Bug 161791 - Package builds unconditionally as -march=pentium4
Summary: Package builds unconditionally as -march=pentium4
Keywords:
Status: CLOSED INSUFFICIENT_DATA
Alias: None
Product: Red Hat Enterprise Linux 4
Classification: Red Hat
Component: gmp
Version: 4.0
Hardware: i386
OS: Linux
medium
medium
Target Milestone: ---
: ---
Assignee: Thomas Woerner
QA Contact:
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2005-06-27 13:27 UTC by Steve Snyder
Modified: 2007-11-30 22:07 UTC (History)
0 users

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2006-08-01 14:37:27 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)

Description Steve Snyder 2005-06-27 13:27:57 UTC
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.8) Gecko/20050520 Fedora/1.7.8-1.icc

Description of problem:
The gmp-4.1.4-3 package unconditionally builds with 'CFLAGS="-O2 -g -march=pentium4"'.  This creates the potential for the compiler to generate instruction that are not supported on the current system.  Seen on a fully-updated Pentium3 machine.


Version-Release number of selected component (if applicable):
gmp-4.1.4-3

How reproducible:
Always

Steps to Reproduce:
1.Rebuild binary packages or edit gmp.spec.
2.Notice compiler is building with Pentium4 instruction set.
3.
  

Actual Results:  Rebuilt on P3 machine, saw -march=pentium4 on compiler command line.  Yikes!

Expected Results:  Build should respect RPM_OPT_FLAGS.

Additional info:

Comment 1 Steve Snyder 2006-01-26 16:24:57 UTC
Seven months and this bug is still New?  I would have guessed that the
generation of potentially illegal instructions merited more attention.



Comment 2 Thomas Woerner 2006-01-26 16:46:45 UTC
Please give an example, when it was built as pentium4?

BTW: Here is an excerpt of the spec file:

# Important for %{ix86}:
# This rpm has to be build on a CPU with sse2 support like Pentium 4 !

The sse2 DSOs gets built additionally to the i386 DSOs.


Note You need to log in before you can comment on or make changes to this bug.