lkml.org 
[lkml]   [2009]   [Feb]   [11]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 0/8] git pull request for tip/tracing/core

* Steven Rostedt <rostedt@goodmis.org> wrote:

>
> On Wed, 11 Feb 2009, Ingo Molnar wrote:
>
> >
> > > > Steven Rostedt (7):
> > > > ring-buffer: add NMI protection for spinlocks
> > > > nmi: add generic nmi tracking state
> >
> > hm, this commit breaks a number of non-x86 architectures the following way:
> >
> > In file included from /home/mingo/tip/include/linux/interrupt.h:12,
> > from /home/mingo/tip/arch/ia64/include/asm/mca.h:16,
> > from /home/mingo/tip/arch/ia64/kernel/asm-offsets.c:17:
> > /home/mingo/tip/include/linux/hardirq.h:67:2: #error PREEMPT_ACTIVE too high!
> > make[2]: *** [arch/ia64/kernel/asm-offsets.s] Error 1
> >
> > Alpha and ARM are broken too. (and possibly others)
>
> I'm glad we had that check.
>
> OK, revert it and I'll take a look to see what other options we have.

Reverting it is neither simple nor desired: 1) it's now embedded in other
tracing commits 2) my test-boxes stopped doing those rare lockups with the
NMI fix in place ...

Any better approach to fix those platforms? I reported this ftrace NMI problem
to you about a year ago ...

Ingo


\
 
 \ /
  Last update: 2009-02-11 17:29    [from the cache]
©2003-2011 Jasper Spaans