lkml.org 
[lkml]   [2021]   [Feb]   [2]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: [RFC][PATCH 11/13] mm/vmscan: Consider anonymous pages without swap
On Tue, Feb 2, 2021 at 1:35 PM Dave Hansen <dave.hansen@intel.com> wrote:
>
> On 2/2/21 10:56 AM, Yang Shi wrote:
> >>
> >> /* If we have no swap space, do not bother scanning anon pages. */
> >> - if (!sc->may_swap || mem_cgroup_get_nr_swap_pages(memcg) <= 0) {
> >> + if (!sc->may_swap || !can_reclaim_anon_pages(memcg, pgdat->node_id)) {
> > Just one minor thing about may_swap. It may be cleared by
> > nr_boost_reclaim. But demotion should be fine for boost_reclaim.
>
> In other words, this if() is here is to avoid generating suboptimal I/O
> during boost_reclaim. But, since demotion doesn't generate any I/O, it
> *should* be fine for boost_reclaim.
>
> I agree with that in theory. Although, I'm tempted to put it in the
> TODO list as something to look at in the future. Do you think it's
> something that's immediately necessary?

No, I don't think so. We could tweak this later on.

\
 
 \ /
  Last update: 2021-02-02 23:37    [W:0.235 / U:0.208 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site