[FFmpeg-devel] [PATCH] define _BSD_SOURCE for bktr.c

Jacob Meuser jakemsr
Sat Dec 13 21:47:32 CET 2008


On Sat, Dec 13, 2008 at 12:46:56PM +0100, Diego Biurrun wrote:
> On Sat, Dec 13, 2008 at 03:01:27AM +0000, Jacob Meuser wrote:
> > On Sat, Dec 13, 2008 at 03:57:50AM +0100, Diego Biurrun wrote:
> > > On Sat, Dec 13, 2008 at 02:46:14AM +0000, Jacob Meuser wrote:
> > > > On Sat, Dec 13, 2008 at 03:33:36AM +0100, Diego Biurrun wrote:
> > > > > On Sat, Dec 13, 2008 at 01:10:38AM +0000, Jacob Meuser wrote:
> > > > > > 
> > > > > > I maintain the openbsd port/package.
> > > > > 
> > > > > So how about submitting some patches to us instead of doing
> > > > > incomprehensible things in your ports tree?
> > > > 
> > > > I tried and was told to fix OpenBSD instead.
> > > 
> > > Somehow Reimar was able to submit a part of your patchset.
> > > 
> > > It's all a matter of picking out the relevant parts and submitting them
> > > in a suitable form.  Some parts of your patchset look like ugly hacks,
> > > some look acceptable if only I could understand what you are trying to
> > > achieve...
> > 
> > most of the ugly is to maintain correct library dependencies.
> 
> Well, you have 8 patches in your tree for a multitude of different
> things...
> 
> > if you have specific questions, just ask.
> 
> patch-libavdevice_v4l2_c: no longer necessary
> 
> patch-version_sh: Please explain.

we don't do svn in our ports.  I mean, we don't pull sources from
cvs, svn, git or whatever as part of building a package.  instead
someone puts a tarball on a server, and we use that.

> patch-libavcodec_ppc_check_altivec_c: could be acceptable.
> 
> patch-configure: I merged some more parts, the rest needs an explanation.

haven't kept up with what is and isn't merged.  what parts still
need to be explained?

> patch-tests_regression_sh: half-merged, half mysterious

ports run the regression tests before installation.

> patch-libavcodec_libfaad_c, patch-libavcodec_dsputil_c,
> patch-subdir_mak: mysterious

libfaad uses e.g. long in headers and then int32_t in the actual
library code.  that doesn't work on, say, sparc64.  so I patched
libfaad to not use long for 32-bit values, along with all the rest
of the ports that use faad.

ff_check_alignment causes crashes on OpenBSD.  I commented it out
and things worked.  at least that is true for that particular
ffmpeg version in ports.

I explained the subdir.mk patch long ago on this very list.

-- 
jakemsr at sdf.lonestar.org
SDF Public Access UNIX System - http://sdf.lonestar.org




More information about the ffmpeg-devel mailing list