Messages in this thread | | | Date | Thu, 11 Dec 2008 08:32:28 +0100 | From | Jens Axboe <> | Subject | Re: [bug?] bio: add experimental support for inlining a number of bio_vecs inside the bio |
| |
On Wed, Dec 10 2008, Andrew Morton wrote: > On Wed, 10 Dec 2008 13:12:38 +0100 > Jens Axboe <jens.axboe@oracle.com> wrote: > > > > experimental?? > > > > Experimental in the sense that it's the first version, not the concept. > > It's held up well for me, but just yesterday I received the first bug > > report on a ppc platform and now you have this on ia64. > > > Well there's another bug report up-list. "next-20081209: pdflush: page > allocation failure (xfs)". > > The clearing of (__GFP_WAIT | __GFP_IO) will call the page allocator in the > weakest possible mode and will of course spew page allocations warnings > with great frequency. It should at least add in __GFP_NOWARN.
OK, that must be for the change titled
"bio: only mempool back the largest bio_vec slab cache"
and yes, that does look like it wants a bit more gfp loving, probably the same as mempool_alloc() would use anyway. I'll get that fixed.
-- Jens Axboe
| |