[CI] Add MacOS-M2-15 as MPS test target on trunk #149900
Conversation
Now that we have runners allocated by AWS
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/149900
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (1 Unrelated Failure)As of commit 27b9921 with merge base 2385539 ( UNSTABLE - The following job is marked as unstable, possibly due to flakiness on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
| @@ -109,6 +109,7 @@ jobs: | |||
| { include: [ | |||
| { config: "mps", shard: 1, num_shards: 1, runner: "macos-m1-13" }, | |||
| { config: "mps", shard: 1, num_shards: 1, runner: "macos-m1-14" }, | |||
| { config: "mps", shard: 1, num_shards: 1, runner: "macos-m2-15" }, | |||
There was a problem hiding this comment.
General question here, how does sharding work for MPS? Does this mean that we replicate the entire test suite for MPS 3 across 3 different hardware shards?
There was a problem hiding this comment.
Yes, MPS implementation at the moment at the OS features, so we run the whole testsuite across different OS versions . I.e. it's similar to linux-focal and linux-jammy tests, but we don't need to rebuild for different OS version, as MacOS makes pretty strong compatibly guarantees and opt-in for new APIs
Also please note that duration of every MPS shard is still under 30 min
|
@pytorchbot merge -i |
Merge startedYour change will be merged while ignoring the following 2 checks: Check Labels, pull / cuda12.4-py3.10-gcc9-sm75 / test (pr_time_benchmarks, 1, 1, linux.g4dn.metal.nvidia.gpu) Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Now that we have runners allocated by AWS Pull Request resolved: pytorch#149900 Approved by: https://github.com/ZainRizvi, https://github.com/seemethere
Now that we have runners allocated by AWS