On Mon, Mar 20, 2006 at 09:20:48AM -0600, Randy McMurchy wrote: > Bruce Dubbs wrote these words on 03/20/06 08:33 CST: > > > Kristian Gavran wrote: > >>2. The configure script insists to configure the CFLAG for itself. So it > >>would be usefull to bypass the CFLAG environment variable when calling > >>the configure script. Further we should use a sed command to correct the > >>autogenerated CFLAG. The reason for this is that the configure script > >>generates an "-O4" switch in the config.mak. This is useless because the > >>highest optimization flag supported by GCC is "-O3". > > The sed sounds like a good idea. As far as CFLAG goes, if the > developer put it there, we usually keep it, as we feel the > developer knows best what makes his program run best. Perhaps > a note saying the MPlayer sets a default CFLAG so that folks know > it and can remove it if they wish.
Unsetting CFLAGS is probably a good idea, though I wouldn't use the
sed to put -O4 to -O3. -O4 is perfectly valid, but equivalent to -O3
(you can check in gcc/opts.c, function decode_options). Actually, -O45
would also be equivalent to -O3 :-)
I agree that it doesn't make sense to use -ON (with N>3), but if the
developers deem that necessary... There are a couple of more projects
using something like -O20, so with 'fixing' MPlayer the work wouldn't
be finish...
And who knows.. one day, gcc might start doing some serious
voodoo optimizations with -O<hugenumber>.
Cheers,
Steffen
--
I say, if your knees aren't green by the end of the day, you ought to
seriously re-examine your life. -- Calvin
pgpZlXOxL1KtU.pgp
Description: PGP signature
-- http://linuxfromscratch.org/mailman/listinfo/blfs-dev FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
