[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <D5BTFE1AXSVK.2YNEV7RO2T5IT@verdict.gg>
Date: Sat, 02 Nov 2024 18:58:39 +0300
From: "Vladimir Vdovin" <deliran@...dict.gg>
To: "Paolo Abeni" <pabeni@...hat.com>, "Jakub Kicinski" <kuba@...nel.org>
Cc: <netdev@...r.kernel.org>, <dsahern@...nel.org>, <davem@...emloft.net>,
<idosch@...sch.org>, <edumazet@...gle.com>,
<linux-kselftest@...r.kernel.org>, <shuah@...nel.org>, <horms@...nel.org>
Subject: Re: [PATCH v5] net: ipv4: Cache pmtu for all packet paths if
multipath enabled
On Sat Nov 2, 2024 at 11:49 AM MSK, Paolo Abeni wrote:
> Hi,
>
> On 11/1/24 18:34, Vladimir Vdovin wrote:
> > On Fri Nov 1, 2024 at 4:45 PM MSK, Jakub Kicinski wrote:
> >> On Fri, 1 Nov 2024 10:48:57 +0000 Vladimir Vdovin wrote:
> >>> + pmtu_ipv4_mp_exceptions ipv4: PMTU multipath nh exceptions 0"
> >>
> >> This new test seems to fail in our CI:
> >>
> >> # TEST: ipv4: PMTU multipath nh exceptions [FAIL]
> >> # there are not enough cached exceptions
> >>
> >> https://netdev-3.bots.linux.dev/vmksft-net/results/840861/3-pmtu-sh/stdout
> >
> > Yes it failed in V4 patch, in this V5 its already ok:
> >
> > # TEST: ipv4: PMTU multipath nh exceptions [ OK ]
> > ok 1 selftests: net: pmtu.sh
> >
> > https://netdev-3.bots.linux.dev/vmksft-net-dbg/results/841042/2-pmtu-sh/stdout
> >
> > But in V5, there is failed test, not sure that this patch causes fail:
> > https://netdev-3.bots.linux.dev/vmksft-net-dbg/results/841042/31-busy-poll-test-sh/stdout
> >
> >>
> >> Also some process notes:
> >> - please don't post multiple versions of the patch a day:
> >> https://www.kernel.org/doc/html/next/process/maintainer-netdev.html#tl-dr
> >> - please avoid posting new versions in-reply-to the old one
> > Thanks, will keep it in mind next time, sorry for my ignorance
>
> Some additional notes:
>
> - please do answer to Ido's question: what about ipv6?
> - move the changelog after the SoB tag and a '---' separator, so that it
> will not be included into the git commit message
> - post new revisions of the patch in a different thread
>
> Thanks,
>
> Paolo
Thanks for your comments,
I will resend patch with fixed commit message as new thread.
Powered by blists - more mailing lists