Re: libstdc++/2644: g++ produces wrong code

Re: libstdc++/2644: g++ produces wrong code

[Phil Edwards]

The following reply was made to PR libstdc++/2644; it has been noted by GNATS.

From: Phil Edwards <pedwards@disaster.jaj.com>
To: Benjamin Kosnik <bkoz@redhat.com>
Cc: gdr@gcc.gnu.org, gcc-gnats@gcc.gnu.org
Subject: Re: libstdc++/2644: g++ produces wrong code
Date: Wed, 2 May 2001 21:23:30 -0400

 On Wed, May 02, 2001 at 07:38:30PM -0700, Benjamin Kosnik wrote:
 > > *sigh* The branch/trunk differences are driving me nuts.  If we just did
 > > a bulk merge, do you think anyone would notice?  :-)
 > 
 > the libsupc++ bits are pretty hard to get around. 
 
 That's the problem with having umpteen thousand customers watching... no
 sneaky stuff permitted during development.
 
 
 -- 
 pedwards at disaster dot jaj dot com  |  pme at sources dot redhat dot com
 devphil at several other less interesting addresses in various dot domains
 The gods do not protect fools.  Fools are protected by more capable fools.

Re: libstdc++/2644: g++ produces wrong code

[Benjamin Kosnik]

The following reply was made to PR libstdc++/2644; it has been noted by GNATS.

From: Benjamin Kosnik <bkoz@redhat.com>
To: Phil Edwards <pedwards@disaster.jaj.com>
Cc: gdr@gcc.gnu.org, gcc-gnats@gcc.gnu.org
Subject: Re: libstdc++/2644: g++ produces wrong code
Date: Wed, 2 May 2001 19:38:30 -0700 (PDT)

 > *sigh* The branch/trunk differences are driving me nuts.  If we just did
 > a bulk merge, do you think anyone would notice?  :-)
 
 the libsupc++ bits are pretty hard to get around. 
 
 -benjamin

Re: libstdc++/2644: g++ produces wrong code

[Phil Edwards]

The following reply was made to PR libstdc++/2644; it has been noted by GNATS.

From: Phil Edwards <pedwards@disaster.jaj.com>
To: bkoz@gcc.gnu.org
Cc: gdr@gcc.gnu.org, gcc-gnats@gcc.gnu.org
Subject: Re: libstdc++/2644: g++ produces wrong code
Date: Wed, 2 May 2001 19:52:58 -0400

 On Wed, May 02, 2001 at 07:16:01PM -0000, bkoz@gcc.gnu.org wrote:
 >      Hey folks, I cannot reproduce this with current CVS gcc. I'm not quite sure what is up, but perhaps this bug was filed before wchar_t support was on by default. Can everybody try with current CVS and see what's up?
 >      here's what I get:
 >      
 >      %COMP.sh "-static" 2644.cc
 >      <bkoz@fillmore> /mnt/hd/bliss/src.gcc/libstdc++-v3/testsuite                    %a.out 
 >      31	wchar_t digits
 >      2147483647	wchar_t max
 >      
 >      Looks right to me.
 
 Ditto for the trunk, but this morning's 3.0 branch produces zeros.
 (The submitter specified 3.0 in the "release" field.)  I thought the
 wchar_t change was on the branch now... *looks*  ugh, there's a bunch of
 wchar-related things in the trunk ChangeLog not mentioned in the branch
 ChangeLog.
 
 *sigh* The branch/trunk differences are driving me nuts.  If we just did
 a bulk merge, do you think anyone would notice?  :-)
 
 
 -- 
 pedwards at disaster dot jaj dot com  |  pme at sources dot redhat dot com
 devphil at several other less interesting addresses in various dot domains
 The gods do not protect fools.  Fools are protected by more capable fools.