2.6.33-rc8-rt1
in [Kernel]
|
Prev: [kernel26] Very high power consumption and cpu usage on laptop with any kernel26-2.6.32.*
Next: [PATCH 3/4] tracing: graph output support for preemptirqsoff/preemptoff tracers
From: Dan Carpenter on 22 Feb 2010 08:40 mm/slab.c 2117 kmem_freepages(cachep, addr, *this_cpu); 2118 if (OFF_SLAB(cachep)) { 2119 if (this_cpu) We already dereferenced "this_cpu" two lines earlier. 2120 __cache_free(cachep->slabp_cache, slabp, this_cpu); 2121 else 2122 kmem_cache_free(cachep->slabp_cache, slabp); regards, dan carpenter -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Thomas Gleixner on 22 Feb 2010 10:10 On Mon, 22 Feb 2010, Dan Carpenter wrote: > mm/slab.c > 2117 kmem_freepages(cachep, addr, *this_cpu); > 2118 if (OFF_SLAB(cachep)) { > 2119 if (this_cpu) > > We already dereferenced "this_cpu" two lines earlier. > > 2120 __cache_free(cachep->slabp_cache, slabp, this_cpu); > 2121 else > 2122 kmem_cache_free(cachep->slabp_cache, slabp); > Good catch. Thanks, tglx -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Thomas Gleixner on 22 Feb 2010 10:20 On Mon, 22 Feb 2010, John Kacur wrote: > That patch fixes the symptom I reported, but unfortunately, I still > can't log-in to my laptop without the machine freezing. > Not sure why, investigating. Found it, at least it make my mouse/keyboard work again and fixes the weird double character problem on VT login. Thanks, tglx ---- diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 25642d1..1efad98 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -563,15 +563,20 @@ irq_wait_for_interrupt(struct irq_desc *desc, struct irqaction *action) * handler finished. unmask if the interrupt has not been disabled and * is marked MASKED. */ -static void irq_finalize_oneshot(unsigned int irq, struct irq_desc *desc) +static void irq_finalize_oneshot(unsigned int irq, struct irq_desc *desc, + struct irqaction *action) { chip_bus_lock(irq, desc); +#ifndef CONFIG_PREEMPT_RT raw_spin_lock_irq(&desc->lock); if (!(desc->status & IRQ_DISABLED) && (desc->status & IRQ_MASKED)) { desc->status &= ~IRQ_MASKED; desc->chip->unmask(irq); } raw_spin_unlock_irq(&desc->lock); +#else + preempt_hardirq_thread_done(desc, action); +#endif chip_bus_sync_unlock(irq, desc); } @@ -645,7 +650,7 @@ static int irq_thread(void *data) action->thread_fn(action->irq, action->dev_id); if (oneshot) - irq_finalize_oneshot(action->irq, desc); + irq_finalize_oneshot(action->irq, desc, action); } wake = atomic_dec_and_test(&desc->threads_active); -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: John Kacur on 22 Feb 2010 11:00 On Mon, 22 Feb 2010, Thomas Gleixner wrote: > On Mon, 22 Feb 2010, John Kacur wrote: > > That patch fixes the symptom I reported, but unfortunately, I still > > can't log-in to my laptop without the machine freezing. > > Not sure why, investigating. > > Found it, at least it make my mouse/keyboard work again and fixes the > weird double character problem on VT login. > > Thanks, > > tglx > ---- > diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c > index 25642d1..1efad98 100644 > --- a/kernel/irq/manage.c > +++ b/kernel/irq/manage.c > @@ -563,15 +563,20 @@ irq_wait_for_interrupt(struct irq_desc *desc, struct irqaction *action) > * handler finished. unmask if the interrupt has not been disabled and > * is marked MASKED. > */ > -static void irq_finalize_oneshot(unsigned int irq, struct irq_desc *desc) > +static void irq_finalize_oneshot(unsigned int irq, struct irq_desc *desc, > + struct irqaction *action) > { > chip_bus_lock(irq, desc); > +#ifndef CONFIG_PREEMPT_RT > raw_spin_lock_irq(&desc->lock); > if (!(desc->status & IRQ_DISABLED) && (desc->status & IRQ_MASKED)) { > desc->status &= ~IRQ_MASKED; > desc->chip->unmask(irq); > } > raw_spin_unlock_irq(&desc->lock); > +#else > + preempt_hardirq_thread_done(desc, action); > +#endif > chip_bus_sync_unlock(irq, desc); > } > > @@ -645,7 +650,7 @@ static int irq_thread(void *data) > action->thread_fn(action->irq, action->dev_id); > > if (oneshot) > - irq_finalize_oneshot(action->irq, desc); > + irq_finalize_oneshot(action->irq, desc, action); > } > > wake = atomic_dec_and_test(&desc->threads_active); > Thanks Thomas The fix seems to work for me. I've got some other issues, but for now I'm going to assume they are the same acpi problems I have reported to lkml for the "vanilla" -rc8 John -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: John Kacur on 22 Feb 2010 11:20
On Mon, Feb 22, 2010 at 4:53 PM, John Kacur <jkacur(a)redhat.com> wrote: > > > On Mon, 22 Feb 2010, Thomas Gleixner wrote: > >> On Mon, 22 Feb 2010, John Kacur wrote: >> > That patch fixes the symptom I reported, but unfortunately, I still >> > can't log-in to my laptop without the machine freezing. >> > Not sure why, investigating. >> >> Found it, at least it make my mouse/keyboard work again and fixes the >> weird double character problem on VT login. >> >> Thanks, >> >> � � � tglx >> ---- >> diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c >> index 25642d1..1efad98 100644 >> --- a/kernel/irq/manage.c >> +++ b/kernel/irq/manage.c >> @@ -563,15 +563,20 @@ irq_wait_for_interrupt(struct irq_desc *desc, struct irqaction *action) >> � * handler finished. unmask if the interrupt has not been disabled and >> � * is marked MASKED. >> � */ >> -static void irq_finalize_oneshot(unsigned int irq, struct irq_desc *desc) >> +static void irq_finalize_oneshot(unsigned int irq, struct irq_desc *desc, >> + � � � � � � � � � � � � � � �struct irqaction *action) >> �{ >> � � � chip_bus_lock(irq, desc); >> +#ifndef CONFIG_PREEMPT_RT >> � � � raw_spin_lock_irq(&desc->lock); >> � � � if (!(desc->status & IRQ_DISABLED) && (desc->status & IRQ_MASKED)) { >> � � � � � � � desc->status &= ~IRQ_MASKED; >> � � � � � � � desc->chip->unmask(irq); >> � � � } >> � � � raw_spin_unlock_irq(&desc->lock); >> +#else >> + � � preempt_hardirq_thread_done(desc, action); >> +#endif >> � � � chip_bus_sync_unlock(irq, desc); >> �} >> >> @@ -645,7 +650,7 @@ static int irq_thread(void *data) >> � � � � � � � � � � � action->thread_fn(action->irq, action->dev_id); >> >> � � � � � � � � � � � if (oneshot) >> - � � � � � � � � � � � � � � irq_finalize_oneshot(action->irq, desc); >> + � � � � � � � � � � � � � � irq_finalize_oneshot(action->irq, desc, action); >> � � � � � � � } >> >> � � � � � � � wake = atomic_dec_and_test(&desc->threads_active); >> > > > Thanks Thomas > > The fix seems to work for me. > I've got some other issues, but for now I'm going to assume they are the > same acpi problems I have reported to lkml for the "vanilla" -rc8 > Sorry, spoke to soon, graphical login still freezing. I did nohz=off and nmi_watchdog=2 as you told me in the irc session, and got a nice oops, here it is. ------------[ cut here ]------------ Kernel BUG at ffffffff81045d73 [verbose debug info unavailable] invalid opcode: 0000 [#1] PREEMPT SMP last sysfs file: /sys/class/power_supply/BAT0/energy_full CPU 0 Pid: 3467, comm: firefox-bin Tainted: G D 2.6.33-rc8-rt1-debug #2 2241B4 8/2241B48 RIP: 0010:[<ffffffff81045d73>] [<ffffffff81045d73>] do_exit+0x776/0x784 RSP: 0018:ffff88002be06e28 EFLAGS: 00010246 RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 RDX: 0000000000000002 RSI: 0000000000000003 RDI: ffff880115638000 RBP: ffff88002be06e98 R08: 0000000000000002 R09: 0000000000000001 R10: 0000000000000002 R11: 0000000000000046 R12: ffff88002be06e48 R13: ffff8801157fd200 R14: ffff88011546d080 R15: ffff8801157fd4c0 FS: 0000000000000000(0000) GS:ffff88002be00000(0000) knlGS:0000000000000000 CS: 0010 DS: 002b ES: 002b CR0: 000000008005003b CR2: 00000034e64a7ce0 CR3: 0000000139bd3000 CR4: 00000000000006f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process firefox-bin (pid: 3467, threadinfo ffff880115638000, task ffff8801157fd2 00) Stack: 0000000081602f00 ffffffff8135cf95 ffff8801157fd4c0 ffff8801157fd1f0 <0> ffff88002be06e48 ffff88002be06e48 ffff88002bfcbb20 0000000000000001 <0> 0000000000000007 ffffffff811aca40 0000000000000046 ffff88002be06f58 Call Trace: <NMI> [<ffffffff8135cf95>] ? _raw_spin_unlock_irqrestore+0x63/0x68 [<ffffffff811aca40>] ? delay_tsc+0x6/0xc6 [<ffffffff8135e31f>] die_nmi+0xef/0xf8 [<ffffffff8135e971>] nmi_watchdog_tick+0x159/0x1d6 [<ffffffff8135de29>] do_nmi+0xc2/0x28a [<ffffffff8135d86a>] nmi+0x1a/0x2c [<ffffffff811aca40>] ? delay_tsc+0x6/0xc6 <<EOE>> [<ffffffff811ac9d5>] __delay+0xa/0xc [<ffffffff811b0d50>] do_raw_spin_lock+0xde/0x146 [<ffffffff8135ce39>] _raw_spin_lock_irqsave+0x67/0x7f [<ffffffff8109e8c0>] ? rcu_sched_qs+0x70/0x16c [<ffffffff8109e8c0>] rcu_sched_qs+0x70/0x16c [<ffffffff8135a3f7>] __schedule+0x32/0x923 [<ffffffff8135d340>] ? restore_args+0x0/0x30 [<ffffffff8135ada8>] preempt_schedule+0x55/0x78 [<ffffffff8135cf95>] _raw_spin_unlock_irqrestore+0x63/0x68 [<ffffffff8135bce5>] rt_spin_lock_slowlock+0x2ad/0x2e0 [<ffffffff8135ba38>] ? rt_spin_lock_slowlock+0x0/0x2e0 [<ffffffff81076dde>] rt_spin_lock_fastlock+0x56/0x5b [<ffffffff8135c625>] rt_spin_lock+0x1f/0x42 [<ffffffff8110616c>] __d_lookup+0xc5/0x189 [<ffffffff811060a7>] ? __d_lookup+0x0/0x189 [<ffffffff810fca20>] do_lookup+0x53/0x187 [<ffffffff810fd27d>] link_path_walk+0x2ca/0x699 [<ffffffff810fd802>] path_walk+0x6d/0xeb [<ffffffff810fdc29>] do_path_lookup+0x2a/0x8a [<ffffffff811004b7>] user_path_at+0x52/0x8c [<ffffffff8106ee74>] ? trace_hardirqs_on+0xd/0xf [<ffffffff8135ba0d>] ? rt_spin_lock_slowunlock+0x4e/0x79 [<ffffffff8110b145>] ? mntput_no_expire+0x24/0xf8 [<ffffffff810f7d6b>] vfs_fstatat+0x35/0x62 [<ffffffff810f7dec>] vfs_lstat+0x19/0x1b [<ffffffff8102bdbb>] sys32_lstat64+0x1a/0x34 [<ffffffff810fc85a>] ? path_put+0x1d/0x22 [<ffffffff81097c8c>] ? audit_syscall_entry+0x114/0x140 [<ffffffff8102aad8>] sysenter_dispatch+0x7/0x2c [<ffffffff8135c1f7>] ? trace_hardirqs_on_thunk+0x3a/0x3f Code: bd 98 15 00 00 48 85 ff 74 05 e8 03 54 0b 00 bf 01 00 00 00 e8 b4 a2 31 00 e8 d2 85 05 00 49 c7 45 00 80 00 00 00 e8 58 50 31 00 <0f> 0b eb fe bb fe ff ff ff e9 1d ff ff ff 55 48 89 e5 41 56 41 RIP [<ffffffff81045d73>] do_exit+0x776/0x784 RSP <ffff88002be06e28> ---[ end trace 6f724503db632ce1 ]--- Fixing recursive fault but reboot is needed! BUG: NMI Watchdog detected LOCKUP on CPU1, ip ffffffff811aca7e, registers: CPU 1 Pid: 0, comm: swapper Tainted: G D 2.6.33-rc8-rt1-debug #2 2241B48/2241B 48 RIP: 0010:[<ffffffff811aca7e>] [<ffffffff811aca7e>] delay_tsc+0x44/0xc6 RSP: 0018:ffff88002c003d28 EFLAGS: 00000002 RAX: 00000000e46481d9 RBX: 00000000e4648146 RCX: 000000000000ab00 RDX: 00000000000000c5 RSI: ffffffff815527c6 RDI: 0000000000000001 RBP: ffff88002c003d58 R08: 0000000000000002 R09: 0000000000000001 R10: ffffffff8175bc00 R11: 0000000000000046 R12: 0000000000000001 R13: 00000000e46481d9 R14: 0000000000000001 R15: ffff88013bbcffd8 FS: 0000000000000000(0000) GS:ffff88002c000000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 0000000000f05fb0 CR3: 000000012c4c8000 CR4: 00000000000006e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process swapper (pid: 0, threadinfo ffff88013bbce000, task ffff88013bbd07c0) Stack: ffff88013bbd07c0 000000000d0fcb31 ffffffff8175c080 0000000000000001 <0> ffff88013bbd07c0 ffff88013bbd0ce8 ffff88002c003d68 ffffffff811ac9d5 <0> ffff88002c003da8 ffffffff811b0d50 00000000a6773048 ffffffff8175c098 Call Trace: <IRQ> [<ffffffff811ac9d5>] __delay+0xa/0xc [<ffffffff811b0d50>] do_raw_spin_lock+0xde/0x146 [<ffffffff8106aa26>] ? tick_sched_timer+0x0/0x9d [<ffffffff8135ce39>] _raw_spin_lock_irqsave+0x67/0x7f [<ffffffff8109ea01>] ? rcu_print_task_stall+0x45/0xb4 [<ffffffff8109ea01>] rcu_print_task_stall+0x45/0xb4 [<ffffffff8106aa26>] ? tick_sched_timer+0x0/0x9d [<ffffffff8109ef6a>] __rcu_pending+0x15b/0x2d3 [<ffffffff8106aa26>] ? tick_sched_timer+0x0/0x9d [<ffffffff8109f151>] rcu_check_callbacks+0x6f/0x17a [<ffffffff8106aa26>] ? tick_sched_timer+0x0/0x9d [<ffffffff81050a69>] update_process_times+0x3e/0x54 [<ffffffff8106aa99>] tick_sched_timer+0x73/0x9d [<ffffffff81060cfb>] __run_hrtimer+0xc9/0x13a [<ffffffff81061256>] hrtimer_interrupt+0xe0/0x1de [<ffffffff81019a3e>] smp_apic_timer_interrupt+0x81/0x94 [<ffffffff81003613>] apic_timer_interrupt+0x13/0x20 <EOI> [<ffffffff813600ee>] ? __atomic_notifier_call_chain+0x0/0x94 [<ffffffff812967fb>] ? poll_idle+0x16/0x5f [<ffffffff81296809>] ? poll_idle+0x24/0x5f [<ffffffff812967fb>] ? poll_idle+0x16/0x5f [<ffffffff813600ee>] ? __atomic_notifier_call_chain+0x0/0x94 [<ffffffff81296d13>] cpuidle_idle_call+0xa0/0x111 [<ffffffff81001f2a>] cpu_idle+0xa4/0x107 [<ffffffff81355fc9>] start_secondary+0x18f/0x194 Code: e8 bd 35 1b 00 e8 87 54 00 00 41 89 c6 66 66 90 0f ae e8 0f 31 89 c3 65 4c 8b 3c 25 48 95 00 00 66 66 90 0f ae e8 0f 31 41 89 c5 <4c> 89 e8 48 29 d8 4c 39 e0 73 49 bf 01 00 00 00 e8 04 35 1b 00 ---[ end trace 6f724503db632ce2 ]--- Kernel panic - not syncing: Aiee, killing interrupt handler! -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ |