[fp8 blockwise] wrap triton quantization kernels in custom ops for torch.compile compatibility

[danielvegamyhre]

Stacked PRs:

• ->[fp8 blockwise] wrap triton quantization kernels in custom ops for torch.compile compatibility #2829
• [fp8 blockwise] load 2d chunks for groupwise quant to enable coalesced gmem accesses #2827
• use shared bench + profile utils in blockwise fwd bwd bench script #2826
• integrate torch._scaled_mm into Float8BlockwiseLinear and add bench script #2785

---

[fp8 blockwise] wrap triton quantization kernels in custom ops for torch.compile compatibility

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/2829

• 📄 Preview Python docs built from this PR

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

❗ 1 Active SEVs

There are 1 currently active SEVs. If your PR is affected, please view them below:

• Multiple CI trunk failures after landing https://github.com/pytorch/pytorch/pull/161002

✅ No Failures

As of commit b1bd84b with merge base 3bf21d0 ():
💚 Looks good so far! There are no failures yet. 💚

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