Skip to contentSkip to conversation

[ INFO: iconsistent lock state ]

INFO: iconsistent lock state In review Justin Mattock 3 participants 10 messages Apr 25, 2008

Subsystems (3)
x86 ACPI Apm driver
lore
Share & tools
lore

Justin Mattock

author
#0 lore raw

attached is dmesg of [ INFO: inconsistent lock state ] whenever I use the boot option nohz=off this does not appear. regards;

Signature · 1 line
Justin P. Mattock
#1 lore raw
Quoted 3 lines · Justin Mattock
[   13.269763] =================================
[   13.270954] [ INFO: inconsistent lock state ]
[   13.271865] 2.6.25-04569-gb69d398 #3

What tree is that? git checkout gb69d398 doesn't want to happen here.

Quoted 38 lines
[   13.272614] ---------------------------------
[   13.273521] inconsistent {in-hardirq-W} -> {hardirq-on-W} usage.
[   13.274745] swapper/0 [HC0[0]:SC0[0]:HE1:SE1] takes:
[   13.275787]  (&rq->rq_lock_key){++..}, at: [<c01208e9>] sched_clock_idle_wakeup_event+0x43/0x74
[   13.276859] {in-hardirq-W} state was registered at:
[   13.276859]   [<c01423d8>] __lock_acquire+0x419/0xb70
[   13.276859]   [<c0142bae>] lock_acquire+0x7f/0xa6
[   13.280188]   [<c0394132>] _spin_lock+0x1c/0x49
[   13.280188]   [<c0121c44>] scheduler_tick+0x43/0x1bd
[   13.280188]   [<c012d99f>] update_process_times+0x3d/0x49
[   13.283524]   [<c013daf6>] tick_periodic+0x66/0x72
[   13.283524]   [<c013db1b>] tick_handle_periodic+0x19/0x6a
[   13.283524]   [<c0105a87>] timer_interrupt+0x47/0x4e
[   13.286855]   [<c015cf64>] handle_IRQ_event+0x1a/0x4f
[   13.286855]   [<c015e3e9>] handle_level_irq+0x7f/0xca
[   13.286855]   [<c0105a27>] do_IRQ+0x71/0x8a
[   13.290190]   [<c01043fe>] common_interrupt+0x2e/0x34
[   13.290190]   [<c038bd6f>] calibrate_delay+0x8f/0x276
[   13.290190]   [<c04f0883>] start_kernel+0x27c/0x2f8
[   13.293520]   [<c04f0008>] __init_begin+0x8/0xa
[   13.293520]   [<ffffffff>] 0xffffffff
[   13.293520] irq event stamp: 253965
[   13.296856] hardirqs last  enabled at (253965): [<c0108069>] native_sched_clock+0xe7/0xff
[   13.296856] hardirqs last disabled at (253964): [<c0107fef>] native_sched_clock+0x6d/0xff
[   13.300190] softirqs last  enabled at (253958): [<c0129cea>] __do_softirq+0xf9/0xff
[   13.300190] softirqs last disabled at (253953): [<c0129d3d>] do_softirq+0x4d/0x79
[   13.303522]
[   13.303522] other info that might help us debug this:
[   13.303522] no locks held by swapper/0.
[   13.303522]
[   13.303522] stack backtrace:
[   13.306852] Pid: 0, comm: swapper Not tainted 2.6.25-04569-gb69d398 #3
[   13.336851]  [<c01409f9>] print_usage_bug+0x106/0x113
[   13.340185]  [<c0141823>] mark_lock+0x1ed/0x3a5
[   13.343519]  [<c014244d>] __lock_acquire+0x48e/0xb70
[   13.346852]  [<c0142bae>] lock_acquire+0x7f/0xa6
[   13.350185]  [<c01208e9>] ? sched_clock_idle_wakeup_event+0x43/0x74
[   13.353519]  [<c0394132>] _spin_lock+0x1c/0x49
Quoted 8 lines
[   13.360185]  [<c01208e9>] ? sched_clock_idle_wakeup_event+0x43/0x74
[   13.363518]  [<c01208e9>] sched_clock_idle_wakeup_event+0x43/0x74
[   13.366851]  [<f883919b>] acpi_idle_enter_bm+0x2b3/0x333 [processor]
[   13.370184]  [<c02e5ec1>] cpuidle_idle_call+0x63/0x92
[   13.373517]  [<c02e5e5e>] ? cpuidle_idle_call+0x0/0x92
[   13.380184]  [<c010257d>] cpu_idle+0xb6/0xd6
[   13.383517]  [<c0387071>] rest_init+0x49/0x4b
[   13.386850]  =======================

looking at current -linus I'm not directly seeing how we can get there with interrupts enabled.

author
#2 lore raw
Quoted 55 lines · Peter Zijlstra
Quoted 3 lines · Justin Mattock
[   13.269763] =================================
[   13.270954] [ INFO: inconsistent lock state ]
[   13.271865] 2.6.25-04569-gb69d398 #3

What tree is that? git checkout gb69d398 doesn't want to happen here.

Quoted 38 lines
[   13.272614] ---------------------------------
[   13.273521] inconsistent {in-hardirq-W} -> {hardirq-on-W} usage.
[   13.274745] swapper/0 [HC0[0]:SC0[0]:HE1:SE1] takes:
[   13.275787]  (&rq->rq_lock_key){++..}, at: [<c01208e9>] sched_clock_idle_wakeup_event+0x43/0x74
[   13.276859] {in-hardirq-W} state was registered at:
[   13.276859]   [<c01423d8>] __lock_acquire+0x419/0xb70
[   13.276859]   [<c0142bae>] lock_acquire+0x7f/0xa6
[   13.280188]   [<c0394132>] _spin_lock+0x1c/0x49
[   13.280188]   [<c0121c44>] scheduler_tick+0x43/0x1bd
[   13.280188]   [<c012d99f>] update_process_times+0x3d/0x49
[   13.283524]   [<c013daf6>] tick_periodic+0x66/0x72
[   13.283524]   [<c013db1b>] tick_handle_periodic+0x19/0x6a
[   13.283524]   [<c0105a87>] timer_interrupt+0x47/0x4e
[   13.286855]   [<c015cf64>] handle_IRQ_event+0x1a/0x4f
[   13.286855]   [<c015e3e9>] handle_level_irq+0x7f/0xca
[   13.286855]   [<c0105a27>] do_IRQ+0x71/0x8a
[   13.290190]   [<c01043fe>] common_interrupt+0x2e/0x34
[   13.290190]   [<c038bd6f>] calibrate_delay+0x8f/0x276
[   13.290190]   [<c04f0883>] start_kernel+0x27c/0x2f8
[   13.293520]   [<c04f0008>] __init_begin+0x8/0xa
[   13.293520]   [<ffffffff>] 0xffffffff
[   13.293520] irq event stamp: 253965
[   13.296856] hardirqs last  enabled at (253965): [<c0108069>] native_sched_clock+0xe7/0xff
[   13.296856] hardirqs last disabled at (253964): [<c0107fef>] native_sched_clock+0x6d/0xff
[   13.300190] softirqs last  enabled at (253958): [<c0129cea>] __do_softirq+0xf9/0xff
[   13.300190] softirqs last disabled at (253953): [<c0129d3d>] do_softirq+0x4d/0x79
[   13.303522]
[   13.303522] other info that might help us debug this:
[   13.303522] no locks held by swapper/0.
[   13.303522]
[   13.303522] stack backtrace:
[   13.306852] Pid: 0, comm: swapper Not tainted 2.6.25-04569-gb69d398 #3
[   13.336851]  [<c01409f9>] print_usage_bug+0x106/0x113
[   13.340185]  [<c0141823>] mark_lock+0x1ed/0x3a5
[   13.343519]  [<c014244d>] __lock_acquire+0x48e/0xb70
[   13.346852]  [<c0142bae>] lock_acquire+0x7f/0xa6
[   13.350185]  [<c01208e9>] ? sched_clock_idle_wakeup_event+0x43/0x74
[   13.353519]  [<c0394132>] _spin_lock+0x1c/0x49
Quoted 8 lines
[   13.360185]  [<c01208e9>] ? sched_clock_idle_wakeup_event+0x43/0x74
[   13.363518]  [<c01208e9>] sched_clock_idle_wakeup_event+0x43/0x74
[   13.366851]  [<f883919b>] acpi_idle_enter_bm+0x2b3/0x333 [processor]
[   13.370184]  [<c02e5ec1>] cpuidle_idle_call+0x63/0x92
[   13.373517]  [<c02e5e5e>] ? cpuidle_idle_call+0x0/0x92
[   13.380184]  [<c010257d>] cpu_idle+0xb6/0xd6
[   13.383517]  [<c0387071>] rest_init+0x49/0x4b
[   13.386850]  =======================
Quoted 4 lines

looking at current -linus I'm not directly seeing how we can get there with interrupts enabled.

Hello; I have been using:
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git

I used the tutorial from kernel hackers, and IBM. If this is what you're asking. All I know is last friday I did git-pull, "2.6.25-xxxx-xxxxx" and my system has not been able to run right,
If I new how to pull code out of my ass I would probably have fixed this problem that I have by now, but I dont,
regards;

Signature · 1 line
Justin P. Mattock
#3 lore raw
Quoted 8 lines · Justin Mattock
Quoted 7 lines · Peter Zijlstra
Quoted 3 lines · Justin Mattock
[   13.269763] =================================
[   13.270954] [ INFO: inconsistent lock state ]
[   13.271865] 2.6.25-04569-gb69d398 #3

What tree is that? git checkout gb69d398 doesn't want to happen here.

Argh, s/g// so the hash is: b69d398

Could we make that -git-# ?

Quoted 32 lines
Quoted 32 lines
Quoted 32 lines
[   13.272614] ---------------------------------
[   13.273521] inconsistent {in-hardirq-W} -> {hardirq-on-W} usage.
[   13.274745] swapper/0 [HC0[0]:SC0[0]:HE1:SE1] takes:
[   13.275787]  (&rq->rq_lock_key){++..}, at: [<c01208e9>] sched_clock_idle_wakeup_event+0x43/0x74
[   13.276859] {in-hardirq-W} state was registered at:
[   13.276859]   [<c01423d8>] __lock_acquire+0x419/0xb70
[   13.276859]   [<c0142bae>] lock_acquire+0x7f/0xa6
[   13.280188]   [<c0394132>] _spin_lock+0x1c/0x49
[   13.280188]   [<c0121c44>] scheduler_tick+0x43/0x1bd
[   13.280188]   [<c012d99f>] update_process_times+0x3d/0x49
[   13.283524]   [<c013daf6>] tick_periodic+0x66/0x72
[   13.283524]   [<c013db1b>] tick_handle_periodic+0x19/0x6a
[   13.283524]   [<c0105a87>] timer_interrupt+0x47/0x4e
[   13.286855]   [<c015cf64>] handle_IRQ_event+0x1a/0x4f
[   13.286855]   [<c015e3e9>] handle_level_irq+0x7f/0xca
[   13.286855]   [<c0105a27>] do_IRQ+0x71/0x8a
[   13.290190]   [<c01043fe>] common_interrupt+0x2e/0x34
[   13.290190]   [<c038bd6f>] calibrate_delay+0x8f/0x276
[   13.290190]   [<c04f0883>] start_kernel+0x27c/0x2f8
[   13.293520]   [<c04f0008>] __init_begin+0x8/0xa
[   13.293520]   [<ffffffff>] 0xffffffff
[   13.293520] irq event stamp: 253965
[   13.296856] hardirqs last  enabled at (253965): [<c0108069>] native_sched_clock+0xe7/0xff
[   13.296856] hardirqs last disabled at (253964): [<c0107fef>] native_sched_clock+0x6d/0xff
[   13.300190] softirqs last  enabled at (253958): [<c0129cea>] __do_softirq+0xf9/0xff
[   13.300190] softirqs last disabled at (253953): [<c0129d3d>] do_softirq+0x4d/0x79
[   13.303522]
[   13.303522] other info that might help us debug this:
[   13.303522] no locks held by swapper/0.
[   13.303522]
[   13.303522] stack backtrace:
[   13.306852] Pid: 0, comm: swapper Not tainted 2.6.25-04569-gb69d398 #3
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
[   13.336851]  [<c01409f9>] print_usage_bug+0x106/0x113
[   13.340185]  [<c0141823>] mark_lock+0x1ed/0x3a5
[   13.343519]  [<c014244d>] __lock_acquire+0x48e/0xb70
[   13.346852]  [<c0142bae>] lock_acquire+0x7f/0xa6
[   13.350185]  [<c01208e9>] ? sched_clock_idle_wakeup_event+0x43/0x74
[   13.353519]  [<c0394132>] _spin_lock+0x1c/0x49
[   13.360185]  [<c01208e9>] ? sched_clock_idle_wakeup_event+0x43/0x74
[   13.363518]  [<c01208e9>] sched_clock_idle_wakeup_event+0x43/0x74

Got it:

acpi_idle_do_entry()

    acpi_processor_ffh_cstate_enter()
      mwait_idle_with_hints() (32 bit)
        local_irq_enable()

sched_clock_idle_wakeup_event()

I think my recent idle patches should address this, no?

Quoted 10 lines
Quoted 10 lines
Quoted 6 lines
[   13.366851]  [<f883919b>] acpi_idle_enter_bm+0x2b3/0x333 [processor]
[   13.370184]  [<c02e5ec1>] cpuidle_idle_call+0x63/0x92
[   13.373517]  [<c02e5e5e>] ? cpuidle_idle_call+0x0/0x92
[   13.380184]  [<c010257d>] cpu_idle+0xb6/0xd6
[   13.383517]  [<c0387071>] rest_init+0x49/0x4b
[   13.386850]  =======================

looking at current -linus I'm not directly seeing how we can get there with interrupts enabled.

#4 lore raw

PATCH 1/1 idle (arch, acpi and apm) and lockdep

Quoted 51 lines · Peter Zijlstra
Quoted 38 lines · Justin Mattock
Quoted 37 lines · Peter Zijlstra
Quoted 35 lines · Justin Mattock
[   13.269763] =================================
[   13.270954] [ INFO: inconsistent lock state ]
[   13.271865] 2.6.25-04569-gb69d398 #3
[   13.272614] ---------------------------------
[   13.273521] inconsistent {in-hardirq-W} -> {hardirq-on-W} usage.
[   13.274745] swapper/0 [HC0[0]:SC0[0]:HE1:SE1] takes:
[   13.275787]  (&rq->rq_lock_key){++..}, at: [<c01208e9>] sched_clock_idle_wakeup_event+0x43/0x74
[   13.276859] {in-hardirq-W} state was registered at:
[   13.276859]   [<c01423d8>] __lock_acquire+0x419/0xb70
[   13.276859]   [<c0142bae>] lock_acquire+0x7f/0xa6
[   13.280188]   [<c0394132>] _spin_lock+0x1c/0x49
[   13.280188]   [<c0121c44>] scheduler_tick+0x43/0x1bd
[   13.280188]   [<c012d99f>] update_process_times+0x3d/0x49
[   13.283524]   [<c013daf6>] tick_periodic+0x66/0x72
[   13.283524]   [<c013db1b>] tick_handle_periodic+0x19/0x6a
[   13.283524]   [<c0105a87>] timer_interrupt+0x47/0x4e
[   13.286855]   [<c015cf64>] handle_IRQ_event+0x1a/0x4f
[   13.286855]   [<c015e3e9>] handle_level_irq+0x7f/0xca
[   13.286855]   [<c0105a27>] do_IRQ+0x71/0x8a
[   13.290190]   [<c01043fe>] common_interrupt+0x2e/0x34
[   13.290190]   [<c038bd6f>] calibrate_delay+0x8f/0x276
[   13.290190]   [<c04f0883>] start_kernel+0x27c/0x2f8
[   13.293520]   [<c04f0008>] __init_begin+0x8/0xa
[   13.293520]   [<ffffffff>] 0xffffffff
[   13.293520] irq event stamp: 253965
[   13.296856] hardirqs last  enabled at (253965): [<c0108069>] native_sched_clock+0xe7/0xff
[   13.296856] hardirqs last disabled at (253964): [<c0107fef>] native_sched_clock+0x6d/0xff
[   13.300190] softirqs last  enabled at (253958): [<c0129cea>] __do_softirq+0xf9/0xff
[   13.300190] softirqs last disabled at (253953): [<c0129d3d>] do_softirq+0x4d/0x79
[   13.303522]
[   13.303522] other info that might help us debug this:
[   13.303522] no locks held by swapper/0.
[   13.303522]
[   13.303522] stack backtrace:
[   13.306852] Pid: 0, comm: swapper Not tainted 2.6.25-04569-gb69d398 #3
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
[   13.336851]  [<c01409f9>] print_usage_bug+0x106/0x113
[   13.340185]  [<c0141823>] mark_lock+0x1ed/0x3a5
[   13.343519]  [<c014244d>] __lock_acquire+0x48e/0xb70
[   13.346852]  [<c0142bae>] lock_acquire+0x7f/0xa6
[   13.350185]  [<c01208e9>] ? sched_clock_idle_wakeup_event+0x43/0x74
[   13.353519]  [<c0394132>] _spin_lock+0x1c/0x49
[   13.360185]  [<c01208e9>] ? sched_clock_idle_wakeup_event+0x43/0x74
[   13.363518]  [<c01208e9>] sched_clock_idle_wakeup_event+0x43/0x74

Got it:

acpi_idle_do_entry()

Quoted 15 lines
    acpi_processor_ffh_cstate_enter()
      mwait_idle_with_hints() (32 bit)
        local_irq_enable()

sched_clock_idle_wakeup_event()

I think my recent idle patches should address this, no?

Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
[   13.366851]  [<f883919b>] acpi_idle_enter_bm+0x2b3/0x333 [processor]
[   13.370184]  [<c02e5ec1>] cpuidle_idle_call+0x63/0x92
[   13.373517]  [<c02e5e5e>] ? cpuidle_idle_call+0x0/0x92
[   13.380184]  [<c010257d>] cpu_idle+0xb6/0xd6
[   13.383517]  [<c0387071>] rest_init+0x49/0x4b
[   13.386850]  =======================

(I just found out I failed to copy LKML on the last discussion about these patches)

---
Subject: [PATCH] idle (arch,acpi and apm) and lockdep
From: Peter Zijlstra <[email protected]>

OK, so 25-mm1 gave a lockdep error which made me look into this.

The first thing that I noticed was the horrible mess; the second thing I saw was hacks like: 71e93d15612c61c2e26a169567becf088e71b8ff

The problem is that arch idle routines are somewhat inconsitent with their IRQ state handling and instead of fixing _that_, we go paper over the problem.

So the thing I've tried to do is set a standard for idle routines and fix them all up to adhere to that. So the rules are:

idle routines are entered with IRQs disabled
idle routines will exit with IRQs enabled

Nearly all already did this in one form or another.

Merge the 32 and 64 bit bits so they no longer have different bugs.

As for the actual lockdep warning; __sti_mwait() did a plainly un-annotated irq-enable.

Signed-off-by Peter Zijlstra <[email protected]>
---
 arch/x86/kernel/apm_32.c      |    3 +
 arch/x86/kernel/process.c     |  117 +++++++++++++++++++++++++++++++++++++++
 arch/x86/kernel/process_32.c  |  118 +---------------------------------------
 arch/x86/kernel/process_64.c  |  123 +-----------------------------------------
 drivers/acpi/processor_idle.c |   19 +++---
 include/asm-x86/processor.h   |    1 
 6 files changed, 137 insertions(+), 244 deletions(-)
arch/x86/kernel/process_32.c +4 -114
Index: linux-2.6-2/arch/x86/kernel/process_32.c
===================================================================
@@ -111,12 +111,10 @@ void default_idle(void)
 		 */
 		smp_mb();
 
-		local_irq_disable();
-		if (!need_resched()) {
+		if (!need_resched())
 			safe_halt();	/* enables interrupts racelessly */
-			local_irq_disable();
-		}
-		local_irq_enable();
+		else
+			local_irq_enable();
 		current_thread_info()->status |= TS_POLLING;
 	} else {
 		local_irq_enable();

@@ -128,17 +126,6 @@ void default_idle(void)
 EXPORT_SYMBOL(default_idle);
 #endif
 
-/*
- * On SMP it's slightly faster (but much more power-consuming!)
- * to poll the ->work.need_resched flag instead of waiting for the
- * cross-CPU IPI to arrive. Use this option with caution.
- */
-static void poll_idle(void)
-{
-	local_irq_enable();
-	cpu_relax();
-}
-
 #ifdef CONFIG_HOTPLUG_CPU
 #include <asm/nmi.h>
 /* We don't actually take CPU down, just spin without interrupts. */

@@ -196,6 +183,7 @@ void cpu_idle(void)
 			if (cpu_is_offline(cpu))
 				play_dead();
 
+			local_irq_disable();
 			__get_cpu_var(irq_stat).idle_timestamp = jiffies;
 			idle();
 		}

@@ -206,104 +194,6 @@ void cpu_idle(void)
 	}
 }
 
-static void do_nothing(void *unused)
-{
-}
-
-/*
- * cpu_idle_wait - Used to ensure that all the CPUs discard old value of
- * pm_idle and update to new pm_idle value. Required while changing pm_idle
- * handler on SMP systems.
- *
- * Caller must have changed pm_idle to the new value before the call. Old
- * pm_idle value will not be used by any CPU after the return of this function.
- */
-void cpu_idle_wait(void)
-{
-	smp_mb();
-	/* kick all the CPUs so that they exit out of pm_idle */
-	smp_call_function(do_nothing, NULL, 0, 1);
-}
-EXPORT_SYMBOL_GPL(cpu_idle_wait);
-
-/*
- * This uses new MONITOR/MWAIT instructions on P4 processors with PNI,
- * which can obviate IPI to trigger checking of need_resched.
- * We execute MONITOR against need_resched and enter optimized wait state
- * through MWAIT. Whenever someone changes need_resched, we would be woken
- * up from MWAIT (without an IPI).
- *
- * New with Core Duo processors, MWAIT can take some hints based on CPU
- * capability.
- */
-void mwait_idle_with_hints(unsigned long ax, unsigned long cx)
-{
-	if (!need_resched()) {
-		__monitor((void *)&current_thread_info()->flags, 0, 0);
-		smp_mb();
-		if (!need_resched())
-			__sti_mwait(ax, cx);
-		else
-			local_irq_enable();
-	} else
-		local_irq_enable();
-}
-
-/* Default MONITOR/MWAIT with no hints, used for default C1 state */
-static void mwait_idle(void)
-{
-	local_irq_enable();
-	mwait_idle_with_hints(0, 0);
-}
-
-static int __cpuinit mwait_usable(const struct cpuinfo_x86 *c)
-{
-	if (force_mwait)
-		return 1;
-	/* Any C1 states supported? */
-	return c->cpuid_level >= 5 && ((cpuid_edx(5) >> 4) & 0xf) > 0;
-}
-
-void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c)
-{
-	static int selected;
-
-	if (selected)
-		return;
-#ifdef CONFIG_X86_SMP
-	if (pm_idle == poll_idle && smp_num_siblings > 1) {
-		printk(KERN_WARNING "WARNING: polling idle and HT enabled,"
-			" performance may degrade.\n");
-	}
-#endif
-	if (cpu_has(c, X86_FEATURE_MWAIT) && mwait_usable(c)) {
-		/*
-		 * Skip, if setup has overridden idle.
-		 * One CPU supports mwait => All CPUs supports mwait
-		 */
-		if (!pm_idle) {
-			printk(KERN_INFO "using mwait in idle threads.\n");
-			pm_idle = mwait_idle;
-		}
-	}
-	selected = 1;
-}
-
-static int __init idle_setup(char *str)
-{
-	if (!strcmp(str, "poll")) {
-		printk("using polling idle threads.\n");
-		pm_idle = poll_idle;
-	} else if (!strcmp(str, "mwait"))
-		force_mwait = 1;
-	else
-		return -1;
-
-	boot_option_idle_override = 1;
-	return 0;
-}
-early_param("idle", idle_setup);
-
 void __show_registers(struct pt_regs *regs, int all)
 {
 	unsigned long cr0 = 0L, cr2 = 0L, cr3 = 0L, cr4 = 0L;
arch/x86/kernel/process_64.c +3 -120

139 lines not on this page · raw message

drivers/acpi/processor_idle.c +9 -10
Index: linux-2.6-2/drivers/acpi/processor_idle.c
===================================================================
@@ -418,13 +418,12 @@ static void acpi_processor_idle(void)
 
 	cx = pr->power.state;
 	if (!cx || acpi_idle_suspend) {
-		if (pm_idle_save)
-			pm_idle_save();
-		else
+		if (pm_idle_save) {
+			pm_idle_save(); /* enables IRQs */
+		} else {
 			acpi_safe_halt();
-
-		if (irqs_disabled())
 			local_irq_enable();
+		}
 
 		return;
 	}

@@ -520,10 +519,12 @@ static void acpi_processor_idle(void)
 		 * Use the appropriate idle routine, the one that would
 		 * be used without acpi C-states.
 		 */
-		if (pm_idle_save)
-			pm_idle_save();
-		else
+		if (pm_idle_save) {
+			pm_idle_save(); /* enables IRQs */
+		} else {
 			acpi_safe_halt();
+			local_irq_enable();
+		}
 
 		/*
 		 * TBD: Can't get time duration while in C1, as resumes

@@ -534,8 +535,6 @@ static void acpi_processor_idle(void)
 		 *       skew otherwise.
 		 */
 		sleep_ticks = 0xFFFFFFFF;
-		if (irqs_disabled())
-			local_irq_enable();
 
 		break;
arch/x86/kernel/apm_32.c +3 -0
Index: linux-2.6-2/arch/x86/kernel/apm_32.c
===================================================================
@@ -904,6 +904,7 @@ recalc:
 			original_pm_idle();
 		else
 			default_idle();
+		local_irq_disable();
 		jiffies_since_last_check = jiffies - last_jiffies;
 		if (jiffies_since_last_check > idle_period)
 			goto recalc;

@@ -911,6 +912,8 @@ recalc:
 
 	if (apm_idle_done)
 		apm_do_busy();
+
+	local_irq_enable();
 }
 
 /**
include/asm-x86/processor.h +1 -0
Index: linux-2.6-2/include/asm-x86/processor.h
===================================================================
@@ -723,6 +723,7 @@ static inline void __mwait(unsigned long
 
 static inline void __sti_mwait(unsigned long eax, unsigned long ecx)
 {
+	trace_hardirqs_on();
 	/* "mwait %eax, %ecx;" */
 	asm volatile("sti; .byte 0x0f, 0x01, 0xc9;"
 		     :: "a" (eax), "c" (ecx));
arch/x86/kernel/process.c +117 -0

126 lines not on this page · raw message

author
#5 lore raw

Re: [PATCH] idle (arch, acpi and apm) and lockdep

O.K. I applied this patch, below are the results:

Quoted 607 lines · Peter Zijlstra
Quoted 51 lines · Peter Zijlstra
Quoted 38 lines · Justin Mattock
Quoted 37 lines · Peter Zijlstra
Quoted 35 lines · Justin Mattock
[   13.269763] =================================
[   13.270954] [ INFO: inconsistent lock state ]
[   13.271865] 2.6.25-04569-gb69d398 #3
[   13.272614] ---------------------------------
[   13.273521] inconsistent {in-hardirq-W} -> {hardirq-on-W} usage.
[   13.274745] swapper/0 [HC0[0]:SC0[0]:HE1:SE1] takes:
[   13.275787]  (&rq->rq_lock_key){++..}, at: [<c01208e9>] sched_clock_idle_wakeup_event+0x43/0x74
[   13.276859] {in-hardirq-W} state was registered at:
[   13.276859]   [<c01423d8>] __lock_acquire+0x419/0xb70
[   13.276859]   [<c0142bae>] lock_acquire+0x7f/0xa6
[   13.280188]   [<c0394132>] _spin_lock+0x1c/0x49
[   13.280188]   [<c0121c44>] scheduler_tick+0x43/0x1bd
[   13.280188]   [<c012d99f>] update_process_times+0x3d/0x49
[   13.283524]   [<c013daf6>] tick_periodic+0x66/0x72
[   13.283524]   [<c013db1b>] tick_handle_periodic+0x19/0x6a
[   13.283524]   [<c0105a87>] timer_interrupt+0x47/0x4e
[   13.286855]   [<c015cf64>] handle_IRQ_event+0x1a/0x4f
[   13.286855]   [<c015e3e9>] handle_level_irq+0x7f/0xca
[   13.286855]   [<c0105a27>] do_IRQ+0x71/0x8a
[   13.290190]   [<c01043fe>] common_interrupt+0x2e/0x34
[   13.290190]   [<c038bd6f>] calibrate_delay+0x8f/0x276
[   13.290190]   [<c04f0883>] start_kernel+0x27c/0x2f8
[   13.293520]   [<c04f0008>] __init_begin+0x8/0xa
[   13.293520]   [<ffffffff>] 0xffffffff
[   13.293520] irq event stamp: 253965
[   13.296856] hardirqs last  enabled at (253965): [<c0108069>] native_sched_clock+0xe7/0xff
[   13.296856] hardirqs last disabled at (253964): [<c0107fef>] native_sched_clock+0x6d/0xff
[   13.300190] softirqs last  enabled at (253958): [<c0129cea>] __do_softirq+0xf9/0xff
[   13.300190] softirqs last disabled at (253953): [<c0129d3d>] do_softirq+0x4d/0x79
[   13.303522]
[   13.303522] other info that might help us debug this:
[   13.303522] no locks held by swapper/0.
[   13.303522]
[   13.303522] stack backtrace:
[   13.306852] Pid: 0, comm: swapper Not tainted 2.6.25-04569-gb69d398 #3
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
[   13.336851]  [<c01409f9>] print_usage_bug+0x106/0x113
[   13.340185]  [<c0141823>] mark_lock+0x1ed/0x3a5
[   13.343519]  [<c014244d>] __lock_acquire+0x48e/0xb70
[   13.346852]  [<c0142bae>] lock_acquire+0x7f/0xa6
[   13.350185]  [<c01208e9>] ? sched_clock_idle_wakeup_event+0x43/0x74
[   13.353519]  [<c0394132>] _spin_lock+0x1c/0x49
[   13.360185]  [<c01208e9>] ? sched_clock_idle_wakeup_event+0x43/0x74
[   13.363518]  [<c01208e9>] sched_clock_idle_wakeup_event+0x43/0x74

Got it:

acpi_idle_do_entry()

Quoted 15 lines
    acpi_processor_ffh_cstate_enter()
      mwait_idle_with_hints() (32 bit)
        local_irq_enable()

sched_clock_idle_wakeup_event()

I think my recent idle patches should address this, no?

Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
[   13.366851]  [<f883919b>] acpi_idle_enter_bm+0x2b3/0x333 [processor]
[   13.370184]  [<c02e5ec1>] cpuidle_idle_call+0x63/0x92
[   13.373517]  [<c02e5e5e>] ? cpuidle_idle_call+0x0/0x92
[   13.380184]  [<c010257d>] cpu_idle+0xb6/0xd6
[   13.383517]  [<c0387071>] rest_init+0x49/0x4b
[   13.386850]  =======================

(I just found out I failed to copy LKML on the last discussion about these patches)

---
Subject: [PATCH] idle (arch,acpi and apm) and lockdep
From: Peter Zijlstra <[email protected]>

OK, so 25-mm1 gave a lockdep error which made me look into this.

The first thing that I noticed was the horrible mess; the second thing I saw was hacks like: 71e93d15612c61c2e26a169567becf088e71b8ff

The problem is that arch idle routines are somewhat inconsitent with their IRQ state handling and instead of fixing _that_, we go paper over the problem.

So the thing I've tried to do is set a standard for idle routines and fix them all up to adhere to that. So the rules are:

idle routines are entered with IRQs disabled
idle routines will exit with IRQs enabled

Nearly all already did this in one form or another.

Merge the 32 and 64 bit bits so they no longer have different bugs.

As for the actual lockdep warning; __sti_mwait() did a plainly un-annotated irq-enable.

Signed-off-by Peter Zijlstra <[email protected]>

---

  arch/x86/kernel/apm_32.c      |    3 +
  arch/x86/kernel/process.c     |  117 +++++++++++++++++++++++++++++++++++++++
  arch/x86/kernel/process_32.c  |  118 +---------------------------------------
  arch/x86/kernel/process_64.c  |  123 +-----------------------------------------
  drivers/acpi/processor_idle.c |   19 +++---
  include/asm-x86/processor.h   |    1
  6 files changed, 137 insertions(+), 244 deletions(-)

Index: linux-2.6-2/arch/x86/kernel/process_32.c
===================================================================
--- linux-2.6-2.orig/arch/x86/kernel/process_32.c +++ linux-2.6-2/arch/x86/kernel/process_32.c @@ -111,12 +111,10 @@ void default_idle(void)

                 */
                smp_mb();
 -               local_irq_disable();
 -               if (!need_resched()) {
 +               if (!need_resched())
                        safe_halt();    /* enables interrupts racelessly */
 -                       local_irq_disable();
 -               }
 -               local_irq_enable();
 +               else
 +                       local_irq_enable();
                current_thread_info()->status |= TS_POLLING;
        } else {
                local_irq_enable();

@@ -128,17 +126,6 @@ void default_idle(void)
EXPORT_SYMBOL(default_idle);
#endif

-/*
- * On SMP it's slightly faster (but much more power-consuming!)
- * to poll the ->work.need_resched flag instead of waiting for the
- * cross-CPU IPI to arrive. Use this option with caution.
- */
-static void poll_idle(void)
-{
- local_irq_enable();
- cpu_relax();
-}
-
#ifdef CONFIG_HOTPLUG_CPU
#include <asm/nmi.h>
/* We don't actually take CPU down, just spin without interrupts. */
@@ -196,6 +183,7 @@ void cpu_idle(void)

                        if (cpu_is_offline(cpu))
                                play_dead();
 +                       local_irq_disable();
                        __get_cpu_var(irq_stat).idle_timestamp = jiffies;
                        idle();
                }
 @@ -206,104 +194,6 @@ void cpu_idle(void)
        }

}

-static void do_nothing(void *unused)
-{
-}
-
-/*
- * cpu_idle_wait - Used to ensure that all the CPUs discard old value of
- * pm_idle and update to new pm_idle value. Required while changing pm_idle
- * handler on SMP systems.
- *
- * Caller must have changed pm_idle to the new value before the call. Old
- * pm_idle value will not be used by any CPU after the return of this function.
- */
-void cpu_idle_wait(void)
-{

 -       smp_mb();
 -       /* kick all the CPUs so that they exit out of pm_idle */
 -       smp_call_function(do_nothing, NULL, 0, 1);
 -}
 -EXPORT_SYMBOL_GPL(cpu_idle_wait);
 -
 -/*
 - * This uses new MONITOR/MWAIT instructions on P4 processors with PNI,
 - * which can obviate IPI to trigger checking of need_resched.
 - * We execute MONITOR against need_resched and enter optimized wait state
 - * through MWAIT. Whenever someone changes need_resched, we would be woken
 - * up from MWAIT (without an IPI).
 - *
 - * New with Core Duo processors, MWAIT can take some hints based on CPU
 - * capability.
 - */
 -void mwait_idle_with_hints(unsigned long ax, unsigned long cx)
 -{
 -       if (!need_resched()) {
 -               __monitor((void *)&current_thread_info()->flags, 0, 0);
 -               smp_mb();
 -               if (!need_resched())
 -                       __sti_mwait(ax, cx);
 -               else
 -                       local_irq_enable();
 -       } else
 -               local_irq_enable();
 -}
 -
 -/* Default MONITOR/MWAIT with no hints, used for default C1 state */
 -static void mwait_idle(void)
 -{
 -       local_irq_enable();
 -       mwait_idle_with_hints(0, 0);
 -}
 -
 -static int __cpuinit mwait_usable(const struct cpuinfo_x86 *c)
 -{
 -       if (force_mwait)
 -               return 1;
 -       /* Any C1 states supported? */
 -       return c->cpuid_level >= 5 && ((cpuid_edx(5) >> 4) & 0xf) > 0;
 -}
 -
 -void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c)
 -{
 -       static int selected;
 -
 -       if (selected)
 -               return;
 -#ifdef CONFIG_X86_SMP
 -       if (pm_idle == poll_idle && smp_num_siblings > 1) {
 -               printk(KERN_WARNING "WARNING: polling idle and HT enabled,"
 -                       " performance may degrade.\n");
 -       }
 -#endif
 -       if (cpu_has(c, X86_FEATURE_MWAIT) && mwait_usable(c)) {
 -               /*
 -                * Skip, if setup has overridden idle.
 -                * One CPU supports mwait => All CPUs supports mwait
 -                */
 -               if (!pm_idle) {
 -                       printk(KERN_INFO "using mwait in idle threads.\n");
 -                       pm_idle = mwait_idle;
 -               }
 -       }
 -       selected = 1;
 -}
 -
 -static int __init idle_setup(char *str)
 -{
 -       if (!strcmp(str, "poll")) {
 -               printk("using polling idle threads.\n");
 -               pm_idle = poll_idle;
 -       } else if (!strcmp(str, "mwait"))
 -               force_mwait = 1;
 -       else
 -               return -1;
 -
 -       boot_option_idle_override = 1;
 -       return 0;
 -}
 -early_param("idle", idle_setup);
 -
  void __show_registers(struct pt_regs *regs, int all)
  {
        unsigned long cr0 = 0L, cr2 = 0L, cr3 = 0L, cr4 = 0L;
 Index: linux-2.6-2/arch/x86/kernel/process_64.c
 ===================================================================
 --- linux-2.6-2.orig/arch/x86/kernel/process_64.c
 +++ linux-2.6-2/arch/x86/kernel/process_64.c
 @@ -106,26 +106,13 @@ void default_idle(void)
         * test NEED_RESCHED:
         */
        smp_mb();
 -       local_irq_disable();
 -       if (!need_resched()) {
 +       if (!need_resched())
                safe_halt();    /* enables interrupts racelessly */
 -               local_irq_disable();
 -       }
 -       local_irq_enable();
 +       else
 +               local_irq_enable();
        current_thread_info()->status |= TS_POLLING;

}

-/*
- * On SMP it's slightly faster (but much more power-consuming!)
- * to poll the ->need_resched flag instead of waiting for the
- * cross-CPU IPI to arrive. Use this option with caution.
- */
-static void poll_idle(void)
-{
- local_irq_enable();
- cpu_relax();
-}
-
#ifdef CONFIG_HOTPLUG_CPU
DECLARE_PER_CPU(int, cpu_state);

@@ -192,110 +179,6 @@ void cpu_idle(void)

        }

}

-static void do_nothing(void *unused)
-{
-}
-
-/*
- * cpu_idle_wait - Used to ensure that all the CPUs discard old value of
- * pm_idle and update to new pm_idle value. Required while changing pm_idle
- * handler on SMP systems.
- *
- * Caller must have changed pm_idle to the new value before the call. Old
- * pm_idle value will not be used by any CPU after the return of this function.
- */
-void cpu_idle_wait(void)
-{

 -       smp_mb();
 -       /* kick all the CPUs so that they exit out of pm_idle */
 -       smp_call_function(do_nothing, NULL, 0, 1);
 -}
 -EXPORT_SYMBOL_GPL(cpu_idle_wait);
 -
 -/*
 - * This uses new MONITOR/MWAIT instructions on P4 processors with PNI,
 - * which can obviate IPI to trigger checking of need_resched.
 - * We execute MONITOR against need_resched and enter optimized wait state
 - * through MWAIT. Whenever someone changes need_resched, we would be woken
 - * up from MWAIT (without an IPI).
 - *
 - * New with Core Duo processors, MWAIT can take some hints based on CPU
 - * capability.
 - */
 -void mwait_idle_with_hints(unsigned long ax, unsigned long cx)
 -{
 -       if (!need_resched()) {
 -               __monitor((void *)&current_thread_info()->flags, 0, 0);
 -               smp_mb();
 -               if (!need_resched())
 -                       __mwait(ax, cx);
 -       }
 -}
 -
 -/* Default MONITOR/MWAIT with no hints, used for default C1 state */
 -static void mwait_idle(void)
 -{
 -       if (!need_resched()) {
 -               __monitor((void *)&current_thread_info()->flags, 0, 0);
 -               smp_mb();
 -               if (!need_resched())
 -                       __sti_mwait(0, 0);
 -               else
 -                       local_irq_enable();
 -       } else {
 -               local_irq_enable();
 -       }
 -}
 -
 -
 -static int __cpuinit mwait_usable(const struct cpuinfo_x86 *c)
 -{
 -       if (force_mwait)
 -               return 1;
 -       /* Any C1 states supported? */
 -       return c->cpuid_level >= 5 && ((cpuid_edx(5) >> 4) & 0xf) > 0;
 -}
 -
 -void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c)
 -{
 -       static int selected;
 -
 -       if (selected)
 -               return;
 -#ifdef CONFIG_X86_SMP
 -       if (pm_idle == poll_idle && smp_num_siblings > 1) {
 -               printk(KERN_WARNING "WARNING: polling idle and HT enabled,"
 -                       " performance may degrade.\n");
 -       }
 -#endif
 -       if (cpu_has(c, X86_FEATURE_MWAIT) && mwait_usable(c)) {
 -               /*
 -                * Skip, if setup has overridden idle.
 -                * One CPU supports mwait => All CPUs supports mwait
 -                */

- if (!pm_idle) {
- printk(KERN_INFO "using mwait in idle threads.\n");
- pm_idle = mwait_idle;
- }
- }
- selected = 1;
-}
-
-static int __init idle_setup(char *str)
-{
- if (!strcmp(str, "poll")) {
- printk("using polling idle threads.\n");
- pm_idle = poll_idle;
- } else if (!strcmp(str, "mwait"))
- force_mwait = 1;
- else
- return -1;
-
- boot_option_idle_override = 1;
- return 0;
-}
-early_param("idle", idle_setup);
-
/* Prints also some state that isn't saved in the pt_regs */ void __show_regs(struct pt_regs * regs)
{
Index: linux-2.6-2/drivers/acpi/processor_idle.c
===================================================================
--- linux-2.6-2.orig/drivers/acpi/processor_idle.c +++ linux-2.6-2/drivers/acpi/processor_idle.c
@@ -418,13 +418,12 @@ static void acpi_processor_idle(void)

        cx = pr->power.state;
        if (!cx || acpi_idle_suspend) {
 -               if (pm_idle_save)
 -                       pm_idle_save();
 -               else
 +               if (pm_idle_save) {
 +                       pm_idle_save(); /* enables IRQs */
 +               } else {
                        acpi_safe_halt();
 -
 -               if (irqs_disabled())
                        local_irq_enable();

+ }

                return;
        }
 @@ -520,10 +519,12 @@ static void acpi_processor_idle(void)
                 * Use the appropriate idle routine, the one that would
                 * be used without acpi C-states.
                 */
 -               if (pm_idle_save)
 -                       pm_idle_save();
 -               else
 +               if (pm_idle_save) {
 +                       pm_idle_save(); /* enables IRQs */
 +               } else {
                        acpi_safe_halt();

+ local_irq_enable();
+ }

/*
* TBD: Can't get time duration while in C1, as resumes
@@ -534,8 +535,6 @@ static void acpi_processor_idle(void)

                 *       skew otherwise.
                 */
                sleep_ticks = 0xFFFFFFFF;

- if (irqs_disabled())
- local_irq_enable();

                break;

Index: linux-2.6-2/arch/x86/kernel/apm_32.c
===================================================================
--- linux-2.6-2.orig/arch/x86/kernel/apm_32.c
+++ linux-2.6-2/arch/x86/kernel/apm_32.c
@@ -904,6 +904,7 @@ recalc:

                        original_pm_idle();
                else
                        default_idle();
 +               local_irq_disable();
                jiffies_since_last_check = jiffies - last_jiffies;
                if (jiffies_since_last_check > idle_period)
                        goto recalc;

@@ -911,6 +912,8 @@ recalc:

        if (apm_idle_done)
                apm_do_busy();

+
+ local_irq_enable();
}

/**
Index: linux-2.6-2/include/asm-x86/processor.h
===================================================================
--- linux-2.6-2.orig/include/asm-x86/processor.h +++ linux-2.6-2/include/asm-x86/processor.h
@@ -723,6 +723,7 @@ static inline void __mwait(unsigned long

static inline void __sti_mwait(unsigned long eax, unsigned long ecx) {

 +       trace_hardirqs_on();
        /* "mwait %eax, %ecx;" */
        asm volatile("sti; .byte 0x0f, 0x01, 0xc9;"
                     :: "a" (eax), "c" (ecx));

Index: linux-2.6-2/arch/x86/kernel/process.c
===================================================================
--- linux-2.6-2.orig/arch/x86/kernel/process.c
+++ linux-2.6-2/arch/x86/kernel/process.c
@@ -4,6 +4,8 @@
#include <linux/smp.h>
#include <linux/slab.h>
#include <linux/sched.h>
+#include <linux/module.h>
+#include <linux/pm.h>

struct kmem_cache *task_xstate_cachep;

@@ -42,3 +44,118 @@ void arch_task_cache_init(void)

                                  __alignof__(union thread_xstate),
                                  SLAB_PANIC, NULL);
  }
 +
 +static void do_nothing(void *unused)
 +{
 +}
 +
 +/*
 + * cpu_idle_wait - Used to ensure that all the CPUs discard old value of
 + * pm_idle and update to new pm_idle value. Required while changing pm_idle
 + * handler on SMP systems.
 + *
 + * Caller must have changed pm_idle to the new value before the call. Old
 + * pm_idle value will not be used by any CPU after the return of this function.
 + */
 +void cpu_idle_wait(void)
 +{
 +       smp_mb();
 +       /* kick all the CPUs so that they exit out of pm_idle */
 +       smp_call_function(do_nothing, NULL, 0, 1);
 +}
 +EXPORT_SYMBOL_GPL(cpu_idle_wait);
 +
 +/*
 + * This uses new MONITOR/MWAIT instructions on P4 processors with PNI,
 + * which can obviate IPI to trigger checking of need_resched.
 + * We execute MONITOR against need_resched and enter optimized wait state
 + * through MWAIT. Whenever someone changes need_resched, we would be woken
 + * up from MWAIT (without an IPI).
 + *
 + * New with Core Duo processors, MWAIT can take some hints based on CPU
 + * capability.
 + */
 +void mwait_idle_with_hints(unsigned long ax, unsigned long cx)
 +{
 +       if (!need_resched()) {
 +               __monitor((void *)&current_thread_info()->flags, 0, 0);
 +               smp_mb();
 +               if (!need_resched())
 +                       __mwait(ax, cx);
 +       }
 +}
 +
 +/* Default MONITOR/MWAIT with no hints, used for default C1 state */
 +static void mwait_idle(void)
 +{
 +       if (!need_resched()) {
 +               __monitor((void *)&current_thread_info()->flags, 0, 0);
 +               smp_mb();
 +               if (!need_resched())
 +                       __sti_mwait(0, 0);
 +               else
 +                       local_irq_enable();
 +       } else
 +               local_irq_enable();
 +}
 +
 +
 +static int __cpuinit mwait_usable(const struct cpuinfo_x86 *c)
 +{
 +       if (force_mwait)
 +               return 1;
 +       /* Any C1 states supported? */
 +       return c->cpuid_level >= 5 && ((cpuid_edx(5) >> 4) & 0xf) > 0;
 +}
 +
 +/*
 + * On SMP it's slightly faster (but much more power-consuming!)
 + * to poll the ->work.need_resched flag instead of waiting for the
 + * cross-CPU IPI to arrive. Use this option with caution.
 + */
 +static void poll_idle(void)
 +{
 +       local_irq_enable();
 +       cpu_relax();
 +}
 +
 +void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c)
 +{
 +       static int selected;
 +
 +       if (selected)
 +               return;
 +#ifdef CONFIG_X86_SMP
 +       if (pm_idle == poll_idle && smp_num_siblings > 1) {
 +               printk(KERN_WARNING "WARNING: polling idle and HT enabled,"
 +                       " performance may degrade.\n");
 +       }
 +#endif
 +       if (cpu_has(c, X86_FEATURE_MWAIT) && mwait_usable(c)) {
 +               /*
 +                * Skip, if setup has overridden idle.
 +                * One CPU supports mwait => All CPUs supports mwait
 +                */

+ if (!pm_idle) {
+ printk(KERN_INFO "using mwait in idle threads.\n");
+ pm_idle = mwait_idle;
+ }
+ }
+ selected = 1;
+}
+
+static int __init idle_setup(char *str)
+{
+ if (!strcmp(str, "poll")) {
+ printk("using polling idle threads.\n");
+ pm_idle = poll_idle;
+ } else if (!strcmp(str, "mwait"))
+ force_mwait = 1;

Quoted 11 lines

+ else
+ return -1;
+
+ boot_option_idle_override = 1;
+ return 0;
+}
+early_param("idle", idle_setup);
+

patching file arch/x86/kernel/process_32.c
Hunk #1 FAILED at 111.
Hunk #2 FAILED at 126.
Hunk #3 FAILED at 183.
Hunk #4 FAILED at 194.
4 out of 4 hunks FAILED -- saving rejects to file arch/x86/kernel/process_32.c.rej
patching file arch/x86/kernel/process_64.c
Hunk #1 FAILED at 106.
Hunk #2 FAILED at 179.
2 out of 2 hunks FAILED -- saving rejects to file arch/x86/kernel/process_64.c.rej
patching file drivers/acpi/processor_idle.c
Hunk #1 FAILED at 418.
Hunk #2 FAILED at 519.
Hunk #3 FAILED at 535.
3 out of 3 FAILED -- saving rejects to file drivers/acpi/processor_idle.c.rej pathing file arch/x86/kernel/apm_32.c
Hunk #1 FAILED at 904.
Hunk #2 FAILED at 912.
2 out of 2 hunks FAILED -- saving rejects to file arch/x86/kernel/apm_32.c.rej patching file include/asm/asm-x86/processor.h
Hunk #1 FAILED at 723.
1 out of 1 hunk FAILE -- saving rejects to file include/asm-x86/processor.h.rej patching file arch/x86/kernel/process.c
Hunk #2 succeeded at 44 with fuzz 2.

When I compile the kernel I'm receiving this.

arch/x86/kernel/process.o: In function `select_idle_routine':
process.c:(.cpuinit.text+0x0): multiple definition of select_idle_routine' arch/x86/kernel/process_32.o:process_32.c(.cpuinit.text+0x0: first defined here arch/x86/kernel/process.o: In function cpu_idle_wait':
process.c:(text+0x16): multiple definition of cpu_idle_wait' arch/x86/kernel/process_32.o:process_32.c:(.text0x432): first defined here arch/x86/kernel/process.o: In function mwait_idle_with_hints':
process.c:(.text+0x113): multiple definition of 'mwait_idle_with_hints' arch/x86/kernel/process_32.o:process_32.c:(.text+0x1db): first defined here make[1]: *** [arch/x86/kernel/built-in.o] Error 1
make: *** [arch/x86/kernel] Error 2

Should I just go ahead and try git-pull to see if these patches are applied over there correctly.
regards;

Signature · 1 line
Justin P. Mattock
#6 lore raw

Re: [PATCH] idle (arch, acpi and apm) and lockdep

Quoted 44 lines · Justin Mattock

O.K. I applied this patch, below are the results:

patching file arch/x86/kernel/process_32.c
Hunk #1 FAILED at 111.
Hunk #2 FAILED at 126.
Hunk #3 FAILED at 183.
Hunk #4 FAILED at 194.
4 out of 4 hunks FAILED -- saving rejects to file arch/x86/kernel/process_32.c.rej
patching file arch/x86/kernel/process_64.c
Hunk #1 FAILED at 106.
Hunk #2 FAILED at 179.
2 out of 2 hunks FAILED -- saving rejects to file arch/x86/kernel/process_64.c.rej
patching file drivers/acpi/processor_idle.c
Hunk #1 FAILED at 418.
Hunk #2 FAILED at 519.
Hunk #3 FAILED at 535.
3 out of 3 FAILED -- saving rejects to file drivers/acpi/processor_idle.c.rej pathing file arch/x86/kernel/apm_32.c
Hunk #1 FAILED at 904.
Hunk #2 FAILED at 912.
2 out of 2 hunks FAILED -- saving rejects to file arch/x86/kernel/apm_32.c.rej patching file include/asm/asm-x86/processor.h
Hunk #1 FAILED at 723.
1 out of 1 hunk FAILE -- saving rejects to file include/asm-x86/processor.h.rej patching file arch/x86/kernel/process.c
Hunk #2 succeeded at 44 with fuzz 2.

When I compile the kernel I'm receiving this.

arch/x86/kernel/process.o: In function `select_idle_routine':
process.c:(.cpuinit.text+0x0): multiple definition of select_idle_routine' arch/x86/kernel/process_32.o:process_32.c(.cpuinit.text+0x0: first defined here arch/x86/kernel/process.o: In function cpu_idle_wait':
process.c:(text+0x16): multiple definition of cpu_idle_wait' arch/x86/kernel/process_32.o:process_32.c:(.text0x432): first defined here arch/x86/kernel/process.o: In function mwait_idle_with_hints':
process.c:(.text+0x113): multiple definition of 'mwait_idle_with_hints' arch/x86/kernel/process_32.o:process_32.c:(.text+0x1db): first defined here make[1]: *** [arch/x86/kernel/built-in.o] Error 1
make: *** [arch/x86/kernel] Error 2

Quoted 3 lines

Should I just go ahead and try git-pull to see if these patches are applied over there correctly.
regards;

That's no good :-/

weird, I have:

# git describe
v2.6.25-4569-gb69d398
# quilt push
Applying patch patches/idle-lockdep-1.patch
patching file arch/x86/kernel/process_32.c
patching file arch/x86/kernel/process_64.c
patching file drivers/acpi/processor_idle.c
patching file arch/x86/kernel/apm_32.c
patching file include/asm-x86/processor.h
patching file arch/x86/kernel/process.c

Now at patch patches/idle-lockdep-1.patch

And I did a i386 and x86_64 defconfig build before I send it out.

author
#7 lore raw

Re: [PATCH] idle (arch, acpi and apm) and lockdep

Quoted 69 lines · Peter Zijlstra
Quoted 46 lines · Justin Mattock

O.K. I applied this patch, below are the results:

patching file arch/x86/kernel/process_32.c
Hunk #1 FAILED at 111.
Hunk #2 FAILED at 126.
Hunk #3 FAILED at 183.
Hunk #4 FAILED at 194.
4 out of 4 hunks FAILED -- saving rejects to file arch/x86/kernel/process_32.c.rej
patching file arch/x86/kernel/process_64.c
Hunk #1 FAILED at 106.
Hunk #2 FAILED at 179.
2 out of 2 hunks FAILED -- saving rejects to file arch/x86/kernel/process_64.c.rej
patching file drivers/acpi/processor_idle.c
Hunk #1 FAILED at 418.
Hunk #2 FAILED at 519.
Hunk #3 FAILED at 535.
3 out of 3 FAILED -- saving rejects to file drivers/acpi/processor_idle.c.rej pathing file arch/x86/kernel/apm_32.c
Hunk #1 FAILED at 904.
Hunk #2 FAILED at 912.
2 out of 2 hunks FAILED -- saving rejects to file arch/x86/kernel/apm_32.c.rej patching file include/asm/asm-x86/processor.h
Hunk #1 FAILED at 723.
1 out of 1 hunk FAILE -- saving rejects to file include/asm-x86/processor.h.rej patching file arch/x86/kernel/process.c
Hunk #2 succeeded at 44 with fuzz 2.

When I compile the kernel I'm receiving this.

arch/x86/kernel/process.o: In function `select_idle_routine':
process.c:(.cpuinit.text+0x0): multiple definition of select_idle_routine' arch/x86/kernel/process_32.o:process_32.c(.cpuinit.text+0x0: first defined here arch/x86/kernel/process.o: In function cpu_idle_wait':
process.c:(text+0x16): multiple definition of cpu_idle_wait' arch/x86/kernel/process_32.o:process_32.c:(.text0x432): first defined here arch/x86/kernel/process.o: In function mwait_idle_with_hints':
process.c:(.text+0x113): multiple definition of 'mwait_idle_with_hints' arch/x86/kernel/process_32.o:process_32.c:(.text+0x1db): first defined here make[1]: *** [arch/x86/kernel/built-in.o] Error 1
make: *** [arch/x86/kernel] Error 2

Quoted 3 lines

Should I just go ahead and try git-pull to see if these patches are applied over there correctly.
regards;

That's no good :-/

weird, I have:

 # git describe
 v2.6.25-4569-gb69d398
 # quilt push
 Applying patch patches/idle-lockdep-1.patch

patching file arch/x86/kernel/process_32.c

patching file arch/x86/kernel/process_64.c

patching file drivers/acpi/processor_idle.c
patching file arch/x86/kernel/apm_32.c
patching file include/asm-x86/processor.h

Quoted 8 lines

patching file arch/x86/kernel/process.c

Now at patch patches/idle-lockdep-1.patch

And I did a i386 and x86_64 defconfig build before I send it out.

I did patch -p1 < xxx.patch

Signature · 1 line
Justin P. Mattock
author
#8 lore raw

Re: [PATCH] idle (arch, acpi and apm) and lockdep

Quoted 76 lines · Justin Mattock
Quoted 74 lines · Peter Zijlstra
Quoted 46 lines · Justin Mattock

O.K. I applied this patch, below are the results:

patching file arch/x86/kernel/process_32.c
Hunk #1 FAILED at 111.
Hunk #2 FAILED at 126.
Hunk #3 FAILED at 183.
Hunk #4 FAILED at 194.
4 out of 4 hunks FAILED -- saving rejects to file arch/x86/kernel/process_32.c.rej
patching file arch/x86/kernel/process_64.c
Hunk #1 FAILED at 106.
Hunk #2 FAILED at 179.
2 out of 2 hunks FAILED -- saving rejects to file arch/x86/kernel/process_64.c.rej
patching file drivers/acpi/processor_idle.c
Hunk #1 FAILED at 418.
Hunk #2 FAILED at 519.
Hunk #3 FAILED at 535.
3 out of 3 FAILED -- saving rejects to file drivers/acpi/processor_idle.c.rej pathing file arch/x86/kernel/apm_32.c
Hunk #1 FAILED at 904.
Hunk #2 FAILED at 912.
2 out of 2 hunks FAILED -- saving rejects to file arch/x86/kernel/apm_32.c.rej patching file include/asm/asm-x86/processor.h
Hunk #1 FAILED at 723.
1 out of 1 hunk FAILE -- saving rejects to file include/asm-x86/processor.h.rej patching file arch/x86/kernel/process.c
Hunk #2 succeeded at 44 with fuzz 2.

When I compile the kernel I'm receiving this.

arch/x86/kernel/process.o: In function `select_idle_routine':
process.c:(.cpuinit.text+0x0): multiple definition of select_idle_routine' arch/x86/kernel/process_32.o:process_32.c(.cpuinit.text+0x0: first defined here arch/x86/kernel/process.o: In function cpu_idle_wait':
process.c:(text+0x16): multiple definition of cpu_idle_wait' arch/x86/kernel/process_32.o:process_32.c:(.text0x432): first defined here arch/x86/kernel/process.o: In function mwait_idle_with_hints':
process.c:(.text+0x113): multiple definition of 'mwait_idle_with_hints' arch/x86/kernel/process_32.o:process_32.c:(.text+0x1db): first defined here make[1]: *** [arch/x86/kernel/built-in.o] Error 1
make: *** [arch/x86/kernel] Error 2

Quoted 3 lines

Should I just go ahead and try git-pull to see if these patches are applied over there correctly.
regards;

That's no good :-/

weird, I have:

 # git describe
 v2.6.25-4569-gb69d398
 # quilt push
 Applying patch patches/idle-lockdep-1.patch

patching file arch/x86/kernel/process_32.c

patching file arch/x86/kernel/process_64.c

patching file drivers/acpi/processor_idle.c
patching file arch/x86/kernel/apm_32.c
patching file include/asm-x86/processor.h

patching file arch/x86/kernel/process.c

Now at patch patches/idle-lockdep-1.patch

Quoted 9 lines
Quoted 3 lines

And I did a i386 and x86_64 defconfig build before I send it out.

I did patch -p1 < xxx.patch

--
Justin P. Mattock

Hello;
Thank you very much for that patch, the system is running and I dont
need to use nohz=off ("knock on wood")
I dont know why the patch failed the first time, so I decide to clean the board and start fresh, then manually go in and add and subtract what
you had from the patch.( "I need to get better at reading and writting") So far I'm not receiving any freezes, but it's only been around 20 minutes. I'll leave the system running for a while to see if there's a freeze.
again thank you very much,
regards;

Signature · 1 line
Justin P. Mattock
author
#9 lore raw
In reply to Peter Zijlstra
Quoted 64 lines · Peter Zijlstra
Quoted 8 lines · Justin Mattock
Quoted 7 lines · Peter Zijlstra
Quoted 3 lines · Justin Mattock
[   13.269763] =================================
[   13.270954] [ INFO: inconsistent lock state ]
[   13.271865] 2.6.25-04569-gb69d398 #3

What tree is that? git checkout gb69d398 doesn't want to happen here.

Argh, s/g// so the hash is: b69d398

Could we make that -git-# ?

Quoted 32 lines
Quoted 32 lines
Quoted 32 lines
[   13.272614] ---------------------------------
[   13.273521] inconsistent {in-hardirq-W} -> {hardirq-on-W} usage.
[   13.274745] swapper/0 [HC0[0]:SC0[0]:HE1:SE1] takes:
[   13.275787]  (&rq->rq_lock_key){++..}, at: [<c01208e9>] sched_clock_idle_wakeup_event+0x43/0x74
[   13.276859] {in-hardirq-W} state was registered at:
[   13.276859]   [<c01423d8>] __lock_acquire+0x419/0xb70
[   13.276859]   [<c0142bae>] lock_acquire+0x7f/0xa6
[   13.280188]   [<c0394132>] _spin_lock+0x1c/0x49
[   13.280188]   [<c0121c44>] scheduler_tick+0x43/0x1bd
[   13.280188]   [<c012d99f>] update_process_times+0x3d/0x49
[   13.283524]   [<c013daf6>] tick_periodic+0x66/0x72
[   13.283524]   [<c013db1b>] tick_handle_periodic+0x19/0x6a
[   13.283524]   [<c0105a87>] timer_interrupt+0x47/0x4e
[   13.286855]   [<c015cf64>] handle_IRQ_event+0x1a/0x4f
[   13.286855]   [<c015e3e9>] handle_level_irq+0x7f/0xca
[   13.286855]   [<c0105a27>] do_IRQ+0x71/0x8a
[   13.290190]   [<c01043fe>] common_interrupt+0x2e/0x34
[   13.290190]   [<c038bd6f>] calibrate_delay+0x8f/0x276
[   13.290190]   [<c04f0883>] start_kernel+0x27c/0x2f8
[   13.293520]   [<c04f0008>] __init_begin+0x8/0xa
[   13.293520]   [<ffffffff>] 0xffffffff
[   13.293520] irq event stamp: 253965
[   13.296856] hardirqs last  enabled at (253965): [<c0108069>] native_sched_clock+0xe7/0xff
[   13.296856] hardirqs last disabled at (253964): [<c0107fef>] native_sched_clock+0x6d/0xff
[   13.300190] softirqs last  enabled at (253958): [<c0129cea>] __do_softirq+0xf9/0xff
[   13.300190] softirqs last disabled at (253953): [<c0129d3d>] do_softirq+0x4d/0x79
[   13.303522]
[   13.303522] other info that might help us debug this:
[   13.303522] no locks held by swapper/0.
[   13.303522]
[   13.303522] stack backtrace:
[   13.306852] Pid: 0, comm: swapper Not tainted 2.6.25-04569-gb69d398 #3
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
[   13.336851]  [<c01409f9>] print_usage_bug+0x106/0x113
[   13.340185]  [<c0141823>] mark_lock+0x1ed/0x3a5
[   13.343519]  [<c014244d>] __lock_acquire+0x48e/0xb70
[   13.346852]  [<c0142bae>] lock_acquire+0x7f/0xa6
[   13.350185]  [<c01208e9>] ? sched_clock_idle_wakeup_event+0x43/0x74
[   13.353519]  [<c0394132>] _spin_lock+0x1c/0x49
[   13.360185]  [<c01208e9>] ? sched_clock_idle_wakeup_event+0x43/0x74
[   13.363518]  [<c01208e9>] sched_clock_idle_wakeup_event+0x43/0x74

Got it:

acpi_idle_do_entry()

    acpi_processor_ffh_cstate_enter()
      mwait_idle_with_hints() (32 bit)
        local_irq_enable()
Quoted 6 lines

sched_clock_idle_wakeup_event()

I think my recent idle patches should address this, no?

From what I'm experiencing over here, as soon as I start the system,if

I turn on glxgears from the terminal, the system wont freeze(until after a while when powersave starts to kick in), but If I
start the system and let it sit there the system will freeze. At this point since I dont know where to look in the kernel, where this problem is
at I'm game to try anything to see if it stops the freezing from happening.

Quoted 10 lines
Quoted 10 lines
Quoted 10 lines
Quoted 9 lines


[   13.366851]  [<f883919b>] acpi_idle_enter_bm+0x2b3/0x333 [processor]
[   13.370184]  [<c02e5ec1>] cpuidle_idle_call+0x63/0x92
[   13.373517]  [<c02e5e5e>] ? cpuidle_idle_call+0x0/0x92
[   13.380184]  [<c010257d>] cpu_idle+0xb6/0xd6
[   13.383517]  [<c0387071>] rest_init+0x49/0x4b
[   13.386850]  =======================
Quoted 5 lines
Quoted 3 lines
Quoted 3 lines

looking at current -linus I'm not directly seeing how we can get there with interrupts enabled.

With -git-# should I put that at the end of
git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git
like this linux-2.6-git-# or linux-2.6.-git-#
regards;

Signature · 1 line
Justin P. Mattock