[llvm] [PAC][CodeGen][ELF][AArch64] Support signed GOT (PR #105798)
Daniil Kovalev via llvm-commits
llvm-commits at lists.llvm.org
Fri Aug 23 09:15:18 PDT 2024
kovdan01 wrote:
I was unable to reproduce the original buildbot failure https://lab.llvm.org/buildbot/#/builders/153/builds/5329. In that build, `ExecutionEngine/JITLink/AArch64/ELF_relocations.s` test fails (see https://github.com/llvm/llvm-project/pull/96164#issuecomment-2275170432).
- When triggering rebuild for that particular build, tests pass: see https://lab.llvm.org/buildbot/#/builders/153/builds/6640
- When triggering force build for branch with changes re-applied, tests pass: see https://lab.llvm.org/buildbot/#/builders/153/builds/6580
- Locally, I've cross-compiled llvm for aarch64 and ran the failing test under qemu for hundreds of times (just in case it's some unstable test which fails not at every time) - the test passes every time
Maybe the failure was somehow related to #100854 (which touched the failing test) being merged several commits earlier than #96164. The build, corresponding to that PR, also passed - see https://lab.llvm.org/buildbot/#/builders/153/builds/5325.
Unfortunately, I can't now say what was the reason of the original failure. Even though the src dir and the build dir were not cleaned before build, that should not make any difference in this case.
Would be glad to see everyone's thoughts on probable reasons of original build failure if you have some. As for now - I just re-submit the patch (currently treating the failure as a strange instability which was not reproduced neither on buildbot nor locally) with a couple of changes described below.
https://github.com/llvm/llvm-project/pull/105798
More information about the llvm-commits
mailing list