Mail Archives: djgpp/1997/11/30/06:17:56
On 26 Nov 1997, Ed Leiby wrote:
> I tried the -pg/gprof angle and was able to identify where a general
> purpose floating pt. complex correlation routine was showing a total
> ms/call of 0.04 ( slow case ) compared to 5.11 total us/call. Assuming
> ms=milliseconds and us =microseconds this is a ~8:1 delta. This routine
> is simple and seems to give the same results whether its running slow or
> fast.
This might be far-fetched, but if the code generated by as.exe from
Binutils 2.8.1 is slightly larger than that from 2.7, it could
overflow your L1 or L2 CPU caches, which might then cause certain
slow-down.
- Raw text -