bug#65310: test failure on Alpine Linux: tests/sort/sort-debug-keys

2023-08-15 Thread Bruno Haible
Hi, Doing "make check" of current coreutils on Alpine Linux 3.18, I see a test failure that I didn't see with the coreutils-9.3 release: FAIL: tests/sort/sort-debug-keys I'm attaching the relevant part of tests/test-suite.log. Bruno ==

bug#65255: uptime's boot time is inconsistent after VM sleep & resume

2023-08-15 Thread Bruno Haible
I wrote: > > I'll provide a NEWS > > entry afterwards, that summarizes the changes in coreutils + gnulib on the > > various platforms. > Paul Eggert wrote: > Thanks, this looks good. A NEWS entry would be welcome. Actually, it's 4 entries: * The 'uptime' program is now being built on FreeBSD,

bug#65310: test failure on Alpine Linux: tests/sort/sort-debug-keys

2023-08-15 Thread Pádraig Brady
On 15/08/2023 14:14, Bruno Haible wrote: Hi, Doing "make check" of current coreutils on Alpine Linux 3.18, I see a test failure that I didn't see with the coreutils-9.3 release: FAIL: tests/sort/sort-debug-keys I'm attaching the relevant part of tests/test-suite.log. In the log we have:

bug#65255: uptime's boot time is inconsistent after VM sleep & resume

2023-08-15 Thread Bruno Haible
Paul Eggert wrote: > Thanks for the further comments. I installed your patch, along with the > attached additional patches. The other patch I mentioned, from https://lists.gnu.org/archive/html/coreutils/2023-08/msg00028.html , is also needed, for the "VM saved/sleep" change on NetBSD, OpenBSD,

bug#65310: test failure on Alpine Linux: tests/sort/sort-debug-keys

2023-08-15 Thread Bruno Haible
Pádraig Brady wrote: > The fact that ',' isn't used as the decimal point is surprising, > and is what's causing the test to fail. Ah, I see. Yes, for some tests one only needs a fr_FR.UTF-8 that supports UTF-8 in LC_CTYPE, whereas for other tests it also needs to support the LC_NUMERIC category

bug#65269: Possible null pointer dereference on the function cycle_check in rm

2023-08-15 Thread Paul Eggert
On 2023-08-14 00:05, Haoxin Tu wrote: if the function `fts_read` get a return value of NULL and the malloc from `fts->fts_cycle.state = malloc (sizeof *fts->fts_cycle.state)` (Line 62 in fts_cycle.c) is NULL, the pointer `fts->fts_cycle.state` will still keep 0 before the free operation `free

bug#65255: uptime's boot time is inconsistent after VM sleep & resume

2023-08-15 Thread Paul Eggert
Thanks for the further comments. I installed your patch, along with the attached additional patches. The first makes 'uptime' a bit more resilient in the case of utmp and other failures, and the second adds NEWS items as per your comments.From 1ea34cbf6a235f2436a3265ab9ded6f04748051e Mon Sep 17

bug#65331: tests/df/skip-rootfs.sh fails on WSL2

2023-08-15 Thread mytec
Hi, On Windows WSL2 (1.2.5.0) running Ubuntu 22.04.3 LTS one test failed. I downloaded a source release (coreutils-9.3) and did: ./configure, make, make check. Testsuite summary for GNU coreutils 9.3

bug#65255: uptime's boot time is inconsistent after VM sleep & resume

2023-08-15 Thread Paul Eggert
On 2023-08-15 14:23, Bruno Haible wrote: The other patch I mentioned, from https://lists.gnu.org/archive/html/coreutils/2023-08/msg00028.html , is also needed, for the "VM saved/sleep" change on NetBSD, OpenBSD, Minix, as far as I understand. Also, a typo in NEWS: s/Minux/Minix/ Thanks, I