On Wed, May 04, 2011 at 12:29:50PM +0200, Julien BLACHE wrote: > "Steve M. Robbins" <st...@sumost.ca> wrote: > > Hi, > > > I'm with Linus on this: let's just revert to the old behaviour. A > > tiny amount of clock cycles saved isn't worth the instability. > > Tiny amount?! The optimized memcpy() variants that break shitty code > bring a 4 to 5x speedup on the processors they've been written for!
Though I didn't see any actual time measurements in the bug thread, I am sure there is a speedup of some kind for the memcpy() call itself. I'm not interested in that. I am interested to know the speedup -- measured in real-world conditions -- for actual, popular applications. But I'm really not interested that much. I'm really interested in having a running system. Yes, even one with buggy software that happens to be important to me. Cheers, -Steve
signature.asc
Description: Digital signature