Pádraig Brady wrote:
On 07/27/2012 05:56 PM, Jim Meyering wrote:
...
When I try to trigger such misbehavior, timeout seems to work fine
on a system (Fedora 17) with a working timer_settime function:
$ timeout 2 date --set=$(LC_ALL=C date -d 10\ sec +@%s); echo $?
0
I.e., when
Ondrej Oprala wrote:
Alright, I've redone it so now the hashes create and expect GNU escaping
in both the BSD and GNU output format.
I left the --tag format indication of text files to be a space before
the algo. name
so far, but I can change that if needed.
...
+static const char *const
Pierre-Jean wrote:
I'm trying to sort a file containing accents and numbers,
but can't find a way to do this correctly:
- Without -V option, accents are correctly sorted, but not
numbers.
- With -V option, numbers are correctly sorted, but not
accents.
Here is an example:
echo
A
Now that util-linux has made a v2.22-rc release with improved su[1],
I'd like to make a coreutils release (which removes su altogether).
There's enough new code in md5sum --tag that I will defer
adding that feature until after 8.18.
I've just remembered that I have one more minor tail fix, and
Pierre-Jean wrote:
Jim Meyering j...@meyering.net wrote:
Pierre-Jean wrote:
I'm trying to sort a file containing accents and numbers,
but can't find a way to do this correctly:
The trick is to specify sorting with -f for the first column
and -V for the second. Then it does what you
Jim Meyering wrote:
Now that util-linux has made a v2.22-rc release with improved su[1],
I'd like to make a coreutils release (which removes su altogether).
There's enough new code in md5sum --tag that I will defer
adding that feature until after 8.18.
I've just remembered that I have one
FYI,
From 81806035310c5c16b1b11c88709bf066305ed160 Mon Sep 17 00:00:00 2001
From: Jim Meyering meyer...@redhat.com
Date: Fri, 27 Jul 2012 23:42:05 +0200
Subject: [PATCH] tail: avoid rare error-path FD leak
* src/tail.c (tail_forever): Close FD to avoid leak after a
failed fstat.
---
src/tail.c
Pádraig Brady wrote:
On 08/02/2012 03:52 PM, Jim Meyering wrote:
FYI,
Subject: [PATCH] tail: avoid rare error-path FD leak
* src/tail.c (tail_forever): Close FD to avoid leak after a
failed fstat.
...
static analysis or eagle eyes?
Change looks good.
Thanks.
I noticed it while working
32f7dd1ad64165cdde0c8d90997c2a6445965c00 Mon Sep 17 00:00:00 2001
From: Jim Meyering meyer...@redhat.com
Date: Thu, 2 Aug 2012 19:31:36 +0200
Subject: [PATCH 1/2] split: free a lines/round-robin chunk processing buffer
* src/split.c (lines_rr): Plug a harmless leak.
---
src/split.c | 1 +
1 file changed
All ulimit-requiring tests would be skipped
when running against valgrind-wrapped tools.
This adjusts the heuristic to avoid that:
From 5367b398b0dc203e84e4210c647f4b03ed4dcfe2 Mon Sep 17 00:00:00 2001
From: Jim Meyering meyer...@redhat.com
Date: Thu, 2 Aug 2012 19:12:18 +0200
Subject: [PATCH
an lseek failure on a known-open-and-valid
read-only file descriptor, maybe it doesn't deserve one.
WDYT?
From a714f9a350a3b69aec03ee2a0c74b627d8138542 Mon Sep 17 00:00:00 2001
From: Jim Meyering meyer...@redhat.com
Date: Wed, 25 Jul 2012 18:26:35 +0200
Subject: [PATCH] tail: do not let failed lseek
FYI,
From 0ee727e612ae202e5b47e39062f6604d65a76074 Mon Sep 17 00:00:00 2001
From: Jim Meyering meyer...@redhat.com
Date: Sat, 4 Aug 2012 11:02:40 +0200
Subject: [PATCH] truncate: don't leak a file descriptor with --ref=PIPE
* src/truncate.c (main): For a user who makes the mistake of
using a non
Pádraig Brady wrote:
On 08/03/2012 01:53 PM, Jim Meyering wrote:
I ran make check on valgrind-wrapped tools (took most of a day on a
6/12-core system) and saw split at the top of the list of definitely lost
buffer leaks. So far, the only ones I've investigated are not important,
but at least
Pádraig Brady wrote:
On 08/04/2012 10:06 AM, Jim Meyering wrote:
...
Subject: [PATCH] truncate: don't leak a file descriptor with --ref=PIPE
* src/truncate.c (main): For a user who makes the mistake of
using a non-seekable file as a reference for the desired length,
truncate would open
Pádraig Brady wrote:
On 08/04/2012 11:26 AM, Jim Meyering wrote:
Pádraig Brady wrote:
On 08/04/2012 10:06 AM, Jim Meyering wrote:
...
Subject: [PATCH] truncate: don't leak a file descriptor with --ref=PIPE
* src/truncate.c (main): For a user who makes the mistake of
using a non-seekable
Pádraig Brady wrote:
On 08/04/2012 12:11 AM, Bernhard Voelker wrote:
...
Subject: [PATCH] df: fix exit code and ensure printing unprocessed msg when
totaling
When the combination of the file system options with given files or
devices does not lead to output, df --total would exit
Pádraig Brady wrote:
Pushed with Jim's improvement.
http://git.sv.gnu.org/gitweb/?p=coreutils.git;a=commit;h=3b8139e
I see you adjusted the test and fixed this for me:
- if (exit_status == 0 ! file_systems_processed)
+ if (exit_status == EXIT_SUCCESS ! file_systems_processed)
Thanks!
The printf-surprise test failed on FreeBSD 9.0 not because of
a printf problem per se, but because the cost of running printf
there has just surpassed our semi-arbitrary ulimit of 10,000KiB.
Here's the fix:
From c3b777e07d175713745a162d60acd11af427e159 Mon Sep 17 00:00:00 2001
From: Jim Meyering
Bernhard Voelker wrote:
On 08/07/2012 12:59 PM, Ondrej Oprala wrote:
Hi,
here's the new patch. Everything should be in order now.
As the new variable explicit_no_preserve is only used for
the mode, I find its name is a bit confusing (because it's
too general).
Good point.
Ondřej, please
and error messages with --total
Bruno Haible (1):
doc: clarify meaning of '-parodd' in stty help
Erik Auerswald (1):
doc: mention uniq(1) in sort(1) man-page and vice versa
Jim Meyering (38):
maint: post-release administrivia
ls: color each symlink-to-relative-name
Bernhard Voelker wrote:
On 08/08/2012 05:43 PM, Pádraig Brady wrote:
+require_mount_list_()
+{
+ df /dev/null 21 ||
+skip_ Unable to read the list of mounted file systems.
+}
+
Hmm, I think this is not enough: df can also fail due to other reasons,
e.g. when a mount point is not
Pádraig Brady wrote:
On 08/08/2012 05:07 PM, Jim Meyering wrote:
Bernhard Voelker wrote:
On 08/08/2012 05:43 PM, Pádraig Brady wrote:
+require_mount_list_()
+{
+ df /dev/null 21 ||
+skip_ Unable to read the list of mounted file systems.
+}
+
Hmm, I think this is not enough: df
Ondrej Oprala wrote:
Hi, I've changed the du util a bit to reflect
the proposed solution here
http://marc.info/?l=coreutils-bugm=134124896022426w=2 .
Thanks for working on this!
I'm not sure about the XDEV checking though. Shouldn't the matching
st_dev numbers be enough to know both files
the beginning) and relatively obscure bugs.
See the NEWS below for a brief summary.
Thanks to everyone who has contributed!
The following people contributed changes to this release:
Andrew D Warshall (1)
Bernhard Voelker (4)
Bruno Haible (1)
Erik Auerswald (1)
Jim Meyering (39)
Joachim
Bernhard Voelker wrote:
On 08/09/2012 01:53 PM, Bernhard Voelker wrote:
Here's a proposed patch.
Oops, sorry, one of the cases in the new test failed.
I fixed it:
-LD_PRELOAD=./k.so df --total || fail=1
+LD_PRELOAD=./k.so df --total fail=1
Additionally, as this patch didn't make it into
Bernhard Voelker wrote:
On 08/14/2012 10:51 AM, Jim Meyering wrote:
Here is the patch that I expect to squash in.
Most are nits, but there was an exploitable bit
of (debugging?) code in the new test.
diff --git a/NEWS b/NEWS
index 0f63ddf..46d0a41 100644
--- a/NEWS
+++ b/NEWS
@@ -2,8
Ondrej Oprala wrote:
...
Subject: [PATCH] tests: Add error checking for root-only tests running
setuidgid
* NEWS: Mention the fix.
* tests/init.cfg: Modify the require_root_ function to check
for setuidgid calls and proper permissions.
---
NEWS | 6 ++
tests/init.cfg | 19
Krzysztof Goj wrote:
Please add a NEWS entry, a description of the new option
in coreutils.texi and adjust for the issues below
[...]
Thanks for the review. I've updated the patch, it's attached.
Thanks for contributing.
I adjusted the wording and punctuation in the commit log and
NEWS and
Ondrej Oprala wrote:
here's the new patch...I hope it's all alright now.
And sorry about not mentioning Bernhard in the first place,
it just didn't occur to me.
...
Subject: [PATCH] tests: Add error checking for root-only tests running
setuidgid
* NEWS: Mention the test improvement.
*
From: Jim Meyering meyer...@redhat.com
* tests/Coreutils.pm (_compare_files): Reverse diff arguments so
that we invoke diff -c $expected $actual, which is consistent with
how init.sh-using tests invoke compare exp out.
---
tests/Coreutils.pm | 6 +++---
1 file changed, 3 insertions(+), 3
Here are three tiny changes:
[PATCH 1/3] maint: fix comment grammar to placate make syntax-check
[PATCH 2/3] maint: correct a stale comment in sort.c
[PATCH 3/3] tests: reverse args in Coreutils.pm-invoked diff, for consistency
src/remove.c |2 +-
src/sort.c |2 +-
Jim Meyering wrote:
Ondrej Oprala wrote:
From d8eab7abdf19b668065ff8be3700c9966eefdc39 Mon Sep 17 00:00:00 2001
From: Ondrej Oprala oopr...@redhat.com
Date: Thu, 9 Aug 2012 17:34:09 +0200
Subject: [PATCH] du: Fix an issue with bogus warnings on bind-mounted
directories
* NEWS: Mention
Jim Meyering (7):
maint: post-release administrivia
maint: fix comment grammar to placate make syntax-check
maint: correct a stale comment in sort.c
tests: reverse args in Coreutils.pm-invoked diff, for consistency
sort: sort --unique (-u) could cause data loss
I saw this test fail when run via make -j2 on a low-memory arm-based system,
but not when run in isolation.
From e44943a25e4a225568b51a411f18ba61134d3eb3 Mon Sep 17 00:00:00 2001
From: Jim Meyering meyer...@redhat.com
Date: Sun, 19 Aug 2012 17:34:04 +0200
Subject: [PATCH] tests: avoid split
Jim Meyering wrote:
sort -u data loss... who would have guessed?
This is not your typical coreutils bug.
Actually, there were *two* sort -u bugs.
How many more are lurking?
Anyhow, here's a snapshot.
Please give this a try, and report any problems.
I will probably release coreutils-8.19
changes to this release:
Bernhard Voelker (1)
Jim Meyering (9)
Krzysztof Goj (1)
Paul Eggert (2)
Special thanks to Rasmus Borup Hansen for reporting the sort -u problem
along with a reproducer, and to Paul Eggert for spotting the second bug
and for his minimal patch that fixed both bugs
Jim Meyering wrote:
Jim Meyering wrote:
Ondrej Oprala wrote:
From d8eab7abdf19b668065ff8be3700c9966eefdc39 Mon Sep 17 00:00:00 2001
From: Ondrej Oprala oopr...@redhat.com
Date: Thu, 9 Aug 2012 17:34:09 +0200
...
From 491914627db04fe789bfdbf796f9be21198a6b09 Mon Sep 17 00:00:00 2001
From
Pádraig Brady wrote:
On 08/18/2012 10:57 PM, Mike Frysinger wrote:
When statically linking pthreads with rt, the current order is:
-lpthread -lrt
But when statically linking, the -lpthread will be discarded as it
isn't used. It needs to come after the -lrt.
* m4/timer_time.m4
Paul Eggert wrote:
On 08/17/2012 01:31 PM, Jim Meyering wrote:
So we definitely have a *second* bug here.
Yes, I noticed. It definitely counts as a double-ouch.
I'm glad the bug report prompted us to read this code
more carefully.
My latest patch should fix both bugs.
Thanks again
[Re: http://bugs.gnu.org/11844
du: continue processing after bind-mount induced dir-cycle ]
Ondrej Oprala wrote:
Sure, that looks great. Thanks for the help.
I added this to the log:
* THANKS.in: Update.
This implements the proposal in http://bugs.gnu.org/11844.
Originally
Bernhard Voelker wrote:
I noticed (too late) that the new tests/du/bind-mount-dir-cycle
calls print_ver_ with rm as argument instead of du.
A little script found other wrong uses, too.
...
Subject: [PATCH] tests: correct print_ver_ arguments
Some tests have been derived from other tests
Bernhard Voelker wrote:
On 08/22/2012 09:57 AM, Jim Meyering wrote:
Bernhard Voelker wrote:
Subject: [PATCH] tests: correct print_ver_ arguments
Nice!
Thanks.
What do you think about adding a syntax-check to run that script?
Never done before, but below is what seems to work - together
Ondrej Oprala wrote:
Hi, I haven't heard from this thread since I posted the last patch,
are there still things to correct or should I consider it closed?
There are still things to fix.
For example, in your new test,
+for i in 'a\b' 'a\' a$'\n'b a$'\t'b; do
+ : $i
+ md5sum --tag $i
Nguyễn Thái Ngọc Duy wrote:
* src/df.c (get_header): Mark %s-%s for translation. Give
translators some hint what it is for.
Thank you!
Pushed with minor tweaks.
Stefano Lattarini wrote:
[NOTE: I'm sending this to the main coreutils for lack of a dedicated
coreutils-patch list; I hope it's OK!]
OK, these are the easy pieces actually, but we have to start somewhere.
This series consists of:
- some preparatory refactorings;
- the merging
Stefano Lattarini wrote:
On 08/30/2012 01:14 AM, Jim Meyering wrote:
Thanks for all the work.
Would you please mail to me (or to the list) the concatenation of those
patches
Sorry to be dense, but: what does concatenation mean in this conext
exactly?
The output of git format-patch -22
Pádraig Brady wrote:
Thanks for the detailed rationale, however
the existing env variables are significant to more utils than sort(1).
I.E. they're generally system level settings, rather than command level.
Also sort -S is very portable, even though not standardised.
solaris' sort(1) has -S
Nick Alcock wrote:
This test requires an unwritable input and an unreadable output,
so will fail if the testsuite is being run as root, and should
not be run in this case.
---
tests/misc/sort-exit-early | 2 ++
1 file changed, 2 insertions(+)
diff --git a/tests/misc/sort-exit-early
Stefano Lattarini wrote:
On 08/30/2012 01:14 AM, Jim Meyering wrote:
Thanks for all the work.
Would you please mail to me (or to the list) the concatenation of those
patches
Thanks again. I've begun looking at these changes, so here's
some preliminary feedback.
As expected, the build now
I noticed a lot of Signed-off-by: lines in Stefano's series.
This makes our no-S-o-b policy more apparent:
From 4f76f9b42ad71df40f1f714f2ecfb1e4b6cd7078 Mon Sep 17 00:00:00 2001
From: Jim Meyering meyer...@redhat.com
Date: Thu, 30 Aug 2012 13:07:10 +0200
Subject: [PATCH] scripts: git commit
Stefano Lattarini wrote:
On 08/30/2012 02:16 PM, Jim Meyering wrote:
I've merged those into the appropriate change sets, removed all
Signed-off-by: lines and made a few minor wording changes.
Please review the differences. If you're happy with the result,
I'll push the lot.
I have few
Stefano Lattarini wrote:
This is not strictly required now (it will be once we make more
parts of the coreutils build system non-recursive), but enabling
it early help to ensure that we don't unwittingly introduce any
incompatibility or subtle breakage that might get to bite us later.
*
Bernhard Voelker wrote:
...
After finally getting autoconf+autmake working (and running bootstrap
etc in coreutils), a `make syntax-check` failed because of the
Thanks for the quick work and sorry I forgot to run
make syntax-check before pushing that.
renaming of the test scripts.
The
Erik Auerswald wrote:
On Fri, Aug 31, 2012 at 09:19:19AM +0200, Jim Meyering wrote:
Bernhard Voelker wrote:
On 08/30/2012 02:13 PM, Stefano Lattarini wrote:
Now that we use AM_TESTS_ENVIRONMENT, we should require at least
Automake = 1.11.2; but since all the Automake version until 1.11.5
c1495a1107871f451fd680d5c23b2c71af6e9971 Mon Sep 17 00:00:00 2001
From: Jim Meyering meyer...@redhat.com
Date: Fri, 31 Aug 2012 10:55:03 +0200
Subject: [PATCH] scripts: add autotools-install, for those stuck with
outdated tools
* scripts/autotools-install: New script, so you can always build
from
Stefano Lattarini wrote:
* tests/Makefile.am (EXTRA_DIST): Drop 'check.mk', which has been
removed (being merged into 'tests/Makefile.am') by recent changes.
Signed-off-by: Stefano Lattarini stefano.lattar...@gmail.com
---
tests/Makefile.am | 1 -
1 file changed, 1 deletion(-)
diff --git
Bernhard Voelker wrote:
On 08/31/2012 10:16 AM, Jim Meyering wrote:
wrote the commit log for you, too.
Here's the result. I'll wait for your ACK:
Thanks for wrapping that into a patch.
I just noticed that there's still a stray
grep: *.1: No such file or directory
Well spotted
Stefano Lattarini wrote:
On 08/31/2012 11:56 AM, Jim Meyering wrote:
Stefano Lattarini wrote:
* tests/Makefile.am (EXTRA_DIST): Drop 'check.mk', which has been
removed (being merged into 'tests/Makefile.am') by recent changes.
Signed-off-by: Stefano Lattarini stefano.lattar...@gmail.com
Stefano Lattarini wrote:
It was useful only back when coreutils used CVS as its version
control system.
* build-aux/cvsu: Delete.
* Makefile.am (EXTRA_DIST): Remove it.
Signed-off-by: Stefano Lattarini stefano.lattar...@gmail.com
---
Makefile.am| 1 -
build-aux/cvsu | 512
Stefano Lattarini wrote:
Found these when trying to convert 'src/' to a non-recursive setup.
I think they should be simple and uncontroversial enough to go into
the repository ASAP.
Thanks,
Stefano
-*-*-*-
Stefano Lattarini (7):
build: move definition of 'all_programs' in cfg.mk
Stefano Lattarini wrote:
* bootstrap.conf ($buildreq): Require gettext = 0.18.1, consistently
with the call to AM_GNU_GETTEXT_VERSION in configure.ac.
Signed-off-by: Stefano Lattarini stefano.lattar...@gmail.com
---
bootstrap.conf | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
Jim Meyering wrote:
Stefano Lattarini wrote:
* bootstrap.conf ($buildreq): Require gettext = 0.18.1, consistently
with the call to AM_GNU_GETTEXT_VERSION in configure.ac.
Signed-off-by: Stefano Lattarini stefano.lattar...@gmail.com
---
bootstrap.conf | 2 +-
1 file changed, 1 insertion
Stefano Lattarini wrote:
On 08/31/2012 12:34 PM, Jim Meyering wrote:
Jim Meyering wrote:
Stefano Lattarini wrote:
* bootstrap.conf ($buildreq): Require gettext = 0.18.1, consistently
with the call to AM_GNU_GETTEXT_VERSION in configure.ac.
Signed-off-by: Stefano Lattarini stefano.lattar
Stefano Lattarini wrote:
Found these when trying to convert 'src/' to a non-recursive setup.
I think they should be simple and uncontroversial enough to go into
the repository ASAP.
Thanks. Pushed.
Jim Meyering wrote:
Stefano Lattarini wrote:
On 08/31/2012 03:34 PM, Jim Meyering wrote:
Stefano Lattarini wrote:
These should go on the top of my other pending series [PATCH 0/7] Minor
cleanup and refactoring:
http://lists.gnu.org/archive/html/coreutils/2012-08/msg00205.html
Thanks
Stefano Lattarini wrote:
On 08/31/2012 03:57 PM, Jim Meyering wrote:
Stefano Lattarini wrote:
On 08/31/2012 03:34 PM, Jim Meyering wrote:
Stefano Lattarini wrote:
These should go on the top of my other pending series [PATCH 0/7] Minor
cleanup and refactoring:
http://lists.gnu.org
Stefano Lattarini wrote:
On 08/31/2012 04:06 PM, Jim Meyering wrote:
Hmm... actually there are a few other problems. See below.
If you don't mind fixing those and reposting, I would be grateful.
I'm stepping out for an hour or so.
Here is an updated series. Notice that there is still one
Stefano Lattarini wrote:
On 08/31/2012 08:12 PM, Jim Meyering wrote:
From ea8043e5b08956b804c1cc3cb79ee54f2d969b36 Mon Sep 17 00:00:00 2001
From: Jim Meyering meyer...@redhat.com
Date: Fri, 31 Aug 2012 20:05:29 +0200
Subject: [PATCH 2/2] maint: check-programs-vs-x: avoid a new syntax-check
Jim Meyering wrote:
...
No problem.
Next step, make distcheck. First failure is because inclusion of
version.h fails in the non-srcdir build:
...
This fixes it:
diff --git a/src/local.mk b/src/local.mk
FYI, here's another make distcheck problem,
in case someone else can get to it before I
Eric Blake wrote:
Newer glibc hates unconditional use of FORTIFY_SOURCE when optimization
is disabled but -Werror is in effect.
* configure.ac (FORTIFY_SOURCE): Make conditional.
---
Copied from a trick I used in libvirt.
configure.ac | 9 +++--
1 file changed, 7 insertions(+), 2
Jim Meyering wrote:
Jim Meyering wrote:
...
No problem.
Next step, make distcheck. First failure is because inclusion of
version.h fails in the non-srcdir build:
...
This fixes it:
diff --git a/src/local.mk b/src/local.mk
FYI, here's another make distcheck problem,
in case someone
Jim Meyering wrote:
Jim Meyering wrote:
Jim Meyering wrote:
...
No problem.
Next step, make distcheck. First failure is because inclusion of
version.h fails in the non-srcdir build:
...
This fixes it:
diff --git a/src/local.mk b/src/local.mk
FYI, here's another make distcheck problem
Jim Meyering wrote:
And here's one more, so I don't forget:
rm -rf src/.deps
rm -f Makefile
ERROR: files left in build directory after distclean:
./man/arch.1
./man/hostname.1
make[1]: *** [distcleancheck] Error 1
make[1]: Leaving directory
`/h/j/w/co/cu/coreutils-8.19.68
Stefano Lattarini wrote:
On 08/31/2012 09:59 PM, Jim Meyering wrote:
Jim Meyering wrote:
Jim Meyering wrote:
...
No problem.
Next step, make distcheck. First failure is because inclusion of
version.h fails in the non-srcdir build:
...
This fixes it:
diff --git a/src/local.mk b/src
Stefano Lattarini wrote:
This is a re-spin, extension and follow-up to this previous series:
http://lists.gnu.org/archive/html/coreutils/2012-08/msg00257.html
plus other follow-up fixlets and refactoring. Thanks to Jim Meyering
for his testing of earlier iterations of this series
Stefano Lattarini wrote:
On 09/01/2012 01:46 AM, Stefano Lattarini wrote:
Subject: maint: avoid distcheck failure by properly cleaning $(EXTRA_MANS)
Erm, it should be s/$(EXTRA_MANS)/$(ALL_MANS)/ here.
Amended. Thanks.
Jim Meyering wrote:
...
FYI, here's another make distcheck problem,
...
And here's another:
...
make[3]: Makefile: No such file or directory
make[3]: *** No rule to make target `Makefile'. Stop.
make[3]: Entering directory
/h/j/w/co/cu/tests/torture/coreutils/test/coreutils-8.19.68
Stefano Lattarini wrote:
From: Jim Meyering meyer...@redhat.com
* man/local.mk (distclean-local): Remove $(ALL_MANS) when doing
a VPATH build. If it's not done, generated manpages can be left
around in the build directory after a make distclean, causing
failures in make distcheck
Stefano Lattarini wrote:
Oh, and now that I notice ...
On 09/01/2012 01:46 AM, Stefano Lattarini wrote:
Some of them can be simplified after the previous changes, some
of them have been downright broken them, and need fixing.
* src/local.mk: Adjust some comments.
(EXTRA_DIST): Avoid
Jim Meyering wrote:
...
With this (final?) change, I expect make distcheck to succeed
using my current set of patches.
Not quite:
...
And fixed by this:
Subject: [PATCH] build: restore handling of space-tainted build directory
name
* man/local.mk: With commit v8.19-84-g08cf455, man page
.
...
[PATCH 19/25] maint: port manpages generation to VPATH builds
* man/local.mk (.x.1): Use '$(MKDIR_P)' rather than bare 'mkdir'
where appropriate.
Reported-by: Jim Meyering meyer...@redhat.com
You might want to delete this 'Reported-by:', if I remember your
preferences correctly
Yes
Stefano Lattarini wrote:
[PATCH 22/25] build: create src/ and man/ via configure
* configure.ac: Create man/ and src/, since, now that there is no
generated Makefile in each, the directory will not exist in a
non-srcdir build.
This patch can be dropped AFAICS: 'man/' is created thanks to
Jim Meyering wrote:
Erik Auerswald wrote:
...
Good point. Thanks. I'm tempted to remove the build instructions from
README-prereq, and instead to include my autotools-install script under
script and referencing it. WDYT?
I prefer written build instructions to a script. The script could
Sep 17 00:00:00 2001
From: Jim Meyering meyer...@redhat.com
Date: Sun, 2 Sep 2012 13:42:39 +0200
Subject: [PATCH] maint: accommodate upcoming, expanded list of warnings from
gnulib
* configure.ac: Disable a new gcc warning, -Wsuggest-attribute=format,
since it triggers on copy.c (which I'm
or so ago,
but somehow managed not to include in the series I pushed:
From 135664d25e09d9b285b59012832f0e8db5802042 Mon Sep 17 00:00:00 2001
From: Jim Meyering meyer...@redhat.com
Date: Sun, 2 Sep 2012 13:46:17 +0200
Subject: [PATCH] maint: disable sc_proper_name_utf8_requires_ICONV test
It would
Stefano Lattarini wrote:
Plus some other preparatory, follow-up or while-at-it changes.
After this, make syntax-check and make distcheck still pass on
my Debian box.
This completes the first important part of the de-recursion of the
coreutils build system: the compilation of the programs,
produces
precisely the one or two lines that I wanted to see. Which do you think
I will be using most of the time? ;-)
From 45bf1ab13682f8842b2b9ed81bb9069f6e5b01b3 Mon Sep 17 00:00:00 2001
From: Jim Meyering meyer...@redhat.com
Date: Tue, 4 Sep 2012 12:12:19 +0200
Subject: [PATCH] tests: depend
Andreas Schwab wrote:
Jim Meyering jim-oxw1nkzkivjk1umjsbk...@public.gmane.org writes:
Stefano Lattarini wrote:
* tests/init.cfg (stty_reversible_init_): Quote '$abs_top_srcdir'
properly.
(fiemap_capable_): Quote '$abs_srcdir' properly.
(require_dirent_d_type_): Likewise.
---
init.cfg
Stefano Lattarini wrote:
On 09/04/2012 05:25 PM, Jim Meyering wrote:
Hi Stefano,
I'm done reviewing.
Here's the 17-cset series I'm ready to push.
[For anyone wondering about the message to which Stefano has
just replied, it was delayed due to size 100KiB, then rejected, since
the only one
Stefano Lattarini wrote:
On 09/04/2012 06:03 PM, Jim Meyering wrote:
Stefano Lattarini wrote:
On 09/04/2012 05:25 PM, Jim Meyering wrote:
Hi Stefano,
I'm done reviewing.
Here's the 17-cset series I'm ready to push.
[For anyone wondering about the message to which Stefano has
just
Bernhard Voelker wrote:
What about making --enable-gcc-warnings the default
for CU developers, i.e. when .git exists?
Good idea.
If someone has gone to the trouble of building from cloned sources
they're probably also using modern enough tools.
Do you feel like writing the patch?
Pádraig Brady wrote:
* src/nproc.c (main): Error if any non option parameters.
* NEWS: Mention the change in behavior.
---
NEWS|4
src/nproc.c |7 +++
2 files changed, 11 insertions(+), 0 deletions(-)
diff --git a/NEWS b/NEWS
index 63fa042..995fafb 100644
---
Pádraig Brady wrote:
On 09/06/2012 11:09 AM, Jim Meyering wrote:
diff --git a/NEWS b/NEWS
index 63fa042..6941721 100644
--- a/NEWS
+++ b/NEWS
@@ -25,6 +25,12 @@ GNU coreutils NEWS -*- outline -*-
** Improvements
+ root-only tests now check for permissions of our dummy user
Bernhard Voelker wrote:
On September 6, 2012 at 4:31 PM Eric Blake ebl...@redhat.com wrote:
$ echo one bar
$ echo two blah
$ cp -n blah bar
$ echo $?
0
$ cat bar
one
that FreeBSD also returns 0 in this case, and I don't want to break
interoperability. Therefore, I'm going to close
Ondrej Oprala wrote:
On 09/06/2012 12:09 PM, Jim Meyering wrote:
I'll push this once you ACK it:
Hi,
it's all ok with me, thanks for taking the time.
Rats! Both of us missed the fact that I changed the Author:
from your name to mine. Grr... Terribly sorry.
This should fix
I've converted lib/ to non-recursive make, too, and thought it was time
to measure the performance change. For now, it's all in a single ugly
commit, so I'm not ready to share. I will tease it into at least a few
separate commits so that it's easier to review.
I didn't expect much improvement,
Stefano Lattarini wrote:
On 09/08/2012 07:46 PM, Jim Meyering wrote:
Stefano Lattarini wrote:
On 09/08/2012 03:29 PM, Jim Meyering wrote:
But, ... surprise! So far there is *no* improvement, and in fact a small
penalty in this very worst case:
My wild guess is that the culprit
Jim Meyering wrote:
...
On the subject of converting to non-recursive make in lib/,
I have at least one problem so far: how can I select different
CFLAGS depending on the directory containing the file I'm compiling?
As you know, we currently have three different sets of warnings:
- strict
Jim Meyering wrote:
Jim Meyering wrote:
...
On the subject of converting to non-recursive make in lib/,
I have at least one problem so far: how can I select different
CFLAGS depending on the directory containing the file I'm compiling?
As you know, we currently have three different sets
same-inode
configure.ac:
-gl_ROOT_DEV_INO
Makefile.am:
+lib_SOURCES += root-dev-ino.c root-dev-ino.h
Include:
root-dev-ino.h
--
1.7.12.289.g0ce9864
From c424de16b1c8607f2caa7c85d69196f9d8e3a103 Mon Sep 17 00:00:00 2001
From: Jim Meyering meyer...@redhat.com
Date: Sat, 8 Sep 2012 15:11:20
601 - 700 of 5559 matches
Mail list logo
