Could we please have a hotfix for sys-devel/gcc-10.2.0 to include the patch for the upstream bug linked in [URL]? The patch[1] applies cleanly to sys-devel/gcc-10.2.0-r3 and resolves the issue in sys-apps/smartmontools discussed at [2]. [1] https://gcc.gnu.org/git/?p=gcc.git;a=patch;h=0dbfa88edafbe913a7a9099246041e0190aa3948;hp=7e53436da1902061af797d0aaa744c52bd9829ae [2] https://bugzilla.redhat.com/show_bug.cgi?id=1896161 To summarize: sys-apps/smartmontools-7.1 is unable to directly address drives behind a 3ware RAID controller due to a miscompilation of a strncmp function call caused by a bug in GCC 10 that has been patched upstream and is to be included in GCC 10.3. While we wait for the release of GCC 10.3, could we get a revbump of GCC 10.2 to include the patch for this bug and maybe also Bug 746749? That bug is currently marked as "wait for upstream release" but, like this bug, is causing miscompilations today and could easily be patched in a hotfix revbump.
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/proj/gcc-patches.git/commit/?id=9b342fdeb8588681bb106e072cefd481a045fb21 commit 9b342fdeb8588681bb106e072cefd481a045fb21 Author: Sergei Trofimovich <slyfox@gentoo.org> AuthorDate: 2020-12-01 19:16:33 +0000 Commit: Sergei Trofimovich <slyfox@gentoo.org> CommitDate: 2020-12-01 19:16:33 +0000 10.2.0: backport strncmp() folding Reported-by: Matt Whitlock Bug: https://bugs.gentoo.org/757792 Bug: https://gcc.gnu.org/PR96758 Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org> 10.2.0/gentoo/41_all_strlen-PR96758.patch | 95 +++++++++++++++++++++++++++++++ 10.2.0/gentoo/README.history | 3 + 2 files changed, 98 insertions(+)
Queued patch for next gcc-10.2.0 patchset. It usually takes a few weeks to cut a new release. Meanwhile you can drop a patch to /etc/portage/patches/sys-devel/gcc:10/ and verify if it's enough to fix the underlying bug.
(In reply to Sergei Trofimovich from comment #2) > Meanwhile you can drop a patch to > /etc/portage/patches/sys-devel/gcc:10/ and verify if it's enough to fix the > underlying bug. I did that. (Actually to /etc/portage/patches/sys-devel/gcc-10.2.0-r3 so I don't have to remember to delete it when -r4 is released.) It does solve the underlying issue, as I noted. My smartctl is again able to address the underlying drives in my 3ware RAID.
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=f0e31bd8e8a4d7a67033600a0331a6deb7db1cc4 commit f0e31bd8e8a4d7a67033600a0331a6deb7db1cc4 Author: Sergei Trofimovich <slyfox@gentoo.org> AuthorDate: 2020-12-18 21:03:32 +0000 Commit: Sergei Trofimovich <slyfox@gentoo.org> CommitDate: 2020-12-18 21:04:24 +0000 sys-devel/gcc: 10.2.0: cut 5 patchset Single new patch: + 41_all_strlen-PR96758.patch: fix string comparison against literals Bug: https://bugs.gentoo.org/757792 Package-Manager: Portage-3.0.12, Repoman-3.0.2 Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org> sys-devel/gcc/Manifest | 1 + sys-devel/gcc/gcc-10.2.0-r4.ebuild | 18 ++++++++++++++++++ 2 files changed, 19 insertions(+)
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/proj/gcc-patches.git/commit/?id=e71e59c25315ebda3ebf4014bb108db9dda08a27 commit e71e59c25315ebda3ebf4014bb108db9dda08a27 Author: Sergei Trofimovich <slyfox@gentoo.org> AuthorDate: 2020-12-18 20:57:44 +0000 Commit: Sergei Trofimovich <slyfox@gentoo.org> CommitDate: 2020-12-18 20:57:44 +0000 10.2.0: cut 5 patchset Single new patch: + 41_all_strlen-PR96758.patch: fix string comparison against literals Bug: https://bugs.gentoo.org/757792 Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org> 10.2.0/gentoo/README.history | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
