On Mon, Feb 27, 2012 at 12:26:45PM +0200, Martin Storsjö wrote:
> On Mon, 27 Feb 2012, Diego Biurrun wrote:
> 
> >On Mon, Feb 27, 2012 at 12:01:51PM +0200, Martin Storsjö wrote:
> >>This fixes building libavformat with OpenSSL on slightly older
> >>distributions (e.g. CentOS 5.5).
> >>---
> >> libavformat/tls.c |    2 ++
> >> 1 files changed, 2 insertions(+), 0 deletions(-)
> >
> >CentOS 6 is out.  I've been through the pains of supporting the old
> >5.5 version with nightmarishly outdated versions of everything at work,
> >I really don't want to go through the pain again.  Do you think this is
> >worth the trouble?  By the time this change has trickled down from HEAD
> >to the distros, everybody will have updated and those that did not,
> >should.
> 
> I occasionally build and test a larger setup including libavformat
> on such a CentOS machine, and that setup enables the openssl code -
> this patch fixes building that setup. So for my case, I wouldn't
> need to wait for this fix to be included in the distros themselves.

So why not change the OpenSSL check to test for the availability of
that particular function?

> If you're opposed, I can live without this, but this particular
> patch is kinda nonintrusive, and I'd probably have written it in
> this way anyway if I had ran into the issue while making the tls
> stuff last year.

To be honest, I disagree slightly about the non-intrusiveness.
ifdefs are ifdefs and we should fight them tooth and nail.

Diego
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to