Re: [RFC][PATCH] sched,mips,ia64: Remove __ARCH_WANT_UNLOCKED_CTXSW

[Kirill Tkhai <tkhai@yandex.ru>]

23.09.2014, 19:06, "Peter Zijlstra" <peterz@infradead.org>:
> On Mon, Sep 22, 2014 at 10:36:18PM +0400, Kirill Tkhai wrote:
>> šFrom: Kirill Tkhai <ktkhai@parallels.com>
>>
>> šArchitectures, which define __ARCH_WANT_UNLOCKED_CTXSW,
>> šmay pull a task when it's in the middle of schedule().
>>
>> šCPU1(task1 calls schedule) šššššššššššCPU2
>> š... ššššššššššššššššššššššššššššššššššschedule()
>> š... šššššššššššššššššššššššššššššššššššššidle_balance()
>> š... ššššššššššššššššššššššššššššššššššššššššload_balance()
>> š... ššššššššššššššššššššššššššššššššššššššššššš...
>> šschedule() šššššššššššššššššššššššššššššššššššš...
>> ššššprepare_lock_switch() šššššššššššššššššššššš...
>> šššššššraw_spin_unlock(&rq1->lock) ššššššššššššš...
>> ššššššš... šššššššššššššššššššššššššššššššššššššraw_spin_lock(&rq1->lock)
>> ššššššš... ššššššššššššššššššššššššššššššššššššššššdetach_tasks();
>> ššššššš... šššššššššššššššššššššššššššššššššššššššššššcan_migrate_task(task1)
>> ššššššš... ššššššššššššššššššššššššššššššššššššššššattach_tasks(); <--- move task1 to rq2
>> ššššššš... šššššššššššššššššššššššššššššššššššššraw_spin_unlock(&rq1->lock)
>> ššššššš... šššššššššššššššššššššššššššššššcontext_switch() <--- switch to task1's stack
>> ššššššš... ššššššššššššššššššššššššššššššš...
>> šššš(using task1's stack) šššššššššššššššš(using task1's stack)
>> šššš... šššššššššššššššššššššššššššššššššš...
>> ššššcontext_switch() ššššššššššššššššššššš...
>>
>> šParallel use of a single stack is not a good idea.
>
> Indeed it is, but how about we do this instead?

Completely agree, looks good for me.

> ---
> Subject: sched,mips,ia64: Remove __ARCH_WANT_UNLOCKED_CTXSW
>
> Kirill found that there's a subtle race in the
> __ARCH_WANT_UNLOCKED_CTXSW code, and instead of fixing it, remove the
> entire exception because neither arch that uses it seems to actually
> still require it.
>
> Boot tested on mips64el (qemu) only.
>
> Cc: Oleg Nesterov <oleg@redhat.com>
> Cc: Guenter Roeck <linux@roeck-us.net>
> Cc: Ralf Baechle <ralf@linux-mips.org>
> Cc: Tony Luck <tony.luck@intel.com>
> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>

Reviewed-by: Kirill Tkhai <ktkhai@parallels.com> (one more review of ia64 part)