netdev - Re: python-eventlet test broken in 5.19 [was: Revert "tcp: change pingpong threshold to 3"]

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]

Date:   Thu, 18 Aug 2022 10:50:31 -0700
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     Jiri Slaby <jirislaby@...nel.org>
Cc:     Neal Cardwell <ncardwell@...gle.com>, Wei Wang <weiwan@...gle.com>,
        David Miller <davem@...emloft.net>,
        Eric Dumazet <edumazet@...gle.com>,
        Jakub Kicinski <kuba@...nel.org>, netdev@...r.kernel.org,
        Soheil Hassas Yeganeh <soheil@...gle.com>,
        Yuchung Cheng <ycheng@...gle.com>,
        LemmyHuang <hlm3280@....com>, stable <stable@...r.kernel.org>,
        temotor@...il.com, jakub@...siak.at
Subject: Re: python-eventlet test broken in 5.19 [was: Revert "tcp: change
 pingpong threshold to 3"]

On Tue, Aug 16, 2022 at 11:44 PM Jiri Slaby <jirislaby@...nel.org> wrote:
>
> Thanks a lot, Neal! So for the time being, until this is resolved in
> eventlet, I pushed a change to disable the test in openSUSE.

Yes, that is the right approach in this case, since it really does
seem to be purely about timing.

Note that "breaks user space" is not about random test suites. For all
we know, a test suite can exist literally to check for known bugs etc.

No user space breakage is about real loads by real people, ie "I used
to do this, and now that no longer works".

The test suites showing behavioral differences are certainly always a
cause of worry - because the test may exist for a very particular
reason, and may be indicative that yes, there are actual real programs
depending on this behavior - but isn't in itself necessarily a problem
in itself.

            Linus