2009-04-15 05:10:36

by Nikanth Karthikesan

Subject: [PATCH 0/7] Cleanup code that think bio_alloc with __GFP_WAIT can fail

bio_alloc will not fail when __GFP_WAIT is specified in the gfp_flags. See
http://lkml.org/lkml/2009/4/14/198 for the discussion. But this not documented
and there are places where there is unwanted code to handle bio_alloc failure
even when __GFP_WAIT is specified. This patch set documents this and cleans up
various callers who assume bio_alloc can fail even with __GFP_WAIT.

Thanks
Nikanth

2009-04-15 06:55:31

by Jens Axboe

Subject: Re: [PATCH 0/7] Cleanup code that think bio_alloc with __GFP_WAIT can fail

On Wed, Apr 15 2009, Nikanth Karthikesan wrote:
> bio_alloc will not fail when __GFP_WAIT is specified in the gfp_flags.
> See http://lkml.org/lkml/2009/4/14/198 for the discussion. But this
> not documented and there are places where there is unwanted code to
> handle bio_alloc failure even when __GFP_WAIT is specified. This patch
> set documents this and cleans up various callers who assume bio_alloc
> can fail even with __GFP_WAIT.

I added your patches 2-7, I substituted 1/7 with one I did myself that
uses proper kerneldoc annotation. Thanks!

--
Jens Axboe