open high priority bugs

open high priority bugs

[Nathan Sidwell]

Hi,
There were some 22 bugs in the open state with high priority. 
http://gcc.gnu.org/cgi-bin/gnatsweb.pl?database=gcc&cmd=submit%20query&category=all&severity=all&priority=high&responsible=all&state=open&ignoreclosed=Ignore%20Closed&class=all&synopsis=&multitext=&columns=category&columns=state&columns=class&columns=responsible&columns=synopsis&sortby=Responsible&.cgifields=columns&.cgifields=originatedbyme&.cgifields=ignoreclosed

By useage, this condition should be unreachable. Open bugs should first
be analyzed to confirm them, and then possibly made high priority.

I have ruthlessly downgraded those that are obviously not 3.0 related,
leaving 18 remaining.

If your name is on the cc list, then you had some input to the remaining
bugs, please take a look and either
a) downgrade them (or ask me to do it for you)
b) make a case to Mark as to why they should be high priority.

nathan
-- 
Dr Nathan Sidwell   ::   http://www.codesourcery.com   ::   CodeSourcery LLC
         'But that's a lie.' - 'Yes it is. What's your point?'
nathan@codesourcery.com : http://www.cs.bris.ac.uk/~nathan/ : nathan@acm.org

Re: open high priority bugs

[Per Bothner]

Nathan Sidwell <nathan@codesourcery.com> writes:

> If your name is on the cc list, then you had some input to the remaining
> bugs, please take a look and either
> a) downgrade them (or ask me to do it for you)
> b) make a case to Mark as to why they should be high priority.

I submitted 3062 yesterday because the branch no longer builds a working
Kawa when compiled with -O.  This is why I want to plead for this
as high priority:

I am trying to get a long-overdue Kawa release ready, and I want to
try to publicise it better than I have in the past.  The most obvious
"selling point" is that you can now compile Scheme (and Emacs Lisp!)
code to native code using Gcc 3.0.  It could help the visibility of
both Kawa and Gcj.  But if I have to weasel and say it only works if
you compile without optimization, the impact of any such annoucement
is much lessened.

Furthermore, this used to work not that long ago.  I assume the
new exception handling changes broke things.  It is plausible
other code could also mis-compile.  (The test-case isn't that
complicated.)  I've tried to debug the problem, but it is getting
deep into things I don't understand well:  exception handling,
rtl, and optimization.  I don't even know if the bug is actually
an optimizer bug or the optimizer exposes some other bug.  So
I need help getting this solved, preferably from Richard H.
-- 
	--Per Bothner
per@bothner.com   http://www.bothner.com/per/

Re: open high priority bugs

[Mark Mitchell]

>>>>> "Per" == Per Bothner <per@bothner.com> writes:

    Per> I submitted 3062 yesterday because the branch no longer
    Per> builds a working Kawa when compiled with -O.  This is why I
    Per> want to plead for this as high priority:

It's up to the Java team -- I'm still happy to have any Java fixes
that y'all think are important.  I can't really see holding the
release for this, though. 

If a Java person will mark this bug as `assigned' to themselves, that
would be good.

Thanks,

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

Re: open high priority bugs

[Tom Tromey]

>>>>> "Mark" == Mark Mitchell <mark@codesourcery.com> writes:

Mark> If a Java person will mark this bug as `assigned' to themselves,
Mark> that would be good.

The Java person who would normally work on it (Andrew) is on vacation.
Alex, is this something you can fix?  I certainly can't.  If Alex
can't then we will need help.

Tom

Re: open high priority bugs

[Philip Blundell]

>If your name is on the cc list, then you had some input to the remaining
>bugs, please take a look and either
>a) downgrade them (or ask me to do it for you)
>b) make a case to Mark as to why they should be high priority.

I still feel #2878 should be high priority: it's definitely a regression from 
2.95, and it will make GCC 3.0 essentially unusable on ARM GNU/Linux systems.
(I realise this is a minority platform, though, so I wouldn't hold out for 
delaying the release until it's fixed.)

If anybody would like to work on this bug but lacks access to a suitable 
machine, send me mail and I'll create you an account.

p.


-- 
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.5 (GNU/Linux)
Comment: Exmh version 2.1.1 10/15/1999 (debian)

iD8DBQE7IH+PVTLPJe9CT30RAqbaAJwJdBLvxBKJ7XOnQo0EOD1h8WhG/gCfUBgp
tcwyT2/FOJcQKpP0Qfp8714=
=gSOS
-----END PGP SIGNATURE-----

Re: open high priority bugs

[Mark Mitchell]

>>>>> "Philip" == Philip Blundell <philb@gnu.org> writes:

    Philip> I still feel #2878 should be high priority: it's

It is already so marked.  I hope someone will fix it!

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