linux-kernel - Re: [PATCH v2] selftest: txtimestamp: fix net ns entry logic

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: <CA+FuTSc0V4TGMF6KhZVkLfv1=OXXdV+J3ufKuXd0GhrHb2hjbw@mail.gmail.com>
Date:   Wed, 22 Jul 2020 08:49:47 -0400
From:   Willem de Bruijn <willemdebruijn.kernel@...il.com>
To:     Paolo Pisati <paolo.pisati@...onical.com>
Cc:     Willem de Bruijn <willemdebruijn.kernel@...il.com>,
        "David S . Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>,
        Shuah Khan <shuah@...nel.org>, Jian Yang <jianyang@...gle.com>,
        Network Development <netdev@...r.kernel.org>,
        "open list:KERNEL SELFTEST FRAMEWORK" 
        <linux-kselftest@...r.kernel.org>,
        linux-kernel <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2] selftest: txtimestamp: fix net ns entry logic

On Wed, Jul 22, 2020 at 4:37 AM Paolo Pisati <paolo.pisati@...onical.com> wrote:
>
> On Tue, Jul 21, 2020 at 6:26 PM Willem de Bruijn
> <willemdebruijn.kernel@...il.com> wrote:
> >
> > Fixes: cda261f421ba ("selftests: add txtimestamp kselftest")
> >
> > Acked-by: Willem de Bruijn <willemb@...gle.com>
>
> Besides, is it just me or this test fails frequently? I've been
> running it on 5.4.x, 5.7.x and 5.8-rcX and it often fails:
>
> ...
>     USR: 1595405084 s 947366 us (seq=0, len=0)
>     SND: 1595405084 s 948686 us (seq=9, len=10)  (USR +1319 us)
> ERROR: 6542 us expected between 6000 and 6500
>     ACK: 1595405084 s 953908 us (seq=9, len=10)  (USR +6541 us)
>     USR: 1595405084 s 997979 us (seq=0, len=0)
>     SND: 1595405084 s 999101 us (seq=19, len=10)  (USR +1121 us)
>     ACK: 1595405085 s 4438 us (seq=19, len=10)  (USR +6458 us)
>     USR: 1595405085 s 49317 us (seq=0, len=0)
>     SND: 1595405085 s 50680 us (seq=29, len=10)  (USR +1363 us)
> ERROR: 6661 us expected between 6000 and 6500
>     ACK: 1595405085 s 55978 us (seq=29, len=10)  (USR +6661 us)
>     USR: 1595405085 s 101049 us (seq=0, len=0)
>     SND: 1595405085 s 102342 us (seq=39, len=10)  (USR +1293 us)
> ERROR: 6578 us expected between 6000 and 6500
>     ACK: 1595405085 s 107627 us (seq=39, len=10)  (USR +6577 us)
>     USR-SND: count=4, avg=1274 us, min=1121 us, max=1363 us
>     USR-ACK: count=4, avg=6559 us, min=6458 us, max=6661 us
>
>
> In particular, "run_test_v4v6 ${args}       # tcp" is the most
> susceptible to failures (though i've seen the udp variant fail too).

Not for me. The interval bounds have been set as is based on previous
experience.

Are you running it inside a VM? Especially qemu without kvm
acceleration could increase jitter.

The reports are not far outside the bounds. They can be extended a bit
if that considerably reduces flakiness.