Re: irq lock inversion

[Peter Zijlstra]

On Fri, 2009-11-06 at 12:03 -0500, Christoph Lameter wrote:
> Atomic allocs are allowed to fail.

All allocs (except __GFP_NOFAIL, but those are a deadlock waiting to
happen) are allowed to fail, !__GFP_WAIT allocs simply fail more easily.

The fact that __GFP_WAIT && order <= PAGE_ALLOC_COSTLY_ORDER allocations
currently don't fail but keep hammering the allocator could be
considered a bug (the recent OOM threads in fact suggest this is a
serious issue for some people).

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/