[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <0f7f3317-7a96-44f2-a3e7-a49f75bcd6aa@tu-bs.de>
Date: Wed, 17 Dec 2025 20:05:23 +0100
From: Thomas Haas <t.haas@...bs.de>
To: Alan Stern <stern@...land.harvard.edu>, Andrea Parri
<parri.andrea@...il.com>, Will Deacon <will@...nel.org>, Peter Zijlstra
<peterz@...radead.org>, Boqun Feng <boqun.feng@...il.com>, Nicholas Piggin
<npiggin@...il.com>, David Howells <dhowells@...hat.com>, Jade Alglave
<j.alglave@....ac.uk>, Luc Maranget <luc.maranget@...ia.fr>, "Paul E.
McKenney" <paulmck@...nel.org>, Akira Yokosawa <akiyks@...il.com>, "Daniel
Lustig" <dlustig@...dia.com>, Joel Fernandes <joelagnelf@...dia.com>,
<linux-kernel@...r.kernel.org>, <linux-arch@...r.kernel.org>,
<lkmm@...ts.linux.dev>
CC: <hernan.poncedeleon@...weicloud.com>, <jonas.oberhauser@...weicloud.com>,
"r.maseli@...bs.de" <r.maseli@...bs.de>
Subject: Re: [RFC] Potential problem in qspinlock due to mixed-size accesses
On 12.06.25 16:55, Thomas Haas wrote:
> We have been taking a look if mixed-size accesses (MSA) can affect the
> correctness of qspinlock.
> We are focusing on aarch64 which is the only memory model with MSA
> support [1].
> For this we extended the dartagnan [2] tool to support MSA and now it
> reports liveness, synchronization, and mutex issues.
--- cut off ---
ARM has recently fixed the issue on their side by strengthening the
memory model (see
https://github.com/herd/herdtools7/commit/2b7921a44a61766e23a1234767d28af696b436a0)
With the updated ARM-MSA model, Dartagnan does no longer find violations
in qspinlock. No patches needed :)
With best regards,
Thomas
Powered by blists - more mailing lists