Messages in this thread Patch in this message | | | Subject | Re: [BUG] "sched: Remove rq->lock from the first half of ttwu()" locks up on ARM | From | Peter Zijlstra <> | Date | Wed, 25 May 2011 19:08:55 +0200 |
| |
On Tue, 2011-05-24 at 23:32 +0200, Peter Zijlstra wrote: > On Tue, 2011-05-24 at 19:13 +0100, Marc Zyngier wrote: > > Peter, > > > > I've experienced all kind of lock-ups on ARM SMP platforms recently, and > > finally tracked it down to the following patch: > > > > e4a52bcb9a18142d79e231b6733cabdbf2e67c1f [sched: Remove rq->lock from the first half of ttwu()]. > > > > Even on moderate load, the machine locks up, often silently, and > > sometimes with a few messages like: > > INFO: rcu_preempt_state detected stalls on CPUs/tasks: { 0} (detected by 1, t=12002 jiffies) > > > > Another side effect of this patch is that the load average is always 0, > > whatever load I throw at the system. > > > > Reverting the sched changes up to that patch (included) gives me a > > working system again, which happily survives parallel kernel > > compilations without complaining. > > > > My knowledge of the scheduler being rather limited, I haven't been able > > to pinpoint the exact problem (though it probably have something to do > > with __ARCH_WANT_INTERRUPTS_ON_CTXSW being defined on ARM). The enclosed > > patch somehow papers over the load average problem, but the system ends > > up locking up anyway: > > Hurm.. I'll try and make x86 use __ARCH_WANT_INTERRUPTS_ON_CTXSW, IIRC > Ingo once said that that is possible and try to see if I can reproduce. > No clear ideas atm.
So I checked out that particular commit and build with the below patch on-top. grep __ARCH_WANT /proc/sched_debug did indeed return those strings so I'm assuming CPP did its magic and I'm indeed running a kernel that enables IRQs around context switches.
The sad news however is that a make -j8 (on a dual core) seems to result in a kernel image, not an oops.
Ooh, shiny, whilst typing this I got an NMI-watchdog error reporting me that CPU1 got stuck in try_to_wake_up(), so it looks like I can indeed reproduce some funnies.
/me goes dig in.
--- arch/x86/include/asm/system.h | 2 ++ kernel/sched_debug.c | 7 +++++++ 2 files changed, 9 insertions(+), 0 deletions(-)
diff --git a/arch/x86/include/asm/system.h b/arch/x86/include/asm/system.h index 12569e6..56103bb 100644 --- a/arch/x86/include/asm/system.h +++ b/arch/x86/include/asm/system.h @@ -10,6 +10,8 @@ #include <linux/kernel.h> #include <linux/irqflags.h> +#define __ARCH_WANT_INTERRUPTS_ON_CTXSW + /* entries in ARCH_DLINFO: */ #if defined(CONFIG_IA32_EMULATION) || !defined(CONFIG_X86_64) # define AT_VECTOR_SIZE_ARCH 2 diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c index 3669bec6..18b4ace 100644 --- a/kernel/sched_debug.c +++ b/kernel/sched_debug.c @@ -335,6 +335,13 @@ static int sched_debug_show(struct seq_file *m, void *v) (int)strcspn(init_utsname()->version, " "), init_utsname()->version); +#ifdef __ARCH_WANT_INTERRUPTS_ON_CTXSW + SEQ_printf(m, "__ARCH_WANT_INTERRUPTS_ON_CTXSW\n"); +#endif +#ifdef __ARCH_WANT_UNLOCKED_CTXSW + SEQ_printf(m, "__ARCH_WANT_UNLOCKED_CTXSW\n"); +#endif + #define P(x) \ SEQ_printf(m, "%-40s: %Ld\n", #x, (long long)(x)) #define PN(x) \
| |