[MPlayer-cvslog] r22091 - trunk/configure

[Dominik 'Rathann' Mierzejewski]

On Wednesday, 31 January 2007 at 17:52, Reimar Döffinger wrote:
> Hello,
> On Wed, Jan 31, 2007 at 03:56:39PM +0100, Dominik 'Rathann' Mierzejewski wrote:
> > On Wednesday, 31 January 2007 at 15:42, Diego Biurrun wrote:
> > > On Wed, Jan 31, 2007 at 03:31:52PM +0100, Dominik 'Rathann' Mierzejewski wrote:
> > > > On Wednesday, 31 January 2007 at 10:25, lorenm wrote:
> [...]
> > > > > -    cc_check -c -O4 $_ld_tmp $_ld_lm && _libs_mencoder="$_libs_mencoder $_ld_faac" && _faac=yes && break
> > > > > +    cc_check -O4 $_ld_tmp $_ld_lm && _libs_mencoder="$_libs_mencoder $_ld_faac" && _faac=yes && break
> > > > 
> > > > Can we drop the "-O4" from test while we're at it?
> 
> I'd agree with that.
> 
> > > Why?  And why is it there in the first place?
> > 
> > Because it crashed on some GCC 3.x flavour while it worked with -O2 (which
> > I specified in CFLAGS). IOW: I don't like configure overriding my CFLAGS choice.
> 
> Well, that is more an argument to keep it. Anything that causes people
> running away screaming from broken compilers instead of producing broken
> code is good...

No. It would still break when the default flags are used, just later
(during compilation). However, second-guessing user overrides IS bad. If I
want to shoot myself in the foot (i.e. use my own CFLAGS), then, by all
means, let me do it! Not to mention that it's inconsistent with other
detection stuff.

Regards,
R.

-- 
MPlayer developer and RPMs maintainer: http://mplayerhq.hu http://rpm.livna.org
There should be a science of discontent. People need hard times and
oppression to develop psychic muscles.
	-- from "Collected Sayings of Muad'Dib" by the Princess Irulan