From: erik quanstrom <quanstro@quanstro.net>
To: 9fans@9fans.net
Subject: Re: [9fans] integer width on AMD64 (was: Re: AMD64 system)
Date: Mon, 7 May 2012 08:44:37 -0400 [thread overview]
Message-ID: <dce977796edc89acaf58226c0a24e2de@kw.quanstro.net> (raw)
In-Reply-To: <3E93FE94-76BC-4B38-9FB2-DEDC5C3CEF9E@quintile.net>
On Mon May 7 04:54:23 EDT 2012, steve@quintile.net wrote:
> sorry for being vague.
>
> treating pixels as 64bit on amd64 as that is the natural size for the
> machine, vs using 32bits per pixel - 10 bits of r, g, and b or y, u,
> and v plus 2 spare leads to a significant speedup; where significant
> is a number lost in the mists of time.
>
> i believe this speedup is due to the reduction in the rate of cache
> line refills, as forsyth described.
i'm confused. (asserting parity error.) which one's faster?
given your problem one would assume that in the absense of
any real gotchas,
processor bw >> memory bw ==> smaller integers faster
memory bw >> processor bw ==> natural integers faster
(this is yet another reason that int_fast* are a half-baked idea.
how does the compiler know this relation for the target machine
ahead of time?)
don't get me wrong, i can easily believe there are gotchas, that's
why i'm confused.
- erik
next prev parent reply other threads:[~2012-05-07 12:44 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <CAL3m8eAf2Yjpg77wtykTSPnyKHxKHYc5YWfWH2L8CNjc86P1kA@mail.gmail.c>
2012-04-25 14:28 ` [9fans] AMD64 system erik quanstrom
2012-04-25 18:04 ` Strake
2012-04-25 18:27 ` Lyndon Nerenberg
2012-04-25 18:49 ` Matthew Veety
2012-04-25 19:09 ` Strake
2012-04-25 19:15 ` John Floren
2012-04-25 19:57 ` Strake
2012-04-25 20:04 ` John Floren
2012-04-25 20:30 ` Strake
2012-04-25 20:43 ` John Floren
2012-04-26 1:11 ` Strake
2012-04-26 1:21 ` andrey mirtchovski
2012-04-26 1:24 ` andy zerger
2012-04-26 1:49 ` John Floren
2012-04-26 3:41 ` Strake
2012-04-25 19:19 ` Lyndon Nerenberg
2012-04-25 20:01 ` Strake
2012-04-25 20:05 ` Gorka Guardiola
2012-04-25 20:08 ` Lyndon Nerenberg
[not found] ` <CAL3m8eCJughpHZ6htHyLyq9vORA+z5ajpWMbWjnCTVZ+YZRv9g@mail.gmail.c>
2012-04-25 19:36 ` erik quanstrom
[not found] ` <CA+4OWrwGgkYfuj2t4dkWsZHFex4oFTveDq_9A+yr=TsNpG0C8g@mail.gmail.c>
2012-04-25 18:56 ` erik quanstrom
2012-04-25 19:16 ` Strake
[not found] ` <CAL3m8eDJEo6o+srxuYpKSGDmM+0KgQhC15u7Qxg4TvnpO57Zgw@mail.gmail.c>
2012-04-25 19:32 ` erik quanstrom
2012-04-25 20:13 ` Strake
2012-04-25 20:20 ` Lyndon Nerenberg
2012-04-26 3:38 ` Russ Cox
2012-04-26 4:04 ` Devon H. O'Dell
2012-04-26 4:13 ` andrey mirtchovski
2012-04-26 4:36 ` Strake
2012-05-05 15:02 ` Ethan Grammatikidis
2012-05-05 15:33 ` [9fans] integer width on AMD64 (was: Re: AMD64 system) dexen deVries
2012-05-05 16:41 ` David du Colombier
2012-05-05 17:42 ` erik quanstrom
2012-05-05 17:48 ` Charles Forsyth
2012-05-05 17:52 ` dexen deVries
2012-05-05 21:06 ` Comeau At9Fans
2012-05-06 2:58 ` [9fans] integer width on AMD64 Joel C. Salomon
2012-05-06 3:13 ` Bruce Ellis
2012-05-06 8:46 ` Comeau At9Fans
[not found] ` <CAE9W7-gmaSYzx+4TK0LKtiHaZYC-4xz_DTbC_wqhsGo_7zXc0Q@mail.gmail.c>
2012-05-06 13:10 ` erik quanstrom
2012-05-06 13:57 ` Comeau At9Fans
2012-05-06 9:20 ` [9fans] integer width on AMD64 (was: Re: AMD64 system) steve
2012-05-06 11:43 ` Comeau At9Fans
2012-05-07 8:53 ` steve
2012-05-07 10:01 ` dexen deVries
2012-05-07 10:16 ` Charles Forsyth
2012-05-07 10:27 ` Charles Forsyth
2012-05-07 10:36 ` dexen deVries
2012-05-08 19:04 ` Comeau At9Fans
[not found] ` <CAOw7k5h9QVUCrKr4jYU++sSWE6DHqbA_oFRwajyKbj7A0AjBWg@mail.gmail.c>
2012-05-07 12:11 ` erik quanstrom
2012-05-07 12:32 ` erik quanstrom
2012-05-07 12:44 ` erik quanstrom [this message]
2012-05-08 19:14 ` Comeau At9Fans
2012-05-05 21:00 ` Comeau At9Fans
[not found] ` <CAL3m8eDMXC2STjFjBUurmdY5DynCOrpqwE-ViRuSubfF6HjBOw@mail.gmail.c>
2012-04-25 18:11 ` [9fans] AMD64 system erik quanstrom
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=dce977796edc89acaf58226c0a24e2de@kw.quanstro.net \
--to=quanstro@quanstro.net \
--cc=9fans@9fans.net \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).