Bug 258812
| Summary: | devel/basu: broken LLD 13 build | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | Ports & Packages | Reporter: | Evgenii Khramtsov <2khramtsov> | ||||||||
| Component: | Individual Port(s) | Assignee: | Jan Beich <jbeich> | ||||||||
| Status: | Closed FIXED | ||||||||||
| Severity: | Affects Only Me | CC: | 2khramtsov, dim, jbeich, moonlapse81 | ||||||||
| Priority: | --- | Keywords: | patch | ||||||||
| Version: | Latest | Flags: | jbeich:
merge-quarterly+
|
||||||||
| Hardware: | Any | ||||||||||
| OS: | Any | ||||||||||
| See Also: | https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=258209 | ||||||||||
| Bug Depends on: | |||||||||||
| Bug Blocks: | 258209 | ||||||||||
| Attachments: |
|
||||||||||
Forgot PORTREVISION again, --amend this, sorry. Comment on attachment 228291 [details]
v1 (use "git am")
I'll probably wait for upstream review. The fix introduces warnings:
../src/libsystemd/sd-bus/bus-error.c:74:31: warning: 'extern' variable has an initializer [-Wextern-initializer]
extern const sd_bus_error_map __start_BUS_ERROR_MAP[] = {};
^
../src/libsystemd/sd-bus/bus-error.c:75:31: warning: 'extern' variable has an initializer [-Wextern-initializer]
extern const sd_bus_error_map __stop_BUS_ERROR_MAP[] = {};
^
> The fix introduces warnings Don't merge this :) tests will also SIGSEGV due to overrun while walking bus_standard_errors[], x11/mako didn't expose this, but tests did. The way to fix this is to rewrite some of bus_standard_errors[] walking logic, I attached WIP at https://lists.sr.ht/~emersion/public-inbox/%3CCANvPQf_5qOdfbe4Tk029yVB6BHipmYfQnuuCVoTUv3N%3D2f8jfw%40mail.gmail.com%3E "test-bus-error" works with standard dbus errors using WIP, but still fails with custom dbus errors. I opted for an upstream issue because I don't know what to do about custom dbus error tests, I don't know dbus stuff. Created attachment 228559 [details]
v2 ("git am")
- test-bus-error passes again;
- package contents did not change on 12.2-13.0/{amd64,i386} and 13.0/aarch64;
- no new warnings on 12.2-13.0/{amd64,i386} and 13.0/aarch64 (via qemu-user-static and native-xtools).
Upstream yet to decide.
(In reply to Evgeniy Khramtsov from comment #4) I believe the maintainer timeout has now expired? (In reply to Dimitry Andric from comment #5) No. - I didn't request maintainer-approval for v2. - I forgot to mark v1 as obsolete after v2. This should be first fixed upstream per comment 2. (In reply to Evgeniy Khramtsov from comment #6) I've pinged upstream but it ended up in a separate thread (despite importing mbox locally to make sure References field is correct). https://lists.sr.ht/~emersion/public-inbox/%3Ca6j4-iinz-wny%40FreeBSD.org%3E Otherwise, v2 looks OK. Let's wait at least 2 days more then land regardless. A commit in branch main references this bug: URL: https://cgit.FreeBSD.org/ports/commit/?id=ffe546db80d95396051300a693d07acdcda457a3 commit ffe546db80d95396051300a693d07acdcda457a3 Author: Evgeniy Khramtsov <evgeniy@khramtsov.org> AuthorDate: 2021-10-09 15:37:45 +0000 Commit: Jan Beich <jbeich@FreeBSD.org> CommitDate: 2021-10-30 02:45:34 +0000 devel/basu: unbreak build with LLD 13 (devel/llvm13 or -CURRENT) ld: error: undefined symbol: __start_BUS_ERROR_MAP >>> referenced by bus-error.c >>> libsystemd_static.a.p/sd-bus_bus-error.c.o:(bus_error_name_to_errno) in archive src/libsystemd/libsystemd_static.a ld: error: undefined symbol: __stop_BUS_ERROR_MAP >>> referenced by bus-error.c >>> libsystemd_static.a.p/sd-bus_bus-error.c.o:(bus_error_name_to_errno) in archive src/libsystemd/libsystemd_static.a >>> referenced by bus-error.c >>> libsystemd_static.a.p/sd-bus_bus-error.c.o:(bus_error_name_to_errno) in archive src/libsystemd/libsystemd_static.a cc: error: linker command failed with exit code 1 (use -v to see invocation) PR: 258812 Obtained from: https://github.com/systemd/systemd/commit/945317a4b69d devel/basu/files/patch-lld13 (new) | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) A commit in branch 2021Q4 references this bug: URL: https://cgit.FreeBSD.org/ports/commit/?id=b1a8e908da76d6f35eef7486f0a30b637cf879b2 commit b1a8e908da76d6f35eef7486f0a30b637cf879b2 Author: Evgeniy Khramtsov <evgeniy@khramtsov.org> AuthorDate: 2021-10-09 15:37:45 +0000 Commit: Jan Beich <jbeich@FreeBSD.org> CommitDate: 2021-10-30 02:46:08 +0000 devel/basu: unbreak build with LLD 13 (devel/llvm13 or -CURRENT) ld: error: undefined symbol: __start_BUS_ERROR_MAP >>> referenced by bus-error.c >>> libsystemd_static.a.p/sd-bus_bus-error.c.o:(bus_error_name_to_errno) in archive src/libsystemd/libsystemd_static.a ld: error: undefined symbol: __stop_BUS_ERROR_MAP >>> referenced by bus-error.c >>> libsystemd_static.a.p/sd-bus_bus-error.c.o:(bus_error_name_to_errno) in archive src/libsystemd/libsystemd_static.a >>> referenced by bus-error.c >>> libsystemd_static.a.p/sd-bus_bus-error.c.o:(bus_error_name_to_errno) in archive src/libsystemd/libsystemd_static.a cc: error: linker command failed with exit code 1 (use -v to see invocation) PR: 258812 Obtained from: https://github.com/systemd/systemd/commit/945317a4b69d (cherry picked from commit ffe546db80d95396051300a693d07acdcda457a3) devel/basu/files/patch-lld13 (new) | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) Created attachment 229762 [details]
use upstream fix ("git am")
A commit in branch main references this bug: URL: https://cgit.FreeBSD.org/ports/commit/?id=175cb372bbfba1f6f56a2710cc8aab40512b1cea commit 175cb372bbfba1f6f56a2710cc8aab40512b1cea Author: Evgeniy Khramtsov <evgeniy@khramtsov.org> AuthorDate: 2021-11-28 03:56:15 +0000 Commit: Jan Beich <jbeich@FreeBSD.org> CommitDate: 2021-11-28 05:13:00 +0000 devel/basu: switch to upstream fix after ffe546db80d9 PR: 258812 devel/basu/Makefile | 1 + devel/basu/distinfo | 2 ++ devel/basu/files/patch-lld13 (gone) | 23 ----------------------- 3 files changed, 3 insertions(+), 23 deletions(-) |
Created attachment 228291 [details] v1 (use "git am") devel/llvm13 as system toolchain can't link devel/basu. The attached workaround (?) fixes this for me. At least x11/mako notifications works fine. I'll file upstream PR today or tomorrow, also need to check if this is the right thing to do.