Patchwork [0/1] Fix build failure of openjade using newer host

login
register
mail settings
Submitter Otavio Salvador
Date June 2, 2012, 5:34 p.m.
Message ID <cover.1338658338.git.otavio@ossystems.com.br>
Download mbox
Permalink /patch/29165/
State New
Headers show

Pull-request

git://github.com/OSSystems/oe-core master

Comments

Otavio Salvador - June 2, 2012, 5:34 p.m.
This has been catch when building denzil on a Debian sid host; the fix
and build failure are documented on the patch. Please evaluate this
patch for inclusion on denzil 1.2.1 release as this will be catch by
users later.

The following changes since commit e3113827810e98eb1b012f0b280fb917199704c1:

  webkit-gtk: Use glib as unicode backend to avoid browser crash (2012-05-30 17:37:58 +0100)

are available in the git repository at:
  git://github.com/OSSystems/oe-core master
  https://github.com/OSSystems/oe-core/tree/HEAD

Otavio Salvador (1):
  openjade: autoreconf to fix building on newer host systems

 .../openjade/openjade-1.3.2/autoreconf.patch       |21657 ++++++++++++++++++++
 .../openjade/openjade-native_1.3.2.bb              |    3 +-
 2 files changed, 21659 insertions(+), 1 deletions(-)
 create mode 100644 meta/recipes-devtools/openjade/openjade-1.3.2/autoreconf.patch
Richard Purdie - June 8, 2012, 10:33 a.m.
On Sat, 2012-06-02 at 14:34 -0300, Otavio Salvador wrote:
> This has been catch when building denzil on a Debian sid host; the fix
> and build failure are documented on the patch. Please evaluate this
> patch for inclusion on denzil 1.2.1 release as this will be catch by
> users later.
> 
> The following changes since commit e3113827810e98eb1b012f0b280fb917199704c1:
> 
>   webkit-gtk: Use glib as unicode backend to avoid browser crash (2012-05-30 17:37:58 +0100)
> 
> are available in the git repository at:
>   git://github.com/OSSystems/oe-core master
>   https://github.com/OSSystems/oe-core/tree/HEAD
> 
> Otavio Salvador (1):
>   openjade: autoreconf to fix building on newer host systems

Why can't we run autoreconf instead of having the huge patch? This might
be the right answer but I'd like to understand the problem here a bit
more...

Cheers,

Richard
Otavio Salvador - June 8, 2012, 11:52 a.m.
On Fri, Jun 8, 2012 at 7:33 AM, Richard Purdie <
richard.purdie@linuxfoundation.org> wrote:

> Why can't we run autoreconf instead of having the huge patch? This might
> be the right answer but I'd like to understand the problem here a bit
> more...
>

The configure scripts are inside config dir and to proper configure it you
need to call it with some params; too hackish! I ended getting this change
from Debian package to not waste time mangling with options for autoreconf
and like.
Richard Purdie - June 12, 2012, 1:13 p.m.
On Fri, 2012-06-08 at 08:52 -0300, Otavio Salvador wrote:
> On Fri, Jun 8, 2012 at 7:33 AM, Richard Purdie
> <richard.purdie@linuxfoundation.org> wrote:
>         Why can't we run autoreconf instead of having the huge patch?
>         This might
>         be the right answer but I'd like to understand the problem
>         here a bit
>         more...
>
> The configure scripts are inside config dir and to proper configure it
> you need to call it with some params; too hackish! I ended getting
> this change from Debian package to not waste time mangling with
> options for autoreconf and like.

I really want to run reautoconf like we do with the rest of the system
for consistency. Its less important as long as this is a native recipe
but if we started building it for the target it would become a problem.

I appreciate you don't want to get distracted with something like that
and I can't really ask you to do it so I've done it myself put out an
alternative patch for review.

Cheers,

Richard
Otavio Salvador - June 12, 2012, 1:16 p.m.
On Tue, Jun 12, 2012 at 10:13 AM, Richard Purdie
<richard.purdie@linuxfoundation.org> wrote:
> I appreciate you don't want to get distracted with something like that
> and I can't really ask you to do it so I've done it myself put out an
> alternative patch for review.

I did try to make it this way and failed. Nice you did find a way for
it; this is indeed the better way to fix it.