Re: 3.2 && 3.3 && 3.4 REGRESSION: PR 9479

[Greg Schafer <gschafer at zip dot com dot au>]

On Wed, Jan 29, 2003 at 04:24:12AM -0500, Jakub Jelinek wrote:
> On Wed, Jan 29, 2003 at 02:43:41AM +0100, Gabriel Dos_Reis wrote:
> > | On Tue, Jan 28, 2003 at 10:12:22AM +0100, Gabriel Dos_Reis wrote:
> > | > 
> > | > I uploaded 3.2.2 pre-release tarballs on the FTP server as
> > | > 
> > | >     ftp://gcc.gnu.org/pub/gnu/snapshots/gcc-*3.2.2-20030128.*
> > | > 
> > | > Please donwload and test them.  Report problems to me and fill GNATS
> > | > PRs. 
> > | 
> > | Hi
> > | 
> > | We've discovered a release critical bug. Gcc does not correctly detect the
> > | latest FSF release of binutils-2.13.2.1 which is quite an embarrasment.
> > | 
> > | The GNATS PR is 9479
> > 
> > I would like to have a binutils expert review the proposed patch in 
> > RP 9479.
> 
> The patch is correct, I even tested it on a bunch of ld --version strings
> ld used to print.
> Just note that while major.minor.patch.subpatch should be parsed into
> non-empty ld_vers, version with 5 numbers separated by dots
> should not (that's hjl's numbering and say 2.12.90.0.1 was still broken).

Thanks Jakub.

So to clarify, hj's releases have a date string which takes preference over
the version string, so the example with 5 numbers doesn't matter at this
point. It will only become an issue if the FSF release ever makes a 5
numbered release.

Gabriel, please can you include this patch in the 3.2.2 release?

Thanks
Greg