On 3/14/21 4:53 AM, Bruno Haible wrote:
On NetBSD, tzalloc() is in libc, and tzalloc("\\") returns NULL.
On other platforms, tzalloc() comes from Gnulib, and tzalloc("\\") returns
non-NULL.
Which behaviour is correct?
Both. The set of supported TZ values is system-dependent.
this. Please try the attached patch, which I
installed into Gnulib.
>From fc6d7d850bdebfed81e9212910f44edf99dd7743 Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Wed, 10 Mar 2021 15:04:55 -0800
Subject: [PATCH] libc-config: port to DragonFlyBSD 5.9
DragonFlyBSD defines __nonnull incompati
On 3/9/21 7:38 PM, Nick Bowler wrote:
The fix is simple; why not fix it?
In this case heirloom-sh is not an actual porting target but is merely a
tool being used to find portability bugs, and apparently it's generating
a false alarm, and I'd rather avoid code churn due to problems that
On 3/9/21 6:45 PM, Nick Bowler wrote:
I routinely use heirloom-sh to suss out shell portability
Does your project also have a problem on Solaris 10 because its /bin/sh
also doesn't grok "if !"?
If so, how/why does that happen? ./configure is supposed to re-exec with
a working shell, and
On 3/9/21 6:29 PM, Nick Bowler wrote:
This syntax is not portable in practice, as discussed in the Autoconf
manual[1].
This shouldn't be a issue in configure scripts, since they should
arrange to use a good shell. If that isn't happening for some reason I'd
rather fix the underlying problem
On 3/7/21 1:58 AM, Bruno Haible wrote:
The mbrtowc modules states:
Link:
$(LIB_MBRTOWC)
Thanks for explaining that. Over time Gnulib has evolved and 'tar'
hadn't caught up with all those Link: lines. So I just now added
$(LIB_ACL), $(LIB_GETRANDOM), $(LIB_HARD_LOCALE), $(LIB_HAS_ACL),
:00 2001
From: Paul Eggert
Date: Sun, 7 Mar 2021 13:47:02 -0800
Subject: [PATCH] manywarnings: modernize documentation
Problem reported by Reuben Thomas in:
https://lists.gnu.org/r/bug-gnulib/2021-03/msg00046.html
* doc/manywarnings.texi (manywarnings): Update in
the light of recent chan
On 1/14/21 2:00 AM, Florian Weimer wrote:
We saw commercial demand for Shift-JIS much later than that.
Yes, I see Red Hat has a product enhancement for that, for coreutils,
dated 2019.[1]
Shift-JIS is still very much alive and kicking in Japan in its own
circles. Even DBCS EBCDIC is still
On 2/21/21 10:41 AM, Bruno Haible wrote:
it seems that some functions have a 'gl_' prefix and some
don't. Does this matter? Maybe it does not matter because
the contents of lib/malloc/dynarray.h is irrelevant and only
lib/malloc/dynarray-skeleton.c matters?
The idea was that the gl_* symbols
On 3/2/21 7:27 PM, Bruno Haible wrote:
a general macro won't cut it e.g. for coreutils. coreutils
has multithreaded programs ('sort') next to single-threaded programs.
But coreutils wants to optimize wcwidth. GNULIB_WCHAR_SINGLE actually
means "assume that the locale has been set before the
special case for licensing. I installed the attached 1-byte patch to
work around the issue.
>From 487b9551b63ef936a6be6df38d1c9484cd97810c Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Sat, 6 Mar 2021 08:23:48 -0800
Subject: [PATCH] parse-datetime2: fix licensing
Problem reported by Bruno Hai
On 3/3/21 5:58 PM, Jeffrey Walton wrote:
I believe Debian collects statistics on program usage. Maybe you
should look at TinyCC usage before you decide to spend a lot of time
on it.
Probably not worth our time as being too far from the GNU target.
Before reading your email I installed a
c defines __attribute__(x) to empty for compilers
like tcc that do not define __GNUC__, but tcc's alignof relies on
__attribute__. I installed the attached patch to try to work around that
problem.
From 7955dc93ef02c0cb90a8c4966abf142a3813ff48 Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date:
I left them alone for now.
---
ChangeLog | 15 +++
lib/mbtowc-lock.h | 10 +-
2 files changed, 24 insertions(+), 1 deletion(-)
diff --git a/ChangeLog b/ChangeLog
index 0304f6958..7ba436925 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,18 @@
+2021-03-02 Paul Eggert
/parse-datetime.y | 122 +++
3 files changed, 102 insertions(+), 46 deletions(-)
diff --git a/ChangeLog b/ChangeLog
index 9bc60f88e..f4bf4fced 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,24 @@
+2021-03-01 Paul Eggert
+
+ parse-datetime2: new
+++--
lib/unlocked-io.h | 26 +-
3 files changed, 51 insertions(+), 27 deletions(-)
diff --git a/ChangeLog b/ChangeLog
index f4bf4fced..c13c6816d 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,16 @@
2021-03-01 Paul Eggert
+ unlocked-io: do not redefine
494fea035166a4947db72bd23161faa1f1733708 Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Wed, 24 Feb 2021 10:52:59 -0800
Subject: [PATCH 1/2] glob: include libc-config.h in a more-standard way
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Inspired by Tom Tromey’s report for RHEL 6
On 2/21/21 10:20 AM, Bruno Haible wrote:
it sounds like a reasonable security measure
to revoke the write access for users who have been inactive for a
certain time, say 4 years.
That sounds reasonable, for people inactive on the GNU project. However,
Sergey (for example) has contributed to
attempted to work around it by
installing the attached patch into Gnulib; please give it a try.
From 4b5bc7fdd55da390f486bafa1e91ba93dd8d887f Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Tue, 16 Feb 2021 11:41:19 -0800
Subject: [PATCH] Port better to macOS Mojave
MIME-Version: 1.0
Content-
On 2/9/21 5:21 PM, Eric Wong wrote:
I know the (C)Ruby VM uses it internally (or eventfd on Linux);
I think CPython's GVL does, too.
GNU Make does too. It's where I learned of the trick.
On 2/9/21 11:42 AM, Adhemerval Zanella wrote:
Trying to sync gnulib with glibc code, this patch trigger some regression
on glibc testcases:
Thanks for letting me know. That's odd, as I didn't find any regressions
when I synced. I do know of remaining bugs and will add these to my list.
* lib/regexec.c (update_regs): Move comment.
---
ChangeLog | 3 +++
lib/regexec.c | 2 +-
2 files changed, 4 insertions(+), 1 deletion(-)
diff --git a/ChangeLog b/ChangeLog
index d838ad0d0..70e357884 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,8 @@
2021-02-05 Paul Eggert
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,9 @@
2021-02-05 Paul Eggert
+ regex-tests: fix typo
+ * tests/test-regex.c (main): Fix typo that would have caused an
+ old test case to report incorrect values on failure.
+
regex: make it easier to merge into glibc
@@
2021-02-05 Paul Eggert
+ regex: make it easier to merge into glibc
+ * lib/regex_internal.h [_LIBC]: Do not include Gnulib’s dynarray.h.
+
regex: minor refactoring
* lib/regexec.c (proceed_next_node): Use more-local decls.
diff --git a/lib/regex_internal.h b
/ChangeLog
index bd7d1fa16..69abc35dc 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,10 @@
2021-02-05 Paul Eggert
+ regex: debug check for set member duplicates
+ * lib/regex_internal.c (re_node_set_insert): Add a DEBUG_ASSERT
+ that would have caught some recently-fixed
+1,8 @@
2021-02-05 Paul Eggert
+ regex-tests: add bug 11053 test
+ * tests/test-regex.c (main): New test case for glibc bug 11053.
+
regex: debug check for set member duplicates
* lib/regex_internal.c (re_node_set_insert): Add a DEBUG_ASSERT
that would have
+1,8 @@
+2021-02-05 Paul Eggert
+
+ regex: improve comments
+ * lib/regexec.c: Add and correct comments about return values.
+
2021-01-31 Bruno Haible
relocatable-prog-wrapper: Tweak today's patch.
diff --git a/lib/regexec.c b/lib/regexec.c
index f7b4f9cfc..fa413dfd5
exec.c | 26 +++
tests/test-regex.c | 113 -
3 files changed, 141 insertions(+), 11 deletions(-)
diff --git a/ChangeLog b/ChangeLog
index 74304474b..bd7d1fa16 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,18 @@
2021-02-05 Paul
/regexec.c | 30 ++
2 files changed, 20 insertions(+), 16 deletions(-)
diff --git a/ChangeLog b/ChangeLog
index fef04a89c..6fcd5819f 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,11 @@
2021-02-05 Paul Eggert
+ regex: avoid undefined behavior
+ * lib
/ChangeLog
index f4583e0af..74304474b 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,10 @@
2021-02-05 Paul Eggert
+ regex: avoid duplicate in espilon closure
+ * lib/regcomp.c (calc_eclosure_iter): Insert NODE into epsilon
+ closure first rather than last. Otherwise
-05 Paul Eggert
+ regex: minor refactoring
+ * lib/regexec.c (proceed_next_node): Use more-local decls.
+
regex: avoid undefined behavior
* lib/regexec.c (pop_fail_stack): If the stack is empty, return -1
instead of indulging in undefined behavior
On 1/25/21 7:55 PM, Jeffrey Walton wrote:
Jeffrey Walton wrote:
Placing flags in CC breaks other scripts. For example, this no longer works:
if [[ ! $(command -v "${CC}") ]]; then
echo "The compiler is not valid. Please install a compiler."
exit 1
fi
You can
On 1/25/21 1:25 PM, Peter Dyballa wrote:
The configure script of wget 1.21.1 seems to have a bug, which becomes visible
at least on old Mac OS X versions which use old GCC 4.2, see this
report:https://trac.macports.org/ticket/62134. It's my assumption that the
tests for C99 and C11
/ChangeLog
@@ -1,3 +1,14 @@
+2021-01-23 Paul Eggert
+
+ libc-config: port to Xcode 7
+ Problem reported by Mattias Engdegård in:
+ https://lists.gnu.org/r/emacs-devel/2021-01/msg01089.html
+ * lib/cdefs.h (__nonnull): If already defined but glibc is not in
+ use
On 1/22/21 1:00 PM, Brian Leung wrote:
Tomohiro Matsui wrote a patch to support lookaround assertions in regexps some
time ago
(http://emacs.1067599.n8.nabble.com/Patch-for-lookaround-assertion-in-regexp-td121057.html#a23863009),
but it never got merged.
Is there still interest in something
thing at a time.
From a0ddbbd97816c52141c85deb4105a749703ce0ba Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Thu, 21 Jan 2021 13:40:16 -0800
Subject: [PATCH] Port FALLTHROUGH to clang 3.4 and earlier
Problem reported by Akim Demaille in:
https://lists.gnu.org/r/bug-gnulib/2021-01/msg00241.html
On 1/20/21 3:27 AM, Adhemerval Zanella wrote:
#if defined __STDC_VERSION__ && 201710L < __STDC_VERSION__
# define __attribute_fallthrough__ [[__fallthrough__]]
#elif __GNUC_PREREQ (7, 0) || __glibc_has_attribute (__fallthrough__)
# define __attribute_fallthrough__ __attribute__
9fe86ec087654893087a83290ee7 Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Tue, 19 Jan 2021 18:35:30 -0800
Subject: [PATCH] fnmatch, regex, fts: FALLTHROUGH consistency
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Be more consistent about how FALLTHROUG
On 1/19/21 10:22 AM, Akim Demaille wrote:
Clang 3.3 and 3.4 cannot compile the new scratch-buffer module. On Bison's CI
(Clang 3.4: https://api.travis-ci.org/v3/job/755133481/log.txt), there is:
...
CC lib/lib_libbison_a-canonicalize.o
In file included from ../lib/canonicalize.c:31:
On 1/17/21 8:04 AM, Bruno Haible wrote:
The tests in the 'for' loop fail for i = 5 ... 10. That is, something
with the ".." handling is not working as expected.
Do you have access to a Cygwin machine for investigation?
Unfortunately not; I'm really not a Cygwin guy.
OK with me if we simply
On 1/15/21 12:55 AM, Bruno Haible wrote:
Would compiling the gnulib
part with options for fewer warnings be OK with you?
Not only is it OK, it's routine. Coreutils, Emacs, etc. do it.
On 1/14/21 3:44 AM, Adhemerval Zanella wrote:
I think we still should fix BZ#14185 with the suggested with the strategy of
falling back to non wide mode in case of encoding error.
For glibc 2.33 that's likely the best we can do.
The full fix will take some time, since it is pretty much a
dex 2123974c8..8991d6a35 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,12 @@
+2021-01-13 Paul Eggert
+
+ verify: simplify static_assert configuration
+ * lib/verify.h (_GL_HAVE__STATIC_ASSERT, _GL_HAVE__STATIC_ASSERT1):
+ Do not define for C++. This should be good enough nowadays,
+ si
attached (I haven't installed it).
From ce90a689fc388c4c612861054adcb09bd3658b1c Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Wed, 13 Jan 2021 15:46:33 -0800
Subject: [PATCH] verify: rely more on __cpp_static_assert
* lib/verify.h (_GL_HAVE__STATIC_ASSERT, _GL_HAVE__STATIC_ASSERT1):
Simplify
for later patches and that we can
switch to them as needed, but I wanted to focus on the actual problem
first rather than worrying about scratch buffers vs dynarrays.
From 7bd5987cf0ed6668d34a921866c94b112594e714 Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Tue, 12 Jan 2021 17:24:17 -0800
On 1/5/21 5:07 AM, Adhemerval Zanella wrote:
It seems that gnulib has added the proposed fix with
aed23714d60d91b2abc74be33635c32ddc5132b5 (done in 2005) and just recently
with a glibc merge with 67306f600fe6a3bcf3fbb6d8bf4b8953b74a8fb7 (done in
2020 to sync back glibc changes) it has fallback
2001
From: Paul Eggert
Date: Mon, 11 Jan 2021 16:46:12 -0800
Subject: [PATCH] tempname: consume less entropy
Derived from a glibc patch proposed by Adhemerval Zanella in:
https://sourceware.org/pipermail/libc-alpha/2021-January/121302.html
* lib/tempname.c: Include stdbool.h.
(random_bits)
problem, I guess. I installed the attached patch; please
give it a try.
>From 077ffc1e416a6be980dd45979547201e572962f6 Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Sat, 9 Jan 2021 14:17:32 -0800
Subject: [PATCH] snippet/_Noreturn: port to pedantic clang
Problem reported by Joe Nel
it. I installed the attached hack into Gnulib to
work around the problem. This should work with glibc as well.
>From 1b50e7d4d63fee04dbe06d4a58f3061c569fede8 Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Fri, 8 Jan 2021 19:50:16 -0800
Subject: [PATCH] =?UTF-8?q?dynarray:=20work=20even=20if
.
Thanks, I installed that part of the proposal into Gnulib by installing
the attached. The idea is for tempname.c to be identical after we get
the abovementioned issue worked out.
From b0ebaf83a49fe4a895a78ddf5b0c4a029e34c566 Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Fri, 8 Jan 2021 17:54:30
with the wrong API.
I installed the attached patch into Gnulib, which should do the change
right for both Gnulib and glibc, the idea being that the Gnulib and
glibc source files can be identical.
From 8eb0e08598b9815f0e85ce1bde676169920087ef Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Fri, 8 Jan 2021
On 1/4/21 4:52 AM, Adhemerval Zanella wrote:
The extra comparison might avoid the scratch_buffer resize that will
fail (since malloc will fail to try allocate PTRDIFF_MAX object), but
it will be used only when such objects are provided (which depending
of the system should not happen).
As you
This patch looks good for glibc, once the previous two regex patches are
done. I installed it into Gnulib by applying the attached, so that
regex_internal.h can stay in lock-step between Gnulib and glibc.
>From 87ed1f9c4bc4ae9e98e1293465a5b3a929b9c3f9 Mon Sep 17 00:00:00 2001
From: Paul Egg
025e89118912adaa309374201c9012f6fb46d583 Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Fri, 8 Jan 2021 14:22:15 -0800
Subject: [PATCH] regexec: remove alloca usage in build_trtable
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Prompted by this different
On 1/8/21 1:49 PM, Bruno Haible wrote:
So, the problem is a signed-integer overflow of constants in a dead
(unreachable) code branch.
Yes, I get many warnings like that in Oracle Developer Studio too, with
some of the Gnulib tests. Either ignore the warnings, or don't use
-Woverflow, or (as
improvement to Gnulib by installing the second attached
patch, so you should be able to copy Gnulib regexec.c to glibc without
changing it.
From ae1984c524f03f8e8bb104d3a2b5c533114cf552 Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Fri, 8 Jan 2021 11:44:19 -0800
Subject: [PATCH 1/2] dynarray: new
+++ b/ChangeLog
@@ -1,5 +1,9 @@
2021-01-08 Paul Eggert
+ scratch_buffer: add scratch_buffer_dupfree macro
+ * lib/scratch_buffer.h (__libc_scratch_buffer_dupfree):
+ New macro, needed to support recent changes in this module.
+
regex: remove alloca usage on regex
On 1/6/21 1:55 PM, Simon Josefsson via Gnulib discussion list wrote:
I'm using the patch below in gsasl, what do you think?
Thanks, looks good to me.
On 1/6/21 3:54 PM, Bernhard Voelker wrote:
And instead of adding yet another way to do the same job, shouldn't the
existing code better be
On 1/6/21 10:06 AM, Paul Eggert wrote:
These fallbacks could be concentrated in libc-config.h, rather than
duplicating them in argp, fnmatch, glob, regex, obstack, etc.
It would be helpful if we arranged for argp etc. to use libc-config.h
and so removed the need for those fallbacks in argp.h
@@
+2021-01-06 Paul Eggert
+
+ libc-config: simplify clang __has_* macros
+ cdefs.h sometimes used names like __glibc_clang_has_attribute and
+ sometimes __glibc_has_attribute. There is no need in glibc or
+ Gnulib to distinguish clang from other compilers here, so
On 1/5/21 2:42 PM, Bruno Haible wrote:
2) Another possibility I can think of is to go through all the macros
defined by lib/cdefs.h, and for those which are actually used in Gnulib
(most aren't), provide fallback definitions in lib/libc-config.h.
Although that would solve the scenario you
---
ChangeLog| 1 +
modules/filename | 2 +-
modules/idx | 2 +-
3 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/ChangeLog b/ChangeLog
index 3d4d5f260..e58f9ff9d 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -41,6 +41,7 @@
* config/srclist.txt: filename.h, idx.h,
of
it is simple reindenting, and you can use "git diff -w" to narrow it down.
>From 7fcdb8f9f2f568b48e8b4d7beab344a735cc2b9a Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Tue, 5 Jan 2021 13:12:39 -0800
Subject: [PATCH] libc-config: merge from glibc
Use a better way of keeping glib
a/ChangeLog b/ChangeLog
index 48aeb44ac..83b2cda2d 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,10 @@
+2021-01-05 Paul Eggert
+
+ maint: note syncs to glibc
+ * config/srclist.txt: filename.h, idx.h, scratch_buffer.h and
+ scratch_buffer_dupfree.c are now identical to glibc
On 1/4/21 6:29 PM, Bruno Haible wrote:
I want the files to be identical (eventually), which is why they have
the same include guard.
If they have the same include guard, then
Having the same include guard was merely a (wrong) means to the goal.
The goal is for the files to be identical. They
On 1/4/21 5:06 PM, Bruno Haible wrote:
I see the problem now: glibc's and Gnulib's have
the same include guard!
I want the files to be identical (eventually), which is why they have
the same include guard. Is there some way to arrange for them to be
identical?
It could be that this
can just ask
people to use --disable-gcc-warnings if their GCC is old.
From f88b033174c4fa816d3b146312820725c658f707 Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Sat, 2 Jan 2021 19:07:53 -0800
Subject: [PATCH] maint: work around GCC -Wreturn-local-addr bug
* configure.ac: Do not use -Wreturn
On 1/1/21 6:30 PM, Bernhard Voelker wrote:
As I see the same on Fedora, the issue seems to be upstream, doesn't it?
I see the problem on Fedora 33 as well. It doesn't appear to be a GCC
bug. unistd.h's declaration expands to this:
extern int getgroups (int __size, __gid_t __list[])
hed Gnulib patch, because it contains an
unnecessary comparison to IDX_MAX and an unnecessary cast to idx_t.
From 8f6b9b66be6672bed1045c27e606dd9fcedcf022 Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Fri, 1 Jan 2021 15:54:43 -0800
Subject: [PATCH] canonicalize: remove NARROW_ADDRESSES optimizat
On 1/1/21 2:00 PM, Karl Berry wrote:
In any case ... I think it's (well past) time for me to bow out of doing
these updates. I gather you're dealing with it anyway, so will you take
over the commits?
I can do it, although it's better if multiple people do it. I don't sync
as often as you do;
On 1/1/21 10:40 AM, Jeffrey Walton wrote:
tempname.c: In function 'random_bits':
tempname.c:90: warning: integer constant is too large for 'long' type
tempname.c:90: warning: this decimal constant is unsigned only in ISO C90
tempname.c: In function 'try_tempname_len':
tempname.c:288: error:
diff --git a/ChangeLog b/ChangeLog
index 69ca053df..c4c44e291 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,10 @@
+2021-01-01 Paul Eggert
+
+ libc-config: port to Sun C 5.9
+ * lib/libc-config.h (libc_hidden_proto): Omit ", ..." in macro
+ definition. It’s not needed fo
;gmp" branch since it
contains bugfixes not in the "gmp-6.2" branch, such as this bugfix:
changeset: 18103:b22f374d0d4c
user:Marco Bodrato
date:Sun Oct 25 17:19:53 2020 +0100
summary: mini-gmp.c (mpz_out_str): Return 0 when base is out of
range (spotted by Paul Egg
:00:00 2001
From: Paul Eggert
Date: Thu, 31 Dec 2020 19:29:08 -0800
Subject: [PATCH] filename: change filename.h comment to LGPLv2.1+
* lib/filename.h: Change license notice to match what should be in
glibc, and what modules/filename specifies. This is to simplify
syncing with glibc.
---
Change
On 12/31/20 3:12 PM, Joseph Myers wrote:
Anything used in glibc libraries should be LGPLv2.1+, not GPL.
That was fixed in the most recently-published version of the patch:
https://sourceware.org/pipermail/libc-alpha/2020-December/121178.html
energetic this
evening so I installed the attached.
>From 35589c5a953da353f9cb56984a161c7c91019e4a Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Thu, 31 Dec 2020 19:14:49 -0800
Subject: [PATCH] doc: mention year-0 bugs in Solaris etc.
* doc/posix-functions/gmtime.texi, doc/posix-functi
and the other abovementioned
glitches, so that you should now be able to sync fnmatch from Gnulib
unchanged.
From a7ad4b110fd6b7dde424dceb46c9c09c31cfbe69 Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Thu, 31 Dec 2020 13:35:53 -0800
Subject: [PATCH 2/2] fnmatch: merge from glibc + proposal
Thanks, this patch looks fine.
Thanks, this patch looks fine.
Sep 17 00:00:00 2001
From: Paul Eggert
Date: Thu, 31 Dec 2020 09:05:13 -0800
Subject: [PATCH 1/2] glob: merge proposed glibc changes
This merges the change proposed by Adhemerval Zanella in:
https://sourceware.org/pipermail/libc-alpha/2020-December/121211.html
* lib/glob.c (__lstat64, __stat64
This patch looks good to me, thanks.
On 12/29/20 7:39 PM, Paul Eggert wrote:
Unfortunately that patch didn't correctly treat size-calculation
overflow like other out-of-memory situations. I installed the attached
further patch into Gnulib.
And even that patch mishandled errno on size-calculation overflow,
so I installed
On 12/29/20 5:21 PM, Paul Eggert wrote:
I installed the attached patch into Gnulib to fix the bug in a way I
hope is better.
Unfortunately that patch didn't correctly treat size-calculation
overflow like other out-of-memory situations. I installed the attached
further patch into Gnulib.
>F
to sync this into
glibc without needing a patch like the above.
From b4e94717557545d613bca58a27d4ef698d551ed2 Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Tue, 29 Dec 2020 17:08:11 -0800
Subject: [PATCH] canonicalize: fix ptrdiff_t overflow bug
Problem reported by Adhemerval Zanella in:
https
On 12/28/20 5:59 AM, Adhemerval Zanella wrote:
The faccessat (..., AT_EACCESS) on Linux fallback will either use
__NR_faccessat for __libc_enable_secure or LFS fstatat call. Neither
fallbacks returns EOVERFLOW in failure case.
Argh, I had misread the glibc source code in this area. So I
On 12/28/20 5:59 AM, Adhemerval Zanella wrote:
It sync with gnulib version bbaba6ce5 with the following difference
require fix a glibc build:
--- stdlib/canonicalize.c
+++ lib/canonicalize-lgpl.c
@@ -52,7 +52,6 @@
# endif
# define GCC_LINT 1
# define _GL_ATTRIBUTE_PURE __attribute__
On 12/28/20 3:42 AM, Adhemerval Zanella wrote:
On 24/12/2020 21:27, Paul Eggert wrote:
Also, shouldn't we merge the already-existing Gnulib scratch_buffer changes
into glibc, along with this new change?
Which changes are you referring? Checking against bbaba6ce5 I see all
glibc files
On 12/28/20 9:39 AM, Ron Eggler wrote:
I cannot find any function that is called "rpl_fseeko"
Presumably an earlier "#define fseeko rpl_fseeko" means that when you
see a definition of the fseeko function, it's really defining rpl_fseeko.
--
m4/regex.m4 | 3 +--
modules/regex | 3 +--
4 files changed, 10 insertions(+), 38 deletions(-)
delete mode 100644 m4/glibc21.m4
diff --git a/ChangeLog b/ChangeLog
index 3f0a51d0e..97966b427 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,3 +1,11 @@
+2020-12-27 Paul Eggert
On 12/26/20 4:07 PM, Alfred M. Szmidt wrote:
install-info does not have an replacement, like say egrep/fgrep --
this is how we install a dir entry for a info manual. Removing
install-info would be a regression.
In practice, GNU installation procedures use install-info in the way
that's
On 12/26/20 1:55 PM, Simon Josefsson wrote:
Looking at that list now, the only obvious tool to remove is egrep,
since 'grep -E' subsumes it nowadays.
Yes, I think that should be dropped.
What about 'install-info'? It is quite GNU-specific. Nothing in
gnulib appears to refer to it (except for
When I run this command:
./gnulib-tool --conditional-dependencies --test regex
on Fedora 33, it fails as shown in the attached file. The problem occurs
because --test brings in binary-io for some reason, but doesn't bring in
the fcntl-h module that binary-io depends on.
A simple workaround
On 12/26/20 10:49 AM, Bruno Haible wrote:
It would be nice to reduce that set of tools.
Well, both 'cmp' and 'diff' are part of POSIX:
Even awk is on that list, and it's much fancier than cmp or diff. And I
wouldn't support removing awk either, as it's a handy tool in Makefiles
when the
in glibc's scratch_buffer.h.
>From e0ea25e27e1f1ed70b5dcc2338b9ede365f9a79a Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Fri, 25 Dec 2020 17:26:51 -0800
Subject: [PATCH] idx: change idx.h comment to LGPLv2.1+
* lib/idx.h: Change license notice to match what should be in glibc.
gnulib-t
the bogus warnings that Adhemerval noticed when
building glibc's stdlib/canonicalize.c.
From abc2b3fe5af97e971f10ebdbeee174f6d87f1fe6 Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Thu, 24 Dec 2020 21:33:18 -0800
Subject: [PATCH] canonicalize, canonicalize-lgpl: remove lint
Pacify GCC. Some
led this into Gnulib master on savannah.
From 5186fb6c9e2840ab921418bf73fe2662e200f89d Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Thu, 24 Dec 2020 16:17:44 -0800
Subject: [PATCH] canonicalize: simplify via scratch_buffer_dupfree
* config/srclist.txt: Adjust accordingly.
* lib/canonica
Thanks, this looks good. Also, the following "[PATCH 3/5] Import
filename.h from gnulib" looks good. I merged both changes into Gnulib here:
http://git.savannah.gnu.org/cgit/gnulib.git/commit/?id=1013adf6bf48627f84a9043cf2631ff13d2a452f
+1,14 @@
+2020-12-24 Paul Eggert
+
+ canonicalize-lgpl: merge proposed libc changes
+ This merges the changes proposed for glibc in:
+ https://sourceware.org/pipermail/libc-alpha/2020-December/121085.html
+ https://sourceware.org/pipermail/libc-alpha/2020-December/121086
On 12/24/20 7:16 AM, Adhemerval Zanella wrote:
It sync with gnulib version d9c121346 with the following difference
require fix a glibc build:
As I mentioned today on bug-gnulib I installed some changes into Gnulib
which are related, and it'd probably be better to sync with the latest
on the Gnulib side. I plan
to look at those Glibc patches next
>From 011dcf74d9c3bf6373a8fea1ce524b2d123b4130 Mon Sep 17 00:00:00 2001
From: Paul Eggert
Date: Thu, 24 Dec 2020 11:38:47 -0800
Subject: [PATCH 01/10] canonicalize, canonicalize-lgpl: fix symlink bug
Problem reported by Adhemerval Zane
1301 - 1400 of 5560 matches
Mail list logo
