Re: [RFC 03/13] mm, page_alloc: don't retry initial attempt in slowpath

[Vlastimil Babka]

On 05/31/2016 08:25 AM, Joonsoo Kim wrote:
> On Tue, May 10, 2016 at 09:35:53AM +0200, Vlastimil Babka wrote:
> > After __alloc_pages_slowpath() sets up new alloc_flags and wakes up kswapd, it
> > first tries get_page_from_freelist() with the new alloc_flags, as it may
> > succeed e.g. due to using min watermark instead of low watermark. This attempt
> > does not have to be retried on each loop, since direct reclaim, direct
> > compaction and oom call get_page_from_freelist() themselves.
>
> Hmm... there is a corner case. If did_some_progress is 0 or compaction
> is deferred, get_page_from_freelist() isn't called. But, we can
> succeed to allocate memory since there is a kswapd that reclaims
> memory in background.

Hmm good point. I think the cleanest solution is to let 
__alloc_pages_direct_reclaim attempt regardless of did_some_progress.

> Thanks.