Re: Random panics seen in 2.6.18-rc1

[Andrew Morton <akpm@osdl.org>]

On Thu, 13 Jul 2006 09:26:35 +0200
Ingo Molnar <mingo@elte.hu> wrote:

> > 
> > But what was that change _for_?  Presumably, to plug some lockdep 
> > problem. Which now will come back.
> 
> correct - but i first wanted to get the fix out, before trying to fix 
> the lockdep thing.

More yup.

> > And the additional arg to __cache_free() was rather a step backwards - 
> > this is fastpath.  With a bit more effort that could have been avoided 
> > (please).
> 
> yeah, i'll fix this. Any suggestions of how to avoid the parameter 
> passing? (without ugly #ifdeffery)

No, I don't see a way apart from inlining __cache_free(), or inlining
cache_free_alien() into both kfree() and kmem_cache_free(), both of which
are unattractive.

Well.  One could do

	local_irq_disable();
	cachep->array[smp_processor_id()]->lockdep_nested = 1;
	__cache_free(...)
	cachep->array[smp_processor_id()]->lockdep_nested = 0;
	local_irq_enable();

then do lots of ifdeffery to make that go away if !LOCKDEP.  But sheesh.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/