Repository navigation
regression: linking with cc failed #130581
Description
- https://crater-reports.s3.amazonaws.com/beta-1.82-1/beta-2024-09-05/reg/dilithium-raw-0.1.0/log.txt
- https://crater-reports.s3.amazonaws.com/beta-1.82-1/beta-2024-09-05/reg/voro_rs-0.3.0/log.txt
- https://crater-reports.s3.amazonaws.com/beta-1.82-1/beta-2024-09-05/reg/pqcrypto-newhope-0.1.2/log.txt
- https://crater-reports.s3.amazonaws.com/beta-1.82-1/beta-2024-09-05/reg/woff-0.3.3/log.txt
- https://crater-reports.s3.amazonaws.com/beta-1.82-1/beta-2024-09-05/reg/mountpoint-s3-crt-sys-0.8.0/log.txt
- https://crater-reports.s3.amazonaws.com/beta-1.82-1/beta-2024-09-05/reg/pblib-rs-0.1.0/log.txt
- https://crater-reports.s3.amazonaws.com/beta-1.82-1/beta-2024-09-05/reg/libwebp-sys-0.9.6/log.txt
- https://crater-reports.s3.amazonaws.com/beta-1.82-1/beta-2024-09-05/reg/nostrdb-0.3.4/log.txt
- https://crater-reports.s3.amazonaws.com/beta-1.82-1/beta-2024-09-05/reg/kangarootwelve_xkcp-0.1.5/log.txt
- https://crater-reports.s3.amazonaws.com/beta-1.82-1/beta-2024-09-05/reg/ion-c-sys-0.4.17/log.txt
there was far more than just these listed crates that had linker errors. It seems like significantly more than prior crater runs which is odd
Activity
ChrisDenton commentedon Sep 19, 2024
cc @NobodyXu in case you have any ideas
saethlin commentedon Sep 19, 2024
Something very fishy is going on here.
libwebp-sysis small, well-maintained and has quite a few downloads, so I'm looking into that.cargo +beta build --testsfails as crater indicates. I've tried every nightly in 2024-08-XX and 2024-09-XX and all of them can build the crate's tests.The
ccversion doesn't seem to matter, I've tried versions back to 1.0.50 which is years old.saethlin commentedon Sep 19, 2024
I also cannot reproduce these linker errors using a toolchain that I've built from the beta branch.
saethlin commentedon Sep 19, 2024
Ah! My confusion is because nightly now uses lld by default, which is a good little linker and takes care of you even if you give it a confusing argument order.
saethlin commentedon Sep 19, 2024
Bisects to #128400
A crater run was done in that PR but the results were ignored because of the same confusion I ran into above #128400 (comment)
apiraino commentedon Oct 7, 2024
As discussed on Zulip is there any actionable here, or can this issue be closed? #128400 was noted in the release notes (unsure if this is enough tho)
lqd commentedon Oct 7, 2024
cc @petrochenkov
apiraino commentedon Nov 25, 2024
Closing since #128400 will be mentioned in the release notes