[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <alpine.DEB.2.21.1807162123380.1693@nanos.tec.linutronix.de>
Date: Mon, 16 Jul 2018 21:23:57 +0200 (CEST)
From: Thomas Gleixner <tglx@...utronix.de>
To: Mark Rutland <mark.rutland@....com>
cc: mingo@...nel.org, mingo@...hat.com, andy.shevchenko@...il.com,
arnd@...db.de, aryabinin@...tuozzo.com, boqun.feng@...il.com,
catalin.marinas@....com, dvyukov@...gle.com, glider@...gle.com,
hpa@...or.com, linux-kernel@...r.kernel.org,
parri.andrea@...il.com, peter@...leysoftware.com,
peterz@...radead.org, will.deacon@....com,
linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCHv4 02/12] atomics/x86: reduce arch_cmpxchg64*()
instrumentation
On Mon, 16 Jul 2018, Mark Rutland wrote:
> Currently x86's arch_cmpxchg64() and arch_cmpxchg64_local() are
> instrumented twice, as they call into instrumented atomics rather than
> their arch_ equivalents.
>
> A call to cmpxchg64() results in:
>
> cmpxchg64()
> kasan_check_write()
> arch_cmpxchg64()
> cmpxchg()
> kasan_check_write()
> arch_cmpxchg()
>
> Let's fix this up and call the arch_ equivalents, resulting in:
>
> cmpxchg64()
> kasan_check_write()
> arch_cmpxchg64()
> arch_cmpxchg()
>
> Signed-off-by: Mark Rutland <mark.rutland@....com>
> Acked-by: Peter Zijlstra (Intel) <peterz@...radead.org>
> Acked-by: Will Deacon <will.deacon@....com>
Acked-by: Thomas Gleixner <tglx@...utronix.de>
Powered by blists - more mailing lists