Branko Badrljica <bran...@avtomatika.com> posted 49fca81d.4040...@avtomatika.com, excerpted below, on Sat, 02 May 2009 22:07:57 +0200:
> BTW: How much of a performance gain would be if one could compile > everything with -profile-generate, run resulting program on some > representative test data and then use profile files to generate final > binary with -fprofile-use ? In theory, that's more likely to give you a decent gain than most of those fancy optimizations you listed. However, few people do it for one- off or only a few users, as the time to compile, profile, and recompile, is way more than the time you'd save by the better optimizations, for just a few users over the life of most packages. OTOH, for something like fold...@home or the like that you'll be running 24/7 at full CPU untilization, it's likely to make sense, as it would for say Red Hat for x86 especially for packages like glibc, xorg, etc, that most folks use, since they likely have tens of thousands of people running the same binaries, often for a much longer time between releases, than Gentoo has. That's an entirely different scenario than every Gentoo user compiling his own, meanwhile, pointing out one of the weaknesses in the Gentoo setup, since that profile data really doesn't make sense to collect and use on an individual basis for the vast majority of usage scenarios. -- Duncan - List replies preferred. No HTML msgs. "Every nonfree program has a lord, a master -- and if you use the program, he is your master." Richard Stallman