[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <Z9BATkiH4LLpt1C6@localhost.localdomain>
Date: Tue, 11 Mar 2025 14:53:18 +0100
From: Frederic Weisbecker <frederic@...nel.org>
To: Thomas Gleixner <tglx@...utronix.de>
Cc: LKML <linux-kernel@...r.kernel.org>,
Anna-Maria Behnsen <anna-maria@...utronix.de>,
Benjamin Segall <bsegall@...gle.com>,
Eric Dumazet <edumazet@...gle.com>,
Andrey Vagin <avagin@...nvz.org>,
Pavel Tikhomirov <ptikhomirov@...tuozzo.com>,
Peter Zijlstra <peterz@...radead.org>,
Cyrill Gorcunov <gorcunov@...il.com>
Subject: Re: [patch V3 14/18] posix-timers: Avoid false cacheline sharing
Le Sat, Mar 08, 2025 at 05:48:42PM +0100, Thomas Gleixner a écrit :
> struct k_itimer has the hlist_node, which is used for lookup in the hash
> bucket, and the timer lock in the same cache line.
>
> That's obviously bad, if one CPU fiddles with a timer and the other is
> walking the hash bucket on which that timer is queued.
>
> Avoid this by restructuring struct k_itimer, so that the read mostly (only
> modified during setup and teardown) fields are in the first cache line and
> the lock and the rest of the fields which get written to are in cacheline
> 2-N.
>
> Reduces cacheline contention in a test case of 64 processes creating and
> accessing 20000 timers each by almost 30% according to perf.
>
> Signed-off-by: Thomas Gleixner <tglx@...utronix.de>
Impressive what a fields reshuffle and alignement can achieve!
Reviewed-by: Frederic Weisbecker <frederic@...nel.org>
Powered by blists - more mailing lists