Date: Tue, 26 May 2015 15:26:34 +0800 From: Lei Zhang <zhanglei.april@...il.com> To: john-dev@...ts.openwall.com Subject: Re: interleaving in SHA256 & SHA512 > On May 25, 2015, at 4:05 PM, magnum <john.magnum@...hmail.com <mailto:john.magnum@...hmail.com>> wrote: > > BTW for interleaving factors, you should not just try multiples like 2, 4, 8. You need to try all of 2, 3, 4, 5 - and more than 5 is probably never worthwhile (if it is, I guess it indicates we should look into the code instead of just bumping para). So maybe you need to retry all those benchmark yet again ;-) Ok, here's a re-test on MIC, with both a OpenMP-enabled and a OpenMP-disabled build: OpenMP-enabled: pwsafe (sha256) x1 Raw: 128000 c/s real, 532 c/s virtual x2 Raw: 104388 c/s real, 435 c/s virtual x3 Raw: 105688 c/s real, 441 c/s virtual x4 Raw: 105411 c/s real, 439 c/s virtual x5 Raw: 105688 c/s real, 441 c/s virtual sha512crypt x1 Raw: 6262 c/s real, 26.1 c/s virtual x2 Raw: 6606 c/s real, 27.9 c/s virtual x3 Raw: 6658 c/s real, 28.1 c/s virtual x4 Raw: 7029 c/s real, 29.6 c/s virtual x5 Raw: 6946 c/s real, 30.3 c/s virtual OpenMP-disabled: pwsafe (sha256) x1 Raw: 1129 c/s real, 1129 c/s virtual x2 Raw: 869 c/s real, 869 c/s virtual x3 Raw: 914 c/s real, 914 c/s virtual x4 Raw: 923 c/s real, 923 c/s virtual x5 Raw: 941 c/s real, 941 c/s virtual sha512crypt x1 Raw: 51.2 c/s real, 51.2 c/s virtual x2 Raw: 50.0 c/s real, 50.0 c/s virtual x3 Raw: 55.1 c/s real, 55.1 c/s virtual x4 Raw: 57.9 c/s real, 57.9 c/s virtual x5 Raw: 57.2 c/s real, 57.2 c/s virtual I think the performance of the OpenMP-disabled build basically matches that of the OpenMP-enabled build under different interleaving factors. And the optimal SIMD_PARA_SHA512 for MIC is 4. Lei Content of type "text/html" skipped
Powered by blists - more mailing lists
Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.