From: Diego Novillo <dnovillo@google.com>
To: Richard Kenner <kenner@vlsi1.ultra.nyu.edu>
Cc: joseph@codesourcery.com, davidxl@google.com, ebotcazou@gnat.com,
gcc-patches@gcc.gnu.org, hubicka@ucw.cz,
richard.guenther@gmail.com
Subject: Re: [google]: initialize language field for clone function struct
Date: Wed, 04 May 2011 12:23:00 -0000 [thread overview]
Message-ID: <BANLkTinq2-5TGD_rZAu7LZzrwa1tE7BuVw@mail.gmail.com> (raw)
In-Reply-To: <20110504114442.1E15033C67@vlsi1.gnat.com>
On Wed, May 4, 2011 at 07:44, Richard Kenner <kenner@vlsi1.ultra.nyu.edu> wrote:
>> That's what we should phase out. The eventual aim should be for (a)
>> folding on GIMPLE (gimple-fold etc. - working with SSA not combined trees)
>> as an optimization and (b) folding done by front ends only when required
>> for language semantics (e.g. constant expressions).
>
> Why? Isn't it always better to optimize as early as you can when it's
> easy? Why keep unfolded constants around at all?
There are pros and cons about early optimization, actually.
Generating extremely optimized IL very early can actually tie up
subsequent passes. For instance, loop unrolling and vectorization.
There are others in the literature.
Diego.
next prev parent reply other threads:[~2011-05-04 12:20 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-04-30 1:34 Xinliang David Li
2011-05-02 22:13 ` Xinliang David Li
2011-05-03 10:55 ` Jan Hubicka
2011-05-03 16:15 ` Xinliang David Li
2011-05-03 16:46 ` Jan Hubicka
2011-05-03 18:55 ` Eric Botcazou
2011-05-03 19:07 ` Joseph S. Myers
2011-05-03 19:16 ` Nathan Froyd
2011-05-03 19:51 ` Joseph S. Myers
2011-05-03 20:05 ` Nathan Froyd
2011-05-03 19:52 ` Eric Botcazou
2011-05-03 20:09 ` Joseph S. Myers
2011-05-03 20:36 ` Eric Botcazou
2011-05-03 21:27 ` Joseph S. Myers
2011-05-04 8:53 ` Eric Botcazou
2011-05-04 17:27 ` Tom Tromey
2011-05-04 9:52 ` Richard Guenther
2011-05-04 10:02 ` Eric Botcazou
2011-05-04 10:30 ` Joseph S. Myers
2011-05-04 11:01 ` Eric Botcazou
2011-05-04 11:50 ` Richard Kenner
2011-05-04 12:23 ` Diego Novillo [this message]
2011-05-04 12:27 ` Richard Kenner
2011-05-04 12:31 ` Diego Novillo
2011-05-04 13:16 ` Michael Matz
2011-05-04 13:22 ` Nathan Froyd
2011-05-04 13:26 ` Diego Novillo
2011-05-04 13:39 ` Richard Kenner
2011-05-04 11:19 ` Richard Guenther
2011-05-04 12:00 ` Michael Matz
2011-05-04 12:08 ` Richard Guenther
2011-05-04 15:33 ` Eric Botcazou
2011-05-04 16:05 ` Richard Guenther
2011-05-04 17:22 ` Eric Botcazou
2011-05-05 9:07 ` Richard Guenther
2011-05-05 9:25 ` Eric Botcazou
2011-05-05 9:44 ` Richard Guenther
2011-05-05 10:22 ` Eric Botcazou
2011-05-05 11:03 ` Richard Guenther
2011-05-05 11:59 ` Eric Botcazou
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=BANLkTinq2-5TGD_rZAu7LZzrwa1tE7BuVw@mail.gmail.com \
--to=dnovillo@google.com \
--cc=davidxl@google.com \
--cc=ebotcazou@gnat.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=hubicka@ucw.cz \
--cc=joseph@codesourcery.com \
--cc=kenner@vlsi1.ultra.nyu.edu \
--cc=richard.guenther@gmail.com \
/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 read-only IMAP folder(s) and NNTP newsgroup(s).