Date: Mon, 14 Sep 2015 21:46:56 +0200 From: magnum <john.magnum@...hmail.com> To: john-dev@...ts.openwall.com Subject: Re: md5crypt mmxput*() On 2015-09-14 19:44, Solar Designer wrote: > On Wed, Sep 09, 2015 at 03:00:01AM +0200, magnum wrote: >> On 2015-09-08 16:18, Solar Designer wrote: >>> Benchmarking: md5crypt, crypt(3) $1$ [MD5 128/128 XOP 4x2]... (8xOMP) DONE >>> Raw: 231680 c/s real, 28923 c/s virtual >> >> Code size, eh? This reminded me there is a "#pragma GCC optimize 3" in >> that file that I always found slightly dubious. We should verify how >> each format reacts to dropping that. >> >> Quick test for now, on bull; Enabled the switch for XOP, dropped that >> pragma: >> >> Benchmarking: md5crypt, crypt(3) $1$ [MD5 128/128 XOP 4x2]... (8xOMP) DONE >> Raw: 233472 c/s real, 29184 c/s virtual > > I think you never committed this change? Since you also confirmed it > helping on AVX2, I think you should, unless you know of a regression. > Have you done more testing since? Yes, but I can't seem to get conclusive results. I guess I should just commit it. BTW the total size of simd-intrinsics.o (after stripping) actually increased. I'm not sure how to get detailed figures (eg. per function or something?). magnum
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.