From f6d87f4bd259cf33e092cd1a8fde05f291c47af1 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 7 Nov 2008 13:18:30 +0100 Subject: genirq: keep affinities set from userspace across free/request_irq() Impact: preserve user-modified affinities on interrupts Kumar Galak noticed that commit 18404756765c713a0be4eb1082920c04822ce588 (genirq: Expose default irq affinity mask (take 3)) overrides an already set affinity setting across a free / request_irq(). Happens e.g. with ifdown/ifup of a network device. Change the logic to mark the affinities as set and keep them intact. This also fixes the unlocked access to irq_desc in irq_select_affinity() when called from irq_affinity_proc_write() Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar --- kernel/irq/internals.h | 2 ++ kernel/irq/manage.c | 58 +++++++++++++++++++++++++++++++++++++++++--------- kernel/irq/migration.c | 11 ---------- kernel/irq/proc.c | 2 +- 4 files changed, 51 insertions(+), 22 deletions(-) (limited to 'kernel') diff --git a/kernel/irq/internals.h b/kernel/irq/internals.h index c9767e64198..64c1c7253da 100644 --- a/kernel/irq/internals.h +++ b/kernel/irq/internals.h @@ -25,6 +25,8 @@ static inline void unregister_handler_proc(unsigned int irq, struct irqaction *action) { } #endif +extern int irq_select_affinity_usr(unsigned int irq); + /* * Debugging printout: */ diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index c498a1b8c62..634a2a95510 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -82,24 +82,27 @@ int irq_can_set_affinity(unsigned int irq) int irq_set_affinity(unsigned int irq, cpumask_t cpumask) { struct irq_desc *desc = irq_to_desc(irq); + unsigned long flags; if (!desc->chip->set_affinity) return -EINVAL; + spin_lock_irqsave(&desc->lock, flags); + #ifdef CONFIG_GENERIC_PENDING_IRQ if (desc->status & IRQ_MOVE_PCNTXT || desc->status & IRQ_DISABLED) { - unsigned long flags; - - spin_lock_irqsave(&desc->lock, flags); desc->affinity = cpumask; desc->chip->set_affinity(irq, cpumask); - spin_unlock_irqrestore(&desc->lock, flags); - } else - set_pending_irq(irq, cpumask); + } else { + desc->status |= IRQ_MOVE_PENDING; + desc->pending_mask = cpumask; + } #else desc->affinity = cpumask; desc->chip->set_affinity(irq, cpumask); #endif + desc->status |= IRQ_AFFINITY_SET; + spin_unlock_irqrestore(&desc->lock, flags); return 0; } @@ -107,24 +110,59 @@ int irq_set_affinity(unsigned int irq, cpumask_t cpumask) /* * Generic version of the affinity autoselector. */ -int irq_select_affinity(unsigned int irq) +int do_irq_select_affinity(unsigned int irq, struct irq_desc *desc) { cpumask_t mask; - struct irq_desc *desc; if (!irq_can_set_affinity(irq)) return 0; cpus_and(mask, cpu_online_map, irq_default_affinity); - desc = irq_to_desc(irq); + /* + * Preserve an userspace affinity setup, but make sure that + * one of the targets is online. + */ + if (desc->status & IRQ_AFFINITY_SET) { + if (cpus_intersects(desc->affinity, cpu_online_map)) + mask = desc->affinity; + else + desc->status &= ~IRQ_AFFINITY_SET; + } + desc->affinity = mask; desc->chip->set_affinity(irq, mask); return 0; } +#else +static inline int do_irq_select_affinity(unsigned int irq, struct irq_desc *d) +{ + return irq_select_affinity(irq); +} #endif +/* + * Called when affinity is set via /proc/irq + */ +int irq_select_affinity_usr(unsigned int irq) +{ + struct irq_desc *desc = irq_to_desc(irq); + unsigned long flags; + int ret; + + spin_lock_irqsave(&desc->lock, flags); + ret = do_irq_select_affinity(irq, desc); + spin_unlock_irqrestore(&desc->lock, flags); + + return ret; +} + +#else +static inline int do_select_irq_affinity(int irq, struct irq_desc *desc) +{ + return 0; +} #endif /** @@ -446,7 +484,7 @@ __setup_irq(unsigned int irq, struct irq_desc * desc, struct irqaction *new) desc->depth = 1; /* Set default affinity mask once everything is setup */ - irq_select_affinity(irq); + do_irq_select_affinity(irq, desc); } else if ((new->flags & IRQF_TRIGGER_MASK) && (new->flags & IRQF_TRIGGER_MASK) diff --git a/kernel/irq/migration.c b/kernel/irq/migration.c index 90b920d3f52..9db681d9581 100644 --- a/kernel/irq/migration.c +++ b/kernel/irq/migration.c @@ -1,17 +1,6 @@ #include -void set_pending_irq(unsigned int irq, cpumask_t mask) -{ - struct irq_desc *desc = irq_to_desc(irq); - unsigned long flags; - - spin_lock_irqsave(&desc->lock, flags); - desc->status |= IRQ_MOVE_PENDING; - desc->pending_mask = mask; - spin_unlock_irqrestore(&desc->lock, flags); -} - void move_masked_irq(int irq) { struct irq_desc *desc = irq_to_desc(irq); diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c index 4d161c70ba5..d257e7d6a8a 100644 --- a/kernel/irq/proc.c +++ b/kernel/irq/proc.c @@ -62,7 +62,7 @@ static ssize_t irq_affinity_proc_write(struct file *file, if (!cpus_intersects(new_value, cpu_online_map)) /* Special case for empty set - allow the architecture code to set default SMP affinity. */ - return irq_select_affinity(irq) ? -EINVAL : count; + return irq_select_affinity_usr(irq) ? -EINVAL : count; irq_set_affinity(irq, new_value); -- cgit v1.2.3-18-g5258 From 612e3684c1b7752d2890510e4f90115fd1eb2afb Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Fri, 7 Nov 2008 13:58:46 +0100 Subject: genirq: fix the affinity setting in setup_irq The affinity setting in setup irq is called before the NO_BALANCING flag is checked and might therefore override affinity settings from the calling code with the default setting. Move the NO_BALANCING flag check before the call to the affinity setting. Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar --- kernel/irq/manage.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 634a2a95510..948a22a2c01 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -123,7 +123,7 @@ int do_irq_select_affinity(unsigned int irq, struct irq_desc *desc) * Preserve an userspace affinity setup, but make sure that * one of the targets is online. */ - if (desc->status & IRQ_AFFINITY_SET) { + if (desc->status & (IRQ_AFFINITY_SET | IRQ_NO_BALANCING)) { if (cpus_intersects(desc->affinity, cpu_online_map)) mask = desc->affinity; else @@ -483,6 +483,10 @@ __setup_irq(unsigned int irq, struct irq_desc * desc, struct irqaction *new) /* Undo nested disables: */ desc->depth = 1; + /* Exclude IRQ from balancing if requested */ + if (new->flags & IRQF_NOBALANCING) + desc->status |= IRQ_NO_BALANCING; + /* Set default affinity mask once everything is setup */ do_irq_select_affinity(irq, desc); @@ -497,10 +501,6 @@ __setup_irq(unsigned int irq, struct irq_desc * desc, struct irqaction *new) *p = new; - /* Exclude IRQ from balancing */ - if (new->flags & IRQF_NOBALANCING) - desc->status |= IRQ_NO_BALANCING; - /* Reset broken irq detection when installing new handler */ desc->irq_count = 0; desc->irqs_unhandled = 0; -- cgit v1.2.3-18-g5258 From f131e2436ddbac2527bb2d6297a823aae4b024f8 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Sat, 8 Nov 2008 09:57:40 +0100 Subject: irq: fix typo Impact: build fix fix build failure on UP. Signed-off-by: Ingo Molnar --- kernel/irq/manage.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 948a22a2c01..435861284e4 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -159,7 +159,7 @@ int irq_select_affinity_usr(unsigned int irq) } #else -static inline int do_select_irq_affinity(int irq, struct irq_desc *desc) +static inline int do_irq_select_affinity(int irq, struct irq_desc *desc) { return 0; } -- cgit v1.2.3-18-g5258 From 3ff68a6a106c362a6811d3e51bced58e6fc87de7 Mon Sep 17 00:00:00 2001 From: Mark Nelson Date: Thu, 13 Nov 2008 21:37:41 +1100 Subject: genirq: __irq_set_trigger: change pr_warning to pr_debug Commit 0c5d1eb77a8be917b638344a22afe1398236482b (genirq: record trigger type) caused powerpc platforms that had no set_type() function in their struct irq_chip to spew out warnings about "No set_type function for IRQ...". This warning isn't necessarily justified though because the generic powerpc platform code calls set_irq_type() (which in turn calls __irq_set_trigger) with information from the device tree to establish the interrupt mappings, regardless of whether the PIC can actually set a type. A platform's irq_chip might not have a set_type function for a variety of reasons, for example: the platform may have the type essentially hard-coded, or as in the case for Cell interrupts are just messages past around that have no real concept of type, or the platform could even have a virtual PIC as on the PS3. Signed-off-by: Mark Nelson Signed-off-by: Ingo Molnar --- kernel/irq/manage.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 435861284e4..801addda3c4 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -365,7 +365,7 @@ int __irq_set_trigger(struct irq_desc *desc, unsigned int irq, * IRQF_TRIGGER_* but the PIC does not support multiple * flow-types? */ - pr_warning("No set_type function for IRQ %d (%s)\n", irq, + pr_debug("No set_type function for IRQ %d (%s)\n", irq, chip ? (chip->name ? : "unknown") : "unknown"); return 0; } -- cgit v1.2.3-18-g5258 From 522a110b42b306d696cf84e34c677ed0e7080194 Mon Sep 17 00:00:00 2001 From: Liming Wang Date: Fri, 21 Nov 2008 11:00:18 +0800 Subject: function tracing: fix wrong position computing of stack_trace Impact: make output of stack_trace complete if buffer overruns When read buffer overruns, the output of stack_trace isn't complete. When printing records with seq_printf in t_show, if the read buffer has overruned by the current record, then this record won't be printed to user space through read buffer, it will just be dropped in this printing. When next printing, t_start should return the "*pos"th record, which is the one dropped by previous printing, but it just returns (m->private + *pos)th record. Here we use a more sane method to implement seq_operations which can be found in kernel code. Thus we needn't initialize m->private. About testing, it's not easy to overrun read buffer, but we can use seq_printf to print more padding bytes in t_show, then it's easy to check whether or not records are lost. This commit has been tested on both condition of overrun and non overrun. Signed-off-by: Liming Wang Acked-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/trace/trace_stack.c | 24 +++++++++++++++--------- 1 file changed, 15 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c index be682b62fe5..3bdb44bde4b 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -184,11 +184,16 @@ static struct file_operations stack_max_size_fops = { static void * t_next(struct seq_file *m, void *v, loff_t *pos) { - long i = (long)m->private; + long i; (*pos)++; - i++; + if (v == SEQ_START_TOKEN) + i = 0; + else { + i = *(long *)v; + i++; + } if (i >= max_stack_trace.nr_entries || stack_dump_trace[i] == ULONG_MAX) @@ -201,12 +206,15 @@ t_next(struct seq_file *m, void *v, loff_t *pos) static void *t_start(struct seq_file *m, loff_t *pos) { - void *t = &m->private; + void *t = SEQ_START_TOKEN; loff_t l = 0; local_irq_disable(); __raw_spin_lock(&max_stack_lock); + if (*pos == 0) + return SEQ_START_TOKEN; + for (; t && l < *pos; t = t_next(m, t, &l)) ; @@ -235,10 +243,10 @@ static int trace_lookup_stack(struct seq_file *m, long i) static int t_show(struct seq_file *m, void *v) { - long i = *(long *)v; + long i; int size; - if (i < 0) { + if (v == SEQ_START_TOKEN) { seq_printf(m, " Depth Size Location" " (%d entries)\n" " ----- ---- --------\n", @@ -246,6 +254,8 @@ static int t_show(struct seq_file *m, void *v) return 0; } + i = *(long *)v; + if (i >= max_stack_trace.nr_entries || stack_dump_trace[i] == ULONG_MAX) return 0; @@ -275,10 +285,6 @@ static int stack_trace_open(struct inode *inode, struct file *file) int ret; ret = seq_open(file, &stack_trace_seq_ops); - if (!ret) { - struct seq_file *m = file->private_data; - m->private = (void *)-1; - } return ret; } -- cgit v1.2.3-18-g5258 From 7ee1768ddb3075ae3a0801cc2d0ea4195530a7db Mon Sep 17 00:00:00 2001 From: Pekka Paalanen Date: Sun, 23 Nov 2008 21:24:30 +0200 Subject: x86, mmiotrace: fix buffer overrun detection Impact: fix mmiotrace overrun tracing When ftrace framework moved to use the ring buffer facility, the buffer overrun detection was broken after 2.6.27 by commit | commit 3928a8a2d98081d1bc3c0a84a2d70e29b90ecf1c | Author: Steven Rostedt | Date: Mon Sep 29 23:02:41 2008 -0400 | | ftrace: make work with new ring buffer | | This patch ports ftrace over to the new ring buffer. The detection is now fixed by using the ring buffer API. When mmiotrace detects a buffer overrun, it will report the number of lost events. People reading an mmiotrace log must know if something was missed, otherwise the data may not make sense. Signed-off-by: Pekka Paalanen Acked-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/trace/trace_mmiotrace.c | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/trace_mmiotrace.c b/kernel/trace/trace_mmiotrace.c index f28484618ff..e62cbf78eab 100644 --- a/kernel/trace/trace_mmiotrace.c +++ b/kernel/trace/trace_mmiotrace.c @@ -18,12 +18,14 @@ struct header_iter { static struct trace_array *mmio_trace_array; static bool overrun_detected; +static unsigned long prev_overruns; static void mmio_reset_data(struct trace_array *tr) { int cpu; overrun_detected = false; + prev_overruns = 0; tr->time_start = ftrace_now(tr->cpu); for_each_online_cpu(cpu) @@ -128,16 +130,12 @@ static void mmio_close(struct trace_iterator *iter) static unsigned long count_overruns(struct trace_iterator *iter) { - int cpu; unsigned long cnt = 0; -/* FIXME: */ -#if 0 - for_each_online_cpu(cpu) { - cnt += iter->overrun[cpu]; - iter->overrun[cpu] = 0; - } -#endif - (void)cpu; + unsigned long over = ring_buffer_overruns(iter->tr->buffer); + + if (over > prev_overruns) + cnt = over - prev_overruns; + prev_overruns = over; return cnt; } -- cgit v1.2.3-18-g5258 From eccdaeafaea3ed115068ba55d01f22e486e5437d Mon Sep 17 00:00:00 2001 From: Petr Tesarik Date: Mon, 24 Nov 2008 15:46:31 +0100 Subject: posix-cpu-timers: fix clock_gettime with CLOCK_PROCESS_CPUTIME_ID Since CLOCK_PROCESS_CPUTIME_ID is in fact translated to -6, the switch statement in cpu_clock_sample_group() must first mask off the irrelevant bits, similar to cpu_clock_sample(). Signed-off-by: Petr Tesarik Signed-off-by: Thomas Gleixner -- posix-cpu-timers.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) --- kernel/posix-cpu-timers.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index 895337b16a2..4e5288a831d 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -311,7 +311,7 @@ static int cpu_clock_sample_group(const clockid_t which_clock, struct task_cputime cputime; thread_group_cputime(p, &cputime); - switch (which_clock) { + switch (CPUCLOCK_WHICH(which_clock)) { default: return -EINVAL; case CPUCLOCK_PROF: -- cgit v1.2.3-18-g5258 From 4f5a7f40ddbae98569acbb99118a98570315579c Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Thu, 27 Nov 2008 10:21:46 +0800 Subject: ftrace: prevent recursion Impact: prevent unnecessary stack recursion if the resched flag was set before we entered, then don't reschedule. Signed-off-by: Lai Jiangshan Acked-by: Steven Rostedt Signed-off-by: Ingo Molnar --- kernel/trace/ring_buffer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index f780e9552f9..668bbb5ef2b 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -1215,7 +1215,7 @@ ring_buffer_lock_reserve(struct ring_buffer *buffer, out: if (resched) - preempt_enable_notrace(); + preempt_enable_no_resched_notrace(); else preempt_enable_notrace(); return NULL; -- cgit v1.2.3-18-g5258 From 4cd4262034849da01eb88659af677b69f8169f06 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Wed, 26 Nov 2008 21:04:24 -0500 Subject: sched: prevent divide by zero error in cpu_avg_load_per_task Impact: fix divide by zero crash in scheduler rebalance irq While testing the branch profiler, I hit this crash: divide error: 0000 [#1] PREEMPT SMP [...] RIP: 0010:[] [] cpu_avg_load_per_task+0x50/0x7f [...] Call Trace: <0> [] find_busiest_group+0x3e5/0xcaa [] rebalance_domains+0x2da/0xa21 [] ? find_next_bit+0x1b2/0x1e6 [] run_rebalance_domains+0x112/0x19f [] __do_softirq+0xa8/0x232 [] call_softirq+0x1c/0x3e [] do_softirq+0x94/0x1cd [] irq_exit+0x6b/0x10e [] smp_apic_timer_interrupt+0xd3/0xff [] apic_timer_interrupt+0x13/0x20 The code for cpu_avg_load_per_task has: if (rq->nr_running) rq->avg_load_per_task = rq->load.weight / rq->nr_running; The runqueue lock is not held here, and there is nothing that prevents the rq->nr_running from going to zero after it passes the if condition. The branch profiler simply made the race window bigger. This patch saves off the rq->nr_running to a local variable and uses that for both the condition and the division. Signed-off-by: Steven Rostedt Peter Zijlstra Signed-off-by: Ingo Molnar --- kernel/sched.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 9b1e79371c2..700aa9a1413 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -1453,9 +1453,10 @@ static int task_hot(struct task_struct *p, u64 now, struct sched_domain *sd); static unsigned long cpu_avg_load_per_task(int cpu) { struct rq *rq = cpu_rq(cpu); + unsigned long nr_running = rq->nr_running; - if (rq->nr_running) - rq->avg_load_per_task = rq->load.weight / rq->nr_running; + if (nr_running) + rq->avg_load_per_task = rq->load.weight / nr_running; else rq->avg_load_per_task = 0; -- cgit v1.2.3-18-g5258 From 1583715ddb61f822041807a0f18b3b4845e88c76 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Tue, 25 Nov 2008 10:27:49 +0100 Subject: sched, cpusets: fix warning in kernel/cpuset.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit this warning: kernel/cpuset.c: In function ‘generate_sched_domains’: kernel/cpuset.c:588: warning: ‘ndoms’ may be used uninitialized in this function triggers because GCC does not recognize that ndoms stays uninitialized only if doms is NULL - but that flow is covered at the end of generate_sched_domains(). Help out GCC by initializing this variable to 0. (that's prudent anyway) Also, this function needs a splitup and code flow simplification: with 160 lines length it's clearly too long. Signed-off-by: Ingo Molnar --- kernel/cpuset.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/cpuset.c b/kernel/cpuset.c index da7ff6137f3..96c0ba13b8c 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -585,7 +585,7 @@ static int generate_sched_domains(cpumask_t **domains, int i, j, k; /* indices for partition finding loops */ cpumask_t *doms; /* resulting partition; i.e. sched domains */ struct sched_domain_attr *dattr; /* attributes for custom domains */ - int ndoms; /* number of sched domains in result */ + int ndoms = 0; /* number of sched domains in result */ int nslot; /* next empty doms[] cpumask_t slot */ doms = NULL; -- cgit v1.2.3-18-g5258 From af6d596fd603219b054c1c90fb16672a9fd441bd Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Sat, 29 Nov 2008 20:45:15 +0100 Subject: sched: prevent divide by zero error in cpu_avg_load_per_task, update Regarding the bug addressed in: 4cd4262: sched: prevent divide by zero error in cpu_avg_load_per_task Linus points out that the fix is not complete: > There's nothing that keeps gcc from deciding not to reload > rq->nr_running. > > Of course, in _practice_, I don't think gcc ever will (if it decides > that it will spill, gcc is likely going to decide that it will > literally spill the local variable to the stack rather than decide to > reload off the pointer), but it's a valid compiler optimization, and > it even has a name (rematerialization). > > So I suspect that your patch does fix the bug, but it still leaves the > fairly unlikely _potential_ for it to re-appear at some point. > > We have ACCESS_ONCE() as a macro to guarantee that the compiler > doesn't rematerialize a pointer access. That also would clarify > the fact that we access something unsafe outside a lock. So make sure our nr_running value is immutable and cannot change after we check it for nonzero. Signed-off-by: Ingo Molnar --- kernel/sched.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index 700aa9a1413..b7480fb5c3d 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -1453,7 +1453,7 @@ static int task_hot(struct task_struct *p, u64 now, struct sched_domain *sd); static unsigned long cpu_avg_load_per_task(int cpu) { struct rq *rq = cpu_rq(cpu); - unsigned long nr_running = rq->nr_running; + unsigned long nr_running = ACCESS_ONCE(rq->nr_running); if (nr_running) rq->avg_load_per_task = rq->load.weight / nr_running; -- cgit v1.2.3-18-g5258 From 8419641450edc838a6ce7cdf0f99d262bf0af2d5 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 22 Nov 2008 17:36:44 +0000 Subject: cpuinit fixes in kernel/* Signed-off-by: Al Viro Signed-off-by: Linus Torvalds --- kernel/cpu.c | 2 +- kernel/profile.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/cpu.c b/kernel/cpu.c index 5a732c5ef08..8ea32e8d68b 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -462,7 +462,7 @@ out: * It must be called by the arch code on the new cpu, before the new cpu * enables interrupts and before the "boot" cpu returns from __cpu_up(). */ -void notify_cpu_starting(unsigned int cpu) +void __cpuinit notify_cpu_starting(unsigned int cpu) { unsigned long val = CPU_STARTING; diff --git a/kernel/profile.c b/kernel/profile.c index 5b7d1ac7124..dc41827fbfe 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -351,7 +351,7 @@ out: put_cpu(); } -static int __devinit profile_cpu_callback(struct notifier_block *info, +static int __cpuinit profile_cpu_callback(struct notifier_block *info, unsigned long action, void *__cpu) { int node, cpu = (unsigned long)__cpu; @@ -596,7 +596,7 @@ out_cleanup: #define create_hash_tables() ({ 0; }) #endif -int create_proc_profile(void) +int __ref create_proc_profile(void) /* false positive from hotcpu_notifier */ { struct proc_dir_entry *entry; -- cgit v1.2.3-18-g5258 From 96b8936a9ed08746e47081458a5eb9e43a751e24 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 25 Nov 2008 08:10:03 +0100 Subject: remove __ARCH_WANT_COMPAT_SYS_PTRACE All architectures now use the generic compat_sys_ptrace, as should every new architecture that needs 32bit compat (if we'll ever get another). Remove the now superflous __ARCH_WANT_COMPAT_SYS_PTRACE define, and also kill a comment about __ARCH_SYS_PTRACE that was added after __ARCH_SYS_PTRACE was already gone. Signed-off-by: Christoph Hellwig Acked-by: David S. Miller Signed-off-by: Linus Torvalds --- kernel/ptrace.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 1e68e4c39e2..4c8bcd7dd8e 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -612,7 +612,7 @@ int generic_ptrace_pokedata(struct task_struct *tsk, long addr, long data) return (copied == sizeof(data)) ? 0 : -EIO; } -#if defined CONFIG_COMPAT && defined __ARCH_WANT_COMPAT_SYS_PTRACE +#if defined CONFIG_COMPAT #include int compat_ptrace_request(struct task_struct *child, compat_long_t request, @@ -709,4 +709,4 @@ asmlinkage long compat_sys_ptrace(compat_long_t request, compat_long_t pid, unlock_kernel(); return ret; } -#endif /* CONFIG_COMPAT && __ARCH_WANT_COMPAT_SYS_PTRACE */ +#endif /* CONFIG_COMPAT */ -- cgit v1.2.3-18-g5258 From 7ef9964e6d1b911b78709f144000aacadd0ebc21 Mon Sep 17 00:00:00 2001 From: Davide Libenzi Date: Mon, 1 Dec 2008 13:13:55 -0800 Subject: epoll: introduce resource usage limits It has been thought that the per-user file descriptors limit would also limit the resources that a normal user can request via the epoll interface. Vegard Nossum reported a very simple program (a modified version attached) that can make a normal user to request a pretty large amount of kernel memory, well within the its maximum number of fds. To solve such problem, default limits are now imposed, and /proc based configuration has been introduced. A new directory has been created, named /proc/sys/fs/epoll/ and inside there, there are two configuration points: max_user_instances = Maximum number of devices - per user max_user_watches = Maximum number of "watched" fds - per user The current default for "max_user_watches" limits the memory used by epoll to store "watches", to 1/32 of the amount of the low RAM. As example, a 256MB 32bit machine, will have "max_user_watches" set to roughly 90000. That should be enough to not break existing heavy epoll users. The default value for "max_user_instances" is set to 128, that should be enough too. This also changes the userspace, because a new error code can now come out from EPOLL_CTL_ADD (-ENOSPC). The EMFILE from epoll_create() was already listed, so that should be ok. [akpm@linux-foundation.org: use get_current_user()] Signed-off-by: Davide Libenzi Cc: Michael Kerrisk Cc: Cc: Cyrill Gorcunov Reported-by: Vegard Nossum Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sysctl.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'kernel') diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 9d048fa2d90..3d56fe7570d 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -176,6 +176,9 @@ extern struct ctl_table random_table[]; #ifdef CONFIG_INOTIFY_USER extern struct ctl_table inotify_table[]; #endif +#ifdef CONFIG_EPOLL +extern struct ctl_table epoll_table[]; +#endif #ifdef HAVE_ARCH_PICK_MMAP_LAYOUT int sysctl_legacy_va_layout; @@ -1325,6 +1328,13 @@ static struct ctl_table fs_table[] = { .child = inotify_table, }, #endif +#ifdef CONFIG_EPOLL + { + .procname = "epoll", + .mode = 0555, + .child = epoll_table, + }, +#endif #endif { .ctl_name = KERN_SETUID_DUMPABLE, -- cgit v1.2.3-18-g5258 From a8005992836434cab6182c6147993d21442184c1 Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Mon, 1 Dec 2008 13:14:00 -0800 Subject: taint: add missing comment The description for 'D' was missing in the comment... (causing me a minute of WTF followed by looking at more of the code) Signed-off-by: Arjan van de Ven Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/panic.c | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel') diff --git a/kernel/panic.c b/kernel/panic.c index 6513aac8e99..4d5088355bf 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -167,6 +167,7 @@ static const struct tnt tnts[] = { * 'M' - System experienced a machine check exception. * 'B' - System has hit bad_page. * 'U' - Userspace-defined naughtiness. + * 'D' - Kernel has oopsed before * 'A' - ACPI table overridden. * 'W' - Taint on warning. * 'C' - modules from drivers/staging are loaded. -- cgit v1.2.3-18-g5258 From 201955463a5c1a70d3f70d1598b27e4c2c402642 Mon Sep 17 00:00:00 2001 From: Roel Kluin Date: Tue, 2 Dec 2008 22:55:38 +0100 Subject: check_hung_task(): unsigned sysctl_hung_task_warnings cannot be less than 0 Impact: fix warnings-limit cutoff check for debug feature unsigned sysctl_hung_task_warnings cannot be less than 0 Signed-off-by: Roel Kluin Signed-off-by: Ingo Molnar --- kernel/softlockup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/softlockup.c b/kernel/softlockup.c index 3953e4aed73..dc0b3be6b7d 100644 --- a/kernel/softlockup.c +++ b/kernel/softlockup.c @@ -188,7 +188,7 @@ static void check_hung_task(struct task_struct *t, unsigned long now) if ((long)(now - t->last_switch_timestamp) < sysctl_hung_task_timeout_secs) return; - if (sysctl_hung_task_warnings < 0) + if (!sysctl_hung_task_warnings) return; sysctl_hung_task_warnings--; -- cgit v1.2.3-18-g5258 From 6c9bacb41c10ba84ff68f238e234d96f35fb64f7 Mon Sep 17 00:00:00 2001 From: john stultz Date: Mon, 1 Dec 2008 18:34:41 -0800 Subject: time: catch xtime_nsec underflows and fix them Impact: fix time warp bug Alex Shi, along with Yanmin Zhang have been noticing occasional time inconsistencies recently. Through their great diagnosis, they found that the xtime_nsec value used in update_wall_time was occasionally going negative. After looking through the code for awhile, I realized we have the possibility for an underflow when three conditions are met in update_wall_time(): 1) We have accumulated a second's worth of nanoseconds, so we incremented xtime.tv_sec and appropriately decrement xtime_nsec. (This doesn't cause xtime_nsec to go negative, but it can cause it to be small). 2) The remaining offset value is large, but just slightly less then cycle_interval. 3) clocksource_adjust() is speeding up the clock, causing a corrective amount (compensating for the increase in the multiplier being multiplied against the unaccumulated offset value) to be subtracted from xtime_nsec. This can cause xtime_nsec to underflow. Unfortunately, since we notify the NTP subsystem via second_overflow() whenever we accumulate a full second, and this effects the error accumulation that has already occured, we cannot simply revert the accumulated second from xtime nor move the second accumulation to after the clocksource_adjust call without a change in behavior. This leaves us with (at least) two options: 1) Simply return from clocksource_adjust() without making a change if we notice the adjustment would cause xtime_nsec to go negative. This would work, but I'm concerned that if a large adjustment was needed (due to the error being large), it may be possible to get stuck with an ever increasing error that becomes too large to correct (since it may always force xtime_nsec negative). This may just be paranoia on my part. 2) Catch xtime_nsec if it is negative, then add back the amount its negative to both xtime_nsec and the error. This second method is consistent with how we've handled earlier rounding issues, and also has the benefit that the error being added is always in the oposite direction also always equal or smaller then the correction being applied. So the risk of a corner case where things get out of control is lessened. This patch fixes bug 11970, as tested by Yanmin Zhang http://bugzilla.kernel.org/show_bug.cgi?id=11970 Reported-by: alex.shi@intel.com Signed-off-by: John Stultz Acked-by: "Zhang, Yanmin" Tested-by: "Zhang, Yanmin" Signed-off-by: Ingo Molnar --- kernel/time/timekeeping.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) (limited to 'kernel') diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index e7acfb482a6..fa05e88aa76 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -518,6 +518,28 @@ void update_wall_time(void) /* correct the clock when NTP error is too big */ clocksource_adjust(offset); + /* + * Since in the loop above, we accumulate any amount of time + * in xtime_nsec over a second into xtime.tv_sec, its possible for + * xtime_nsec to be fairly small after the loop. Further, if we're + * slightly speeding the clocksource up in clocksource_adjust(), + * its possible the required corrective factor to xtime_nsec could + * cause it to underflow. + * + * Now, we cannot simply roll the accumulated second back, since + * the NTP subsystem has been notified via second_overflow. So + * instead we push xtime_nsec forward by the amount we underflowed, + * and add that amount into the error. + * + * We'll correct this error next time through this function, when + * xtime_nsec is not as small. + */ + if (unlikely((s64)clock->xtime_nsec < 0)) { + s64 neg = -(s64)clock->xtime_nsec; + clock->xtime_nsec = 0; + clock->error += neg << (NTP_SCALE_SHIFT - clock->shift); + } + /* store full nanoseconds into xtime after rounding it up and * add the remainder to the error difference. */ -- cgit v1.2.3-18-g5258 From 50c396d38c1a7f0c693579ec88cb4be3c0b0645e Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sun, 30 Nov 2008 01:47:12 -0500 Subject: [PATCH] kill obsolete temporary comment in swsusp_close() it had been put there to mark the call of blkdev_put() that needed proper argument propagated to it; later patch in the same series had done just that. Signed-off-by: Al Viro --- kernel/power/swap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/power/swap.c b/kernel/power/swap.c index b7713b53d07..6da14358537 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -633,7 +633,7 @@ void swsusp_close(fmode_t mode) return; } - blkdev_put(resume_bdev, mode); /* move up */ + blkdev_put(resume_bdev, mode); } static int swsusp_header_init(void) -- cgit v1.2.3-18-g5258 From a3f07114e3359fb98683069ae397220e8992a24a Mon Sep 17 00:00:00 2001 From: Eric Paris Date: Wed, 5 Nov 2008 12:47:09 -0500 Subject: [PATCH] Audit: make audit=0 actually turn off audit Currently audit=0 on the kernel command line does absolutely nothing. Audit always loads and always uses its resources such as creating the kernel netlink socket. This patch causes audit=0 to actually disable audit. Audit will use no resources and starting the userspace auditd daemon will not cause the kernel audit system to activate. Signed-off-by: Eric Paris Signed-off-by: Al Viro --- kernel/audit.c | 28 +++++++++++++++++++++------- 1 file changed, 21 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/kernel/audit.c b/kernel/audit.c index 4414e93d875..d8646c23b42 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -61,8 +61,11 @@ #include "audit.h" -/* No auditing will take place until audit_initialized != 0. +/* No auditing will take place until audit_initialized == AUDIT_INITIALIZED. * (Initialization happens after skb_init is called.) */ +#define AUDIT_DISABLED -1 +#define AUDIT_UNINITIALIZED 0 +#define AUDIT_INITIALIZED 1 static int audit_initialized; #define AUDIT_OFF 0 @@ -965,6 +968,9 @@ static int __init audit_init(void) { int i; + if (audit_initialized == AUDIT_DISABLED) + return 0; + printk(KERN_INFO "audit: initializing netlink socket (%s)\n", audit_default ? "enabled" : "disabled"); audit_sock = netlink_kernel_create(&init_net, NETLINK_AUDIT, 0, @@ -976,7 +982,7 @@ static int __init audit_init(void) skb_queue_head_init(&audit_skb_queue); skb_queue_head_init(&audit_skb_hold_queue); - audit_initialized = 1; + audit_initialized = AUDIT_INITIALIZED; audit_enabled = audit_default; audit_ever_enabled |= !!audit_default; @@ -999,13 +1005,21 @@ __initcall(audit_init); static int __init audit_enable(char *str) { audit_default = !!simple_strtol(str, NULL, 0); - printk(KERN_INFO "audit: %s%s\n", - audit_default ? "enabled" : "disabled", - audit_initialized ? "" : " (after initialization)"); - if (audit_initialized) { + if (!audit_default) + audit_initialized = AUDIT_DISABLED; + + printk(KERN_INFO "audit: %s", audit_default ? "enabled" : "disabled"); + + if (audit_initialized == AUDIT_INITIALIZED) { audit_enabled = audit_default; audit_ever_enabled |= !!audit_default; + } else if (audit_initialized == AUDIT_UNINITIALIZED) { + printk(" (after initialization)"); + } else { + printk(" (until reboot)"); } + printk("\n"); + return 1; } @@ -1146,7 +1160,7 @@ struct audit_buffer *audit_log_start(struct audit_context *ctx, gfp_t gfp_mask, int reserve; unsigned long timeout_start = jiffies; - if (!audit_initialized) + if (audit_initialized != AUDIT_INITIALIZED) return NULL; if (unlikely(audit_filter_type(type))) -- cgit v1.2.3-18-g5258 From a64e64944f4b8ce3288519555dbaa0232414b8ac Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 12 Nov 2008 18:37:41 -0500 Subject: [PATCH] return records for fork() both to child and parent Signed-off-by: Al Viro --- kernel/auditsc.c | 17 +++++++++++++++++ kernel/fork.c | 1 + 2 files changed, 18 insertions(+) (limited to 'kernel') diff --git a/kernel/auditsc.c b/kernel/auditsc.c index cf5bc2f5f9c..de8468050af 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -1548,6 +1548,23 @@ void audit_syscall_entry(int arch, int major, context->ppid = 0; } +void audit_finish_fork(struct task_struct *child) +{ + struct audit_context *ctx = current->audit_context; + struct audit_context *p = child->audit_context; + if (!p || !ctx || !ctx->auditable) + return; + p->arch = ctx->arch; + p->major = ctx->major; + memcpy(p->argv, ctx->argv, sizeof(ctx->argv)); + p->ctime = ctx->ctime; + p->dummy = ctx->dummy; + p->auditable = ctx->auditable; + p->in_syscall = ctx->in_syscall; + p->filterkey = kstrdup(ctx->filterkey, GFP_KERNEL); + p->ppid = current->pid; +} + /** * audit_syscall_exit - deallocate audit context after a system call * @tsk: task being audited diff --git a/kernel/fork.c b/kernel/fork.c index 2a372a0e206..8d6a7dd9282 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1398,6 +1398,7 @@ long do_fork(unsigned long clone_flags, init_completion(&vfork); } + audit_finish_fork(p); tracehook_report_clone(trace, regs, clone_flags, nr, p); /* -- cgit v1.2.3-18-g5258 From 7f0ed77d241b60f70136f15b8eef30a3de1fa249 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Mon, 1 Dec 2008 14:16:06 -0800 Subject: [patch 1/1] audit: remove excess kernel-doc Delete excess kernel-doc notation in kernel/auditsc.c: Warning(linux-2.6.27-git10//kernel/auditsc.c:1481): Excess function parameter or struct member 'tsk' description in 'audit_syscall_entry' Warning(linux-2.6.27-git10//kernel/auditsc.c:1564): Excess function parameter or struct member 'tsk' description in 'audit_syscall_exit' Signed-off-by: Randy Dunlap Cc: Al Viro Cc: Eric Paris Signed-off-by: Andrew Morton Signed-off-by: Al Viro --- kernel/auditsc.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'kernel') diff --git a/kernel/auditsc.c b/kernel/auditsc.c index de8468050af..0a13d689549 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -1459,7 +1459,6 @@ void audit_free(struct task_struct *tsk) /** * audit_syscall_entry - fill in an audit record at syscall entry - * @tsk: task being audited * @arch: architecture type * @major: major syscall type (function) * @a1: additional syscall register 1 @@ -1567,7 +1566,6 @@ void audit_finish_fork(struct task_struct *child) /** * audit_syscall_exit - deallocate audit context after a system call - * @tsk: task being audited * @valid: success/failure flag * @return_code: syscall return value * -- cgit v1.2.3-18-g5258 From 48887e63d6e057543067327da6b091297f7fe645 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Sat, 6 Dec 2008 01:05:50 -0500 Subject: [PATCH] fix broken timestamps in AVC generated by kernel threads Timestamp in audit_context is valid only if ->in_syscall is set. Signed-off-by: Al Viro --- kernel/audit.c | 4 +--- kernel/auditsc.c | 5 ++++- 2 files changed, 5 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/audit.c b/kernel/audit.c index d8646c23b42..ce6d8ea3131 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -1121,9 +1121,7 @@ unsigned int audit_serial(void) static inline void audit_get_stamp(struct audit_context *ctx, struct timespec *t, unsigned int *serial) { - if (ctx) - auditsc_get_stamp(ctx, t, serial); - else { + if (!ctx || !auditsc_get_stamp(ctx, t, serial)) { *t = CURRENT_TIME; *serial = audit_serial(); } diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 0a13d689549..2a3f0afc4d2 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -1957,15 +1957,18 @@ EXPORT_SYMBOL_GPL(__audit_inode_child); * * Also sets the context as auditable. */ -void auditsc_get_stamp(struct audit_context *ctx, +int auditsc_get_stamp(struct audit_context *ctx, struct timespec *t, unsigned int *serial) { + if (!ctx->in_syscall) + return 0; if (!ctx->serial) ctx->serial = audit_serial(); t->tv_sec = ctx->ctime.tv_sec; t->tv_nsec = ctx->ctime.tv_nsec; *serial = ctx->serial; ctx->auditable = 1; + return 1; } /* global counter which is incremented every time something logs in */ -- cgit v1.2.3-18-g5258 From 9a2bd244e18ffbb96c8b783210fda4eded7c7e6f Mon Sep 17 00:00:00 2001 From: Brian King Date: Tue, 9 Dec 2008 08:47:00 -0600 Subject: sched: CPU remove deadlock fix Impact: fix possible deadlock in CPU hot-remove path This patch fixes a possible deadlock scenario in the CPU remove path. migration_call grabs rq->lock, then wakes up everything on rq->migration_queue with the lock held. Then one of the tasks on the migration queue ends up calling tg_shares_up which then also tries to acquire the same rq->lock. [c000000058eab2e0] c000000000502078 ._spin_lock_irqsave+0x98/0xf0 [c000000058eab370] c00000000008011c .tg_shares_up+0x10c/0x20c [c000000058eab430] c00000000007867c .walk_tg_tree+0xc4/0xfc [c000000058eab4d0] c0000000000840c8 .try_to_wake_up+0xb0/0x3c4 [c000000058eab590] c0000000000799a0 .__wake_up_common+0x6c/0xe0 [c000000058eab640] c00000000007ada4 .complete+0x54/0x80 [c000000058eab6e0] c000000000509fa8 .migration_call+0x5fc/0x6f8 [c000000058eab7c0] c000000000504074 .notifier_call_chain+0x68/0xe0 [c000000058eab860] c000000000506568 ._cpu_down+0x2b0/0x3f4 [c000000058eaba60] c000000000506750 .cpu_down+0xa4/0x108 [c000000058eabb10] c000000000507e54 .store_online+0x44/0xa8 [c000000058eabba0] c000000000396260 .sysdev_store+0x3c/0x50 [c000000058eabc10] c0000000001a39b8 .sysfs_write_file+0x124/0x18c [c000000058eabcd0] c00000000013061c .vfs_write+0xd0/0x1bc [c000000058eabd70] c0000000001308a4 .sys_write+0x68/0x114 [c000000058eabe30] c0000000000086b4 syscall_exit+0x0/0x40 Signed-off-by: Brian King Acked-by: Peter Zijlstra Signed-off-by: Ingo Molnar --- kernel/sched.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index b7480fb5c3d..e4bb1dd7b30 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -6587,7 +6587,9 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) req = list_entry(rq->migration_queue.next, struct migration_req, list); list_del_init(&req->list); + spin_unlock_irq(&rq->lock); complete(&req->done); + spin_lock_irq(&rq->lock); } spin_unlock_irq(&rq->lock); break; -- cgit v1.2.3-18-g5258 From fbb5b7ae4b442f1923513dc6165a66c7a7f29073 Mon Sep 17 00:00:00 2001 From: Tom Zanussi Date: Tue, 9 Dec 2008 13:14:10 -0800 Subject: relayfs: fix infinite loop with splice() Running kmemtraced, which uses splice() on relayfs, causes a hard lock on x86-64 SMP. As described by Tom Zanussi: It looks like you hit the same problem as described here: commit 8191ecd1d14c6914c660dfa007154860a7908857 splice: fix infinite loop in generic_file_splice_read() relay uses the same loop but it never got noticed or fixed. Cc: Mathieu Desnoyers Tested-by: Pekka Enberg Signed-off-by: Tom Zanussi Signed-off-by: Pekka Enberg Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/relay.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) (limited to 'kernel') diff --git a/kernel/relay.c b/kernel/relay.c index 32b0befdcb6..09ac2008f77 100644 --- a/kernel/relay.c +++ b/kernel/relay.c @@ -1317,12 +1317,9 @@ static ssize_t relay_file_splice_read(struct file *in, if (ret < 0) break; else if (!ret) { - if (spliced) - break; - if (flags & SPLICE_F_NONBLOCK) { + if (flags & SPLICE_F_NONBLOCK) ret = -EAGAIN; - break; - } + break; } *ppos += ret; -- cgit v1.2.3-18-g5258 From 9c24624727f6d6c460e45762a408ca5f5b9b8ef2 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Tue, 9 Dec 2008 13:14:27 -0800 Subject: KSYM_SYMBOL_LEN fixes Miles Lane tailing /sys files hit a BUG which Pekka Enberg has tracked to my 966c8c12dc9e77f931e2281ba25d2f0244b06949 sprint_symbol(): use less stack exposing a bug in slub's list_locations() - kallsyms_lookup() writes a 0 to namebuf[KSYM_NAME_LEN-1], but that was beyond the end of page provided. The 100 slop which list_locations() allows at end of page looks roughly enough for all the other stuff it might print after the symbol before it checks again: break out KSYM_SYMBOL_LEN earlier than before. Latencytop and ftrace and are using KSYM_NAME_LEN buffers where they need KSYM_SYMBOL_LEN buffers, and vmallocinfo a 2*KSYM_NAME_LEN buffer where it wants a KSYM_SYMBOL_LEN buffer: fix those before anyone copies them. [akpm@linux-foundation.org: ftrace.h needs module.h] Signed-off-by: Hugh Dickins Cc: Christoph Lameter Cc Miles Lane Acked-by: Pekka Enberg Acked-by: Steven Rostedt Acked-by: Frederic Weisbecker Cc: Rusty Russell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/latencytop.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/latencytop.c b/kernel/latencytop.c index 5e7b45c5692..449db466bdb 100644 --- a/kernel/latencytop.c +++ b/kernel/latencytop.c @@ -191,7 +191,7 @@ static int lstats_show(struct seq_file *m, void *v) latency_record[i].time, latency_record[i].max); for (q = 0; q < LT_BACKTRACEDEPTH; q++) { - char sym[KSYM_NAME_LEN]; + char sym[KSYM_SYMBOL_LEN]; char *c; if (!latency_record[i].backtrace[q]) break; -- cgit v1.2.3-18-g5258 From b88ed20594db2c685555b68c52b693b75738b2f5 Mon Sep 17 00:00:00 2001 From: Hugh Dickins Date: Wed, 10 Dec 2008 20:48:52 +0000 Subject: fix mapping_writably_mapped() Lee Schermerhorn noticed yesterday that I broke the mapping_writably_mapped test in 2.6.7! Bad bad bug, good good find. The i_mmap_writable count must be incremented for VM_SHARED (just as i_writecount is for VM_DENYWRITE, but while holding the i_mmap_lock) when dup_mmap() copies the vma for fork: it has its own more optimal version of __vma_link_file(), and I missed this out. So the count was later going down to 0 (dangerous) when one end unmapped, then wrapping negative (inefficient) when the other end unmapped. The only impact on x86 would have been that setting a mandatory lock on a file which has at some time been opened O_RDWR and mapped MAP_SHARED (but not necessarily PROT_WRITE) across a fork, might fail with -EAGAIN when it should succeed, or succeed when it should fail. But those architectures which rely on flush_dcache_page() to flush userspace modifications back into the page before the kernel reads it, may in some cases have skipped the flush after such a fork - though any repetitive test will soon wrap the count negative, in which case it will flush_dcache_page() unnecessarily. Fix would be a two-liner, but mapping variable added, and comment moved. Reported-by: Lee Schermerhorn Signed-off-by: Hugh Dickins Signed-off-by: Linus Torvalds --- kernel/fork.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/fork.c b/kernel/fork.c index 8d6a7dd9282..495da2e9a8b 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -315,17 +315,20 @@ static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) file = tmp->vm_file; if (file) { struct inode *inode = file->f_path.dentry->d_inode; + struct address_space *mapping = file->f_mapping; + get_file(file); if (tmp->vm_flags & VM_DENYWRITE) atomic_dec(&inode->i_writecount); - - /* insert tmp into the share list, just after mpnt */ - spin_lock(&file->f_mapping->i_mmap_lock); + spin_lock(&mapping->i_mmap_lock); + if (tmp->vm_flags & VM_SHARED) + mapping->i_mmap_writable++; tmp->vm_truncate_count = mpnt->vm_truncate_count; - flush_dcache_mmap_lock(file->f_mapping); + flush_dcache_mmap_lock(mapping); + /* insert tmp into the share list, just after mpnt */ vma_prio_tree_add(tmp, mpnt); - flush_dcache_mmap_unlock(file->f_mapping); - spin_unlock(&file->f_mapping->i_mmap_lock); + flush_dcache_mmap_unlock(mapping); + spin_unlock(&mapping->i_mmap_lock); } /* -- cgit v1.2.3-18-g5258 From ca7e716c7833aeaeb8fedd6d004c5f5d5e14d325 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 14 Dec 2008 15:46:01 -0800 Subject: Revert "sched_clock: prevent scd->clock from moving backwards" This reverts commit 5b7dba4ff834259a5623e03a565748704a8fe449, which caused a regression in hibernate, reported by and bisected by Fabio Comolli. This revert fixes http://bugzilla.kernel.org/show_bug.cgi?id=12155 http://bugzilla.kernel.org/show_bug.cgi?id=12149 Bisected-by: Fabio Comolli Requested-by: Rafael J. Wysocki Acked-by: Dave Kleikamp Cc: Peter Zijlstra Cc: Ingo Molnar Signed-off-by: Linus Torvalds --- kernel/sched_clock.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/sched_clock.c b/kernel/sched_clock.c index 81787248b60..e8ab096ddfe 100644 --- a/kernel/sched_clock.c +++ b/kernel/sched_clock.c @@ -118,13 +118,13 @@ static u64 __update_sched_clock(struct sched_clock_data *scd, u64 now) /* * scd->clock = clamp(scd->tick_gtod + delta, - * max(scd->tick_gtod, scd->clock), - * max(scd->clock, scd->tick_gtod + TICK_NSEC)); + * max(scd->tick_gtod, scd->clock), + * scd->tick_gtod + TICK_NSEC); */ clock = scd->tick_gtod + delta; min_clock = wrap_max(scd->tick_gtod, scd->clock); - max_clock = wrap_max(scd->clock, scd->tick_gtod + TICK_NSEC); + max_clock = scd->tick_gtod + TICK_NSEC; clock = wrap_max(clock, min_clock); clock = wrap_min(clock, max_clock); -- cgit v1.2.3-18-g5258 From 307257cf475aac25db30b669987f13d90c934e3a Mon Sep 17 00:00:00 2001 From: Paul Menage Date: Mon, 15 Dec 2008 13:54:22 -0800 Subject: cgroups: fix a race between rmdir and remount When a cgroup is removed, it's unlinked from its parent's children list, but not actually freed until the last dentry on it is released (at which point cgrp->root->number_of_cgroups is decremented). Currently rebind_subsystems checks for the top cgroup's child list being empty in order to rebind subsystems into or out of a hierarchy - this can result in the set of subsystems bound to a hierarchy being removed-but-not-freed cgroup. The simplest fix for this is to forbid remounts that change the set of subsystems on a hierarchy that has removed-but-not-freed cgroups. This bug can be reproduced via: mkdir /mnt/cg mount -t cgroup -o ns,freezer cgroup /mnt/cg mkdir /mnt/cg/foo sleep 1h < /mnt/cg/foo & rmdir /mnt/cg/foo mount -t cgroup -o remount,ns,devices,freezer cgroup /mnt/cg kill $! Though the above will cause oops in -mm only but not mainline, but the bug can cause memory leak in mainline (and even oops) Signed-off-by: Paul Menage Reviewed-by: Li Zefan Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/cgroup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/cgroup.c b/kernel/cgroup.c index fe00b3b983a..8185a0f0959 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -702,7 +702,7 @@ static int rebind_subsystems(struct cgroupfs_root *root, * any child cgroups exist. This is theoretically supportable * but involves complex error handling, so it's being left until * later */ - if (!list_empty(&cgrp->children)) + if (root->number_of_cgroups > 1) return -EBUSY; /* Process each subsystem */ -- cgit v1.2.3-18-g5258 From 3d44cc3e01ee1b40317f79ed54324e25c4f848df Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Sat, 20 Dec 2008 21:27:34 +0100 Subject: Null pointer deref with hrtimer_try_to_cancel() Impact: Prevent kernel crash with posix timer clockid CLOCK_MONOTONIC_RAW commit 2d42244ae71d6c7b0884b5664cf2eda30fb2ae68 (clocksource: introduce CLOCK_MONOTONIC_RAW) introduced a new clockid, which is only available to read out the raw not NTP adjusted system time. The above commit did not prevent that a posix timer can be created with that clockid. The timer_create() syscall succeeds and initializes the timer to a non existing hrtimer base. When the timer is deleted either by timer_delete() or by the exit() cleanup the kernel crashes. Prevent the creation of timers for CLOCK_MONOTONIC_RAW by setting the posix clock function to no_timer_create which returns an error code. Reported-and-tested-by: Eric Sesterhenn Signed-off-by: Thomas Gleixner Acked-by: Oleg Nesterov Signed-off-by: Linus Torvalds --- kernel/posix-timers.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'kernel') diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c index 5e79c662294..a140e44eebb 100644 --- a/kernel/posix-timers.c +++ b/kernel/posix-timers.c @@ -197,6 +197,11 @@ static int common_timer_create(struct k_itimer *new_timer) return 0; } +static int no_timer_create(struct k_itimer *new_timer) +{ + return -EOPNOTSUPP; +} + /* * Return nonzero if we know a priori this clockid_t value is bogus. */ @@ -248,6 +253,7 @@ static __init int init_posix_timers(void) .clock_getres = hrtimer_get_res, .clock_get = posix_get_monotonic_raw, .clock_set = do_posix_clock_nosettime, + .timer_create = no_timer_create, }; register_posix_clock(CLOCK_REALTIME, &clock_realtime); -- cgit v1.2.3-18-g5258 From e368d3a836797ddf193b1ec18c97407a791d2451 Mon Sep 17 00:00:00 2001 From: Sharyathi Nagesh Date: Tue, 23 Dec 2008 13:57:12 -0800 Subject: cgroups: suppress bogus warning messages Remove spurious warning messages that are thrown onto the console during cgroup operations. Signed-off-by: Alexey Dobriyan Signed-off-by: Sharyathi Nagesh Acked-by: Serge E. Hallyn Cc: Paul Menage Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/cgroup.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'kernel') diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 8185a0f0959..a3415507bd0 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -2934,9 +2934,6 @@ int cgroup_clone(struct task_struct *tsk, struct cgroup_subsys *subsys, again: root = subsys->root; if (root == &rootnode) { - printk(KERN_INFO - "Not cloning cgroup for unused subsystem %s\n", - subsys->name); mutex_unlock(&cgroup_mutex); return 0; } -- cgit v1.2.3-18-g5258 From 20ca9b3f4c6dfa0af8dd5b18a64df17eb994b54d Mon Sep 17 00:00:00 2001 From: Li Zefan Date: Tue, 23 Dec 2008 13:57:14 -0800 Subject: cgroups: avoid accessing uninitialized data in failure path If cgroup_get_rootdir() failed, free_cg_links() will be called in the failure path, but tmp_cg_links hasn't been initialized at that time. I introduced this bug in the 2.6.27 merge window. Signed-off-by: Li Zefan Acked-by: Serge Hallyn Cc: Paul Menage Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/cgroup.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/cgroup.c b/kernel/cgroup.c index a3415507bd0..2606d0fb4e5 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -1024,7 +1024,7 @@ static int cgroup_get_sb(struct file_system_type *fs_type, if (ret == -EBUSY) { mutex_unlock(&cgroup_mutex); mutex_unlock(&inode->i_mutex); - goto drop_new_super; + goto free_cg_links; } /* EBUSY should be the only error here */ @@ -1073,10 +1073,11 @@ static int cgroup_get_sb(struct file_system_type *fs_type, return simple_set_mnt(mnt, sb); + free_cg_links: + free_cg_links(&tmp_cg_links); drop_new_super: up_write(&sb->s_umount); deactivate_super(sb); - free_cg_links(&tmp_cg_links); return ret; } -- cgit v1.2.3-18-g5258