Re: spontaneous reboots and hangs on x86_64

Re: spontaneous reboots and hangs on x86_64

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1154 bytes --]

Hi Ingo,

On Fri, 16 May 2008 20:53:36 +0200 Mariusz Kozlowski <m.kozlowski@tuxland.pl> wrote:
>
> Ok I got bisection working finally, now I'm at d5be23e2 I have something like 80 revisions left and
> I'm stuck with this:
> 
> In file included from kernel/sched.c:1861:
> kernel/sched_rt.c: In function 'check_preempt_curr_rt':
> kernel/sched_rt.c:653: error: implicit declaration of function 'cpupri_find'
> kernel/sched_rt.c:653: error: 'struct root_domain' has no member named 'cpupri'
> make[1]: *** [kernel/sched.o] Error 1
> make: *** [kernel] Error 2

This error is caused by commit 677ceea32cf1d8ba9a4870bc58079cd7e7f77558
("sched: prioritize non-migratable tasks over migratable ones") in the
sched tree (sched/for-akpm).  This commit is either just wrong, is was
misrebased, or is just badly out of order.  It introduces a use of
cpupri_find, but the function is not defined until commit
ea9b3c95b7c666d4658a139b88587269e5baa647 ("sched: use a 2-d bitmap for
searching lowest-pri CPU") about 146 commits later ...

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: spontaneous reboots and hangs on x86_64

[Mariusz Kozlowski]

Hi,

> Hi Ingo,
> 
> On Fri, 16 May 2008 20:53:36 +0200 Mariusz Kozlowski <m.kozlowski@tuxland.pl> wrote:
> >
> > Ok I got bisection working finally, now I'm at d5be23e2 I have something like 80 revisions left and
> > I'm stuck with this:
> > 
> > In file included from kernel/sched.c:1861:
> > kernel/sched_rt.c: In function 'check_preempt_curr_rt':
> > kernel/sched_rt.c:653: error: implicit declaration of function 'cpupri_find'
> > kernel/sched_rt.c:653: error: 'struct root_domain' has no member named 'cpupri'
> > make[1]: *** [kernel/sched.o] Error 1
> > make: *** [kernel] Error 2
> 
> This error is caused by commit 677ceea32cf1d8ba9a4870bc58079cd7e7f77558
> ("sched: prioritize non-migratable tasks over migratable ones") in the
> sched tree (sched/for-akpm).  This commit is either just wrong, is was
> misrebased, or is just badly out of order.  It introduces a use of
> cpupri_find, but the function is not defined until commit
> ea9b3c95b7c666d4658a139b88587269e5baa647 ("sched: use a 2-d bitmap for
> searching lowest-pri CPU") about 146 commits later ...

Whatever it was its gone in recent linux-next releases, box boots without
problems.

	Mariusz