From: Artur Skawina <art.08.09@gmail.com>
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Git Mailing List <git@vger.kernel.org>
Subject: Re: [PATCH 0/7] block-sha1: improved SHA1 hashing
Date: Fri, 07 Aug 2009 01:19:13 +0200 [thread overview]
Message-ID: <4A7B64F1.2000309@gmail.com> (raw)
In-Reply-To: <alpine.LFD.2.01.0908061531310.3390@localhost.localdomain>
Linus Torvalds wrote:
>
> Yeah, verified. Google for
>
> northwood "barrel shifter"
>
> and you'll find a lot of it.
>
> Basically, older P4's will I think shift one bit at a time. So while even
> Prescott is relatively weak in the shifter department, pre-prescott
> (Willamette and Northwood) are _really_ weak. If your P4 is one of those,
> you really shouldn't use it to decide on optimizations.
Actually that's even more of a reason to make sure the code doesn't suck :)
The difference on less perverse cpus will usually be small, but on P4 it
can be huge.
A few years back I found my old ip checksum microbenchmark, and when I ran
it on a P4 (prescott iirc) i didn't believe my eyes. The straightforward
32-bit C implementation was running circles around the in-kernel one...
And a few tweaks to the assembler version got me another ~100% speedup.[1]
After that the P4 became the very first cpu to test any code on... :)
artur
[1] just reran the benchmark on this p4; true on northwood too:
IACCK 0.9.30 Artur Skawina <...>
[ exec time; lower is better ] [speed ] [ time ] [ok?]
TIME-N+S TIME32 TIME33 TIME1480 MBYTES/S TIMEXXXX CSUM FUNCTION ( rdtsc_overhead=0 null=0 )
17901 510 557 3010 393.36 59772 56dd csum_partial_cdumb16
3019 154 156 431 2747.10 43106 56dd csum_partial_c32
2413 170 177 328 3609.76 37501 56dd csum_partial_c32l
2437 170 170 328 3609.76 37488 56dd csum_partial_c32i
5078 205 254 767 1543.68 48117 56dd csum_partial_std
5612 299 291 851 1391.30 53673 56dd csum_partial_686
1584 99 127 227 5215.86 14495 56dd csum_partial_586f
1738 107 121 229 5170.31 14785 56dd csum_partial_586fs
4893 175 171 759 1559.95 52347 56dd csum_partial_copy_generic_std
4949 151 189 756 1566.14 67847 56dd csum_partial_copy_generic_686
2072 110 134 302 3920.53 39061 56dd csum_partial_copy_generic_p4as1
next prev parent reply other threads:[~2009-08-06 23:19 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-08-06 15:13 [PATCH 0/7] block-sha1: improved SHA1 hashing Linus Torvalds
2009-08-06 15:15 ` [PATCH 1/7] block-sha1: add new optimized C 'block-sha1' routines Linus Torvalds
2009-08-06 15:16 ` [PATCH 2/7] block-sha1: try to use rol/ror appropriately Linus Torvalds
2009-08-06 15:18 ` [PATCH 3/7] block-sha1: make the 'ntohl()' part of the first SHA1 loop Linus Torvalds
2009-08-06 15:20 ` [PATCH 4/7] block-sha1: re-use the temporary array as we calculate the SHA1 Linus Torvalds
2009-08-06 15:22 ` [PATCH 5/7] block-sha1: macroize the rounds a bit further Linus Torvalds
2009-08-06 15:24 ` [PATCH 6/7] block-sha1: Use '(B&C)+(D&(B^C))' instead of '(B&C)|(D&(B|C))' in round 3 Linus Torvalds
2009-08-06 15:25 ` [PATCH 7/7] block-sha1: get rid of redundant 'lenW' context Linus Torvalds
2009-08-06 18:25 ` [PATCH 2/7] block-sha1: try to use rol/ror appropriately Bert Wesarg
2009-08-06 17:22 ` [PATCH 0/7] block-sha1: improved SHA1 hashing Artur Skawina
2009-08-06 18:09 ` Linus Torvalds
2009-08-06 19:10 ` Artur Skawina
2009-08-06 19:41 ` Linus Torvalds
2009-08-06 20:08 ` Artur Skawina
2009-08-06 20:53 ` Linus Torvalds
2009-08-06 21:24 ` Linus Torvalds
2009-08-06 21:39 ` Artur Skawina
2009-08-06 21:52 ` Artur Skawina
2009-08-06 22:27 ` Linus Torvalds
2009-08-06 22:33 ` Linus Torvalds
2009-08-06 23:19 ` Artur Skawina [this message]
2009-08-06 23:42 ` Linus Torvalds
2009-08-06 22:55 ` Artur Skawina
2009-08-06 23:04 ` Linus Torvalds
2009-08-06 23:25 ` Linus Torvalds
2009-08-07 0:13 ` Linus Torvalds
2009-08-07 1:30 ` Artur Skawina
2009-08-07 1:55 ` Linus Torvalds
2009-08-07 0:53 ` Artur Skawina
2009-08-07 2:23 ` Linus Torvalds
2009-08-07 4:16 ` Artur Skawina
[not found] ` <alpine.LFD.2.01.0908071614310.3288@localhost.localdomain>
[not found] ` <4A7CBD28.6070306@gmail.com>
[not found] ` <4A7CBF47.9000903@gmail.com>
[not found] ` <alpine.LFD.2.01.0908071700290.3288@localhost.localdomain>
[not found] ` <4A7CC380.3070008@gmail.com>
2009-08-08 4:16 ` Linus Torvalds
2009-08-08 5:34 ` Artur Skawina
2009-08-08 17:10 ` Linus Torvalds
2009-08-08 18:12 ` Artur Skawina
2009-08-08 22:58 ` Artur Skawina
2009-08-08 23:36 ` Artur Skawina
-- strict thread matches above, loose matches on Subject: below --
2009-08-07 7:36 George Spelvin
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4A7B64F1.2000309@gmail.com \
--to=art.08.09@gmail.com \
--cc=git@vger.kernel.org \
--cc=torvalds@linux-foundation.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.