[<prev] [next>] [day] [month] [year] [list]
Message-ID: <CAAmzW4NxHW5boy6OiPcwzamwrJGXnEVNgn27ACtfDCoqQ3F24w@mail.gmail.com>
Date: Tue, 7 Apr 2020 09:42:54 +0900
From: Joonsoo Kim <js1304@...il.com>
To: Hillf Danton <hdanton@...a.com>
Cc: Andrew Morton <akpm@...ux-foundation.org>,
Linux Memory Management List <linux-mm@...ck.org>,
LKML <linux-kernel@...r.kernel.org>,
Johannes Weiner <hannes@...xchg.org>,
Michal Hocko <mhocko@...nel.org>,
Hugh Dickins <hughd@...gle.com>,
Minchan Kim <minchan@...nel.org>,
Vlastimil Babka <vbabka@...e.cz>,
Mel Gorman <mgorman@...hsingularity.net>
Subject: Re: [PATCH v5 05/10] mm/swap: charge the page when adding to the swap cache
2020년 4월 6일 (월) 오후 8:58, Hillf Danton <hdanton@...a.com>님이 작성:
>
>
> On Fri, 3 Apr 2020 14:40:43 +0900 Joonsoo Kim wrote:
> >
> > @@ -153,11 +165,16 @@ int add_to_swap_cache(struct page *page, swp_entry_t entry,
> > xas_unlock_irq(&xas);
> > } while (xas_nomem(&xas, gfp));
> >
> > - if (!xas_error(&xas))
> > + if (!xas_error(&xas)) {
> > + mem_cgroup_commit_charge(page, memcg, false, compound);
>
> Add a tp
> trace_mm_add_to_swap_cache(page);
Please let me know the reason of this comment. IMO, adding a tracepoint isn't
good idea until it is really necessary since it is considered as
kernel ABI and it
cannot be changed easily.
Thanks.
Powered by blists - more mailing lists