This is the mail archive of the
libstdc++@gcc.gnu.org
mailing list for the libstdc++ project.
Re: gcc-3.0.3: libstdc++/5045 and /3720 unsuffciently resolved?
- From: Paolo Carlini <pcarlini at unitus dot it>
- To: Markus Werle <numerical dot simulation at web dot de>
- Cc: libstdc++ at gcc dot gnu dot org
- Date: Tue, 05 Feb 2002 15:22:32 +0100
- Subject: Re: gcc-3.0.3: libstdc++/5045 and /3720 unsuffciently resolved?
- References: <3C5FE385.3C9704E6@web.de>
Markus Werle wrote:
> Hi folks!
>
> The problems described at
> http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&pr=3720&database=gcc
> and
> http://gcc.gnu.org/ml/libstdc++/2001-12/msg00158.html
> seem not to be resolved with gcc-3.0.3 on i686-pc-linux-gnu and hppa-2.0w
Hi,
I cannot reproduce your problems on i686-pc-linux-gnu. Indeed both 3.0.3 and 3.1 20020205 compile the testcase and behave this way:
3.1 20020205
------------
paolo:~> a.out
1.2345678901234567890123456789012345678901234567890123456
value: 1.23457
3.0.3
-----
paolo:~> a.out
1.2345678901234567890123456789012345678901234567890123456
value: -1.99827
(it is *well* known that the issue is only partially solved for 3.0.x in order *not* to break ABI compatibility. Notice however, that the failbit is set)
Indeed:
3.0.3
-----
paolo:~> a.out
1.23456789012345678901234567
value: 1.23457
If breaking the ABI compatibility is not a problem for you, why don't you test the just released 3.0.96 for 3.0.x ??
Ciao,
Paolo.