From: Benedikt Meurer <benedikt.meurer@googlemail.com>
To: Gerd Stolpmann <info@gerd-stolpmann.de>
Cc: caml-list@inria.fr
Subject: Re: [Caml-list] Linear Scan Register Allocator for ocamlopt/ocamlnat
Date: Thu, 25 Aug 2011 10:02:39 +0200 [thread overview]
Message-ID: <11577B43-4270-4BA4-AA77-7FDEFB4B563B@googlemail.com> (raw)
In-Reply-To: <1314218451.3496.42.camel@thinkpad>
On Aug 24, 2011, at 22:40 , Gerd Stolpmann wrote:
>>> - the performance cost of this new allocator in the generated code? I
>>> suppose the results may vary between different architectures (eg. x86
>>> is probably more sensitive to good allocation decisions than x86_64).
>>
>> - http://ps.informatik.uni-siegen.de/~meurer/tmp/compiletime_timings.pdf contains a comparison of the ocamlopt invocations.
>> - http://ps.informatik.uni-siegen.de/~meurer/tmp/runtime_timings.pdf contains comparison of the generated code.
>>
>> As can be seen from the results, amd64 is more sensitive to register allocator changes than i386.
>
> Well, this particular i386 CPU model is a strange guy - Northwoods have
> this extremely long pipeline, which is very sensitive to unforeseen
> jumps. It would be more interesting to see this test on a modern CPU in
> i386 mode. My guess is that it behaves then more like amd64.
Modern CPUs most probably don't run ocaml in 32bit mode, but more likely in long mode. That's why we choose to run the i386 on "real 32bit hardware", where the ocaml i386 port is actually used.
I'll rerun the benchmark in 32bit mode on a modern cpu to see how things change.
> Gerd
Benedikt
next prev parent reply other threads:[~2011-08-25 8:02 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-08-01 14:53 Benedikt Meurer
2011-08-01 15:03 ` [Caml-list] " Benedikt Meurer
2011-08-01 15:04 ` [Caml-list] " Gabriel Scherer
2011-08-01 16:57 ` Benedikt Meurer
2011-08-24 19:35 ` Benedikt Meurer
2011-08-24 20:40 ` Gerd Stolpmann
2011-08-25 8:02 ` Benedikt Meurer [this message]
2011-08-25 9:34 ` Benedikt Meurer
2011-08-25 10:21 ` Gerd Stolpmann
2011-08-25 10:25 ` Pierre-Alexandre Voye
2011-08-26 10:58 ` Guillaume Yziquel
2011-08-26 12:29 ` Erik de Castro Lopo
[not found] ` <CANnJ5GfmnOhk8mJor8PfrGoC1-5vZWOgUKNZMT9k69iu0TBCeA@mail.gmail.com>
[not found] ` <CANnJ5Gf4AnKPDvrKiYx0joiPbQVOCdFWgPR24Y1ZYYL1Qv5=fg@mail.gmail.com>
2011-08-26 18:33 ` Pierre-Alexandre Voye
2011-08-30 21:07 ` Fermin Reig
2011-08-30 21:21 ` Pierre-Alexandre Voye
2011-08-25 10:43 ` Benedikt Meurer
2011-08-24 20:47 ` malc
2011-08-26 12:40 ` Roberto Di Cosmo
2011-08-01 15:29 ` Wojciech Meyer
2011-08-01 17:03 ` Benedikt Meurer
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=11577B43-4270-4BA4-AA77-7FDEFB4B563B@googlemail.com \
--to=benedikt.meurer@googlemail.com \
--cc=caml-list@inria.fr \
--cc=info@gerd-stolpmann.de \
/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).