public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "rguenth at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug rtl-optimization/59429] Missed optimization opportunity in qsort-style comparison functions
Date: Mon, 05 May 2014 08:48:00 -0000	[thread overview]
Message-ID: <bug-59429-4-ALNZpcJfrY@http.gcc.gnu.org/bugzilla/> (raw)
In-Reply-To: <bug-59429-4@http.gcc.gnu.org/bugzilla/>

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59429

--- Comment #12 from Richard Biener <rguenth at gcc dot gnu.org> ---
My idea was to have sth like tree-affine for conditions and that condition
optimizing passes (like ifcombine or ifcvt or phiopt) build an on-the-side
representation for this, also catching straight-line code

  a = b < 0;
  c = d > 1;
  d = a | c;
  if (d)
    ...

the tree-affine like condition can be normalized and simplified on-the-side
and you'd do simplifications/lookup using that machinery, only emitting
changed computes in a transform stage.  Basically you collect BB or edge
predicates (like ifcvt does) but with a more elaborate machinery that
allows for simplification (ifcombine has its own as well).

I wouldn't try to tackle this testcase from a pattern matching side.
In principle comment#4 looks like sth for phiopt.


  parent reply	other threads:[~2014-05-05  8:48 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-12-09 11:35 [Bug rtl-optimization/59429] New: " jengelh at inai dot de
2013-12-09 13:16 ` [Bug rtl-optimization/59429] " hjl.tools at gmail dot com
2013-12-09 14:02 ` jengelh at inai dot de
2013-12-09 14:05 ` jengelh at inai dot de
2014-04-04 11:37 ` ktietz at gcc dot gnu.org
2014-05-02 15:58 ` law at redhat dot com
2014-05-02 15:59 ` law at redhat dot com
2014-05-02 16:03 ` ktietz at gcc dot gnu.org
2014-05-02 16:09 ` law at redhat dot com
2014-05-02 16:14 ` ktietz at gcc dot gnu.org
2014-05-02 16:33 ` law at redhat dot com
2014-05-02 16:42 ` ktietz at gcc dot gnu.org
2014-05-05  8:48 ` rguenth at gcc dot gnu.org [this message]
2014-05-30  8:33 ` ktietz at gcc dot gnu.org
2021-07-25  0:22 ` [Bug tree-optimization/59429] " pinskia at gcc dot gnu.org
2024-01-24 16:56 ` pinskia at gcc dot gnu.org
2024-03-22 10:55 ` redbeard0531 at gmail dot com

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=bug-59429-4-ALNZpcJfrY@http.gcc.gnu.org/bugzilla/ \
    --to=gcc-bugzilla@gcc.gnu.org \
    --cc=gcc-bugs@gcc.gnu.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 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).