[Bug fortran/33252] GCC-4.3.0 Bootstrap testsuite error increase

["michelin60 at gmail dot com" <gcc-bugzilla at gcc dot gnu dot org>]

------- Comment #10 from michelin60 at gmail dot com  2007-08-31 04:33 -------
(In reply to comment #7)
> Subject: Re:  GCC-4.3.0 Bootstrap testsuite error increase
> 
> > Second I mentioned Mark Mitchell because he, as relesae manager, put a stop to backporting definitely aggravating productive use of GCC.

> This is the trunk we are talking about,  I am seriously thinking you
> need to understand what that means.  There is no backporting, it is
> And again, this is the trunk we are talking about, not some release
> branch and we are not even in stage 3 yet so there will be issues.

> It was just fixed so update and try again.
> 
> Yes we know about those, if did a quick search, you would find that I
> filed the bug about those before I added those testcase.



Regarding the last quote I am led to believe that Mr. Pinski is taking undue
credit. PR30758 (marked as a duplicate) is the first addressing the
re-appearance of mayalias. there are another 5 PR, all appearing before Mr.
Pinski's filing. 
This reminds me of what happened regarding PR33125 (filed by myself) and
PR33126. Neither made to the bug list; neither has been officially closed.
Anyhow the patch submitted seems to have resolved the issue.

I updated, and I am now bootstrapping with check. I am sure Mr Delisle patch
originally worked and passed tests; but, that some other later patch gave rise
to what I, and seemingly others, saw in PR33252. I took issue with Mr. Pinski's
confusing PR32225 with PR33125 not Mr Delisle volunteer work.

Now to the more confusing issue of backporting. I managed to find the
following:

> IMO, for the Fortran front end, "regressions-only" is an inappropriate
> criterion for this sort of mode, given that the front end does not have
> a long history for the wrong-code and ICE-on-valid bugs to be
> regressions against.

I think that's less true than it used to be.  It's true that gfortran is
new-ish, and I've given the Fortran team a lot of flexibility in past
release cycles.  But, congratulations are in order: gfortran is now a
real Fortran compiler and people really are using it!  But,
"regressions, wrong-code, and ICEs" isn't a bad criteria for this
intermediate lockdown.

I do expect Fortran to honor the regressions-only rule once the 4.3
release branch has been made.

Thanks,
-- 
Mark Mitchell
CodeSourcery
mark@codesourcery.com

Mr Kargl explains what happened from an maintainer's and release manager's view
but does not provide a paliative to the user. G77 relative to g90 is relatively
much more out of date than C1989 to C1994. Anyhow, if memory serves, there was
significant back-porting from 4.3 to 4.2.x to to solve some memory usage
problems. 

Michelin


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=33252