* NEWS: Mention the bug fix.
* src/nice.c (MIN_ADJUSTMENT): Set to 0 on the Hurd with glibc ≤ 2.42.
(MAX_ADJUSTMENT): Set to (2 * NZERO - 2) on the Hurd with glibc ≤ 2.42.
(main): Clamp the niceness to be greater or equal to MIN_ADJUSTMENT and
less than or equal to MAX_ADJUSTMENT.
* tests/nice/nice.sh: Add some tests for the Hurd's ranges.
* gnulib: Update to the latest gnulib to pull in the fix
(v1.0-2406-g89f63027de) to allocate the $PATH processing memory
before the vfork call, which is required on ppc and sparc.
* tests/install/basic-1.sh: Ensure posix_spawnp() suports a large $PATH,
which needs careful handling with vfork() as detailed in gnulib.
Make the documentation more accurate following v9.3-80-g5e1e0993b
* doc/coreutils.texi (cksum common options): State that --text and
--binary are actually supported, but only to support emulation
of the legacy utils.
Many people are used to seeing ISO 8601 dates using a period separating
seconds and nanoseconds. This behavior seems to be worth documenting
given the bug reports:
https://bugs.gnu.org/63119https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1118970
* doc/coreutils.texi (Options for date): Mention that
'date --iso-8601=ns' uses a comma as a separator, following the
preference of ISO 8601. Give an example of how to get an ISO 8601 date
with a period separator.
* src/copy-file-data.c (lseek_copy): hole_start is initialized
only when ext_start == ipos.
(infer_scantype): Update the hole_start initialization to
the more logically correct POS, even though that init
is only needed to suppress a -Wmaybe-uninitialized warning.
Note gcc 15.2 at least doesn't seem to need that suppression.
It was seen that gnulib's read_file_system_list may use fopen instead
of open. Adjust the df(1) tests to replace both library functions.
* tests/df/no-mtab-status.sh: Change the shared library code invoked
via LD_PRELOAD to override both fopen and open. While at it, perform
varargs processing only when path is not "/proc/self/mountinfo".
* tests/df/skip-duplicates.sh: Likewise.
Two df(1) tests were skipped (since commit ee367bd38dac), because
the build of the shared library in those tests failed.
+ gcc -Wall -shared --std=gnu99 -fPIC -O2 k.c -o k.so -ldl
k.c: In function 'open':
k.c:37:7: error: implicit declaration of function 'streq'; did you \
mean 'strsep'? [-Wimplicit-function-declaration]
37 | if (streq (path, "/proc/self/mountinfo"))
| ^~~~~
| strsep
Gnulib streq is not available in the tests.
* tests/df/no-mtab-status.sh: Replace "streq" by "0==strcmp" in the
shared library source.
* tests/df/skip-duplicates.sh: Likewise.
This improvement is due to changes to Gnulib's fprintftime module.
* NEWS: Mention the improvement.
* src/getlimits.c (OFF64_T_MAX, OFF64_T_MIN): New macros.
(main): Print them.
* tests/misc/write-errors.sh: Call getlimits_. Add a date invocation.
* src/pr.c (init_header): Do not report an nstrftime EOVERFLOW
error as memory exhaustion. Instead, output the time as an
integer. Also, work even if nstrftime (nullptr, SIZE_MAX, ...)
would return PTRDIFF_MAX which means adding 1 would overflow..
* src/local.mk: Revert v9.7-322-gc2e1816a5, instead relying
on the more focused v9.8-79-g532cd66af. When built with
--disable-nls on macOS this will result in only some commands
being linked with INTL_MACOSX_LIBS, thus resulting in env(1)
at least not setting a __CF_USER_TEXT_ENCODING envirnoment variable.
A more precise fix than commit v9.7-322-gc2e1816a5
* configure.ac (USE_NLS): new AM_CONDITIONAL.
* src/local.mk: add @INTL_MACOSX_LIBS@ to <prog>_LDADD that need it.
* src/copy-file-data.c (copy_file_data): Only give the
POSIX_FADV_SEQUENTIAL hint when we _know_ we'll definitely
use a read/write loop to copy the data. Also only apply
the hint to the whole file, as we've seen OpenZFS at least
special case that.
(sparse_copy): Update stale comment.
* src/copy-file-data.c (infer_scantype): Fall back to a plain copy
if SEEK_HOLE indicates non-sparse, as zero copy avoids copy offload.
This was seen with transparently compressed files on OpenZFS.
* tests/cp/sparse-perf.sh: Add a test case even though it might
only trigger on compressed file systems that don't support reflink.
* NEWS: Mention the bug fix.
Addresses https://github.com/coreutils/coreutils/issues/122
* src/copy-file-data.c (): pass 0 to posix_fadvise to indicate to EOF.
coreutils 9.8 used OFF_T_MAX instead, which triggered OpenZFS 2.2.2
at least to synchronously (decompress and) populate the page cache.
Addresses https://github.com/coreutils/coreutils/issues/122
* NEWS: Remove timeout from the list of programs that use posix_spawn.
* bootstrap.conf (gnulib_modules): Remove posix_spawnattr_setsigmask.
* src/timeout.c: Don't include spawn.h.
(main): Use fork and execvp instead of posix_spawn.
This reverts commit dac96ce3e3a2424919fea6a17978b6e91e575d86.
* NEWS: Mention that we don't fall-back to executing /bin/sh <script>
for malformed scripts that don't start with #!..., or any executable
that returns ENOEXEC in general.
* bootstrap.conf (gnulib_modules): Add findprog-in.
* src/sort.c: Include findprog.h.
(pipe_child): Look up the compress_program in $PATH and report errors
such as ENOENT or EACCES before invoking posix_spawnp.
This avoids inconsistency on systems that emulate posix_spawn through
fork/exec, as they would otherwise treat such a failure as a generic
failure and fail the sort, rather than continuing without compression.
* src/sort.c (main): Ignore SIGPIPE so we've more control over
how we handle for stdout and compression programs.
(sort_die): Handle EPIPE from stdout and mimic a standard SIGPIPE,
otherwise reverting to a standard exit(SORT_FAILURE);
* tests/sort/sort-compress-proc.sh: Add a test case.
* NEWS: Mention the bug fix.
* tests/date/resolution.sh: Fix comparison on systems with less than
nano second reslution, where we use sed to discard the redundant
trailing zeros output by date --resolution.
Reported by Bruno Haible on macOS.
* src/numfmt.c (newline_or_blank): Explicitly ensure
we don't match NBSP as on platforms like NetBSD 10 or Solaris 11,
NBSP is considered a blank character.
This should have been part of commit v9.8-39-g8bc11f80a
Solaris 11 test failure reported by Bruno Haible.
* NEWS: Mention the improvement. Consolidate the posix_spawn
improvements into one item.
* bootstrap.conf (gnulib_modules): Add posix_spawnattr_setsigmask.
* src/timeout.c: Include spawn.h.
(main): Setup signals using a posix_spawnattr_t object. Use posix_spawn
instead of fork and execvp.
The build can be force enabled with --with-selinux and vice versa.
* build-aux/gen-lists-of-programs.sh: Move chcon and runcon
to the list of optional programs.
* configure.ac: Only enable chcon and runcon if selinux.h is available.
* NEWS: Mention the Build-related change.
Fixes https://github.com/coreutils/coreutils/issues/121
* NEWS: Mention the improvement. Mention that 'sort' will continue
without compressing temporary files if the program specified by
--compress-program cannot be executed.
* doc/coreutils.texi (sort invocation): Document the behavior when the
program specified by --compress-program cannot be executed.
* src/sort.c: Include spawn.h.
(MAX_FORK_TRIES_COMPRESS, MAX_FORK_TRIES_DECOMPRESS): Remove definition.
(MAX_TRIES_COMPRESS, MAX_TRIES_DECOMPRESS): New definitions based on
MAX_FORK_TRIES_COMPRESS and MAX_FORK_TRIES_DECOMPRESS.
(async_safe_die): Remove function.
(posix_spawn_file_actions_move_fd): New function.
(pipe_fork): Remove function.
(pipe_child): New function based on pipe_fork. Return an error number
instead of a pid. Use posix_spawnp instead of calling fork and expecting
the caller to exec.
(maybe_create_temp): Call pipe_child instead of pipe_fork. Print a
warning to standard error if --compress-program cannot be executed and
the error is different than the previous call. Remove code for the child
process.
(open_temp): Remove code for the child process. Improve error message.
* tests/sort/sort-compress.sh: Add a test case for when the program
specified by --compress-program does not exist.
* src/install.c: Remove ignore-value.h include.
(strip): Don't use ignore_value on posix_spawnattr_destroy since it
isn't declared with the warn_unused_result attribute. Pass the checked
pointer to posix_spawnattr_destroy instead of the variable it points to.
