From: Zack Weinberg <zack@codesourcery.com>
To: Bruce Korb <bkorb@pacbell.net>
Cc: gcc@gcc.gnu.org
Subject: Re: module level flags
Date: Sat, 28 Sep 2002 20:56:00 -0000 [thread overview]
Message-ID: <20020929010601.GB1288@codesourcery.com> (raw)
In-Reply-To: <3D9642A2.77470B36@pacbell.net>
On Sat, Sep 28, 2002 at 05:00:34PM -0700, Bruce Korb wrote:
> Zack Weinberg wrote:
> >
> > On Sat, Sep 28, 2002 at 04:46:21PM -0700, Bruce Korb wrote:
>
> > If you've found an optimizer bug in 3.[12] not present in 3.0 and
> > prior, you could have the courtesy to *tell us what it is* so we can
> > *fix* it.
>
> OK: :-) Meanwhile, I have determined it was caused by failing
> to recognize a valid alias.
I'm afraid the compiler is working as designed.
> YYSTYPE def,
> YYSTYPE list )
> {
> + tDefEntry* ret = (tDefEntry*)list;
> tDefEntry* pDef = (tDefEntry*)def;
> - tDefEntry* pScn = (tDefEntry*)list;
> - tDefEntry** ppT = (tDefEntry**)&list;
> + tDefEntry* pScn = ret;
> + tDefEntry** ppT = &ret;
In the original code, the type-based aliasing rules permit the
compiler to assume that assignment to ppT does not modify list
EVEN THOUGH IT IS OBVIOUS TO A HUMAN THAT IT DOES.
Your fix is the right way to correct your code. By immediately
assigning list to a temporary with its true type, you show the
compiler what's really going on.
zw
next prev parent reply other threads:[~2002-09-29 1:06 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2002-09-28 16:41 Bruce Korb
2002-09-28 16:52 ` Tim Prince
2002-09-28 16:53 ` Bruce Korb
2002-09-28 18:06 ` Zack Weinberg
2002-09-28 18:20 ` Bruce Korb
2002-09-28 20:56 ` Zack Weinberg [this message]
2002-09-28 22:08 ` Bruce Korb
2002-09-29 15:46 ` Geoff Keating
2002-09-29 3:20 ` Robert Dewar
2002-09-29 11:55 ` Zack Weinberg
2002-09-29 13:10 ` Bruce Korb
2002-09-29 13:59 ` Zack Weinberg
2002-09-29 17:21 ` Tom Tromey
2002-09-29 6:40 Robert Dewar
2002-09-29 10:05 ` Bruce Korb
2002-09-29 10:17 ` Gabriel Dos Reis
2002-09-29 10:45 ` Bruce Korb
2002-09-29 11:36 ` Gabriel Dos Reis
2002-09-29 13:06 ` Daniel Jacobowitz
2002-09-29 13:08 ` Bruce Korb
2002-09-29 13:21 ` Daniel Jacobowitz
2002-09-29 10:49 ` Russ Allbery
2002-09-30 21:33 ` Alexandre Oliva
2002-09-30 21:42 ` Bruce Korb
2002-09-29 10:31 Robert Dewar
2002-09-29 10:45 Robert Dewar
2002-09-29 13:36 Robert Dewar
2002-09-29 13:38 ` Bruce Korb
2002-09-29 14:05 ` Russ Allbery
2002-09-29 13:38 Robert Dewar
2002-09-29 13:42 Robert Dewar
2002-09-29 14:52 ` Russ Allbery
2002-09-29 14:03 Robert Dewar
[not found] <orn0py67ug.fsf@free.redhat.lsd.ic.unicamp.br>
2002-10-01 11:33 ` Joe Buck
2002-10-01 12:16 ` Geoff Keating
2002-10-05 21:03 ` Kai Henningsen
2002-10-06 6:36 ` Gabriel Dos Reis
2002-10-06 8:25 ` Nathan Sidwell
2002-10-06 8:29 ` Gabriel Dos Reis
2002-10-01 12:29 ` Bruce Korb
2002-10-01 14:14 ` Alexandre Oliva
2002-10-01 14:25 ` Alexandre Oliva
2002-10-03 16:46 ` Joe Buck
2002-10-03 17:34 ` Zack Weinberg
2002-10-03 18:07 ` Richard Henderson
2002-10-01 13:58 Robert Dewar
2002-10-01 14:54 ` Joe Buck
2002-10-01 15:40 ` Michael Matz
2002-10-05 11:36 ` Toon Moene
2002-10-02 0:53 Robert Dewar
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=20020929010601.GB1288@codesourcery.com \
--to=zack@codesourcery.com \
--cc=bkorb@pacbell.net \
--cc=gcc@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).