[MPS] Extend torch.mm/torch.bmm to integral types
#145809
Closed
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Closed
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/145809
Note: Links to docs will display an error until the docs builds have been completed. ⏳ No Failures, 49 PendingAs of commit b15f4af with merge base 6b41f31 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
dcci
approved these changes
Jan 28, 2025
|
Debug script import torch
import os
os.environ["MTL_CAPTURE_ENABLED"]="1"
torch.manual_seed(42)
x = torch.testing.make_tensor([10,], dtype=torch.int16, device="mps")
y = torch.testing.make_tensor([5, 10, 5], dtype=torch.int16, device="mps")
# x = torch.ones((5, 1, 10,), dtype=torch.int16, device="mps")
#x = torch.randint(-1, 3, (10,), dtype=torch.int16, device="mps")
#y = torch.arange(250, dtype=torch.int16, device="mps").reshape(5, 10, 5)
#with torch.device("mps"):
# x=torch.arange(15, dtype=torch.int16).reshape(3, 5, 1)
# y=torch.arange(18, dtype=torch.int16).reshape(3, 1, 6)
# print(x.stride(), y.stride())
#with torch.mps.profiler.metal_capture("bmm"):
# z = torch.matmul(x, y)
print(x)
print(y)
z = torch.matmul(x, y)
print(x.stride(), y.stride())
print(z)
print(torch.matmul(x.cpu(), y.cpu())) |
torch.mm to integral typestorch.mm/torch.bmm to integral types
|
@pytorchbot merge -f "Lint + MPS are green" |
Merge startedYour change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Please use Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Stack from ghstack (oldest at bottom):
torch.mm/torch.bmmto integral types #145809By using
naive_mmkernel, but make sure that accumulation is done over int32 for smaller int types (and float for half and bfloat) as well as addingnavie_bmmthat follows the same pattern.Remove stale restriction on
torch.dot(which works fine on MacOS-14/15)This also enables integer op flavors for:
addmveinsuminnerlinalg.multi_dotmatmulmvtensordot