[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250708074124.GE1613376@noisy.programming.kicks-ass.net>
Date: Tue, 8 Jul 2025 09:41:24 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Sebastian Andrzej Siewior <bigeasy@...utronix.de>
Cc: linux-kernel@...r.kernel.org,
André Almeida <andrealmeid@...lia.com>,
Darren Hart <dvhart@...radead.org>,
Davidlohr Bueso <dave@...olabs.net>, Ingo Molnar <mingo@...hat.com>,
Juri Lelli <juri.lelli@...hat.com>,
Thomas Gleixner <tglx@...utronix.de>,
Valentin Schneider <vschneid@...hat.com>,
Waiman Long <longman@...hat.com>
Subject: Re: [PATCH 1/3] selftests/futex: Adapt the private hash test to RCU
related changes
On Mon, Jul 07, 2025 at 04:36:21PM +0200, Sebastian Andrzej Siewior wrote:
> The auto scaling on create creation used to automatically assign the new
> hash because there was the private hash was unused and could be replaced
> right away.
>
> With the upcoming change to wait for a RCU grace period before the hash
> can be assigned, the test will always fail.
>
> If the reported number of hash buckets is not updated after an
> auto scaling event, block on an acquired lock with a timeout. The timeout
> is the delay to wait towards a grace period and locking and a locked
> pthread_mutex_t ensure that glibc calls into kernel using futex
> operation which will assign new private hash if available.
> This will retry every 100ms up to 2 seconds in total.
So the auto scaling thing is 'broken' in that if you do a 'final'
pthread_create() it will try and stage this new hash. If for whatever
reason the refcount isn't '0' -- and this can already happen today due
to a concurrent futex operation. Nothing will re-try the rehash.
This RCU business made this all but a certainty, but the race was there.
I briefly wondered if we should have a workqueue re-try the rehash.
Powered by blists - more mailing lists