Openembedded Core Discussions
 help / color / mirror / Atom feed
From: Markus Volk <f_l_k@t-online.de>
To: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Cc: openembedded-core@lists.openembedded.org
Subject: Re: [OE-core] [oe-ore][PATCHv4] libsecret: add pam support depending on DISTRO_FEATURES
Date: Mon, 02 Dec 2024 11:37:05 +0100	[thread overview]
Message-ID: <T54VNS.XA4ZGCLO7PS3@t-online.de> (raw)
In-Reply-To: <Z013Th944PYfoHa-@bootlin-mathieu>

[-- Attachment #1: Type: text/plain, Size: 2025 bytes --]

On Mon, Dec 2 2024 at 10:01:02 +01:00:00, Mathieu Dubois-Briand 
<mathieu.dubois-briand@bootlin.com> wrote:
> ERROR: Nothing PROVIDES 'libpam' (but 
> /srv/pokybuild/yocto-worker/check-layer/build/meta/recipes-devtools/pam-wrapper/pam-wrapper_1.1.7.bb 
> DEPENDS on or otherwise requires it)
> libpam was skipped: one of 'pam systemd' needs to be in 
> DISTRO_FEATURES
> ERROR: Required build target 'meta-world-pkgdata' has no buildable 
> providers.
> Missing or unbuildable dependency chain was: ['meta-world-pkgdata', 
> 'pam-wrapper', 'libpam']
> 
> <https://valkyrie.yoctoproject.org/#/builders/27/builds/566/steps/11/logs/stdio>
> 
> Can you have a look at this issue, please ?

Hi,
I've been testing a patch for a few days that should also fix this 
problem. In libpam the autotools buildsystem was removed. This patch 
switches to meson. The pam recipe was split into two parts and now 
compiles twice. First only the libs are packaged and then the libpam 
modules. This breaks a dependency loop and it is now possible to link 
against systemd and enable 'logind' support for the modules.

The patch is invasive, but my tests so far are promising. It compiles 
for x86-64/arm/arm64 and mutlilib and it seems to fix the build for 
libpam-native (and this missing DISTRO_FEATURES issue). And, although 
not yet compilable, it also reduces the dependency loops between 
systemd<->shadow

One file is still packed into libpam-modules. This should be moved to 
the appropriate module, but I'm not sure what the /run/sepermit 
directory is needed for. There is also a patch added that fixes a 
problem in meson when checking for keyring. But when I wanted to send 
it upstream, I saw that the problem was already solved and submitted 
nicer by Alexander Kanavin

The ptest part is also still missing. When xtests is enabled, the tests 
compile without errors, but by default they are neither built nor 
installed. So I wanted to ask first, what of this is still needed? I 
will therefore send the patch as RFC for review


[-- Attachment #2: Type: text/html, Size: 2343 bytes --]

  reply	other threads:[~2024-12-02 10:37 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-11-30  8:52 [oe-ore][PATCHv4] libsecret: add pam support depending on DISTRO_FEATURES Markus Volk
2024-11-30  9:01 ` Patchtest results for " patchtest
2024-12-02  9:01 ` [OE-core] " Mathieu Dubois-Briand
2024-12-02 10:37   ` Markus Volk [this message]
2024-12-02 10:42     ` Alexander Kanavin
2024-12-03 12:42 ` Ross Burton
2024-12-03 14:07   ` Markus Volk
2024-12-03 14:35     ` Ross Burton
     [not found] <180CB3010466226B.7301@lists.openembedded.org>
2024-11-30  8:54 ` Markus Volk

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=T54VNS.XA4ZGCLO7PS3@t-online.de \
    --to=f_l_k@t-online.de \
    --cc=mathieu.dubois-briand@bootlin.com \
    --cc=openembedded-core@lists.openembedded.org \
    /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