- Add upstream patch (coreutils-copy-fix-selinux-existing-dirs.patch):
cp -a: set the correct SELinux context on already existing
destination directories (rh#1045122).
- Merge I18n fixes from Fedora (coreutils-i18n.patch):
* sort: fix sorting by non-first field (rh#1003544)
* cut: avoid using slower multi-byte code in non-UTF-8 locales
(rh#1021403, rh#499220).
- Testsuite: skip some tests:
* coreutils-skip-some-sort-tests-on-ppc.patch: Add patch to
skip 2 valgrind'ed sort tests on ppc/ppc64.
* coreutils-skip-gnulib-test-tls.patch: Add patch to skip
the gnulib test 'test-tls' on i586, x86_64, ppc and ppc64.
* coreutils-tests-avoid-FP-cp-cpuinfo.patch: Add patch to skip a
test when cp fails for /proc/cpuinfo which happens on aarch64.
* coreutils-tests-shorten-extreme-factor-tests.patch: Add patch
to skip most of the extreme-expensive factor tests.
OBS-URL: https://build.opensuse.org/request/show/213254
OBS-URL: https://build.opensuse.org/package/show/Base:System/coreutils?expand=0&rev=221
The "gnulib" package contains a number of floating-point test cases
that do not work correctly with the PowerPC long double ("double double")
format. These tests "accidentally" succeeded anyway in the big-endian
variant, but are now actually failing in little-endian mode.
As is usual for gnulib, those tests end up copied into the source code
of various packages that use gnulib, including coreutils, findutils,
grep, and libunistring.
A patch to fix the tests for ppc64le has been submitted to upstream
to the bug-gnulib mailing list. We'll work with upstream of the
other affected packages to make sure the copies are refreshed.
- coreutils-gnulib-tests-ppc64le.patch: Fix imported gnulib long double
math tests for little-endian PowerPC. (forwarded request 211829 from uweigand)
OBS-URL: https://build.opensuse.org/request/show/212043
OBS-URL: https://build.opensuse.org/package/show/openSUSE:Factory/coreutils?expand=0&rev=98
The "gnulib" package contains a number of floating-point test cases
that do not work correctly with the PowerPC long double ("double double")
format. These tests "accidentally" succeeded anyway in the big-endian
variant, but are now actually failing in little-endian mode.
As is usual for gnulib, those tests end up copied into the source code
of various packages that use gnulib, including coreutils, findutils,
grep, and libunistring.
A patch to fix the tests for ppc64le has been submitted to upstream
to the bug-gnulib mailing list. We'll work with upstream of the
other affected packages to make sure the copies are refreshed.
- coreutils-gnulib-tests-ppc64le.patch: Fix imported gnulib long double
math tests for little-endian PowerPC.
OBS-URL: https://build.opensuse.org/request/show/211829
OBS-URL: https://build.opensuse.org/package/show/Base:System/coreutils?expand=0&rev=217
- Update I18N patch from Fedora:
(coreutils-i18n.patch)
* sort: fix multibyte incompabilities (rh#821264)
* pr -e, with a mix of backspaces and TABs, could corrupt the
heap in multibyte locales (analyzed by J.Koncicky)
* path in the testsuite to cover i18n regressions
* Enable cut and sort-merge perl tests for multibyte as well
- Refresh longlong-aarch64.patch. (forwarded request 209118 from bernhard-voelker)
OBS-URL: https://build.opensuse.org/request/show/209317
OBS-URL: https://build.opensuse.org/package/show/openSUSE:Factory/coreutils?expand=0&rev=95
- Update I18N patch from Fedora:
(coreutils-i18n.patch)
* sort: fix multibyte incompabilities (rh#821264)
* pr -e, with a mix of backspaces and TABs, could corrupt the
heap in multibyte locales (analyzed by J.Koncicky)
* path in the testsuite to cover i18n regressions
* Enable cut and sort-merge perl tests for multibyte as well
- Refresh longlong-aarch64.patch.
OBS-URL: https://build.opensuse.org/request/show/209118
OBS-URL: https://build.opensuse.org/package/show/Base:System/coreutils?expand=0&rev=209
- Undo the previous change.
Remove configure options gl_cv_func_printf_directive_n and
gl_cv_func_printf_infinite_long_double again because of constant
factory build failures on x86_64 and i586. The argument for
adding them was that the fortify checks would be bypassed
by the gnulib "reimplementation of printf", but that is not
the case: instead, gnulib just adds some wrapping code to ensure
a consistent behaviour on all supported platforms. (forwarded request 184097 from bernhard-voelker)
OBS-URL: https://build.opensuse.org/request/show/184102
OBS-URL: https://build.opensuse.org/package/show/openSUSE:Factory/coreutils?expand=0&rev=92
- Undo the previous change.
Remove configure options gl_cv_func_printf_directive_n and
gl_cv_func_printf_infinite_long_double again because of constant
factory build failures on x86_64 and i586. The argument for
adding them was that the fortify checks would be bypassed
by the gnulib "reimplementation of printf", but that is not
the case: instead, gnulib just adds some wrapping code to ensure
a consistent behaviour on all supported platforms.
OBS-URL: https://build.opensuse.org/request/show/184097
OBS-URL: https://build.opensuse.org/package/show/Base:System/coreutils?expand=0&rev=201
- Add systemd to build and use requires
- Make pam configuration for command su using the systemd login
manager for ordinary users as this allows to use all services
provided by systemd login manager
- change the buildrequire system to pkgconfig(systemd) to fix bootstrap
OBS-URL: https://build.opensuse.org/package/show/Base:System/coreutils?expand=0&rev=192
* src/join.c: Instead of usig unreliable alloca() stack allocation,
use heap allocation via xmalloc()+free().
(coreutils-i18n.patch, from Philipp Thomas <pth@suse.de>)
- Avoid segmentation fault in "sort -d" and "sort -M" with long line input
(bnc#798538, VUL-1)
* src/sort.c: Instead of usig unreliable alloca() stack allocation,
use heap allocation via xmalloc()+free().
(coreutils-i18n.patch, from Philipp Thomas <pth@suse.de>)
OBS-URL: https://build.opensuse.org/package/show/Base:System/coreutils?expand=0&rev=173
