From: Henning Schild <henning.schild@siemens.com>
To: Jan Kiszka <jan.kiszka@siemens.com>
Cc: isar-users <isar-users@googlegroups.com>
Subject: Re: [PATCH] linux-custom: skip linux-libc-dev deployment on downgrade
Date: Wed, 23 Jan 2019 19:13:41 +0100 [thread overview]
Message-ID: <20190123191341.683395ab@md1za8fc.ad001.siemens.net> (raw)
In-Reply-To: <20190123190323.5e346a25@md1za8fc.ad001.siemens.net>
Am Wed, 23 Jan 2019 19:03:23 +0100
schrieb "[ext] Henning Schild" <henning.schild@siemens.com>:
> Am Wed, 23 Jan 2019 18:43:08 +0100
> schrieb Jan Kiszka <jan.kiszka@siemens.com>:
>
> > On 23.01.19 18:30, Henning Schild wrote:
> > > Am Wed, 23 Jan 2019 18:26:08 +0100
> > > schrieb Jan Kiszka <jan.kiszka@siemens.com>:
> > >
> > >> On 23.01.19 18:23, [ext] Henning Schild wrote:
> > >>> Ping
> > >>>
> > >>
> > >> Looks good to me - what scenarios did you test?
> > >
> > > I know the breakages of buildchroot from layers, and the not
> > > deploying the kernel downgrade helped fix the issues.
> > > In this repo i honestly only tested the logic and whether the
> > > warning will pop up. Tested a qemuamd64 with the cip kernel inside
> > > Isar.
> >
> > OK, but we also have a "should upgrade" kernel recipe in Isar:
> > linux-mainline_4.19.0.
>
> I just tested the "gt" operator of the comparator against the 4.9 that
> we get in debian 9. It worked there so there is no reason to think it
> will not work for 4.19.
>
> >
> > One question on second thought:
> >
> > > +
> > > +# linux-libc-dev causes dependency problems if we downgrade
> > > +# remove it after the build so the downgraded version does not
> > > get deployed +LINUX_LIBC_DEV_V=`dpkg-query --show --showformat
> > > '${Version}' linux-libc-dev` +if dpkg --compare-versions
> > > $LINUX_LIBC_DEV_V gt $PV; then
> > > + rm -f linux-libc-dev_${PV}*.deb
> > > +fi
> >
> > Isn't this assuming we have linux-libc-dev from Debian installed? Is
> > that always valid? Wouldn't it be better to query the Debian repo
> > specifically?
>
> Assuming you are installing only one kernel in your image, the command
> will operate on linux-libc-dev from all repos but ours ... debian.
> I guess if we are looking at images with multiple kernels, or
> incremental builds that downgrade against their own upgrade ... things
> could get tricky. Or actually not, in this case we want to get
> notified about the downgrade in our repo.
>
> Say your debian official kernel is 4.9, your layer adds 4.14.42 and
> you go back to 4.14.40 in an incremental build step. The upgrade from
> 4.9 to 4.14.42 will be done as usual, and you should now receive the
> warning about a downgrade that will not be performed. ... Not
> tested ...
In fact that might not be true. reprepro will probably drop the 4.14.42
so it will not be seen at install time of 4.14.40. But still we will
always detect if we fall below the greatest version coming from a repo
we do not manage ... the debian default or your other mirror.
Henning
> So i am guessing the logic is correct. For the multiple kernel case
> (does that even exist?) one might have to ignore the warning or
> enforce an install order with artificial dependencies.
>
> > Also: $(...) - `...` is old-style.
>
> Ok will fix.
>
> Henning
>
> > Jan
> >
>
next prev parent reply other threads:[~2019-01-23 18:13 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-17 13:03 Henning Schild
2019-01-23 17:23 ` Henning Schild
2019-01-23 17:26 ` Jan Kiszka
2019-01-23 17:30 ` Henning Schild
2019-01-23 17:43 ` Jan Kiszka
2019-01-23 18:03 ` Henning Schild
2019-01-23 18:13 ` Henning Schild [this message]
2019-01-28 11:51 ` [PATCHv2] " Henning Schild
2019-01-28 11:54 ` Jan Kiszka
2019-01-30 10:49 ` Maxim Yu. Osipov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190123191341.683395ab@md1za8fc.ad001.siemens.net \
--to=henning.schild@siemens.com \
--cc=isar-users@googlegroups.com \
--cc=jan.kiszka@siemens.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox