linux-kernel - Re: [PATCH 04/11] sched_ext: Fix processs_ddsp_deferred_locals() by unifying DTL

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: <20240901153548.GK70166@maniforge>
Date: Sun, 1 Sep 2024 10:35:48 -0500
From: David Vernet <void@...ifault.com>
To: Tejun Heo <tj@...nel.org>
Cc: kernel-team@...a.com, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 04/11] sched_ext: Fix processs_ddsp_deferred_locals() by
 unifying DTL_INVALID handling

On Sat, Aug 31, 2024 at 10:03:58PM -1000, Tejun Heo wrote:
> Hello,
> 
> On Sat, Aug 31, 2024 at 07:56:39PM -0500, David Vernet wrote:
> ...
> > Sorry, should have been more clear: the testcase dispatched all tasks to
> > the wrong CPU, which is why it's a kworker in the print output below. I
> > believe that ksoftiqrd hit the same path as well and just wasn't printed
> > in the output because it lost the race to scx_bpf_error(). Let me know
> > if you want the testcase to repro and I can send it, or send a separate
> > patch to add it to selftests.
> 
> Yeah, please share the repro.

See the attached patch. You can run the test as follows after rebuilding
selftests:

./runner -t ddsp_local_on_invalid

You may have to run the test a few times to see the repro. Worth noting
is that the repro doesn't seem to hit if we don't explicitly set the
fallback target to 0 in ddsp_local_on_invalid_enqueue().

Thanks,
David

View attachment "0001-scx-Add-test-validating-direct-dispatch-with-LOCAL_O.patch" of type "text/x-patch" (4491 bytes)

Download attachment "signature.asc" of type "application/pgp-signature" (229 bytes)