linux-kernel - Re: [PATCH 4/4] [x86] perf: fix accidentally ack'ing a second event on intel perf counter

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <AANLkTin2KVPpkQWUa_GQ66q+wvZpfYHOKAfWbG-+K5FD@mail.gmail.com>
Date:	Thu, 2 Sep 2010 16:39:02 +0200
From:	Stephane Eranian <eranian@...gle.com>
To:	Don Zickus <dzickus@...hat.com>
Cc:	Robert Richter <robert.richter@....com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"mingo@...e.hu" <mingo@...e.hu>,
	Peter Zijlstra <peterz@...radead.org>
Subject: Re: [PATCH 4/4] [x86] perf: fix accidentally ack'ing a second event
 on intel perf counter

Don,

On Thu, Sep 2, 2010 at 4:19 PM, Don Zickus <dzickus@...hat.com> wrote:
> On Thu, Sep 02, 2010 at 10:13:19AM +0200, Stephane Eranian wrote:
>> Robert,
>>
>> Do you have the test program you used to test this?
>> I believe the NHM hack does not solve the problem, it
>> just makes it harder to appear.
>
> Could be.
>
>>
>> I suspect the real issue is that the GLOBAL_STATUS
>> bitmask cannot be trusted. I'd like to verify this.
>>
>> Has the problem appear only on Nehalem or also on
>> Westmere?
>
> I was able to duplicate on
>
> Intel(R) Core(TM) i5 CPU         650  @ 3.20GHz
> Intel(R) Xeon(R) CPU           X5560  @ 2.80GHz
>
I managed to reproduce on core i7 860 (without patch4).
Looking at the code again, I am dubious you ever execute
the retry goto. If the PMU is disabled and you've just
cleared the OVF_STAT, then I don't see where the new
overflows would come from. But that's a separate problem.

One thing I did is to compare status obtained via OVFL_STATUS
with one that I build manually by inspecting each individual
counter. The two returned bitmasks should always be identical
(with PEBS disabled).  When I got the spurious NMI, it did not
trip my status validation. So the OVFL_STATUS is valid.

I found something else that looked fishy. I am experimenting
with it. I will report back.


> with just running 'perf top' for about 60 seconds.
>
> You would need the first three patches to expose the problem.
>
> Reading the code, it seemed like the perf counters should be disabled and
> this patch should be unecessary, but after playing around with the code
> for a few hours, I came up with this patch to trap the issue.
>
> I read through the cpu errata and could not find anything related but I
> might have missed something.
>
> I am willing to help test if you have a more targeted patch.
>
Thanks.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/