[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20191114132848.55atqtjshjmi2udl@gondor.apana.org.au>
Date: Thu, 14 Nov 2019 21:28:48 +0800
From: Herbert Xu <herbert@...dor.apana.org.au>
To: Will Deacon <will@...nel.org>
Cc: linux@...linux.org.uk, torvalds@...ux-foundation.org,
catalin.marinas@....com, vincent.whitchurch@...s.com,
axboe@...nel.dk, linux-kernel@...r.kernel.org,
linux-arm-kernel@...ts.infradead.org, rabinv@...s.com,
Richard.Earnshaw@....com
Subject: Re: [PATCH v2] buffer: Fix I/O error due to ARM read-after-read
hazard
Will Deacon <will@...nel.org> wrote:
>
> which is what can happen due to this erratum. It's generally good practice
> to use READ_ONCE() when reading something which can be updated concurrently
> because:
>
> * It ensures that the value is (re-)loaded from memory
>
> * It prevents the compiler from performing harmful optimisations,
> such as merging or tearing (although in this case I suspect
> these are ok because we're dealing with a single bit)
>
> * On Alpha, it gives you a barrier so that dependency ordering
> can be relied upon from the load
The Alpha barrier matters for pointers, how could it make a
difference for individual bits?
Cheers,
--
Email: Herbert Xu <herbert@...dor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
Powered by blists - more mailing lists