Add F8 BLAS data types conversion

[malfet]

Stack from ghstack (oldest at bottom):

• Introduce CUDA-only _scaled_mm op #106844
• [BE] Move common logic into cublasCommonArgs #106842
• -> Add F8 BLAS data types conversion #106843

🤖 Generated by Copilot at 54e843d

Float8 types added
to switch on ScalarType -
CUDA version checked.

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/106843

• 📄 Preview Python docs built from this PR
• 📄 Preview C++ docs built from this PR
• ❓ Need help or want to give feedback on the CI? Visit the bot commands wiki or our office hours

Note: Links to docs will display an error until the docs builds have been completed.

❌ 1 New Failure, 2 Unrelated Failures

As of commit 54e843d:

NEW FAILURE - The following job has failed:

• win-vs2019-cpu-py3 / test (default, 3, 3, windows.4xlarge.nonephemeral) (gh)

UNSTABLE - The following jobs failed but were likely due to flakiness present on trunk and has been marked as unstable:

• linux-focal-rocm5.6-py3.8 / test (default, 1, 3, linux.rocm.gpu, unstable) (gh)
• linux-focal-rocm5.6-py3.8 / test (default, 3, 3, linux.rocm.gpu, unstable) (gh)

This comment was automatically generated by Dr. CI and updates every 15 minutes.

[vkuzo]

to clarify, does this bind the unsaturated casts?

[vkuzo]

i.e. if someone does torch.randn(4, 4).cuda().to(torch.float8_e4m3fn), will the casts be hardware accelerated or go through the bit shifting path?

[malfet]

to clarify, does this bind the unsaturated casts?

No, this is a no-op that simple adds BLAS dtypes, let me change the title to clarify what it does a bit better.

Wow, CoPilot did a really great job of summarizing the change.

[malfet]

@pytorchbot merge

[pytorchmergebot]

Merge failed

Reason: This PR needs a release notes: label
If your changes are user facing and intended to be a part of release notes, please use a label starting with release notes:.

If not, please add the topic: not user facing label.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "topic: not user facing"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

Details for Dev Infra team

Raised by workflow job

[malfet]

@pytorchbot merge

[pytorchmergebot]

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here

[pytorchmergebot]

Merge failed

Reason: 1 jobs have failed, first few of them are: trunk / win-vs2019-cpu-py3 / test (default, 3, 3, windows.4xlarge.nonephemeral)

Details for Dev Infra team

Raised by workflow job

[malfet]

@pytorchbot merge -i

[pytorchmergebot]

Merge started

Your change will be merged while ignoring the following 3 checks: trunk / linux-focal-rocm5.6-py3.8 / test (default, 1, 3, linux.rocm.gpu, unstable), trunk / linux-focal-rocm5.6-py3.8 / test (default, 3, 3, linux.rocm.gpu, unstable), trunk / win-vs2019-cpu-py3 / test (default, 3, 3, windows.4xlarge.nonephemeral)

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here