This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Fwd: Patch: RTL loop opt bug

From: Dale Johannesen <dalej at apple dot com>
To: "gcc-patches at gcc dot gnu dot org Patches" <gcc-patches at gcc dot gnu dot org>
Cc: Dale Johannesen <dalej at apple dot com>
Date: Thu, 29 Apr 2004 16:55:44 -0700
Subject: Fwd: Patch: RTL loop opt bug

Accidentally hit Send before attaching ChangeLog, sorry.

2004-04-29 Dale Johannesen <dalej@apple.com>

        * loop.c (loop_givs_reduce): Avoid miscompilation of
        loops entered at bottom.

This fixes a loop optimization bug I encountered running equake (SPEC) on the LNO branch. While I don't know of any way to get a FE to generate the failing RTL pattern (see patch) in mainline, I believe it is pretty obviously wrong there also, and that's where I'd like to fix it. Bootstrapped and tested on darwin.

Attachment: diff1.txt
Description: Text document

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]