61913 – Building gcc3.3 with gcc3.4 - Need more intelligent CFLAGS handling

Bug 61913 - Building gcc3.3 with gcc3.4 - Need more intelligent CFLAGS handling

Summary: Building gcc3.3 with gcc3.4 - Need more intelligent CFLAGS handling

Status:	RESOLVED WONTFIX

Alias:	None

Product:	Gentoo Linux
Classification:	Unclassified
Component:	[OLD] Core system (show other bugs)
Hardware:	x86 Linux

Importance:	High enhancement (vote)
Assignee:	Gentoo Toolchain Maintainers

URL:
Whiteboard:
Keywords:

Depends on:
Blocks:

Reported:	2004-08-27 05:22 UTC by Stuart Shelton
Modified:	2004-09-18 21:50 UTC (History)
CC List:	0 users

See Also:
Package list:
Runtime testing required:	---

Attachments
Add an attachment (proposed patch, testcase, etc.)

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

Description Stuart Shelton 2004-08-27 05:22:11 UTC

I've switched to the gcc34-x86-2004.2 profile to use gcc3.4.1-r2, which (finally) supports -march=pentium-m.

Building gcc3.3 on this is interesting - gcc compiles, but then fails when it attempts to recompile itself, because (obviously) pentium-m is not a valid architecture.

This can be easily solved by altering the ebuild to replace pentium-m with pentium4 (pentium3?) - but going forward, would it not be better to define, say, GCC295CFLAGS, GCC33CFLAGS, and GCC34CFLAGS (or similar) in make.conf - allowing the correct values to be chosen on a per-compiler basis.

Any thoughts?  Is this a valid enhancement?

Comment 1 Stuart Shelton 2004-08-27 07:16:14 UTC

Actually, scratch that - the eBuild isn't trivially modifiable afterall :(

It does, of course, build fine by switching to gcc33 with gcc-config - but it'd still be nice to be able to do it from gcc34...

Comment 2 Travis Tilley (RETIRED) gentoo-dev

2004-09-18 21:50:07 UTC

...