> For all OpenSSL rc4-amd64s RC4_INT is int, not long. And for *all*
> rc4-amd64 RC4_CHUNK is effectively long.

Yes, it's my fault. 

I also tested your new rc4-amd64.pl patch. But found performance down
compared with v1.3 version both on Opteron and Nocona.

Opteron 2.2G:
rc4-amd64.v1.3   371MB/s
rc4-amd64.new   306MB/s

Nocona 3.6G:
rc4-amd64.v1.3   290MB/s
rc4-amd64.new   265MB/s

Did you find the same phenomena? 

> >
> > On AMD64, 'dec' is as fast as 'sub $1', but replacing 'ror' by 'shl+bswap'
> > is indeed about 1% faster (thanks, I have updated rc4-amd64).
Yes, I also found that. 

Thx, Xuekun
______________________________________________________________________
OpenSSL Project                                 http://www.openssl.org
Development Mailing List                       [EMAIL PROTECTED]
Automated List Manager                           [EMAIL PROTECTED]

Reply via email to