[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAHk-=wiz739MaJQdHdK85WbBChKjt7PDavW15saX-NkWYEo0Lg@mail.gmail.com>
Date: Thu, 27 Jun 2024 10:23:27 -0700
From: Linus Torvalds <torvalds@...ux-foundation.org>
To: Mateusz Guzik <mjguzik@...il.com>
Cc: kernel test robot <oliver.sang@...el.com>, oe-lkp@...ts.linux.dev, lkp@...el.com,
Linux Memory Management List <linux-mm@...ck.org>, Christian Brauner <brauner@...nel.org>, linux-kernel@...r.kernel.org,
ying.huang@...el.com, feng.tang@...el.com, fengwei.yin@...el.com
Subject: Re: [linux-next:master] [lockref] d042dae6ad: unixbench.throughput
-33.7% regression
On Thu, 27 Jun 2024 at 09:57, Linus Torvalds
<torvalds@...ux-foundation.org> wrote:
>
> .dput.terminate_walk.path_openat.do_filp_open
Hmm. This may be due to the benchmark being in /proc/net, and
retain_dentry() returning false.
Then the fastpath in fast_dput() won't trigger, and we'll always end
up doing the dentry lock if we're the last user (which presumably we
will be, there's probably not anybody else there?)
Linus
Powered by blists - more mailing lists