[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAKEwX=NQmNpFNf=8OLSWVp-JsRPgo90n4DR68k0Y6+nFUfXw4Q@mail.gmail.com>
Date: Thu, 12 Sep 2024 16:04:07 -0700
From: Nhat Pham <nphamcs@...il.com>
To: Yosry Ahmed <yosryahmed@...gle.com>
Cc: Jesper Dangaard Brouer <hawk@...nel.org>, tj@...nel.org, cgroups@...r.kernel.org,
shakeel.butt@...ux.dev, hannes@...xchg.org, lizefan.x@...edance.com,
longman@...hat.com, kernel-team@...udflare.com, linux-mm@...ck.org,
linux-kernel@...r.kernel.org, mfleming@...udflare.com,
joshua.hahnjy@...il.com
Subject: Re: [PATCH V10] cgroup/rstat: Avoid flushing if there is an ongoing
root flush
On Thu, Sep 12, 2024 at 11:51 AM Yosry Ahmed <yosryahmed@...gle.com> wrote:
>
> Do you mean per-lruvec or per-memcg?
Ah yeah ignore that part. I was originally thinking of adding a
per-lruvec zswap atomic counter just to fix zswap shrinker, but now
we'll need 2 atomic counters per-memcg to replace MEMCG_ZSWAP_B and
MEMCG_ZSWAPPED. The zswap_shrinker_count() will also use these two new
counters too. My apologies for the confusions...
Powered by blists - more mailing lists