Message-ID: <c2aee7e6b9096556aab9b47156e91082c9345a90.camel@gmx.de>
Subject: Re: [PATCH 1/1] mm: vmscan: Reduce throttling due to a failure to
 make progress
From:   Mike Galbraith <efault@gmx.de>
To:     Mel Gorman <mgorman@techsingularity.net>,
        Alexey Avramov <hakavlad@inbox.lv>
Cc:     Andrew Morton <akpm@linux-foundation.org>,
        Michal Hocko <mhocko@suse.com>,
        Vlastimil Babka <vbabka@suse.cz>,
        Rik van Riel <riel@surriel.com>,
        Darrick Wong <djwong@kernel.org>, regressions@lists.linux.dev,
        Linux-fsdevel <linux-fsdevel@vger.kernel.org>,
        Linux-MM <linux-mm@kvack.org>,
        LKML <linux-kernel@vger.kernel.org>
Date:   Tue, 30 Nov 2021 18:59:58 +0100
In-Reply-To: <20211130172754.GS3366@techsingularity.net>
References: <20211125151853.8540-1-mgorman@techsingularity.net>
         <20211127011246.7a8ac7b8@mail.inbox.lv>
         <20211129150117.GO3366@techsingularity.net>
         <20211201010348.31e99637@mail.inbox.lv>
         <20211130172754.GS3366@techsingularity.net>
Content-Type: text/plain; charset="UTF-8"
User-Agent: Evolution 3.42.1 
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Tue, 2021-11-30 at 17:27 +0000, Mel Gorman wrote:
>
> Obviously a fairly different experience and most likely due to the
> underlying storage.

I bet a virtual nickle this is the sore spot.

=2D--
 mm/vmscan.c |   10 ++++++++++
 1 file changed, 10 insertions(+)

=2D-- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -3360,7 +3360,17 @@ static void shrink_node(pg_data_t *pgdat
 	if (!current_is_kswapd() && current_may_throttle() &&
 	    !sc->hibernation_mode &&
 	    test_bit(LRUVEC_CONGESTED, &target_lruvec->flags))
+#if 0
 		reclaim_throttle(pgdat, VMSCAN_THROTTLE_WRITEBACK);
+#else
+		/*
+		 * wait_iff_congested() used to live here and was
+		 * _apparently_ a misspelled cond_resched()?????
+		 * In any case, we are most definitely NOT starting
+		 * any IO in reclaim_throttle(), so why bother?
+		 */
+		cond_resched();
+#endif

 	if (should_continue_reclaim(pgdat, sc->nr_reclaimed -
nr_reclaimed,
 				    sc))