linux-kernel - Re: [PATCH] perf: Fix sys_perf_event

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] [day] [month] [year] [list]

Message-ID: <7d830cc8-1463-d3ae-74cb-a908ac19a400@amd.com>
Date:   Thu, 17 Nov 2022 19:08:43 +0530
From:   Ravi Bangoria <ravi.bangoria@....com>
To:     Borislav Petkov <bp@...en8.de>
Cc:     Peter Zijlstra <peterz@...radead.org>, nslusarek@....net,
        Linus Torvalds <torvalds@...ux-foundation.org>, x86@...nel.org,
        linux-kernel@...r.kernel.org, Ravi Bangoria <ravi.bangoria@....com>
Subject: Re: [PATCH] perf: Fix sys_perf_event_open() race against self

On 17-Nov-22 5:29 PM, Borislav Petkov wrote:
> On Thu, Nov 17, 2022 at 04:29:31PM +0530, Ravi Bangoria wrote:
>> I do agree that more tests are always better. Though, this simple program
>> was to test a _specific race condition_ which no longer exists in kernel.
>> So even if we add it, what would it test?
> 
> It would make sure that race doesn't happen again. Or are you saying it
> will never ever happen so no need? Because we never ever rewrite the
> kernel?
> 
> Lemme save us some time: this test is dirt cheap. It is good to have so
> that multithreaded sys_perf_event_open() is exercised. And once it is
> there, someone else might have a look at it and improve it more or reuse
> it for another test.
> 
> And there are no downsides.
> 
> If you're still not convinced, lemme know and I'll turn it into a proper
> patch and submit it.

I'll do it :)

Thanks,
Ravi