From 6ec3cfeca04622e3d80c9270191cd7f5f88214af Mon Sep 17 00:00:00 2001 From: "Pallipadi, Venkatesh" Date: Mon, 13 Apr 2009 15:20:58 -0700 Subject: x86, irq: Remove IRQ_DISABLED check in process context IRQ move As discussed in the thread here: http://marc.info/?l=linux-kernel&m=123964468521142&w=2 Eric W. Biederman observed: > It looks like some additional bugs have slipped in since last I looked. > > set_irq_affinity does this: > ifdef CONFIG_GENERIC_PENDING_IRQ > if (desc->status & IRQ_MOVE_PCNTXT || desc->status & IRQ_DISABLED) { > cpumask_copy(desc->affinity, cpumask); > desc->chip->set_affinity(irq, cpumask); > } else { > desc->status |= IRQ_MOVE_PENDING; > cpumask_copy(desc->pending_mask, cpumask); > } > #else > > That IRQ_DISABLED case is a software state and as such it has nothing to > do with how safe it is to move an irq in process context. [...] > > The only reason we migrate MSIs in interrupt context today is that there > wasn't infrastructure for support migration both in interrupt context > and outside of it. Yes. The idea here was to force the MSI migration to happen in process context. One of the patches in the series did disable_irq(dev->irq); irq_set_affinity(dev->irq, cpumask_of(dev->cpu)); enable_irq(dev->irq); with the above patch adding irq/manage code check for interrupt disabled and moving the interrupt in process context. IIRC, there was no IRQ_MOVE_PCNTXT when we were developing this HPET code and we ended up having this ugly hack. IRQ_MOVE_PCNTXT was there when we eventually submitted the patch upstream. But, looks like I did a blind rebasing instead of using IRQ_MOVE_PCNTXT in hpet MSI code. Below patch fixes this. i.e., revert commit 932775a4ab622e3c99bd59f14cc and add PCNTXT to HPET MSI setup. Also removes copying of desc->affinity in generic code as set_affinity routines are doing it internally. Reported-by: "Eric W. Biederman" Signed-off-by: Venkatesh Pallipadi Acked-by: "Eric W. Biederman" Cc: "Li Shaohua" Cc: Gary Hade Cc: "lcm@us.ibm.com" Cc: suresh.b.siddha@intel.com LKML-Reference: <20090413222058.GB8211@linux-os.sc.intel.com> Signed-off-by: Ingo Molnar --- kernel/irq/manage.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 7e2e7dd4cd2f..2734eca59243 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -109,10 +109,9 @@ int irq_set_affinity(unsigned int irq, const struct cpumask *cpumask) spin_lock_irqsave(&desc->lock, flags); #ifdef CONFIG_GENERIC_PENDING_IRQ - if (desc->status & IRQ_MOVE_PCNTXT || desc->status & IRQ_DISABLED) { - cpumask_copy(desc->affinity, cpumask); + if (desc->status & IRQ_MOVE_PCNTXT) desc->chip->set_affinity(irq, cpumask); - } else { + else { desc->status |= IRQ_MOVE_PENDING; cpumask_copy(desc->pending_mask, cpumask); } -- cgit v1.2.3 From c8a250058656495be02c00de61e26b017c86ef00 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 17 Apr 2009 09:40:49 +0200 Subject: lockdep: more robust lockdep_map init sequence Steven Rostedt reported: > OK, I think I figured this bug out. This is a lockdep issue with respect > to tracepoints. > > The trace points in lockdep are called all the time. Outside the lockdep > logic. But if lockdep were to trigger an error / warning (which this run > did) we might be in trouble. For new locks, like the dentry->d_lock, that > are created, they will not get a name: > > void lockdep_init_map(struct lockdep_map *lock, const char *name, > struct lock_class_key *key, int subclass) > { > if (unlikely(!debug_locks)) > return; > > When a problem is found by lockdep, debug_locks becomes false. Thus we > stop allocating names for locks. This dentry->d_lock I had, now has no > name. Worse yet, I have CONFIG_DEBUG_VM set, that scrambles non > initialized memory. Thus, when the trace point was hit, it had junk for > the lock->name, and the machine crashed. Ah, nice catch. I think we should put at least the name in regardless. Ensure we at least initialize the trivial entries of the depmap so that they can be relied upon, even when lockdep itself decided to pack up and go home. [ Impact: fix lock tracing after lockdep warnings. ] Reported-by: Steven Rostedt Signed-off-by: Peter Zijlstra Acked-by: Steven Rostedt Cc: Andrew Morton Cc: Frederic Weisbecker LKML-Reference: <1239954049.23397.4156.camel@laptop> Signed-off-by: Ingo Molnar --- kernel/lockdep.c | 22 ++++++++++++++-------- 1 file changed, 14 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/kernel/lockdep.c b/kernel/lockdep.c index b0f011866969..accb40cdb12a 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -2490,13 +2490,20 @@ static int mark_lock(struct task_struct *curr, struct held_lock *this, void lockdep_init_map(struct lockdep_map *lock, const char *name, struct lock_class_key *key, int subclass) { - if (unlikely(!debug_locks)) + lock->class_cache = NULL; +#ifdef CONFIG_LOCK_STAT + lock->cpu = raw_smp_processor_id(); +#endif + + if (DEBUG_LOCKS_WARN_ON(!name)) { + lock->name = "NULL"; return; + } + + lock->name = name; if (DEBUG_LOCKS_WARN_ON(!key)) return; - if (DEBUG_LOCKS_WARN_ON(!name)) - return; /* * Sanity check, the lock-class key must be persistent: */ @@ -2505,12 +2512,11 @@ void lockdep_init_map(struct lockdep_map *lock, const char *name, DEBUG_LOCKS_WARN_ON(1); return; } - lock->name = name; lock->key = key; - lock->class_cache = NULL; -#ifdef CONFIG_LOCK_STAT - lock->cpu = raw_smp_processor_id(); -#endif + + if (unlikely(!debug_locks)) + return; + if (subclass) register_lock_class(lock, subclass, 1); } -- cgit v1.2.3 From b48ccb095a0c9257241261ec2bd1cbb1bdabc48b Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 23 Apr 2009 09:36:52 +0200 Subject: locking: clarify kernel-taint warning message Andi Kleen reported this message triggering on non-lockdep kernels: Disabling lockdep due to kernel taint Clarify the message to say 'lock debugging' - debug_locks_off() turns off all things lock debugging, not just lockdep. [ Impact: change kernel warning message text ] Reported-by: Andi Kleen Cc: Peter Zijlstra Cc: Andrew Morton Signed-off-by: Ingo Molnar --- kernel/panic.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/panic.c b/kernel/panic.c index 934fb377f4b3..3dcaa1661357 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -221,7 +221,7 @@ void add_taint(unsigned flag) * post-warning case. */ if (flag != TAINT_CRAP && flag != TAINT_WARN && __debug_locks_off()) - printk(KERN_WARNING "Disabling lockdep due to kernel taint\n"); + printk(KERN_WARNING "Disabling lock debugging due to kernel taint\n"); set_bit(flag, &tainted_mask); } -- cgit v1.2.3 From cad81bc2529ab8c62b6fdc83a1c0c7f4a87209eb Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Mon, 27 Apr 2009 01:41:34 +0200 Subject: ptrace: ptrace_attach: fix the usage of ->cred_exec_mutex ptrace_attach() needs task->cred_exec_mutex, not current->cred_exec_mutex. Signed-off-by: Oleg Nesterov Acked-by: Roland McGrath Acked-by: David Howells Signed-off-by: James Morris --- 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 dfcd83ceee3b..0692ab5a0d67 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -188,7 +188,7 @@ int ptrace_attach(struct task_struct *task) /* Protect exec's credential calculations against our interference; * SUID, SGID and LSM creds get determined differently under ptrace. */ - retval = mutex_lock_interruptible(¤t->cred_exec_mutex); + retval = mutex_lock_interruptible(&task->cred_exec_mutex); if (retval < 0) goto out; @@ -232,7 +232,7 @@ repeat: bad: write_unlock_irqrestore(&tasklist_lock, flags); task_unlock(task); - mutex_unlock(¤t->cred_exec_mutex); + mutex_unlock(&task->cred_exec_mutex); out: return retval; } -- cgit v1.2.3 From 9e4a5bda89034502fb144331e71a0efdfd5fae97 Mon Sep 17 00:00:00 2001 From: Andrea Righi Date: Thu, 30 Apr 2009 15:08:57 -0700 Subject: mm: prevent divide error for small values of vm_dirty_bytes Avoid setting less than two pages for vm_dirty_bytes: this is necessary to avoid potential division by 0 (like the following) in get_dirty_limits(). [ 49.951610] divide error: 0000 [#1] PREEMPT SMP [ 49.952195] last sysfs file: /sys/devices/pci0000:00/0000:00:01.1/host0/target0:0:0/0:0:0:0/block/sda/uevent [ 49.952195] CPU 1 [ 49.952195] Modules linked in: pcspkr [ 49.952195] Pid: 3064, comm: dd Not tainted 2.6.30-rc3 #1 [ 49.952195] RIP: 0010:[] [] get_dirty_limits+0xe9/0x2c0 [ 49.952195] RSP: 0018:ffff88001de03a98 EFLAGS: 00010202 [ 49.952195] RAX: 00000000000000c0 RBX: ffff88001de03b80 RCX: 28f5c28f5c28f5c3 [ 49.952195] RDX: 0000000000000000 RSI: 00000000000000c0 RDI: 0000000000000000 [ 49.952195] RBP: ffff88001de03ae8 R08: 0000000000000000 R09: 0000000000000000 [ 49.952195] R10: ffff88001ddda9a0 R11: 0000000000000001 R12: 0000000000000001 [ 49.952195] R13: ffff88001fbc8218 R14: ffff88001de03b70 R15: ffff88001de03b78 [ 49.952195] FS: 00007fe9a435b6f0(0000) GS:ffff8800025d9000(0000) knlGS:0000000000000000 [ 49.952195] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 49.952195] CR2: 00007fe9a39ab000 CR3: 000000001de38000 CR4: 00000000000006e0 [ 49.952195] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 49.952195] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 [ 49.952195] Process dd (pid: 3064, threadinfo ffff88001de02000, task ffff88001ddda250) [ 49.952195] Stack: [ 49.952195] ffff88001fa0de00 ffff88001f2dbd70 ffff88001f9fe800 000080b900000000 [ 49.952195] 00000000000000c0 ffff8800027a6100 0000000000000400 ffff88001fbc8218 [ 49.952195] 0000000000000000 0000000000000600 ffff88001de03bb8 ffffffff802d3ed7 [ 49.952195] Call Trace: [ 49.952195] [] balance_dirty_pages_ratelimited_nr+0x1d7/0x3f0 [ 49.952195] [] ? ext3_writeback_write_end+0x9e/0x120 [ 49.952195] [] generic_file_buffered_write+0x12f/0x330 [ 49.952195] [] __generic_file_aio_write_nolock+0x26d/0x460 [ 49.952195] [] ? generic_file_aio_write+0x52/0xd0 [ 49.952195] [] generic_file_aio_write+0x69/0xd0 [ 49.952195] [] ext3_file_write+0x26/0xc0 [ 49.952195] [] do_sync_write+0xf1/0x140 [ 49.952195] [] ? get_lock_stats+0x2a/0x60 [ 49.952195] [] ? autoremove_wake_function+0x0/0x40 [ 49.952195] [] vfs_write+0xcb/0x190 [ 49.952195] [] sys_write+0x50/0x90 [ 49.952195] [] system_call_fastpath+0x16/0x1b [ 49.952195] Code: 00 00 00 2b 05 09 1c 17 01 48 89 c6 49 0f af f4 48 c1 ee 02 48 89 f0 48 f7 e1 48 89 d6 31 d2 48 c1 ee 02 48 0f af 75 d0 48 89 f0 <48> f7 f7 41 8b 95 ac 01 00 00 48 89 c7 49 0f af d4 48 c1 ea 02 [ 49.952195] RIP [] get_dirty_limits+0xe9/0x2c0 [ 49.952195] RSP [ 50.096523] ---[ end trace 008d7aa02f244d7b ]--- Signed-off-by: Andrea Righi Cc: Peter Zijlstra Cc: David Rientjes Cc: Dave Chinner Cc: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sysctl.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/sysctl.c b/kernel/sysctl.c index e3d2c7dd59b9..ea78fa101ad6 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -103,6 +103,9 @@ static unsigned long one_ul = 1; static int one_hundred = 100; static int one_thousand = 1000; +/* this is needed for the proc_doulongvec_minmax of vm_dirty_bytes */ +static unsigned long dirty_bytes_min = 2 * PAGE_SIZE; + /* this is needed for the proc_dointvec_minmax for [fs_]overflow UID and GID */ static int maxolduid = 65535; static int minolduid; @@ -1006,7 +1009,7 @@ static struct ctl_table vm_table[] = { .mode = 0644, .proc_handler = &dirty_bytes_handler, .strategy = &sysctl_intvec, - .extra1 = &one_ul, + .extra1 = &dirty_bytes_min, }, { .procname = "dirty_writeback_centisecs", -- cgit v1.2.3