Re: [RFC 3/4] mm, page_alloc: move cpuset seqcount checking to slowpath

[Vlastimil Babka]

On 01/18/2017 08:22 AM, Hillf Danton wrote:
> On Wednesday, January 18, 2017 6:16 AM Vlastimil Babka wrote:
> > This is a preparation for the following patch to make review simpler. While
> > the primary motivation is a bug fix, this could also save some cycles in the
> > fast path.
> This also gets kswapd involved.
> Dunno how frequent cpuset is changed in real life.

I don't think the extra kswapd wakeups due to retry_cpuset would be noticeable. 
Such frequent cpuset changes would likely have their own associated overhead 
larger than the wakeups.

> Hillf