I have been working through some gcc-3.4 libstdc++-v3 testsuite failures on cygwin.
The tests:
- 26_numerics/c99_classification_macros_c++.cc
- 26_numerics/c99_classification_macros_c.cc
fail because C99 and C++ disagree over fpclassify()
The following patch to /usr/include/math.h does the job,
I have IM installed. Busy right now but I will have a look at it tonight
and report back.
-Original Message-
From: Bill McCormick [mailto:[EMAIL PROTECTED]
Sent: Monday, 21 July 2003 3:39 PM
To: Cygwin
Subject: ImageMagick
Hi all,
Anybody have ImageMagick installed?
I'm having instal
I can build and test gcc HEAD with 1.5.0 and latest (well yesterday's latest)
test packages. The testsuite results are similar to 1.3, but I haven't
checked the output line by line.
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problem
ImageMagick-5.5.3 builds out of the box for me with shared libraries, and I have
been using it for months without problem.
I did re-autotool it before I built, but I am not sure if that is required
libtoolize --copy --force && \
aclocal && \
automake && \
autoconf
I then configured wit
You can build gnat on cygwin. See
- http://gcc.gnu.org/ml/gcc/2002-06/msg00161.html
- http://sources.redhat.com/ml/cygwin/2002-09/msg01212.html
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cy
I cannot bootstrap gcc-3.3 with binutils-20021107-1. The problem
goes away if I revert to binutils-20020706-2, and reoccurs
if I reinstall binutils-20021107-1.
The error that occurs is a Signal 11 running gengtype.exe
gcc -g -DIN_GCC -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-p
This is a libtool problem. If you do not find a more sophisticated
fix you might try renaming or deleting
/usr/lib/gcc-lib/i686-pc-cygwin/3.2/../../../libstdc++.la.
-Original Message-
From: Braden McDaniel [mailto:braden.n.mcdaniel@;saic.com]
Sent: Thursday, 14 November 2002 10:59 AM
To
You do need a gnat compiler, as found http://www.geocities.com/billingd2002/gnat/
Works fine for gcc-3.2 but not for gcc-3.3 (gcc PR 6942)
-Original Message-
From: Robert McNulty Junior [mailto:[EMAIL PROTECTED]]
Sent: Thursday, 6 February 2003 4:05 PM
To: Dockeen; [EMAIL PROTECTED]
Subjec
I just tried this too. Same result. I have been chasing (without success) some
libstdc++-v3 testsuite failures that are similar. That discusion is on the libstdc++
mailing list. The advice was to try recompiling libtsdc++-v3 with CXXFLAGS="-g -O0"
and use the debugger. This got me a little
There has been some discussion of this on [EMAIL PROTECTED] recently. Check the
archives at http://mail.gnu.org/pipermail/gcl-devel/ for the details.
working from memory here, so check this, but:
- you will get better results on cygwin and mingw using the CVS version
2.50 from savannah.gcc.
The file texi2html, in the tetex-beta package, starts
#!/usr/local/bin/perl
Of course, it should be
#!/usr/bin/perl
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ:
Tim Prince wrote:
My dejagnu runs reasonably well, although I haven't figured out
how David Billinghurst was able to bring his up to date.
dejagnu-1.4.2 builds OOTB for me.
- download from any gnu mirror
- build out of source tree with
srcdir/configure --prefix=/usr
make
make in
conds).
mkdir dejagnu-1.4.2
cd dejagnu-1.4.2
/usr/local/src/dejagnu-1.4.2/configure --prefix=/usr
make
make install
-Original Message-
From: Tim Prince [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, 3 April 2002 8:05
To: Billinghurst, David (CRTS)
Cc: [EMAIL PROTECTED]
Subjec
> From: Michael A Chase [mailto:[EMAIL PROTECTED]]
> Sent: Wednesday, 27 March 2002 8:49
> To: Alexei Lioubimov; [EMAIL PROTECTED]
> Subject: Re: setup 2.194.2.21: Questions about the "per-mirror directories" feature
> I've written a Perl script
> (http://home.ix.netcom.com/~mchase/zip/clean_set
with mingw.
-Original Message-
From: Dershowitz, Bill [mailto:[EMAIL PROTECTED]]
Sent: Friday, 3 May 2002 2:20
To: Billinghurst, David (CRTS)
Subject: Macsyma
Dear David:
I understand from the Cygwin site that you have compiled Macsyma. Is it
possible to obtain the binary for Macsyma from you (or
hives,
which I will obviously have to re-read in the next little while.
-Original Message-
From: Tom Tromey [mailto:[EMAIL PROTECTED]]
Sent: Saturday, 4 May 2002 6:31
To: Billinghurst, David (CRTS)
Cc: [EMAIL PROTECTED]
Subject: Re: cygwin failures - assertion "!(addr &
Danny,
Can you elaborate on this. Now would be a good time to get
dwarf2 EH working on cygwin. It would be painful to release
a cygwin gcc-3.1 with sjlj exceptions, then switch.
I don't mind hacking away at the gcc build process, but I
get confused if I get too deep into the internals.
It can be done. I have a working (or so it seems) GNAT for cygwin.
Let me know if you are interested.
-Original Message-
From: David A. Cobb [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, 18 September 2002 3:21
To: Dockeen
Cc: [EMAIL PROTECTED]
Subject: Re: New gcc-3.1.1 Release - g77
B
se it was there".
-Original Message-
From: David A. Cobb [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, 18 September 2002 9:32
To: Billinghurst, David (CRTS); [EMAIL PROTECTED]
Subject: GNAT in gcc-3.2 [was Re: New gcc-3.1.1 Release - g77]
Billinghurst, David (CRTS) wrote:
>It can be done. I h
I have build an ada compiler on cygwin using cygwin
branch of gcc-3.2. The original bootstrap was done
a few months ago using a native PC version of GNAT.
I forget the details but they were posted to the list
back then. I have no real interest in ada, but do use
it for testing gcc.
The only
You wrote that you had problems compiling octave-2.1.35 on cygwin, due to a
missing definition of _rl_clear_screen.
The problem is that _rl_clear_screen is "private" and is not exported in
cygreadline.dll. As you and I both found out, simply removing the leading
underscore is the "Wrong Thing".
Jan,
I have successfully installed your texmf packages using setup. I just put
your files in my existing downloaded files and edited my existing setup.ini
to add the texmf-* hints.
Simple tests for tex, latex and dvips work. I will have a play over the
weekend and see what I can break.
-
Cygwin has recently acquired the ualarm function. The function is declared
in as
useconds_t _EXFUN(ualarm, (useconds_t __useconds));
This doesn't match the The Single UNIX Specification Version 2 at
http://www.opengroup.org/onlinepubs/007908799/xsh/ualarm.html, which wants
useconds_t ualarm(
nds)
unsigned int seconds;
{
@@ -616,6 +617,7 @@
}
return remaining;
}
+#endif
#
-
-Original Message-
From: Dan Stanger [mailto:[EMAIL PROTECTED]]
Sent: Monday, 10 December 2001 2:44
To: Billi
Thanks. This works nicely with gcc-3.1 (which has had broken nls
support for a while)
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Bug reporting: http://cygwin.com/bugs.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com
To install gcc-3 into /usr/local
install dejagnu-1.4.2 if you want to run the g++ testsuite
if you want to build objc, install patch
http://gcc.gnu.org/ml/gcc-bugs/2002-01/msg00578.html
cd to somewhere out of the gcc source tree
mkdir ${build}
cd ${build}
${srcdir}/configure --enable-threads=pos
Mack,
A couple of things to try:
Unpack the gcc source tree using cygwin tar and gzip/bzip2. Some windows utilities
are "smart" enough to convert \n to \r\n
Try using binary mounts for the source and build trees.
Use the configure options I posted earlier.
-Original Message-
From:
The fastjar install problem was fixed in gcc-3.1.
zlib, fastjar and gcj are built on the mainline to prevent bitrot. One could make a
case for disabling them for a release.
-Original Message-
From: Alexandre Oliva [mailto:[EMAIL PROTECTED]]
Sent: Monday, 4 February 2002 10:59
To: [E
cygwin cvs (on win2k) copes with this ok. It happens with gcc and maxima sources. It
just overwrites on file and complains.
-Original Message-
From: Charles Wilson [mailto:[EMAIL PROTECTED]]
Sent: Thursday, 7 February 2002 12:56
To: Volker Quetschke
Cc: [EMAIL PROTECTED]
Subject: Re:
29 matches
Mail list logo
