From c79ee4e466dd12347f112e2af306dca35198458f Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Tue, 16 Jun 2009 12:23:58 +0200 Subject: [PATCH 1/7] dma-debug: fix off-by-one error in overlap function This patch fixes a bug in the overlap function which returned true if one region ends exactly before the second region begins. This is no overlap but the function returned true in that case. Cc: stable@kernel.org Reported-by: Andrew Randrianasulu Signed-off-by: Joerg Roedel --- lib/dma-debug.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/dma-debug.c b/lib/dma-debug.c index 3b93129a968..a9b6b5c9e09 100644 --- a/lib/dma-debug.c +++ b/lib/dma-debug.c @@ -862,7 +862,7 @@ static inline bool overlap(void *addr, u64 size, void *start, void *end) return ((addr >= start && addr < end) || (addr2 >= start && addr2 < end) || - ((addr < start) && (addr2 >= end))); + ((addr < start) && (addr2 > end))); } static void check_for_illegal_area(struct device *dev, void *addr, u64 size) From b0a5b83ee0fce9dbf8ff5fe1f8c9ae7dfafe458c Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 16 Jun 2009 16:11:14 +0200 Subject: [PATCH 2/7] dma-debug: Put all hash-chain locks into the same lock class Alan Cox reported that lockdep runs out of its stack-trace entries with certain configs: BUG: MAX_STACK_TRACE_ENTRIES too low This happens because there are 1024 hash buckets, each with a separate lock. Lockdep puts each lock into a separate lock class and tracks them independently. But in reality we never take more than one of the buckets, so they really belong into a single lock-class. Annotate the has bucket lock init accordingly. [ Impact: reduce the lockdep footprint of dma-debug ] Reported-by: Alan Cox Signed-off-by: Ingo Molnar Signed-off-by: Joerg Roedel --- lib/dma-debug.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/lib/dma-debug.c b/lib/dma-debug.c index a9b6b5c9e09..c9187fed0b9 100644 --- a/lib/dma-debug.c +++ b/lib/dma-debug.c @@ -716,7 +716,7 @@ void dma_debug_init(u32 num_entries) for (i = 0; i < HASH_SIZE; ++i) { INIT_LIST_HEAD(&dma_entry_hash[i].list); - dma_entry_hash[i].lock = SPIN_LOCK_UNLOCKED; + spin_lock_init(&dma_entry_hash[i].lock); } if (dma_debug_fs_init() != 0) { From f6faac71d502be1c29c81b2f45657662c3b84470 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Tue, 23 Jun 2009 17:24:40 -0700 Subject: [PATCH 3/7] rcu: Mark Hierarchical RCU no longer experimental Removes the warnings about Hierarchical RCU being experimental, given that it has gone through almost six months of being the default RCU in mainline for the x86 with very little trouble. This makes hierarchical-RCU bootup look less scary. Signed-off-by: Paul E. McKenney Cc: akpm@linux-foundation.org Cc: niv@us.ibm.com Cc: dvhltc@us.ibm.com Cc: dipankar@in.ibm.com Cc: dhowells@redhat.com Cc: lethal@linux-sh.org Cc: kernel@wantstofly.org Cc: cl@linux-foundation.org Cc: schamp@sgi.com Signed-off-by: Ingo Molnar --- kernel/rcutree.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/kernel/rcutree.c b/kernel/rcutree.c index 0dccfbba6d2..7717b95c202 100644 --- a/kernel/rcutree.c +++ b/kernel/rcutree.c @@ -1533,7 +1533,7 @@ void __init __rcu_init(void) int j; struct rcu_node *rnp; - printk(KERN_WARNING "Experimental hierarchical RCU implementation.\n"); + printk(KERN_INFO "Hierarchical RCU implementation.\n"); #ifdef CONFIG_RCU_CPU_STALL_DETECTOR printk(KERN_INFO "RCU-based detection of stalled CPUs is enabled.\n"); #endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ @@ -1546,7 +1546,6 @@ void __init __rcu_init(void) rcu_cpu_notify(&rcu_nb, CPU_UP_PREPARE, (void *)(long)i); /* Register notifier for non-boot CPUs */ register_cpu_notifier(&rcu_nb); - printk(KERN_WARNING "Experimental hierarchical RCU init done.\n"); } module_param(blimit, int, 0); From 7afdbf23c3acdec3eaf1b94f87132fff3d81ce73 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 7 Jul 2009 10:23:43 +0200 Subject: [PATCH 4/7] signals: declare sys_rt_tgsigqueueinfo in syscalls.h sys_rt_tgsigqueueinfo needs to be declared in linux/syscalls.h so that architectures defining the system call table in C can reference it. Signed-off-by: Arnd Bergmann LKML-Reference: <200907071023.44008.arnd@arndb.de> Signed-off-by: Thomas Gleixner --- include/linux/syscalls.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index fa4242cdade..80de7003d8c 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -321,6 +321,8 @@ asmlinkage long sys_rt_sigtimedwait(const sigset_t __user *uthese, siginfo_t __user *uinfo, const struct timespec __user *uts, size_t sigsetsize); +asmlinkage long sys_rt_tgsigqueueinfo(pid_t tgid, pid_t pid, int sig, + siginfo_t __user *uinfo); asmlinkage long sys_kill(int pid, int sig); asmlinkage long sys_tgkill(int tgid, int pid, int sig); asmlinkage long sys_tkill(int pid, int sig); From 8d7ff4f2a0b22b7d6d7bc3982257d1dadea22824 Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Tue, 23 Jun 2009 11:48:14 +0200 Subject: [PATCH 5/7] x86/oprofile: rename kernel parameter for architectural perfmon to arch_perfmon The short name of the achitecture is 'arch_perfmon'. This patch changes the kernel parameter to use this name. Cc: Andi Kleen Signed-off-by: Robert Richter Signed-off-by: Ingo Molnar --- Documentation/kernel-parameters.txt | 4 ++-- arch/x86/oprofile/nmi_int.c | 2 +- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 92e1ab8178a..c59e965a748 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1728,8 +1728,8 @@ and is between 256 and 4096 characters. It is defined in the file oprofile.cpu_type= Force an oprofile cpu type This might be useful if you have an older oprofile userland or if you want common events. - Format: { archperfmon } - archperfmon: [X86] Force use of architectural + Format: { arch_perfmon } + arch_perfmon: [X86] Force use of architectural perfmon on Intel CPUs instead of the CPU specific event set. diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c index b07dd8d0b32..89b9a5cd63d 100644 --- a/arch/x86/oprofile/nmi_int.c +++ b/arch/x86/oprofile/nmi_int.c @@ -390,7 +390,7 @@ static int __init p4_init(char **cpu_type) static int force_arch_perfmon; static int force_cpu_type(const char *str, struct kernel_param *kp) { - if (!strcmp(str, "archperfmon")) { + if (!strcmp(str, "arch_perfmon")) { force_arch_perfmon = 1; printk(KERN_INFO "oprofile: forcing architectural perfmon\n"); } From 2b8777ca0c944bf6498c45ed9c5c246bd63a719e Mon Sep 17 00:00:00 2001 From: Maynard Johnson Date: Wed, 27 May 2009 10:15:08 -0500 Subject: [PATCH 6/7] oprofile: reset bt_lost_no_mapping with other stats The bt_lost_no_mapping is not getting reset at the start of a profiling run, thus the oprofiled.log shows erroneous values for this statistic. The attached patch fixes this problem. Signed-off-by: Maynard Johnson Signed-off-by: Robert Richter Signed-off-by: Ingo Molnar --- drivers/oprofile/oprofile_stats.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/oprofile/oprofile_stats.c b/drivers/oprofile/oprofile_stats.c index e1f6ce03705..3c2270a8300 100644 --- a/drivers/oprofile/oprofile_stats.c +++ b/drivers/oprofile/oprofile_stats.c @@ -33,6 +33,7 @@ void oprofile_reset_stats(void) atomic_set(&oprofile_stats.sample_lost_no_mm, 0); atomic_set(&oprofile_stats.sample_lost_no_mapping, 0); atomic_set(&oprofile_stats.event_lost_overflow, 0); + atomic_set(&oprofile_stats.bt_lost_no_mapping, 0); } From f39d1b9792881ce4eb982ec8cc65258bf95674b5 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 10 Jul 2009 21:38:02 +0200 Subject: [PATCH 7/7] dma-debug: Fix the overlap() function to be correct and readable Linus noticed how unclean and buggy the overlap() function is: - It uses convoluted (and bug-causing) positive checks for range overlap - instead of using a more natural negative check. - Even the positive checks are buggy: a positive intersection check has four natural cases while we checked only for three, missing the (addr < start && addr2 == end) case for example. - The variables are mis-named, making it non-obvious how the check was done. - It needlessly uses u64 instead of unsigned long. Since these are kernel memory pointers and we explicitly exclude highmem ranges anyway we cannot ever overflow 32 bits, even if we could. (and on 64-bit it doesnt matter anyway) All in one, this function needs a total revamp. I used Linus's suggestions minus the paranoid checks (we cannot overflow really because if we get totally bad DMA ranges passed far more things break in the systems than just DMA debugging). I also fixed a few other small details i noticed. Reported-by: Linus Torvalds Cc: Joerg Roedel Signed-off-by: Ingo Molnar --- lib/dma-debug.c | 24 ++++++++++++------------ 1 file changed, 12 insertions(+), 12 deletions(-) diff --git a/lib/dma-debug.c b/lib/dma-debug.c index c9187fed0b9..65b0d99b6d0 100644 --- a/lib/dma-debug.c +++ b/lib/dma-debug.c @@ -856,22 +856,21 @@ static void check_for_stack(struct device *dev, void *addr) "stack [addr=%p]\n", addr); } -static inline bool overlap(void *addr, u64 size, void *start, void *end) +static inline bool overlap(void *addr, unsigned long len, void *start, void *end) { - void *addr2 = (char *)addr + size; + unsigned long a1 = (unsigned long)addr; + unsigned long b1 = a1 + len; + unsigned long a2 = (unsigned long)start; + unsigned long b2 = (unsigned long)end; - return ((addr >= start && addr < end) || - (addr2 >= start && addr2 < end) || - ((addr < start) && (addr2 > end))); + return !(b1 <= a2 || a1 >= b2); } -static void check_for_illegal_area(struct device *dev, void *addr, u64 size) +static void check_for_illegal_area(struct device *dev, void *addr, unsigned long len) { - if (overlap(addr, size, _text, _etext) || - overlap(addr, size, __start_rodata, __end_rodata)) - err_printk(dev, NULL, "DMA-API: device driver maps " - "memory from kernel text or rodata " - "[addr=%p] [size=%llu]\n", addr, size); + if (overlap(addr, len, _text, _etext) || + overlap(addr, len, __start_rodata, __end_rodata)) + err_printk(dev, NULL, "DMA-API: device driver maps memory from kernel text or rodata [addr=%p] [len=%lu]\n", addr, len); } static void check_sync(struct device *dev, @@ -969,7 +968,8 @@ void debug_dma_map_page(struct device *dev, struct page *page, size_t offset, entry->type = dma_debug_single; if (!PageHighMem(page)) { - void *addr = ((char *)page_address(page)) + offset; + void *addr = page_address(page) + offset; + check_for_stack(dev, addr); check_for_illegal_area(dev, addr, size); }