Adds extra supported architectures that were available for vanilla
scudo, in preparation for D102543. Hopefully the dust has settled and
7d0a81ca38e427de9b7fb0961ec643b757028131 is no longer an issue.
Details
- Reviewers
cryptoad vitalybuka - Commits
- rGf4ccbaf310f1: [scudo] Add supported architectures.
Diff Detail
- Repository
- rG LLVM Github Monorepo
Event Timeline
Now the bot should cover this archs https://lab.llvm.org/buildbot/#/builders/70/builds/7283
Thanks for that. I'll rip some of the cmake commands out of that and try and drill down the bug on our aarch64 bot cross-compiling for armv7.
Consider disabling the test for arm and landing the patch to avoid regression on working arm tests.
Disable the buggy secondary test until we can figure out what's going on. I don't have access to any hardware that can reproduce the problem, and it doesn't seem to pop up on qemu arm32 user mode emulation. I tried an aarch64 machine, but apparently arm32 support is optional in armv8 and the CPU on that bot doesn't support arm32.
compiler-rt/lib/scudo/standalone/tests/secondary_test.cpp | ||
---|---|---|
36–44 ↗ | (On Diff #346551) | there is a small change that config-ix.cmake will be reverted, so I would keep secondary_test.cpp in a separate patch |
compiler-rt/lib/scudo/standalone/tests/secondary_test.cpp | ||
---|---|---|
36–44 ↗ | (On Diff #346551) | Sounds good to me, I'll split this patch and submit them separately (test first, then cmake). |
Partial commit in 577a80bff8bdf9b26c0f4ff6d1807e43da66ec6a with just the secondary_test.cpp changes.