[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <352704e1-2ef7-4141-a6a9-896981d8afa1@rivosinc.com>
Date: Tue, 3 Jun 2025 09:32:15 +0200
From: Clément Léger <cleger@...osinc.com>
To: David Laight <david.laight.linux@...il.com>
Cc: linux-riscv@...ts.infradead.org, linux-kernel@...r.kernel.org,
Paul Walmsley <paul.walmsley@...ive.com>, Palmer Dabbelt
<palmer@...belt.com>, Albert Ou <aou@...s.berkeley.edu>,
Alexandre Ghiti <alex@...ti.fr>, "Maciej W . Rozycki" <macro@...am.me.uk>
Subject: Re: [PATCH v2 0/3] riscv: misaligned: fix misaligned accesses
handling in put/get_user()
On 02/06/2025 23:08, David Laight wrote:
> On Mon, 2 Jun 2025 21:39:13 +0200
> Clément Léger <cleger@...osinc.com> wrote:
>
> ...
>> The second solution was the one chosen as there are too many callsites to
>> put/get_user() that could potentially do misaligned accesses. We tried
>> two approaches for that, either split access in two aligned accesses
>> (and do RMW for put_user())
>
> You can't do RMW because it is visible to other threads.
Ahem yeah indeed, so the only way was to do individual lb at least up to
some alignment.
Thanks,
Clément
>
> David
Powered by blists - more mailing lists