Re: Ada bugs

Re: Ada bugs

[Robert Dewar]

Actually, let me check on this one, I am afraid I may be mixing it up
with another report.

Re: Ada bugs

[Arnaud Charlet]

Just one small typo:

>    Please submit bug a report, see http://gcc.gnu.org/bugs.html.

Please submit a bug report, see ...

Arno

Re: Ada bugs

[Robert Dewar]

sorry for missing context on previous message. Geert said

<<Right now, I'm testing a patch changing the bug reporting instructions
produced by GNAT. Text will read:

   Please submit bug a report, see http://gcc.gnu.org/bugs.html.
   Include the entire contents of this bug box in the report.
   Include the exact gcc or gnatmake command that you entered.
   Also include sources listed below in gnatchop format
   concatenated together with no headers between files.

OK?
>>

And I was replying "sounds fine to me" in this context.

Re: Ada bugs

[Robert Dewar]

sounds fine to me

Re: Ada bugs

[Gerald Pfeifer]

On Wed, 1 May 2002, Florian Weimer wrote:
> Looks good.

Yes, thanks.

> I'm going to commit my last proposed change for bugs.thml, but I hope
> to rewrite this document to follow the three-stage process soon.

Would be great to have this before 3.1 is released, thanks!

Gerald
-- 
Gerald "Jerry" pfeifer@dbai.tuwien.ac.at http://www.dbai.tuwien.ac.at/~pfeifer/

Re: Ada bugs

[Florian Weimer]

Geert Bosch <bosch@gnat.com> writes:

> Right now, I'm testing a patch changing the bug reporting instructions
> produced by GNAT. Text will read:
>
>    Please submit bug a report, see http://gcc.gnu.org/bugs.html.
>    Include the entire contents of this bug box in the report.
>    Include the exact gcc or gnatmake command that you entered.
>    Also include sources listed below in gnatchop format
>    concatenated together with no headers between files.
>
> OK?

Looks good.  I'm going to commit my last proposed change for
bugs.thml, but I hope to rewrite this document to follow the
three-stage process soon.

Re: Ada bugs

[Geert Bosch]

On Wednesday, May 1, 2002, at 05:45 , Gerald Pfeifer wrote:

> On Tue, 30 Apr 2002, Florian Weimer wrote:
>>> It is perfectly reasonable to label the GNAT part of the release "work
>>> in progress".
>> Where could we put such kind of information?
>
> wwwdocs/htdocs/gcc-3.1/changes.html has already has subsection for the
> other front ends; just go ahead and add Ada there. ;-)

Right now, I'm testing a patch changing the bug reporting instructions
produced by GNAT. Text will read:

   Please submit bug a report, see http://gcc.gnu.org/bugs.html.
   Include the entire contents of this bug box in the report.
   Include the exact gcc or gnatmake command that you entered.
   Also include sources listed below in gnatchop format
   concatenated together with no headers between files.

OK?

Re: Ada bugs

[Gerald Pfeifer]

On Tue, 30 Apr 2002, Florian Weimer wrote:
>> It is perfectly reasonable to label the GNAT part of the release "work
>> in progress".
> Where could we put such kind of information?

wwwdocs/htdocs/gcc-3.1/changes.html has already has subsection for the
other front ends; just go ahead and add Ada there. ;-)

Gerald
-- 
Gerald "Jerry" pfeifer@dbai.tuwien.ac.at http://www.dbai.tuwien.ac.at/~pfeifer/

Re: Ada bugs

[Florian Weimer]

dewar@gnat.com (Robert Dewar) writes:

> It is perfectly reasonable to label the GNAT part of the release "work
> in progress".

Where could we put such kind of information?

Re: Ada bugs

[Robert Dewar]

It is perfectly reasonable to label the GNAT part of the release "work
in progress".

Re: Ada bugs

[Florian Weimer]

dewar@gnat.com (Robert Dewar) writes:

> We were not planning on a big Ada merge from the mailine to the branch before
> the release, since we consider that this will diminish stability. But we
> should and will check into the issue of reversions.

I think we should clearly label the GNAT part of the release as "work
in progress".  People will soon complain that the following things are
missing or not usable with GNAT 5.0 in GCC 3.1:

        * ASIS (I know, hardly anybody needs it ;-)
        * GLADE
        * FLORIST
        * GDB (no debugging *at all*, -gstabs might cure it)
        * GtkAda (if I recall correctly)

From a FSF point of view, these aren't regressions because this is the
first GNAT release, but former users of the public ACT version might
have different expectations.

Re: Ada bugs

[Robert Dewar]

<<It will then be necessary to fix on the branch various issues already
fixed on the mainline, such as the bug box (only partially fixed on the
mainline) and the "For information on GLADE, contact Ada Core
Technologies" and similar comments in the manual.  (Simply merging the
mainline manual isn't necessarily the right solution; the branch manual
should describe the version of the code that is on the branch, not some
later or earlier version.)
>>

Right, the bug box in particular is in the list of issues identified that
must be fixed.

Re: Ada bugs

[Joseph S. Myers]

On Wed, 17 Apr 2002, Robert Dewar wrote:

> We were not planning on a big Ada merge from the mailine to the branch before
> the release, since we consider that this will diminish stability. But we

It will then be necessary to fix on the branch various issues already
fixed on the mainline, such as the bug box (only partially fixed on the
mainline) and the "For information on GLADE, contact Ada Core
Technologies" and similar comments in the manual.  (Simply merging the
mainline manual isn't necessarily the right solution; the branch manual
should describe the version of the code that is on the branch, not some
later or earlier version.)

-- 
Joseph S. Myers
jsm28@cam.ac.uk

Re: Ada bugs

[Robert Dewar]

We were not planning on a big Ada merge from the mailine to the branch before
the release, since we consider that this will diminish stability. But we
should and will check into the issue of reversions.

Re: Ada bugs

[Joseph S. Myers]

If there is to be an Ada merge from mainline to branch before the release
(as part of fixing any of these bugs), note that some patches reverted in
the last big merge to mainline still need to be unreverted.  (There isn't
currently a PR open for these.)  The list I sent at
http://gcc.gnu.org/ml/gcc-patches/2002-03/msg00402.html may not be
complete - the person who did the merge should make a proper check of all
patches from outside ACT in the relevant timeframe that might have been
affected.  (The unreversion at least for the last patch ought to be done
by someone who understands Ada well enough to be sure that the patch
reverted is still applicable and doesn't need changes to account for other
changes in the front end.)

-- 
Joseph S. Myers
jsm28@cam.ac.uk

Re: Ada bugs

[Robert Dewar]

Note that the patch for PR5903 is already there. All that is needed is for
someone to copy it over. Florian, you are most welcome to make this change,
otherwise we will try to do it in the next day or two.

Re: Ada bugs

[Mark Mitchell]

--On Wednesday, April 17, 2002 11:54:37 AM +0200 Florian Weimer 
<fw@deneb.enyo.de> wrote:

>
>
> IMHO, PR5903 should be fixed before the release of GCC 3.1, too.
>
> (On a second thought, however, it would be fun to write a security
> advisory: a buffer overflow bug in the GNU Ada run-time library. ;-)

Time is running out.  We're already beyond the original release date, so
if we're going to fix these things it needs to get done right away.
Since we have no previous Ada release to benchmark against, none of these
things are regressions, but they're still  important, so please make an
effort to deal with them ASAP.

Thanks,

--
Mark Mitchell                   mark@codesourcery.com
CodeSourcery, LLC               http://www.codesourcery.com

Re: Ada bugs

[Robert Dewar]

Sorry, please ignore my previous comments on PR5903. This should indeed
be fixed before the release, and we will make sure this is done.

Re: Ada bugs

[Robert Dewar]

> IMHO, PR5903 should be fixed before the release of GCC 3.1, too.

I think it is fine for Florian to go ahead and fix this before the 3.1
release (it's not something that we can get very excited about, but
since Florian thinks it is critical, it makes sense for him to fix it!)

So Florian, if you want to go ahead and fix this, by all means do so.

Re: Ada bugs

[Florian Weimer]

Mark Mitchell <mark@codesourcery.com> writes:

> They are in GNATS, and have the followin PRs:
>
>   PR5907
>   PR5904
>   PR5856

IMHO, PR5903 should be fixed before the release of GCC 3.1, too.

(On a second thought, however, it would be fun to write a security
advisory: a buffer overflow bug in the GNU Ada run-time library. ;-)

Re: Ada bugs

[Robert Dewar]

We will investigate status of these reports and recommend appropriate action.

Ada bugs

[Mark Mitchell]

Robert --

There are three Ada bugs that are marked as high-priority issues
for the GCC 3.1 release.  Would you please look at these issues and
try to resolve them, or ask someone else to do so?

They are in GNATS, and have the followin PRs:

  PR5907
  PR5904
  PR5856

Thanks,

--
Mark Mitchell                   mark@codesourcery.com
CodeSourcery, LLC               http://www.codesourcery.com