the last function to fill the read buffer, if overflowed, will be dropped. This also applies to set_ftrace_filter if set_ftrace_filter has more bytes than read buffer. fix: Through checking return value of seq_printf, if less than 0, we know this function doesn't be printed. Then we decrease position to force this function to be printed next time, in next read buffer. Another little fix is to show correct allocating pages count. Signed-off-by: walimis Acked-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/trace/ftrace.c | 26 +++++++++++++++----------- 1 file changed, 15 insertions(+), 11 deletions(-) diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 14fa52297b2..e60205722d0 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -673,7 +673,7 @@ static int __init ftrace_dyn_table_alloc(unsigned long num_to_init) cnt = num_to_init / ENTRIES_PER_PAGE; pr_info("ftrace: allocating %ld entries in %d pages\n", - num_to_init, cnt); + num_to_init, cnt + 1); for (i = 0; i < cnt; i++) { pg->next = (void *)get_zeroed_page(GFP_KERNEL); @@ -753,13 +753,11 @@ static void *t_start(struct seq_file *m, loff_t *pos) void *p = NULL; loff_t l = -1; - if (*pos != iter->pos) { - for (p = t_next(m, p, &l); p && l < *pos; p = t_next(m, p, &l)) - ; - } else { - l = *pos; - p = t_next(m, p, &l); - } + if (*pos > iter->pos) + *pos = iter->pos; + + l = *pos; + p = t_next(m, p, &l); return p; } @@ -770,15 +768,21 @@ static void t_stop(struct seq_file *m, void *p) static int t_show(struct seq_file *m, void *v) { + struct ftrace_iterator *iter = m->private; struct dyn_ftrace *rec = v; char str[KSYM_SYMBOL_LEN]; + int ret = 0; if (!rec) return 0; kallsyms_lookup(rec->ip, NULL, NULL, NULL, str); - seq_printf(m, "%s\n", str); + ret = seq_printf(m, "%s\n", str); + if (ret < 0) { + iter->pos--; + iter->idx--; + } return 0; } @@ -804,7 +808,7 @@ ftrace_avail_open(struct inode *inode, struct file *file) return -ENOMEM; iter->pg = ftrace_pages_start; - iter->pos = -1; + iter->pos = 0; ret = seq_open(file, &show_ftrace_seq_ops); if (!ret) { @@ -891,7 +895,7 @@ ftrace_regex_open(struct inode *inode, struct file *file, int enable) if (file->f_mode & FMODE_READ) { iter->pg = ftrace_pages_start; - iter->pos = -1; + iter->pos = 0; iter->flags = enable ? FTRACE_ITER_FILTER : FTRACE_ITER_NOTRACE; From d3c6aa1e69f705ac3ab64584101b1d38435b1353 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Sun, 16 Nov 2008 00:49:31 -0800 Subject: [PATCH 028/160] x86: fix es7000 compiling Impact: fix es7000 build CC arch/x86/kernel/es7000_32.o arch/x86/kernel/es7000_32.c: In function find_unisys_acpi_oem_table: arch/x86/kernel/es7000_32.c:255: error: implicit declaration of function acpi_get_table_with_size arch/x86/kernel/es7000_32.c:261: error: implicit declaration of function early_acpi_os_unmap_memory arch/x86/kernel/es7000_32.c: In function unmap_unisys_acpi_oem_table: arch/x86/kernel/es7000_32.c:277: error: implicit declaration of function __acpi_unmap_table make[1]: *** [arch/x86/kernel/es7000_32.o] Error 1 we applied one patch out of order... | commit a73aaedd95703bd49f4c3f9df06fb7b7373ba905 | Author: Yinghai Lu | Date: Sun Sep 14 02:33:14 2008 -0700 | | x86: check dsdt before find oem table for es7000, v2 | | v2: use __acpi_unmap_table() that patch need: x86: use early_ioremap in __acpi_map_table x86: always explicitly map acpi memory acpi: remove final __acpi_map_table mapping before setting acpi_gbl_permanent_mmap acpi/x86: introduce __apci_map_table, v4 submitted to the ACPI tree but not upstream yet. fix it until those patches applied, need to revert this one Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/kernel/es7000_32.c | 9 +-------- 1 file changed, 1 insertion(+), 8 deletions(-) diff --git a/arch/x86/kernel/es7000_32.c b/arch/x86/kernel/es7000_32.c index f454c78fcef..0aa2c443d60 100644 --- a/arch/x86/kernel/es7000_32.c +++ b/arch/x86/kernel/es7000_32.c @@ -250,31 +250,24 @@ int __init find_unisys_acpi_oem_table(unsigned long *oem_addr) { struct acpi_table_header *header = NULL; int i = 0; - acpi_size tbl_size; - while (ACPI_SUCCESS(acpi_get_table_with_size("OEM1", i++, &header, &tbl_size))) { + while (ACPI_SUCCESS(acpi_get_table("OEM1", i++, &header))) { if (!memcmp((char *) &header->oem_id, "UNISYS", 6)) { struct oem_table *t = (struct oem_table *)header; oem_addrX = t->OEMTableAddr; oem_size = t->OEMTableSize; - early_acpi_os_unmap_memory(header, tbl_size); *oem_addr = (unsigned long)__acpi_map_table(oem_addrX, oem_size); return 0; } - early_acpi_os_unmap_memory(header, tbl_size); } return -1; } void __init unmap_unisys_acpi_oem_table(unsigned long oem_addr) { - if (!oem_addr) - return; - - __acpi_unmap_table((char *)oem_addr, oem_size); } #endif From 1e74f3000b86969de421ca0da08f42e7d21cbd99 Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Mon, 17 Nov 2008 16:24:34 +0900 Subject: [PATCH 029/160] swiotlb: use coherent_dma_mask in alloc_coherent Impact: fix DMA buffer allocation coherency bug in certain configs This patch fixes swiotlb to use dev->coherent_dma_mask in swiotlb_alloc_coherent(). coherent_dma_mask is a subset of dma_mask (equal to it most of the time), enumerating the address range that a given device is able to DMA to/from in a cache-coherent way. But currently, swiotlb uses dev->dma_mask in alloc_coherent() implicitly via address_needs_mapping(), but alloc_coherent is really supposed to use coherent_dma_mask. This bug could break drivers that uses smaller coherent_dma_mask than dma_mask (though the current code works for the majority that use the same mask for coherent_dma_mask and dma_mask). Signed-off-by: FUJITA Tomonori Cc: tony.luck@intel.com Signed-off-by: Ingo Molnar --- lib/swiotlb.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/lib/swiotlb.c b/lib/swiotlb.c index 78330c37a61..5f6c629a924 100644 --- a/lib/swiotlb.c +++ b/lib/swiotlb.c @@ -467,9 +467,13 @@ swiotlb_alloc_coherent(struct device *hwdev, size_t size, dma_addr_t dev_addr; void *ret; int order = get_order(size); + u64 dma_mask = DMA_32BIT_MASK; + + if (hwdev && hwdev->coherent_dma_mask) + dma_mask = hwdev->coherent_dma_mask; ret = (void *)__get_free_pages(flags, order); - if (ret && address_needs_mapping(hwdev, virt_to_bus(ret), size)) { + if (ret && !is_buffer_dma_capable(dma_mask, virt_to_bus(ret), size)) { /* * The allocated memory isn't reachable by the device. * Fall back on swiotlb_map_single(). @@ -493,9 +497,9 @@ swiotlb_alloc_coherent(struct device *hwdev, size_t size, dev_addr = virt_to_bus(ret); /* Confirm address can be DMA'd by device */ - if (address_needs_mapping(hwdev, dev_addr, size)) { + if (!is_buffer_dma_capable(dma_mask, dev_addr, size)) { printk("hwdev DMA mask = 0x%016Lx, dev_addr = 0x%016Lx\n", - (unsigned long long)*hwdev->dma_mask, + (unsigned long long)dma_mask, (unsigned long long)dev_addr); /* DMA_TO_DEVICE to avoid memcpy in unmap_single */ From ad133ba3dc283300e5b62b5b7211d2f39fbf6ee7 Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Mon, 17 Nov 2008 15:39:47 +0100 Subject: [PATCH 030/160] sched, signals: fix the racy usage of ->signal in account_group_xxx/run_posix_cpu_timers Impact: fix potential NULL dereference Contrary to ad474caca3e2a0550b7ce0706527ad5ab389a4d4 changelog, other acct_group_xxx() helpers can be called after exit_notify() by timer tick. Thanks to Roland for pointing out this. Somehow I missed this simple fact when I read the original patch, and I am afraid I confused Frank during the discussion. Sorry. Fortunately, these helpers work with current, we can check ->exit_state to ensure that ->signal can't go away under us. Also, add the comment and compiler barrier to account_group_exec_runtime(), to make sure we load ->signal only once. Signed-off-by: Oleg Nesterov Signed-off-by: Ingo Molnar --- kernel/posix-cpu-timers.c | 7 +++++-- kernel/sched_stats.h | 15 +++++++++++---- 2 files changed, 16 insertions(+), 6 deletions(-) diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index 153dcb2639c..895337b16a2 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -1308,9 +1308,10 @@ static inline int task_cputime_expired(const struct task_cputime *sample, */ static inline int fastpath_timer_check(struct task_struct *tsk) { - struct signal_struct *sig = tsk->signal; + struct signal_struct *sig; - if (unlikely(!sig)) + /* tsk == current, ensure it is safe to use ->signal/sighand */ + if (unlikely(tsk->exit_state)) return 0; if (!task_cputime_zero(&tsk->cputime_expires)) { @@ -1323,6 +1324,8 @@ static inline int fastpath_timer_check(struct task_struct *tsk) if (task_cputime_expired(&task_sample, &tsk->cputime_expires)) return 1; } + + sig = tsk->signal; if (!task_cputime_zero(&sig->cputime_expires)) { struct task_cputime group_sample; diff --git a/kernel/sched_stats.h b/kernel/sched_stats.h index ee71bec1da6..7dbf72a2b02 100644 --- a/kernel/sched_stats.h +++ b/kernel/sched_stats.h @@ -298,9 +298,11 @@ static inline void account_group_user_time(struct task_struct *tsk, { struct signal_struct *sig; - sig = tsk->signal; - if (unlikely(!sig)) + /* tsk == current, ensure it is safe to use ->signal */ + if (unlikely(tsk->exit_state)) return; + + sig = tsk->signal; if (sig->cputime.totals) { struct task_cputime *times; @@ -325,9 +327,11 @@ static inline void account_group_system_time(struct task_struct *tsk, { struct signal_struct *sig; - sig = tsk->signal; - if (unlikely(!sig)) + /* tsk == current, ensure it is safe to use ->signal */ + if (unlikely(tsk->exit_state)) return; + + sig = tsk->signal; if (sig->cputime.totals) { struct task_cputime *times; @@ -353,8 +357,11 @@ static inline void account_group_exec_runtime(struct task_struct *tsk, struct signal_struct *sig; sig = tsk->signal; + /* see __exit_signal()->task_rq_unlock_wait() */ + barrier(); if (unlikely(!sig)) return; + if (sig->cputime.totals) { struct task_cputime *times; From a4a16beadea041ab601e65b264b568e8b6b4f68d Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 10 Nov 2008 09:05:37 +0100 Subject: [PATCH 031/160] oprofile: fix an overflow in ppro code reset_value was changed from long to u64 in commit b99170288421c79f0c2efa8b33e26e65f4bb7fb8 (oprofile: Implement Intel architectural perfmon support) But dynamic allocation of this array use a wrong type (long instead of u64) Cc: Andi Kleen Signed-off-by: Eric Dumazet Signed-off-by: Robert Richter --- arch/x86/oprofile/op_model_ppro.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c index 3f1b81a83e2..716d26f0e5d 100644 --- a/arch/x86/oprofile/op_model_ppro.c +++ b/arch/x86/oprofile/op_model_ppro.c @@ -69,7 +69,7 @@ static void ppro_setup_ctrs(struct op_msrs const * const msrs) int i; if (!reset_value) { - reset_value = kmalloc(sizeof(unsigned) * num_counters, + reset_value = kmalloc(sizeof(reset_value[0]) * num_counters, GFP_ATOMIC); if (!reset_value) return; From 26a3e99160cfb06a0a33e25b9fb0d516e2cc680d Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Mon, 17 Nov 2008 09:35:00 -0800 Subject: [PATCH 032/160] xen: fix scrub_page() Impact: fix guest kernel crash with CONFIG_XEN_SCRUB_PAGES=y Jens noticed that scrub_page() has a buggy unmap of the wrong thing. (virtual address instead of page) Linus pointed out that the whole scrub_page() code is an unnecessary reimplementation of clear_highpage() to begin with. Just use clear_highpage() rather than reimplementing it poorly. Reported-by: Jens Axboe Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar --- drivers/xen/balloon.c | 9 +-------- 1 file changed, 1 insertion(+), 8 deletions(-) diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c index a0fb5eac407..526c191e84e 100644 --- a/drivers/xen/balloon.c +++ b/drivers/xen/balloon.c @@ -122,14 +122,7 @@ static struct timer_list balloon_timer; static void scrub_page(struct page *page) { #ifdef CONFIG_XEN_SCRUB_PAGES - if (PageHighMem(page)) { - void *v = kmap(page); - clear_page(v); - kunmap(v); - } else { - void *v = page_address(page); - clear_page(v); - } + clear_highpage(page); #endif } From 93ce99e849433ede4ce8b410b749dc0cad1100b2 Mon Sep 17 00:00:00 2001 From: Venki Pallipadi Date: Mon, 17 Nov 2008 14:43:58 -0800 Subject: [PATCH 033/160] x86: add rdtsc barrier to TSC sync check Impact: fix incorrectly marked unstable TSC clock Patch (commit 0d12cdd "sched: improve sched_clock() performance") has a regression on one of the test systems here. With the patch, I see: checking TSC synchronization [CPU#0 -> CPU#1]: Measured 28 cycles TSC warp between CPUs, turning off TSC clock. Marking TSC unstable due to check_tsc_sync_source failed Whereas, without the patch syncs pass fine on all CPUs: checking TSC synchronization [CPU#0 -> CPU#1]: passed. Due to this, TSC is marked unstable, when it is not actually unstable. This is because syncs in check_tsc_wrap() goes away due to this commit. As per the discussion on this thread, correct way to fix this is to add explicit syncs as below? Signed-off-by: Venkatesh Pallipadi Signed-off-by: Ingo Molnar --- arch/x86/kernel/tsc_sync.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/x86/kernel/tsc_sync.c b/arch/x86/kernel/tsc_sync.c index 9ffb01c31c4..1c0dfbca87c 100644 --- a/arch/x86/kernel/tsc_sync.c +++ b/arch/x86/kernel/tsc_sync.c @@ -46,7 +46,9 @@ static __cpuinit void check_tsc_warp(void) cycles_t start, now, prev, end; int i; + rdtsc_barrier(); start = get_cycles(); + rdtsc_barrier(); /* * The measurement runs for 20 msecs: */ @@ -61,7 +63,9 @@ static __cpuinit void check_tsc_warp(void) */ __raw_spin_lock(&sync_lock); prev = last_tsc; + rdtsc_barrier(); now = get_cycles(); + rdtsc_barrier(); last_tsc = now; __raw_spin_unlock(&sync_lock); From 700018e0a77b4113172257fcdaa1c58e27a5074f Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Tue, 18 Nov 2008 14:02:03 +0800 Subject: [PATCH 034/160] cpuset: fix regression when failed to generate sched domains Impact: properly rebuild sched-domains on kmalloc() failure When cpuset failed to generate sched domains due to kmalloc() failure, the scheduler should fallback to the single partition 'fallback_doms' and rebuild sched domains, but now it only destroys but not rebuilds sched domains. The regression was introduced by: | commit dfb512ec4834116124da61d6c1ee10fd0aa32bd6 | Author: Max Krasnyansky | Date: Fri Aug 29 13:11:41 2008 -0700 | | sched: arch_reinit_sched_domains() must destroy domains to force rebuild After the above commit, partition_sched_domains(0, NULL, NULL) will only destroy sched domains and partition_sched_domains(1, NULL, NULL) will create the default sched domain. Signed-off-by: Li Zefan Cc: Max Krasnyansky Cc: Signed-off-by: Ingo Molnar --- kernel/cpuset.c | 12 ++++++++---- kernel/sched.c | 13 +++++++------ 2 files changed, 15 insertions(+), 10 deletions(-) diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 3e00526f52e..81fc6791a29 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -587,7 +587,6 @@ static int generate_sched_domains(cpumask_t **domains, int ndoms; /* number of sched domains in result */ int nslot; /* next empty doms[] cpumask_t slot */ - ndoms = 0; doms = NULL; dattr = NULL; csa = NULL; @@ -674,10 +673,8 @@ restart: * Convert to and populate cpu masks. */ doms = kmalloc(ndoms * sizeof(cpumask_t), GFP_KERNEL); - if (!doms) { - ndoms = 0; + if (!doms) goto done; - } /* * The rest of the code, including the scheduler, can deal with @@ -732,6 +729,13 @@ restart: done: kfree(csa); + /* + * Fallback to the default domain if kmalloc() failed. + * See comments in partition_sched_domains(). + */ + if (doms == NULL) + ndoms = 1; + *domains = doms; *attributes = dattr; return ndoms; diff --git a/kernel/sched.c b/kernel/sched.c index c94baf2969e..9b1e79371c2 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -7789,13 +7789,14 @@ static int dattrs_equal(struct sched_domain_attr *cur, int idx_cur, * * The passed in 'doms_new' should be kmalloc'd. This routine takes * ownership of it and will kfree it when done with it. If the caller - * failed the kmalloc call, then it can pass in doms_new == NULL, - * and partition_sched_domains() will fallback to the single partition - * 'fallback_doms', it also forces the domains to be rebuilt. + * failed the kmalloc call, then it can pass in doms_new == NULL && + * ndoms_new == 1, and partition_sched_domains() will fallback to + * the single partition 'fallback_doms', it also forces the domains + * to be rebuilt. * - * If doms_new==NULL it will be replaced with cpu_online_map. - * ndoms_new==0 is a special case for destroying existing domains. - * It will not create the default domain. + * If doms_new == NULL it will be replaced with cpu_online_map. + * ndoms_new == 0 is a special case for destroying existing domains, + * and it will not create the default domain. * * Call with hotplug lock held */ From e270219f4372b58bd3eeac12bd9f7edc592b8f6b Mon Sep 17 00:00:00 2001 From: Rakib Mullick Date: Tue, 18 Nov 2008 10:15:24 +0600 Subject: [PATCH 035/160] kernel/profile.c: fix section mismatch warning Impact: fix section mismatch warning in kernel/profile.c Here, profile_nop function has been called from a non-init function create_hash_tables(void). Which generetes a section mismatch warning. Previously, create_hash_tables(void) was a init function. So, removing __init from create_hash_tables(void) requires profile_nop to be non-init. This patch makes profile_nop function inline and fixes the following warning: WARNING: vmlinux.o(.text+0x6ebb6): Section mismatch in reference from the function create_hash_tables() to the function .init.text:profile_nop() The function create_hash_tables() references the function __init profile_nop(). This is often because create_hash_tables lacks a __init annotation or the annotation of profile_nop is wrong. Signed-off-by: Rakib Mullick Signed-off-by: Ingo Molnar --- kernel/profile.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/profile.c b/kernel/profile.c index 9830a037d8d..5b7d1ac7124 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -544,7 +544,7 @@ static const struct file_operations proc_profile_operations = { }; #ifdef CONFIG_SMP -static void __init profile_nop(void *unused) +static inline void profile_nop(void *unused) { } From 41c3b648bd4cdc34fd1918e288f8afe78903432e Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Tue, 18 Nov 2008 10:45:15 +0100 Subject: [PATCH 036/160] ALSA: hda - Fix GPIO initialization in patch_stac92hd71bxx() Fixed the GPIO mask and co initialization in patch_stac92hd71bxx() so that the gpio_maks for HP_M4 model is set properly. Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_sigmatel.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c index 0df6f979f2a..b254e90b47c 100644 --- a/sound/pci/hda/patch_sigmatel.c +++ b/sound/pci/hda/patch_sigmatel.c @@ -4478,6 +4478,13 @@ again: stac92xx_set_config_regs(codec); } + if (spec->board_config > STAC_92HD71BXX_REF) { + /* GPIO0 = EAPD */ + spec->gpio_mask = 0x01; + spec->gpio_dir = 0x01; + spec->gpio_data = 0x01; + } + switch (codec->vendor_id) { case 0x111d76b6: /* 4 Port without Analog Mixer */ case 0x111d76b7: @@ -4537,13 +4544,6 @@ again: spec->aloopback_mask = 0x50; spec->aloopback_shift = 0; - if (spec->board_config > STAC_92HD71BXX_REF) { - /* GPIO0 = EAPD */ - spec->gpio_mask = 0x01; - spec->gpio_dir = 0x01; - spec->gpio_data = 0x01; - } - spec->powerdown_adcs = 1; spec->digbeep_nid = 0x26; spec->mux_nids = stac92hd71bxx_mux_nids; From 4213cb64004e38b3e78424f30e1e638f8004c7a8 Mon Sep 17 00:00:00 2001 From: Graf Yang Date: Tue, 18 Nov 2008 17:48:22 +0800 Subject: [PATCH 037/160] Blackfin arch: fix bug - Turn on DEBUG_DOUBLEFAULT, booting SMP kernel crash Signed-off-by: Graf Yang Signed-off-by: Bryan Wu --- arch/blackfin/mach-common/entry.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/blackfin/mach-common/entry.S b/arch/blackfin/mach-common/entry.S index c13fa8da28c..bde6dc4e261 100644 --- a/arch/blackfin/mach-common/entry.S +++ b/arch/blackfin/mach-common/entry.S @@ -277,7 +277,7 @@ ENTRY(_bfin_return_from_exception) p5.h = hi(ILAT); r6 = [p5]; r7 = 0x20; /* Did I just cause anther HW error? */ - r7 = r7 & r1; + r6 = r7 & r6; CC = R7 == R6; if CC JUMP _double_fault; #endif From 62273eeb6ac516ab0abf49417378726ad8875b03 Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Tue, 18 Nov 2008 17:48:22 +0800 Subject: [PATCH 038/160] Blackfin arch: fix a broken define in dma-mapping dma_mapping_error is an actual function, so fix broken define with a real inline stub Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu --- arch/blackfin/include/asm/dma-mapping.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/blackfin/include/asm/dma-mapping.h b/arch/blackfin/include/asm/dma-mapping.h index ede748d67ef..d7d9148e433 100644 --- a/arch/blackfin/include/asm/dma-mapping.h +++ b/arch/blackfin/include/asm/dma-mapping.h @@ -15,7 +15,11 @@ void dma_free_coherent(struct device *dev, size_t size, void *vaddr, #define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f) #define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h) -#define dma_mapping_error +static inline +int dma_mapping_error(struct device *dev, dma_addr_t dma_addr) +{ + return 0; +} /* * Map a single buffer of the indicated size for DMA in streaming mode. From 80bf272468a8b63f5550304363e55c1ba06c13db Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Tue, 18 Nov 2008 10:48:41 +0100 Subject: [PATCH 039/160] ALSA: hda - Add quirks for HP Pavilion DV models Added the quirk entries for HP Pavilion DV5 and DV7 with model=hp-m4. Reference: Novell bnc#445321, bnc#445161 https://bugzilla.novell.com/show_bug.cgi?id=445321 https://bugzilla.novell.com/show_bug.cgi?id=445161 Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_sigmatel.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c index b254e90b47c..25d4bf8e422 100644 --- a/sound/pci/hda/patch_sigmatel.c +++ b/sound/pci/hda/patch_sigmatel.c @@ -1688,6 +1688,10 @@ static struct snd_pci_quirk stac92hd71bxx_cfg_tbl[] = { /* SigmaTel reference board */ SND_PCI_QUIRK(PCI_VENDOR_ID_INTEL, 0x2668, "DFI LanParty", STAC_92HD71BXX_REF), + SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x30f2, + "HP dv5", STAC_HP_M4), + SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x30f4, + "HP dv7", STAC_HP_M4), SND_PCI_QUIRK(PCI_VENDOR_ID_HP, 0x361a, "unknown HP", STAC_HP_M4), SND_PCI_QUIRK(PCI_VENDOR_ID_DELL, 0x0233, From c5d08bb567874e410210cf7d21a25f28fcceb0a7 Mon Sep 17 00:00:00 2001 From: Takashi Iwai Date: Tue, 18 Nov 2008 10:55:36 +0100 Subject: [PATCH 040/160] ALSA: hda - Fix resume of GPIO unsol event for STAC/IDT Use cached write for setting the GPIO unsolicited event mask to be restored properly at resume. Signed-off-by: Takashi Iwai --- sound/pci/hda/patch_sigmatel.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c index 25d4bf8e422..9563b5bbb27 100644 --- a/sound/pci/hda/patch_sigmatel.c +++ b/sound/pci/hda/patch_sigmatel.c @@ -4502,7 +4502,7 @@ again: switch (spec->board_config) { case STAC_HP_M4: /* Enable VREF power saving on GPIO1 detect */ - snd_hda_codec_write(codec, codec->afg, 0, + snd_hda_codec_write_cache(codec, codec->afg, 0, AC_VERB_SET_GPIO_UNSOLICITED_RSP_MASK, 0x02); snd_hda_codec_write_cache(codec, codec->afg, 0, AC_VERB_SET_UNSOLICITED_ENABLE, @@ -4895,7 +4895,7 @@ static int patch_stac9205(struct hda_codec *codec) stac92xx_set_config_reg(codec, 0x20, 0x1c410030); /* Enable unsol response for GPIO4/Dock HP connection */ - snd_hda_codec_write(codec, codec->afg, 0, + snd_hda_codec_write_cache(codec, codec->afg, 0, AC_VERB_SET_GPIO_UNSOLICITED_RSP_MASK, 0x10); snd_hda_codec_write_cache(codec, codec->afg, 0, AC_VERB_SET_UNSOLICITED_ENABLE, From eb60fa1066622ddb2278732cf61e0c4544e82c6f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 10 Nov 2008 15:28:59 +0900 Subject: [PATCH 041/160] block: fix add_partition() error path Partition stats structure was not freed on devt allocation failure path. Fix it. Signed-off-by: Tejun Heo Signed-off-by: Jens Axboe --- fs/partitions/check.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/fs/partitions/check.c b/fs/partitions/check.c index 633f7a0ebb2..90bcf136a9d 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -395,7 +395,7 @@ int add_partition(struct gendisk *disk, int partno, err = blk_alloc_devt(p, &devt); if (err) - goto out_free; + goto out_free_stats; pdev->devt = devt; /* delay uevent until 'holders' subdir is created */ @@ -426,6 +426,8 @@ int add_partition(struct gendisk *disk, int partno, return 0; +out_free_stats: + free_part_stats(p); out_free: kfree(p); return err; From ba32929a91fe2c0628f5be62d1597b379c8d3062 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 10 Nov 2008 15:29:58 +0900 Subject: [PATCH 042/160] block: make add_partition() return pointer to hd_struct Make add_partition() return pointer to the new hd_struct on success and ERR_PTR() value on failure. This change will be used to fix md autodetection bug. Signed-off-by: Tejun Heo Cc: Neil Brown Signed-off-by: Jens Axboe --- block/ioctl.c | 7 +++---- fs/partitions/check.c | 25 +++++++++++++------------ include/linux/genhd.h | 4 +++- 3 files changed, 19 insertions(+), 17 deletions(-) diff --git a/block/ioctl.c b/block/ioctl.c index c832d639b6e..d03985b04d6 100644 --- a/block/ioctl.c +++ b/block/ioctl.c @@ -18,7 +18,6 @@ static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user struct disk_part_iter piter; long long start, length; int partno; - int err; if (!capable(CAP_SYS_ADMIN)) return -EACCES; @@ -61,10 +60,10 @@ static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user disk_part_iter_exit(&piter); /* all seems OK */ - err = add_partition(disk, partno, start, length, - ADDPART_FLAG_NONE); + part = add_partition(disk, partno, start, length, + ADDPART_FLAG_NONE); mutex_unlock(&bdev->bd_mutex); - return err; + return IS_ERR(part) ? PTR_ERR(part) : 0; case BLKPG_DEL_PARTITION: part = disk_get_part(disk, partno); if (!part) diff --git a/fs/partitions/check.c b/fs/partitions/check.c index 90bcf136a9d..63302534023 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -348,8 +348,8 @@ static ssize_t whole_disk_show(struct device *dev, static DEVICE_ATTR(whole_disk, S_IRUSR | S_IRGRP | S_IROTH, whole_disk_show, NULL); -int add_partition(struct gendisk *disk, int partno, - sector_t start, sector_t len, int flags) +struct hd_struct *add_partition(struct gendisk *disk, int partno, + sector_t start, sector_t len, int flags) { struct hd_struct *p; dev_t devt = MKDEV(0, 0); @@ -361,15 +361,15 @@ int add_partition(struct gendisk *disk, int partno, err = disk_expand_part_tbl(disk, partno); if (err) - return err; + return ERR_PTR(err); ptbl = disk->part_tbl; if (ptbl->part[partno]) - return -EBUSY; + return ERR_PTR(-EBUSY); p = kzalloc(sizeof(*p), GFP_KERNEL); if (!p) - return -ENOMEM; + return ERR_PTR(-EBUSY); if (!init_part_stats(p)) { err = -ENOMEM; @@ -424,20 +424,20 @@ int add_partition(struct gendisk *disk, int partno, if (!ddev->uevent_suppress) kobject_uevent(&pdev->kobj, KOBJ_ADD); - return 0; + return p; out_free_stats: free_part_stats(p); out_free: kfree(p); - return err; + return ERR_PTR(err); out_del: kobject_put(p->holder_dir); device_del(pdev); out_put: put_device(pdev); blk_free_devt(devt); - return err; + return ERR_PTR(err); } /* Not exported, helper to add_disk(). */ @@ -568,10 +568,11 @@ int rescan_partitions(struct gendisk *disk, struct block_device *bdev) disk->disk_name, p, (unsigned long long) size); size = get_capacity(disk) - from; } - res = add_partition(disk, p, from, size, state->parts[p].flags); - if (res) { - printk(KERN_ERR " %s: p%d could not be added: %d\n", - disk->disk_name, p, -res); + part = add_partition(disk, p, from, size, + state->parts[p].flags); + if (IS_ERR(part)) { + printk(KERN_ERR " %s: p%d could not be added: %ld\n", + disk->disk_name, p, -PTR_ERR(part)); continue; } #ifdef CONFIG_BLK_DEV_MD diff --git a/include/linux/genhd.h b/include/linux/genhd.h index e439e6aed83..3df7742ce24 100644 --- a/include/linux/genhd.h +++ b/include/linux/genhd.h @@ -522,7 +522,9 @@ extern char *disk_name (struct gendisk *hd, int partno, char *buf); extern int disk_expand_part_tbl(struct gendisk *disk, int target); extern int rescan_partitions(struct gendisk *disk, struct block_device *bdev); -extern int __must_check add_partition(struct gendisk *, int, sector_t, sector_t, int); +extern struct hd_struct * __must_check add_partition(struct gendisk *disk, + int partno, sector_t start, + sector_t len, int flags); extern void delete_partition(struct gendisk *, int); extern void printk_all_partitions(void); From 55e8e30c382d25c34f8aafcc78efec948571a941 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 10 Nov 2008 15:30:47 +0900 Subject: [PATCH 043/160] block/md: fix md autodetection Block ext devt conversion missed md_autodetect_dev() call in rescan_partitions() leaving md autodetect unable to see partitions. Fix it. Signed-off-by: Tejun Heo Cc: Neil Brown Signed-off-by: Jens Axboe --- fs/partitions/check.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/partitions/check.c b/fs/partitions/check.c index 63302534023..6d5b213b8a9 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -577,7 +577,7 @@ int rescan_partitions(struct gendisk *disk, struct block_device *bdev) } #ifdef CONFIG_BLK_DEV_MD if (state->parts[p].flags & ADDPART_FLAG_RAID) - md_autodetect_dev(bdev->bd_dev+p); + md_autodetect_dev(part_to_dev(part)->devt); #endif } kfree(state); From 561ec68e4de7947167937c49c451728e6b19e63b Mon Sep 17 00:00:00 2001 From: "Zhang, Yanmin" Date: Fri, 14 Nov 2008 08:26:30 +0100 Subject: [PATCH 044/160] block: fix boot failure with CONFIG_DEBUG_BLOCK_EXT_DEVT=y and nash We run into system boot failure with kernel 2.6.28-rc. We found it on a couple of machines, including T61 notebook, nehalem machine, and another HPC NX6325 notebook. All the machines use FedoraCore 8 or FedoraCore 9. With kernel prior to 2.6.28-rc, system boot doesn't fail. I debug it and locate the root cause. Pls. see http://bugzilla.kernel.org/show_bug.cgi?id=11899 https://bugzilla.redhat.com/show_bug.cgi?id=471517 As a matter of fact, there are 2 bugs. 1)root=/dev/sda1, system boot randomly fails. Mostly, boot for 5 times and fails once. nash has a bug. Some of its functions misuse return value 0. Sometimes, 0 means timeout and no uevent available. Sometimes, 0 means nash gets an uevent, but the uevent isn't block-related (for exmaple, usb). If by coincidence, kernel tells nash that uevents are available, but kernel also set timeout, nash might stops collecting other uevents in queue if current uevent isn't block-related. I work out a patch for nash to fix it. http://bugzilla.kernel.org/attachment.cgi?id=18858 2) root=LABEL=/, system always can't boot. initrd init reports switchroot fails. Here is an executation branch of nash when booting: (1) nash read /sys/block/sda/dev; Assume major is 8 (on my desktop) (2) nash query /proc/devices with the major number; It found line "8 sd"; (3) nash use 'sd' to search its own probe table to find device (DISK) type for the device and add it to its own list; (4) Later on, it probes all devices in its list to get filesystem labels; scsi register "8 sd" always. When major is 259, nash fails to find the device(DISK) type. I enables CONFIG_DEBUG_BLOCK_EXT_DEVT=y when compiling kernel, so 259 is picked up for device /dev/sda1, which causes nash to fail to find device (DISK) type. To fixing issue 2), I create a patch for nash and another patch for kernel. http://bugzilla.kernel.org/attachment.cgi?id=18859 http://bugzilla.kernel.org/attachment.cgi?id=18837 Below is the patch for kernel 2.6.28-rc4. It registers blkext, a new block device in proc/devices. With 2 patches on nash and 1 patch on kernel, I boot my machines for dozens of times without failure. Signed-off-by Zhang Yanmin Acked-by: Tejun Heo Signed-off-by: Jens Axboe --- block/genhd.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/block/genhd.c b/block/genhd.c index 4e5e7493f67..27549e470da 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -768,6 +768,8 @@ static int __init genhd_device_init(void) bdev_map = kobj_map_init(base_probe, &block_class_lock); blk_dev_init(); + register_blkdev(BLOCK_EXT_MAJOR, "blkext"); + #ifndef CONFIG_SYSFS_DEPRECATED /* create top-level block dir */ block_depr = kobject_create_and_add("block", NULL); From 68aee07f9bad2c830a898cf6d6bfc11ea24efc40 Mon Sep 17 00:00:00 2001 From: Zhaolei Date: Fri, 14 Nov 2008 09:44:33 +0100 Subject: [PATCH 045/160] Release old elevator on change elevator We should release old elevator when change to use a new one. Signed-off-by: Zhao Lei Signed-off-by: Jens Axboe --- drivers/block/xen-blkfront.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index b220c686089..2d19f0cc47f 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -338,12 +338,18 @@ wait: static int xlvbd_init_blk_queue(struct gendisk *gd, u16 sector_size) { struct request_queue *rq; + elevator_t *old_e; rq = blk_init_queue(do_blkif_request, &blkif_io_lock); if (rq == NULL) return -1; - elevator_init(rq, "noop"); + old_e = rq->elevator; + if (IS_ERR_VALUE(elevator_init(rq, "noop"))) + printk(KERN_WARNING + "blkfront: Switch elevator failed, use default\n"); + else + elevator_exit(old_e); /* Hard sector size and max sectors impersonate the equiv. hardware. */ blk_queue_hardsect_size(rq, sector_size); From 98ba4031ab2adc8b394295e68aa4c8fe9d5060db Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Fri, 14 Nov 2008 10:44:59 +0100 Subject: [PATCH 046/160] relay: fix cpu offline problem relay_open() will close allocated buffers when failed. but if cpu offlined, some buffer will not be closed. this patch fixed it. and did cleanup for relay_reset() too. Signed-off-by: Lai Jiangshan Signed-off-by: Jens Axboe --- kernel/relay.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/kernel/relay.c b/kernel/relay.c index 8d13a7855c0..32b0befdcb6 100644 --- a/kernel/relay.c +++ b/kernel/relay.c @@ -400,7 +400,7 @@ void relay_reset(struct rchan *chan) } mutex_lock(&relay_channels_mutex); - for_each_online_cpu(i) + for_each_possible_cpu(i) if (chan->buf[i]) __relay_reset(chan->buf[i], 0); mutex_unlock(&relay_channels_mutex); @@ -611,10 +611,9 @@ struct rchan *relay_open(const char *base_filename, return chan; free_bufs: - for_each_online_cpu(i) { - if (!chan->buf[i]) - break; - relay_close_buf(chan->buf[i]); + for_each_possible_cpu(i) { + if (chan->buf[i]) + relay_close_buf(chan->buf[i]); } kref_put(&chan->kref, relay_destroy_channel); From c26156b2534c75bb3cdedf76f6ad1340971cf5bd Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Tue, 18 Nov 2008 15:07:05 +0100 Subject: [PATCH 047/160] block: hold extra reference to bio in blk_rq_map_user_iov() If the size passed in is OK but we end up mapping too many segments, we call the unmap path directly like from IO completion. But from IO completion we have an extra reference to the bio, so this error case goes OOPS when it attempts to free and already free bio. Fix it by getting an extra reference to the bio before calling the unmap failure case. Reported-by: Petr Vandrovec Signed-off-by: Jens Axboe --- block/blk-map.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/block/blk-map.c b/block/blk-map.c index 4849fa36161..0f4b4b88181 100644 --- a/block/blk-map.c +++ b/block/blk-map.c @@ -217,6 +217,12 @@ int blk_rq_map_user_iov(struct request_queue *q, struct request *rq, return PTR_ERR(bio); if (bio->bi_size != len) { + /* + * Grab an extra reference to this bio, as bio_unmap_user() + * expects to be able to drop it twice as it happens on the + * normal IO completion path + */ + bio_get(bio); bio_endio(bio, 0); bio_unmap_user(bio); return -EINVAL; From 10db4ef7b9a65b86e4d047671a1886f4c101a859 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 18 Nov 2008 15:23:08 +0100 Subject: [PATCH 048/160] x86, PEBS/DS: fix code flow in ds_request() this compiler warning: arch/x86/kernel/ds.c: In function 'ds_request': arch/x86/kernel/ds.c:368: warning: 'context' may be used uninitialized in this function Shows that the code flow in ds_request() is buggy - it goes into the unlock+release-context path even when the context is not allocated yet. First allocate the context, then do the other checks. Also, take care with GFP allocations under the ds_lock spinlock. Cc: Signed-off-by: Ingo Molnar --- arch/x86/kernel/ds.c | 26 +++++++++++++++++++++----- 1 file changed, 21 insertions(+), 5 deletions(-) diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c index ac1d5b0586b..d1a121443bd 100644 --- a/arch/x86/kernel/ds.c +++ b/arch/x86/kernel/ds.c @@ -236,17 +236,33 @@ static inline struct ds_context *ds_alloc_context(struct task_struct *task) struct ds_context *context = *p_context; if (!context) { + spin_unlock(&ds_lock); + context = kzalloc(sizeof(*context), GFP_KERNEL); - if (!context) + if (!context) { + spin_lock(&ds_lock); return NULL; + } context->ds = kzalloc(ds_cfg.sizeof_ds, GFP_KERNEL); if (!context->ds) { kfree(context); + spin_lock(&ds_lock); return NULL; } + spin_lock(&ds_lock); + /* + * Check for race - another CPU could have allocated + * it meanwhile: + */ + if (*p_context) { + kfree(context->ds); + kfree(context); + return *p_context; + } + *p_context = context; context->this = p_context; @@ -384,15 +400,15 @@ static int ds_request(struct task_struct *task, void *base, size_t size, spin_lock(&ds_lock); - error = -EPERM; - if (!check_tracer(task)) - goto out_unlock; - error = -ENOMEM; context = ds_alloc_context(task); if (!context) goto out_unlock; + error = -EPERM; + if (!check_tracer(task)) + goto out_unlock; + error = -EALREADY; if (context->owner[qual] == current) goto out_unlock; From e5e1f606ecbf67e52ebe2df5d14f8b94ec6544d0 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Mon, 17 Nov 2008 15:07:17 +0100 Subject: [PATCH 049/160] AMD IOMMU: add parameter to disable device isolation Impact: add a new AMD IOMMU kernel command line parameter Signed-off-by: Joerg Roedel --- Documentation/kernel-parameters.txt | 2 ++ arch/x86/kernel/amd_iommu_init.c | 2 ++ 2 files changed, 4 insertions(+) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 9fa6508892c..b56ee02d251 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -295,6 +295,8 @@ and is between 256 and 4096 characters. It is defined in the file isolate - enable device isolation (each device, as far as possible, will get its own protection domain) + share - put every device behind one IOMMU into the + same protection domain fullflush - enable flushing of IO/TLB entries when they are unmapped. Otherwise they are flushed before they will be reused, which diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 0cdcda35a05..838a2e1d5bb 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -1213,6 +1213,8 @@ static int __init parse_amd_iommu_options(char *str) for (; *str; ++str) { if (strncmp(str, "isolate", 7) == 0) amd_iommu_isolate = 1; + if (strncmp(str, "share", 5) == 0) + amd_iommu_isolate = 0; if (strncmp(str, "fullflush", 11) == 0) amd_iommu_unmap_flush = true; } From 3ce1f93c6d53c3f91c3846cf66b018276c8ac2e7 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Mon, 17 Nov 2008 15:09:20 +0100 Subject: [PATCH 050/160] AMD IOMMU: enable device isolation per default Impact: makes device isolation the default for AMD IOMMU Some device drivers showed double-free bugs of DMA memory while testing them with AMD IOMMU. If all devices share the same protection domain this can lead to data corruption and data loss. Prevent this by putting each device into its own protection domain per default. Signed-off-by: Joerg Roedel --- Documentation/kernel-parameters.txt | 2 +- arch/x86/kernel/amd_iommu_init.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index b56ee02d251..f2e1e7fc0ff 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -294,7 +294,7 @@ and is between 256 and 4096 characters. It is defined in the file Possible values are: isolate - enable device isolation (each device, as far as possible, will get its own protection - domain) + domain) [default] share - put every device behind one IOMMU into the same protection domain fullflush - enable flushing of IO/TLB entries when diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 838a2e1d5bb..595edd2befc 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -121,7 +121,7 @@ u16 amd_iommu_last_bdf; /* largest PCI device id we have LIST_HEAD(amd_iommu_unity_map); /* a list of required unity mappings we find in ACPI */ unsigned amd_iommu_aperture_order = 26; /* size of aperture in power of 2 */ -int amd_iommu_isolate; /* if 1, device isolation is enabled */ +int amd_iommu_isolate = 1; /* if 1, device isolation is enabled */ bool amd_iommu_unmap_flush; /* if true, flush on every unmap */ LIST_HEAD(amd_iommu_list); /* list of all AMD IOMMUs in the From 695b5676c727d80921a2dc8737d5b3322222db85 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Mon, 17 Nov 2008 15:16:43 +0100 Subject: [PATCH 051/160] AMD IOMMU: fix fullflush comparison length Impact: fix comparison length for 'fullflush' Signed-off-by: Joerg Roedel --- arch/x86/kernel/amd_iommu_init.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 595edd2befc..30ae2701b3d 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -1215,7 +1215,7 @@ static int __init parse_amd_iommu_options(char *str) amd_iommu_isolate = 1; if (strncmp(str, "share", 5) == 0) amd_iommu_isolate = 0; - if (strncmp(str, "fullflush", 11) == 0) + if (strncmp(str, "fullflush", 9) == 0) amd_iommu_unmap_flush = true; } From 8501c45cc32c311ae755a2d5ac8c4a5f04908d42 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Mon, 17 Nov 2008 19:11:46 +0100 Subject: [PATCH 052/160] AMD IOMMU: check for next_bit also in unmapped area Impact: fix possible use of stale IO/TLB entries Signed-off-by: Joerg Roedel --- arch/x86/kernel/amd_iommu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index 331b318304e..e4899e0e878 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -537,7 +537,7 @@ static void dma_ops_free_addresses(struct dma_ops_domain *dom, address >>= PAGE_SHIFT; iommu_area_free(dom->bitmap, address, pages); - if (address + pages >= dom->next_bit) + if (address >= dom->next_bit) dom->need_flush = true; } From 0af40a4b1050c050e62eb1dc30b82d5ab22bf221 Mon Sep 17 00:00:00 2001 From: Philipp Kohlbecher Date: Sun, 16 Nov 2008 12:11:01 +0100 Subject: [PATCH 053/160] x86: more general identifier for Phoenix BIOS Impact: widen the reach of the low-memory-protect DMI quirk Phoenix BIOSes variously identify their vendor as "Phoenix Technologies, LTD" or "Phoenix Technologies LTD" (without the comma.) This patch makes the identification string in the bad_bios_dmi_table more general (following a suggestion by Ingo Molnar), so that both versions are handled. Again, the patched file compiles cleanly and the patch has been tested successfully on my machine. Signed-off-by: Philipp Kohlbecher Cc: Signed-off-by: Ingo Molnar --- arch/x86/kernel/setup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 0fa6790c1dd..9d5674f7b6c 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -764,7 +764,7 @@ static struct dmi_system_id __initdata bad_bios_dmi_table[] = { .callback = dmi_low_memory_corruption, .ident = "Phoenix BIOS", .matches = { - DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix Technologies, LTD"), + DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix Technologies"), }, }, #endif From 0bb943c7a2136716757a263f604d26309fd98042 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Fri, 14 Nov 2008 19:05:31 +0100 Subject: [PATCH 054/160] tracing: kernel/trace/trace.c: introduce missing kfree() Impact: fix memory leak Error handling code following a kzalloc should free the allocated data. The semantic match that finds the problem is as follows: (http://www.emn.fr/x-info/coccinelle/) // @r exists@ local idexpression x; statement S; expression E; identifier f,l; position p1,p2; expression *ptr != NULL; @@ ( if ((x@p1 = \(kmalloc\|kzalloc\|kcalloc\)(...)) == NULL) S | x@p1 = \(kmalloc\|kzalloc\|kcalloc\)(...); ... if (x == NULL) S ) <... when != x when != if (...) { <+...x...+> } x->f = E ...> ( return \(0\|<+...x...+>\|ptr\); | return@p2 ...; ) @script:python@ p1 << r.p1; p2 << r.p2; @@ print "* file: %s kmalloc %s return %s" % (p1[0].file,p1[0].line,p2[0].line) // Signed-off-by: Julia Lawall Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/trace/trace.c | 1 + 1 file changed, 1 insertion(+) diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index 697eda36b86..d86e3252f30 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -1936,6 +1936,7 @@ __tracing_open(struct inode *inode, struct file *file, int *ret) ring_buffer_read_finish(iter->buffer_iter[cpu]); } mutex_unlock(&trace_types_lock); + kfree(iter); return ERR_PTR(-ENOMEM); } From a6a0c4ca7edb378a8a7332501f097089cb1051c4 Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Tue, 18 Nov 2008 06:56:51 -0800 Subject: [PATCH 055/160] suspend: use WARN not WARN_ON to print the message By using WARN(), kerneloops.org can collect which component is causing the delay and make statistics about that. suspend_test_finish() is currently the number 2 item but unless we can collect who's causing it we're not going to be able to fix the hot topic ones.. Signed-off-by: Arjan van de Ven Signed-off-by: Linus Torvalds --- kernel/power/main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/power/main.c b/kernel/power/main.c index 19122cf6d82..b8f7ce9473e 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -174,7 +174,7 @@ static void suspend_test_finish(const char *label) * has some performance issues. The stack dump of a WARN_ON * is more likely to get the right attention than a printk... */ - WARN_ON(msec > (TEST_SUSPEND_SECONDS * 1000)); + WARN(msec > (TEST_SUSPEND_SECONDS * 1000), "Component: %s\n", label); } #else From 66359f849074c55fb80d15235ea5b99f1918663c Mon Sep 17 00:00:00 2001 From: Mike Frysinger Date: Tue, 18 Nov 2008 15:12:42 +0800 Subject: [PATCH 056/160] Parport driver: disable pc-style parport on Blackfin systems Signed-off-by: Mike Frysinger Signed-off-by: Bryan Wu Signed-off-by: Linus Torvalds --- drivers/parport/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/parport/Kconfig b/drivers/parport/Kconfig index 209b4a464bc..855f389eea4 100644 --- a/drivers/parport/Kconfig +++ b/drivers/parport/Kconfig @@ -36,7 +36,7 @@ if PARPORT config PARPORT_PC tristate "PC-style hardware" depends on (!SPARC64 || PCI) && !SPARC32 && !M32R && !FRV && \ - (!M68K || ISA) && !MN10300 && !AVR32 + (!M68K || ISA) && !MN10300 && !AVR32 && !BLACKFIN ---help--- You should say Y here if you have a PC-style parallel port. All IBM PC compatible computers and some Alphas have PC-style From bfb59820ee46616a7bdb4af6b8f7e109646de6ec Mon Sep 17 00:00:00 2001 From: Steve French Date: Tue, 18 Nov 2008 16:33:48 +0000 Subject: [PATCH 057/160] [CIFS] fix check for dead tcon in smb_init This was recently changed to check for need_reconnect, but should actually be a check for a tidStatus of CifsExiting. Signed-off-by: Jeff Layton Signed-off-by: Steve French --- fs/cifs/CHANGES | 6 +++++- fs/cifs/cifssmb.c | 2 +- 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index 8855331b2fb..e078b7aea14 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -8,7 +8,11 @@ handling fcntl(F_SETLEASE). Convert cifs to using blocking tcp sends, and also let tcp autotune the socket send and receive buffers. This reduces the number of EAGAIN errors returned by TCP/IP in high stress workloads (and the number of retries on socket writes -when sending large SMBWriteX requests). +when sending large SMBWriteX requests). Fix case in which a portion of +data can in some cases not get written to the file on the server before the +file is closed. Fix DFS parsing to properly handle path consumed field, +and to handle certain codepage conversions better. Fix mount and +umount race that can cause oops in mount or umount or reconnect. Version 1.54 ------------ diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index bdda46dd435..2af8626ced4 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -295,7 +295,7 @@ smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, check for tcp and smb session status done differently for those three - in the calling routine */ if (tcon) { - if (tcon->need_reconnect) { + if (tcon->tidStatus == CifsExiting) { /* only tree disconnect, open, and write, (and ulogoff which does not have tcon) are allowed as we start force umount */ From 641d2f63cfe24539e154efa2f932937934c27dde Mon Sep 17 00:00:00 2001 From: Vegard Nossum Date: Tue, 18 Nov 2008 19:22:13 +0100 Subject: [PATCH 058/160] trace: introduce missing mutex_unlock() Impact: fix tracing buffer mutex leak in case of allocation failure This error was spotted by this semantic patch: http://www.emn.fr/x-info/coccinelle/mut.html It looks correct as far as I can tell. Please review. Signed-off-by: Vegard Nossum Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/trace/ring_buffer.c | 1 + 1 file changed, 1 insertion(+) diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 036456cbb4f..f780e9552f9 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -617,6 +617,7 @@ int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size) list_del_init(&page->list); free_buffer_page(page); } + mutex_unlock(&buffer->mutex); return -ENOMEM; } From 093bac154c142fa1fb31a3ac69ae1bc08930231b Mon Sep 17 00:00:00 2001 From: Steve Conklin Date: Fri, 14 Nov 2008 00:55:51 -0600 Subject: [PATCH 059/160] x86: quirk for reboot stalls on a Dell Optiplex 330 Dell Optiplex 330 appears to hang on reboot. This is resolved by adding a quirk to set bios reboot. Signed-off-by: Leann Ogasawara Signed-off-by: Steve Conklin Signed-off-by: Ingo Molnar --- arch/x86/kernel/reboot.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 724adfc63cb..cc5a2545dd4 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -169,6 +169,15 @@ static struct dmi_system_id __initdata reboot_dmi_table[] = { DMI_MATCH(DMI_BOARD_NAME, "0KW626"), }, }, + { /* Handle problems with rebooting on Dell Optiplex 330 with 0KP561 */ + .callback = set_bios_reboot, + .ident = "Dell OptiPlex 330", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "OptiPlex 330"), + DMI_MATCH(DMI_BOARD_NAME, "0KP561"), + }, + }, { /* Handle problems with rebooting on Dell 2400's */ .callback = set_bios_reboot, .ident = "Dell PowerEdge 2400", From 20a4a236c7de5c915551cdc562482aa53eaff40e Mon Sep 17 00:00:00 2001 From: Hiroshi Shimamoto Date: Thu, 13 Nov 2008 18:06:04 -0800 Subject: [PATCH 060/160] x86: uaccess_64: fix return value in __copy_from_user() __copy_from_user() will return invalid value 16 when it fails to access user space and the size is 10. Signed-off-by: Hiroshi Shimamoto Signed-off-by: Ingo Molnar --- arch/x86/include/asm/uaccess_64.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/include/asm/uaccess_64.h b/arch/x86/include/asm/uaccess_64.h index 664f15280f1..f8cfd00db45 100644 --- a/arch/x86/include/asm/uaccess_64.h +++ b/arch/x86/include/asm/uaccess_64.h @@ -46,7 +46,7 @@ int __copy_from_user(void *dst, const void __user *src, unsigned size) return ret; case 10: __get_user_asm(*(u64 *)dst, (u64 __user *)src, - ret, "q", "", "=r", 16); + ret, "q", "", "=r", 10); if (unlikely(ret)) return ret; __get_user_asm(*(u16 *)(8 + (char *)dst), From 9b44fb89cab6e01816cdc05d6b59fdcf8100b8c3 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 29 Oct 2008 23:24:14 +0100 Subject: [PATCH 061/160] libertas_tf: fix skb tail pointer skb->tail can't be meant here because it's not the same across 32/64 bit compilations. This means there's no way the current driver can work on 64-bit architectures. Signed-off-by: Johannes Berg Cc: stable@kernel.org [2.6.27] Signed-off-by: John W. Linville --- drivers/net/wireless/libertas_tf/if_usb.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/wireless/libertas_tf/if_usb.c b/drivers/net/wireless/libertas_tf/if_usb.c index 1cc03a8dd67..59634c33b1f 100644 --- a/drivers/net/wireless/libertas_tf/if_usb.c +++ b/drivers/net/wireless/libertas_tf/if_usb.c @@ -331,7 +331,7 @@ static int __if_usb_submit_rx_urb(struct if_usb_card *cardp, /* Fill the receive configuration URB and initialise the Rx call back */ usb_fill_bulk_urb(cardp->rx_urb, cardp->udev, usb_rcvbulkpipe(cardp->udev, cardp->ep_in), - (void *) (skb->tail), + skb_tail_pointer(skb), MRVDRV_ETH_RX_PACKET_BUFFER_SIZE, callbackfn, cardp); cardp->rx_urb->transfer_flags |= URB_ZERO_PACKET; From 8e3bad65a59915f2ddc40f62a180ad81695d8440 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Mon, 17 Nov 2008 10:59:59 +0100 Subject: [PATCH 062/160] mac80211: remove ieee80211_notify_mac Before ieee80211_notify_mac() was added, it was presented with the use case of using it to tell mac80211 that the association may have been lost because the firmware crashed/reset. Since then, it has also been used by iwlwifi to (slightly) speed up re-association after resume, a workaround around the fact that mac80211 has no suspend/resume handling yet. It is also not used by any other drivers, so clearly it cannot be necessary for "good enough" suspend/resume. Unfortunately, the callback suffers from a severe problem: It only works for station mode. If suspend/resume happens while in IBSS or any other mode (but station), then the callback is pointless. Recently, it has created a number of locking issues, first because it required rtnl locking rather than RCU due to calling sleeping functions within the critical section, and now because it's called by iwlwifi from the mac80211 workqueue that may not use the rtnl because it is flushed under rtnl. (cf. http://bugzilla.kernel.org/show_bug.cgi?id=12046) I think, therefore, that we should take a step back, remove it entirely for now and add the small feature it provided properly. For suspend and resume we will need to introduce new hooks, and for the case where the firmware was reset the driver will probably simply just pretend it has done a suspend/resume cycle to get mac80211 to reprogram the hardware completely, not just try to connect to the current AP again in station mode. When doing so, we will need to take into account locking issues and possibly defer to schedule_work from within mac80211 for the resume operation, while the suspend operation must be done directly. Proper suspend/resume should also not necessarily try to reconnect to the current AP, the time spent in suspend may have been short enough to not be disconnected from the AP, mac80211 will detect that the AP went out of range quickly if it did, and if the association is lost then the AP will disassoc as soon as a data frame is sent. We might also take into account WWOL then, and have mac80211 program the hardware into such a mode where it is available and requested. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- drivers/net/wireless/iwlwifi/iwl-agn.c | 1 - drivers/net/wireless/iwlwifi/iwl3945-base.c | 1 - include/net/mac80211.h | 20 ------------------- net/mac80211/mlme.c | 22 --------------------- 4 files changed, 44 deletions(-) diff --git a/drivers/net/wireless/iwlwifi/iwl-agn.c b/drivers/net/wireless/iwlwifi/iwl-agn.c index 8d690a0eb1a..6751bb2b8ae 100644 --- a/drivers/net/wireless/iwlwifi/iwl-agn.c +++ b/drivers/net/wireless/iwlwifi/iwl-agn.c @@ -2341,7 +2341,6 @@ static void iwl_bg_alive_start(struct work_struct *data) mutex_lock(&priv->mutex); iwl_alive_start(priv); mutex_unlock(&priv->mutex); - ieee80211_notify_mac(priv->hw, IEEE80211_NOTIFY_RE_ASSOC); } static void iwl4965_bg_rf_kill(struct work_struct *work) diff --git a/drivers/net/wireless/iwlwifi/iwl3945-base.c b/drivers/net/wireless/iwlwifi/iwl3945-base.c index 285b53e7e26..45a6b0c3569 100644 --- a/drivers/net/wireless/iwlwifi/iwl3945-base.c +++ b/drivers/net/wireless/iwlwifi/iwl3945-base.c @@ -6012,7 +6012,6 @@ static void iwl3945_bg_alive_start(struct work_struct *data) mutex_lock(&priv->mutex); iwl3945_alive_start(priv); mutex_unlock(&priv->mutex); - ieee80211_notify_mac(priv->hw, IEEE80211_NOTIFY_RE_ASSOC); } static void iwl3945_bg_rf_kill(struct work_struct *work) diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 8856e2d60e9..73d81bc6aa7 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -73,14 +73,6 @@ * not do so then mac80211 may add this under certain circumstances. */ -/** - * enum ieee80211_notification_type - Low level driver notification - * @IEEE80211_NOTIFY_RE_ASSOC: start the re-association sequence - */ -enum ieee80211_notification_types { - IEEE80211_NOTIFY_RE_ASSOC, -}; - /** * struct ieee80211_ht_bss_info - describing BSS's HT characteristics * @@ -1797,18 +1789,6 @@ void ieee80211_stop_tx_ba_cb(struct ieee80211_hw *hw, u8 *ra, u8 tid); void ieee80211_stop_tx_ba_cb_irqsafe(struct ieee80211_hw *hw, const u8 *ra, u16 tid); -/** - * ieee80211_notify_mac - low level driver notification - * @hw: pointer as obtained from ieee80211_alloc_hw(). - * @notif_type: enum ieee80211_notification_types - * - * This function must be called by low level driver to inform mac80211 of - * low level driver status change or force mac80211 to re-assoc for low - * level driver internal error that require re-assoc. - */ -void ieee80211_notify_mac(struct ieee80211_hw *hw, - enum ieee80211_notification_types notif_type); - /** * ieee80211_find_sta - find a station * diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c index 14d165f0df7..409bb771623 100644 --- a/net/mac80211/mlme.c +++ b/net/mac80211/mlme.c @@ -2560,25 +2560,3 @@ void ieee80211_mlme_notify_scan_completed(struct ieee80211_local *local) ieee80211_restart_sta_timer(sdata); rcu_read_unlock(); } - -/* driver notification call */ -void ieee80211_notify_mac(struct ieee80211_hw *hw, - enum ieee80211_notification_types notif_type) -{ - struct ieee80211_local *local = hw_to_local(hw); - struct ieee80211_sub_if_data *sdata; - - switch (notif_type) { - case IEEE80211_NOTIFY_RE_ASSOC: - rtnl_lock(); - list_for_each_entry(sdata, &local->interfaces, list) { - if (sdata->vif.type != NL80211_IFTYPE_STATION) - continue; - - ieee80211_sta_req_auth(sdata, &sdata->u.sta); - } - rtnl_unlock(); - break; - } -} -EXPORT_SYMBOL(ieee80211_notify_mac); From 4018517a1a69a85c3d61b20fa02f187b80773137 Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Tue, 18 Nov 2008 01:47:21 +0100 Subject: [PATCH 063/160] iwlagn: fix RX skb alignment So I dug deeper into the DMA problems I had with iwlagn and a kind soul helped me in that he said something about pci-e alignment and mentioned the iwl_rx_allocate function to check for crossing 4KB boundaries. Since there's 8KB A-MPDU support, crossing 4k boundaries didn't seem like something the device would fail with, but when I looked into the function for a minute anyway I stumbled over this little gem: BUG_ON(rxb->dma_addr & (~DMA_BIT_MASK(36) & 0xff)); Clearly, that is a totally bogus check, one would hope the compiler removes it entirely. (Think about it) After fixing it, I obviously ran into it, nothing guarantees the alignment the way you want it, because of the way skbs and their headroom are allocated. I won't explain that here nor double-check that I'm right, that goes beyond what most of the CC'ed people care about. So then I came up with the patch below, and so far my system has survived minutes with 64K pages, when it would previously fail in seconds. And I haven't seen a single instance of the TX bug either. But when you see the patch it'll be pretty obvious to you why. This should fix the following reported kernel bugs: http://bugzilla.kernel.org/show_bug.cgi?id=11596 http://bugzilla.kernel.org/show_bug.cgi?id=11393 http://bugzilla.kernel.org/show_bug.cgi?id=11983 I haven't checked if there are any elsewhere, but I suppose RHBZ will have a few instances too... I'd like to ask anyone who is CC'ed (those are people I know ran into the bug) to try this patch. I am convinced that this patch is correct in spirit, but I haven't understood why, for example, there are so many unmap calls. I'm not entirely convinced that this is the only bug leading to the TX reply errors. Signed-off-by: Johannes Berg Signed-off-by: John W. Linville --- drivers/net/wireless/iwlwifi/iwl-agn.c | 6 +++--- drivers/net/wireless/iwlwifi/iwl-dev.h | 3 ++- drivers/net/wireless/iwlwifi/iwl-rx.c | 26 +++++++++++++++++--------- 3 files changed, 22 insertions(+), 13 deletions(-) diff --git a/drivers/net/wireless/iwlwifi/iwl-agn.c b/drivers/net/wireless/iwlwifi/iwl-agn.c index 6751bb2b8ae..444c5cc05f0 100644 --- a/drivers/net/wireless/iwlwifi/iwl-agn.c +++ b/drivers/net/wireless/iwlwifi/iwl-agn.c @@ -1384,7 +1384,7 @@ void iwl_rx_handle(struct iwl_priv *priv) rxq->queue[i] = NULL; - pci_dma_sync_single_for_cpu(priv->pci_dev, rxb->dma_addr, + pci_dma_sync_single_for_cpu(priv->pci_dev, rxb->aligned_dma_addr, priv->hw_params.rx_buf_size, PCI_DMA_FROMDEVICE); pkt = (struct iwl_rx_packet *)rxb->skb->data; @@ -1436,8 +1436,8 @@ void iwl_rx_handle(struct iwl_priv *priv) rxb->skb = NULL; } - pci_unmap_single(priv->pci_dev, rxb->dma_addr, - priv->hw_params.rx_buf_size, + pci_unmap_single(priv->pci_dev, rxb->real_dma_addr, + priv->hw_params.rx_buf_size + 256, PCI_DMA_FROMDEVICE); spin_lock_irqsave(&rxq->lock, flags); list_add_tail(&rxb->list, &priv->rxq.rx_used); diff --git a/drivers/net/wireless/iwlwifi/iwl-dev.h b/drivers/net/wireless/iwlwifi/iwl-dev.h index c018121085e..9966d4e384c 100644 --- a/drivers/net/wireless/iwlwifi/iwl-dev.h +++ b/drivers/net/wireless/iwlwifi/iwl-dev.h @@ -89,7 +89,8 @@ extern struct iwl_cfg iwl5100_abg_cfg; #define DEFAULT_LONG_RETRY_LIMIT 4U struct iwl_rx_mem_buffer { - dma_addr_t dma_addr; + dma_addr_t real_dma_addr; + dma_addr_t aligned_dma_addr; struct sk_buff *skb; struct list_head list; }; diff --git a/drivers/net/wireless/iwlwifi/iwl-rx.c b/drivers/net/wireless/iwlwifi/iwl-rx.c index 7cde9d76ff5..0509c16dbe7 100644 --- a/drivers/net/wireless/iwlwifi/iwl-rx.c +++ b/drivers/net/wireless/iwlwifi/iwl-rx.c @@ -204,7 +204,7 @@ int iwl_rx_queue_restock(struct iwl_priv *priv) list_del(element); /* Point to Rx buffer via next RBD in circular buffer */ - rxq->bd[rxq->write] = iwl_dma_addr2rbd_ptr(priv, rxb->dma_addr); + rxq->bd[rxq->write] = iwl_dma_addr2rbd_ptr(priv, rxb->aligned_dma_addr); rxq->queue[rxq->write] = rxb; rxq->write = (rxq->write + 1) & RX_QUEUE_MASK; rxq->free_count--; @@ -251,7 +251,7 @@ void iwl_rx_allocate(struct iwl_priv *priv) rxb = list_entry(element, struct iwl_rx_mem_buffer, list); /* Alloc a new receive buffer */ - rxb->skb = alloc_skb(priv->hw_params.rx_buf_size, + rxb->skb = alloc_skb(priv->hw_params.rx_buf_size + 256, __GFP_NOWARN | GFP_ATOMIC); if (!rxb->skb) { if (net_ratelimit()) @@ -266,9 +266,17 @@ void iwl_rx_allocate(struct iwl_priv *priv) list_del(element); /* Get physical address of RB/SKB */ - rxb->dma_addr = - pci_map_single(priv->pci_dev, rxb->skb->data, - priv->hw_params.rx_buf_size, PCI_DMA_FROMDEVICE); + rxb->real_dma_addr = pci_map_single( + priv->pci_dev, + rxb->skb->data, + priv->hw_params.rx_buf_size + 256, + PCI_DMA_FROMDEVICE); + /* dma address must be no more than 36 bits */ + BUG_ON(rxb->real_dma_addr & ~DMA_BIT_MASK(36)); + /* and also 256 byte aligned! */ + rxb->aligned_dma_addr = ALIGN(rxb->real_dma_addr, 256); + skb_reserve(rxb->skb, rxb->aligned_dma_addr - rxb->real_dma_addr); + list_add_tail(&rxb->list, &rxq->rx_free); rxq->free_count++; } @@ -300,8 +308,8 @@ void iwl_rx_queue_free(struct iwl_priv *priv, struct iwl_rx_queue *rxq) for (i = 0; i < RX_QUEUE_SIZE + RX_FREE_BUFFERS; i++) { if (rxq->pool[i].skb != NULL) { pci_unmap_single(priv->pci_dev, - rxq->pool[i].dma_addr, - priv->hw_params.rx_buf_size, + rxq->pool[i].real_dma_addr, + priv->hw_params.rx_buf_size + 256, PCI_DMA_FROMDEVICE); dev_kfree_skb(rxq->pool[i].skb); } @@ -354,8 +362,8 @@ void iwl_rx_queue_reset(struct iwl_priv *priv, struct iwl_rx_queue *rxq) * to an SKB, so we need to unmap and free potential storage */ if (rxq->pool[i].skb != NULL) { pci_unmap_single(priv->pci_dev, - rxq->pool[i].dma_addr, - priv->hw_params.rx_buf_size, + rxq->pool[i].real_dma_addr, + priv->hw_params.rx_buf_size + 256, PCI_DMA_FROMDEVICE); priv->alloc_rxb_skb--; dev_kfree_skb(rxq->pool[i].skb); From f10ed36ec1118c6f9523cd7e53cb0aadb53efe9f Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Fri, 7 Nov 2008 22:36:02 -0500 Subject: [PATCH 064/160] ftrace: fix set_ftrace_filter Impact: fix of output of set_ftrace_filter The commit "ftrace: do not show freed records in available_filter_functions" Removed a bit too much from the set_ftrace_filter code, where we now see all functions in the set_ftrace_filter file even when we set a filter. Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/trace/ftrace.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 4a39d24568c..dcac7418f68 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -738,6 +738,9 @@ t_next(struct seq_file *m, void *v, loff_t *pos) ((iter->flags & FTRACE_ITER_FAILURES) && !(rec->flags & FTRACE_FL_FAILED)) || + ((iter->flags & FTRACE_ITER_FILTER) && + !(rec->flags & FTRACE_FL_FILTER)) || + ((iter->flags & FTRACE_ITER_NOTRACE) && !(rec->flags & FTRACE_FL_NOTRACE))) { rec = NULL; From 820432783190b4096499e38a4a4d7095c511913d Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Tue, 18 Nov 2008 23:57:14 -0500 Subject: [PATCH 065/160] ftrace: make filtered functions effective on setting Impact: fix filter selection to apply when set It can be confusing when the set_filter_functions is set (or cleared) and the functions being recorded by the dynamic tracer does not match. This patch causes the code to be updated if the function tracer is enabled and the filter is changed. Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/trace/ftrace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index dcac7418f68..5cbddb59e99 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -1189,7 +1189,7 @@ ftrace_regex_release(struct inode *inode, struct file *file, int enable) mutex_lock(&ftrace_sysctl_lock); mutex_lock(&ftrace_start_lock); - if (iter->filtered && ftrace_start && ftrace_enabled) + if (ftrace_start && ftrace_enabled) ftrace_run_update_code(FTRACE_ENABLE_CALLS); mutex_unlock(&ftrace_start_lock); mutex_unlock(&ftrace_sysctl_lock); From 32464779a1b8c15e9aa9aa0306b2f735080df9d8 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Tue, 18 Nov 2008 20:33:02 -0500 Subject: [PATCH 066/160] ftrace: fix dyn ftrace filter selection Impact: clean up and fix for dyn ftrace filter selection The previous logic of the dynamic ftrace selection of enabling or disabling functions was complex and incorrect. This patch simplifies the code and corrects the usage. This simplification also makes the code more robust. Here is the correct logic: Given a function that can be traced by dynamic ftrace: If the function is not to be traced, disable it if it was enabled. (this is if the function is in the set_ftrace_notrace file) (filter is on if there exists any functions in set_ftrace_filter file) If the filter is on, and we are enabling functions: If the function is in set_ftrace_filter, enable it if it is not already enabled. If the function is not in set_ftrace_filter, disable it if it is not already disabled. Otherwise, if the filter is off and we are enabling function tracing: Enable the function if it is not already enabled. Otherwise, if we are disabling function tracing: Disable the function if it is not already disabled. This code now sets or clears the ENABLED flag in the record, and at the end it will enable the function if the flag is set, or disable the function if the flag is cleared. The parameters for the function that does the above logic is also simplified. Instead of passing in confusing "new" and "old" where they might be swapped if the "enabled" flag is not set. The old logic even had one of the above always NULL and had to be filled in. The new logic simply passes in one parameter called "nop". A "call" is calculated in the code, and at the end of the logic, when we know we need to either disable or enable the function, we can then use the "nop" and "call" properly. This code is more robust than the previous version. Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/trace/ftrace.c | 112 ++++++++++++++++++++---------------------- 1 file changed, 52 insertions(+), 60 deletions(-) diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 5cbddb59e99..fdaab04a028 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -327,96 +327,89 @@ ftrace_record_ip(unsigned long ip) static int __ftrace_replace_code(struct dyn_ftrace *rec, - unsigned char *old, unsigned char *new, int enable) + unsigned char *nop, int enable) { unsigned long ip, fl; + unsigned char *call, *old, *new; ip = rec->ip; - if (ftrace_filtered && enable) { - /* - * If filtering is on: - * - * If this record is set to be filtered and - * is enabled then do nothing. - * - * If this record is set to be filtered and - * it is not enabled, enable it. - * - * If this record is not set to be filtered - * and it is not enabled do nothing. - * - * If this record is set not to trace then - * do nothing. - * - * If this record is set not to trace and - * it is enabled then disable it. - * - * If this record is not set to be filtered and - * it is enabled, disable it. - */ - - fl = rec->flags & (FTRACE_FL_FILTER | FTRACE_FL_NOTRACE | - FTRACE_FL_ENABLED); - - if ((fl == (FTRACE_FL_FILTER | FTRACE_FL_ENABLED)) || - (fl == (FTRACE_FL_FILTER | FTRACE_FL_NOTRACE)) || - !fl || (fl == FTRACE_FL_NOTRACE)) + /* + * If this record is not to be traced and + * it is not enabled then do nothing. + * + * If this record is not to be traced and + * it is enabled then disabled it. + * + */ + if (rec->flags & FTRACE_FL_NOTRACE) { + if (rec->flags & FTRACE_FL_ENABLED) + rec->flags &= ~FTRACE_FL_ENABLED; + else return 0; + } else if (ftrace_filtered && enable) { /* - * If it is enabled disable it, - * otherwise enable it! + * Filtering is on: */ - if (fl & FTRACE_FL_ENABLED) { - /* swap new and old */ - new = old; - old = ftrace_call_replace(ip, FTRACE_ADDR); + + fl = rec->flags & (FTRACE_FL_FILTER | FTRACE_FL_ENABLED); + + /* Record is filtered and enabled, do nothing */ + if (fl == (FTRACE_FL_FILTER | FTRACE_FL_ENABLED)) + return 0; + + /* Record is not filtered and is not enabled do nothing */ + if (!fl) + return 0; + + /* Record is not filtered but enabled, disable it */ + if (fl == FTRACE_FL_ENABLED) rec->flags &= ~FTRACE_FL_ENABLED; - } else { - new = ftrace_call_replace(ip, FTRACE_ADDR); + else + /* Otherwise record is filtered but not enabled, enable it */ rec->flags |= FTRACE_FL_ENABLED; - } } else { + /* Disable or not filtered */ if (enable) { - /* - * If this record is set not to trace and is - * not enabled, do nothing. - */ - fl = rec->flags & (FTRACE_FL_NOTRACE | FTRACE_FL_ENABLED); - if (fl == FTRACE_FL_NOTRACE) - return 0; - - new = ftrace_call_replace(ip, FTRACE_ADDR); - } else - old = ftrace_call_replace(ip, FTRACE_ADDR); - - if (enable) { + /* if record is enabled, do nothing */ if (rec->flags & FTRACE_FL_ENABLED) return 0; + rec->flags |= FTRACE_FL_ENABLED; + } else { + + /* if record is not enabled do nothing */ if (!(rec->flags & FTRACE_FL_ENABLED)) return 0; + rec->flags &= ~FTRACE_FL_ENABLED; } } + call = ftrace_call_replace(ip, FTRACE_ADDR); + + if (rec->flags & FTRACE_FL_ENABLED) { + old = nop; + new = call; + } else { + old = call; + new = nop; + } + return ftrace_modify_code(ip, old, new); } static void ftrace_replace_code(int enable) { int i, failed; - unsigned char *new = NULL, *old = NULL; + unsigned char *nop = NULL; struct dyn_ftrace *rec; struct ftrace_page *pg; - if (enable) - old = ftrace_nop_replace(); - else - new = ftrace_nop_replace(); + nop = ftrace_nop_replace(); for (pg = ftrace_pages_start; pg; pg = pg->next) { for (i = 0; i < pg->index; i++) { @@ -434,7 +427,7 @@ static void ftrace_replace_code(int enable) unfreeze_record(rec); } - failed = __ftrace_replace_code(rec, old, new, enable); + failed = __ftrace_replace_code(rec, nop, enable); if (failed && (rec->flags & FTRACE_FL_CONVERTED)) { rec->flags |= FTRACE_FL_FAILED; if ((system_state == SYSTEM_BOOTING) || @@ -538,8 +531,7 @@ static void ftrace_startup(void) mutex_lock(&ftrace_start_lock); ftrace_start++; - if (ftrace_start == 1) - command |= FTRACE_ENABLE_CALLS; + command |= FTRACE_ENABLE_CALLS; if (saved_ftrace_func != ftrace_trace_function) { saved_ftrace_func = ftrace_trace_function; From b4e0f9eb8aeceb22c48fee005378bd19e25216fc Mon Sep 17 00:00:00 2001 From: FUJITA Tomonori Date: Wed, 19 Nov 2008 13:53:42 +0900 Subject: [PATCH 067/160] intel-iommu: fix compile warnings Impact: cleanup I got the following warnings on IA64: linux-2.6/drivers/pci/intel-iommu.c: In function 'init_dmars': linux-2.6/drivers/pci/intel-iommu.c:1658: warning: format '%Lx' expects type 'long long unsigned int', but argument 2 has type 'u64' linux-2.6/drivers/pci/intel-iommu.c:1663: warning: format '%Lx' expects type 'long long unsigned int', but argument 2 has type 'u64' Another victim of int-ll64.h versus int-l64.h confusion between platforms. ->reg_base_addr has a type of u64 - which can only be printed out consistently if we cast its type up to LL. [ Eventually reg_base_addr should be converted to phys_addr_t, for which we have the %pR printk helper - but that is out of the scope of late -rc's. ] Signed-off-by: FUJITA Tomonori Signed-off-by: Ingo Molnar --- drivers/pci/intel-iommu.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c index a2692724b68..5c8baa43ac9 100644 --- a/drivers/pci/intel-iommu.c +++ b/drivers/pci/intel-iommu.c @@ -1655,12 +1655,14 @@ int __init init_dmars(void) iommu->flush.flush_context = __iommu_flush_context; iommu->flush.flush_iotlb = __iommu_flush_iotlb; printk(KERN_INFO "IOMMU 0x%Lx: using Register based " - "invalidation\n", drhd->reg_base_addr); + "invalidation\n", + (unsigned long long)drhd->reg_base_addr); } else { iommu->flush.flush_context = qi_flush_context; iommu->flush.flush_iotlb = qi_flush_iotlb; printk(KERN_INFO "IOMMU 0x%Lx: using Queued " - "invalidation\n", drhd->reg_base_addr); + "invalidation\n", + (unsigned long long)drhd->reg_base_addr); } } From c3e388964baa00d8c3960f23e8c8a1fb3966759e Mon Sep 17 00:00:00 2001 From: Benjamin Thery Date: Wed, 19 Nov 2008 14:07:41 -0800 Subject: [PATCH 068/160] net: fix ip_mr_init() error path Similarly to IPv6 ip6_mr_init() (fixed last week), the order of cleanup operations in the error/exit section of ip_mr_init() is completely inversed. It should be the other way around. Also a del_timer() is missing in the error path. I should have guessed last week that this same error existed in ipmr.c too, as ip6mr.c is largely inspired by ipmr.c. Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- net/ipv4/ipmr.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c index b42e082cc17..25924b1eb2e 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -1945,13 +1945,14 @@ int __init ip_mr_init(void) goto proc_cache_fail; #endif return 0; -reg_notif_fail: - kmem_cache_destroy(mrt_cachep); #ifdef CONFIG_PROC_FS -proc_vif_fail: - unregister_netdevice_notifier(&ip_mr_notifier); proc_cache_fail: proc_net_remove(&init_net, "ip_mr_vif"); +proc_vif_fail: + unregister_netdevice_notifier(&ip_mr_notifier); #endif +reg_notif_fail: + del_timer(&ipmr_expire_timer); + kmem_cache_destroy(mrt_cachep); return err; } From bfdbc0acadcc761b94814d78f0acec90f0d760de Mon Sep 17 00:00:00 2001 From: Robert Olsson Date: Wed, 19 Nov 2008 14:09:47 -0800 Subject: [PATCH 069/160] pktgen: fix multiple queue warning As number of TX queues in unrelated to number of CPU's we remove this test and just make sure nxtq never gets exceeded. Signed-off-by: Robert Olsson Signed-off-by: David S. Miller --- net/core/pktgen.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) diff --git a/net/core/pktgen.c b/net/core/pktgen.c index a47f5bad110..8997e912aaa 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -1973,13 +1973,7 @@ static void pktgen_setup_inject(struct pktgen_dev *pkt_dev) /* make sure that we don't pick a non-existing transmit queue */ ntxq = pkt_dev->odev->real_num_tx_queues; - if (ntxq > num_online_cpus() && (pkt_dev->flags & F_QUEUE_MAP_CPU)) { - printk(KERN_WARNING "pktgen: WARNING: QUEUE_MAP_CPU " - "disabled because CPU count (%d) exceeds number " - "of tx queues (%d) on %s\n", num_online_cpus(), ntxq, - pkt_dev->odev->name); - pkt_dev->flags &= ~F_QUEUE_MAP_CPU; - } + if (ntxq <= pkt_dev->queue_map_min) { printk(KERN_WARNING "pktgen: WARNING: Requested " "queue_map_min (zero-based) (%d) exceeds valid range " @@ -2202,6 +2196,7 @@ static void set_cur_queue_map(struct pktgen_dev *pkt_dev) } pkt_dev->cur_queue_map = t; } + pkt_dev->cur_queue_map = pkt_dev->cur_queue_map % pkt_dev->odev->real_num_tx_queues; } /* Increment/randomize headers according to flags and current values From 1df8fb3d5f078f9cab901b6106ef2c9b74eef7df Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Tue, 11 Nov 2008 17:17:45 +0800 Subject: [PATCH 070/160] PCI: Fix disable IRQ 0 in pci_reset_function() Before initialization, dev->irq may be zero. Make sure we don't disable it at reset time in that case. Reviewed-by: Matthew Wilcox Signed-off-by: Sheng Yang Signed-off-by: Jesse Barnes --- drivers/pci/pci.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 21f2ac639ca..28af496b441 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -1832,7 +1832,7 @@ int pci_reset_function(struct pci_dev *dev) if (!(cap & PCI_EXP_DEVCAP_FLR)) return -ENOTTY; - if (!dev->msi_enabled && !dev->msix_enabled) + if (!dev->msi_enabled && !dev->msix_enabled && dev->irq != 0) disable_irq(dev->irq); pci_save_state(dev); @@ -1841,7 +1841,7 @@ int pci_reset_function(struct pci_dev *dev) r = pci_execute_reset_function(dev); pci_restore_state(dev); - if (!dev->msi_enabled && !dev->msix_enabled) + if (!dev->msi_enabled && !dev->msix_enabled && dev->irq != 0) enable_irq(dev->irq); return r; From 79aefa45b20940cbb9104464548ff74f80f2395e Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Wed, 19 Nov 2008 14:17:02 -0800 Subject: [PATCH 071/160] ixgbe: fix compilation with gcc-3.4 CC [M] drivers/net/ixgbe/ixgbe_main.o drivers/net/ixgbe/ixgbe_main.c: In function `ixgbe_intr': drivers/net/ixgbe/ixgbe_main.c:1290: sorry, unimplemented: inlining failed in call to 'ixgbe_irq_enable': function body not available drivers/net/ixgbe/ixgbe_main.c:1312: sorry, unimplemented: called from here make[4]: *** [drivers/net/ixgbe/ixgbe_main.o] Error 1 Signed-off-by: Alexey Dobriyan Acked-by: Peter P Waskiewicz Jr Signed-off-by: David S. Miller --- drivers/net/ixgbe/ixgbe_main.c | 58 ++++++++++++++++------------------ 1 file changed, 28 insertions(+), 30 deletions(-) diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c index 7548fb7360d..36f2bb666bf 100644 --- a/drivers/net/ixgbe/ixgbe_main.c +++ b/drivers/net/ixgbe/ixgbe_main.c @@ -1287,7 +1287,34 @@ static void ixgbe_set_itr(struct ixgbe_adapter *adapter) return; } -static inline void ixgbe_irq_enable(struct ixgbe_adapter *adapter); +/** + * ixgbe_irq_disable - Mask off interrupt generation on the NIC + * @adapter: board private structure + **/ +static inline void ixgbe_irq_disable(struct ixgbe_adapter *adapter) +{ + IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMC, ~0); + IXGBE_WRITE_FLUSH(&adapter->hw); + if (adapter->flags & IXGBE_FLAG_MSIX_ENABLED) { + int i; + for (i = 0; i < adapter->num_msix_vectors; i++) + synchronize_irq(adapter->msix_entries[i].vector); + } else { + synchronize_irq(adapter->pdev->irq); + } +} + +/** + * ixgbe_irq_enable - Enable default interrupt generation settings + * @adapter: board private structure + **/ +static inline void ixgbe_irq_enable(struct ixgbe_adapter *adapter) +{ + u32 mask; + mask = IXGBE_EIMS_ENABLE_MASK; + IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMS, mask); + IXGBE_WRITE_FLUSH(&adapter->hw); +} /** * ixgbe_intr - legacy mode Interrupt Handler @@ -1393,35 +1420,6 @@ static void ixgbe_free_irq(struct ixgbe_adapter *adapter) } } -/** - * ixgbe_irq_disable - Mask off interrupt generation on the NIC - * @adapter: board private structure - **/ -static inline void ixgbe_irq_disable(struct ixgbe_adapter *adapter) -{ - IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMC, ~0); - IXGBE_WRITE_FLUSH(&adapter->hw); - if (adapter->flags & IXGBE_FLAG_MSIX_ENABLED) { - int i; - for (i = 0; i < adapter->num_msix_vectors; i++) - synchronize_irq(adapter->msix_entries[i].vector); - } else { - synchronize_irq(adapter->pdev->irq); - } -} - -/** - * ixgbe_irq_enable - Enable default interrupt generation settings - * @adapter: board private structure - **/ -static inline void ixgbe_irq_enable(struct ixgbe_adapter *adapter) -{ - u32 mask; - mask = IXGBE_EIMS_ENABLE_MASK; - IXGBE_WRITE_REG(&adapter->hw, IXGBE_EIMS, mask); - IXGBE_WRITE_FLUSH(&adapter->hw); -} - /** * ixgbe_configure_msi_and_legacy - Initialize PIN (INTA...) and MSI interrupts * From 566521d63720ab47576afb85147e5652993bf1e6 Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Wed, 19 Nov 2008 14:17:41 -0800 Subject: [PATCH 072/160] phonet: fix compilation with gcc-3.4 CC [M] net/phonet/af_phonet.o net/phonet/af_phonet.c: In function `pn_socket_create': net/phonet/af_phonet.c:38: sorry, unimplemented: inlining failed in call to 'phonet_proto_put': function body not available net/phonet/af_phonet.c:99: sorry, unimplemented: called from here make[3]: *** [net/phonet/af_phonet.o] Error 1 Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- net/phonet/af_phonet.c | 52 +++++++++++++++++++----------------------- 1 file changed, 24 insertions(+), 28 deletions(-) diff --git a/net/phonet/af_phonet.c b/net/phonet/af_phonet.c index 7ab30f668b5..9d211f12582 100644 --- a/net/phonet/af_phonet.c +++ b/net/phonet/af_phonet.c @@ -33,9 +33,30 @@ #include #include -static struct net_proto_family phonet_proto_family; -static struct phonet_protocol *phonet_proto_get(int protocol); -static inline void phonet_proto_put(struct phonet_protocol *pp); +/* Transport protocol registration */ +static struct phonet_protocol *proto_tab[PHONET_NPROTO] __read_mostly; +static DEFINE_SPINLOCK(proto_tab_lock); + +static struct phonet_protocol *phonet_proto_get(int protocol) +{ + struct phonet_protocol *pp; + + if (protocol >= PHONET_NPROTO) + return NULL; + + spin_lock(&proto_tab_lock); + pp = proto_tab[protocol]; + if (pp && !try_module_get(pp->prot->owner)) + pp = NULL; + spin_unlock(&proto_tab_lock); + + return pp; +} + +static inline void phonet_proto_put(struct phonet_protocol *pp) +{ + module_put(pp->prot->owner); +} /* protocol family functions */ @@ -375,10 +396,6 @@ static struct packet_type phonet_packet_type = { .func = phonet_rcv, }; -/* Transport protocol registration */ -static struct phonet_protocol *proto_tab[PHONET_NPROTO] __read_mostly; -static DEFINE_SPINLOCK(proto_tab_lock); - int __init_or_module phonet_proto_register(int protocol, struct phonet_protocol *pp) { @@ -412,27 +429,6 @@ void phonet_proto_unregister(int protocol, struct phonet_protocol *pp) } EXPORT_SYMBOL(phonet_proto_unregister); -static struct phonet_protocol *phonet_proto_get(int protocol) -{ - struct phonet_protocol *pp; - - if (protocol >= PHONET_NPROTO) - return NULL; - - spin_lock(&proto_tab_lock); - pp = proto_tab[protocol]; - if (pp && !try_module_get(pp->prot->owner)) - pp = NULL; - spin_unlock(&proto_tab_lock); - - return pp; -} - -static inline void phonet_proto_put(struct phonet_protocol *pp) -{ - module_put(pp->prot->owner); -} - /* Module registration */ static int __init phonet_init(void) { From b47300168e770b60ab96c8924854c3b0eb4260eb Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 19 Nov 2008 15:33:54 -0800 Subject: [PATCH 073/160] net: Do not fire linkwatch events until the device is registered. Several device drivers try to do things like netif_carrier_off() before register_netdev() is invoked. This is bogus, but too many drivers do this to fix them all up in one go. Reported-by: Folkert van Heusden Signed-off-by: David S. Miller --- net/sched/sch_generic.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 93cd30ce650..cdcd16fcfed 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -270,6 +270,8 @@ static void dev_watchdog_down(struct net_device *dev) void netif_carrier_on(struct net_device *dev) { if (test_and_clear_bit(__LINK_STATE_NOCARRIER, &dev->state)) { + if (dev->reg_state == NETREG_UNINITIALIZED) + return; linkwatch_fire_event(dev); if (netif_running(dev)) __netdev_watchdog_up(dev); @@ -285,8 +287,11 @@ EXPORT_SYMBOL(netif_carrier_on); */ void netif_carrier_off(struct net_device *dev) { - if (!test_and_set_bit(__LINK_STATE_NOCARRIER, &dev->state)) + if (!test_and_set_bit(__LINK_STATE_NOCARRIER, &dev->state)) { + if (dev->reg_state == NETREG_UNINITIALIZED) + return; linkwatch_fire_event(dev); + } } EXPORT_SYMBOL(netif_carrier_off); From 9d6ada9f4141fb8ab3c5d7dffe382f6a68b8e961 Mon Sep 17 00:00:00 2001 From: Paulius Zaleckas Date: Wed, 19 Nov 2008 15:38:24 -0800 Subject: [PATCH 074/160] phylib: fix phy name example in documentation All MDIO bus drivers currently name bus with "%x" format. There is one exception where mv643xx_eth driver is using "%d". Phy address on the bus uses format "%02x". Fixing phy name example to match all real life MDIO drivers. Signed-off-by: Paulius Zaleckas Signed-off-by: David S. Miller --- Documentation/networking/phy.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/networking/phy.txt b/Documentation/networking/phy.txt index 8df6a7b0e66..88bb71b46da 100644 --- a/Documentation/networking/phy.txt +++ b/Documentation/networking/phy.txt @@ -96,7 +96,7 @@ Letting the PHY Abstraction Layer do Everything static void adjust_link(struct net_device *dev); Next, you need to know the device name of the PHY connected to this device. - The name will look something like, "phy0:0", where the first number is the + The name will look something like, "0:00", where the first number is the bus id, and the second is the PHY's address on that bus. Typically, the bus is responsible for making its ID unique. From 31c221c49f92d17632e0d662eb62a27e8b425805 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Wed, 19 Nov 2008 15:50:59 -0800 Subject: [PATCH 075/160] net: jme.c rxdesc.flags is __le16, other missing endian swaps This is the minimal patch to fix endian mismatches. These are probably bugs on big-endian arches, noops on little endian. jme_rxsum_ok could be improved to directly take a __le16 and change all of the masks/sets to be in little-endian, but has not been done here to keep the patch small. Signed-off-by: Harvey Harrison Signed-off-by: David S. Miller --- drivers/net/jme.c | 21 ++++++++++----------- 1 file changed, 10 insertions(+), 11 deletions(-) diff --git a/drivers/net/jme.c b/drivers/net/jme.c index 81c6cdc3851..665e70d620f 100644 --- a/drivers/net/jme.c +++ b/drivers/net/jme.c @@ -912,23 +912,23 @@ jme_alloc_and_feed_skb(struct jme_adapter *jme, int idx) skb_put(skb, framesize); skb->protocol = eth_type_trans(skb, jme->dev); - if (jme_rxsum_ok(jme, rxdesc->descwb.flags)) + if (jme_rxsum_ok(jme, le16_to_cpu(rxdesc->descwb.flags))) skb->ip_summed = CHECKSUM_UNNECESSARY; else skb->ip_summed = CHECKSUM_NONE; - if (rxdesc->descwb.flags & RXWBFLAG_TAGON) { + if (rxdesc->descwb.flags & cpu_to_le16(RXWBFLAG_TAGON)) { if (jme->vlgrp) { jme->jme_vlan_rx(skb, jme->vlgrp, - le32_to_cpu(rxdesc->descwb.vlan)); + le16_to_cpu(rxdesc->descwb.vlan)); NET_STAT(jme).rx_bytes += 4; } } else { jme->jme_rx(skb); } - if ((le16_to_cpu(rxdesc->descwb.flags) & RXWBFLAG_DEST) == - RXWBFLAG_DEST_MUL) + if ((rxdesc->descwb.flags & cpu_to_le16(RXWBFLAG_DEST)) == + cpu_to_le16(RXWBFLAG_DEST_MUL)) ++(NET_STAT(jme).multicast); jme->dev->last_rx = jiffies; @@ -961,7 +961,7 @@ jme_process_receive(struct jme_adapter *jme, int limit) rxdesc = rxring->desc; rxdesc += i; - if ((rxdesc->descwb.flags & RXWBFLAG_OWN) || + if ((rxdesc->descwb.flags & cpu_to_le16(RXWBFLAG_OWN)) || !(rxdesc->descwb.desccnt & RXWBDCNT_WBCPL)) goto out; @@ -1763,10 +1763,9 @@ jme_expand_header(struct jme_adapter *jme, struct sk_buff *skb) } static int -jme_tx_tso(struct sk_buff *skb, - u16 *mss, u8 *flags) +jme_tx_tso(struct sk_buff *skb, __le16 *mss, u8 *flags) { - *mss = skb_shinfo(skb)->gso_size << TXDESC_MSS_SHIFT; + *mss = cpu_to_le16(skb_shinfo(skb)->gso_size << TXDESC_MSS_SHIFT); if (*mss) { *flags |= TXFLAG_LSEN; @@ -1826,11 +1825,11 @@ jme_tx_csum(struct jme_adapter *jme, struct sk_buff *skb, u8 *flags) } static inline void -jme_tx_vlan(struct sk_buff *skb, u16 *vlan, u8 *flags) +jme_tx_vlan(struct sk_buff *skb, __le16 *vlan, u8 *flags) { if (vlan_tx_tag_present(skb)) { *flags |= TXFLAG_TAGON; - *vlan = vlan_tx_tag_get(skb); + *vlan = cpu_to_le16(vlan_tx_tag_get(skb)); } } From de339c2aa7fea18410b1abeab5674bfbd4073a63 Mon Sep 17 00:00:00 2001 From: Trent Piepho Date: Wed, 19 Nov 2008 15:52:41 -0800 Subject: [PATCH 076/160] phylib: Fix auto-negotiation restart avoidance A previous patch, 51e2a3846eab18711f4eb59cd0a4c33054e2980a, made genphy_config_aneg() not restart aneg by calling genphy_restart_aneg() if the advertisement hadn't changed. But, genphy_restart_aneg() doesn't just restart aneg, it may also *enable* aneg or un-isolate the PHY from the MII (those functions are controlled by the same register). The code to avoid calling genphy_restart_aneg() didn't consider this. So, modify genphy_config_aneg() to also check if the PHY needs to have aneg enabled or be un-isolated before deciding not to restart aneg. This caused a problem with certain Davicom PHYs, as that driver isolates the PHY (why?) before calling genphy_config_aneg() and expects the PHY to be un-isolated by that function. Signed-off-by: Trent Piepho Reported-by: Scott Wood Signed-off-by: David S. Miller --- drivers/net/phy/phy_device.c | 34 +++++++++++++++++++++++----------- 1 file changed, 23 insertions(+), 11 deletions(-) diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c index 8fb1faca883..55bc24b234e 100644 --- a/drivers/net/phy/phy_device.c +++ b/drivers/net/phy/phy_device.c @@ -564,20 +564,32 @@ EXPORT_SYMBOL(genphy_restart_aneg); */ int genphy_config_aneg(struct phy_device *phydev) { - int result = 0; + int result; - if (AUTONEG_ENABLE == phydev->autoneg) { - int result = genphy_config_advert(phydev); + if (AUTONEG_ENABLE != phydev->autoneg) + return genphy_setup_forced(phydev); - if (result < 0) /* error */ - return result; + result = genphy_config_advert(phydev); - /* Only restart aneg if we are advertising something different - * than we were before. */ - if (result > 0) - result = genphy_restart_aneg(phydev); - } else - result = genphy_setup_forced(phydev); + if (result < 0) /* error */ + return result; + + if (result == 0) { + /* Advertisment hasn't changed, but maybe aneg was never on to + * begin with? Or maybe phy was isolated? */ + int ctl = phy_read(phydev, MII_BMCR); + + if (ctl < 0) + return ctl; + + if (!(ctl & BMCR_ANENABLE) || (ctl & BMCR_ISOLATE)) + result = 1; /* do restart aneg */ + } + + /* Only restart aneg if we are advertising something different + * than we were before. */ + if (result > 0) + result = genphy_restart_aneg(phydev); return result; } From b88ed5cc884e4c28d787cf0912b8635376f16a87 Mon Sep 17 00:00:00 2001 From: Harvey Harrison Date: Wed, 19 Nov 2008 15:54:38 -0800 Subject: [PATCH 077/160] net: ipg.c fix bracing on endian swapping rxfd->frag_info is a __le64, IPG_RFI_FRAGLEN is a cpu-endian constant and wants to be outside of the le64_to_cpu. Fixed in multiple places. Also an occurrence where le64_to_cpu was used instead of cpu_to_le64 Signed-off-by: Harvey Harrison Signed-off-by: David S. Miller --- drivers/net/ipg.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/ipg.c b/drivers/net/ipg.c index 7373dafbb3f..059369885be 100644 --- a/drivers/net/ipg.c +++ b/drivers/net/ipg.c @@ -1112,7 +1112,7 @@ static void ipg_nic_rx_free_skb(struct net_device *dev) struct ipg_rx *rxfd = sp->rxd + entry; pci_unmap_single(sp->pdev, - le64_to_cpu(rxfd->frag_info & ~IPG_RFI_FRAGLEN), + le64_to_cpu(rxfd->frag_info) & ~IPG_RFI_FRAGLEN, sp->rx_buf_sz, PCI_DMA_FROMDEVICE); dev_kfree_skb_irq(sp->rx_buff[entry]); sp->rx_buff[entry] = NULL; @@ -1179,7 +1179,7 @@ static int ipg_nic_rx_check_error(struct net_device *dev) */ if (sp->rx_buff[entry]) { pci_unmap_single(sp->pdev, - le64_to_cpu(rxfd->frag_info & ~IPG_RFI_FRAGLEN), + le64_to_cpu(rxfd->frag_info) & ~IPG_RFI_FRAGLEN, sp->rx_buf_sz, PCI_DMA_FROMDEVICE); dev_kfree_skb_irq(sp->rx_buff[entry]); @@ -1246,7 +1246,7 @@ static void ipg_nic_rx_with_start(struct net_device *dev, if (jumbo->found_start) dev_kfree_skb_irq(jumbo->skb); - pci_unmap_single(pdev, le64_to_cpu(rxfd->frag_info & ~IPG_RFI_FRAGLEN), + pci_unmap_single(pdev, le64_to_cpu(rxfd->frag_info) & ~IPG_RFI_FRAGLEN, sp->rx_buf_sz, PCI_DMA_FROMDEVICE); skb_put(skb, sp->rxfrag_size); @@ -1349,7 +1349,7 @@ static int ipg_nic_rx_jumbo(struct net_device *dev) unsigned int entry = curr % IPG_RFDLIST_LENGTH; struct ipg_rx *rxfd = sp->rxd + entry; - if (!(rxfd->rfs & le64_to_cpu(IPG_RFS_RFDDONE))) + if (!(rxfd->rfs & cpu_to_le64(IPG_RFS_RFDDONE))) break; switch (ipg_nic_rx_check_frame_type(dev)) { From 9ca791bbd464d7968db1530b433cc397a4c348c1 Mon Sep 17 00:00:00 2001 From: Henrik Rydberg Date: Wed, 19 Nov 2008 15:36:06 -0800 Subject: [PATCH 078/160] hwmon: applesmc: Add support for iMac 6 Add temperature sensor support for iMac 6. Signed-off-by: Henrik Rydberg Tested-by: Caleb Hyde Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/hwmon/applesmc.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/hwmon/applesmc.c b/drivers/hwmon/applesmc.c index 488e45cd43d..10977b3d201 100644 --- a/drivers/hwmon/applesmc.c +++ b/drivers/hwmon/applesmc.c @@ -128,6 +128,9 @@ static const char* temperature_sensors_sets[][36] = { /* Set 13: iMac 8,1 */ { "TA0P", "TC0D", "TC0H", "TC0P", "TG0D", "TG0H", "TG0P", "TH0P", "TL0P", "TO0P", "TW0P", "Tm0P", "Tp0P", NULL }, +/* Set 14: iMac 6,1 */ + { "TA0P", "TC0D", "TC0H", "TC0P", "TG0D", "TG0H", "TG0P", "TH0P", + "TO0P", "Tp0P", NULL }, }; /* List of keys used to read/write fan speeds */ @@ -1296,6 +1299,8 @@ static __initdata struct dmi_match_data applesmc_dmi_data[] = { { .accelerometer = 1, .light = 1, .temperature_set = 12 }, /* iMac 8: light sensor only, temperature set 13 */ { .accelerometer = 0, .light = 0, .temperature_set = 13 }, +/* iMac 6: light sensor only, temperature set 14 */ + { .accelerometer = 0, .light = 0, .temperature_set = 14 }, }; /* Note that DMI_MATCH(...,"MacBook") will match "MacBookPro1,1". @@ -1353,6 +1358,10 @@ static __initdata struct dmi_system_id applesmc_whitelist[] = { DMI_MATCH(DMI_BOARD_VENDOR, "Apple"), DMI_MATCH(DMI_PRODUCT_NAME, "iMac8") }, &applesmc_dmi_data[13]}, + { applesmc_dmi_match, "Apple iMac 6", { + DMI_MATCH(DMI_BOARD_VENDOR, "Apple"), + DMI_MATCH(DMI_PRODUCT_NAME, "iMac6") }, + &applesmc_dmi_data[14]}, { applesmc_dmi_match, "Apple iMac 5", { DMI_MATCH(DMI_BOARD_VENDOR, "Apple"), DMI_MATCH(DMI_PRODUCT_NAME, "iMac5") }, From cf7b9a1e11993a064f445d332fecf22819b87a5e Mon Sep 17 00:00:00 2001 From: Julien Brunel Date: Wed, 19 Nov 2008 15:36:07 -0800 Subject: [PATCH 079/160] drivers/video: bad error test before a dereference The error test that follows the call to backlight_device_register semms not to concern the right variable. A simplified version of the semantic match that finds this problem is as follows: (http://www.emn.fr/x-info/coccinelle/) // @def0@ expression x; position p0; @@ x@p0 = backlight_device_register(...) @protected@ expression def0.x,E; position def0.p0; position p; statement S; @@ x@p0 ... when != x = E if (!IS_ERR(x) && ...) {<... x@p ...>} else S @unprotected@ expression def0.x; identifier fld; position def0.p0; position p != protected.p; @@ x@p0 ... when != x = E * x@p->fld // Signed-off-by: Julien Brunel Signed-off-by: Julia Lawall Acked-by: Nicolas Ferre Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/atmel_lcdfb.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/video/atmel_lcdfb.c b/drivers/video/atmel_lcdfb.c index f8d0a57a07c..9a577a800db 100644 --- a/drivers/video/atmel_lcdfb.c +++ b/drivers/video/atmel_lcdfb.c @@ -132,7 +132,7 @@ static void init_backlight(struct atmel_lcdfb_info *sinfo) bl = backlight_device_register("backlight", &sinfo->pdev->dev, sinfo, &atmel_lcdc_bl_ops); - if (IS_ERR(sinfo->backlight)) { + if (IS_ERR(bl)) { dev_err(&sinfo->pdev->dev, "error %ld on backlight register\n", PTR_ERR(bl)); return; From e00b4ff7ebf098b11b11be403921c1cf41d9e321 Mon Sep 17 00:00:00 2001 From: Nadia Derbey Date: Wed, 19 Nov 2008 15:36:08 -0800 Subject: [PATCH 080/160] sysvipc: fix the ipc structures initialization A problem was found while reviewing the code after Bugzilla bug http://bugzilla.kernel.org/show_bug.cgi?id=11796. In ipc_addid(), the newly allocated ipc structure is inserted into the ipcs tree (i.e made visible to readers) without locking it. This is not correct since its initialization continues after it has been inserted in the tree. This patch moves the ipc structure lock initialization + locking before the actual insertion. Signed-off-by: Nadia Derbey Reported-by: Clement Calmels Cc: Manfred Spraul Cc: [2.6.27.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- ipc/util.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/ipc/util.c b/ipc/util.c index 49b3ea615dc..361fd1c96fc 100644 --- a/ipc/util.c +++ b/ipc/util.c @@ -266,9 +266,17 @@ int ipc_addid(struct ipc_ids* ids, struct kern_ipc_perm* new, int size) if (ids->in_use >= size) return -ENOSPC; + spin_lock_init(&new->lock); + new->deleted = 0; + rcu_read_lock(); + spin_lock(&new->lock); + err = idr_get_new(&ids->ipcs_idr, new, &id); - if (err) + if (err) { + spin_unlock(&new->lock); + rcu_read_unlock(); return err; + } ids->in_use++; @@ -280,10 +288,6 @@ int ipc_addid(struct ipc_ids* ids, struct kern_ipc_perm* new, int size) ids->seq = 0; new->id = ipc_buildid(id, new->seq); - spin_lock_init(&new->lock); - new->deleted = 0; - rcu_read_lock(); - spin_lock(&new->lock); return id; } From cf7ee554f3a324e98181b0ea249d9d5be3a0acb8 Mon Sep 17 00:00:00 2001 From: Clemens Ladisch Date: Wed, 19 Nov 2008 15:36:10 -0800 Subject: [PATCH 081/160] fbdev: clean the penguin's dirty feet When booting in a direct color mode, the penguin has dirty feet, i.e., some pixels have the wrong color. This is caused by fb_set_logo_directpalette() which does not initialize the last 32 palette entries. Signed-off-by: Clemens Ladisch Acked-by: Geert Uytterhoeven Cc: Krzysztof Helt Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/fbmem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c index 1d5ae39cb27..3c65b0d6761 100644 --- a/drivers/video/fbmem.c +++ b/drivers/video/fbmem.c @@ -230,7 +230,7 @@ static void fb_set_logo_directpalette(struct fb_info *info, greenshift = info->var.green.offset; blueshift = info->var.blue.offset; - for (i = 32; i < logo->clutsize; i++) + for (i = 32; i < 32 + logo->clutsize; i++) palette[i] = i << redshift | i << greenshift | i << blueshift; } From de11defebf00007677fb7ee91d9b089b78786fbb Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 19 Nov 2008 15:36:14 -0800 Subject: [PATCH 082/160] reintroduce accept4 Introduce a new accept4() system call. The addition of this system call matches analogous changes in 2.6.27 (dup3(), evenfd2(), signalfd4(), inotify_init1(), epoll_create1(), pipe2()) which added new system calls that differed from analogous traditional system calls in adding a flags argument that can be used to access additional functionality. The accept4() system call is exactly the same as accept(), except that it adds a flags bit-mask argument. Two flags are initially implemented. (Most of the new system calls in 2.6.27 also had both of these flags.) SOCK_CLOEXEC causes the close-on-exec (FD_CLOEXEC) flag to be enabled for the new file descriptor returned by accept4(). This is a useful security feature to avoid leaking information in a multithreaded program where one thread is doing an accept() at the same time as another thread is doing a fork() plus exec(). More details here: http://udrepper.livejournal.com/20407.html "Secure File Descriptor Handling", Ulrich Drepper). The other flag is SOCK_NONBLOCK, which causes the O_NONBLOCK flag to be enabled on the new open file description created by accept4(). (This flag is merely a convenience, saving the use of additional calls fcntl(F_GETFL) and fcntl (F_SETFL) to achieve the same result. Here's a test program. Works on x86-32. Should work on x86-64, but I (mtk) don't have a system to hand to test with. It tests accept4() with each of the four possible combinations of SOCK_CLOEXEC and SOCK_NONBLOCK set/clear in 'flags', and verifies that the appropriate flags are set on the file descriptor/open file description returned by accept4(). I tested Ulrich's patch in this thread by applying against 2.6.28-rc2, and it passes according to my test program. /* test_accept4.c Copyright (C) 2008, Linux Foundation, written by Michael Kerrisk Licensed under the GNU GPLv2 or later. */ #define _GNU_SOURCE #include #include #include #include #include #include #include #include #define PORT_NUM 33333 #define die(msg) do { perror(msg); exit(EXIT_FAILURE); } while (0) /**********************************************************************/ /* The following is what we need until glibc gets a wrapper for accept4() */ /* Flags for socket(), socketpair(), accept4() */ #ifndef SOCK_CLOEXEC #define SOCK_CLOEXEC O_CLOEXEC #endif #ifndef SOCK_NONBLOCK #define SOCK_NONBLOCK O_NONBLOCK #endif #ifdef __x86_64__ #define SYS_accept4 288 #elif __i386__ #define USE_SOCKETCALL 1 #define SYS_ACCEPT4 18 #else #error "Sorry -- don't know the syscall # on this architecture" #endif static int accept4(int fd, struct sockaddr *sockaddr, socklen_t *addrlen, int flags) { printf("Calling accept4(): flags = %x", flags); if (flags != 0) { printf(" ("); if (flags & SOCK_CLOEXEC) printf("SOCK_CLOEXEC"); if ((flags & SOCK_CLOEXEC) && (flags & SOCK_NONBLOCK)) printf(" "); if (flags & SOCK_NONBLOCK) printf("SOCK_NONBLOCK"); printf(")"); } printf("\n"); #if USE_SOCKETCALL long args[6]; args[0] = fd; args[1] = (long) sockaddr; args[2] = (long) addrlen; args[3] = flags; return syscall(SYS_socketcall, SYS_ACCEPT4, args); #else return syscall(SYS_accept4, fd, sockaddr, addrlen, flags); #endif } /**********************************************************************/ static int do_test(int lfd, struct sockaddr_in *conn_addr, int closeonexec_flag, int nonblock_flag) { int connfd, acceptfd; int fdf, flf, fdf_pass, flf_pass; struct sockaddr_in claddr; socklen_t addrlen; printf("=======================================\n"); connfd = socket(AF_INET, SOCK_STREAM, 0); if (connfd == -1) die("socket"); if (connect(connfd, (struct sockaddr *) conn_addr, sizeof(struct sockaddr_in)) == -1) die("connect"); addrlen = sizeof(struct sockaddr_in); acceptfd = accept4(lfd, (struct sockaddr *) &claddr, &addrlen, closeonexec_flag | nonblock_flag); if (acceptfd == -1) { perror("accept4()"); close(connfd); return 0; } fdf = fcntl(acceptfd, F_GETFD); if (fdf == -1) die("fcntl:F_GETFD"); fdf_pass = ((fdf & FD_CLOEXEC) != 0) == ((closeonexec_flag & SOCK_CLOEXEC) != 0); printf("Close-on-exec flag is %sset (%s); ", (fdf & FD_CLOEXEC) ? "" : "not ", fdf_pass ? "OK" : "failed"); flf = fcntl(acceptfd, F_GETFL); if (flf == -1) die("fcntl:F_GETFD"); flf_pass = ((flf & O_NONBLOCK) != 0) == ((nonblock_flag & SOCK_NONBLOCK) !=0); printf("nonblock flag is %sset (%s)\n", (flf & O_NONBLOCK) ? "" : "not ", flf_pass ? "OK" : "failed"); close(acceptfd); close(connfd); printf("Test result: %s\n", (fdf_pass && flf_pass) ? "PASS" : "FAIL"); return fdf_pass && flf_pass; } static int create_listening_socket(int port_num) { struct sockaddr_in svaddr; int lfd; int optval; memset(&svaddr, 0, sizeof(struct sockaddr_in)); svaddr.sin_family = AF_INET; svaddr.sin_addr.s_addr = htonl(INADDR_ANY); svaddr.sin_port = htons(port_num); lfd = socket(AF_INET, SOCK_STREAM, 0); if (lfd == -1) die("socket"); optval = 1; if (setsockopt(lfd, SOL_SOCKET, SO_REUSEADDR, &optval, sizeof(optval)) == -1) die("setsockopt"); if (bind(lfd, (struct sockaddr *) &svaddr, sizeof(struct sockaddr_in)) == -1) die("bind"); if (listen(lfd, 5) == -1) die("listen"); return lfd; } int main(int argc, char *argv[]) { struct sockaddr_in conn_addr; int lfd; int port_num; int passed; passed = 1; port_num = (argc > 1) ? atoi(argv[1]) : PORT_NUM; memset(&conn_addr, 0, sizeof(struct sockaddr_in)); conn_addr.sin_family = AF_INET; conn_addr.sin_addr.s_addr = htonl(INADDR_LOOPBACK); conn_addr.sin_port = htons(port_num); lfd = create_listening_socket(port_num); if (!do_test(lfd, &conn_addr, 0, 0)) passed = 0; if (!do_test(lfd, &conn_addr, SOCK_CLOEXEC, 0)) passed = 0; if (!do_test(lfd, &conn_addr, 0, SOCK_NONBLOCK)) passed = 0; if (!do_test(lfd, &conn_addr, SOCK_CLOEXEC, SOCK_NONBLOCK)) passed = 0; close(lfd); exit(passed ? EXIT_SUCCESS : EXIT_FAILURE); } [mtk.manpages@gmail.com: rewrote changelog, updated test program] Signed-off-by: Ulrich Drepper Tested-by: Michael Kerrisk Acked-by: Michael Kerrisk Cc: Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/x86/include/asm/unistd_64.h | 4 +- include/linux/net.h | 6 +-- include/linux/syscalls.h | 3 +- kernel/sys_ni.c | 2 +- net/compat.c | 50 ++------------------ net/socket.c | 80 ++++---------------------------- 6 files changed, 21 insertions(+), 124 deletions(-) diff --git a/arch/x86/include/asm/unistd_64.h b/arch/x86/include/asm/unistd_64.h index 834b2c1d89f..d2e415e6666 100644 --- a/arch/x86/include/asm/unistd_64.h +++ b/arch/x86/include/asm/unistd_64.h @@ -639,8 +639,8 @@ __SYSCALL(__NR_fallocate, sys_fallocate) __SYSCALL(__NR_timerfd_settime, sys_timerfd_settime) #define __NR_timerfd_gettime 287 __SYSCALL(__NR_timerfd_gettime, sys_timerfd_gettime) -#define __NR_paccept 288 -__SYSCALL(__NR_paccept, sys_paccept) +#define __NR_accept4 288 +__SYSCALL(__NR_accept4, sys_accept4) #define __NR_signalfd4 289 __SYSCALL(__NR_signalfd4, sys_signalfd4) #define __NR_eventfd2 290 diff --git a/include/linux/net.h b/include/linux/net.h index 6dc14a24004..4515efae4c3 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -40,7 +40,7 @@ #define SYS_GETSOCKOPT 15 /* sys_getsockopt(2) */ #define SYS_SENDMSG 16 /* sys_sendmsg(2) */ #define SYS_RECVMSG 17 /* sys_recvmsg(2) */ -#define SYS_PACCEPT 18 /* sys_paccept(2) */ +#define SYS_ACCEPT4 18 /* sys_accept4(2) */ typedef enum { SS_FREE = 0, /* not allocated */ @@ -100,7 +100,7 @@ enum sock_type { * remaining bits are used as flags. */ #define SOCK_TYPE_MASK 0xf -/* Flags for socket, socketpair, paccept */ +/* Flags for socket, socketpair, accept4 */ #define SOCK_CLOEXEC O_CLOEXEC #ifndef SOCK_NONBLOCK #define SOCK_NONBLOCK O_NONBLOCK @@ -223,8 +223,6 @@ extern int sock_map_fd(struct socket *sock, int flags); extern struct socket *sockfd_lookup(int fd, int *err); #define sockfd_put(sock) fput(sock->file) extern int net_ratelimit(void); -extern long do_accept(int fd, struct sockaddr __user *upeer_sockaddr, - int __user *upeer_addrlen, int flags); #define net_random() random32() #define net_srandom(seed) srandom32((__force u32)seed) diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index d6ff145919c..04fb47bfb92 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -410,8 +410,7 @@ asmlinkage long sys_getsockopt(int fd, int level, int optname, asmlinkage long sys_bind(int, struct sockaddr __user *, int); asmlinkage long sys_connect(int, struct sockaddr __user *, int); asmlinkage long sys_accept(int, struct sockaddr __user *, int __user *); -asmlinkage long sys_paccept(int, struct sockaddr __user *, int __user *, - const __user sigset_t *, size_t, int); +asmlinkage long sys_accept4(int, struct sockaddr __user *, int __user *, int); asmlinkage long sys_getsockname(int, struct sockaddr __user *, int __user *); asmlinkage long sys_getpeername(int, struct sockaddr __user *, int __user *); asmlinkage long sys_send(int, void __user *, size_t, unsigned); diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index a77b27b11b0..e14a2328170 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -31,7 +31,7 @@ cond_syscall(sys_socketpair); cond_syscall(sys_bind); cond_syscall(sys_listen); cond_syscall(sys_accept); -cond_syscall(sys_paccept); +cond_syscall(sys_accept4); cond_syscall(sys_connect); cond_syscall(sys_getsockname); cond_syscall(sys_getpeername); diff --git a/net/compat.c b/net/compat.c index 6ce1a1cadcc..a3a2ba0fac0 100644 --- a/net/compat.c +++ b/net/compat.c @@ -725,7 +725,7 @@ EXPORT_SYMBOL(compat_mc_getsockopt); static unsigned char nas[19]={AL(0),AL(3),AL(3),AL(3),AL(2),AL(3), AL(3),AL(3),AL(4),AL(4),AL(4),AL(6), AL(6),AL(2),AL(5),AL(5),AL(3),AL(3), - AL(6)}; + AL(4)}; #undef AL asmlinkage long compat_sys_sendmsg(int fd, struct compat_msghdr __user *msg, unsigned flags) @@ -738,52 +738,13 @@ asmlinkage long compat_sys_recvmsg(int fd, struct compat_msghdr __user *msg, uns return sys_recvmsg(fd, (struct msghdr __user *)msg, flags | MSG_CMSG_COMPAT); } -asmlinkage long compat_sys_paccept(int fd, struct sockaddr __user *upeer_sockaddr, - int __user *upeer_addrlen, - const compat_sigset_t __user *sigmask, - compat_size_t sigsetsize, int flags) -{ - compat_sigset_t ss32; - sigset_t ksigmask, sigsaved; - int ret; - - if (sigmask) { - if (sigsetsize != sizeof(compat_sigset_t)) - return -EINVAL; - if (copy_from_user(&ss32, sigmask, sizeof(ss32))) - return -EFAULT; - sigset_from_compat(&ksigmask, &ss32); - - sigdelsetmask(&ksigmask, sigmask(SIGKILL)|sigmask(SIGSTOP)); - sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved); - } - - ret = do_accept(fd, upeer_sockaddr, upeer_addrlen, flags); - - if (ret == -ERESTARTNOHAND) { - /* - * Don't restore the signal mask yet. Let do_signal() deliver - * the signal on the way back to userspace, before the signal - * mask is restored. - */ - if (sigmask) { - memcpy(¤t->saved_sigmask, &sigsaved, - sizeof(sigsaved)); - set_restore_sigmask(); - } - } else if (sigmask) - sigprocmask(SIG_SETMASK, &sigsaved, NULL); - - return ret; -} - asmlinkage long compat_sys_socketcall(int call, u32 __user *args) { int ret; u32 a[6]; u32 a0, a1; - if (call < SYS_SOCKET || call > SYS_PACCEPT) + if (call < SYS_SOCKET || call > SYS_ACCEPT4) return -EINVAL; if (copy_from_user(a, args, nas[call])) return -EFAULT; @@ -804,7 +765,7 @@ asmlinkage long compat_sys_socketcall(int call, u32 __user *args) ret = sys_listen(a0, a1); break; case SYS_ACCEPT: - ret = do_accept(a0, compat_ptr(a1), compat_ptr(a[2]), 0); + ret = sys_accept4(a0, compat_ptr(a1), compat_ptr(a[2]), 0); break; case SYS_GETSOCKNAME: ret = sys_getsockname(a0, compat_ptr(a1), compat_ptr(a[2])); @@ -844,9 +805,8 @@ asmlinkage long compat_sys_socketcall(int call, u32 __user *args) case SYS_RECVMSG: ret = compat_sys_recvmsg(a0, compat_ptr(a1), a[2]); break; - case SYS_PACCEPT: - ret = compat_sys_paccept(a0, compat_ptr(a1), compat_ptr(a[2]), - compat_ptr(a[3]), a[4], a[5]); + case SYS_ACCEPT4: + ret = sys_accept4(a0, compat_ptr(a1), compat_ptr(a[2]), a[3]); break; default: ret = -EINVAL; diff --git a/net/socket.c b/net/socket.c index 57550c3bcab..92764d83689 100644 --- a/net/socket.c +++ b/net/socket.c @@ -1426,8 +1426,8 @@ asmlinkage long sys_listen(int fd, int backlog) * clean when we restucture accept also. */ -long do_accept(int fd, struct sockaddr __user *upeer_sockaddr, - int __user *upeer_addrlen, int flags) +asmlinkage long sys_accept4(int fd, struct sockaddr __user *upeer_sockaddr, + int __user *upeer_addrlen, int flags) { struct socket *sock, *newsock; struct file *newfile; @@ -1510,66 +1510,10 @@ out_fd: goto out_put; } -#if 0 -#ifdef HAVE_SET_RESTORE_SIGMASK -asmlinkage long sys_paccept(int fd, struct sockaddr __user *upeer_sockaddr, - int __user *upeer_addrlen, - const sigset_t __user *sigmask, - size_t sigsetsize, int flags) -{ - sigset_t ksigmask, sigsaved; - int ret; - - if (sigmask) { - /* XXX: Don't preclude handling different sized sigset_t's. */ - if (sigsetsize != sizeof(sigset_t)) - return -EINVAL; - if (copy_from_user(&ksigmask, sigmask, sizeof(ksigmask))) - return -EFAULT; - - sigdelsetmask(&ksigmask, sigmask(SIGKILL)|sigmask(SIGSTOP)); - sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved); - } - - ret = do_accept(fd, upeer_sockaddr, upeer_addrlen, flags); - - if (ret < 0 && signal_pending(current)) { - /* - * Don't restore the signal mask yet. Let do_signal() deliver - * the signal on the way back to userspace, before the signal - * mask is restored. - */ - if (sigmask) { - memcpy(¤t->saved_sigmask, &sigsaved, - sizeof(sigsaved)); - set_restore_sigmask(); - } - } else if (sigmask) - sigprocmask(SIG_SETMASK, &sigsaved, NULL); - - return ret; -} -#else -asmlinkage long sys_paccept(int fd, struct sockaddr __user *upeer_sockaddr, - int __user *upeer_addrlen, - const sigset_t __user *sigmask, - size_t sigsetsize, int flags) -{ - /* The platform does not support restoring the signal mask in the - * return path. So we do not allow using paccept() with a signal - * mask. */ - if (sigmask) - return -EINVAL; - - return do_accept(fd, upeer_sockaddr, upeer_addrlen, flags); -} -#endif -#endif - asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr, int __user *upeer_addrlen) { - return do_accept(fd, upeer_sockaddr, upeer_addrlen, 0); + return sys_accept4(fd, upeer_sockaddr, upeer_addrlen, 0); } /* @@ -2096,7 +2040,7 @@ static const unsigned char nargs[19]={ AL(0),AL(3),AL(3),AL(3),AL(2),AL(3), AL(3),AL(3),AL(4),AL(4),AL(4),AL(6), AL(6),AL(2),AL(5),AL(5),AL(3),AL(3), - AL(6) + AL(4) }; #undef AL @@ -2115,7 +2059,7 @@ asmlinkage long sys_socketcall(int call, unsigned long __user *args) unsigned long a0, a1; int err; - if (call < 1 || call > SYS_PACCEPT) + if (call < 1 || call > SYS_ACCEPT4) return -EINVAL; /* copy_from_user should be SMP safe. */ @@ -2143,9 +2087,8 @@ asmlinkage long sys_socketcall(int call, unsigned long __user *args) err = sys_listen(a0, a1); break; case SYS_ACCEPT: - err = - do_accept(a0, (struct sockaddr __user *)a1, - (int __user *)a[2], 0); + err = sys_accept4(a0, (struct sockaddr __user *)a1, + (int __user *)a[2], 0); break; case SYS_GETSOCKNAME: err = @@ -2192,12 +2135,9 @@ asmlinkage long sys_socketcall(int call, unsigned long __user *args) case SYS_RECVMSG: err = sys_recvmsg(a0, (struct msghdr __user *)a1, a[2]); break; - case SYS_PACCEPT: - err = - sys_paccept(a0, (struct sockaddr __user *)a1, - (int __user *)a[2], - (const sigset_t __user *) a[3], - a[4], a[5]); + case SYS_ACCEPT4: + err = sys_accept4(a0, (struct sockaddr __user *)a1, + (int __user *)a[2], a[3]); break; default: err = -EINVAL; From f8b2256e9c11a825899345de06b39a4bdf44911d Mon Sep 17 00:00:00 2001 From: David Miller Date: Wed, 19 Nov 2008 15:36:15 -0800 Subject: [PATCH 083/160] sparc64: wire up accept4() This adds the sparc syscall hookups. Signed-off-by: David S. Miller Cc: Ulrich Drepper Cc: Michael Kerrisk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/sparc/include/asm/unistd_32.h | 3 ++- arch/sparc/include/asm/unistd_64.h | 3 ++- arch/sparc/kernel/systbls.S | 2 +- arch/sparc64/kernel/sys32.S | 13 ++++++++++++- arch/sparc64/kernel/systbls.S | 4 ++-- 5 files changed, 19 insertions(+), 6 deletions(-) diff --git a/arch/sparc/include/asm/unistd_32.h b/arch/sparc/include/asm/unistd_32.h index 648643a9f13..0d13d2a4c76 100644 --- a/arch/sparc/include/asm/unistd_32.h +++ b/arch/sparc/include/asm/unistd_32.h @@ -338,8 +338,9 @@ #define __NR_dup3 320 #define __NR_pipe2 321 #define __NR_inotify_init1 322 +#define __NR_accept4 323 -#define NR_SYSCALLS 323 +#define NR_SYSCALLS 324 /* Sparc 32-bit only has the "setresuid32", "getresuid32" variants, * it never had the plain ones and there is no value to adding those diff --git a/arch/sparc/include/asm/unistd_64.h b/arch/sparc/include/asm/unistd_64.h index c5cc0e05232..fa5d3c0343c 100644 --- a/arch/sparc/include/asm/unistd_64.h +++ b/arch/sparc/include/asm/unistd_64.h @@ -340,8 +340,9 @@ #define __NR_dup3 320 #define __NR_pipe2 321 #define __NR_inotify_init1 322 +#define __NR_accept4 323 -#define NR_SYSCALLS 323 +#define NR_SYSCALLS 324 #ifdef __KERNEL__ #define __ARCH_WANT_IPC_PARSE_VERSION diff --git a/arch/sparc/kernel/systbls.S b/arch/sparc/kernel/systbls.S index e1b9233b90a..7d080758644 100644 --- a/arch/sparc/kernel/systbls.S +++ b/arch/sparc/kernel/systbls.S @@ -81,4 +81,4 @@ sys_call_table: /*305*/ .long sys_set_mempolicy, sys_kexec_load, sys_move_pages, sys_getcpu, sys_epoll_pwait /*310*/ .long sys_utimensat, sys_signalfd, sys_timerfd_create, sys_eventfd, sys_fallocate /*315*/ .long sys_timerfd_settime, sys_timerfd_gettime, sys_signalfd4, sys_eventfd2, sys_epoll_create1 -/*320*/ .long sys_dup3, sys_pipe2, sys_inotify_init1 +/*320*/ .long sys_dup3, sys_pipe2, sys_inotify_init1, sys_accept4 diff --git a/arch/sparc64/kernel/sys32.S b/arch/sparc64/kernel/sys32.S index ade18ba0c68..f061c4dda9e 100644 --- a/arch/sparc64/kernel/sys32.S +++ b/arch/sparc64/kernel/sys32.S @@ -150,7 +150,7 @@ sys32_mmap2: sys32_socketcall: /* %o0=call, %o1=args */ cmp %o0, 1 bl,pn %xcc, do_einval - cmp %o0, 17 + cmp %o0, 18 bg,pn %xcc, do_einval sub %o0, 1, %o0 sllx %o0, 5, %o0 @@ -319,6 +319,15 @@ do_sys_recvmsg: /* compat_sys_recvmsg(int, struct compat_msghdr *, unsigned int) nop nop nop +do_sys_accept4: /* sys_accept4(int, struct sockaddr *, int *, int) */ +63: ldswa [%o1 + 0x0] %asi, %o0 + sethi %hi(sys_accept4), %g1 +64: lduwa [%o1 + 0x8] %asi, %o2 +65: ldswa [%o1 + 0xc] %asi, %o3 + jmpl %g1 + %lo(sys_accept4), %g0 +66: lduwa [%o1 + 0x4] %asi, %o1 + nop + nop .section __ex_table,"a" .align 4 @@ -353,4 +362,6 @@ do_sys_recvmsg: /* compat_sys_recvmsg(int, struct compat_msghdr *, unsigned int) .word 57b, __retl_efault, 58b, __retl_efault .word 59b, __retl_efault, 60b, __retl_efault .word 61b, __retl_efault, 62b, __retl_efault + .word 63b, __retl_efault, 64b, __retl_efault + .word 65b, __retl_efault, 66b, __retl_efault .previous diff --git a/arch/sparc64/kernel/systbls.S b/arch/sparc64/kernel/systbls.S index b2fa4c16363..9fc78cf354b 100644 --- a/arch/sparc64/kernel/systbls.S +++ b/arch/sparc64/kernel/systbls.S @@ -82,7 +82,7 @@ sys_call_table32: .word compat_sys_set_mempolicy, compat_sys_kexec_load, compat_sys_move_pages, sys_getcpu, compat_sys_epoll_pwait /*310*/ .word compat_sys_utimensat, compat_sys_signalfd, sys_timerfd_create, sys_eventfd, compat_sys_fallocate .word compat_sys_timerfd_settime, compat_sys_timerfd_gettime, compat_sys_signalfd4, sys_eventfd2, sys_epoll_create1 -/*320*/ .word sys_dup3, sys_pipe2, sys_inotify_init1 +/*320*/ .word sys_dup3, sys_pipe2, sys_inotify_init1, sys_accept4 #endif /* CONFIG_COMPAT */ @@ -156,4 +156,4 @@ sys_call_table: .word sys_set_mempolicy, sys_kexec_load, sys_move_pages, sys_getcpu, sys_epoll_pwait /*310*/ .word sys_utimensat, sys_signalfd, sys_timerfd_create, sys_eventfd, sys_fallocate .word sys_timerfd_settime, sys_timerfd_gettime, sys_signalfd4, sys_eventfd2, sys_epoll_create1 -/*320*/ .word sys_dup3, sys_pipe2, sys_inotify_init1 +/*320*/ .word sys_dup3, sys_pipe2, sys_inotify_init1, sys_accept4 From 1c207f952a3a39f9d07a3e9de0cf6b9a9464eeaa Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 19 Nov 2008 15:36:16 -0800 Subject: [PATCH 084/160] Documentation/kernel-parameters.txt: add min_addr, fix max_addr Add "min_addr" documentation. For "max_addr", add nn before [KMG] since a number is needed and this is consistent with other uses of [KMG]. Signed-off-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/kernel-parameters.txt | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 9fa6508892c..899650c20cc 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1193,8 +1193,8 @@ and is between 256 and 4096 characters. It is defined in the file it is equivalent to "nosmp", which also disables the IO APIC. - max_addr=[KMG] [KNL,BOOT,ia64] All physical memory greater than or - equal to this physical address is ignored. + max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater than + or equal to this physical address is ignored. max_luns= [SCSI] Maximum number of LUNs to probe. Should be between 1 and 2^32-1. @@ -1294,6 +1294,9 @@ and is between 256 and 4096 characters. It is defined in the file mga= [HW,DRM] + min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this + physical address is ignored. + mminit_loglevel= [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this parameter allows control of the logging verbosity for From 6e8ba729b6332f2a75572e02480936d2b51665aa Mon Sep 17 00:00:00 2001 From: Jarkko Nikula Date: Wed, 19 Nov 2008 15:36:17 -0800 Subject: [PATCH 085/160] gpiolib: extend gpio label column width in debugfs file There are already various drivers having bigger label than 12 bytes. Most of them fit well under 20 bytes but make column width exact so that oversized labels don't mess up output alignment. Signed-off-by: Jarkko Nikula Acked-by: David Brownell Cc: [2.6.26.x, 2.6.26.x, 2.6.27.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/gpio/gpiolib.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c index faa1cc66e9c..82020abc329 100644 --- a/drivers/gpio/gpiolib.c +++ b/drivers/gpio/gpiolib.c @@ -1134,7 +1134,7 @@ static void gpiolib_dbg_show(struct seq_file *s, struct gpio_chip *chip) continue; is_out = test_bit(FLAG_IS_OUT, &gdesc->flags); - seq_printf(s, " gpio-%-3d (%-12s) %s %s", + seq_printf(s, " gpio-%-3d (%-20.20s) %s %s", gpio, gdesc->label, is_out ? "out" : "in ", chip->get From 415d8cfa845ec9dac42e7b354b1f80485805455b Mon Sep 17 00:00:00 2001 From: Jack Steiner Date: Wed, 19 Nov 2008 15:36:18 -0800 Subject: [PATCH 086/160] GRU: fix for debug option Enable -D DEBUG in the GRU Makefile if CONFIG_SGI_GRU_DEBUG is selected. Signed-off-by: Jack Steiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/misc/sgi-gru/Makefile | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/misc/sgi-gru/Makefile b/drivers/misc/sgi-gru/Makefile index d03597a521b..9e9170b3599 100644 --- a/drivers/misc/sgi-gru/Makefile +++ b/drivers/misc/sgi-gru/Makefile @@ -1,3 +1,7 @@ +ifdef CONFIG_SGI_GRU_DEBUG + EXTRA_CFLAGS += -DDEBUG +endif + obj-$(CONFIG_SGI_GRU) := gru.o gru-y := grufile.o grumain.o grufault.o grutlbpurge.o gruprocfs.o grukservices.o From a495a6d35a026826d54a69d01f373681e14f59ac Mon Sep 17 00:00:00 2001 From: Dmitry Baryshkov Date: Wed, 19 Nov 2008 15:36:19 -0800 Subject: [PATCH 087/160] tmiofb: fix compilation with ACCEL disabled Restore support for compiling tmiofb with acceleration disabled. Signed-off-by: Dmitry Baryshkov Cc: Krzysztof Helt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/tmiofb.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/drivers/video/tmiofb.c b/drivers/video/tmiofb.c index 2a380011e9b..7baf2dd12d5 100644 --- a/drivers/video/tmiofb.c +++ b/drivers/video/tmiofb.c @@ -222,6 +222,9 @@ static irqreturn_t tmiofb_irq(int irq, void *__info) unsigned int bbisc = tmio_ioread16(par->lcr + LCR_BBISC); + tmio_iowrite16(bbisc, par->lcr + LCR_BBISC); + +#ifdef CONFIG_FB_TMIO_ACCELL /* * We were in polling mode and now we got correct irq. * Switch back to IRQ-based sync of command FIFO @@ -231,9 +234,6 @@ static irqreturn_t tmiofb_irq(int irq, void *__info) par->use_polling = false; } - tmio_iowrite16(bbisc, par->lcr + LCR_BBISC); - -#ifdef CONFIG_FB_TMIO_ACCELL if (bbisc & 1) wake_up(&par->wait_acc); #endif @@ -938,7 +938,9 @@ static void tmiofb_dump_regs(struct platform_device *dev) static int tmiofb_suspend(struct platform_device *dev, pm_message_t state) { struct fb_info *info = platform_get_drvdata(dev); +#ifdef CONFIG_FB_TMIO_ACCELL struct tmiofb_par *par = info->par; +#endif struct mfd_cell *cell = dev->dev.platform_data; int retval = 0; @@ -950,12 +952,14 @@ static int tmiofb_suspend(struct platform_device *dev, pm_message_t state) info->fbops->fb_sync(info); +#ifdef CONFIG_FB_TMIO_ACCELL /* * The fb should be usable even if interrupts are disabled (and they are * during suspend/resume). Switch temporary to forced polling. */ printk(KERN_INFO "tmiofb: switching to polling\n"); par->use_polling = true; +#endif tmiofb_hw_stop(dev); if (cell->suspend) From f652c521e0bec2e70cf123f47e80117a7e6ed139 Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Wed, 19 Nov 2008 15:36:19 -0800 Subject: [PATCH 088/160] lib/scatterlist.c: fix kunmap() argument in sg_miter_stop() kunmap() takes as argument the struct page that orginally got kmap()'d, however the sg_miter_stop() function passed it the kernel virtual address instead, resulting in weird stuff. Somehow I ended up fixing this bug by accident while looking for a bug in the same area. Reported-by: kerneloops.org Acked-by: Tejun Heo Signed-off-by: Arjan van de Ven Cc: Hugh Dickins Cc: [2.6.27.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- lib/scatterlist.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/scatterlist.c b/lib/scatterlist.c index 8d2688ff135..b7b449dafbe 100644 --- a/lib/scatterlist.c +++ b/lib/scatterlist.c @@ -395,7 +395,7 @@ void sg_miter_stop(struct sg_mapping_iter *miter) WARN_ON(!irqs_disabled()); kunmap_atomic(miter->addr, KM_BIO_SRC_IRQ); } else - kunmap(miter->addr); + kunmap(miter->page); miter->page = NULL; miter->addr = NULL; From 393df744e056ba24e9531d0657d09fc3c7c0dd22 Mon Sep 17 00:00:00 2001 From: Ned Forrester Date: Wed, 19 Nov 2008 15:36:21 -0800 Subject: [PATCH 089/160] pxa2xx_spi: bugfix full duplex dma data corruption Fixes a data corruption bug in pxa2xx_spi.c when operating in full duplex mode with DMA and using buffers that overlap. SPI transmit and receive buffers are allowed to be the same or to overlap. However, this driver fails if such overlap is attempted in DMA mode because it maps the rx and tx buffers in the wrong order. By mapping DMA_FROM_DEVICE (read) before DMA_TO_DEVICE (write), it invalidates the cache before flushing it, thus discarding data which should have been transmitted. The patch corrects the order of mapping. This bug exists in all versions of pxa2xx_spi.c; similar bugs are in the drivers for two other SPI controllers (au1500, imx). A version of this patch has been tested on kernel 2.6.20 using verification of loopback data with: random transfer length, random bits-per-word, random positive offsets (both larger and smaller than transfer length) between the start of the rx and tx buffers, and varying clock rates. Signed-off-by: Ned Forrester Cc: Vernon Sauder Cc: J. Scott Merritt Signed-off-by: David Brownell Cc: [2.6.27.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/spi/pxa2xx_spi.c | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/drivers/spi/pxa2xx_spi.c b/drivers/spi/pxa2xx_spi.c index dae87b1a4c6..cf12f2d84be 100644 --- a/drivers/spi/pxa2xx_spi.c +++ b/drivers/spi/pxa2xx_spi.c @@ -352,21 +352,21 @@ static int map_dma_buffers(struct driver_data *drv_data) } else drv_data->tx_map_len = drv_data->len; - /* Stream map the rx buffer */ - drv_data->rx_dma = dma_map_single(dev, drv_data->rx, - drv_data->rx_map_len, - DMA_FROM_DEVICE); - if (dma_mapping_error(dev, drv_data->rx_dma)) + /* Stream map the tx buffer. Always do DMA_TO_DEVICE first + * so we flush the cache *before* invalidating it, in case + * the tx and rx buffers overlap. + */ + drv_data->tx_dma = dma_map_single(dev, drv_data->tx, + drv_data->tx_map_len, DMA_TO_DEVICE); + if (dma_mapping_error(dev, drv_data->tx_dma)) return 0; - /* Stream map the tx buffer */ - drv_data->tx_dma = dma_map_single(dev, drv_data->tx, - drv_data->tx_map_len, - DMA_TO_DEVICE); - - if (dma_mapping_error(dev, drv_data->tx_dma)) { - dma_unmap_single(dev, drv_data->rx_dma, + /* Stream map the rx buffer */ + drv_data->rx_dma = dma_map_single(dev, drv_data->rx, drv_data->rx_map_len, DMA_FROM_DEVICE); + if (dma_mapping_error(dev, drv_data->rx_dma)) { + dma_unmap_single(dev, drv_data->tx_dma, + drv_data->tx_map_len, DMA_TO_DEVICE); return 0; } From c267fd777a478d74fa8959628538b64088f67fd2 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Bruno=20Pr=C3=A9mont?= Date: Wed, 19 Nov 2008 15:36:23 -0800 Subject: [PATCH 090/160] viafb: fix releasing of /proc/viafb/ subtree MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When unloading viafb module it does not remove it's /proc/viafb/ subtree which causes multiple viafb directories to appear below proc when mobprobing viafb and also lets kernel WARN() on duplicate proc entries: [ 145.458387] WARNING: at /usr/src/linux-2.6.28-rc3-git6/fs/proc/generic.c:551 proc_register+0xe6/0x160() [ 145.458945] proc_dir_entry '/proc/viafb' already registered [ 145.459278] Modules linked in: viafb(+) i2c_algo_bit cfbcopyarea cfbimgblt cfbfillrect snd_hda_intel snd_pcm snd_timer snd soundcore snd_page_alloc sg via_agp agpgart [last unloaded: drm] [ 145.460647] Pid: 1904, comm: modprobe Tainted: G W 2.6.28-rc3-git6 #4 [ 145.461064] Call Trace: [ 145.461248] [] ? dump_stack+0x1/0x80 [ 145.461533] [] warn_slowpath+0x63/0x80 [ 145.461851] [] ? idr_get_empty_slot+0xe9/0x250 [ 145.462186] [] ? ida_get_new_above+0xf0/0x150 [ 145.462528] [] proc_register+0xe6/0x160 [ 145.462827] [] proc_mkdir_mode+0x36/0x50 [ 145.463135] [] proc_mkdir+0xf/0x20 [ 145.463457] [] viafb_init+0x73c/0xc86 [viafb] [ 145.463823] [] ? viafb_init+0x0/0xc86 [viafb] [ 145.464147] [] do_one_initcall+0x2d/0x160 [ 145.464460] [] ? sysfs_add_file+0x13/0x20 [ 145.464786] [] ? vfree+0x21/0x30 [ 145.465049] [] ? load_module+0x1215/0x1500 [ 145.465381] [] ? __alloc_pages_internal+0x95/0x400 [ 145.465755] [] sys_init_module+0x83/0x1a0 [ 145.466065] [] ? sys_read+0x3d/0x70 [ 145.466354] [] sysenter_do_call+0x12/0x25 [ 145.466653] ---[ end trace c84b37826e16748c ]--- Signed-off-by: Bruno Prémont Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/via/viafbdev.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) diff --git a/drivers/video/via/viafbdev.c b/drivers/video/via/viafbdev.c index 0132eae06f5..73ac754ad80 100644 --- a/drivers/video/via/viafbdev.c +++ b/drivers/video/via/viafbdev.c @@ -2036,30 +2036,30 @@ static int viafb_vt1636_proc_write(struct file *file, return count; } -static void viafb_init_proc(struct proc_dir_entry *viafb_entry) +static void viafb_init_proc(struct proc_dir_entry **viafb_entry) { struct proc_dir_entry *entry; - viafb_entry = proc_mkdir("viafb", NULL); + *viafb_entry = proc_mkdir("viafb", NULL); if (viafb_entry) { - entry = create_proc_entry("dvp0", 0, viafb_entry); + entry = create_proc_entry("dvp0", 0, *viafb_entry); if (entry) { entry->owner = THIS_MODULE; entry->read_proc = viafb_dvp0_proc_read; entry->write_proc = viafb_dvp0_proc_write; } - entry = create_proc_entry("dvp1", 0, viafb_entry); + entry = create_proc_entry("dvp1", 0, *viafb_entry); if (entry) { entry->owner = THIS_MODULE; entry->read_proc = viafb_dvp1_proc_read; entry->write_proc = viafb_dvp1_proc_write; } - entry = create_proc_entry("dfph", 0, viafb_entry); + entry = create_proc_entry("dfph", 0, *viafb_entry); if (entry) { entry->owner = THIS_MODULE; entry->read_proc = viafb_dfph_proc_read; entry->write_proc = viafb_dfph_proc_write; } - entry = create_proc_entry("dfpl", 0, viafb_entry); + entry = create_proc_entry("dfpl", 0, *viafb_entry); if (entry) { entry->owner = THIS_MODULE; entry->read_proc = viafb_dfpl_proc_read; @@ -2068,7 +2068,7 @@ static void viafb_init_proc(struct proc_dir_entry *viafb_entry) if (VT1636_LVDS == viaparinfo->chip_info->lvds_chip_info. lvds_chip_name || VT1636_LVDS == viaparinfo->chip_info->lvds_chip_info2.lvds_chip_name) { - entry = create_proc_entry("vt1636", 0, viafb_entry); + entry = create_proc_entry("vt1636", 0, *viafb_entry); if (entry) { entry->owner = THIS_MODULE; entry->read_proc = viafb_vt1636_proc_read; @@ -2087,6 +2087,7 @@ static void viafb_remove_proc(struct proc_dir_entry *viafb_entry) remove_proc_entry("dfpl", viafb_entry); remove_proc_entry("vt1636", viafb_entry); remove_proc_entry("vt1625", viafb_entry); + remove_proc_entry("viafb", NULL); } static int __devinit via_pci_probe(void) @@ -2348,7 +2349,7 @@ static int __devinit via_pci_probe(void) viafbinfo->node, viafbinfo->fix.id, default_var.xres, default_var.yres, default_var.bits_per_pixel); - viafb_init_proc(viaparinfo->proc_entry); + viafb_init_proc(&viaparinfo->proc_entry); viafb_init_dac(IGA2); return 0; } From b3b4dc8840a8fdbe495723d35cd976d781fd42fa Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Wed, 19 Nov 2008 15:36:25 -0800 Subject: [PATCH 091/160] lcd: fix oops if driver only interested in .set_power The LCD driver core calls LCD drivers when either the blanking state or the display mode has changed, but does not make any check to see if the called driver has a .set_mode method. This means if a driver only has a .set_power method then the system will OOPS on changing mode (and with the console semaphore held so you cannot easily see the problem). Fix the problem by ensuring that either callback is valid before use. Signed-off-by: Ben Dooks Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/backlight/lcd.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/drivers/video/backlight/lcd.c b/drivers/video/backlight/lcd.c index 8e1731d3b22..680e57b616c 100644 --- a/drivers/video/backlight/lcd.c +++ b/drivers/video/backlight/lcd.c @@ -42,10 +42,13 @@ static int fb_notifier_callback(struct notifier_block *self, mutex_lock(&ld->ops_lock); if (!ld->ops->check_fb || ld->ops->check_fb(ld, evdata->info)) { - if (event == FB_EVENT_BLANK) - ld->ops->set_power(ld, *(int *)evdata->data); - else - ld->ops->set_mode(ld, evdata->data); + if (event == FB_EVENT_BLANK) { + if (ld->ops->set_power) + ld->ops->set_power(ld, *(int *)evdata->data); + } else { + if (ld->ops->set_mode) + ld->ops->set_mode(ld, evdata->data); + } } mutex_unlock(&ld->ops_lock); return 0; From 3b45d6380c392e402adc460e4ccf7d41e0caf82a Mon Sep 17 00:00:00 2001 From: Andrea Paterniani Date: Wed, 19 Nov 2008 15:36:26 -0800 Subject: [PATCH 092/160] spi_imx: full duplex dma corruption bugfix Fix unsafe order in dma mapping operation: always flush data from the cache *BEFORE* invalidating it, to allow full duplex transfers where the same buffer may be used for both writes and reads. Signed-off-by: Andrea Paterniani Signed-off-by: David Brownell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/spi/spi_imx.c | 45 +++++++++++++++++++++---------------------- 1 file changed, 22 insertions(+), 23 deletions(-) diff --git a/drivers/spi/spi_imx.c b/drivers/spi/spi_imx.c index 61ba147e384..0b4db0ce78d 100644 --- a/drivers/spi/spi_imx.c +++ b/drivers/spi/spi_imx.c @@ -506,20 +506,6 @@ static int map_dma_buffers(struct driver_data *drv_data) if (!IS_DMA_ALIGNED(drv_data->rx) || !IS_DMA_ALIGNED(drv_data->tx)) return -1; - /* NULL rx means write-only transfer and no map needed - since rx DMA will not be used */ - if (drv_data->rx) { - buf = drv_data->rx; - drv_data->rx_dma = dma_map_single( - dev, - buf, - drv_data->len, - DMA_FROM_DEVICE); - if (dma_mapping_error(dev, drv_data->rx_dma)) - return -1; - drv_data->rx_dma_needs_unmap = 1; - } - if (drv_data->tx == NULL) { /* Read only message --> use drv_data->dummy_dma_buf for dummy writes to achive reads */ @@ -533,18 +519,31 @@ static int map_dma_buffers(struct driver_data *drv_data) buf, drv_data->tx_map_len, DMA_TO_DEVICE); - if (dma_mapping_error(dev, drv_data->tx_dma)) { - if (drv_data->rx_dma) { - dma_unmap_single(dev, - drv_data->rx_dma, - drv_data->len, - DMA_FROM_DEVICE); - drv_data->rx_dma_needs_unmap = 0; - } + if (dma_mapping_error(dev, drv_data->tx_dma)) return -1; - } drv_data->tx_dma_needs_unmap = 1; + /* NULL rx means write-only transfer and no map needed + * since rx DMA will not be used */ + if (drv_data->rx) { + buf = drv_data->rx; + drv_data->rx_dma = dma_map_single(dev, + buf, + drv_data->len, + DMA_FROM_DEVICE); + if (dma_mapping_error(dev, drv_data->rx_dma)) { + if (drv_data->tx_dma) { + dma_unmap_single(dev, + drv_data->tx_dma, + drv_data->tx_map_len, + DMA_TO_DEVICE); + drv_data->tx_dma_needs_unmap = 0; + } + return -1; + } + drv_data->rx_dma_needs_unmap = 1; + } + return 0; } From ac97b9f9a2d0b83488e0bbcb8517b229d5c9b142 Mon Sep 17 00:00:00 2001 From: Michael Halcrow Date: Wed, 19 Nov 2008 15:36:28 -0800 Subject: [PATCH 093/160] eCryptfs: Allocate up to two scatterlists for crypto ops on keys I have received some reports of out-of-memory errors on some older AMD architectures. These errors are what I would expect to see if crypt_stat->key were split between two separate pages. eCryptfs should not assume that any of the memory sent through virt_to_scatterlist() is all contained in a single page, and so this patch allocates two scatterlist structs instead of one when processing keys. I have received confirmation from one person affected by this bug that this patch resolves the issue for him, and so I am submitting it for inclusion in a future stable release. Note that virt_to_scatterlist() runs sg_init_table() on the scatterlist structs passed to it, so the calls to sg_init_table() in decrypt_passphrase_encrypted_session_key() are redundant. Signed-off-by: Michael Halcrow Reported-by: Paulo J. S. Silva Cc: "Leon Woestenberg" Cc: Tim Gardner Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/ecryptfs/keystore.c | 31 ++++++++++++++----------------- 1 file changed, 14 insertions(+), 17 deletions(-) diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c index e22bc396134..0d713b69194 100644 --- a/fs/ecryptfs/keystore.c +++ b/fs/ecryptfs/keystore.c @@ -1037,17 +1037,14 @@ static int decrypt_passphrase_encrypted_session_key(struct ecryptfs_auth_tok *auth_tok, struct ecryptfs_crypt_stat *crypt_stat) { - struct scatterlist dst_sg; - struct scatterlist src_sg; + struct scatterlist dst_sg[2]; + struct scatterlist src_sg[2]; struct mutex *tfm_mutex; struct blkcipher_desc desc = { .flags = CRYPTO_TFM_REQ_MAY_SLEEP }; int rc = 0; - sg_init_table(&dst_sg, 1); - sg_init_table(&src_sg, 1); - if (unlikely(ecryptfs_verbosity > 0)) { ecryptfs_printk( KERN_DEBUG, "Session key encryption key (size [%d]):\n", @@ -1066,8 +1063,8 @@ decrypt_passphrase_encrypted_session_key(struct ecryptfs_auth_tok *auth_tok, } rc = virt_to_scatterlist(auth_tok->session_key.encrypted_key, auth_tok->session_key.encrypted_key_size, - &src_sg, 1); - if (rc != 1) { + src_sg, 2); + if (rc < 1 || rc > 2) { printk(KERN_ERR "Internal error whilst attempting to convert " "auth_tok->session_key.encrypted_key to scatterlist; " "expected rc = 1; got rc = [%d]. " @@ -1079,8 +1076,8 @@ decrypt_passphrase_encrypted_session_key(struct ecryptfs_auth_tok *auth_tok, auth_tok->session_key.encrypted_key_size; rc = virt_to_scatterlist(auth_tok->session_key.decrypted_key, auth_tok->session_key.decrypted_key_size, - &dst_sg, 1); - if (rc != 1) { + dst_sg, 2); + if (rc < 1 || rc > 2) { printk(KERN_ERR "Internal error whilst attempting to convert " "auth_tok->session_key.decrypted_key to scatterlist; " "expected rc = 1; got rc = [%d]\n", rc); @@ -1096,7 +1093,7 @@ decrypt_passphrase_encrypted_session_key(struct ecryptfs_auth_tok *auth_tok, rc = -EINVAL; goto out; } - rc = crypto_blkcipher_decrypt(&desc, &dst_sg, &src_sg, + rc = crypto_blkcipher_decrypt(&desc, dst_sg, src_sg, auth_tok->session_key.encrypted_key_size); mutex_unlock(tfm_mutex); if (unlikely(rc)) { @@ -1539,8 +1536,8 @@ write_tag_3_packet(char *dest, size_t *remaining_bytes, size_t i; size_t encrypted_session_key_valid = 0; char session_key_encryption_key[ECRYPTFS_MAX_KEY_BYTES]; - struct scatterlist dst_sg; - struct scatterlist src_sg; + struct scatterlist dst_sg[2]; + struct scatterlist src_sg[2]; struct mutex *tfm_mutex = NULL; u8 cipher_code; size_t packet_size_length; @@ -1619,8 +1616,8 @@ write_tag_3_packet(char *dest, size_t *remaining_bytes, ecryptfs_dump_hex(session_key_encryption_key, 16); } rc = virt_to_scatterlist(crypt_stat->key, key_rec->enc_key_size, - &src_sg, 1); - if (rc != 1) { + src_sg, 2); + if (rc < 1 || rc > 2) { ecryptfs_printk(KERN_ERR, "Error generating scatterlist " "for crypt_stat session key; expected rc = 1; " "got rc = [%d]. key_rec->enc_key_size = [%d]\n", @@ -1629,8 +1626,8 @@ write_tag_3_packet(char *dest, size_t *remaining_bytes, goto out; } rc = virt_to_scatterlist(key_rec->enc_key, key_rec->enc_key_size, - &dst_sg, 1); - if (rc != 1) { + dst_sg, 2); + if (rc < 1 || rc > 2) { ecryptfs_printk(KERN_ERR, "Error generating scatterlist " "for crypt_stat encrypted session key; " "expected rc = 1; got rc = [%d]. " @@ -1651,7 +1648,7 @@ write_tag_3_packet(char *dest, size_t *remaining_bytes, rc = 0; ecryptfs_printk(KERN_DEBUG, "Encrypting [%d] bytes of the key\n", crypt_stat->key_size); - rc = crypto_blkcipher_encrypt(&desc, &dst_sg, &src_sg, + rc = crypto_blkcipher_encrypt(&desc, dst_sg, src_sg, (*key_rec).enc_key_size); mutex_unlock(tfm_mutex); if (rc) { From f481891fdc49d3d1b8a9674a1825d183069a805f Mon Sep 17 00:00:00 2001 From: Miao Xie Date: Wed, 19 Nov 2008 15:36:30 -0800 Subject: [PATCH 094/160] cpuset: update top cpuset's mems after adding a node After adding a node into the machine, top cpuset's mems isn't updated. By reviewing the code, we found that the update function cpuset_track_online_nodes() was invoked after node_states[N_ONLINE] changes. It is wrong because N_ONLINE just means node has pgdat, and if node has/added memory, we use N_HIGH_MEMORY. So, We should invoke the update function after node_states[N_HIGH_MEMORY] changes, just like its commit says. This patch fixes it. And we use notifier of memory hotplug instead of direct calling of cpuset_track_online_nodes(). Signed-off-by: Miao Xie Acked-by: Yasunori Goto Cc: David Rientjes Cc: Paul Menage Signed-off-by: Linus Torvalds --- include/linux/cpuset.h | 4 ---- kernel/cpuset.c | 19 ++++++++++++++++--- mm/memory_hotplug.c | 3 --- 3 files changed, 16 insertions(+), 10 deletions(-) diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h index 2691926fb50..8e540d32c9f 100644 --- a/include/linux/cpuset.h +++ b/include/linux/cpuset.h @@ -74,8 +74,6 @@ static inline int cpuset_do_slab_mem_spread(void) return current->flags & PF_SPREAD_SLAB; } -extern void cpuset_track_online_nodes(void); - extern int current_cpuset_is_being_rebound(void); extern void rebuild_sched_domains(void); @@ -151,8 +149,6 @@ static inline int cpuset_do_slab_mem_spread(void) return 0; } -static inline void cpuset_track_online_nodes(void) {} - static inline int current_cpuset_is_being_rebound(void) { return 0; diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 81fc6791a29..da7ff6137f3 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -36,6 +36,7 @@ #include #include #include +#include #include #include #include @@ -2015,12 +2016,23 @@ static int cpuset_track_online_cpus(struct notifier_block *unused_nb, * Call this routine anytime after node_states[N_HIGH_MEMORY] changes. * See also the previous routine cpuset_track_online_cpus(). */ -void cpuset_track_online_nodes(void) +static int cpuset_track_online_nodes(struct notifier_block *self, + unsigned long action, void *arg) { cgroup_lock(); - top_cpuset.mems_allowed = node_states[N_HIGH_MEMORY]; - scan_for_empty_cpusets(&top_cpuset); + switch (action) { + case MEM_ONLINE: + top_cpuset.mems_allowed = node_states[N_HIGH_MEMORY]; + break; + case MEM_OFFLINE: + top_cpuset.mems_allowed = node_states[N_HIGH_MEMORY]; + scan_for_empty_cpusets(&top_cpuset); + break; + default: + break; + } cgroup_unlock(); + return NOTIFY_OK; } #endif @@ -2036,6 +2048,7 @@ void __init cpuset_init_smp(void) top_cpuset.mems_allowed = node_states[N_HIGH_MEMORY]; hotcpu_notifier(cpuset_track_online_cpus, 0); + hotplug_memory_notifier(cpuset_track_online_nodes, 10); } /** diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 6837a101437..b5b2b15085a 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -22,7 +22,6 @@ #include #include #include -#include #include #include #include @@ -498,8 +497,6 @@ int add_memory(int nid, u64 start, u64 size) /* we online node here. we can't roll back from here. */ node_set_online(nid); - cpuset_track_online_nodes(); - if (new_pgdat) { ret = register_one_node(nid); /* From f011c2dae6cffc50ef67d9bd937b488ba5db8913 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Wed, 19 Nov 2008 15:36:32 -0800 Subject: [PATCH 095/160] mm: vmalloc allocator off by one Fix off by one bug in the KVA allocator that can leave gaps in the address space. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/vmalloc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/vmalloc.c b/mm/vmalloc.c index ba6b0f5f7fa..46aab4dbf61 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -362,7 +362,7 @@ retry: goto found; } - while (addr + size >= first->va_start && addr + size <= vend) { + while (addr + size > first->va_start && addr + size <= vend) { addr = ALIGN(first->va_end + PAGE_SIZE, align); n = rb_next(&first->rb_node); From 496850e5f5a372029ceb2b35c811770a9bb073b6 Mon Sep 17 00:00:00 2001 From: Nick Piggin Date: Wed, 19 Nov 2008 15:36:33 -0800 Subject: [PATCH 096/160] mm: vmalloc failure flush fix An initial vmalloc failure should start off a synchronous flush of lazy areas, in case someone is in progress flushing them already, which could cause us to return an allocation failure even if there is plenty of KVA free. Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/vmalloc.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 46aab4dbf61..04f5e320e74 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -521,6 +521,17 @@ static void __purge_vmap_area_lazy(unsigned long *start, unsigned long *end, spin_unlock(&purge_lock); } +/* + * Kick off a purge of the outstanding lazy areas. Don't bother if somebody + * is already purging. + */ +static void try_purge_vmap_area_lazy(void) +{ + unsigned long start = ULONG_MAX, end = 0; + + __purge_vmap_area_lazy(&start, &end, 0, 0); +} + /* * Kick off a purge of the outstanding lazy areas. */ @@ -528,7 +539,7 @@ static void purge_vmap_area_lazy(void) { unsigned long start = ULONG_MAX, end = 0; - __purge_vmap_area_lazy(&start, &end, 0, 0); + __purge_vmap_area_lazy(&start, &end, 1, 0); } /* @@ -539,7 +550,7 @@ static void free_unmap_vmap_area(struct vmap_area *va) va->flags |= VM_LAZY_FREE; atomic_add((va->va_end - va->va_start) >> PAGE_SHIFT, &vmap_lazy_nr); if (unlikely(atomic_read(&vmap_lazy_nr) > lazy_max_pages())) - purge_vmap_area_lazy(); + try_purge_vmap_area_lazy(); } static struct vmap_area *find_vmap_area(unsigned long addr) From 0ae15132a4f5c758a6ffcde74495641dc3f62ba1 Mon Sep 17 00:00:00 2001 From: Glauber Costa Date: Wed, 19 Nov 2008 15:36:33 -0800 Subject: [PATCH 097/160] mm: vmalloc search restart fix Current vmalloc restart search for a free area in case we can't find one. The reason is there are areas which are lazily freed, and could be possibly freed now. However, current implementation start searching the tree from the last failing address, which is pretty much by definition at the end of address space. So, we fail. The proposal of this patch is to restart the search from the beginning of the requested vstart address. This fixes the regression in running KVM virtual machines for me, described in http://lkml.org/lkml/2008/10/28/349, caused by commit db64fe02258f1507e13fe5212a989922323685ce. Signed-off-by: Glauber Costa Signed-off-by: Nick Piggin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/vmalloc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/mm/vmalloc.c b/mm/vmalloc.c index 04f5e320e74..30f826d484f 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -324,14 +324,14 @@ static struct vmap_area *alloc_vmap_area(unsigned long size, BUG_ON(size & ~PAGE_MASK); - addr = ALIGN(vstart, align); - va = kmalloc_node(sizeof(struct vmap_area), gfp_mask & GFP_RECLAIM_MASK, node); if (unlikely(!va)) return ERR_PTR(-ENOMEM); retry: + addr = ALIGN(vstart, align); + spin_lock(&vmap_area_lock); /* XXX: could have a last_hole cache */ n = vmap_area_root.rb_node; From 3fa59dfbc3b223f02c26593be69ce6fc9a940405 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Wed, 19 Nov 2008 15:36:34 -0800 Subject: [PATCH 098/160] cgroup: fix potential deadlock in pre_destroy As Balbir pointed out, memcg's pre_destroy handler has potential deadlock. It has following lock sequence. cgroup_mutex (cgroup_rmdir) -> pre_destroy -> mem_cgroup_pre_destroy-> force_empty -> cpu_hotplug.lock. (lru_add_drain_all-> schedule_work-> get_online_cpus) But, cpuset has following. cpu_hotplug.lock (call notifier) -> cgroup_mutex. (within notifier) Then, this lock sequence should be fixed. Considering how pre_destroy works, it's not necessary to holding cgroup_mutex() while calling it. As a side effect, we don't have to wait at this mutex while memcg's force_empty works.(it can be long when there are tons of pages.) Signed-off-by: KAMEZAWA Hiroyuki Acked-by: Balbir Singh Cc: Li Zefan Cc: Paul Menage Cc: Daisuke Nishimura Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/cgroup.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 358e77564e6..1a06be61dcd 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -2472,10 +2472,7 @@ static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry) mutex_unlock(&cgroup_mutex); return -EBUSY; } - - parent = cgrp->parent; - root = cgrp->root; - sb = root->sb; + mutex_unlock(&cgroup_mutex); /* * Call pre_destroy handlers of subsys. Notify subsystems @@ -2483,7 +2480,14 @@ static int cgroup_rmdir(struct inode *unused_dir, struct dentry *dentry) */ cgroup_call_pre_destroy(cgrp); - if (cgroup_has_css_refs(cgrp)) { + mutex_lock(&cgroup_mutex); + parent = cgrp->parent; + root = cgrp->root; + sb = root->sb; + + if (atomic_read(&cgrp->count) + || !list_empty(&cgrp->children) + || cgroup_has_css_refs(cgrp)) { mutex_unlock(&cgroup_mutex); return -EBUSY; } From 966c8c12dc9e77f931e2281ba25d2f0244b06949 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 19 Nov 2008 15:36:36 -0800 Subject: [PATCH 099/160] sprint_symbol(): use less stack sprint_symbol(), itself used when dumping stacks, has been wasting 128 bytes of stack: lookup the symbol directly into the buffer supplied by the caller, instead of using a locally declared namebuf. I believe the name != buffer strcpy() is obsolete: the design here dates from when module symbol lookup pointed into a supposedly const but sadly volatile table; nowadays it copies, but an uncalled strcpy() looks better here than the risk of a recursive BUG_ON(). Signed-off-by: Hugh Dickins Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/kallsyms.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c index 5072cf1685a..7b8b0f21a5b 100644 --- a/kernel/kallsyms.c +++ b/kernel/kallsyms.c @@ -304,17 +304,24 @@ int sprint_symbol(char *buffer, unsigned long address) char *modname; const char *name; unsigned long offset, size; - char namebuf[KSYM_NAME_LEN]; + int len; - name = kallsyms_lookup(address, &size, &offset, &modname, namebuf); + name = kallsyms_lookup(address, &size, &offset, &modname, buffer); if (!name) return sprintf(buffer, "0x%lx", address); + if (name != buffer) + strcpy(buffer, name); + len = strlen(buffer); + buffer += len; + if (modname) - return sprintf(buffer, "%s+%#lx/%#lx [%s]", name, offset, - size, modname); + len += sprintf(buffer, "+%#lx/%#lx [%s]", + offset, size, modname); else - return sprintf(buffer, "%s+%#lx/%#lx", name, offset, size); + len += sprintf(buffer, "+%#lx/%#lx", offset, size); + + return len; } /* Look up a kernel symbol and print it to the kernel messages. */ From bda8550deed96687f29992d711a88ea21cff4d26 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 19 Nov 2008 15:36:36 -0800 Subject: [PATCH 100/160] migration: fix writepage error Page migration's writeout() has got understandably confused by the nasty AOP_WRITEPAGE_ACTIVATE case: as in normal success, a writepage() error has unlocked the page, so writeout() then needs to relock it. Signed-off-by: Hugh Dickins Cc: KAMEZAWA Hiroyuki Cc: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/migrate.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/mm/migrate.c b/mm/migrate.c index 385db89f0c3..1e0d6b237f4 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -522,15 +522,12 @@ static int writeout(struct address_space *mapping, struct page *page) remove_migration_ptes(page, page); rc = mapping->a_ops->writepage(page, &wbc); - if (rc < 0) - /* I/O Error writing */ - return -EIO; if (rc != AOP_WRITEPAGE_ACTIVATE) /* unlocked. Relock */ lock_page(page); - return -EAGAIN; + return (rc < 0) ? -EIO : -EAGAIN; } /* From 63eb6b93ce725e4c5f38fc85dd703d49465b03cb Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 19 Nov 2008 15:36:37 -0800 Subject: [PATCH 101/160] vmscan: let GFP_NOFS go to swap again In the past, GFP_NOFS (but of course not GFP_NOIO) was allowed to reclaim by writing to swap. That got partially broken in 2.6.23, when may_enter_fs initialization was moved up before the allocation of swap, so its PageSwapCache test was failing the first time around, Fix it by setting may_enter_fs when add_to_swap() succeeds with __GFP_IO. In fact, check __GFP_IO before calling add_to_swap(): allocating swap we're not ready to use just increases disk seeking. Signed-off-by: Hugh Dickins Cc: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/vmscan.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/mm/vmscan.c b/mm/vmscan.c index c141b3e7807..f83a7ed5c6c 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -623,6 +623,8 @@ static unsigned long shrink_page_list(struct list_head *page_list, * Try to allocate it some swap space here. */ if (PageAnon(page) && !PageSwapCache(page)) { + if (!(sc->gfp_mask & __GFP_IO)) + goto keep_locked; switch (try_to_munlock(page)) { case SWAP_FAIL: /* shouldn't happen */ case SWAP_AGAIN: @@ -634,6 +636,7 @@ static unsigned long shrink_page_list(struct list_head *page_list, } if (!add_to_swap(page, GFP_ATOMIC)) goto activate_locked; + may_enter_fs = 1; } #endif /* CONFIG_SWAP */ From f9454548e17cd56bad081bd7d55a09b001950cbb Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 19 Nov 2008 15:36:38 -0800 Subject: [PATCH 102/160] don't unlink an active swapfile Peter Cordes is sorry that he rm'ed his swapfiles while they were in use, he then had no pathname to swapoff. It's a curious little oversight, but not one worth a lot of hackery. Kudos to Willy Tarreau for turning this around from a discussion of synthetic pathnames to how to prevent unlink. Mimic immutable: prohibit unlinking an active swapfile in may_delete() (and don't worry my little head over the tiny race window). Signed-off-by: Hugh Dickins Cc: Willy Tarreau Acked-by: Christoph Hellwig Cc: Peter Cordes Cc: Bodo Eggert <7eggert@gmx.de> Cc: David Newall Cc: Peter Zijlstra Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/namei.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/namei.c b/fs/namei.c index 09ce58e49e7..d34e0f9681c 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1378,7 +1378,7 @@ static int may_delete(struct inode *dir,struct dentry *victim,int isdir) if (IS_APPEND(dir)) return -EPERM; if (check_sticky(dir, victim->d_inode)||IS_APPEND(victim->d_inode)|| - IS_IMMUTABLE(victim->d_inode)) + IS_IMMUTABLE(victim->d_inode) || IS_SWAPFILE(victim->d_inode)) return -EPERM; if (isdir) { if (!S_ISDIR(victim->d_inode->i_mode)) From c0d861afa5c986f7fe23647fbe411cd300f7c927 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Wed, 19 Nov 2008 15:36:41 -0800 Subject: [PATCH 103/160] drivers/video/backlight/da903x.c: introduce missing kfree Error handling code following a kzalloc should free the allocated data. The semantic match that finds the problem is as follows: (http://www.emn.fr/x-info/coccinelle/) // @r exists@ local idexpression x; statement S; expression E; identifier f,l; position p1,p2; expression *ptr != NULL; @@ ( if ((x@p1 = \(kmalloc\|kzalloc\|kcalloc\)(...)) == NULL) S | x@p1 = \(kmalloc\|kzalloc\|kcalloc\)(...); ... if (x == NULL) S ) <... when != x when != if (...) { <+...x...+> } x->f = E ...> ( return \(0\|<+...x...+>\|ptr\); | return@p2 ...; ) @script:python@ p1 << r.p1; p2 << r.p2; @@ print "* file: %s kmalloc %s return %s" % (p1[0].file,p1[0].line,p2[0].line) // Signed-off-by: Julia Lawall Cc: Mike Rapoport Cc: Richard Purdie Cc: Eric Miao Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/backlight/da903x.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/video/backlight/da903x.c b/drivers/video/backlight/da903x.c index 242c3825016..75388b95943 100644 --- a/drivers/video/backlight/da903x.c +++ b/drivers/video/backlight/da903x.c @@ -119,6 +119,7 @@ static int da903x_backlight_probe(struct platform_device *pdev) default: dev_err(&pdev->dev, "invalid backlight device ID(%d)\n", pdev->id); + kfree(data); return -EINVAL; } From 45a3a36b6333b9ddfb0c57602d959a8d05bc47e4 Mon Sep 17 00:00:00 2001 From: Henrik Rydberg Date: Wed, 19 Nov 2008 15:36:42 -0800 Subject: [PATCH 104/160] drivers/hwmon/applesmc.c: add generic MacPro support In order to analyze the SMC of the newer MacPros, applesmc needs to recognize the machine. This patch adds the missing generic dmi_match entry for MacPro models. Signed-off-by: Henrik Rydberg Cc: Nicolas Boichat Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/hwmon/applesmc.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/hwmon/applesmc.c b/drivers/hwmon/applesmc.c index 10977b3d201..f7dce8b9f64 100644 --- a/drivers/hwmon/applesmc.c +++ b/drivers/hwmon/applesmc.c @@ -1354,6 +1354,10 @@ static __initdata struct dmi_system_id applesmc_whitelist[] = { DMI_MATCH(DMI_BOARD_VENDOR,"Apple"), DMI_MATCH(DMI_PRODUCT_NAME,"MacPro2") }, &applesmc_dmi_data[4]}, + { applesmc_dmi_match, "Apple MacPro", { + DMI_MATCH(DMI_BOARD_VENDOR, "Apple"), + DMI_MATCH(DMI_PRODUCT_NAME, "MacPro") }, + &applesmc_dmi_data[4]}, { applesmc_dmi_match, "Apple iMac 8", { DMI_MATCH(DMI_BOARD_VENDOR, "Apple"), DMI_MATCH(DMI_PRODUCT_NAME, "iMac8") }, From 8e324c19f57cc05922e9cc0f338b58108da45539 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Wed, 19 Nov 2008 15:36:43 -0800 Subject: [PATCH 105/160] MAINTAINERS: change email address for rostedt I find that I answer my email quicker on my home email account, than I do on my work email. Not to mention that I never check my work email while traveling. Please change my email address in the MAINTAINERS file from srostedt@redhat.com to rostedt@goodmis.org. Signed-off-by: Steven Rostedt Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 627e4c89328..618c1ef4a39 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1809,7 +1809,7 @@ S: Maintained FTRACE P: Steven Rostedt -M: srostedt@redhat.com +M: rostedt@goodmis.org S: Maintained FUJITSU FR-V (FRV) PORT From 00d8089c54867053a5aae062b765f257ca419e27 Mon Sep 17 00:00:00 2001 From: Rik van Riel Date: Wed, 19 Nov 2008 15:36:44 -0800 Subject: [PATCH 106/160] vmscan: fix get_scan_ratio() comment Fix the old comment on the scan ratio calculations. Signed-off-by: Rik van Riel Cc: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/vmscan.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/mm/vmscan.c b/mm/vmscan.c index f83a7ed5c6c..7ea1440b53d 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1389,9 +1389,9 @@ static void get_scan_ratio(struct zone *zone, struct scan_control *sc, file_prio = 200 - sc->swappiness; /* - * anon recent_rotated[0] - * %anon = 100 * ----------- / ----------------- * IO cost - * anon + file rotate_sum + * The amount of pressure on anon vs file pages is inversely + * proportional to the fraction of recently scanned pages on + * each list that were recently referenced and in active use. */ ap = (anon_prio + 1) * (zone->recent_scanned[0] + 1); ap /= zone->recent_rotated[0] + 1; From ee11940f8e7a2f064af22d52180cb5f9643eef61 Mon Sep 17 00:00:00 2001 From: Vlada Peric Date: Wed, 19 Nov 2008 15:36:45 -0800 Subject: [PATCH 107/160] cirrusfb: remove unused variables MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit After commit a1d35a7a (cirrusfb: use modedb and add mode_option parameter), these variables are no longer used, so remove them to fix compilation warning. Signed-off-by: Vlada Periæ Acked-by: Krzysztof Helt Cc: Stephen Rothwell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/cirrusfb.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/video/cirrusfb.c b/drivers/video/cirrusfb.c index 8a8760230bc..a2aa6ddffbe 100644 --- a/drivers/video/cirrusfb.c +++ b/drivers/video/cirrusfb.c @@ -2462,8 +2462,7 @@ static int __init cirrusfb_init(void) #ifndef MODULE static int __init cirrusfb_setup(char *options) { - char *this_opt, s[32]; - int i; + char *this_opt; DPRINTK("ENTER\n"); From ea7e743e49b94749fc739baaf160809ed279aeda Mon Sep 17 00:00:00 2001 From: WANG Cong Date: Wed, 19 Nov 2008 15:36:46 -0800 Subject: [PATCH 108/160] hostfs: fix a duplicated global function name fs/hostfs/hostfs_user.c defines do_readlink() as non-static, and so does fs/xfs/linux-2.6/xfs_ioctl.c when CONFIG_XFS_DEBUG=y. So rename do_readlink() in hostfs to hostfs_do_readlink(). I think it's better if XFS guys will also rename their do_readlink(), it's not necessary to use such a general name. Signed-off-by: WANG Cong Cc: Jeff Dike Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/hostfs/hostfs.h | 2 +- fs/hostfs/hostfs_kern.c | 4 ++-- fs/hostfs/hostfs_user.c | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) diff --git a/fs/hostfs/hostfs.h b/fs/hostfs/hostfs.h index 6ae9011b95e..2f34f8f2134 100644 --- a/fs/hostfs/hostfs.h +++ b/fs/hostfs/hostfs.h @@ -81,7 +81,7 @@ extern int do_rmdir(const char *file); extern int do_mknod(const char *file, int mode, unsigned int major, unsigned int minor); extern int link_file(const char *from, const char *to); -extern int do_readlink(char *file, char *buf, int size); +extern int hostfs_do_readlink(char *file, char *buf, int size); extern int rename_file(char *from, char *to); extern int do_statfs(char *root, long *bsize_out, long long *blocks_out, long long *bfree_out, long long *bavail_out, diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index 7f34f4385de..3a31451ac17 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -168,7 +168,7 @@ static char *follow_link(char *link) if (name == NULL) goto out; - n = do_readlink(link, name, len); + n = hostfs_do_readlink(link, name, len); if (n < len) break; len *= 2; @@ -943,7 +943,7 @@ int hostfs_link_readpage(struct file *file, struct page *page) name = inode_name(page->mapping->host, 0); if (name == NULL) return -ENOMEM; - err = do_readlink(name, buffer, PAGE_CACHE_SIZE); + err = hostfs_do_readlink(name, buffer, PAGE_CACHE_SIZE); kfree(name); if (err == PAGE_CACHE_SIZE) err = -E2BIG; diff --git a/fs/hostfs/hostfs_user.c b/fs/hostfs/hostfs_user.c index 53fd0a67c11..b79424f9328 100644 --- a/fs/hostfs/hostfs_user.c +++ b/fs/hostfs/hostfs_user.c @@ -377,7 +377,7 @@ int link_file(const char *to, const char *from) return 0; } -int do_readlink(char *file, char *buf, int size) +int hostfs_do_readlink(char *file, char *buf, int size) { int n; From 33d283bef23132c48195eafc21449f8ba88fce6b Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Wed, 19 Nov 2008 15:36:48 -0800 Subject: [PATCH 109/160] cgroups: fix a serious bug in cgroupstats Try this, and you'll get oops immediately: # cd Documentation/accounting/ # gcc -o getdelays getdelays.c # mount -t cgroup -o debug xxx /mnt # ./getdelays -C /mnt/tasks Because a normal file's dentry->d_fsdata is a pointer to struct cftype, not struct cgroup. After the patch, it returns EINVAL if we try to get cgroupstats from a normal file. Cc: Balbir Singh Signed-off-by: Li Zefan Acked-by: Paul Menage Cc: [2.6.25.x, 2.6.26.x, 2.6.27.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/cgroup.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 1a06be61dcd..fe00b3b983a 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -2039,10 +2039,13 @@ int cgroupstats_build(struct cgroupstats *stats, struct dentry *dentry) struct cgroup *cgrp; struct cgroup_iter it; struct task_struct *tsk; + /* - * Validate dentry by checking the superblock operations + * Validate dentry by checking the superblock operations, + * and make sure it's a directory. */ - if (dentry->d_sb->s_op != &cgroup_ops) + if (dentry->d_sb->s_op != &cgroup_ops || + !S_ISDIR(dentry->d_inode->i_mode)) goto err; ret = 0; From f55491a4bcbe8bab337bc00830ca12d703ea2613 Mon Sep 17 00:00:00 2001 From: Mike Rapoport <[mailto:mike@compulab.co.il]> Date: Wed, 19 Nov 2008 15:36:49 -0800 Subject: [PATCH 110/160] drivers/video/backlight/da903x.c: introduce one more missing kfree One more error handling code should have kfree as well Signed-off-by: Mike Rapoport Acked-by: Eric Miao Cc: Richard Purdie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/video/backlight/da903x.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/video/backlight/da903x.c b/drivers/video/backlight/da903x.c index 75388b95943..93bb4340cc6 100644 --- a/drivers/video/backlight/da903x.c +++ b/drivers/video/backlight/da903x.c @@ -131,6 +131,7 @@ static int da903x_backlight_probe(struct platform_device *pdev) data, &da903x_backlight_ops); if (IS_ERR(bl)) { dev_err(&pdev->dev, "failed to register backlight\n"); + kfree(data); return PTR_ERR(bl); } From 06b0d4dc14a44dd9b57321c24f7eeb10b345abd8 Mon Sep 17 00:00:00 2001 From: "Stanley.Miao" Date: Wed, 19 Nov 2008 15:36:50 -0800 Subject: [PATCH 111/160] W1 OMAP: Fix OMAP LDP boot crash OMAP LDP boot crash. This is because w1 subsystem changed the search interface, so update omap_hdq's search interface to follow the change. Signed-off-by: Stanley.Miao Signed-off-by: Evgeniy Polyakov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/w1/masters/omap_hdq.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/w1/masters/omap_hdq.c b/drivers/w1/masters/omap_hdq.c index 1295625c482..c973889110c 100644 --- a/drivers/w1/masters/omap_hdq.c +++ b/drivers/w1/masters/omap_hdq.c @@ -86,8 +86,8 @@ static struct platform_driver omap_hdq_driver = { static u8 omap_w1_read_byte(void *_hdq); static void omap_w1_write_byte(void *_hdq, u8 byte); static u8 omap_w1_reset_bus(void *_hdq); -static void omap_w1_search_bus(void *_hdq, u8 search_type, - w1_slave_found_callback slave_found); +static void omap_w1_search_bus(void *_hdq, struct w1_master *master_dev, + u8 search_type, w1_slave_found_callback slave_found); static struct w1_bus_master omap_w1_master = { @@ -231,8 +231,8 @@ static u8 omap_w1_reset_bus(void *_hdq) } /* W1 search callback function */ -static void omap_w1_search_bus(void *_hdq, u8 search_type, - w1_slave_found_callback slave_found) +static void omap_w1_search_bus(void *_hdq, struct w1_master *master_dev, + u8 search_type, w1_slave_found_callback slave_found) { u64 module_id, rn_le, cs, id; @@ -249,7 +249,7 @@ static void omap_w1_search_bus(void *_hdq, u8 search_type, cs = w1_calc_crc8((u8 *)&rn_le, 7); id = (cs << 56) | module_id; - slave_found(_hdq, id); + slave_found(master_dev, id); } static int _omap_hdq_reset(struct hdq_data *hdq_data) From 9f92f4719764acf1c9185a5958200887a43e3483 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Wed, 19 Nov 2008 15:36:51 -0800 Subject: [PATCH 112/160] cciss: fix DEBUG printk formats Fix printk format warnings when CCISS_DEBUG is defined. drivers/block/cciss.c:2856: warning: format '%d' expects type 'int', but argument 2 has type 'long unsigned int' drivers/block/cciss.c:3205: warning: format '%x' expects type 'unsigned int', but argument 2 has type 'long unsigned int' drivers/block/cciss.c:3236: warning: format '%x' expects type 'unsigned int', but argument 2 has type '__u64' drivers/block/cciss.c:3246: warning: format '%x' expects type 'unsigned int', but argument 2 has type '__u64' Signed-off-by: Randy Dunlap Cc: Mike Miller Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- drivers/block/cciss.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c index 12de1fdaa6c..9364dc55425 100644 --- a/drivers/block/cciss.c +++ b/drivers/block/cciss.c @@ -2847,7 +2847,7 @@ static void do_cciss_request(struct request_queue *q) h->maxSG = seg; #ifdef CCISS_DEBUG - printk(KERN_DEBUG "cciss: Submitting %d sectors in %d segments\n", + printk(KERN_DEBUG "cciss: Submitting %lu sectors in %d segments\n", creq->nr_sectors, seg); #endif /* CCISS_DEBUG */ @@ -3197,7 +3197,7 @@ static int __devinit cciss_pci_init(ctlr_info_t *c, struct pci_dev *pdev) c->paddr = pci_resource_start(pdev, 0); /* addressing mode bits already removed */ #ifdef CCISS_DEBUG - printk("address 0 = %x\n", c->paddr); + printk("address 0 = %lx\n", c->paddr); #endif /* CCISS_DEBUG */ c->vaddr = remap_pci_mem(c->paddr, 0x250); @@ -3224,7 +3224,8 @@ static int __devinit cciss_pci_init(ctlr_info_t *c, struct pci_dev *pdev) #endif /* CCISS_DEBUG */ cfg_base_addr_index = find_PCI_BAR_index(pdev, cfg_base_addr); #ifdef CCISS_DEBUG - printk("cfg base address index = %x\n", cfg_base_addr_index); + printk("cfg base address index = %llx\n", + (unsigned long long)cfg_base_addr_index); #endif /* CCISS_DEBUG */ if (cfg_base_addr_index == -1) { printk(KERN_WARNING "cciss: Cannot find cfg_base_addr_index\n"); @@ -3234,7 +3235,7 @@ static int __devinit cciss_pci_init(ctlr_info_t *c, struct pci_dev *pdev) cfg_offset = readl(c->vaddr + SA5_CTMEM_OFFSET); #ifdef CCISS_DEBUG - printk("cfg offset = %x\n", cfg_offset); + printk("cfg offset = %llx\n", (unsigned long long)cfg_offset); #endif /* CCISS_DEBUG */ c->cfgtable = remap_pci_mem(pci_resource_start(pdev, cfg_base_addr_index) + From b09bc6cbae4dd3a2d35722668ef2c502a7b8b093 Mon Sep 17 00:00:00 2001 From: Andiry Xu Date: Fri, 14 Nov 2008 11:42:29 +0800 Subject: [PATCH 113/160] USB: fix SB700 usb subsystem hang bug This patch is required for AMD SB700 south bridge revision A12 and A13 to avoid USB subsystem hang symptom. The USB subsystem hang symptom is observed when the system has multiple USB devices connected to it. In some cases a USB hub may be required to observe this symptom. This patch works around the problem by correcting the internal register setting that will help by changing the behavior of the internal logic to avoid the USB subsystem hang issue. The change in the behavior of the logic does not impact the normal operation of the USB subsystem. Reported-by: Volker Armin Hemmann Tested-by: Volker Armin Hemmann Signed-off-by: Andiry Xu Signed-off-by: Libin Yang Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/ehci-pci.c | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+) diff --git a/drivers/usb/host/ehci-pci.c b/drivers/usb/host/ehci-pci.c index c46a58f9181..9d0ea573aef 100644 --- a/drivers/usb/host/ehci-pci.c +++ b/drivers/usb/host/ehci-pci.c @@ -66,6 +66,8 @@ static int ehci_pci_setup(struct usb_hcd *hcd) { struct ehci_hcd *ehci = hcd_to_ehci(hcd); struct pci_dev *pdev = to_pci_dev(hcd->self.controller); + struct pci_dev *p_smbus; + u8 rev; u32 temp; int retval; @@ -166,6 +168,25 @@ static int ehci_pci_setup(struct usb_hcd *hcd) pci_write_config_byte(pdev, 0x4b, tmp | 0x20); } break; + case PCI_VENDOR_ID_ATI: + /* SB700 old version has a bug in EHCI controller, + * which causes usb devices lose response in some cases. + */ + if (pdev->device == 0x4396) { + p_smbus = pci_get_device(PCI_VENDOR_ID_ATI, + PCI_DEVICE_ID_ATI_SBX00_SMBUS, + NULL); + if (!p_smbus) + break; + rev = p_smbus->revision; + if ((rev == 0x3a) || (rev == 0x3b)) { + u8 tmp; + pci_read_config_byte(pdev, 0x53, &tmp); + pci_write_config_byte(pdev, 0x53, tmp | (1<<3)); + } + pci_dev_put(p_smbus); + } + break; } ehci_reset(ehci); From aa5cbbecd903e5692b64f871c385ece1c5508eac Mon Sep 17 00:00:00 2001 From: Felipe Balbi Date: Mon, 17 Nov 2008 09:08:16 +0200 Subject: [PATCH 114/160] usb: musb: fix bug in musb_schedule This bug was introduced recently. Fix it before bigger problems appear. Signed-off-by: Felipe Balbi Cc: Sergei Shtylyov Signed-off-by: Greg Kroah-Hartman --- drivers/usb/musb/musb_host.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/musb/musb_host.c b/drivers/usb/musb/musb_host.c index e45e70bcc5e..cc64462d4c4 100644 --- a/drivers/usb/musb/musb_host.c +++ b/drivers/usb/musb/musb_host.c @@ -1757,7 +1757,7 @@ static int musb_schedule( } } /* use bulk reserved ep1 if no other ep is free */ - if (best_end > 0 && qh->type == USB_ENDPOINT_XFER_BULK) { + if (best_end < 0 && qh->type == USB_ENDPOINT_XFER_BULK) { hw_ep = musb->bulk_ep; if (is_in) head = &musb->in_bulk; From 9beba53dc5c330d781ecc0ad8ea081c2d100ff9f Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Mon, 17 Nov 2008 16:12:32 -0500 Subject: [PATCH 115/160] USB: storage: updates unusual_devs entry for the Nokia 6300 This patch (as1169) modifies the unusual_devs entry for the Nokia 6300. According to Maciej Gierok and David McBride , the revision limits need to be wider. This fixes Bugzilla #11768. Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman --- drivers/usb/storage/unusual_devs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index d4e5fc86e43..096a439baa3 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -240,7 +240,7 @@ UNUSUAL_DEV( 0x0421, 0x04b9, 0x0551, 0x0551, US_FL_FIX_CAPACITY ), /* Reported by Richard Nauber */ -UNUSUAL_DEV( 0x0421, 0x04fa, 0x0601, 0x0601, +UNUSUAL_DEV( 0x0421, 0x04fa, 0x0550, 0x0660, "Nokia", "6300", US_SC_DEVICE, US_PR_DEVICE, NULL, From 589afd3bec907f02c133d7b8185b8af534f14a8e Mon Sep 17 00:00:00 2001 From: Alan Stern Date: Mon, 17 Nov 2008 14:32:16 -0500 Subject: [PATCH 116/160] USB: storage: update unusual_devs entries for Nokia 5300 and 5310 This patch (as1168) updates the unusual_devs entry for the Nokia 5300. According to Jorge Lucangeli Obes , some existing models have a revision number lower than the lower limit of the current entry. The patch also moves the entry for the Nokia 5310 to its correct place in the file. Signed-off-by: Alan Stern Signed-off-by: Greg Kroah-Hartman --- drivers/usb/storage/unusual_devs.h | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index 096a439baa3..6da9a7a962a 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -167,6 +167,13 @@ UNUSUAL_DEV( 0x0421, 0x005d, 0x0001, 0x0600, US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_FIX_CAPACITY ), +/* Patch for Nokia 5310 capacity */ +UNUSUAL_DEV( 0x0421, 0x006a, 0x0000, 0x0591, + "Nokia", + "5310", + US_SC_DEVICE, US_PR_DEVICE, NULL, + US_FL_FIX_CAPACITY ), + /* Reported by Mario Rettig */ UNUSUAL_DEV( 0x0421, 0x042e, 0x0100, 0x0100, "Nokia", @@ -233,7 +240,7 @@ UNUSUAL_DEV( 0x0421, 0x0495, 0x0370, 0x0370, US_FL_MAX_SECTORS_64 ), /* Reported by Cedric Godin */ -UNUSUAL_DEV( 0x0421, 0x04b9, 0x0551, 0x0551, +UNUSUAL_DEV( 0x0421, 0x04b9, 0x0500, 0x0551, "Nokia", "5300", US_SC_DEVICE, US_PR_DEVICE, NULL, From 9c264521a9f836541c122b00f505cfd60cc5bbb5 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Sat, 15 Nov 2008 19:53:21 -0800 Subject: [PATCH 117/160] USB: gadget rndis: stop windows self-immolation MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Somewhere in the conversion of the RNDIS gadget code to the new framework, the descriptor of its data interface seems to have been copied from the CDC Ethernet driver. Unfortunately that means it got a nonzero altsetting ... which is incorrect. Issue uncovered by Richard Röjfors . This patch fixes that problem, and resolves at least some cases of Windows XP bluescreening itself. Tested-by: Richard Röjfors . Signed-off-by: David Brownell Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/f_rndis.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/usb/gadget/f_rndis.c b/drivers/usb/gadget/f_rndis.c index 659b3d9671c..8afb14a4a72 100644 --- a/drivers/usb/gadget/f_rndis.c +++ b/drivers/usb/gadget/f_rndis.c @@ -172,7 +172,6 @@ static struct usb_interface_descriptor rndis_data_intf __initdata = { .bDescriptorType = USB_DT_INTERFACE, /* .bInterfaceNumber = DYNAMIC */ - .bAlternateSetting = 1, .bNumEndpoints = 2, .bInterfaceClass = USB_CLASS_CDC_DATA, .bInterfaceSubClass = 0, From ff3495052af48f7a2bf7961b131dc9e161dae19c Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Richard=20R=C3=B6jfors?= Date: Sat, 15 Nov 2008 19:53:24 -0800 Subject: [PATCH 118/160] USB: gadget rndis: send notifications MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit It turns out that atomic_inc_return() returns the *new* value not the original one, so the logic in rndis_response_available() kept the first RNDIS response notification from getting out. This prevented interoperation with MS-Windows (but not Linux). Fix this to make RNDIS behave again. Signed-off-by: Richard Röjfors Signed-off-by: David Brownell Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/f_rndis.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/gadget/f_rndis.c b/drivers/usb/gadget/f_rndis.c index 8afb14a4a72..428b5993575 100644 --- a/drivers/usb/gadget/f_rndis.c +++ b/drivers/usb/gadget/f_rndis.c @@ -302,7 +302,7 @@ static void rndis_response_available(void *_rndis) __le32 *data = req->buf; int status; - if (atomic_inc_return(&rndis->notify_count)) + if (atomic_inc_return(&rndis->notify_count) != 1) return; /* Send RNDIS RESPONSE_AVAILABLE notification; a From f1c0a2a3aff53698f4855968d576464041d49b39 Mon Sep 17 00:00:00 2001 From: Pete Zaitcev Date: Fri, 14 Nov 2008 09:47:41 -0700 Subject: [PATCH 119/160] USB: usbmon: fix read(2) There's a bug in the usbmon binary reader: When using read() to fetch the packets and a packet's data is partially read, the next read call will once again return up to len_cap bytes of data. The b_read counter is not regarded when determining the remaining chunk size. So, when dumping USB data with "cat /dev/usbmon0 > usbmon.trace" while reading from a USB storage device and analyzing the dump file afterwards it will get out of sync after a couple of packets. Signed-off-by: Ingo van Lil Signed-off-by: Pete Zaitcev Cc: stable Signed-off-by: Greg Kroah-Hartman --- drivers/usb/mon/mon_bin.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/usb/mon/mon_bin.c b/drivers/usb/mon/mon_bin.c index c9de3f027aa..e06810aef2d 100644 --- a/drivers/usb/mon/mon_bin.c +++ b/drivers/usb/mon/mon_bin.c @@ -687,7 +687,10 @@ static ssize_t mon_bin_read(struct file *file, char __user *buf, } if (rp->b_read >= sizeof(struct mon_bin_hdr)) { - step_len = min(nbytes, (size_t)ep->len_cap); + step_len = ep->len_cap; + step_len -= rp->b_read - sizeof(struct mon_bin_hdr); + if (step_len > nbytes) + step_len = nbytes; offset = rp->b_out + PKT_SIZE; offset += rp->b_read - sizeof(struct mon_bin_hdr); if (offset >= rp->b_size) From 5091b58751d9bf51e92aa9e07f3332dcacc2af1c Mon Sep 17 00:00:00 2001 From: M Kondrin Date: Fri, 14 Nov 2008 13:02:45 +0300 Subject: [PATCH 120/160] USB: ACE1001 patch for cp2101.c The patch which adds IDs for AKTAKOM USB->RS232 cable (http://www.aktakom.ru/product/kio/ace-1001.htm) is attached. From: M Kondrin Signed-off-by: Greg Kroah-Hartman --- drivers/usb/serial/cp2101.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/serial/cp2101.c b/drivers/usb/serial/cp2101.c index 9035d7256b0..cfaf1f08553 100644 --- a/drivers/usb/serial/cp2101.c +++ b/drivers/usb/serial/cp2101.c @@ -56,6 +56,7 @@ static void cp2101_shutdown(struct usb_serial *); static int debug; static struct usb_device_id id_table [] = { + { USB_DEVICE(0x0471, 0x066A) }, /* AKTAKOM ACE-1001 cable */ { USB_DEVICE(0x0489, 0xE000) }, /* Pirelli Broadband S.p.A, DP-L10 SIP/GSM Mobile */ { USB_DEVICE(0x08e6, 0x5501) }, /* Gemalto Prox-PU/CU contactless smartcard reader */ { USB_DEVICE(0x0FCF, 0x1003) }, /* Dynastream ANT development board */ From a134f85c131ffd56720e38af2967ec6265480757 Mon Sep 17 00:00:00 2001 From: Balazs Scheidler Date: Thu, 20 Nov 2008 01:07:24 -0800 Subject: [PATCH 121/160] TPROXY: fill struct flowi->flags in udp_sendmsg() udp_sendmsg() didn't fill struct flowi->flags, which means that the route lookup would fail for non-local IPs even if the IP_TRANSPARENT sockopt was set. This prevents sendto() to work properly for UDP sockets, whereas bind(foreign-ip) + connect() + send() worked fine. Signed-off-by: Balazs Scheidler Signed-off-by: David S. Miller --- net/ipv4/udp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index cf02701ced4..98c1fd09be8 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -633,6 +633,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, .saddr = saddr, .tos = tos } }, .proto = sk->sk_protocol, + .flags = inet_sk_flowi_flags(sk), .uli_u = { .ports = { .sport = inet->sport, .dport = dport } } }; From c82838458200ec4167ce7083b0a17474150c5bf7 Mon Sep 17 00:00:00 2001 From: Balazs Scheidler Date: Thu, 20 Nov 2008 01:08:06 -0800 Subject: [PATCH 122/160] TPROXY: supply a struct flowi->flags argument in inet_sk_rebuild_header() inet_sk_rebuild_header() does a new route lookup if the dst_entry associated with a socket becomes stale. However inet_sk_rebuild_header() didn't use struct flowi->flags, causing the route lookup to fail for foreign-bound IP_TRANSPARENT sockets, causing an error state to be set for the sockets in question. Signed-off-by: Balazs Scheidler Signed-off-by: David S. Miller --- net/ipv4/af_inet.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index 1fbff5fa424..1aa2dc9e380 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -1117,6 +1117,7 @@ int inet_sk_rebuild_header(struct sock *sk) }, }, .proto = sk->sk_protocol, + .flags = inet_sk_flowi_flags(sk), .uli_u = { .ports = { .sport = inet->sport, From fb5e2f9b9410a4362897d12dc1ed4f7cec1b0d45 Mon Sep 17 00:00:00 2001 From: Nobuhiro Iwamatsu Date: Mon, 17 Nov 2008 20:29:58 +0000 Subject: [PATCH 123/160] sh: sh_eth: Update to change of mii_bus Update to change of mii_bus interface and fix some warning. Signed-off-by: Nobuhiro Iwamatsu Signed-off-by: David S. Miller --- drivers/net/sh_eth.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/sh_eth.c b/drivers/net/sh_eth.c index a24bb68887a..59f242a6771 100644 --- a/drivers/net/sh_eth.c +++ b/drivers/net/sh_eth.c @@ -927,7 +927,7 @@ static int sh_eth_start_xmit(struct sk_buff *skb, struct net_device *ndev) struct sh_eth_private *mdp = netdev_priv(ndev); struct sh_eth_txdesc *txdesc; u32 entry; - int flags; + unsigned long flags; spin_lock_irqsave(&mdp->lock, flags); if ((mdp->cur_tx - mdp->dirty_tx) >= (TX_RING_SIZE - 4)) { @@ -1141,7 +1141,7 @@ static int sh_mdio_init(struct net_device *ndev, int id) /* Hook up MII support for ethtool */ mdp->mii_bus->name = "sh_mii"; mdp->mii_bus->parent = &ndev->dev; - mdp->mii_bus->id[0] = id; + snprintf(mdp->mii_bus->id, MII_BUS_ID_SIZE, "%x", id); /* PHY IRQ */ mdp->mii_bus->irq = kmalloc(sizeof(int)*PHY_MAX_ADDR, GFP_KERNEL); From bcb3336ce4354395e38df7e2da110fca61f86a84 Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Tue, 18 Nov 2008 04:28:35 +0000 Subject: [PATCH 124/160] mv643xx_eth: fix the order of mdiobus_{unregister, free}() calls Signed-off-by: Lennert Buytenhek Signed-off-by: David S. Miller --- drivers/net/mv643xx_eth.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c index b9dcdbd369f..baa7e845a01 100644 --- a/drivers/net/mv643xx_eth.c +++ b/drivers/net/mv643xx_eth.c @@ -2435,8 +2435,8 @@ static int mv643xx_eth_shared_remove(struct platform_device *pdev) struct mv643xx_eth_shared_platform_data *pd = pdev->dev.platform_data; if (pd == NULL || pd->shared_smi == NULL) { - mdiobus_free(msp->smi_bus); mdiobus_unregister(msp->smi_bus); + mdiobus_free(msp->smi_bus); } if (msp->err_interrupt != NO_IRQ) free_irq(msp->err_interrupt, msp); From 11b4aa03b212cbaa969df25e723fbd400a6e24b9 Mon Sep 17 00:00:00 2001 From: Lennert Buytenhek Date: Tue, 18 Nov 2008 04:28:58 +0000 Subject: [PATCH 125/160] mv643xx_eth: fix recycle check bound When mv643xx_eth allocates skbuffs, it adds 'dma_get_cache_alignment() - 1' to the length it needs, so that it can align the skb's ->data pointer to a cache boundary. When checking whether a transmitted skbuff can be reused as a receive buffer, these bytes needs to be included into the minimum bound for the recycle check. Signed-off-by: Lennert Buytenhek Signed-off-by: David S. Miller --- drivers/net/mv643xx_eth.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c index baa7e845a01..e513f76f2a9 100644 --- a/drivers/net/mv643xx_eth.c +++ b/drivers/net/mv643xx_eth.c @@ -899,7 +899,8 @@ static int txq_reclaim(struct tx_queue *txq, int budget, int force) if (skb != NULL) { if (skb_queue_len(&mp->rx_recycle) < mp->default_rx_ring_size && - skb_recycle_check(skb, mp->skb_size)) + skb_recycle_check(skb, mp->skb_size + + dma_get_cache_alignment() - 1)) __skb_queue_head(&mp->rx_recycle, skb); else dev_kfree_skb(skb); From a7f75c0c9cfb628512b30795bcba02c8a97e03a0 Mon Sep 17 00:00:00 2001 From: Pantelis Koukousoulas Date: Thu, 20 Nov 2008 01:48:46 -0800 Subject: [PATCH 126/160] asix: Fix asix-based cards connecting to 10/100Mbs LAN. Add AX_MEDIUM_ENCK also when speed = 10/100Mbps. This allows my belkin f5d5055 to work with my 100Mbps switch and with an old 10Mbps ISA card. Without this patch, the card is recognized and the interface is brought up fine, but no packets actually flow through the interface. Signed-off-by: Pantelis Koukousoulas Acked-by: David Hollis Signed-off-by: David S. Miller --- drivers/net/usb/asix.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/usb/asix.c b/drivers/net/usb/asix.c index e12cdb4543b..de57490103f 100644 --- a/drivers/net/usb/asix.c +++ b/drivers/net/usb/asix.c @@ -1102,12 +1102,14 @@ static int ax88178_link_reset(struct usbnet *dev) mode = AX88178_MEDIUM_DEFAULT; if (ecmd.speed == SPEED_1000) - mode |= AX_MEDIUM_GM | AX_MEDIUM_ENCK; + mode |= AX_MEDIUM_GM; else if (ecmd.speed == SPEED_100) mode |= AX_MEDIUM_PS; else mode &= ~(AX_MEDIUM_PS | AX_MEDIUM_GM); + mode |= AX_MEDIUM_ENCK; + if (ecmd.duplex == DUPLEX_FULL) mode |= AX_MEDIUM_FD; else From eafdcb433f854c8cb06747a80fa3df7c72147be6 Mon Sep 17 00:00:00 2001 From: Vernon Sauder Date: Thu, 20 Nov 2008 01:56:08 -0800 Subject: [PATCH 127/160] smc911x: Fix printf format typo in smc911x driver. Signed-off-by: Vernon Sauder Signed-off-by: David S. Miller --- drivers/net/smc911x.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/smc911x.c b/drivers/net/smc911x.c index 1f26ab0e798..b185cd12269 100644 --- a/drivers/net/smc911x.c +++ b/drivers/net/smc911x.c @@ -1813,7 +1813,7 @@ static int __init smc911x_probe(struct net_device *dev) val = SMC_GET_BYTE_TEST(lp); DBG(SMC_DEBUG_MISC, "%s: endian probe returned 0x%04x\n", CARDNAME, val); if (val != 0x87654321) { - printk(KERN_ERR "Invalid chip endian 0x08%x\n",val); + printk(KERN_ERR "Invalid chip endian 0x%08x\n",val); retval = -ENODEV; goto err_out; } From 3aa4614da741f10b09559a5675c79e2eff5cccd8 Mon Sep 17 00:00:00 2001 From: Patrick McHardy Date: Thu, 20 Nov 2008 04:07:14 -0800 Subject: [PATCH 128/160] pkt_sched: fix missing check for packet overrun in qdisc_dump_stab() nla_nest_start() might return NULL, causing a NULL pointer dereference. Signed-off-by: Patrick McHardy Signed-off-by: David S. Miller --- net/sched/sch_api.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c index b16ad2972c6..6ab4a2f92ca 100644 --- a/net/sched/sch_api.c +++ b/net/sched/sch_api.c @@ -417,6 +417,8 @@ static int qdisc_dump_stab(struct sk_buff *skb, struct qdisc_size_table *stab) struct nlattr *nest; nest = nla_nest_start(skb, TCA_STAB); + if (nest == NULL) + goto nla_put_failure; NLA_PUT(skb, TCA_STAB_BASE, sizeof(stab->szopts), &stab->szopts); nla_nest_end(skb, nest); From eedd726efbc439dbed94fb8577e5533a986b341f Mon Sep 17 00:00:00 2001 From: Benjamin Thery Date: Thu, 20 Nov 2008 04:16:12 -0800 Subject: [PATCH 129/160] ipv6: use seq_release_private for ip6mr.c /proc entries In ip6mr.c, /proc entries /proc/net/ip6_mr_cache and /proc/net/ip6_mr_vif are opened with seq_open_private(), thus seq_release_private() should be used to release them. Should fix a small memory leak. Signed-off-by: Benjamin Thery Signed-off-by: David S. Miller --- net/ipv6/ip6mr.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c index 52a7eb0e2c2..0524769632e 100644 --- a/net/ipv6/ip6mr.c +++ b/net/ipv6/ip6mr.c @@ -224,7 +224,7 @@ static struct file_operations ip6mr_vif_fops = { .open = ip6mr_vif_open, .read = seq_read, .llseek = seq_lseek, - .release = seq_release, + .release = seq_release_private, }; static void *ipmr_mfc_seq_start(struct seq_file *seq, loff_t *pos) @@ -338,7 +338,7 @@ static struct file_operations ip6mr_mfc_fops = { .open = ipmr_mfc_open, .read = seq_read, .llseek = seq_lseek, - .release = seq_release, + .release = seq_release_private, }; #endif From a849854fc7e200d5e5a437dec9c1d0e8b105e50f Mon Sep 17 00:00:00 2001 From: Alan Jenkins Date: Thu, 20 Nov 2008 04:18:25 -0800 Subject: [PATCH 130/160] atl2: don't request irq on resume if netif running If the device is suspended with the cable disconnected, then resumed with the cable connected, dev->open is called before resume. During resume, we request an IRQ, but the IRQ was already assigned during dev->open, resulting in the warning shown below. Don't request an IRQ if the device is running. Call Trace: [] warn_on_slowpath+0x40/0x59 [] raw_pci_read+0x4d/0x55 [] pci_read+0x1c/0x21 [] __pci_find_next_cap_ttl+0x44/0x70 [] __pci_find_next_cap+0x1a/0x1f [] pci_find_capability+0x28/0x2c [] pci_msi_check_device+0x53/0x62 [] pci_enable_msi+0x3a/0x1cd [] atl2_write_phy_reg+0x40/0x5f [atl2] [] dma_generic_alloc_coherent+0x0/0xd7 [] atl2_request_irq+0x15/0x49 [atl2] [] atl2_open+0x20b/0x297 [atl2] [] dev_open+0x62/0x91 [] dev_change_flags+0x93/0x141 [] do_setlink+0x238/0x2d5 [] rtnl_setlink+0xa9/0xbf [] mutex_lock+0xb/0x19 [] rtnl_dump_ifinfo+0x0/0x69 [] rtnl_setlink+0x0/0xbf [] rtnetlink_rcv_msg+0x185/0x19f [] sock_rmalloc+0x23/0x57 [] rtnetlink_rcv_msg+0x0/0x19f [] netlink_rcv_skb+0x2d/0x71 [] rtnetlink_rcv+0x14/0x1a [] netlink_unicast+0x184/0x1e4 [] netlink_sendmsg+0x233/0x240 [] sock_sendmsg+0xb7/0xd0 [] autoremove_wake_function+0x0/0x2b [] autoremove_wake_function+0x0/0x2b [] mempool_alloc+0x2d/0x9e [] scsi_pool_alloc_command+0x35/0x4f [] mutex_lock+0xb/0x19 [] unix_stream_recvmsg+0x357/0x3e2 [] copy_from_user+0x23/0x4f [] verify_iovec+0x3e/0x6c [] sys_sendmsg+0x18d/0x1f0 [] sys_recvmsg+0x146/0x1c8 [] sys_recvmsg+0x1b4/0x1c8 [] __wake_up+0xf/0x15 [] netlink_table_ungrab+0x17/0x19 [] copy_to_user+0x25/0x3b [] move_addr_to_user+0x50/0x68 [] sys_getsockname+0x6f/0x9a [] sys_getsockname+0x89/0x9a [] do_wp_page+0x3ae/0x41a [] handle_mm_fault+0x4c5/0x540 [] sys_socketcall+0x176/0x1b0 [] sysenter_do_call+0x12/0x21 Signed-off-by: Alan Jenkins Signed-off-by: Jay Cliburn Signed-off-by: David S. Miller --- drivers/net/atlx/atl2.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/net/atlx/atl2.c b/drivers/net/atlx/atl2.c index f5bdc92c1a6..8571e8c0bc6 100644 --- a/drivers/net/atlx/atl2.c +++ b/drivers/net/atlx/atl2.c @@ -1690,9 +1690,11 @@ static int atl2_resume(struct pci_dev *pdev) ATL2_WRITE_REG(&adapter->hw, REG_WOL_CTRL, 0); - err = atl2_request_irq(adapter); - if (netif_running(netdev) && err) - return err; + if (netif_running(netdev)) { + err = atl2_request_irq(adapter); + if (err) + return err; + } atl2_reset_hw(&adapter->hw); From 5ece6c2ddd6f7da9e95dc325c742c0f5afbcecbe Mon Sep 17 00:00:00 2001 From: Alexey Dobriyan Date: Thu, 20 Nov 2008 04:20:10 -0800 Subject: [PATCH 131/160] net: fix tiny output corruption of /proc/net/snmp6 Because "name" is static, it can be occasionally be filled with somewhat garbage if two processes read /proc/net/snmp6. Also, remove useless casts and "-1" -- snprintf() correctly terminates it's output. Signed-off-by: Alexey Dobriyan Signed-off-by: David S. Miller --- net/ipv6/proc.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/net/ipv6/proc.c b/net/ipv6/proc.c index 07f0b76e742..97c17fdd6f7 100644 --- a/net/ipv6/proc.c +++ b/net/ipv6/proc.c @@ -132,7 +132,7 @@ static struct snmp_mib snmp6_udplite6_list[] = { static void snmp6_seq_show_icmpv6msg(struct seq_file *seq, void **mib) { - static char name[32]; + char name[32]; int i; /* print by name -- deprecated items */ @@ -144,7 +144,7 @@ static void snmp6_seq_show_icmpv6msg(struct seq_file *seq, void **mib) p = icmp6type2name[icmptype]; if (!p) /* don't print un-named types here */ continue; - (void) snprintf(name, sizeof(name)-1, "Icmp6%s%s", + snprintf(name, sizeof(name), "Icmp6%s%s", i & 0x100 ? "Out" : "In", p); seq_printf(seq, "%-32s\t%lu\n", name, snmp_fold_field(mib, i)); @@ -157,7 +157,7 @@ static void snmp6_seq_show_icmpv6msg(struct seq_file *seq, void **mib) val = snmp_fold_field(mib, i); if (!val) continue; - (void) snprintf(name, sizeof(name)-1, "Icmp6%sType%u", + snprintf(name, sizeof(name), "Icmp6%sType%u", i & 0x100 ? "Out" : "In", i & 0xff); seq_printf(seq, "%-32s\t%lu\n", name, val); } From 9bc646f163b136684390081262fab0fd8f5343ca Mon Sep 17 00:00:00 2001 From: Rakib Mullick Date: Thu, 20 Nov 2008 19:08:45 +0600 Subject: [PATCH 132/160] x86: fix __cpuinit/__init tangle in init_thread_xstate() Impact: fix incorrect __init annotation This patch removes the following section mismatch warning. A patch set was send previously (http://lkml.org/lkml/2008/11/10/407). But introduce some other problem, reported by Rufus (http://lkml.org/lkml/2008/11/11/46). Then Ingo Molnar suggest that, it's best to remove __init from xsave_cntxt_init(void). Which is the second patch in this series. Now, this one removes the following warning. WARNING: arch/x86/kernel/built-in.o(.cpuinit.text+0x2237): Section mismatch in reference from the function cpu_init() to the function .init.text:init_thread_xstate() The function __cpuinit cpu_init() references a function __init init_thread_xstate(). If init_thread_xstate is only used by cpu_init then annotate init_thread_xstate with a matching annotation. Signed-off-by: Rakib Mullick Signed-off-by: Ingo Molnar --- arch/x86/kernel/i387.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index 1f20608d4ca..b0f61f0dcd0 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -58,7 +58,7 @@ void __cpuinit mxcsr_feature_mask_init(void) stts(); } -void __init init_thread_xstate(void) +void __cpuinit init_thread_xstate(void) { if (!HAVE_HWFP) { xstate_size = sizeof(struct i387_soft_struct); From bfe085f62f98a49e1b864e4950389c7205174e4f Mon Sep 17 00:00:00 2001 From: Rakib Mullick Date: Thu, 20 Nov 2008 19:12:50 +0600 Subject: [PATCH 133/160] x86: fixing __cpuinit/__init tangle, xsave_cntxt_init() Annotate xsave_cntxt_init() as "can be called outside of __init". Signed-off-by: Rakib Mullick Signed-off-by: Ingo Molnar --- arch/x86/kernel/xsave.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kernel/xsave.c b/arch/x86/kernel/xsave.c index b13acb75e82..15c3e699918 100644 --- a/arch/x86/kernel/xsave.c +++ b/arch/x86/kernel/xsave.c @@ -310,7 +310,7 @@ static void __init setup_xstate_init(void) /* * Enable and initialize the xsave feature. */ -void __init xsave_cntxt_init(void) +void __ref xsave_cntxt_init(void) { unsigned int eax, ebx, ecx, edx; From 9e86786a4b17ad186f456dc6ac0508a17556731b Mon Sep 17 00:00:00 2001 From: David Daney Date: Sat, 20 Sep 2008 10:16:36 -0700 Subject: [PATCH 134/160] MIPS: Malta: Fix include paths in malta-amon.c On linux-queue, malta doesn't build after the include file relocation. This should fix it. There some occurrences of 'asm-mips' in the comments of quite a few files, but this is the only place I found it in any code. Signed-off-by: David Daney Signed-off-by: Ralf Baechle --- arch/mips/mti-malta/malta-amon.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/mips/mti-malta/malta-amon.c b/arch/mips/mti-malta/malta-amon.c index 96236bf3383..df9e526312a 100644 --- a/arch/mips/mti-malta/malta-amon.c +++ b/arch/mips/mti-malta/malta-amon.c @@ -22,9 +22,9 @@ #include #include -#include -#include -#include +#include +#include +#include int amon_cpu_avail(int cpu) { From 1b432840d0a4740020e29ae7a00717ef8f44954b Mon Sep 17 00:00:00 2001 From: Florian Fainelli Date: Fri, 31 Oct 2008 14:24:29 +0100 Subject: [PATCH 135/160] MIPS: RB532: GPIO register offsets are relative to GPIOBASE This patch fixes the wrong use of GPIO register offsets in devices.c. To avoid further problems, use gpio_get_value to return the NAND status instead of our own expanded code. Also define the zero offset of the alternate function register to allow consistent access. Signed-off-by: Florian Fainelli Signed-off-by: Phil Sutter Signed-off-by: Ralf Baechle --- arch/mips/include/asm/mach-rc32434/rb.h | 14 ++++++++------ arch/mips/rb532/devices.c | 2 +- 2 files changed, 9 insertions(+), 7 deletions(-) diff --git a/arch/mips/include/asm/mach-rc32434/rb.h b/arch/mips/include/asm/mach-rc32434/rb.h index 79e8ef67d0d..f25a8491670 100644 --- a/arch/mips/include/asm/mach-rc32434/rb.h +++ b/arch/mips/include/asm/mach-rc32434/rb.h @@ -40,12 +40,14 @@ #define BTCS 0x010040 #define BTCOMPARE 0x010044 #define GPIOBASE 0x050000 -#define GPIOCFG 0x050004 -#define GPIOD 0x050008 -#define GPIOILEVEL 0x05000C -#define GPIOISTAT 0x050010 -#define GPIONMIEN 0x050014 -#define IMASK6 0x038038 +/* Offsets relative to GPIOBASE */ +#define GPIOFUNC 0x00 +#define GPIOCFG 0x04 +#define GPIOD 0x08 +#define GPIOILEVEL 0x0C +#define GPIOISTAT 0x10 +#define GPIONMIEN 0x14 +#define IMASK6 0x38 #define LO_WPX (1 << 0) #define LO_ALE (1 << 1) #define LO_CLE (1 << 2) diff --git a/arch/mips/rb532/devices.c b/arch/mips/rb532/devices.c index 2f22d714d5b..c1c29181bd4 100644 --- a/arch/mips/rb532/devices.c +++ b/arch/mips/rb532/devices.c @@ -118,7 +118,7 @@ static struct platform_device cf_slot0 = { /* Resources and device for NAND */ static int rb532_dev_ready(struct mtd_info *mtd) { - return readl(IDT434_REG_BASE + GPIOD) & GPIO_RDY; + return gpio_get_value(GPIO_RDY); } static void rb532_cmd_ctrl(struct mtd_info *mtd, int cmd, unsigned int ctrl) From f43909dfb39c63ce54a598cbd7921643029afdeb Mon Sep 17 00:00:00 2001 From: Dmitri Vorobiev Date: Fri, 31 Oct 2008 19:54:11 +0200 Subject: [PATCH 136/160] MIPS: IP22: Make indy_sc_ops variable static The indy_sc_ops variable in arch/mips/mm/sc-ip22.c is needlessly defined global, and this patch makes it static. Signed-off-by: Dmitri Vorobiev Signed-off-by: Ralf Baechle --- --- arch/mips/mm/sc-ip22.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/mm/sc-ip22.c b/arch/mips/mm/sc-ip22.c index 1f602a110e1..13adb578211 100644 --- a/arch/mips/mm/sc-ip22.c +++ b/arch/mips/mm/sc-ip22.c @@ -161,7 +161,7 @@ static inline int __init indy_sc_probe(void) /* XXX Check with wje if the Indy caches can differenciate between writeback + invalidate and just invalidate. */ -struct bcache_ops indy_sc_ops = { +static struct bcache_ops indy_sc_ops = { .bc_enable = indy_sc_enable, .bc_disable = indy_sc_disable, .bc_wback_inv = indy_sc_wback_invalidate, From 2e373952cc893207a8b47a5e68c2f5155f912449 Mon Sep 17 00:00:00 2001 From: Phil Sutter Date: Sat, 1 Nov 2008 15:13:21 +0100 Subject: [PATCH 137/160] MIPS: RB532: Provide functions for gpio configuration As gpiolib doesn't support pin multiplexing, it provides no way to access the GPIOFUNC register. Also there is no support for setting interrupt status and level. These functions provide access to them and are needed by the CompactFlash driver. Signed-off-by: Phil Sutter Signed-off-by: Ralf Baechle --- arch/mips/include/asm/mach-rc32434/gpio.h | 2 + arch/mips/rb532/gpio.c | 197 ++++++++-------------- 2 files changed, 77 insertions(+), 122 deletions(-) diff --git a/arch/mips/include/asm/mach-rc32434/gpio.h b/arch/mips/include/asm/mach-rc32434/gpio.h index c8e554eafce..b5cf6457305 100644 --- a/arch/mips/include/asm/mach-rc32434/gpio.h +++ b/arch/mips/include/asm/mach-rc32434/gpio.h @@ -84,5 +84,7 @@ extern void set_434_reg(unsigned reg_offs, unsigned bit, unsigned len, unsigned extern unsigned get_434_reg(unsigned reg_offs); extern void set_latch_u5(unsigned char or_mask, unsigned char nand_mask); extern unsigned char get_latch_u5(void); +extern void rb532_gpio_set_ilevel(int bit, unsigned gpio); +extern void rb532_gpio_set_istat(int bit, unsigned gpio); #endif /* _RC32434_GPIO_H_ */ diff --git a/arch/mips/rb532/gpio.c b/arch/mips/rb532/gpio.c index 70c4a672637..0e84c8ab6a3 100644 --- a/arch/mips/rb532/gpio.c +++ b/arch/mips/rb532/gpio.c @@ -39,10 +39,6 @@ struct rb532_gpio_chip { struct gpio_chip chip; void __iomem *regbase; - void (*set_int_level)(struct gpio_chip *chip, unsigned offset, int value); - int (*get_int_level)(struct gpio_chip *chip, unsigned offset); - void (*set_int_status)(struct gpio_chip *chip, unsigned offset, int value); - int (*get_int_status)(struct gpio_chip *chip, unsigned offset); }; struct mpmc_device dev3; @@ -111,15 +107,47 @@ unsigned char get_latch_u5(void) } EXPORT_SYMBOL(get_latch_u5); +/* rb532_set_bit - sanely set a bit + * + * bitval: new value for the bit + * offset: bit index in the 4 byte address range + * ioaddr: 4 byte aligned address being altered + */ +static inline void rb532_set_bit(unsigned bitval, + unsigned offset, void __iomem *ioaddr) +{ + unsigned long flags; + u32 val; + + bitval = !!bitval; /* map parameter to {0,1} */ + + local_irq_save(flags); + + val = readl(ioaddr); + val &= ~( ~bitval << offset ); /* unset bit if bitval == 0 */ + val |= ( bitval << offset ); /* set bit if bitval == 1 */ + writel(val, ioaddr); + + local_irq_restore(flags); +} + +/* rb532_get_bit - read a bit + * + * returns the boolean state of the bit, which may be > 1 + */ +static inline int rb532_get_bit(unsigned offset, void __iomem *ioaddr) +{ + return (readl(ioaddr) & (1 << offset)); +} + /* * Return GPIO level */ static int rb532_gpio_get(struct gpio_chip *chip, unsigned offset) { - u32 mask = 1 << offset; struct rb532_gpio_chip *gpch; gpch = container_of(chip, struct rb532_gpio_chip, chip); - return readl(gpch->regbase + GPIOD) & mask; + return rb532_get_bit(offset, gpch->regbase + GPIOD); } /* @@ -128,23 +156,10 @@ static int rb532_gpio_get(struct gpio_chip *chip, unsigned offset) static void rb532_gpio_set(struct gpio_chip *chip, unsigned offset, int value) { - unsigned long flags; - u32 mask = 1 << offset; - u32 tmp; struct rb532_gpio_chip *gpch; - void __iomem *gpvr; gpch = container_of(chip, struct rb532_gpio_chip, chip); - gpvr = gpch->regbase + GPIOD; - - local_irq_save(flags); - tmp = readl(gpvr); - if (value) - tmp |= mask; - else - tmp &= ~mask; - writel(tmp, gpvr); - local_irq_restore(flags); + rb532_set_bit(value, offset, gpch->regbase + GPIOD); } /* @@ -152,21 +167,14 @@ static void rb532_gpio_set(struct gpio_chip *chip, */ static int rb532_gpio_direction_input(struct gpio_chip *chip, unsigned offset) { - unsigned long flags; - u32 mask = 1 << offset; - u32 value; struct rb532_gpio_chip *gpch; - void __iomem *gpdr; gpch = container_of(chip, struct rb532_gpio_chip, chip); - gpdr = gpch->regbase + GPIOCFG; - local_irq_save(flags); - value = readl(gpdr); - value &= ~mask; - writel(value, gpdr); - local_irq_restore(flags); + if (rb532_get_bit(offset, gpch->regbase + GPIOFUNC)) + return 1; /* alternate function, GPIOCFG is ignored */ + rb532_set_bit(0, offset, gpch->regbase + GPIOCFG); return 0; } @@ -176,99 +184,20 @@ static int rb532_gpio_direction_input(struct gpio_chip *chip, unsigned offset) static int rb532_gpio_direction_output(struct gpio_chip *chip, unsigned offset, int value) { - unsigned long flags; - u32 mask = 1 << offset; - u32 tmp; struct rb532_gpio_chip *gpch; - void __iomem *gpdr; gpch = container_of(chip, struct rb532_gpio_chip, chip); - writel(mask, gpch->regbase + GPIOD); - gpdr = gpch->regbase + GPIOCFG; - local_irq_save(flags); - tmp = readl(gpdr); - tmp |= mask; - writel(tmp, gpdr); - local_irq_restore(flags); + if (rb532_get_bit(offset, gpch->regbase + GPIOFUNC)) + return 1; /* alternate function, GPIOCFG is ignored */ + /* set the initial output value */ + rb532_set_bit(value, offset, gpch->regbase + GPIOD); + + rb532_set_bit(1, offset, gpch->regbase + GPIOCFG); return 0; } -/* - * Set the GPIO interrupt level - */ -static void rb532_gpio_set_int_level(struct gpio_chip *chip, - unsigned offset, int value) -{ - unsigned long flags; - u32 mask = 1 << offset; - u32 tmp; - struct rb532_gpio_chip *gpch; - void __iomem *gpil; - - gpch = container_of(chip, struct rb532_gpio_chip, chip); - gpil = gpch->regbase + GPIOILEVEL; - - local_irq_save(flags); - tmp = readl(gpil); - if (value) - tmp |= mask; - else - tmp &= ~mask; - writel(tmp, gpil); - local_irq_restore(flags); -} - -/* - * Get the GPIO interrupt level - */ -static int rb532_gpio_get_int_level(struct gpio_chip *chip, unsigned offset) -{ - u32 mask = 1 << offset; - struct rb532_gpio_chip *gpch; - - gpch = container_of(chip, struct rb532_gpio_chip, chip); - return readl(gpch->regbase + GPIOILEVEL) & mask; -} - -/* - * Set the GPIO interrupt status - */ -static void rb532_gpio_set_int_status(struct gpio_chip *chip, - unsigned offset, int value) -{ - unsigned long flags; - u32 mask = 1 << offset; - u32 tmp; - struct rb532_gpio_chip *gpch; - void __iomem *gpis; - - gpch = container_of(chip, struct rb532_gpio_chip, chip); - gpis = gpch->regbase + GPIOISTAT; - - local_irq_save(flags); - tmp = readl(gpis); - if (value) - tmp |= mask; - else - tmp &= ~mask; - writel(tmp, gpis); - local_irq_restore(flags); -} - -/* - * Get the GPIO interrupt status - */ -static int rb532_gpio_get_int_status(struct gpio_chip *chip, unsigned offset) -{ - u32 mask = 1 << offset; - struct rb532_gpio_chip *gpch; - - gpch = container_of(chip, struct rb532_gpio_chip, chip); - return readl(gpch->regbase + GPIOISTAT) & mask; -} - static struct rb532_gpio_chip rb532_gpio_chip[] = { [0] = { .chip = { @@ -280,13 +209,35 @@ static struct rb532_gpio_chip rb532_gpio_chip[] = { .base = 0, .ngpio = 32, }, - .get_int_level = rb532_gpio_get_int_level, - .set_int_level = rb532_gpio_set_int_level, - .get_int_status = rb532_gpio_get_int_status, - .set_int_status = rb532_gpio_set_int_status, }, }; +/* + * Set GPIO interrupt level + */ +void rb532_gpio_set_ilevel(int bit, unsigned gpio) +{ + rb532_set_bit(bit, gpio, rb532_gpio_chip->regbase + GPIOILEVEL); +} +EXPORT_SYMBOL(rb532_gpio_set_ilevel); + +/* + * Set GPIO interrupt status + */ +void rb532_gpio_set_istat(int bit, unsigned gpio) +{ + rb532_set_bit(bit, gpio, rb532_gpio_chip->regbase + GPIOISTAT); +} +EXPORT_SYMBOL(rb532_gpio_set_istat); + +/* + * Configure GPIO alternate function + */ +static void rb532_gpio_set_func(int bit, unsigned gpio) +{ + rb532_set_bit(bit, gpio, rb532_gpio_chip->regbase + GPIOFUNC); +} + int __init rb532_gpio_init(void) { struct resource *r; @@ -310,9 +261,11 @@ int __init rb532_gpio_init(void) return -ENXIO; } - /* Set the interrupt status and level for the CF pin */ - rb532_gpio_set_int_level(&rb532_gpio_chip->chip, CF_GPIO_NUM, 1); - rb532_gpio_set_int_status(&rb532_gpio_chip->chip, CF_GPIO_NUM, 0); + /* configure CF_GPIO_NUM as CFRDY IRQ source */ + rb532_gpio_set_func(0, CF_GPIO_NUM); + rb532_gpio_direction_input(&rb532_gpio_chip->chip, CF_GPIO_NUM); + rb532_gpio_set_ilevel(1, CF_GPIO_NUM); + rb532_gpio_set_istat(0, CF_GPIO_NUM); return 0; } From 664c4bbb73ec53e4f81f7d80a09571b49bad1f96 Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 3 Nov 2008 11:31:54 +0000 Subject: [PATCH 138/160] MIPS: csrc-r4k: Fix spelling mistake. Signed-off-by: Ralf Baechle --- arch/mips/kernel/csrc-r4k.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/kernel/csrc-r4k.c b/arch/mips/kernel/csrc-r4k.c index 86e026f067b..74fb74583b4 100644 --- a/arch/mips/kernel/csrc-r4k.c +++ b/arch/mips/kernel/csrc-r4k.c @@ -27,7 +27,7 @@ int __init init_mips_clocksource(void) if (!cpu_has_counter || !mips_hpt_frequency) return -ENXIO; - /* Calclate a somewhat reasonable rating value */ + /* Calculate a somewhat reasonable rating value */ clocksource_mips.rating = 200 + mips_hpt_frequency / 10000000; clocksource_set_clock(&clocksource_mips, mips_hpt_frequency); From a24e849c019f15796984be9fe301fa9ead6f0f9e Mon Sep 17 00:00:00 2001 From: Ralf Baechle Date: Mon, 3 Nov 2008 11:32:34 +0000 Subject: [PATCH 139/160] MIPS: csrc-r4k: Fix declaration depending on the wrong CONFIG_ symbol. Signed-off-by: Ralf Baechle --- arch/mips/include/asm/time.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/mips/include/asm/time.h b/arch/mips/include/asm/time.h index d3bd5c5aa2e..9601ea95054 100644 --- a/arch/mips/include/asm/time.h +++ b/arch/mips/include/asm/time.h @@ -63,7 +63,7 @@ static inline int mips_clockevent_init(void) /* * Initialize the count register as a clocksource */ -#ifdef CONFIG_CEVT_R4K +#ifdef CONFIG_CSRC_R4K extern int init_mips_clocksource(void); #else static inline int init_mips_clocksource(void) From ddb4cbfc53aa0913ee8da059fcbf628d14f40f63 Mon Sep 17 00:00:00 2001 From: Steve French Date: Thu, 20 Nov 2008 20:00:44 +0000 Subject: [PATCH 140/160] [CIFS] Do not attempt to close invalidated file handles If a connection with open file handles has gone down and come back up and reconnected without reopening the file handle yet, do not attempt to send an SMB close request for this handle in cifs_close. We were checking for the connection being invalid in cifs_close but since the connection may have been reconnected we also need to check whether the file handle was marked invalid (otherwise we could close the wrong file handle by accident). Acked-by: Jeff Layton Signed-off-by: Steve French --- fs/cifs/cifsglob.h | 10 +++++++++- fs/cifs/file.c | 21 ++++++++++++++------- fs/cifs/misc.c | 3 +++ fs/cifs/readdir.c | 5 ++++- 4 files changed, 30 insertions(+), 9 deletions(-) diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index f1ae1f57c30..c57c0565547 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -606,7 +606,15 @@ GLOBAL_EXTERN struct list_head cifs_tcp_ses_list; * changes to the tcon->tidStatus should be done while holding this lock. */ GLOBAL_EXTERN rwlock_t cifs_tcp_ses_lock; -GLOBAL_EXTERN rwlock_t GlobalSMBSeslock; /* protects list inserts on 3 above */ + +/* + * This lock protects the cifs_file->llist and cifs_file->flist + * list operations, and updates to some flags (cifs_file->invalidHandle) + * It will be moved to either use the tcon->stat_lock or equivalent later. + * If cifs_tcp_ses_lock and the lock below are both needed to be held, then + * the cifs_tcp_ses_lock must be grabbed first and released last. + */ +GLOBAL_EXTERN rwlock_t GlobalSMBSeslock; GLOBAL_EXTERN struct list_head GlobalOplock_Q; diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 6449e1aae62..b691b893a84 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -488,12 +488,13 @@ int cifs_close(struct inode *inode, struct file *file) pTcon = cifs_sb->tcon; if (pSMBFile) { struct cifsLockInfo *li, *tmp; - + write_lock(&GlobalSMBSeslock); pSMBFile->closePend = true; if (pTcon) { /* no sense reconnecting to close a file that is already closed */ if (!pTcon->need_reconnect) { + write_unlock(&GlobalSMBSeslock); timeout = 2; while ((atomic_read(&pSMBFile->wrtPending) != 0) && (timeout <= 2048)) { @@ -510,12 +511,15 @@ int cifs_close(struct inode *inode, struct file *file) timeout *= 4; } if (atomic_read(&pSMBFile->wrtPending)) - cERROR(1, - ("close with pending writes")); - rc = CIFSSMBClose(xid, pTcon, + cERROR(1, ("close with pending write")); + if (!pTcon->need_reconnect && + !pSMBFile->invalidHandle) + rc = CIFSSMBClose(xid, pTcon, pSMBFile->netfid); - } - } + } else + write_unlock(&GlobalSMBSeslock); + } else + write_unlock(&GlobalSMBSeslock); /* Delete any outstanding lock records. We'll lose them when the file is closed anyway. */ @@ -587,15 +591,18 @@ int cifs_closedir(struct inode *inode, struct file *file) pTcon = cifs_sb->tcon; cFYI(1, ("Freeing private data in close dir")); + write_lock(&GlobalSMBSeslock); if (!pCFileStruct->srch_inf.endOfSearch && !pCFileStruct->invalidHandle) { pCFileStruct->invalidHandle = true; + write_unlock(&GlobalSMBSeslock); rc = CIFSFindClose(xid, pTcon, pCFileStruct->netfid); cFYI(1, ("Closing uncompleted readdir with rc %d", rc)); /* not much we can do if it fails anyway, ignore rc */ rc = 0; - } + } else + write_unlock(&GlobalSMBSeslock); ptmp = pCFileStruct->srch_inf.ntwrk_buf_start; if (ptmp) { cFYI(1, ("closedir free smb buf in srch struct")); diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index addd1dcc2d7..9ee3f689c2b 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -555,12 +555,14 @@ is_valid_oplock_break(struct smb_hdr *buf, struct TCP_Server_Info *srv) continue; cifs_stats_inc(&tcon->num_oplock_brks); + write_lock(&GlobalSMBSeslock); list_for_each(tmp2, &tcon->openFileList) { netfile = list_entry(tmp2, struct cifsFileInfo, tlist); if (pSMB->Fid != netfile->netfid) continue; + write_unlock(&GlobalSMBSeslock); read_unlock(&cifs_tcp_ses_lock); cFYI(1, ("file id match, oplock break")); pCifsInode = CIFS_I(netfile->pInode); @@ -576,6 +578,7 @@ is_valid_oplock_break(struct smb_hdr *buf, struct TCP_Server_Info *srv) return true; } + write_unlock(&GlobalSMBSeslock); read_unlock(&cifs_tcp_ses_lock); cFYI(1, ("No matching file for oplock break")); return true; diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 58d57299f2a..9f51f9bf029 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -741,11 +741,14 @@ static int find_cifs_entry(const int xid, struct cifsTconInfo *pTcon, (index_to_find < first_entry_in_buffer)) { /* close and restart search */ cFYI(1, ("search backing up - close and restart search")); + write_lock(&GlobalSMBSeslock); if (!cifsFile->srch_inf.endOfSearch && !cifsFile->invalidHandle) { cifsFile->invalidHandle = true; + write_unlock(&GlobalSMBSeslock); CIFSFindClose(xid, pTcon, cifsFile->netfid); - } + } else + write_unlock(&GlobalSMBSeslock); if (cifsFile->srch_inf.ntwrk_buf_start) { cFYI(1, ("freeing SMB ff cache buf on search rewind")); if (cifsFile->srch_inf.smallBuf) From ed79b86d8acf1f3d3bb83f04dc216c8dfa1d5970 Mon Sep 17 00:00:00 2001 From: Helge Deller Date: Thu, 20 Nov 2008 10:54:09 +0100 Subject: [PATCH 141/160] parisc: fix bug in compat_arch_ptrace Commit 81e192d6ce303b6792aa38ff35f41a1a7357f23a ("parisc: convert to generic compat_sys_ptrace") introduced a bug which segfaults the parisc 64bit kernel when stracing 32bit applications: Kernel Fault: Code=15 regs=00000000bafa42b0 (Addr=00000001baf5ab57) YZrvWESTHLNXBCVMcbcbcbcbOGFRQPDI PSW: 00001000000001101111111100001011 Tainted: G W r00-03 000000ff0806ff0b 000000004068edc0 00000000401203f8 00000000fb3e2508 r04-07 0000000040686dc0 00000000baf5a800 fffffffffffffffc fffffffffb3e2508 r08-11 00000000baf5a800 000000000004b068 00000000000402b0 0000000000040d68 r12-15 0000000000042a9c 0000000000040a9c 0000000000040d60 0000000000042e9c r16-19 000000000004b060 000000000004b058 0000000000042d9c ffffffffffffffff r20-23 000000000800000b 0000000000000000 000000000800000b fffffffffb3e2508 r24-27 00000000fffffffc 0000000000000003 00000000fffffffc 0000000040686dc0 r28-31 00000001baf5a7ff 00000000bafa4280 00000000bafa42b0 00000000000001d7 sr00-03 0000000000fca000 0000000000000000 0000000000000000 0000000000fca000 sr04-07 0000000000000000 0000000000000000 0000000000000000 0000000000000000 IASQ: 0000000000000000 0000000000000000 IAOQ: 0000000040120400 0000000040120404 IIR: 4b9a06b0 ISR: 0000000000000000 IOR: 00000001baf5ab57 CPU: 0 CR30: 00000000bafa4000 CR31: 00000000d22344e0 ORIG_R28: 00000000fb3e2248 IAOQ[0]: compat_arch_ptrace+0xb8/0x160 IAOQ[1]: compat_arch_ptrace+0xbc/0x160 RP(r2): compat_arch_ptrace+0xb0/0x160 Backtrace: [<00000000401612ac>] compat_sys_ptrace+0x15c/0x180 [<0000000040104ef8>] syscall_exit+0x0/0x14 The problem is that compat_arch_ptrace() enters with an addr value of type compat_ulong_t and calls translate_usr_offset() to translate the address offset into a struct pt_regs offset like this: addr = translate_usr_offset(addr) this means that any return value of translate_usr_offset() is stored back as compat_ulong_t type into the addr variable. But since translate_usr_offset() returns -1 for invalid offsets, addr can now get the value 0xffffffff which then fails the next return-value sanity check and thus the kernel tries to access invalid memory: if (addr < 0) break; Fix this bug by modifying translate_usr_offset() to take and return values of type compat_ulong_t, and by returning the value "sizeof(struct pt_regs)" as an error indicator. Additionally change the sanity check to check for return values for >= sizeof(struct pt_regs). This patch survived my compile and run-tests. Signed-off-by: Helge Deller Signed-off-by: Linus Torvalds --- arch/parisc/kernel/ptrace.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/arch/parisc/kernel/ptrace.c b/arch/parisc/kernel/ptrace.c index 90904f9dfc5..927db3668b6 100644 --- a/arch/parisc/kernel/ptrace.c +++ b/arch/parisc/kernel/ptrace.c @@ -183,10 +183,10 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) * being 64 bit in both cases. */ -static long translate_usr_offset(long offset) +static compat_ulong_t translate_usr_offset(compat_ulong_t offset) { if (offset < 0) - return -1; + return sizeof(struct pt_regs); else if (offset <= 32*4) /* gr[0..31] */ return offset * 2 + 4; else if (offset <= 32*4+32*8) /* gr[0..31] + fr[0..31] */ @@ -194,7 +194,7 @@ static long translate_usr_offset(long offset) else if (offset < sizeof(struct pt_regs)/2 + 32*4) return offset * 2 + 4 - 32*8; else - return -1; + return sizeof(struct pt_regs); } long compat_arch_ptrace(struct task_struct *child, compat_long_t request, @@ -209,7 +209,7 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request, if (addr & (sizeof(compat_uint_t)-1)) break; addr = translate_usr_offset(addr); - if (addr < 0) + if (addr >= sizeof(struct pt_regs)) break; tmp = *(compat_uint_t *) ((char *) task_regs(child) + addr); @@ -236,7 +236,7 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request, if (addr & (sizeof(compat_uint_t)-1)) break; addr = translate_usr_offset(addr); - if (addr < 0) + if (addr >= sizeof(struct pt_regs)) break; if (addr >= PT_FR0 && addr <= PT_FR31 + 4) { /* Special case, fp regs are 64 bits anyway */ From 23918b03060f6e572168fdde1798a905679d2e06 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 20 Nov 2008 16:06:21 -0500 Subject: [PATCH 142/160] SUNRPC: Fix a performance regression in the RPC authentication code Fix a regression reported by Max Kellermann whereby kernel profiling showed that his clients were spending 45% of their time in rpcauth_lookup_credcache. It turns out that although his processes had identical uid/gid/groups, generic_match() was failing to detect this, because the task->group_info pointers were not shared. This again lead to the creation of a huge number of identical credentials at the RPC layer. The regression is fixed by comparing the contents of task->group_info if the actual pointers are not identical. Signed-off-by: Trond Myklebust Signed-off-by: Linus Torvalds --- net/sunrpc/auth_generic.c | 20 ++++++++++++++++++-- 1 file changed, 18 insertions(+), 2 deletions(-) diff --git a/net/sunrpc/auth_generic.c b/net/sunrpc/auth_generic.c index 744b79fdcb1..4028502f052 100644 --- a/net/sunrpc/auth_generic.c +++ b/net/sunrpc/auth_generic.c @@ -133,13 +133,29 @@ static int generic_match(struct auth_cred *acred, struct rpc_cred *cred, int flags) { struct generic_cred *gcred = container_of(cred, struct generic_cred, gc_base); + int i; if (gcred->acred.uid != acred->uid || gcred->acred.gid != acred->gid || - gcred->acred.group_info != acred->group_info || gcred->acred.machine_cred != acred->machine_cred) - return 0; + goto out_nomatch; + + /* Optimisation in the case where pointers are identical... */ + if (gcred->acred.group_info == acred->group_info) + goto out_match; + + /* Slow path... */ + if (gcred->acred.group_info->ngroups != acred->group_info->ngroups) + goto out_nomatch; + for (i = 0; i < gcred->acred.group_info->ngroups; i++) { + if (GROUP_AT(gcred->acred.group_info, i) != + GROUP_AT(acred->group_info, i)) + goto out_nomatch; + } +out_match: return 1; +out_nomatch: + return 0; } void __init rpc_init_generic_auth(void) From 0ca4b6b00113b064c080d26d803d0d7c80fb5dc8 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Thu, 20 Nov 2008 14:09:33 -0700 Subject: [PATCH 143/160] x86: Fix interrupt leak due to migration When we migrate an interrupt from one CPU to another, we set the move_in_progress flag and clean up the vectors later once they're not being used. If you're unlucky and call destroy_irq() before the vectors become un-used, the move_in_progress flag is never cleared, which causes the interrupt to become unusable. This was discovered by Jesse Brandeburg for whom it manifested as an MSI-X device refusing to use MSI-X mode when the driver was unloaded and reloaded repeatedly. Signed-off-by: Matthew Wilcox Signed-off-by: Linus Torvalds --- arch/x86/kernel/io_apic.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/io_apic.c index 7a3f2028e2e..c9513e1ff28 100644 --- a/arch/x86/kernel/io_apic.c +++ b/arch/x86/kernel/io_apic.c @@ -1140,6 +1140,20 @@ static void __clear_irq_vector(int irq) cfg->vector = 0; cpus_clear(cfg->domain); + + if (likely(!cfg->move_in_progress)) + return; + cpus_and(mask, cfg->old_domain, cpu_online_map); + for_each_cpu_mask_nr(cpu, mask) { + for (vector = FIRST_EXTERNAL_VECTOR; vector < NR_VECTORS; + vector++) { + if (per_cpu(vector_irq, cpu)[vector] != irq) + continue; + per_cpu(vector_irq, cpu)[vector] = -1; + break; + } + } + cfg->move_in_progress = 0; } void __setup_vector_irq(int cpu) From b704882e70d87d7f56db5ff17e2253f3fa90e4f3 Mon Sep 17 00:00:00 2001 From: Tony Luck Date: Thu, 20 Nov 2008 13:27:12 -0800 Subject: [PATCH 144/160] [IA64] Rationalize kernel mode alignment checking Itanium processors can handle some misaligned data accesses. They also provide a mode where all such accesses are forced to trap. The kernel was schizophrenic about use of this mode: * Base kernel code ran in permissive mode where the only traps generated were from those cases that the h/w could not handle. * Interrupt, syscall and trap code ran in strict mode where all unaligned accesses caused traps to the 0x5a00 unaligned reference vector. Use strict alignment checking throughout the kernel, but make sure that we continue to let user mode use more relaxed mode as the default. Signed-off-by: Tony Luck --- arch/ia64/kernel/entry.S | 1 + arch/ia64/kernel/head.S | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S index 7ef0c594f5e..d435f4a7a96 100644 --- a/arch/ia64/kernel/entry.S +++ b/arch/ia64/kernel/entry.S @@ -499,6 +499,7 @@ GLOBAL_ENTRY(prefetch_stack) END(prefetch_stack) GLOBAL_ENTRY(kernel_execve) + rum psr.ac mov r15=__NR_execve // put syscall number in place break __BREAK_SYSCALL br.ret.sptk.many rp diff --git a/arch/ia64/kernel/head.S b/arch/ia64/kernel/head.S index 66e491d8baa..59301c47280 100644 --- a/arch/ia64/kernel/head.S +++ b/arch/ia64/kernel/head.S @@ -260,7 +260,7 @@ start_ap: * Switch into virtual mode: */ movl r16=(IA64_PSR_IT|IA64_PSR_IC|IA64_PSR_DT|IA64_PSR_RT|IA64_PSR_DFH|IA64_PSR_BN \ - |IA64_PSR_DI) + |IA64_PSR_DI|IA64_PSR_AC) ;; mov cr.ipsr=r16 movl r17=1f From ef23cdbefc3ab7f2ee9ee6dc6d4a94d7d6ec5e2b Mon Sep 17 00:00:00 2001 From: Hidetoshi Seto Date: Mon, 17 Nov 2008 10:18:08 +0900 Subject: [PATCH 145/160] [IA64] use mprintk instead of printk, in ia64_mca_modify_original_stack Using printk from MCA/INIT context is unsafe since it can cause deadlock. The ia64_mca_modify_original_stack is called from both of mca handler and init handler, so it should use mprintk instead of printk. Signed-off-by: Hidetoshi Seto Signed-off-by: Tony Luck --- arch/ia64/kernel/mca.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/ia64/kernel/mca.c b/arch/ia64/kernel/mca.c index 7dd96c12717..bab1de2d2f6 100644 --- a/arch/ia64/kernel/mca.c +++ b/arch/ia64/kernel/mca.c @@ -1139,7 +1139,7 @@ ia64_mca_modify_original_stack(struct pt_regs *regs, return previous_current; no_mod: - printk(KERN_INFO "cpu %d, %s %s, original stack not modified\n", + mprintk(KERN_INFO "cpu %d, %s %s, original stack not modified\n", smp_processor_id(), type, msg); return previous_current; } From d5964107763d7155e9bea658098a337507b3e928 Mon Sep 17 00:00:00 2001 From: Huang Weiyi Date: Thu, 20 Nov 2008 13:38:16 -0800 Subject: [PATCH 146/160] [IA64] remove duplicate include iommu.h arch/ia64/kernel/pci-dma.c only needs to include iommu once. Signed-off-by: Huang Weiyi Signed-off-by: Tony Luck --- arch/ia64/kernel/pci-dma.c | 1 - 1 file changed, 1 deletion(-) diff --git a/arch/ia64/kernel/pci-dma.c b/arch/ia64/kernel/pci-dma.c index dbdb778efa0..2a92f637431 100644 --- a/arch/ia64/kernel/pci-dma.c +++ b/arch/ia64/kernel/pci-dma.c @@ -19,7 +19,6 @@ #include #include -#include dma_addr_t bad_dma_address __read_mostly; EXPORT_SYMBOL(bad_dma_address); From 93fe10b670a7a6a1dc9649c7860f452dc7bbbb9d Mon Sep 17 00:00:00 2001 From: Isaku Yamahata Date: Tue, 18 Nov 2008 19:19:50 +0900 Subject: [PATCH 147/160] [IA64] ia64/pv_ops/pv_cpu_ops: fix _IA64_REG_IP case. pv_cpu_ops.getreg(_IA64_REG_IP) returned constant. But the returned ip valued should be the one in the caller, not of the callee. This patch fixes that. Signed-off-by: Isaku Yamahata Signed-off-by: Tony Luck --- arch/ia64/include/asm/intrinsics.h | 2 +- arch/ia64/include/asm/paravirt_privop.h | 13 +++++++++++++ arch/ia64/kernel/paravirt.c | 2 +- 3 files changed, 15 insertions(+), 2 deletions(-) diff --git a/arch/ia64/include/asm/intrinsics.h b/arch/ia64/include/asm/intrinsics.h index 47d686dba1e..a3e44a5ed49 100644 --- a/arch/ia64/include/asm/intrinsics.h +++ b/arch/ia64/include/asm/intrinsics.h @@ -226,7 +226,7 @@ extern long ia64_cmpxchg_called_with_bad_pointer (void); /************************************************/ #define ia64_ssm IA64_INTRINSIC_MACRO(ssm) #define ia64_rsm IA64_INTRINSIC_MACRO(rsm) -#define ia64_getreg IA64_INTRINSIC_API(getreg) +#define ia64_getreg IA64_INTRINSIC_MACRO(getreg) #define ia64_setreg IA64_INTRINSIC_API(setreg) #define ia64_set_rr IA64_INTRINSIC_API(set_rr) #define ia64_get_rr IA64_INTRINSIC_API(get_rr) diff --git a/arch/ia64/include/asm/paravirt_privop.h b/arch/ia64/include/asm/paravirt_privop.h index d577aac1183..0b597424fcf 100644 --- a/arch/ia64/include/asm/paravirt_privop.h +++ b/arch/ia64/include/asm/paravirt_privop.h @@ -78,6 +78,19 @@ extern unsigned long ia64_native_getreg_func(int regnum); ia64_native_rsm(mask); \ } while (0) +/* returned ip value should be the one in the caller, + * not in __paravirt_getreg() */ +#define paravirt_getreg(reg) \ + ({ \ + unsigned long res; \ + BUILD_BUG_ON(!__builtin_constant_p(reg)); \ + if ((reg) == _IA64_REG_IP) \ + res = ia64_native_getreg(_IA64_REG_IP); \ + else \ + res = pv_cpu_ops.getreg(reg); \ + res; \ + }) + /****************************************************************************** * replacement of hand written assembly codes. */ diff --git a/arch/ia64/kernel/paravirt.c b/arch/ia64/kernel/paravirt.c index de35d8e8b7d..9f14c16f636 100644 --- a/arch/ia64/kernel/paravirt.c +++ b/arch/ia64/kernel/paravirt.c @@ -130,7 +130,7 @@ ia64_native_getreg_func(int regnum) unsigned long res = -1; switch (regnum) { CASE_GET_REG(GP); - CASE_GET_REG(IP); + /*CASE_GET_REG(IP);*/ /* returned ip value shouldn't be constant */ CASE_GET_REG(PSR); CASE_GET_REG(TP); CASE_GET_REG(SP); From 0090d481ee4c5d9a54ae2c457ab3ef9111dd4a91 Mon Sep 17 00:00:00 2001 From: Isaku Yamahata Date: Tue, 18 Nov 2008 19:20:51 +0900 Subject: [PATCH 148/160] [IA64] xen: fix xen_get_eflags. fix xen_get_eflags. It doesn't take any argument. Signed-off-by: Isaku Yamahata Signed-off-by: Tony Luck --- arch/ia64/xen/hypercall.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/ia64/xen/hypercall.S b/arch/ia64/xen/hypercall.S index d4ff0b9e79f..45e02bb64a9 100644 --- a/arch/ia64/xen/hypercall.S +++ b/arch/ia64/xen/hypercall.S @@ -58,7 +58,7 @@ __HCALL2(xen_set_rr, HYPERPRIVOP_SET_RR) __HCALL2(xen_set_kr, HYPERPRIVOP_SET_KR) #ifdef CONFIG_IA32_SUPPORT -__HCALL1(xen_get_eflag, HYPERPRIVOP_GET_EFLAG) +__HCALL0(xen_get_eflag, HYPERPRIVOP_GET_EFLAG) __HCALL1(xen_set_eflag, HYPERPRIVOP_SET_EFLAG) // refer SDM vol1 3.1.8 #endif /* CONFIG_IA32_SUPPORT */ From 13d428afc007fcfcd6deeb215618f54cf9c0cae6 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 20 Nov 2008 15:19:22 -0800 Subject: [PATCH 149/160] Linux 2.6.28-rc6 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index a9ae5dc0aa1..7b1f2384094 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 28 -EXTRAVERSION = -rc5 +EXTRAVERSION = -rc6 NAME = Killer Bat of Doom # *DOCUMENTATION* From a1967d64414dab500e86cbbddf8eae6ad2047903 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 21 Nov 2008 11:16:48 -0800 Subject: [PATCH 150/160] x86: revert irq number limitation Impact: fix MSIx not enough irq numbers available regression The manual revert of the sparse_irq patches missed to bring the number of possible irqs back to the .27 status. This resulted in a regression when two multichannel network cards were placed in a system with only one IO_APIC - causing the networking driver to not have the right IRQ and the device not coming up. Remove the dynamic allocation logic leftovers and simply return NR_IRQS in probe_nr_irqs() for now. Fixes: http://lkml.org/lkml/2008/11/19/354 Reported-by: Jesper Dangaard Brouer Signed-off-by: Thomas Gleixner Tested-by: Jesper Dangaard Brouer Acked-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/kernel/io_apic.c | 22 +--------------------- 1 file changed, 1 insertion(+), 21 deletions(-) diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/io_apic.c index c9513e1ff28..1fec0f9b150 100644 --- a/arch/x86/kernel/io_apic.c +++ b/arch/x86/kernel/io_apic.c @@ -3608,27 +3608,7 @@ int __init io_apic_get_redir_entries (int ioapic) int __init probe_nr_irqs(void) { - int idx; - int nr = 0; -#ifndef CONFIG_XEN - int nr_min = 32; -#else - int nr_min = NR_IRQS; -#endif - - for (idx = 0; idx < nr_ioapics; idx++) - nr += io_apic_get_redir_entries(idx) + 1; - - /* double it for hotplug and msi and nmi */ - nr <<= 1; - - /* something wrong ? */ - if (nr < nr_min) - nr = nr_min; - if (WARN_ON(nr > NR_IRQS)) - nr = NR_IRQS; - - return nr; + return NR_IRQS; } /* -------------------------------------------------------------------------- From 86bbc2c235e500957b213e7e64ce2e0ccb8bc131 Mon Sep 17 00:00:00 2001 From: Ian Campbell Date: Fri, 21 Nov 2008 10:21:33 +0000 Subject: [PATCH 151/160] xen: pin correct PGD on suspend Impact: fix Xen guest boot failure commit eefb47f6a1e855653d275cb90592a3587ea93a09 ("xen: use spin_lock_nest_lock when pinning a pagetable") changed xen_pgd_walk to walk over mm->pgd rather than taking pgd as an argument. This breaks xen_mm_(un)pin_all() because it makes init_mm.pgd readonly instead of the pgd we are interested in and therefore the pin subsequently fails. (XEN) mm.c:2280:d15 Bad type (saw 00000000e8000001 != exp 0000000060000000) for mfn bc464 (pfn 21ca7) (XEN) mm.c:2665:d15 Error while pinning mfn bc464 [ 14.586913] 1 multicall(s) failed: cpu 0 [ 14.586926] Pid: 14, comm: kstop/0 Not tainted 2.6.28-rc5-x86_32p-xenU-00172-gee2f6cc #200 [ 14.586940] Call Trace: [ 14.586955] [] ? printk+0x18/0x1e [ 14.586972] [] xen_mc_flush+0x163/0x1d0 [ 14.586986] [] __xen_pgd_pin+0xa1/0x110 [ 14.587000] [] ? stop_cpu+0x0/0xf0 [ 14.587015] [] xen_mm_pin_all+0x4b/0x70 [ 14.587029] [] xen_suspend+0x39/0xe0 [ 14.587042] [] ? stop_cpu+0x0/0xf0 [ 14.587054] [] stop_cpu+0x9d/0xf0 [ 14.587067] [] run_workqueue+0x8d/0x150 [ 14.587080] [] ? _spin_unlock_irqrestore+0x23/0x40 [ 14.587094] [] ? prepare_to_wait+0x3a/0x70 [ 14.587107] [] worker_thread+0x88/0xf0 [ 14.587120] [] ? autoremove_wake_function+0x0/0x50 [ 14.587133] [] ? worker_thread+0x0/0xf0 [ 14.587146] [] kthread+0x3c/0x70 [ 14.587157] [] ? kthread+0x0/0x70 [ 14.587170] [] kernel_thread_helper+0x7/0x10 [ 14.587181] call 1/3: op=14 arg=[c0415000] result=0 [ 14.587192] call 2/3: op=14 arg=[e1ca2000] result=0 [ 14.587204] call 3/3: op=26 arg=[c1808860] result=-22 Signed-off-by: Ian Campbell Acked-by: Jeremy Fitzhardinge Signed-off-by: Ingo Molnar --- arch/x86/xen/mmu.c | 21 ++++++++++++++------- 1 file changed, 14 insertions(+), 7 deletions(-) diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index 688936044dc..636ef4caa52 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -661,12 +661,11 @@ void xen_set_pgd(pgd_t *ptr, pgd_t val) * For 64-bit, we must skip the Xen hole in the middle of the address * space, just after the big x86-64 virtual hole. */ -static int xen_pgd_walk(struct mm_struct *mm, - int (*func)(struct mm_struct *mm, struct page *, - enum pt_level), - unsigned long limit) +static int __xen_pgd_walk(struct mm_struct *mm, pgd_t *pgd, + int (*func)(struct mm_struct *mm, struct page *, + enum pt_level), + unsigned long limit) { - pgd_t *pgd = mm->pgd; int flush = 0; unsigned hole_low, hole_high; unsigned pgdidx_limit, pudidx_limit, pmdidx_limit; @@ -753,6 +752,14 @@ out: return flush; } +static int xen_pgd_walk(struct mm_struct *mm, + int (*func)(struct mm_struct *mm, struct page *, + enum pt_level), + unsigned long limit) +{ + return __xen_pgd_walk(mm, mm->pgd, func, limit); +} + /* If we're using split pte locks, then take the page's lock and return a pointer to it. Otherwise return NULL. */ static spinlock_t *xen_pte_lock(struct page *page, struct mm_struct *mm) @@ -854,7 +861,7 @@ static void __xen_pgd_pin(struct mm_struct *mm, pgd_t *pgd) xen_mc_batch(); - if (xen_pgd_walk(mm, xen_pin_page, USER_LIMIT)) { + if (__xen_pgd_walk(mm, pgd, xen_pin_page, USER_LIMIT)) { /* re-enable interrupts for flushing */ xen_mc_issue(0); @@ -998,7 +1005,7 @@ static void __xen_pgd_unpin(struct mm_struct *mm, pgd_t *pgd) PT_PMD); #endif - xen_pgd_walk(mm, xen_unpin_page, USER_LIMIT); + __xen_pgd_walk(mm, pgd, xen_unpin_page, USER_LIMIT); xen_mc_issue(0); } From 5cf02b7bafddb6c3c16ddfb23d3ce187f70528ba Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Tue, 25 Nov 2008 00:42:37 -0500 Subject: [PATCH 152/160] x86: use limited register constraint for setnz Impact: build fix with certain compilers GCC can decide to use %dil when "r" is used, which is not valid for setnz. This bug was brought out by Stephen Rothwell's merging of the branch tracer into linux-next. [ Thanks to Uros Bizjak for recommending 'q' over 'Q' ] Signed-off-by: Steven Rostedt Signed-off-by: Ingo Molnar --- arch/x86/boot/tty.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/boot/tty.c b/arch/x86/boot/tty.c index 0be77b39328..7e8e8b25f5f 100644 --- a/arch/x86/boot/tty.c +++ b/arch/x86/boot/tty.c @@ -74,7 +74,7 @@ static int kbd_pending(void) { u8 pending; asm volatile("int $0x16; setnz %0" - : "=rm" (pending) + : "=qm" (pending) : "a" (0x0100)); return pending; } From eff79aee91dd07e944df65fa448c8baeee7709d8 Mon Sep 17 00:00:00 2001 From: Julia Lawall Date: Tue, 25 Nov 2008 14:13:03 +0100 Subject: [PATCH 153/160] arch/x86/kernel/pci-calgary_64.c: change simple_strtol to simple_strtoul Impact: fix theoretical option string parsing overflow Since bridge is unsigned, it would seem better to use simple_strtoul that simple_strtol. A simplified version of the semantic patch that makes this change is as follows: (http://www.emn.fr/x-info/coccinelle/) // @r2@ long e; position p; @@ e = simple_strtol@p(...) @@ position p != r2.p; type T; T e; @@ e = - simple_strtol@p + simple_strtoul (...) // Signed-off-by: Julia Lawall Cc: muli@il.ibm.com Cc: jdmason@kudzu.us Cc: discuss@x86-64.org Signed-off-by: Ingo Molnar --- arch/x86/kernel/pci-calgary_64.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c index e1e731d78f3..d28bbdc35e4 100644 --- a/arch/x86/kernel/pci-calgary_64.c +++ b/arch/x86/kernel/pci-calgary_64.c @@ -1567,7 +1567,7 @@ static int __init calgary_parse_options(char *p) ++p; if (*p == '\0') break; - bridge = simple_strtol(p, &endp, 0); + bridge = simple_strtoul(p, &endp, 0); if (p == endp) break; From 292c669cd7087a090d6420e223eb1072f3e3c50b Mon Sep 17 00:00:00 2001 From: Markus Metzger Date: Tue, 25 Nov 2008 08:45:13 +0100 Subject: [PATCH 154/160] x86, bts: exclude ds.c from build when disabled Impact: cleanup Move the CONFIG guard from the .c file into the makefile. Reported-by: Andi Kleen Signed-off-by: Markus Metzger Signed-off-by: Ingo Molnar --- arch/x86/kernel/Makefile | 2 +- arch/x86/kernel/ds.c | 3 --- 2 files changed, 1 insertion(+), 4 deletions(-) diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index e489ff9cb3e..b62a7667828 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -41,7 +41,7 @@ obj-$(CONFIG_X86_TRAMPOLINE) += trampoline.o obj-y += process.o obj-y += i387.o xsave.o obj-y += ptrace.o -obj-y += ds.o +obj-$(CONFIG_X86_DS) += ds.o obj-$(CONFIG_X86_32) += tls.o obj-$(CONFIG_IA32_EMULATION) += tls.o obj-y += step.o diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c index d1a121443bd..4c8d57ec966 100644 --- a/arch/x86/kernel/ds.c +++ b/arch/x86/kernel/ds.c @@ -21,8 +21,6 @@ */ -#ifdef CONFIG_X86_DS - #include #include @@ -878,4 +876,3 @@ void ds_free(struct ds_context *context) while (leftovers--) ds_put_context(context); } -#endif /* CONFIG_X86_DS */ From e5e8ca633bbe972eff6f84e064a63c0c08ed6c3d Mon Sep 17 00:00:00 2001 From: Markus Metzger Date: Tue, 25 Nov 2008 08:47:19 +0100 Subject: [PATCH 155/160] x86, bts: turn macro into static inline function Impact: cleanup Replace a macro with a static inline function. Signed-off-by: Markus Metzger Signed-off-by: Ingo Molnar --- arch/x86/include/asm/ds.h | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/arch/x86/include/asm/ds.h b/arch/x86/include/asm/ds.h index 72c5a190bf4..a95008457ea 100644 --- a/arch/x86/include/asm/ds.h +++ b/arch/x86/include/asm/ds.h @@ -23,12 +23,13 @@ #ifndef _ASM_X86_DS_H #define _ASM_X86_DS_H -#ifdef CONFIG_X86_DS #include #include +#ifdef CONFIG_X86_DS + struct task_struct; /* @@ -232,7 +233,8 @@ extern void ds_free(struct ds_context *context); #else /* CONFIG_X86_DS */ -#define ds_init_intel(config) do {} while (0) +struct cpuinfo_x86; +static inline void __cpuinit ds_init_intel(struct cpuinfo_x86 *ignored) {} #endif /* CONFIG_X86_DS */ #endif /* _ASM_X86_DS_H */ From c4858ffc8f2dc850cb1f609c679b1ac1ad36ef0c Mon Sep 17 00:00:00 2001 From: Markus Metzger Date: Tue, 25 Nov 2008 08:49:06 +0100 Subject: [PATCH 156/160] x86, pebs: fix PEBS record size configuration Impact: fix DS hw enablement on 64-bit x86 Fix the PEBS record size in the DS configuration. Reported-by: Stephane Eranian Signed-off-by: Markus Metzger Signed-off-by: Ingo Molnar --- arch/x86/kernel/ds.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c index 4c8d57ec966..04e38ef646a 100644 --- a/arch/x86/kernel/ds.c +++ b/arch/x86/kernel/ds.c @@ -816,13 +816,21 @@ static const struct ds_configuration ds_cfg_var = { .sizeof_ds = sizeof(long) * 12, .sizeof_field = sizeof(long), .sizeof_rec[ds_bts] = sizeof(long) * 3, +#ifdef __i386__ .sizeof_rec[ds_pebs] = sizeof(long) * 10 +#else + .sizeof_rec[ds_pebs] = sizeof(long) * 18 +#endif }; static const struct ds_configuration ds_cfg_64 = { .sizeof_ds = 8 * 12, .sizeof_field = 8, .sizeof_rec[ds_bts] = 8 * 3, +#ifdef __i386__ .sizeof_rec[ds_pebs] = 8 * 10 +#else + .sizeof_rec[ds_pebs] = 8 * 18 +#endif }; static inline void From de90add30e79261c3b5be68bb0f22d2ef98e8113 Mon Sep 17 00:00:00 2001 From: Markus Metzger Date: Tue, 25 Nov 2008 08:52:56 +0100 Subject: [PATCH 157/160] x86, bts: fix wrmsr and spinlock over kmalloc Impact: fix sleeping-with-spinlock-held bugs/crashes - Turn a wrmsr to write the DS_AREA MSR into a wrmsrl. - Use irqsave variants of spinlocks. - Do not allocate memory while holding spinlocks. Reported-by: Stephane Eranian Reported-by: Ingo Molnar Signed-off-by: Markus Metzger Signed-off-by: Ingo Molnar --- arch/x86/kernel/ds.c | 83 +++++++++++++++++++++++--------------------- 1 file changed, 43 insertions(+), 40 deletions(-) diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c index 04e38ef646a..a2d1176c38e 100644 --- a/arch/x86/kernel/ds.c +++ b/arch/x86/kernel/ds.c @@ -209,14 +209,15 @@ static DEFINE_PER_CPU(struct ds_context *, system_context); static inline struct ds_context *ds_get_context(struct task_struct *task) { struct ds_context *context; + unsigned long irq; - spin_lock(&ds_lock); + spin_lock_irqsave(&ds_lock, irq); context = (task ? task->thread.ds_ctx : this_system_context); if (context) context->count++; - spin_unlock(&ds_lock); + spin_unlock_irqrestore(&ds_lock, irq); return context; } @@ -224,55 +225,46 @@ static inline struct ds_context *ds_get_context(struct task_struct *task) /* * Same as ds_get_context, but allocates the context and it's DS * structure, if necessary; returns NULL; if out of memory. - * - * pre: requires ds_lock to be held */ static inline struct ds_context *ds_alloc_context(struct task_struct *task) { struct ds_context **p_context = (task ? &task->thread.ds_ctx : &this_system_context); struct ds_context *context = *p_context; + unsigned long irq; if (!context) { - spin_unlock(&ds_lock); - context = kzalloc(sizeof(*context), GFP_KERNEL); - - if (!context) { - spin_lock(&ds_lock); + if (!context) return NULL; - } context->ds = kzalloc(ds_cfg.sizeof_ds, GFP_KERNEL); if (!context->ds) { kfree(context); - spin_lock(&ds_lock); return NULL; } - spin_lock(&ds_lock); - /* - * Check for race - another CPU could have allocated - * it meanwhile: - */ + spin_lock_irqsave(&ds_lock, irq); + if (*p_context) { kfree(context->ds); kfree(context); - return *p_context; + + context = *p_context; + } else { + *p_context = context; + + context->this = p_context; + context->task = task; + + if (task) + set_tsk_thread_flag(task, TIF_DS_AREA_MSR); + + if (!task || (task == current)) + wrmsrl(MSR_IA32_DS_AREA, + (unsigned long)context->ds); } - - *p_context = context; - - context->this = p_context; - context->task = task; - - if (task) - set_tsk_thread_flag(task, TIF_DS_AREA_MSR); - - if (!task || (task == current)) - wrmsr(MSR_IA32_DS_AREA, (unsigned long)context->ds, 0); - - get_tracer(task); + spin_unlock_irqrestore(&ds_lock, irq); } context->count++; @@ -286,10 +278,12 @@ static inline struct ds_context *ds_alloc_context(struct task_struct *task) */ static inline void ds_put_context(struct ds_context *context) { + unsigned long irq; + if (!context) return; - spin_lock(&ds_lock); + spin_lock_irqsave(&ds_lock, irq); if (--context->count) goto out; @@ -311,7 +305,7 @@ static inline void ds_put_context(struct ds_context *context) kfree(context->ds); kfree(context); out: - spin_unlock(&ds_lock); + spin_unlock_irqrestore(&ds_lock, irq); } @@ -382,6 +376,7 @@ static int ds_request(struct task_struct *task, void *base, size_t size, struct ds_context *context; unsigned long buffer, adj; const unsigned long alignment = (1 << 3); + unsigned long irq; int error = 0; if (!ds_cfg.sizeof_ds) @@ -396,26 +391,27 @@ static int ds_request(struct task_struct *task, void *base, size_t size, return -EOPNOTSUPP; - spin_lock(&ds_lock); - - error = -ENOMEM; context = ds_alloc_context(task); if (!context) - goto out_unlock; + return -ENOMEM; + + spin_lock_irqsave(&ds_lock, irq); error = -EPERM; if (!check_tracer(task)) goto out_unlock; + get_tracer(task); + error = -EALREADY; if (context->owner[qual] == current) - goto out_unlock; + goto out_put_tracer; error = -EPERM; if (context->owner[qual] != NULL) - goto out_unlock; + goto out_put_tracer; context->owner[qual] = current; - spin_unlock(&ds_lock); + spin_unlock_irqrestore(&ds_lock, irq); error = -ENOMEM; @@ -463,10 +459,17 @@ static int ds_request(struct task_struct *task, void *base, size_t size, out_release: context->owner[qual] = NULL; ds_put_context(context); + put_tracer(task); + return error; + + out_put_tracer: + spin_unlock_irqrestore(&ds_lock, irq); + ds_put_context(context); + put_tracer(task); return error; out_unlock: - spin_unlock(&ds_lock); + spin_unlock_irqrestore(&ds_lock, irq); ds_put_context(context); return error; } From ffd565a8b817d1eb4b25184e8418e8d96c3f56f6 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Tue, 25 Nov 2008 17:18:03 +0100 Subject: [PATCH 158/160] x86: fixup config space size of CPU functions for AMD family 11h Impact: extend allowed configuration space access on 11h CPUs from 256 to 4K Signed-off-by: Andreas Herrmann Acked-by: Jesse Barnes Signed-off-by: Ingo Molnar --- arch/x86/pci/fixup.c | 25 ++++++++++++++----------- 1 file changed, 14 insertions(+), 11 deletions(-) diff --git a/arch/x86/pci/fixup.c b/arch/x86/pci/fixup.c index 3c27a809393..2051dc96b8e 100644 --- a/arch/x86/pci/fixup.c +++ b/arch/x86/pci/fixup.c @@ -496,21 +496,24 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_SIEMENS, 0x0015, pci_siemens_interrupt_controller); /* - * Regular PCI devices have 256 bytes, but AMD Family 10h Opteron ext config - * have 4096 bytes. Even if the device is capable, that doesn't mean we can - * access it. Maybe we don't have a way to generate extended config space - * accesses. So check it + * Regular PCI devices have 256 bytes, but AMD Family 10h/11h CPUs have + * 4096 bytes configuration space for each function of their processor + * configuration space. */ -static void fam10h_pci_cfg_space_size(struct pci_dev *dev) +static void amd_cpu_pci_cfg_space_size(struct pci_dev *dev) { dev->cfg_size = pci_cfg_space_size_ext(dev); } - -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1200, fam10h_pci_cfg_space_size); -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1201, fam10h_pci_cfg_space_size); -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1202, fam10h_pci_cfg_space_size); -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1203, fam10h_pci_cfg_space_size); -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1204, fam10h_pci_cfg_space_size); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1200, amd_cpu_pci_cfg_space_size); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1201, amd_cpu_pci_cfg_space_size); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1202, amd_cpu_pci_cfg_space_size); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1203, amd_cpu_pci_cfg_space_size); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1204, amd_cpu_pci_cfg_space_size); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1300, amd_cpu_pci_cfg_space_size); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1301, amd_cpu_pci_cfg_space_size); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1302, amd_cpu_pci_cfg_space_size); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1303, amd_cpu_pci_cfg_space_size); +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_AMD, 0x1304, amd_cpu_pci_cfg_space_size); /* * SB600: Disable BAR1 on device 14.0 to avoid HPET resources from From b627c8b17ccacba38c975bc0f69a49fc4e5261c9 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Thu, 20 Nov 2008 20:49:56 +0100 Subject: [PATCH 159/160] x86: always define DECLARE_PCI_UNMAP* macros Impact: fix boot crash on AMD IOMMU if CONFIG_GART_IOMMU is off Currently these macros evaluate to a no-op except the kernel is compiled with GART or Calgary support. But we also need these macros when we have SWIOTLB, VT-d or AMD IOMMU in the kernel. Since we always compile at least with SWIOTLB we can define these macros always. This patch is also for stable backport for the same reason the SWIOTLB default selection patch is. Signed-off-by: Joerg Roedel Cc: Signed-off-by: Ingo Molnar --- arch/x86/include/asm/pci_64.h | 14 -------------- 1 file changed, 14 deletions(-) diff --git a/arch/x86/include/asm/pci_64.h b/arch/x86/include/asm/pci_64.h index 5b28995d664..d02d936840a 100644 --- a/arch/x86/include/asm/pci_64.h +++ b/arch/x86/include/asm/pci_64.h @@ -34,8 +34,6 @@ extern void pci_iommu_alloc(void); */ #define PCI_DMA_BUS_IS_PHYS (dma_ops->is_phys) -#if defined(CONFIG_GART_IOMMU) || defined(CONFIG_CALGARY_IOMMU) - #define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME) \ dma_addr_t ADDR_NAME; #define DECLARE_PCI_UNMAP_LEN(LEN_NAME) \ @@ -49,18 +47,6 @@ extern void pci_iommu_alloc(void); #define pci_unmap_len_set(PTR, LEN_NAME, VAL) \ (((PTR)->LEN_NAME) = (VAL)) -#else -/* No IOMMU */ - -#define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME) -#define DECLARE_PCI_UNMAP_LEN(LEN_NAME) -#define pci_unmap_addr(PTR, ADDR_NAME) (0) -#define pci_unmap_addr_set(PTR, ADDR_NAME, VAL) do { } while (0) -#define pci_unmap_len(PTR, LEN_NAME) (0) -#define pci_unmap_len_set(PTR, LEN_NAME, VAL) do { } while (0) - -#endif - #endif /* __KERNEL__ */ #endif /* _ASM_X86_PCI_64_H */ From 7b1dedca42ac0d0d0be01e39d8461bb53a2389b3 Mon Sep 17 00:00:00 2001 From: Thomas Bogendoerfer Date: Sat, 29 Nov 2008 13:46:27 +0100 Subject: [PATCH 160/160] x86: fix dma_mapping_error for 32bit x86 Devices like b44 ethernet can't dma from addresses above 1GB. The driver handles this cases by falling back to GFP_DMA allocation. But for detecting the problem it needs to get an indication from dma_mapping_error. The bug is triggered by using a VMSPLIT option of 2G/2G. Signed-off-by: Thomas Bogendoerfer Acked-by: FUJITA Tomonori Signed-off-by: Ingo Molnar --- arch/x86/include/asm/dma-mapping.h | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/arch/x86/include/asm/dma-mapping.h b/arch/x86/include/asm/dma-mapping.h index 7f225a4b2a2..097794ff6b7 100644 --- a/arch/x86/include/asm/dma-mapping.h +++ b/arch/x86/include/asm/dma-mapping.h @@ -71,15 +71,13 @@ static inline struct dma_mapping_ops *get_dma_ops(struct device *dev) /* Make sure we keep the same behaviour */ static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr) { -#ifdef CONFIG_X86_32 - return 0; -#else +#ifdef CONFIG_X86_64 struct dma_mapping_ops *ops = get_dma_ops(dev); if (ops->mapping_error) return ops->mapping_error(dev, dma_addr); - return (dma_addr == bad_dma_address); #endif + return (dma_addr == bad_dma_address); } #define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)