[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20161005204452.GB26721@remoulade>
Date: Wed, 5 Oct 2016 21:44:53 +0100
From: Mark Rutland <mark.rutland@....com>
To: Fredrik Markström <fredrik.markstrom@...il.com>
Cc: Robin Murphy <robin.murphy@....com>,
Kees Cook <keescook@...omium.org>,
Arnd Bergmann <arnd@...db.de>,
Ard Biesheuvel <ard.biesheuvel@...aro.org>,
Linus Walleij <linus.walleij@...aro.org>,
Nicolas Pitre <nico@...aro.org>,
Will Deacon <will.deacon@....com>,
Russell King <linux@...linux.org.uk>,
kristina.martsenko@....com, linux-kernel@...r.kernel.org,
Masahiro Yamada <yamada.masahiro@...ionext.com>,
Chris Brandt <chris.brandt@...esas.com>,
Michal Marek <mmarek@...e.com>,
Zhaoxiu Zeng <zhaoxiu.zeng@...il.com>,
linux-arm-kernel@...ts.infradead.org,
Jonathan Austin <jonathan.austin@....com>
Subject: Re: [PATCH v2] arm: Added support for getcpu() vDSO using TPIDRURW
On Wed, Oct 05, 2016 at 08:00:38PM +0000, Fredrik Markström wrote:
> On Wed, Oct 5, 2016 at 7:48 PM Robin Murphy <robin.murphy@....com> wrote:
> As far as I understand TPIDRURW isn't anything else then an architecture
> specific piece of tls since the last patch, possibly slightly faster then a
> "__thread u32 x;"
>
> The irony is that the two different ways it was handled earlier (not context
> switched or always set to zero on swap in) would have made it useful for this
> purpose.
The "not context switched" case was also arbitrarily corrupted, and could not
have been relied upon.
The zeroing case is similar to the restartable sequences design. So that's
probably worth looking into.
Thanks,
Mark.
Powered by blists - more mailing lists