AIX 3 declared wcwidth() in the wrong header file. This is no longer
an issue with newer versions of AIX.
2020-12-20 Bruno Haible
Remove support for broken in AIX 3.
* m4/wcwidth.m4 (gl_FUNC_WCWIDTH): Simplify test programs.
diff --git a/m4/wcwidth.m4 b/m4/wcwidth.m4
index 2
In many files we have written
/* Tru64 with Desktop Toolkit C has a bug: must be included before
.
BSD/OS 4.0.1 has a bug: , and must be
included before . */
#include
#include
#include
#include
But Tru64 = OSF/1 is not supported by Gnulib (since 2019), since its latest
release wa
A gcc warning uncovered this mistake:
2020-12-20 Bruno Haible
isnanf: Fix autoconf test.
* m4/isnanf.m4 (gl_ISNANF_WORKS): Fix one of the NaN tests.
diff --git a/m4/isnanf.m4 b/m4/isnanf.m4
index 3863aa1..c040bfe 100644
--- a/m4/isnanf.m4
+++ b/m4/isnanf.m4
@@ -1,4 +1,4 @@
-#
We don't need to support IRIX 4 and 5 any more, since IRIX 6 was released
in 1994 [1]. This patch removes some of the support for these systems,
where it is clear that it has no impact on IRIX 6. Namely, IRIX 5.3 and 6.5
have getmntent() in libc.
[1] https://en.wikipedia.org/wiki/IRIX#History
20
With the recent changes proposed for glibc I'm reasonably confident its
'free' will be fixed in 2.33, so I installed the attached further patch into
Gnulib. Doing the check with AC_COMPILE_IFELSE should improve the quality of
the guesses when cross-compiling. (If I'm wrong about 2.33 we can chan
The AC_STRUCT_ST_DM_MODE macro caught my eye: its name starts with AC_ but
it is not part of Autoconf.
Looking deeper, I find that a web search for
"checking for st_dm_mode in struct stat... yes"
finds 0 hits. That's because st_dm_mode existed only on Cray [1]. We missed
it when removing Cray su
On Sun, Dec 20, 2020 at 12:09 PM Zack Weinberg wrote:
> I'm not happy about needing to kludge backward compatibility with the
> older std-gnu11.m4 into autoconf 2.70.1 but I'm going to do it.
It occurs to me that this would be less of a problem in the future if
aclocal/autoreconf could update M
On Sun, 20 Dec 2020 at 16:46, Bruno Haible wrote:
> This patch is already in Gnulib since 2020-12-09. But when people
> run 'autoreconf' on an existing released tarball, they are effectively
> combining an older Gnulib with a newest Autoconf.
>
> Why do people do that? The point of tarballs is th
On Sun, Dec 20, 2020 at 11:54 AM Ross Burton wrote:
> On Sun, 20 Dec 2020 at 16:46, Bruno Haible wrote:
>> This patch is already in Gnulib since 2020-12-09. But when people
>> run 'autoreconf' on an existing released tarball, they are effectively
>> combining an older Gnulib with a newest Autocon
On Sun, Dec 20, 2020 at 11:46 AM Bruno Haible wrote:
>
> Zack Weinberg wrote:
> > $ diff -u m4/std-gnu11.m4{~,}
> > --- m4/std-gnu11.m4~2020-08-30 11:27:01.0 -0400
> > +++ m4/std-gnu11.m42020-12-20 09:43:13.001477099 -0500
> > @@ -27,6 +27,8 @@
> > # François Pinard, Karl Berry, R
Zack Weinberg wrote:
> $ diff -u m4/std-gnu11.m4{~,}
> --- m4/std-gnu11.m4~2020-08-30 11:27:01.0 -0400
> +++ m4/std-gnu11.m42020-12-20 09:43:13.001477099 -0500
> @@ -27,6 +27,8 @@
> # François Pinard, Karl Berry, Richard Pixley, Ian Lance Taylor,
> # Roland McGrath, Noah Friedman,
Bruno Haible wrote:
Oh, I see. So, you mean native Windows.
Fixed:
Wonderful, thank you.
On Wed, Dec 16, 2020 at 2:02 PM Zack Weinberg wrote:
> On Wed, Dec 16, 2020 at 1:52 PM Ross Burton wrote:
> > | checking for gcc option to enable C11 features... none needed
> > | ../bison-3.7.4/configure: line 6187: syntax error near unexpected
> > token `ac_cv_prog_cc_stdc=$ac_cv_prog_cc_c89'
Adrian Ebeling wrote:
> I'm talking about systems that have *neither* pathconf() nor fpathconf().
Oh, I see. So, you mean native Windows.
Fixed:
2020-12-20 Bruno Haible
backupfile: Fix compilation error on native Windows (regr. 2018-10-23).
Reported by Adrian Ebeling in
> 2020-12-20 Bruno Haible
>
> float: Fix compilation error when gnulib's float.h exists twice.
> Reported by Santiago Vila .
> * lib/float.in.h (GNULIB_defined_long_double_union): New macro.
Oops, that was only half of the necessary patch. Part 2 is:
diff --git a/lib/float.
Santiago Vila reported a build of gettext-0.21, which failed due to gnulib's
generated float.h:
/bin/bash ../libtool --tag=CC --mode=compile gcc
-DLOCALEDIR=\"/usr/share/locale\" -DBISON_LOCALEDIR=\"/usr/share/locale\"
-DLOCALE_ALIAS_PATH=\"/usr/share/locale\" -DUSEJAVA=0
-DGETTEXTJAR=\"/usr
16 matches
Mail list logo
