Content-Type: multipart/mixed; boundary="===============0498936853148487056==" MIME-Version: 1.0 From: Al Viro To: lkp@lists.01.org Subject: Re: [PATCH] fsnotify: fix unlink performance regression Date: Sun, 05 May 2019 14:05:29 +0100 Message-ID: <20190505130528.GA23075@ZenIV.linux.org.uk> In-Reply-To: <20190505091549.1934-1-amir73il@gmail.com> List-Id: --===============0498936853148487056== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable On Sun, May 05, 2019 at 12:15:49PM +0300, Amir Goldstein wrote: > __fsnotify_parent() has an optimization in place to avoid unneeded > take_dentry_name_snapshot(). When fsnotify_nameremove() was changed > not to call __fsnotify_parent(), we left out the optimization. > Kernel test robot reported a 5% performance regression in concurrent > unlink() workload. > = > Reported-by: kernel test robot > Link: https://lore.kernel.org/lkml/20190505062153.GG29809(a)shao2-debian/ > Link: https://lore.kernel.org/linux-fsdevel/20190104090357.GD22409(a)quac= k2.suse.cz/ > Fixes: 5f02a8776384 ("fsnotify: annotate directory entry modification eve= nts") > Signed-off-by: Amir Goldstein > --- > = > Jan, > = > The linked 5.1-rc1 performance regression report came with bad timing. > Not sure if Linus is planning an rc8. If not, you will probably not > see this before the 5.1 release and we shall have to queue it for 5.2 > and backport to stable 5.1. > = > I crafted the patch so it applies cleanly both to master and Al's > for-next branch (there are some fsnotify changes in work.dcache). Charming... What about rename() and matching regressions there? --===============0498936853148487056==--