[FlexAttention] FlexDecoding accuracy discrepancy between XPU and CUDA while compiling torch.ops.higher_order.flex_attention

[hoshibara]

Describe the bug

Issue Description:
While running the XPU FlexDecoding UT, we found a test failure due to a tensor mismatch.

python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod1_head_dims1_page_size_256

We captured the compiled results on both XPU and CUDA, unified their outputs, and found that running the Triton codes generated by UT will get different results.

triton-code.zip

Environment details

You can refer to this issue to setup a reproducing environment:
#3518

[liangan1]

@vlad-penkin is there any updates about this issue?

[chengjunlu]

Is this issue overlapped with #3631
Can we close this to as a duplicated issue?

[hoshibara]

Is this issue overlapped with #3631 Can we close this to as a duplicated issue?

Yes, I kept this issue open only because it has already been tagged.