On Wed, May 20, 2015 at 11:58 AM, Jeff Moyer <jmoyer@xxxxxxxxxx> wrote: > Ming Lin <mlin@xxxxxxxxxx> writes: > >> On Mon, May 18, 2015 at 9:52 AM, Jeff Moyer <jmoyer@xxxxxxxxxx> wrote: >>> Ming Lin <mlin@xxxxxxxxxx> writes: >>> >>>> diff --git a/block/blk-core.c b/block/blk-core.c >>>> index fd154b9..909f317 100644 >>>> --- a/block/blk-core.c >>>> +++ b/block/blk-core.c >>>> @@ -617,6 +617,10 @@ struct request_queue *blk_alloc_queue_node(gfp_t gfp_mask, int node_id) >>>> if (q->id < 0) >>>> goto fail_q; >>>> >>>> + q->bio_split = bioset_create(4, 0); >>>> + if (!q->bio_split) >>>> + goto fail_id; >>>> + >>> >>> Arbitrary numbers should be documented. >> >> Kent, >> >> Is there specific reason to choose number 4? >> If no, I may change it to BIO_POOL_SIZE which is 2. > > Here's what he had to say last time around: > > On Mon, Nov 25, 2013 at 10:09:21PM -0800, Christoph Hellwig wrote: > > > + q->bio_split = bioset_create(4, 0); > > > + if (!q->bio_split) > > > + goto fail_id; > > > > How did we arrive at a mempool size of 4 to make sure we can always make > > progress with arbitrarily sized bios? Shouldn't we document the design > > decision somewhere? > > It just has to be nonzero to guarantee forward progress - the Then I'll update it to BIO_POOL_SIZE. > bio_alloc_bioset() rescuer thing I did awhile back guarantees that. > > link: https://lkml.org/lkml/2013/11/26/47 > > Cheers, > Jeff -- dm-devel mailing list dm-devel@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/dm-devel
