Bug 56715 - "gcc-2.96-99 " -O2 -mcpu=i686 -ffast-math" produces broken code
Summary: "gcc-2.96-99 " -O2 -mcpu=i686 -ffast-math" produces broken code
Keywords:
Status: CLOSED WORKSFORME
Alias: None
Product: Red Hat Raw Hide
Classification: Retired
Component: gcc
Version: 1.0
Hardware: i686
OS: Linux
medium
medium
Target Milestone: ---
Assignee: Jakub Jelinek
QA Contact:
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2001-11-26 11:14 UTC by ellson
Modified: 2007-04-18 16:38 UTC (History)
0 users

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2004-10-02 00:56:14 UTC
Embargoed:

Attachments (Terms of Use)
source file (570 bytes, text/plain)
2001-11-26 11:16 UTC, ellson 		no flags Details
Makefile (157 bytes, text/plain)
2001-11-26 11:17 UTC, ellson 		no flags Details
View All

Description ellson 2001-11-26 11:14:21 UTC 
Description of Problem:
"-O2 -mcpu=i686 -ffast-math" produces broken code

Version-Release number of selected component (if applicable):
gcc-2.96-99

How Reproducible:
100%


Steps to Reproduce:
1.make attached example 
2. 
3. 

Actual Results:


Expected Results:


Additional Information:
Problem does not exist in gcc-3.0.2 or in egcs-2.91.66
Comment 1 ellson 2001-11-26 11:16:23 UTC 
Created attachment 38545 [details]
source file
Comment 2 ellson 2001-11-26 11:17:35 UTC 
Created attachment 38547 [details]
Makefile
Comment 3 Richard Henderson 2004-10-02 00:56:14 UTC 
I can't reproduce this with gcc 3.2, but I suspect that this is

  http://gcc.gnu.org/PR323

which is not taking into account (allowed) extra computational
precision.
Note You need to log in before you can comment on or make changes to this bug.