lkml.org 
[lkml]   [2012]   [Sep]   [24]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: RCU idle CPU detection is broken in linux-next
Hi Frederic,

On 09/24/2012 11:29 PM, Frederic Weisbecker wrote:
> Sasha,
>
> Can you please test the following branch:
>
> git://github.com/fweisbec/linux-dynticks.git rcu/idle-for-v3.7-take3
>
> with CONFIG_RCU_USER_QS and CONFIG_RCU_USER_QS_FORCE enabled.
>
> I hope this fixes the warning.
> The changes are:
>
> * add "x86: Unspaghettize do_general_protection()"
> * updated "x86: Exception hooks for userspace RCU extended QS" to
> handle some missed trap handlers. Especially do_general_protection()
> because I can see the problem triggered there in Sasha's warnings. I
> fixed more handlers in the way.

I've tested the branch above, and noticed two things:

- I had merge conflicts when pulling it on top of latest linux-next which I've resolved:

diff --cc arch/x86/kernel/traps.c
index 1ba4850,cb20776..386b079
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@@ -644,8 -646,7 +647,9 @@@ EXPORT_SYMBOL_GPL(math_state_restore)
dotraplinkage void __kprobes
do_device_not_available(struct pt_regs *regs, long error_code)
{
+ BUG_ON(use_eager_fpu());
+
+ exception_enter(regs);
#ifdef CONFIG_MATH_EMULATION
if (read_cr0() & X86_CR0_EM) {
struct math_emu_info info = { };

- While I no longer see the warnings I've originally noticed, if I run with Paul's last debug patch I see the following warning:

[ 212.200137] WARNING: at arch/x86/kernel/process.c:392 cpu_idle+0x12a/0x1f0()
[ 212.200141] Pid: 0, comm: swapper/3 Tainted: G W 3.6.0-rc6-next-20120924-sasha-00030-g71f256c #5
[ 212.200142] Call Trace:
[ 212.200146] [<ffffffff810799da>] ? cpu_idle+0x12a/0x1f0
[ 212.200150] [<ffffffff811078b6>] warn_slowpath_common+0x86/0xb0
[ 212.200153] [<ffffffff811079a5>] warn_slowpath_null+0x15/0x20
[ 212.200156] [<ffffffff810799da>] cpu_idle+0x12a/0x1f0
[ 212.200160] [<ffffffff839a3ee8>] ? setup_APIC_timer+0xb2/0xb6

Which is triggered by:
if (cpuidle_idle_call())
pm_idle();

rcu_idle_exit();
WARN_ON(rcu_is_cpu_idle()); <---- THIS
start_critical_timings();

/* In many cases the interrupt that ended idle
has already called exit_idle. But some idle
loops can be woken up without interrupt. */
WARN_ON(rcu_is_cpu_idle());
__exit_idle();
WARN_ON(rcu_is_cpu_idle());


Thanks,
Sasha


\
 
 \ /
  Last update: 2012-09-25 01:41    [W:0.081 / U:0.048 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site