Add AArch64 support to the premerge tests #155274
Conversation
|
Did we get quota from AWS for this? |
Yes, we did. Right now, I'm just trying to get a prototype working. |
|
There is still one failing test, but the implementation is complete (including sccache setup). |
Due to a fallback in GDBRemoteCommunicationClient.cpp, on Linux we will assume a PID of 1 if the remote does not respond in some way that tells us the real PID. So if PID 1 happened to be a process that the current user could read, we would try to debug that instead. On my current machine, PID 1 was sshd run by root so we would ignore it. If I changed the fallback to some process ID run by my user, the test would fail. To prevent this, return something to LLDB that tells it there is no remote PID, preventing the host lookup. Fixes llvm#155895
|
This is ready for review. I set the continue-on-error property for the aarch64 tests for now to allow for a stabilization period that doesn't impact users. |
|
Also, because I had to change the name of the tests to include the arch, we will need to update the rulset which requires the precommit tests to pass. |
| - depot-ubuntu-24.04-arm-16 | ||
| - llvm-premerge-linux-runners | ||
| runs-on: ${{ matrix.runs-on }} | ||
| container: |
There was a problem hiding this comment.
A comment here explaining we only explicitly spawn a container on AArch64?
There was a problem hiding this comment.
Is it because the GCP agents are containers themselves and we can't have nested containers?
|
How long does the aarch64 job usually take compared to the x86 job? |
It looks like we're about +40% on a cold build. It's generally not easy to see how fast things will run until we start collecting metrics at scale (although tens of percent should translate). |
Co-authored-by: Louis Dionne <[email protected]>
The tests are configured to always report a passing result temporarily until we are confident that they are stable. --------- Co-authored-by: ultimateforce21 <[email protected]> Co-authored-by: David Spickett <[email protected]> Co-authored-by: Louis Dionne <[email protected]>
The tests are configured to always report a passing result temporarily until we are confident that they are stable. --------- Co-authored-by: ultimateforce21 <[email protected]> Co-authored-by: David Spickett <[email protected]> Co-authored-by: Louis Dionne <[email protected]>
|
Great news! Thank you folks, and of course AWS for providing a quota. ❤️ |
7 participants
The tests are configured to always report a passing result temporarily until we are confident that they are stable.