[mxfp8 moe training] integrate triton quant/dequant kernels into mxfp8 all to all

[danielvegamyhre]

Stacked PRs:

• [mxfp8 moe training] bench and profile mxfp8 a2a fwd and bwd separately #3203
• [mxfp8 moe training] improve unit tests #3201
• ->[mxfp8 moe training] integrate triton quant/dequant kernels into mxfp8 all to all #3197
• [mxfp8 moe training] add triton kernel for mxfp8 dequantization #3195

---

[mxfp8 moe training] integrate triton quant/dequant kernels into mxfp8 all to all

Test plan

• pytest test/prototype/moe_training/mxfp8/test_mxfp8_a2a.py -k ToMXFP8AllToAllVDequantTest -s

Benchmarks

Flat perf

input_shape         num_splits    fwd_bf16_ms    fwd_mxfp8_ms    bwd_bf16_ms    bwd_mxfp8_ms
----------------  ------------  -------------  --------------  -------------  --------------
(16, 8192, 5120)             4        2.38517         2.59848         7.5735          7.7693

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/3197

• 📄 Preview Python docs built from this PR

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

❌ 1 New Failure, 1 Unrelated Failure

As of commit 710192d with merge base b644211 ():

NEW FAILURE - The following job has failed:

• Run 1xH100 Tests / test (H100, linux.aws.h100, --pre torch torchvision torchaudio fbgemm-gpu-genai --index-url https... / linux-job (gh)
  RuntimeError: Command docker exec -t b351dffcd246a8f0dc5bd50bd9ca5a0df47c1e52f3ace8a8df66011ddba780d4 /exec failed with exit code 1

BROKEN TRUNK - The following job failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

• Run Regression Tests / test-nightly (CUDA Nightly, linux.g5.12xlarge.nvidia.gpu, --pre torch --index-url https://downloa... / linux-job (gh) (trunk failure)
  test/dtypes/test_nf4.py::TestComm::test_comm

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