public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "frederic.riss at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/48295] Incorrect code generated with dynamic floating point rounding mode switches
Date: Mon, 28 Mar 2011 10:15:00 -0000	[thread overview]
Message-ID: <bug-48295-4-98uUU65aGK@http.gcc.gnu.org/bugzilla/> (raw)
In-Reply-To: <bug-48295-4@http.gcc.gnu.org/bugzilla/>

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48295

--- Comment #3 from Frederic Riss <frederic.riss at gmail dot com> 2011-03-28 09:58:50 UTC ---
In bug #34678 I thought you were agreeing with Joseph that -frounding-math had
a more general meaning than the one you are expressing here, but I get your
point that a more general solution should be implemented.

Regarding the validity of the testcase, I took glibc's implementation of
fesetround that I inlined. I did that because I saw that /usr/include/fenv.h
has provisions for including inline versions of the fenv routines, although
it's not the case yet for x86. So I guess it's possible that someday you get
exactly that situation even though people are using the standard routines to
switch rounding mode. In fact my private port does exactly this.

Can you think of other places than CSE that would exhibit these issues?
Unfortunately, I have to put some band-aids into place for our numerical
analysts to be able to do their jobs.


  parent reply	other threads:[~2011-03-28  9:58 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-03-26 16:56 [Bug tree-optimization/48295] New: " frederic.riss at gmail dot com
2011-03-26 16:57 ` [Bug tree-optimization/48295] " frederic.riss at gmail dot com
2011-03-28  9:40 ` rguenth at gcc dot gnu.org
2011-03-28 10:15 ` frederic.riss at gmail dot com [this message]
2011-03-28 11:09 ` joseph at codesourcery dot com
2011-04-01 19:22 ` pinskia at gcc dot gnu.org

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-48295-4-98uUU65aGK@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).