RE: [RFC][PATCH 2/6] perf, arch: Rework perf_event_index()

[Oleg Strikov]

> Actually, the registers for controlling interrupts aren't visible to userspace with ARMv7. On ARMv6, this is a different story, which is why I didn't mention doing this > for older platforms earlier :)

Yes. But the profiling tool enables the interrupts.
I mean that with the profiler running the bad application can break the system.

Oleg

-----Original Message-----
From: Will Deacon [mailto:will.deacon@xxxxxxx] 
Sent: Tuesday, November 22, 2011 4:14 PM
To: Oleg Strikov
Cc: Peter Zijlstra; mingo@xxxxxxx; William Cohen; linux-kernel@xxxxxxxxxxxxxxx; Michael Cree; Deng-Cheng Zhu; Anton Blanchard; Eric B Munson; Heiko Carstens; Paul Mundt; David S. Miller; Richard Kuo; Stephane Eranian; Arun Sharma; Vince Weaver
Subject: Re: [RFC][PATCH 2/6] perf, arch: Rework perf_event_index()

On Tue, Nov 22, 2011 at 12:00:33PM +0000, Oleg Strikov wrote:
> Also, please do not forget that enabling the userspace access to the 
> counters may not only create some problems to the profilers like Perf/OProfile but could also trash the kernel with the interrupts.

Actually, the registers for controlling interrupts aren't visible to userspace with ARMv7. On ARMv6, this is a different story, which is why I didn't mention doing this for older platforms earlier :)

Anyway, this thread has digressed now so you can always jump onto the ARM list (linux-arm-kernel) if you want to discuss this more.

Cheers,

Will
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/