Re: [platform-testers] new snapshot available: coreutils-9.5.239-a836.tar.xz

Test results of a recent coreutils snapshot from some other platforms (ignoring the issues already reported this week): Hurd 2022/i386 5 FAIL: ls/ls-misc.pl, wc/wc-proc, wc/wc-total, factor/factor-parallel, nice/nice Hurd 2024/x86_641 HANG NetBSD 7.1

Re: new snapshot available, with AIX make

Collin Funk wrote: > The AIX 'make' command didn't build '[' causing the --help test to fail. > It has 'gmake' which builds it though. AIX 'make' also does not support the '!=' operator that is required by POSIX [1]. You should make it a habit to use GNU make on AIX. Bruno [1] https://pubs.open

Re: new snapshot available, ls -Z crash

On - OpenBSD 6.0, - NetBSD 7.1 I see a test failure: FAIL tests/ls/selinux-segfault.sh (exit status: 1) Log is attached. The cause is a crash that is easy to reproduce interactively: $ src/ls -Z . In gdb, I get this stack trace (same on both platforms): #0 strlen #1 gobble_file (na

Re: new snapshot available, with AIX xlc

On AIX 7.1, with xlc as compiler, there is a build failure: $ xlc -q64 -qthreaded -qtls -I. -I.. -I./lib -Ilib -I../lib -Isrc -I../src -I/home/haible/prefix64/include -D_THREAD_SAFE -g -c -o src/ls.o ../src/ls.c "../src/ls.c", line 636.27: 1506-196 (W) Initialization between types "const cha

Re: new snapshot available, on Linux

Test results of a recent snapshot on various Linux systems (other than the issues already reported): Ubuntu with clang+asan+ubsan OK Alpine Linux 3.20 OK Debian 12/m68kOK CentOS 5 3 FAIL (was 4 FAIL in coreutils-9.3.156-b3afb) CentOS 6

Re: new snapshot available

Building a coreutils + gnulib snapshot from yesterday on GNU/Hurd (x86_64) [1]: The test 'wc/wc-total' hangs the file system. The first time I had to reboot. The second time, Ctrl-C was sufficient to interrupt it. [1] https://cdimage.debian.org/cdimage/ports/latest/hurd-amd64/debian-hurd-2024111

Re: [PATCH] fix consistency issues in set_program_name()

Pádraig Brady wrote: > > Can you describe the circumstances where you observed the discrepancy? > > It impacts the tests when we `./configure --enable-single-binary=shebangs`. Oh, I see. In this case, 'cp' for example is an executable script #!/bin/coreutils --coreutils-prog-shebang=cp This c

Re: [PATCH] fix consistency issues in set_program_name()

Pádraig Brady wrote: > Updated version attached. This patch looks good. Bruno

Re: [PATCH] fix consistency issues in set_program_name()

Hi Pádraig, > I noticed a discrepancy in the program name output by error(), > which looks to be due to whether the gnulib or glibc version is used. > If we use gnulib's error() it uses getprogname() > which defaults to program_invocation_short_name. > The glibc error() instead uses program_invoca

Re: behavioral change in coreutils `pwd` handling '//' (double slash) caused by gnulib update

t;\n", iov_len=1}], 2) = 6 close(1)= 0 close(2)= 0 exit_group(0) = ? +++ exited with 0 +++ So, the readlink on the root directory returns "/", not "//". The double slash comes from not-s

Re: Please include the local gnulib modules in releases

Michael Pratt wrote: > > > What's needed is just the "modules" files which are missing > > > > > > No, the gl/lib/ and gl/tests/ subdirectories are input to gnulib-tool as > > well. > > Not only gl/modules/. > > > > And you should list the files to add to the tarball, not directories. > > > T

Re: Please include the local gnulib modules in releases

gnu.org/software/automake/manual/html_node/Basics-of-Distribution.html [3] https://lists.gnu.org/archive/html/automake-patches/2024-10/msg3.html From: Bruno Haible build: Distribute the gl/ directory in the tarballs. Suggested by Michael Pratt in <https://mail.gnu.org/archive/html/coreutil

Re: Building from the git repository fails with "alias_table defined but not used"

ULIB_WARNINGS. Done through the patch below. 2024-10-02 Bruno Haible Silence -Wunused-const-variable warnings in Gnulib code. Reported by Yuri Kanivetsky via Pádraig Brady in <https://lists.gnu.org/archive/html/bug-gnulib/2024-10/msg7.html>.

translator comment in src/stat.c lost

duced in commit d3227eeb90c8308abd1e6bf08ee253b7a4e78d1d on 2012-01-09. Here's patch that fixes it. >From 5fa41bfb49118376c39c1baf8786c2300a336004 Mon Sep 17 00:00:00 2001 From: Bruno Haible Date: Tue, 25 Jun 2024 12:18:40 +0200 Subject: [PATCH] stat: Reactivate a translator comment (lost on 20

Re: coreutils-9.4.185-541b02 od/od-multiple-t test failure

Pádraig Brady wrote: > > I think it is enough to remove the 'long double' part from the test > > (option letter 'L')... > > Pushed that in your name at: > https://git.sv.gnu.org/gitweb/?p=coreutils.git;a=commitdiff;h=bec53850b Thank you. Yes, that's what I meant :-) I confirm that it fixes the t

Re: coreutils-9.4.185-541b02 od/od-multiple-t test failure

Pádraig Brady wrote: > The original bug report was not specific to floats, so I'll just remove them > from the test. I think it is enough to remove the 'long double' part from the test (option letter 'L'). 'float' and 'double' are standardized by IEEE 754 (except on Linux/m68k) and don't have und

coreutils-9.4.185-541b02 tail/retry test failure

On two machines (Linux/sparc64 and Linux/riscv64, real hardware, Gentoo) I see reproducible test failures of tail/retry. See attached log files. I do *not* see these failures with architectures in the same farm (matoro.tk). I also do *not* see these failures in QEMU VMs with a local (surely ext4)

Re: uniq for unsorted input

Pádraig Brady wrote: > A documentation patch would be very useful. > If you search for "Decorate Sort Undecorate" in coreutils.texi > you can see an existing example of this DSU pattern. > I would also mention DSU if adding another example. OK, I'll provide doc input for this (post 9.5). > Re add

Re: [platform-testers] new snapshot available: coreutils-9.4.184-606f5.tar.xz

Pádraig Brady wrote: > OK done. Please use coreutils-9.4.185-541b02.tar.xz Thanks. With this, all tests pass on CentOS 8 Stream.

Re: [platform-testers] new snapshot available: coreutils-9.4.184-606f5.tar.xz

Pádraig Brady wrote: > We plan to release coreutils-9.5 in the next few days I would suggest to update to the current gnulib, in order to get these two fixes: https://git.savannah.gnu.org/gitweb/?p=gnulib.git;a=commit;h=3978cc5adb34f3152f119d46f53e6e8e107158fa https://git.savannah.gnu.org/gitweb/?

Re: coreutils-9.4.170-7b206 misc/numfmt test failure

Pádraig Brady wrote: > >>> The misc/numfmt test fails on > >>> - Alpine Linux 3.18, > >>> - macOS 12.5, > >>> - FreeBSD 14.0, > >>> - Solaris 11 OpenIndiana. > >>> > >>> Find attached the test-suite.log files. > >>> ... > FreeBSD 14 at least looks to be counting bytes rather than c

Re: coreutils-9.4.170-7b206 help/help-version test failure

I wrote: > The help/help-version test fails on AIX 7.1 and AIX 7.3.1. > > test-suite.log on AIX 7.1: > > > FAIL: tests/help/help-version > = > > env: '[': No such file or directory > env: '[': No such file or directory > env: '[': No such file or directory > FAIL: [

Re: coreutils-9.4.170-7b206 misc/numfmt test failure

Pádraig Brady wrote: > > The misc/numfmt test fails on > >- Alpine Linux 3.18, > >- macOS 12.5, > >- FreeBSD 14.0, > >- Solaris 11 OpenIndiana. > > > > Find attached the test-suite.log files. > > > > If these test failures are due to platform-specific *printf bugs, please > > pro

Re: coreutils-9.4.170-7b206 chmod/symlinks test failure

I confirm: this unit test now passes on all mentioned platforms.

Re: coreutils-9.4.170-7b206 mv/mv-exchange test failure

Pádraig Brady wrote: > > The mv/mv-exchange test fails on: > >- NetBSD 9.3 > >- OpenBSD 7.4 > >- AIX 7.1 > >- AIX 7.3.1 > > The attached should fix the issue with this new test. I confirm: It fixes the test failure on all 4 platforms.

Re: coreutils-9.4.170-7b206 ls/removed-directory test failure

Pádraig Brady wrote: > > The ls/removed-directory test fails on FreeBSD 14.0. > > > > tests-suite.log from FreeBSD 14.0: > > > > > > FAIL: tests/ls/removed-directory > > > > > > diff -u /dev/null err > > --- /dev/null 1970-01-01 > > +++ err 1970-01-01

Re: coreutils-9.4.170-7b206 ls/dired test failure

Pádraig Brady wrote: > > The ls/dired test fails on: > >- FreeBSD 14.0, > >- Solaris 11 OpenIndiana. > > https://github.com/coreutils/coreutils/commit/a51af4329 Thanks. I confirm that it fixes the test failure on both platforms.

Re: coreutils-9.4.170-7b206 cp/preserve-mode test failure

Pádraig Brady wrote: > The attached fixes this by including the setgid bit > in the fallback mode if appropriate. Thanks. I confirm that it fixes the test failure.

Re: coreutils-9.4.170-7b206 uniq/uniq test failure

Pádraig Brady wrote: > I'll apply the attached anyway which should avoid this issue. Thanks. I confirm that it fixes the failure.

uniq for unsorted input

The documentation of 'uniq' [1] says: "The input need not be sorted, but repeated input lines are detected only if they are adjacent. If you want to discard non-adjacent duplicate lines, perhaps you want to use 'sort -u'." When I wrote gnulib-tool many years ago, as a shell script, I natu

Re: coreutils-9.4.170-7b206 uniq/uniq test failure

Pádraig Brady wrote: > > You should also have access to cfarm215.cfarm.net and cfarm216.cfarm.net > > (just announced today). > > Nice machines. > Though I can't repo there, I'm guessing because they've > both multi and uni byte fr locales installed. I had seen the issue on Solaris 11 OpenIndiana

Re: coreutils-9.4.170-7b206 uniq/uniq test failure

Pádraig Brady wrote: > while uniq (c32isblank) now determines > it is not blank (which seems more correct). I agree that U+00A0 NO-BREAK SPACE should better be considered to be non-blank (and Gnulib's c32isblank does so). > The only solaris 11 system I have access to, only has the fr_FR.UTF-8 loc

coreutils-9.4.170-7b206 test failures on AIX 7.3

On AIX 7.3, I also see three other test failures: - misc/stdbuf - install/basic-1 - mkdir/p-3 Here's the test-suite.log. FAIL: tests/misc/stdbuf === stdbuf: invalid mode 'l' stdbuf: invalid mode '18446744073709551616': Value too large to be stored in data type stdbuf:

coreutils-9.4.170-7b206 misc/numfmt test failure

The misc/numfmt test fails on - Alpine Linux 3.18, - macOS 12.5, - FreeBSD 14.0, - Solaris 11 OpenIndiana. Find attached the test-suite.log files. If these test failures are due to platform-specific *printf bugs, please provide a test case for *printf at the C level. Then I can add a work

coreutils-9.4.170-7b206 chmod/symlinks test failure

The chmod/symlinks test fails on: - Alpine Linux 3.18, - macOS 12.5, - FreeBSD 14.0, - NetBSD 9.3 and 10.0-rc4, - OpenBSD 7.4, - Solaris 11.4, - AIX 7.1 and 7.3.1. Find attached the test-suite.log files. FAIL: tests/chmod/symlinks == + initial_cwd_=/home/brun

coreutils-9.4.170-7b206 cp/preserve-mode test failure

The cp/preserve-mode test fails on Alpine Linux 3.18. test-suite.log from Alpine Linux 3.18: FAIL: tests/cp/preserve-mode + initial_cwd_=/home/bruno/coreutils-9.4.170-7b206/build + testdir_prefix_ + printf gt + pfx_=gt + mktempd_ /home/bruno/coreutils-9.4.170-7b206/

coreutils-9.4.170-7b206 chown/preserve-root test failure

The chown/preserve-root test fails on macOS 12.5. test-suite.log from macOS 12.5: FAIL: tests/chown/preserve-root === ++ initial_cwd_=/Users/haible/STORAGE/coreutils-9.4.170-7b206/build +++ testdir_prefix_ +++ printf gt ++ pfx_=gt +++ mktempd_ /Users/haible/STORAGE/c

coreutils-9.4.170-7b206 ls/dired test failure

The ls/dired test fails on: - FreeBSD 14.0, - Solaris 11 OpenIndiana. test-suite.log on FreeBSD 14.0: FAIL: tests/ls/dired tail: cannot open '+53' for reading: No such file or directory Mismatch! Expected: 1a, Found: tail: cannot open '+98' for reading: No such file or

coreutils-9.4.170-7b206 ls/removed-directory test failure

The ls/removed-directory test fails on FreeBSD 14.0. tests-suite.log from FreeBSD 14.0: FAIL: tests/ls/removed-directory diff -u /dev/null err --- /dev/null 1970-01-01 +++ err 1970-01-01 +ls: reading directory '.': No such file or directory FAIL tests/ls/remov

coreutils-9.4.170-7b206 date/date-debug, date/date-tz test failures

The date/date-debug, date/date-tz tests fail on NetBSD 9.3 and 10.0-rc4. test-suite.log from NetBSD 9.3: FAIL: tests/date/date-debug === --- exp22024-03-24 12:44:31.942851024 +0100 +++ out22024-03-24 12:44:31.947773870 +0100 @@ -5,9 +5,8 @@ date: error:

coreutils-9.4.170-7b206 mv/mv-exchange test failure

The mv/mv-exchange test fails on: - NetBSD 9.3 - OpenBSD 7.4 - AIX 7.1 - AIX 7.3.1 test-suite.log from NetBSD 9.3: FAIL: tests/mv/mv-exchange == mv: cannot exchange 'a' and 'b': Not supported FAIL tests/mv/mv-exchange.sh (exit status: 1) test-suite.log from Ope

coreutils-9.4.170-7b206 tail/pipe-f2 test failure

The tail/pipe-f2 test fails on NetBSD 9.3 and 10.0-rc4. test-suite.log on NetBSD 9.3: FAIL: tests/tail/pipe-f2 --- exp 2024-03-24 12:45:54.627129684 +0100 +++ out 2024-03-24 12:45:54.628038908 +0100 @@ -1 +0,0 @@ -1 kill: 20415: No such process FAIL tests/tail/pipe-f2.s

coreutils-9.4.170-7b206 stty/stty-row-col test failure

The stty/stty-row-col test fails on Solaris 11.4. test-suite.log on Solaris 11.4: FAIL: tests/stty/stty-row-col = ++ initial_cwd_=/export/home/bruno/coreutils-9.4.170-7b206/build-64 +++ testdir_prefix_ +++ printf gt ++ pfx_=gt +++ mktempd_ /export/home/bruno/coreutil

coreutils-9.4.170-7b206 uniq/uniq test failure

The uniq/uniq test fails on Solaris 11 OpenIndiana. test-suite.log on Solaris 11 OpenIndiana: FAIL: tests/uniq/uniq = uniq (GNU coreutils) 9.4.170-7b206 Copyright (C) 2024 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later

coreutils-9.4.170-7b206 help/help-version test failure

The help/help-version test fails on AIX 7.1 and AIX 7.3.1. test-suite.log on AIX 7.1: FAIL: tests/help/help-version = env: '[': No such file or directory env: '[': No such file or directory env: '[': No such file or directory FAIL: [ 0+1 records in 0+1 records out FA

coreutils-9.4.170-7b206 FTBFS on AIX 7.3.1

On the machine gcc119.fsffrance.org (AIX 7.3.1), with gcc (CC="/opt/freeware/bin/gcc -maix64"), there is a compilation error: CC lib/libcoreutils_a-md5-stream.o In file included from /usr/include/openssl/opensslv.h:125, from ../lib/md5.h:36, from ../lib/md

coreutils-9.4.170-7b206 undefined behaviour in tests

Hi, Building coreutils-9.4.170-7b206 on a glibc system, with clang and ASAN and UBSAN options, produces 4 test failures. I'm using clang 17.0.4 with the options CC="$CC -fsanitize=address,undefined,signed-integer-overflow,shift,integer-divide-by-zero" CFLAGS="-O0 -fno-omit-frame-pointer -ggdb" A

Re: sort dynamic linking overhead

Pádraig Brady wrote: > > Does this work for all the various names of libcrypto in various distros? > > > > Debian 12 libcrypto.so.3 > > Ubuntu 22.04libcrypto.so.1.1 libcrypto.so.3 > > Slackware 15libcrypto.so.1.1 > > openSUSE 15.5 libcrypto.so.1.1 > > CentOS Stream 9 libcrypto.so.3

Re: sort dynamic linking overhead

Paul Eggert wrote: > Thanks for the patch. I was hoping that we didn't need to worry about > older platforms needing -ldl. Oh well. Distros with glibc < 2.34 include: Debian 11 CentOS 8 stream Fedora 34 openSUSE 15.5 Slackware 15 These are not obsolete, so far. > The patch causes 'conf

Re: sort dynamic linking overhead

s: 'ldd' shows that without the patch, 'sort' loads the libraries linux-vdso.so.1 libcrypto.so.3 libpthread.so.0 libc.so.6 libz.so.1 libdl.so.2 /lib64/ld-linux-x86-64.so.2 and with the patch, 'sort' loads the libraries linux-vdso.so.1 libdl.so.2

Re: sort dynamic linking overhead

Pádraig Brady wrote: > >> + void *handle = dlopen ("libcrypto.so", RTLD_LAZY | RTLD_GLOBAL); > > > > That only works if libopenssl-devel is installed. > > Good spot. > I'd already pushed a fix for this at: > https://git.sv.gnu.org/gitweb/?p=coreutils.git;a=commitdiff;h=3bbdb3938 Does this work

Re: Uptime question

> I actually thought something like the CLOCK_MONOTONIC or CLOCK_BOOTTIME were > used for providing uptime to the user, but those do not allow date changes. You are assuming a certain implementation, and then thinking what properties the program "must" have. Try thinking the other way around: Wha

Re: coreutils-9.3.156-b3afb on AIX, CentOS 5, Alpine Linux, FreeBSD, Cygwin

Pádraig Brady wrote: > I don't think there is anything blocking here anyway. > I'll prepare the release... It's your decision. But I happen to agree :) Bruno

Re: coreutils-9.3.147-d553ab on Android

Pádraig Brady wrote: > A lot of the test failures are just because > curly quotes are being used in diagnostics, > thus causing expected messages not to match. > Searching for curly quotes in the logs show many instances like: > >-sort: text ordering performed using simple byte comparison >

Re: coreutils-9.3.147-d553ab on Android

st-suite.log.gz Description: application/gzip From 31c7d3ae272d54badba9f432db5068d52c26dda1 Mon Sep 17 00:00:00 2001 From: Bruno Haible Date: Mon, 28 Aug 2023 12:07:18 +0200 Subject: [PATCH] maint: Avoid test failure on Android * gl/tests/test-mbsalign.c (main): Skip the unibyte truncation test o

Re: coreutils-9.3.147-d553ab on Cygwin

Pádraig Brady wrote: > Weird. I guess the `: > missing\dir` construct causes the script > to exit early in failure. Perhaps `if touch missing\dir` would work better. No, that does not make a difference. Both ": > 'missing\dir'" and "touch 'missing\dir'" succeed when 'missing' is an existing direc

Re: coreutils-9.3.147-d553ab on Cygwin

Pádraig Brady wrote: > > Here is a comparison of the failing tests: > > > > === Cygwin 2.9.0 Cygwin 3.4.6 === > > FAIL: tests/rm/cycleFAIL: tests/rm/cycle > > FAIL: tests/chmod/no-x FAIL: tests/chmod/no-x > > FAIL: tests/

Re: coreutils-9.3.147-d553ab on Cygwin

Pádraig Brady wrote: > The cksum failures are regressions, > which should be handled by the two attached patches. After applying these two patches, only the 'FAIL: tests/cksum/b2sum' is gone. The 'FAIL: tests/cksum/md5sum-bsd' is still there. Find attached the new log file. cygwin346-test-suite

Re: coreutils-9.3.147-d553ab on Cygwin

> On Cygwin 2.9.0/x86_64, I see 8 ERROR and 30 FAIL: Whereas on Cygwin 3.4.6/x86_64, I see 8 ERROR and 29 FAIL: ERROR: tests/tail/inotify-hash-abuse ERROR: tests/rm/rm-readdir-fail ERROR: tests/cp/nfs-removal-race ERROR: tests/df/no-mtab-status ERROR: tests/df/skip-duplicates ERROR: tests/du/inod

Re: coreutils-9.3.147-d553ab on other platforms

The build succeeds and all tests pass on the following platforms: Linux Guix 1.4 Ubuntu 22.04 Ubuntu 22.04 / 32 Ubuntu 22.04 / x32 Debian 9.1 Debian 11.1 Debian 12 / powerpc64 CentOS 8-stream CentOS Stream 9 Fedora 28 / riscv64 Fedora Rawhide Cinnamon OpenSUSE Leap 15.2 M

Re: coreutils-9.3.147-d553ab on FreeBSD

Pádraig Brady wrote: > Perhaps fpurge() is needed on FreeBSD to avoid the repeated warning? > I don't have easy access to a FreeBSD machine at present, > to test this change: > > diff --git a/src/head.c b/src/head.c > index 7bba1420c..da32c886f 100644 > --- a/src/head.c > +++ b/src/head.c > @@ -18

Re: coreutils-9.3.147-d553ab on Hurd

On the Hurd, I got weird results during "make check"; see attached screenshot. Thus I restarted from a clean Hurd environment, and now the "make check" run terminated reasonably, with 18 FAILures. Find attached the log file. hurd-test-suite.log.gz Description: application/gzip

Re: coreutils-9.3.147-d553ab on Alpine Linux

Pádraig Brady wrote: > Perhaps 'sleep' is a shell builtin here which is confusing things? Yes: $ type sleep sleep is a shell builtin > Does the following help (I don't have easy access to an alpine system): > diff --git a/tests/misc/usage_vs_getopt.sh b/tests/misc/usage_vs_getopt.sh > index 3ffb

Re: coreutils-9.3.147-d553ab on NetBSD

On NetBSD 9.3/x86_64, I see 3 test failures: FAIL: tests/date/date-debug FAIL: tests/date/date-tz FAIL: tests/tail/pipe-f2 On NetBSD 8.0/sparc64, I see these test failures, and additionally: FAIL: tests/touch/not-owner This is a subset of what I reported in https://lists.gnu.org/archive/html/co

Re: coreutils-9.3.147-d553ab on Solaris 11.4

On Solaris 11.4, in 64-bit mode, I still get 1 failure: FAIL: tests/stty/stty-row-col like already reported in https://lists.gnu.org/archive/html/coreutils/2023-03/msg00035.html The log file is attached. solaris114-test-suite.log.gz Description: application/gzip

Re: coreutils-9.3.147-d553ab on FreeBSD

There are no test failures on - FreeBSD 12.0/x86_64 - FreeBSD 12.2/arm64 - FreeBSD 13.0/x86_64, FreeBSD 13.0/arm64 However: * There is 1 test failure on FreeBSD 13.1/x86_64: FAIL: tests/head/head-write-error Same as my earlier report in https://lists.gnu.org/archive/html/coreutils/202

Re: coreutils-9.3.147-d553ab on Alpine Linux

On Alpine Linux 3.18 (x86), I see two test failures: FAIL: tests/misc/usage_vs_getopt FAIL: tests/cp/preserve-mode The first one is new. The second one I already reported in https://lists.gnu.org/archive/html/coreutils/2023-03/msg00034.html and you analyzed it in https://lists.gnu.org/archive/ht

Re: coreutils-9.3.147-d553ab on CentOS 5

On CentOS 5 (64-bit mode), I see 4 test failures: FAIL: tests/sort/sort-continue FAIL: tests/sort/sort-merge-fdlimit FAIL: tests/cp/attr-existing FAIL: tests/mv/part-symlink Find attached the log file. centos5-test-suite.log.gz Description: application/gzip

Re: coreutils-9.3.147-d553ab on CentOS 5 and Solaris 10

2928f335beac2bc6baa5f31 Mon Sep 17 00:00:00 2001 From: Bruno Haible Date: Sun, 27 Aug 2023 01:10:13 +0200 Subject: [PATCH] maint: Fix link errors of sort, split on CentOS 5 and Solaris 10 * src/local.mk (src_sort_LDADD, src_split_LDADD): Add $(CLOCK_TIME_LIB). --- src/local.mk | 2 ++ 1 file

Re: coreutils-9.3.147-d553ab on AIX 7.3.1

On AIX 7.3.1, after fixing the compilation errors https://lists.gnu.org/archive/html/bug-gnulib/2023-08/msg00179.html https://lists.gnu.org/archive/html/coreutils/2023-08/msg00062.html there are test failures. I did 3 builds: - 64-bit with gcc - 64-bit with ibm-clang - 32-bit with ibm-clang

[PATCH] uptime: Simplify following gnulib changes

49333ef26ee91f92e8 Mon Sep 17 00:00:00 2001 From: Bruno Haible Date: Sat, 12 Aug 2023 23:08:58 +0200 Subject: [PATCH] uptime: Simplify following gnulib changes * gnulib: Update to newest. * src/uptime.c: Don't include , . (print_uptime): Don't call sysctl, get_system_info, as Gnulib

[PATCH] pinky,users,who: Optimize read_utmp invocation

boot time. Thus it is worth using now. Find attached a patch. Bruno >From bb38f6e5cc9fbf600f698866e268cdcfba636ea6 Mon Sep 17 00:00:00 2001 From: Bruno Haible Date: Thu, 10 Aug 2023 13:50:57 +0200 Subject: [PATCH] pinky,users,who: Optimize read_utmp invocation When we are only interested

Re: Debugging du v9.3

Pádraig Brady wrote: > With the attached gnulib patch applied you should be able to build with: The patch looks good, except that in this .h file, function parameter names are typically omitted. So, the declaration should better read: void fts_cross_check (FTS const *); Bruno

Re: error: Support the compiler's control flow analysis better (was: copy-file: Silence gcc warnings)

Pádraig Brady wrote: > > Was there a reason to prefer curly braces there, rather than the more > > conventional parentheses? > > > > '$(error_fns) \(.*%s[:"], .*(name|file)[^"]*\);$$' *.c; \ > > I had a slight preference for the curly braces since it was used in a shell > pipeline >

Re: nstrftime.c fails to build due to memset overflow

Pádraig Brady wrote: > I'm going to keep this one. > ... > I've lost many hours to analyzing false positives from this one ... > and I've never found a real issue identified by this warning. But can you please remove the line # FP wth -O0 in nstrftime.c w/gcc 12, and 13 at least since we now hav

Re: nstrftime.c fails to build due to memset overflow

Paul Eggert wrote: > The goal here is software reliability not pacifying compilers But when "gcc -Wall" reports 10 warnings to me, and I don't notice that one of them is an actual bug because I mentally discard all of them "oh these new warnings must all be false positives by gcc", there is someth

Re: coreutils CI

Pádraig Brady wrote: > > @Padraig: you're at least doing some pre-release tests on several platforms, > > but that's no real CI, is it? > > Right. > No real CI at present. > It would be good though, directly for coreutils > and indirectly for gnulib. We have a CI for gnulib at Gitlab [1], but it

Re: coreutils CI

Bernhard Voelker wrote: > Gnulib commit 3f0950f65abb (2023-04-26) not only lead to build time > issues, but also made e.g. coreutils' rm(1) fail: > ... > I can confirm that this gnulib commit d4d8abb39eb0 fixes the issue again. Yep, mistakes do happen. Gnulib has stable branches that can protect s

Re: coreutils-9.3 test-strto[u]ll

Hi Kamil, > I had to revert the following gnulib commit to be able to build it > in the Fedora Rawhide buildroot: > https://git.savannah.gnu.org/gitweb/?p=gnulib.git;a=commitdiff;h=4e38f4a0c65d4d19883b404a84169088b84b60d2 > > ... where it was failing with: > > FAIL: test-strtoll > ==

Re: coreutils-9.1.198-e68b1.tar.xz on Linux/sparc64

Hi Sam, > The test-pthread-cond test fails for me. > > gnulib-tests/test-pthread-cond.log doesn't have much detail: > ``` > Starting test_pthread_cond_wait ... OK > Starting test_pthread_cond_timedwait ...FAIL test-pthread-cond (exit > status: 134) > ``` Is it reproducible? I'm asking because o

Re: nstrftime.c fails to build due to memset overflow

Hi Marcus, Please try to understand Paul's arguments. > Sorry to respectfully contradict you here: introducing a macro really > doesn't do readability and immediate clarity of effect any better than > a commented zero-initialization. Apparently Paul and you have different ways of reading code,

Re: coreutils-9.1.198-e68b1.tar.xz on AIX 7.1

Pádraig Brady wrote: > A fix for the cksum-raw issue is attached > (and tested to work on sparc64). It fixes it also on AIX. Bruno

Re: coreutils-9.1.198-e68b1.tar.xz on AIX 7.1

Pádraig Brady wrote: > It would be good to get output from > make TESTS=tests/misc/tee.sh VERBOSE=yes SUBDIRS=. check > to get more info on the tee one, Attached here. = GNU coreutils 9.1.198-e68b1: ./tests/test-suite.log

Re: coreutils-9.1.198-e68b1.tar.xz on other platforms

All tests pass on the following platforms: Linux Ubuntu 22.04 Debian 9.1 Debian 11.1 CentOS 8-stream CentOS Stream 9 OpenSUSE Leap 15.2 Manjaro 17 macOS 12.5 OpenBSD 7.2 with cc Solaris 11 OpenIndiana (*) (*) Here there 1 gnulib-tests failure, already fixed in gnulib: https://git.sa

Re: coreutils-9.1.198-e68b1.tar.xz on Solaris 11

On Solaris 11.4, I get * 1 test failure in both 32-bit and 64-bit mode: FAIL: tests/misc/stty-row-col * 2 test failures only in 32-bit mode: FAIL: tests/misc/sort-continue FAIL: tests/misc/sort-merge-fdlimit Find attached the log files. solaris114-32-test-suite.log.gz Description: app

Re: coreutils-9.1.198-e68b1.tar.xz on FreeBSD

On FreeBSD 13.1 (x86_64), there is 1 test failure: FAIL: tests/misc/head-write-error The log file is attached. = GNU coreutils 9.1.198-e68b1: ./tests/test-suite.log = # TOTAL: 641 #

Re: coreutils-9.1.198-e68b1.tar.xz on NetBSD

On NetBSD 9.0 (x86_64), I see 5 errors: FAIL: tests/misc/date-debug FAIL: tests/misc/date-tz FAIL: tests/misc/expr-multibyte FAIL: tests/tail-2/pipe-f2 FAIL: tests/touch/not-owner The log file is attached. = GNU coreutils 9.1.198-e68b1: .

Re: coreutils-9.1.198-e68b1.tar.xz on AIX 7.2

On AIX 7.2 (32-bit mode, with gcc 7.2.0), I see 8 test failures: # FAIL: 7 FAIL: tests/misc/help-version FAIL: [ FAIL: tests/misc/cksum-raw FAIL: tests/misc/sort-continue FAIL: tests/misc/stdbuf FAIL: tests/misc/tee FAIL: tests/install/basic-1 FAIL: tests/mkdir/p-3 Find attached the log file. ==

Re: coreutils-9.1.198-e68b1.tar.xz on AIX 7.1

On AIX 7.1 (64-bit mode, with AIX xlc), there are - 5 test failures, reported as 'FAIL: 4', - 5 errors. # FAIL: 4 FAIL: tests/misc/help-version FAIL: [ FAIL: tests/misc/cksum-raw FAIL: tests/misc/sort-continue FAIL: tests/misc/tee # ERROR: 5 ERROR: tests/rm/rm-readdir-fail ERROR: tests/cp/nf

Re: coreutils-9.1.198-e68b1.tar.xz on Solaris 10

On Solaris 10 (x86_64, 64-bit mode, Sun cc) 3 tests fail: FAIL: tests/misc/sort-continue FAIL: tests/misc/sort-merge-fdlimit FAIL: tests/misc/tac Find attached the log file. = GNU coreutils 9.1.198-e68b1: ./tests/test-suite.log ==

Re: [platform-testers] new snapshot available: coreutils-9.1.198-e68b1.tar.xz

#x27; is used uninitialized whenever '&&' condition is false [-Wsometimes-uninitialized] ../src/test.c:595:12: warning: variable 'value' is used uninitialized whenever '&&' condition is false [-Wsometimes-uninitialized] They go away with the attached patc

Re: nstrftime.c fails to build due to memset overflow

Pádraig Brady wrote: >return cnt == digest_bin_bytes; > } > > +# pragma GCC diagnostic ignored "-Wmaybe-uninitialized" > static bool > digest_check (char const *checkfile_name) > { Note that this pragma produces a warning with GCC versions < 4.7: GCC 4.6.4: foo.c:1:10: warning: unknown

Re: Improve support for ACLs in coreutils (ls & chmod) following the Solaris way

Pádraig Brady wrote: > So as we see there are lots of "additional attributes" > with dedicated programs to manipulate them. > What's the big advantage of merging with ls and chmod, > over the current situation of separate utilities? In [1] I'm arguing: * A feature that has impact on security, a

Re: Improve support for ACLs in coreutils (ls & chmod) following the Solaris way

Ondrej Valousek wrote: > Is this something I would find support in both coreutils and Gnulib? Regarding Gnulib: Yes, for code that has its proper place in Gnulib, you have my support. Bruno

Re: coreutils-9.0.193-54bec on Solaris 11.4

Pádraig Brady wrote: > The attached fixes this issue in my testing. [tail-f-pipe-solaris.patch] I confirm, on Solaris 11.4: It fixes the test failure and does not cause test regressions. Bruno

Re: coreutils-9.0.193-54bec on Solaris 11.4: stty

Pádraig Brady wrote: > > > > FAIL: tests/misc/stty-row-col > > = > > Any of the commands like `stty rows 40 columns 80` are being silently ignored. 'stty size' retrieves the info from an ioctl(1,TIOCGWINSZ) call and an ioctl(0,TIOCGWINSZ) call. Even the value of envir

Re: coreutils-9.0.207-b411f on Solaris 11.4 32-bit

> On Solaris 11.4 (x86_64, in 64-bit mode), there are 4 test failures: On the same machine, in 32-bit mode, configured with option --disable-year2038, there are two additional test failures: FAIL: tests/misc/sort-continue == ++ initial_cwd_=/export/home/bruno/coreuti

Re: coreutils-9.0.207-b411f on Solaris 11.4

Paul Eggert wrote: > This should all work on Solaris 11. Yes, I confirm that with these patches included, on Solaris 11.4 * the test failures > > FAIL: tests/install/basic-1 > > === > > FAIL: tests/mv/diag > > === are now gone, * ther

Re: coreutils-9.0.193-54bec on Solaris 11.4

Pádraig Brady wrote: > FAIL: tests/install/basic-1 > === > FAIL: tests/mv/diag > === > The attached addresses these two test failures in my testing. Confirmed: It fixes these two test failures also on Solaris 11.4. Bruno

Re: coreutils-9.0.193-54bec on Cygwin

Pádraig Brady wrote: > Hardwired SYSTEMROOT and WINDIR env vars are causing this to fail. > The attached explicitly unsets these for the test, Indeed; this patch fixes the failure. Bruno

  1   2   3   >