($INBOX_DIR/description missing)
 help / color / mirror / Atom feed
From: Steve Sakoman <steve@sakoman.com>
To: seungkyun.kim@lge.com
Cc: openembedded-core@lists.openembedded.org
Subject: Re: [OE-core] [kirkstone][PATCH] ipk/rootfs: run sanity test of multilib in parallel
Date: Mon, 13 May 2024 05:02:02 -0700	[thread overview]
Message-ID: <CAOSpxdY+Ym5=fq4QdOXa=Wyq9ueiYEov_X7CCDQOKhb027toGg@mail.gmail.com> (raw)
In-Reply-To: <20240513015242.1233919-1-seungkyun.kim@lge.com>

This patch would need to be submitted (and merged)  for master before
it can be considered for kirkstone.  And since it is a significant
change in behaviour, even then it might not be suitable for an LTS
branch.

Thanks,

Steve

On Sun, May 12, 2024 at 6:52 PM Seungkyun Kim via
lists.openembedded.org <seungkyun.kim=lge.com@lists.openembedded.org>
wrote:
>
> From: "seungkyun.kim" <seungkyun.kim@lge.com>
>
> For multilib type packages, there is an additional temporary
> installation before the actual installation. It makes almost doubles
> the do_rootfs time if having many multilib type packages.
> To avoid this overhead, run sanity test in parallel.
> Installing package groups through opkg takes much more time than
> copying directory.
>
> Signed-off-by: seungkyun.kim <seungkyun.kim@lge.com>
> ---
>  meta/lib/oe/package_manager/ipk/rootfs.py | 21 +++++++++++++++++----
>  1 file changed, 17 insertions(+), 4 deletions(-)
>
> diff --git a/meta/lib/oe/package_manager/ipk/rootfs.py b/meta/lib/oe/package_manager/ipk/rootfs.py
> index 10a831994e..88108bb312 100644
> --- a/meta/lib/oe/package_manager/ipk/rootfs.py
> +++ b/meta/lib/oe/package_manager/ipk/rootfs.py
> @@ -4,6 +4,7 @@
>
>  import re
>  import filecmp
> +import multiprocessing
>  import shutil
>  from oe.rootfs import Rootfs
>  from oe.manifest import Manifest
> @@ -196,10 +197,16 @@ class PkgRootfs(DpkgOpkgRootfs):
>
>      def _multilib_test_install(self, pkgs):
>          ml_temp = self.d.getVar("MULTILIB_TEMP_ROOTFS")
> +        rootfs_temp = os.path.join(ml_temp, "rootfs")
>          bb.utils.mkdirhier(ml_temp)
>
> -        dirs = [self.image_rootfs]
> +        bb.utils.remove(rootfs_temp, True)
> +        shutil.copytree(self.image_rootfs, rootfs_temp)
> +        dirs = [rootfs_temp]
> +        return multiprocessing.Process(target=self._multilib_test_pkg_install, \
> +                                       args=(pkgs, ml_temp, dirs, False))
>
> +    def _multilib_test_pkg_install(self, pkgs, ml_temp, dirs):
>          for variant in self.d.getVar("MULTILIB_VARIANTS").split():
>              ml_target_rootfs = os.path.join(ml_temp, variant)
>
> @@ -216,6 +223,8 @@ class PkgRootfs(DpkgOpkgRootfs):
>              dirs.append(ml_target_rootfs)
>
>          self._multilib_sanity_test(dirs)
> +        rootfs_temp = os.path.join(ml_temp, "rootfs")
> +        bb.utils.remove(rootfs_temp)
>
>      '''
>      While ipk incremental image generation is enabled, it will remove the
> @@ -298,14 +307,18 @@ class PkgRootfs(DpkgOpkgRootfs):
>
>          for pkg_type in self.install_order:
>              if pkg_type in pkgs_to_install:
> +                sanity_test = None
>                  # For multilib, we perform a sanity test before final install
>                  # If sanity test fails, it will automatically do a bb.fatal()
>                  # and the installation will stop
>                  if pkg_type == Manifest.PKG_TYPE_MULTILIB:
> -                    self._multilib_test_install(pkgs_to_install[pkg_type])
> +                    sanity_test= self._multilib_test_install(pkgs_to_install[pkg_type])
> +                    sanity_test.start()
>
> -                self.pm.install(pkgs_to_install[pkg_type],
> -                                [False, True][pkg_type == Manifest.PKG_TYPE_ATTEMPT_ONLY])
> +                self.pm.install(pkgs_to_install[pkg_type], [False, True][pkg_type == Manifest.PKG_TYPE_ATTEMPT_ONLY])
> +
> +                if sanity_test is not None:
> +                    sanity_test.join()
>
>          if self.progress_reporter:
>              self.progress_reporter.next_stage()
> --
> 2.43.0
>
>
> -=-=-=-=-=-=-=-=-=-=-=-
> Links: You receive all messages sent to this group.
> View/Reply Online (#199222): https://lists.openembedded.org/g/openembedded-core/message/199222
> Mute This Topic: https://lists.openembedded.org/mt/106065500/3620601
> Group Owner: openembedded-core+owner@lists.openembedded.org
> Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [steve@sakoman.com]
> -=-=-=-=-=-=-=-=-=-=-=-
>


      reply	other threads:[~2024-05-13 12:02 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-05-13  1:52 [kirkstone][PATCH] ipk/rootfs: run sanity test of multilib in parallel seungkyun.kim
2024-05-13 12:02 ` Steve Sakoman [this message]

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='CAOSpxdY+Ym5=fq4QdOXa=Wyq9ueiYEov_X7CCDQOKhb027toGg@mail.gmail.com' \
    --to=steve@sakoman.com \
    --cc=openembedded-core@lists.openembedded.org \
    --cc=seungkyun.kim@lge.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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).