> > I wonder how is it even possible? I mean improvement coefficient should > be higher for faster algorithms, while what we see is reverse: > improvement is higher for slower algorithms. Specifically for sha1 and > sha256 overhead is exactly same, while sha256 is significantly slower > than sha1, and then share of "inefficiency" is lower, which in turn > should result in lower improvement coefficient... > > There are several other places where bdnz is used with hint, what effect > would dropping hint have there?
It is interesting, I did not see much changes in performance when dropping the hint in other places (sha, bn, aes). > I assume you run on Linux, and on Linux Yes, I run on RHEL 6.2. > it's actually possible to drop hint "universally", i.e. without looking > for bdnz and modifying all occurrences. In other words such experiment > would be easy to conduct. To do so open crypto/perlasm/ppc-xlate.pl in > text editor, locate $bdnz, then $bo and replace 16+9 with 16. It's not > "official solution", just a way to conduct experiment... > Per your suggestion, I replaced 16+9 with 16 and got the results attached in the spreedsheet. It does not show much changes in the performance. If you have any suggestions or other experiment that you would like to try please let me know. Thanks, -- Ashley Lai
opensslPerfExp.ods
Description: application/vnd.oasis.opendocument.spreadsheet
