[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZlQ/Ks3I2BYybykD@andrea>
Date: Mon, 27 May 2024 10:07:06 +0200
From: Andrea Parri <parri.andrea@...il.com>
To: Alan Stern <stern@...land.harvard.edu>
Cc: will@...nel.org, peterz@...radead.org, boqun.feng@...il.com,
npiggin@...il.com, dhowells@...hat.com, j.alglave@....ac.uk,
luc.maranget@...ia.fr, paulmck@...nel.org, akiyks@...il.com,
dlustig@...dia.com, joel@...lfernandes.org,
linux-kernel@...r.kernel.org, linux-arch@...r.kernel.org,
hernan.poncedeleon@...weicloud.com,
jonas.oberhauser@...weicloud.com
Subject: Re: [PATCH] tools/memory-model: Document herd7 (internal)
representation
> It turns out the problem lies in the way lock.cat tries to calculate the
> rf relation for RU events (a spin_is_locked() that returns False). The
> method it uses amounts to requiring that such events must read from the
> lock's initial value or an LU event (a spin_unlock()) in a different
> thread. This clearly is wrong, and glaringly so in this litmus test
> since there are no other threads!
>
> A patch to fix the problem and reorganize the code a bit for greater
> readability is below. I'd appreciate it if people could try it out on
> various locking litmus tests in our archives.
Thanks for the quick solution, Alan. The results from our archives look
good.
Andrea
Powered by blists - more mailing lists