Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Tue, 19 Apr 2022 14:46:43 -0400
From:   Johannes Weiner <hannes@cmpxchg.org>
To:     Nico Pache <npache@redhat.com>
Cc:     linux-mm@kvack.org, akpm@linux-foundation.org,
        linux-kernel@vger.kernel.org, aquini@redhat.com,
        shakeelb@google.com, llong@redhat.com, mhocko@suse.com,
        hakavlad@inbox.lv
Subject: Re: [PATCH v3] vm_swappiness=0 should still try to avoid swapping
 anon memory
Message-ID: <Yl8Dk+a1YDwupLMv@cmpxchg.org>
References: <20210809223740.59009-1-npache@redhat.com>
 <b5575449-08a2-2901-e7e1-a97150e4878c@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <b5575449-08a2-2901-e7e1-a97150e4878c@redhat.com>
Precedence: bulk

Hi Nico,

On Tue, Apr 19, 2022 at 02:11:53PM -0400, Nico Pache wrote:
> I think its is important to note the issue we are seeing has greatly improved
> since the initial posting. However we have noticed that the issue is still
> present (and significantly worse) when cgroupV1 is set.
> 
> We were initially testing with CgroupV1 and later found that the issue was not
> as bad in CgroupV2 (but was still an noticeable issue). This is also resulting
> in the splitting of THPs in the host kernel.

When swappiness is 0, cgroup limit reclaim has a fixed SCAN_FILE
branch, so it shouldn't ever look at anon. I'm assuming you're getting
global reclaim mixed in. Indeed, I think we can try harder not to swap
for global reclaim if the user asks for that.

Can you try the below patch?

diff --git a/mm/vmscan.c b/mm/vmscan.c
index ba19edbc2452..0aa929151386 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -2729,11 +2729,9 @@ static void get_scan_count(struct lruvec *lruvec, struct scan_control *sc,
 	}
 
 	/*
-	 * Global reclaim will swap to prevent OOM even with no
-	 * swappiness, but memcg users want to use this knob to
-	 * disable swapping for individual groups completely when
-	 * using the memory controller's swap limit feature would be
-	 * too expensive.
+	 * For cgroups, swappiness=0 means never swap. Historically,
+	 * cgroup users have relied on this as it was cheaper than
+	 * setting a swap limit of 0.
 	 */
 	if (cgroup_reclaim(sc) && !swappiness) {
 		scan_balance = SCAN_FILE;
@@ -2754,7 +2752,7 @@ static void get_scan_count(struct lruvec *lruvec, struct scan_control *sc,
 	 * If the system is almost out of file pages, force-scan anon.
 	 */
 	if (sc->file_is_tiny) {
-		scan_balance = SCAN_ANON;
+		scan_balance = SCAN_EQUAL;
 		goto out;
 	}
 
@@ -2767,6 +2765,12 @@ static void get_scan_count(struct lruvec *lruvec, struct scan_control *sc,
 		goto out;
 	}
 
+	/* swappiness=0 means no swap until OOM is imminent */
+	if (!swappiness && sc->priority) {
+		scan_balance = SCAN_FILE;
+		goto out;
+	}
+
 	scan_balance = SCAN_FRACT;
 	/*
 	 * Calculate the pressure balance between anon and file pages.