Doesn't help. (Though it's not a bad idea in general.) The other interesting artifact is that the code is noticably slower in terms of cycles per iteration running on a Pentium 4, clocking in around 2700 cycles per iteration. As a side check I'm also using gettimeofday() to get wall-clock times, and those agree very closely with the cycle counts I'm getting.
![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
no subject
Date: 2003-11-05 01:54 pm (UTC)