caml-list - the Caml user's mailing list
 help / color / mirror / Atom feed
From: "Yuanchen Zhu" <yzhu@fas.harvard.edu>
To: caml-list@yquem.inria.fr
Subject: Re: [Caml-list] Comparison of OCaml and MLton for numerics
Date: Thu, 31 May 2007 17:22:34 +0800	[thread overview]
Message-ID: <5195a210705310222p6aa8482fr70e7bf2b2b631b72@mail.gmail.com> (raw)
In-Reply-To: <200705311008.16662.jon@ffconsultancy.com>

>
> You are doing all this computation in your inner loop unnecessarily:
>
> let getReflected (img:t) y x yr xr =
>   let w = width img and h = height img in
>     if x >= 0 && x < w && y >= 0 && y < h then
>       img.(y).(x)
>     else
>       let x = abs x and y = abs y in
>       let xx = x/w and x' = x mod w and yy = y/h and y' = y mod h in
>       let x',xr = if xx mod 2 = 0 then (x',1.0) else (w - 1 - x',xr) in
>       let y',yr = if yy mod 2 = 0 then (y',1.0) else (h - 1 - y',yr) in
>         img.(y').(x') *. xr *.yr
>
> Hoist as much as you can from the inner loop and this program will run much
> faster in any language.

Yes, I am aware of the inefficiency here. Although because of branch
prediction on modern CPUs, the actual running time overhead would not
be that big.

My point, however, is that MLton and OCaml are being fed the same
code, and if OCaml performs specializing and proper inlining, it will
get almost twice its current performance.


  reply	other threads:[~2007-05-31  9:22 UTC|newest]

Thread overview: 71+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-05-31  5:50 Yuanchen Zhu
2007-05-31  6:17 ` [Caml-list] " Jon Harrop
2007-05-31  6:32   ` skaller
2007-05-31  7:31   ` Yuanchen Zhu
2007-05-31  9:08     ` Jon Harrop
2007-05-31  9:22       ` Yuanchen Zhu [this message]
2007-05-31 10:27         ` Jon Harrop
2007-05-31 21:30           ` Alain Frisch
2007-06-01  1:22             ` skaller
2007-06-01  1:36               ` Erik de Castro Lopo
2007-06-01  2:21                 ` skaller
2007-06-01  2:49                   ` Erick Tryzelaar
2007-06-01  3:05                     ` skaller
2007-06-01  5:30               ` Alain Frisch
2007-06-01  5:39                 ` Jon Harrop
2007-06-01  6:36                   ` Yuanchen Zhu
2007-06-01  8:09                 ` skaller
2007-06-01  8:53                   ` Richard Jones
2007-06-01  8:59                     ` Richard Jones
2007-06-01  9:22                       ` Stephan Tolksdorf
2007-06-01  9:49                         ` Richard Jones
2007-06-01  9:32                       ` Stephan Tolksdorf
2007-06-01 10:02                     ` skaller
2007-06-01 11:29                 ` Yaron Minsky
2007-06-01 11:43                   ` Erik de Castro Lopo
2007-06-01 11:58                     ` Jon Harrop
2007-06-01 13:49                       ` Julien Signoles
2007-06-01 14:18                         ` Stephen Weeks
2007-06-01 14:43                           ` Julien Signoles
2007-06-01 14:57                           ` Brian Hurt
2007-06-01 15:40                             ` Alain Frisch
2007-06-01 15:58                               ` Brian Hurt
2007-06-01 16:25                                 ` Markus Mottl
2007-06-01 16:47                               ` Jon Harrop
2007-06-01 23:26                             ` skaller
2007-06-01 23:49                               ` Brian Hurt
2007-06-02  3:26                                 ` skaller
2007-06-01 12:40                     ` Erik de Castro Lopo
2007-06-01 13:56                       ` Julien Signoles
2007-06-01 11:49                   ` David MENTRE
2007-06-01 14:41                     ` skaller
2007-06-01 16:52                       ` Jon Harrop
2007-06-01 23:33                         ` skaller
2007-06-01 16:14                     ` Markus Mottl
2007-06-01 16:46                       ` Jon Harrop
2007-06-01 17:13                       ` Jon Harrop
2007-06-04 14:03                         ` Mike Furr
2007-06-04 14:39                           ` Jon Harrop
2007-06-04 15:33                             ` Mike Furr
2007-06-04 18:08                             ` skaller
     [not found]                               ` <9d3ec8300706041518y115d22bdsa120d4010261d841@mail.gmail.com>
2007-06-04 22:19                                 ` Fwd: " Till Varoquaux
2007-06-04 23:40                                   ` Jon Harrop
2007-06-05  2:24                                   ` skaller
2007-06-04 22:44                               ` Pierre Etchemaïté
2007-06-05  1:42                                 ` Jon Harrop
2007-06-05 10:30                                   ` Jon Harrop
2007-06-10 12:10                           ` Jon Harrop
2007-06-10 12:58                             ` skaller
2007-06-01 14:15                 ` Stephen Weeks
2007-06-01 14:37                   ` Brian Hurt
2007-06-01 14:39                   ` Eric Cooper
2007-05-31  9:24       ` Yuanchen Zhu
2007-05-31 10:25       ` Loup Vaillant
2007-05-31 10:30         ` Jon Harrop
2007-05-31 12:12     ` skaller
2007-05-31  7:11 ` Daniel Bünzli
2007-05-31 15:15 ` Christophe Raffalli
2007-05-31 15:23   ` Jon Harrop
2007-05-31 15:35     ` Christophe Raffalli
     [not found]       ` <604682010705310923o5a1ee0eiee5ae697da9d3c60@mail.gmail.com>
2007-05-31 20:14         ` Stephen Weeks
2007-05-31 15:16 ` Christophe Raffalli

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=5195a210705310222p6aa8482fr70e7bf2b2b631b72@mail.gmail.com \
    --to=yzhu@fas.harvard.edu \
    --cc=caml-list@yquem.inria.fr \
    /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).