All of lore.kernel.org
 help / color / mirror / Atom feed
From: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
To: buildroot@busybox.net
Subject: [Buildroot] [PATCH v2] package/pkgconf: add patch to restore pre-1.5.3 behavior for sysroot prefixing
Date: Fri, 21 Dec 2018 16:39:15 +0100	[thread overview]
Message-ID: <20181221163915.53685772@windsurf> (raw)
In-Reply-To: <20181218214443.19888-1-thomas.petazzoni@bootlin.com>

Hello,

On Tue, 18 Dec 2018 22:44:43 +0100, Thomas Petazzoni wrote:
> Prior to the bump to version 1.5.3 in commit
> 4e423669399ad8389edd81761ea5c9cc26bf312d, we had a patch on pkgconf
> that ensures only some variables containing paths were prefixed by the
> sysroot directory when queried through pkg-config. This patch was
> dropped as part of the 1.5.3 bump, but it turns out we really need
> something like this, or a significant number of changes need to be
> done to existing packages.
> 
> Indeed, pkg-config has no notion of which variable/path gets used at
> build time vs. which variable/path gets used at runtime. Prefixing
> with the sysroot the paths used at build time works and is desirable,
> but prefixing the paths used at runtime doesn't work.
> 
> This commit should fix a large number of remaining build failures
> related to pkgconf 1.5.3, and should allow reverting a significant
> number of workarounds.
> 
> Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
> ---
> Changes since v1:
>  - Fix the pkgconf patch description, which was meaningless, as
>    noticed by Yann.
>  - Don't initialize the add_sysroot variable to true to later set its
>    value unconditionally to false, as noticed by Yann.
> ---
>  ...th-the-sysroot-a-subset-of-variables.patch | 142 ++++++++++++++++++
>  1 file changed, 142 insertions(+)
>  create mode 100644 package/pkgconf/0001-Only-prefix-with-the-sysroot-a-subset-of-variables.patch

Since this patch has been on the mailing list for a few days, and that
Arnout, Yann and Peter all gave their approval for going in this
direction to try to fix the pkg-config related issues, I've applied.
This will allow us to see if the autobuilders are happier.

If it is successful, my personal preference would be to revert all the
pkg-config 1.5.3 related fixes from Fabrice that we have applied so
far, since they should normally no longer be needed. Does everyone
agree with this proposal ?

However, I would probably wait for a few days before doing this.
Fabrice fixes will not cause any problem in combination with my
pkg-config change, so let's first see if my pkg-config change really
solves the remaining problems. If it does, then we can proceed with
reverting the fixes from Fabrice.

However, I will mark the remaining pkg-config related fixes from
Fabrice pending in patchwork as Superseded.

Best regards,


Thomas
-- 
Thomas Petazzoni, CTO, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com

      parent reply	other threads:[~2018-12-21 15:39 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-12-18 21:44 [Buildroot] [PATCH v2] package/pkgconf: add patch to restore pre-1.5.3 behavior for sysroot prefixing Thomas Petazzoni
2018-12-19 20:58 ` Yann E. MORIN
2018-12-21 15:39 ` Thomas Petazzoni [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=20181221163915.53685772@windsurf \
    --to=thomas.petazzoni@bootlin.com \
    --cc=buildroot@busybox.net \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.