public inbox for gcc-prs@sourceware.org
help / color / mirror / Atom feed
From: Nathan Myers <ncm@itasoftware.com>
To: gcc-gnats@gcc.gnu.org
Cc: jimrees@itasoftware.com
Subject: c++/6546: Optimization regression
Date: Thu, 02 May 2002 22:26:00 -0000	[thread overview]
Message-ID: <E173VZl-0003LQ-00@ipc> (raw)


>Number:         6546
>Category:       c++
>Synopsis:       static const members optimization failure
>Confidential:   no
>Severity:       non-critical
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          pessimizes-code
>Submitter-Id:   net
>Arrival-Date:   Thu May 02 22:26:01 PDT 2002
>Closed-Date:
>Last-Modified:
>Originator:     Nathan Myers
>Release:        3.1 20020422 (prerelease)
>Organization:
ITA Software
>Environment:
>Description:
  The compiler fails to elide loading the value of a static const class 
  member initialized in the class definition when passing it to a function 
  that takes a const& to that type.  This is a regression from 2.95.

  In particular, it fails to generate a compare-immediate operation,
  instead loading the value from main memory.

>How-To-Repeat:
  struct Foo { static const int maxSegments = 10; };
  template< typename T > 
    inline const T& amin(const T& t1, const T& t2) { return t2 < t1 ? t2 : t1; }
  template< typename T > 
    inline T bmin(const T& t1, const T& t2) { return t2 < t1 ? t2 : t1; }
  
  int do_something(int argument)
  {
    return amin(argument, Foo::maxSegments) + bmin(argument, Foo::maxSegments);
  }
  const int Foo::maxSegments;

  $ gcc -S -O3 bug.cc 

>Fix:

>Release-Note:
>Audit-Trail:
>Unformatted:
 >System: Linux ipc 2.4.19-pre4 #3 SMP Fri Mar 29 02:57:10 EST 2002 i686 unknown
 >Architecture: i686
 >host: i686-pc-linux-gnu
 >build: i686-pc-linux-gnu
 >target: i686-pc-linux-gnu
 >configured with: ../gcc-3.1/configure --enable-languages=c++ --enable-threads=posix --prefix=/ncm/gnu-3.1 --enable-concept-checks


                 reply	other threads:[~2002-05-03  5:26 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=E173VZl-0003LQ-00@ipc \
    --to=ncm@itasoftware.com \
    --cc=gcc-gnats@gcc.gnu.org \
    --cc=jimrees@itasoftware.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).