stable.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Tejun Heo <tj@kernel.org>,
	Mel Gorman <mgorman@techsingularity.net>,
	Sasha Levin <sashal@kernel.org>
Subject: [PATCH 5.4 01/38] Revert "cgroup: Add memory barriers to plug cgroup_rstat_updated() race window"
Date: Fri,  5 Jun 2020 16:14:44 +0200	[thread overview]
Message-ID: <20200605140252.623231453@linuxfoundation.org> (raw)
In-Reply-To: <20200605140252.542768750@linuxfoundation.org>

From: Tejun Heo <tj@kernel.org>

[ Upstream commit d8ef4b38cb69d907f9b0e889c44d05fc0f890977 ]

This reverts commit 9a9e97b2f1f2 ("cgroup: Add memory barriers to plug
cgroup_rstat_updated() race window").

The commit was added in anticipation of memcg rstat conversion which needed
synchronous accounting for the event counters (e.g. oom kill count). However,
the conversion didn't get merged due to percpu memory overhead concern which
couldn't be addressed at the time.

Unfortunately, the patch's addition of smp_mb() to cgroup_rstat_updated()
meant that every scheduling event now had to go through an additional full
barrier and Mel Gorman noticed it as 1% regression in netperf UDP_STREAM test.

There's no need to have this barrier in tree now and even if we need
synchronous accounting in the future, the right thing to do is separating that
out to a separate function so that hot paths which don't care about
synchronous behavior don't have to pay the overhead of the full barrier. Let's
revert.

Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Mel Gorman <mgorman@techsingularity.net>
Link: http://lkml.kernel.org/r/20200409154413.GK3818@techsingularity.net
Cc: v4.18+
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 kernel/cgroup/rstat.c |   16 +++-------------
 1 file changed, 3 insertions(+), 13 deletions(-)

--- a/kernel/cgroup/rstat.c
+++ b/kernel/cgroup/rstat.c
@@ -33,12 +33,9 @@ void cgroup_rstat_updated(struct cgroup
 		return;
 
 	/*
-	 * Paired with the one in cgroup_rstat_cpu_pop_upated().  Either we
-	 * see NULL updated_next or they see our updated stat.
-	 */
-	smp_mb();
-
-	/*
+	 * Speculative already-on-list test. This may race leading to
+	 * temporary inaccuracies, which is fine.
+	 *
 	 * Because @parent's updated_children is terminated with @parent
 	 * instead of NULL, we can tell whether @cgrp is on the list by
 	 * testing the next pointer for NULL.
@@ -134,13 +131,6 @@ static struct cgroup *cgroup_rstat_cpu_p
 		*nextp = rstatc->updated_next;
 		rstatc->updated_next = NULL;
 
-		/*
-		 * Paired with the one in cgroup_rstat_cpu_updated().
-		 * Either they see NULL updated_next or we see their
-		 * updated stat.
-		 */
-		smp_mb();
-
 		return pos;
 	}
 



  reply	other threads:[~2020-06-05 14:18 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-06-05 14:14 [PATCH 5.4 00/38] 5.4.45-rc1 review Greg Kroah-Hartman
2020-06-05 14:14 ` Greg Kroah-Hartman [this message]
2020-06-05 14:14 ` [PATCH 5.4 02/38] mm: Fix mremap not considering huge pmd devmap Greg Kroah-Hartman
2020-06-05 14:14 ` [PATCH 5.4 03/38] HID: sony: Fix for broken buttons on DS3 USB dongles Greg Kroah-Hartman
2020-06-05 14:14 ` [PATCH 5.4 04/38] HID: multitouch: enable multi-input as a quirk for some devices Greg Kroah-Hartman
2020-06-05 14:14 ` [PATCH 5.4 05/38] HID: i2c-hid: add Schneider SCL142ALM to descriptor override Greg Kroah-Hartman
2020-06-05 14:14 ` [PATCH 5.4 06/38] p54usb: add AirVasT USB stick device-id Greg Kroah-Hartman
2020-06-05 14:14 ` [PATCH 5.4 07/38] mt76: mt76x02u: Add support for newer versions of the XBox One wifi adapter Greg Kroah-Hartman
2020-06-05 14:14 ` [PATCH 5.4 08/38] kernel/relay.c: handle alloc_percpu returning NULL in relay_open Greg Kroah-Hartman
2020-06-05 14:14 ` [PATCH 5.4 09/38] mmc: fix compilation of user API Greg Kroah-Hartman
2020-06-05 14:14 ` [PATCH 5.4 10/38] media: Revert "staging: imgu: Address a compiler warning on alignment" Greg Kroah-Hartman
2020-06-05 14:14 ` [PATCH 5.4 11/38] media: staging: ipu3-imgu: Move alignment attribute to field Greg Kroah-Hartman
2020-06-05 14:14 ` [PATCH 5.4 12/38] scsi: ufs: Release clock if DMA map fails Greg Kroah-Hartman
2020-06-05 14:14 ` [PATCH 5.4 13/38] net: dsa: mt7530: set CPU port to fallback mode Greg Kroah-Hartman
2020-06-05 14:14 ` [PATCH 5.4 14/38] airo: Fix read overflows sending packets Greg Kroah-Hartman
2020-06-05 14:14 ` [PATCH 5.4 15/38] drm/i915: fix port checks for MST support on gen >= 11 Greg Kroah-Hartman
2020-06-05 14:14 ` [PATCH 5.4 16/38] scsi: hisi_sas: Check sas_port before using it Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 17/38] powerpc/powernv: Avoid re-registration of imc debugfs directory Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 18/38] powerpc/xmon: Restrict when kernel is locked down Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 19/38] spi: dw: use "smp_mb()" to avoid sending spi data error Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 20/38] ASoC: intel - fix the card names Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 21/38] s390/ftrace: save traced function caller Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 22/38] RDMA/qedr: Fix qpids xarray api used Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 23/38] RDMA/qedr: Fix synchronization methods and memory leaks in qedr Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 24/38] ARC: Fix ICCM & DCCM runtime size checks Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 25/38] ARC: [plat-eznps]: Restrict to CONFIG_ISA_ARCOMPACT Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 26/38] evm: Fix RCU list related warnings Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 27/38] scsi: pm: Balance pm_only counter of request queue during system resume Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 28/38] i2c: altera: Fix race between xfer_msg and isr thread Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 29/38] io_uring: initialize ctx->sqo_wait earlier Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 30/38] x86/mmiotrace: Use cpumask_available() for cpumask_var_t variables Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 31/38] net: bmac: Fix read of MAC address from ROM Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 32/38] drm/edid: Add Oculus Rift S to non-desktop list Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 33/38] s390/mm: fix set_huge_pte_at() for empty ptes Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 34/38] null_blk: return error for invalid zone size Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 35/38] net/ethernet/freescale: rework quiesce/activate for ucc_geth Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 36/38] net: ethernet: stmmac: Enable interface clocks on probe for IPQ806x Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 37/38] selftests: mlxsw: qos_mc_aware: Specify arping timeout as an integer Greg Kroah-Hartman
2020-06-05 14:15 ` [PATCH 5.4 38/38] net: smsc911x: Fix runtime PM imbalance on error Greg Kroah-Hartman
2020-06-05 22:16 ` [PATCH 5.4 00/38] 5.4.45-rc1 review Shuah Khan
2020-06-06  6:22 ` Jon Hunter
2020-06-06 13:33 ` Guenter Roeck
2020-06-06 16:26 ` Naresh Kamboju

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200605140252.623231453@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mgorman@techsingularity.net \
    --cc=sashal@kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=tj@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).