GCC 4.0.1 Status (2005-06-27)

GCC 4.0.1 Status (2005-06-27)

[Mark Mitchell]

As stated earlier, the only patches I'm considering for 4.0.1 at present 
are wrong-code cases on primary platforms.  There are several open, but 
the only one I consider a show-stopper is PR 22051, which Jeff Law is 
working on, and hopes to fix Tuesday.  As soon as that's in, I'll build 
RC3, and then, hopefully, a few days later, put out the final release.

I'm sorry this is dragging out, but I think it's worth getting this bug 
fixed.

FYI,

-- 
Mark Mitchell
CodeSourcery, LLC
mark@codesourcery.com
(916) 791-8304

Re: GCC 4.0.1 Status (2005-06-27)

[Scott Robert Ladd]

Mark Mitchell wrote:
> I'm sorry this is dragging out, but I think it's worth getting this bug
> fixed.

No need for apologies; you're doing the "right thing".

..Scott

Re: GCC 4.0.1 Status (2005-06-27)

[Jeffrey A Law]

On Tue, 2005-06-28 at 00:20 -0700, Mark Mitchell wrote:
> As stated earlier, the only patches I'm considering for 4.0.1 at present 
> are wrong-code cases on primary platforms.  There are several open, but 
> the only one I consider a show-stopper is PR 22051, which Jeff Law is 
> working on, and hopes to fix Tuesday.  As soon as that's in, I'll build 
> RC3, and then, hopefully, a few days later, put out the final release.
> 
> I'm sorry this is dragging out, but I think it's worth getting this bug 
> fixed.
I'm working on it right now.    My PA box is experiencing cpu faults
on its second cpu while trying to bootstrap for a set of baseline
results....   I suspect it'll take until late tomorrow before I can get
a set of baseline test results, fix the bug, then get a set of new test
results for comparison purposes.


jeff

Re: GCC 4.0.1 Status (2005-06-27)

[Mark Mitchell]

Jeffrey A Law wrote:
> On Tue, 2005-06-28 at 00:20 -0700, Mark Mitchell wrote:
> 
>>As stated earlier, the only patches I'm considering for 4.0.1 at present 
>>are wrong-code cases on primary platforms.  There are several open, but 
>>the only one I consider a show-stopper is PR 22051, which Jeff Law is 
>>working on, and hopes to fix Tuesday.  As soon as that's in, I'll build 
>>RC3, and then, hopefully, a few days later, put out the final release.
>>
>>I'm sorry this is dragging out, but I think it's worth getting this bug 
>>fixed.
> 
> I'm working on it right now.    My PA box is experiencing cpu faults
> on its second cpu while trying to bootstrap for a set of baseline
> results....   I suspect it'll take until late tomorrow before I can get
> a set of baseline test results, fix the bug, then get a set of new test
> results for comparison purposes.

Perhaps you could get a patch put together, test it by staring 
atassembly output, and then ask for a volunteer to test it?  I expect 
that Joseph could do a test run on PA-HPUX for you.

-- 
Mark Mitchell
CodeSourcery, LLC
mark@codesourcery.com
(916) 791-8304

Re: GCC 4.0.1 Status (2005-06-27)

[Jeffrey A Law]

On Tue, 2005-06-28 at 08:50 -0700, Mark Mitchell wrote:
> Perhaps you could get a patch put together, test it by staring 
> atassembly output, and then ask for a volunteer to test it?  I expect 
> that Joseph could do a test run on PA-HPUX for you.
It might be the best bet.   I'm going to clean out the bloody box
to hopefully improve its airflow and give it one more shot.


jeff

Re: GCC 4.0.1 Status (2005-06-27)

[Joe Buck]

On Tue, Jun 28, 2005 at 10:06:35AM -0600, Jeffrey A Law wrote:
> On Tue, 2005-06-28 at 08:50 -0700, Mark Mitchell wrote:
> > Perhaps you could get a patch put together, test it by staring 
> > atassembly output, and then ask for a volunteer to test it?  I expect 
> > that Joseph could do a test run on PA-HPUX for you.
> It might be the best bet.   I'm going to clean out the bloody box
> to hopefully improve its airflow and give it one more shot.

I can also test hppa-hpux.

Re: GCC 4.0.1 Status (2005-06-27)

[Jeffrey A Law]

On Tue, 2005-06-28 at 08:50 -0700, Mark Mitchell wrote:
> Jeffrey A Law wrote:
> > On Tue, 2005-06-28 at 00:20 -0700, Mark Mitchell wrote:
> > 
> >>As stated earlier, the only patches I'm considering for 4.0.1 at present 
> >>are wrong-code cases on primary platforms.  There are several open, but 
> >>the only one I consider a show-stopper is PR 22051, which Jeff Law is 
> >>working on, and hopes to fix Tuesday.  As soon as that's in, I'll build 
> >>RC3, and then, hopefully, a few days later, put out the final release.
> >>
> >>I'm sorry this is dragging out, but I think it's worth getting this bug 
> >>fixed.
> > 
> > I'm working on it right now.    My PA box is experiencing cpu faults
> > on its second cpu while trying to bootstrap for a set of baseline
> > results....   I suspect it'll take until late tomorrow before I can get
> > a set of baseline test results, fix the bug, then get a set of new test
> > results for comparison purposes.
> 
> Perhaps you could get a patch put together, test it by staring 
> atassembly output, and then ask for a volunteer to test it?  I expect 
> that Joseph could do a test run on PA-HPUX for you.
Someone is going to have to do the PA testing.  The PA box isn't going
to recover.  Oh well.

There's actually two closely related bugs in this code.  In one case
we can eliminate necessary canonicalizations and in others we can 
introduce unwanted canonicalizations.  Whee, fun.

Jeff

Re: GCC 4.0.1 Status (2005-06-27)

[Jeffrey A Law]

On Tue, 2005-06-28 at 08:50 -0700, Mark Mitchell wrote:
> Jeffrey A Law wrote:
> > On Tue, 2005-06-28 at 00:20 -0700, Mark Mitchell wrote:
> > 
> >>As stated earlier, the only patches I'm considering for 4.0.1 at present 
> >>are wrong-code cases on primary platforms.  There are several open, but 
> >>the only one I consider a show-stopper is PR 22051, which Jeff Law is 
> >>working on, and hopes to fix Tuesday.  As soon as that's in, I'll build 
> >>RC3, and then, hopefully, a few days later, put out the final release.
> >>
> >>I'm sorry this is dragging out, but I think it's worth getting this bug 
> >>fixed.
> > 
> > I'm working on it right now.    My PA box is experiencing cpu faults
> > on its second cpu while trying to bootstrap for a set of baseline
> > results....   I suspect it'll take until late tomorrow before I can get
> > a set of baseline test results, fix the bug, then get a set of new test
> > results for comparison purposes.
> 
> Perhaps you could get a patch put together, test it by staring 
> atassembly output, and then ask for a volunteer to test it?  I expect 
> that Joseph could do a test run on PA-HPUX for you.
Eric C was kind enough to point me to a couple HPUX boxes elsewhere
in Red Hat.  I'm prepping a baserun right now and I've got a patch
that I'll test immediately afterwards.

jeff