Skip to content

[FlexAttention] Accuracy issues during running FlexDecoding UT #3631

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
hoshibara opened this issue Mar 10, 2025 · 9 comments · Fixed by #3999
Closed

[FlexAttention] Accuracy issues during running FlexDecoding UT #3631

hoshibara opened this issue Mar 10, 2025 · 9 comments · Fixed by #3999

Comments

@hoshibara
Copy link

hoshibara commented Mar 10, 2025

Describe the bug

python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod1_head_dims1_page_size_256

We captured this case's 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


While running the XPU FlexDecoding UT, we found some failure due to tensors mismatch.

AssertionError: False is not true : XXX Compiled error XXX is greater than ref error XXX by more than XXX.

  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_float16_score_mod1_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims0_score_mod1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims3_score_mod0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims3_score_mod1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims3_score_mod2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims3_score_mod5
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims3_score_mod8
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_max_autotune_with_captured
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_equal_head_dims_score_mod2_float16_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_equal_head_dims_score_mod4_float16_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_equal_head_dims_score_mod7_float16_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod0_BLOCK_SIZE2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod0_BLOCK_SIZE3
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod0_BLOCK_SIZE_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod0_BLOCK_SIZE_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod1_BLOCK_SIZE2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod1_BLOCK_SIZE3
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod1_BLOCK_SIZE_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod1_BLOCK_SIZE_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod2_BLOCK_SIZE2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod2_BLOCK_SIZE3
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod2_BLOCK_SIZE_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod2_BLOCK_SIZE_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod4_BLOCK_SIZE2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod4_BLOCK_SIZE3
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod4_BLOCK_SIZE_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod4_BLOCK_SIZE_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod5_BLOCK_SIZE2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod5_BLOCK_SIZE3
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod5_BLOCK_SIZE_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod5_BLOCK_SIZE_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod7_BLOCK_SIZE2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod7_BLOCK_SIZE3
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod7_BLOCK_SIZE_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod7_BLOCK_SIZE_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod8_BLOCK_SIZE2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod8_BLOCK_SIZE3
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod8_BLOCK_SIZE_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod8_BLOCK_SIZE_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_float16_score_mod0_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_float16_score_mod0_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_float16_score_mod1_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_float16_score_mod2_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_float16_score_mod2_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_float16_score_mod4_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_float16_score_mod4_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_float16_score_mod5_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_float16_score_mod5_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_float16_score_mod7_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_float16_score_mod7_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_float16_score_mod8_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_float16_score_mod8_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_captured_buffers_all_dims_float16
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_captured_buffers_float16
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_captured_scale_float16
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_function_composition_float16
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims0_score_mod0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims0_score_mod2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims0_score_mod4
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims0_score_mod5
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims0_score_mod7
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims0_score_mod8
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims1_score_mod0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims1_score_mod1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims1_score_mod2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims1_score_mod4
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims1_score_mod5
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims1_score_mod7
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims1_score_mod8
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims2_score_mod0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims2_score_mod1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims2_score_mod2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims2_score_mod4
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims2_score_mod5
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims2_score_mod7
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims2_score_mod8
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims3_score_mod0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims3_score_mod1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims3_score_mod2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims3_score_mod4
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims3_score_mod5
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims3_score_mod7
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims3_score_mod8
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims0_score_mod0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims0_score_mod1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims0_score_mod2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims0_score_mod4
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims0_score_mod5
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims0_score_mod7
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims0_score_mod8
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims1_score_mod0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims1_score_mod1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims1_score_mod2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims1_score_mod4
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims1_score_mod5
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims1_score_mod7
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims1_score_mod8
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims2_score_mod0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims2_score_mod1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims2_score_mod2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims2_score_mod4
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims2_score_mod5
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims2_score_mod7
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims2_score_mod8
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims3_score_mod4
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims3_score_mod7
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_load_from_bias_head_seq_batch_float16
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_load_from_bias_seq_batch_float16
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_load_from_bias_seq_only_float16
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_max_autotune
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_divisible_multi_token_offset_mask
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_divisible_multi_token_offset_mask_with_captured_buffer
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_divisible_offset_mask
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_divisible_offset_mask_with_captured_buffer
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_equal_head_dims_score_mod0_float16_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_equal_head_dims_score_mod0_float16_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_equal_head_dims_score_mod1_float16_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_equal_head_dims_score_mod1_float16_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_equal_head_dims_score_mod2_float16_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_equal_head_dims_score_mod4_float16_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_equal_head_dims_score_mod5_float16_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_equal_head_dims_score_mod5_float16_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_equal_head_dims_score_mod7_float16_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_equal_head_dims_score_mod8_float16_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_equal_head_dims_score_mod8_float16_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod0_head_dims1_page_size_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod0_head_dims2_page_size_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod0_head_dims2_page_size_256
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod0_head_dims2_page_size_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod1_head_dims1_page_size_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod1_head_dims2_page_size_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod1_head_dims2_page_size_256
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod1_head_dims2_page_size_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod2_head_dims1_page_size_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod2_head_dims2_page_size_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod2_head_dims2_page_size_256
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod2_head_dims2_page_size_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod4_head_dims1_page_size_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod4_head_dims2_page_size_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod4_head_dims2_page_size_256
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod4_head_dims2_page_size_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod5_head_dims1_page_size_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod5_head_dims2_page_size_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod5_head_dims2_page_size_256
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod5_head_dims2_page_size_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod7_head_dims1_page_size_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod7_head_dims2_page_size_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod7_head_dims2_page_size_256
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod7_head_dims2_page_size_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod8_head_dims1_page_size_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod8_head_dims2_page_size_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod8_head_dims2_page_size_256
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod8_head_dims2_page_size_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_recompile_changed_score_mod_float16
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_seq_masking_float16
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_silu_on_score_float16
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_skip_odd_keys_float16
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_windowed_full_mask_vs_sdpa
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_windowed_full_mask_vs_sdpa_paged_attention
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_windowed_no_mask_vs_sdpa_paged_attention
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_windowed_partial_block_vs_sdpa
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_windowed_partial_block_vs_sdpa_paged_attention

AssertionError: Tensor-likes are not close!

  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod3_BLOCK_SIZE2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod3_BLOCK_SIZE3
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod3_BLOCK_SIZE_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod3_BLOCK_SIZE_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod6_BLOCK_SIZE2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod6_BLOCK_SIZE3
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod6_BLOCK_SIZE_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_different_block_size_float16_score_mod6_BLOCK_SIZE_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_float16_score_mod3_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_float16_score_mod3_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_float16_score_mod6_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_builtin_score_mods_float16_score_mod6_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_do_not_trigger_dynamic_shapes_on_empty_block_mask
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims0_score_mod3
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims0_score_mod6
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims1_score_mod3
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims1_score_mod6
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims2_score_mod3
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims2_score_mod6
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims3_score_mod3
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims1_batch_dims3_score_mod6
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims0_score_mod3
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims0_score_mod6
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims1_score_mod3
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims1_score_mod6
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims2_score_mod3
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims2_score_mod6
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims3_score_mod3
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_kv_batch_broadcast_float16_head_dims2_batch_dims3_score_mod6
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_logsumexp_correctness_float16_score_mod0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_logsumexp_correctness_float16_score_mod1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_multiple_score_mod_calls
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_multiple_score_mod_calls_paged_attention
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_multiple_score_mod_calls_paged_attention2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_equal_head_dims_score_mod3_float16_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_equal_head_dims_score_mod3_float16_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_equal_head_dims_score_mod6_float16_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_non_equal_head_dims_score_mod6_float16_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_padded_dense_causal_float16
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod3_head_dims1_page_size_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod3_head_dims2_page_size_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod3_head_dims2_page_size_256
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod3_head_dims2_page_size_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod6_head_dims1_page_size_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod6_head_dims2_page_size_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod6_head_dims2_page_size_256
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod6_head_dims2_page_size_64
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s0_v_s0_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s0_v_s0_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s0_v_s0_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s0_v_s1_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s0_v_s1_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s0_v_s1_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s0_v_s2_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s0_v_s2_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s0_v_s2_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s0_v_s3_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s0_v_s3_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s0_v_s3_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s1_v_s0_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s1_v_s0_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s1_v_s0_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s1_v_s1_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s1_v_s1_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s1_v_s1_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s1_v_s2_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s1_v_s2_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s1_v_s2_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s1_v_s3_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s1_v_s3_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s1_v_s3_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s2_v_s0_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s2_v_s0_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s2_v_s0_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s2_v_s1_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s2_v_s1_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s2_v_s1_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s2_v_s2_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s2_v_s2_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s2_v_s3_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s2_v_s3_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s2_v_s3_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s3_v_s0_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s3_v_s0_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s3_v_s0_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s3_v_s1_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s3_v_s1_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s3_v_s1_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s3_v_s2_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s3_v_s2_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s3_v_s2_head_dims2
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s3_v_s3_head_dims0
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s3_v_s3_head_dims1
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_strided_inputs_float16_k_s3_v_s3_head_dims2

AssertionError: False is not true : Output/Grad with NaN

  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_njt_causal_float16
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod0_head_dims1_page_size_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod0_head_dims1_page_size_256
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod1_head_dims1_page_size_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod1_head_dims1_page_size_256
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod2_head_dims1_page_size_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod2_head_dims1_page_size_256
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod3_head_dims1_page_size_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod3_head_dims1_page_size_256
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod4_head_dims1_page_size_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod4_head_dims1_page_size_256
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod5_head_dims1_page_size_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod5_head_dims1_page_size_256
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod6_head_dims1_page_size_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod6_head_dims1_page_size_256
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod7_head_dims1_page_size_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod7_head_dims1_page_size_256
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod8_head_dims1_page_size_128
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_paged_attention_page_size_float16_score_mod8_head_dims1_page_size_256
  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_windowed_no_mask_vs_sdpa

AssertionError: tensor(False, device='xpu:0') is not true

  • python test/inductor/test_flex_decoding.py TestFlexDecoding.test_multiple_score_mod_calls2

Environment details

You can setup test env according this issue:
#3518

Collecting environment information...
PyTorch version: N/A
Is debug build: N/A
CUDA used to build PyTorch: N/A
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.4 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version: Could not collect
CMake version: version 3.31.4
Libc version: glibc-2.35

Python version: 3.10.16 | packaged by conda-forge | (main, Dec  5 2024, 14:16:10) [GCC 13.3.0] (64-bit runtime)
Python platform: Linux-5.15.50-051550-generic-x86_64-with-glibc2.35
Is CUDA available: N/A
CUDA runtime version: Could not collect
CUDA_MODULE_LOADING set to: N/A
GPU models and configuration: Could not collect
Nvidia driver version: Could not collect
cuDNN version: Could not collect
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: N/A

CPU:
Architecture:                    x86_64
CPU op-mode(s):                  32-bit, 64-bit
Address sizes:                   52 bits physical, 57 bits virtual
Byte Order:                      Little Endian
CPU(s):                          224
On-line CPU(s) list:             0-223
Vendor ID:                       GenuineIntel
Model name:                      Intel(R) Xeon(R) Platinum 8480+
CPU family:                      6
Model:                           143
Thread(s) per core:              2
Core(s) per socket:              56
Socket(s):                       2
Stepping:                        6
CPU max MHz:                     3800.0000
CPU min MHz:                     800.0000
BogoMIPS:                        4000.00
Flags:                           fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cat_l2 cdp_l3 invpcid_single intel_ppin cdp_l2 ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb intel_pt avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local split_lock_detect avx_vnni avx512_bf16 wbnoinvd dtherm ida arat pln pts hwp hwp_act_window hwp_epp hwp_pkg_req avx512vbmi umip pku ospke waitpkg avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg tme avx512_vpopcntdq la57 rdpid bus_lock_detect cldemote movdiri movdir64b enqcmd fsrm md_clear serialize tsxldtrk pconfig arch_lbr avx512_fp16 flush_l1d arch_capabilities
Virtualization:                  VT-x
L1d cache:                       5.3 MiB (112 instances)
L1i cache:                       3.5 MiB (112 instances)
L2 cache:                        224 MiB (112 instances)
L3 cache:                        210 MiB (2 instances)
NUMA node(s):                    2
NUMA node0 CPU(s):               0-55,112-167
NUMA node1 CPU(s):               56-111,168-223
Vulnerability Itlb multihit:     Not affected
Vulnerability L1tf:              Not affected
Vulnerability Mds:               Not affected
Vulnerability Meltdown:          Not affected
Vulnerability Mmio stale data:   Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1:        Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:        Mitigation; Enhanced IBRS, IBPB conditional, RSB filling
Vulnerability Srbds:             Not affected
Vulnerability Tsx async abort:   Not affected

Versions of relevant libraries:
[pip3] numpy==2.2.3
[pip3] optree==0.14.0
[pip3] torch==2.7.0a0+gite0be178
[pip3] torchaudio==2.6.0.dev20250213+xpu
[pip3] torchinfo==1.8.0
[pip3] torchvision==0.22.0.dev20250213+xpu
[pip3] triton==3.2.0+gitb3f9c6e5
[conda] numpy                     2.2.3                    pypi_0    pypi
[conda] optree                    0.14.0                   pypi_0    pypi
[conda] torch                     2.7.0a0+gite0be178           dev_0    <develop>
[conda] torchaudio                2.6.0.dev20250213+xpu          pypi_0    pypi
[conda] torchinfo                 1.8.0                    pypi_0    pypi
[conda] torchvision               0.22.0.dev20250213+xpu          pypi_0    pypi
[conda] triton                    3.2.0+gitb3f9c6e5           dev_0    <develop>
@chengjunlu
Copy link
Contributor

We find two issues related to Triton XPU: #3672, #3704
And two issues related to IGC.
With the quick work around for theses issues, we can pass all the test cases of FlexDecoding.

@chengjunlu
Copy link
Contributor

chengjunlu commented Apr 2, 2025

All the FlexDecoding UT cases passed with the latest Triton XPU and IGC driver with fix.

This is the total number of the cases tested.

=== 396 passed, 2 xfailed in 263.64s (0:04:23) ===

Please wait the next agama release in April.

@ESI-SYD
Copy link
Contributor

ESI-SYD commented Apr 24, 2025

Accuracy regression:

pytorch/13339ce0869 with the patch from PR pytorch/pytorch#143553

Triton: 0cf724f

==== 430 failed, 142 passed, 2 xfailed in 1481.01s (0:24:41) ===
Cases
================================================================================================= short test summary info =================================================================================================
FAILED [17.9622s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod1_BLOCK_SIZE3_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [17.9906s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod1_BLOCK_SIZE_64_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 1.3772139206003757 is greater than ref error 0.001172120550190836 by more than 10.0X.
FAILED [18.0351s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod6_BLOCK_SIZE2_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.001368046840070739 by more than 10.0X.
FAILED [18.1033s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod5_BLOCK_SIZE_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 192.2712266010517 is greater than ref error 0.00012206174377040455 by more than 10.0X.
FAILED [24.7640s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod5_BLOCK_SIZE2_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [26.6851s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_bfloat16_score_mod5_head_dims2_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.5804479765670332 is greater than ref error 0.000940347721551697 by more than 10.0X.
FAILED [12.5934s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod1_BLOCK_SIZE_128_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.8581s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod2_BLOCK_SIZE2_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 4.261599421707896 is greater than ref error 0.02999668672602221 by more than 10.0X.
FAILED [12.8366s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod6_BLOCK_SIZE3_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.001368046840070739 by more than 10.0X.
FAILED [12.7740s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod5_BLOCK_SIZE_64_xpu_float16 - AssertionError: False is not true : Output/Grad with NaN
FAILED [12.5185s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod1_BLOCK_SIZE_64_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.6567s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod6_BLOCK_SIZE2_xpu_float16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.0001557088130255083 by more than 10.0X.
FAILED [12.7173s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod6_BLOCK_SIZE_128_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.001368046840070739 by more than 10.0X.
FAILED [12.7464s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod2_BLOCK_SIZE3_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 4.659975255867778 is greater than ref error 0.02999668672602221 by more than 10.0X.
FAILED [19.4404s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod1_BLOCK_SIZE2_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [19.5999s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod5_BLOCK_SIZE3_xpu_float32 - AssertionError: False is not true : Output/Grad with NaN
FAILED [12.6237s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod2_BLOCK_SIZE2_xpu_float16 - AssertionError: False is not true : Out Compiled error 4.780477720187941 is greater than ref error 0.003364109592890685 by more than 10.0X.
FAILED [12.4806s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod6_BLOCK_SIZE_64_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.001368046840070739 by more than 10.0X.
FAILED [12.5484s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod6_BLOCK_SIZE3_xpu_float16 - AssertionError: False is not true : Output/Grad with NaN
FAILED [12.8642s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod2_BLOCK_SIZE_128_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 3.3838966223010574 is greater than ref error 0.02999668672602221 by more than 10.0X.
FAILED [19.1620s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod5_BLOCK_SIZE_128_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [19.5688s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod1_BLOCK_SIZE3_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [26.0847s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_bfloat16_score_mod0_head_dims1_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.024974583998020494 is greater than ref error 0.0001258655349551468 by more than 1.1X.
FAILED [12.8099s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod2_BLOCK_SIZE3_xpu_float16 - AssertionError: False is not true : Out Compiled error 4.936314228712467 is greater than ref error 0.003364109592890685 by more than 10.0X.
FAILED [12.9156s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod6_BLOCK_SIZE_128_xpu_float16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.0001557088130255083 by more than 10.0X.
FAILED [12.9570s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod7_BLOCK_SIZE2_xpu_bfloat16 - AssertionError: Tensor-likes are not close!
FAILED [13.0761s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod2_BLOCK_SIZE_64_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 4.37026071338645 is greater than ref error 0.02999668672602221 by more than 10.0X.
FAILED [12.5015s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod2_BLOCK_SIZE_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 4.939242909215347 is greater than ref error 0.003364109592890685 by more than 10.0X.
FAILED [12.5596s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod6_BLOCK_SIZE_64_xpu_float16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.0001557088130255083 by more than 10.0X.
FAILED [12.5827s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod7_BLOCK_SIZE3_xpu_bfloat16 - AssertionError: Tensor-likes are not close!
FAILED [12.8455s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod3_BLOCK_SIZE2_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.001368046840070739 by more than 10.0X.
FAILED [19.1847s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod5_BLOCK_SIZE_64_xpu_float32 - AssertionError: False is not true : Output/Grad with NaN
FAILED [19.8597s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod1_BLOCK_SIZE_128_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [25.5578s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_bfloat16_score_mod6_head_dims1_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.6851061763591133 is greater than ref error 0.0010671715135685913 by more than 10.0X.
FAILED [22.0791s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_bfloat16_score_mod0_head_dims2_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.025629299566817035 is greater than ref error 0.00013171543179015056 by more than 1.1X.
FAILED [13.1505s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod2_BLOCK_SIZE_64_xpu_float16 - AssertionError: False is not true : Out Compiled error 5.23385377424216 is greater than ref error 0.003364109592890685 by more than 10.0X.
FAILED [13.2770s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod7_BLOCK_SIZE_128_xpu_bfloat16 - AssertionError: Tensor-likes are not close!
FAILED [13.4936s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod7_BLOCK_SIZE2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [13.3625s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod3_BLOCK_SIZE3_xpu_bfloat16 - AssertionError: False is not true : Output/Grad with NaN
FAILED [19.4898s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod6_BLOCK_SIZE2_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [19.8046s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod1_BLOCK_SIZE_64_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [12.6721s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod3_BLOCK_SIZE2_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.7274676725150755 is greater than ref error 0.0001557088130255083 by more than 10.0X.
FAILED [12.6993s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod7_BLOCK_SIZE_64_xpu_bfloat16 - AssertionError: Tensor-likes are not close!
FAILED [12.8769s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod7_BLOCK_SIZE3_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.7803s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod3_BLOCK_SIZE_128_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.001368046840070739 by more than 10.0X.
FAILED [21.9133s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_bfloat16_score_mod6_head_dims2_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.6734452085147495 is greater than ref error 0.0011583906671148718 by more than 10.0X.
FAILED [12.7767s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod3_BLOCK_SIZE3_xpu_float16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.0001557088130255083 by more than 10.0X.
FAILED [12.6086s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod8_BLOCK_SIZE2_xpu_bfloat16 - AssertionError: Tensor-likes are not close!
FAILED [12.8380s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod7_BLOCK_SIZE_128_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.8819s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod3_BLOCK_SIZE_64_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.001368046840070739 by more than 10.0X.
FAILED [19.3669s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod6_BLOCK_SIZE3_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [19.5213s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod2_BLOCK_SIZE2_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [12.6728s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod3_BLOCK_SIZE_128_xpu_float16 - AssertionError: False is not true : Output/Grad with NaN
FAILED [12.5794s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod8_BLOCK_SIZE3_xpu_bfloat16 - AssertionError: Tensor-likes are not close!
FAILED [12.7565s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod7_BLOCK_SIZE_64_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.9557s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod4_BLOCK_SIZE2_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.02318579619634426 is greater than ref error 0.0001265045794619076 by more than 1.1X.
FAILED [20.0828s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod6_BLOCK_SIZE_128_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [20.1456s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod2_BLOCK_SIZE3_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [12.5450s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod3_BLOCK_SIZE_64_xpu_float16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.0001557088130255083 by more than 10.0X.
FAILED [12.8047s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod8_BLOCK_SIZE_128_xpu_bfloat16 - AssertionError: Tensor-likes are not close!
FAILED [12.4767s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod8_BLOCK_SIZE2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.9188s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod4_BLOCK_SIZE3_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.024195902113405082 is greater than ref error 0.0001265045794619076 by more than 1.1X.
FAILED [26.1814s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_bfloat16_score_mod1_head_dims1_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 1.673792401821704 is greater than ref error 0.0011210634562040522 by more than 10.0X.
FAILED [12.6618s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod4_BLOCK_SIZE2_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.02312123672732928 is greater than ref error 1.6445839539482343e-05 by more than 1.1X.
FAILED [12.5541s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod8_BLOCK_SIZE_64_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.02944172791085891 is greater than ref error 0.00017743198545133333 by more than 1.1X.
FAILED [12.6339s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod8_BLOCK_SIZE3_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.8324s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod4_BLOCK_SIZE_128_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.022635155570894988 is greater than ref error 0.0001265045794619076 by more than 1.1X.
FAILED [19.3694s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod6_BLOCK_SIZE_64_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [19.7703s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod2_BLOCK_SIZE_128_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [25.8297s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_bfloat16_score_mod7_head_dims1_xpu_bfloat16 - AssertionError: Tensor-likes are not close!
FAILED [12.6935s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod4_BLOCK_SIZE3_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.023464093843967155 is greater than ref error 1.6445839539482343e-05 by more than 1.1X.
FAILED [12.5196s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod0_BLOCK_SIZE2_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.024887906867072734 is greater than ref error 1.6445839539482343e-05 by more than 1.1X.
FAILED [21.8911s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_bfloat16_score_mod1_head_dims2_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 1.7897804189395348 is greater than ref error 0.0014491960071635335 by more than 10.0X.
FAILED [12.5208s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod8_BLOCK_SIZE_128_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.8697s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod4_BLOCK_SIZE_64_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.0231584407888173 is greater than ref error 0.0001265045794619076 by more than 1.1X.
FAILED [19.7337s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod7_BLOCK_SIZE2_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [19.6023s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod2_BLOCK_SIZE_64_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [12.6526s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod4_BLOCK_SIZE_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.021677843727590084 is greater than ref error 1.6445839539482343e-05 by more than 1.1X.
FAILED [12.8135s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod0_BLOCK_SIZE3_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.0237040997407569 is greater than ref error 1.6445839539482343e-05 by more than 1.1X.
FAILED [12.8201s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod8_BLOCK_SIZE_64_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.8850s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod5_BLOCK_SIZE2_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 32.49805874793885 is greater than ref error 0.0010757839871797922 by more than 10.0X.
FAILED [21.8156s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_bfloat16_score_mod7_head_dims2_xpu_bfloat16 - AssertionError: Tensor-likes are not close!
FAILED [12.8536s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod4_BLOCK_SIZE_64_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.021034527313338693 is greater than ref error 1.6445839539482343e-05 by more than 1.1X.
FAILED [12.4149s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod0_BLOCK_SIZE_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.02095820293123383 is greater than ref error 1.6445839539482343e-05 by more than 1.1X.
FAILED [13.0575s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod5_BLOCK_SIZE3_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 16.42792061744348 is greater than ref error 0.0010757839871797922 by more than 10.0X.
FAILED [19.5636s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod7_BLOCK_SIZE3_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [19.4512s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod3_BLOCK_SIZE2_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [12.6967s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod5_BLOCK_SIZE2_xpu_float16 - AssertionError: False is not true : Out Compiled error 144.32912371752917 is greater than ref error 0.00012206174377040455 by more than 10.0X.
FAILED [12.6647s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod0_BLOCK_SIZE_64_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.9098s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod5_BLOCK_SIZE_128_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 176.33689619741787 is greater than ref error 0.0010757839871797922 by more than 10.0X.
FAILED [19.9843s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod7_BLOCK_SIZE_128_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [12.4476s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod5_BLOCK_SIZE3_xpu_float16 - AssertionError: False is not true : Out Compiled error 160.20624710264593 is greater than ref error 0.00012206174377040455 by more than 10.0X.
FAILED [12.6605s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float16_score_mod1_BLOCK_SIZE2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [19.9471s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod3_BLOCK_SIZE3_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [12.8082s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod5_BLOCK_SIZE_64_xpu_bfloat16 - AssertionError: False is not true : Output/Grad with NaN
FAILED [26.6842s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_bfloat16_score_mod2_head_dims1_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 5.256413500053755 is greater than ref error 0.026659345875464757 by more than 10.0X.
FAILED [19.4903s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod7_BLOCK_SIZE_64_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [19.9197s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod3_BLOCK_SIZE_128_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [26.0558s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float16_score_mod0_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.0257460104956291 is greater than ref error 1.7247964328523697e-05 by more than 1.1X.
FAILED [25.6867s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_bfloat16_score_mod8_head_dims1_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.03399005910604222 is greater than ref error 0.0001696530130961467 by more than 1.1X.
FAILED [21.7414s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_bfloat16_score_mod2_head_dims2_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 5.959584893122922 is greater than ref error 0.030714254250005624 by more than 10.0X.
FAILED [19.3152s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod8_BLOCK_SIZE2_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [19.8778s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod3_BLOCK_SIZE_64_xpu_float32 - AssertionError: False is not true : Output/Grad with NaN
FAILED [21.9085s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_function_composition_bfloat16_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 80.28912029107087 is greater than ref error 0.0010757839871797922 by more than 10.0X.
FAILED [21.5669s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float16_score_mod0_head_dims2_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.024001595463885757 is greater than ref error 1.5890373837252596e-05 by more than 1.1X.
FAILED [21.5615s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_bfloat16_score_mod8_head_dims2_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.034704112718518 is greater than ref error 0.00018147907077219753 by more than 1.1X.
FAILED [19.7158s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod8_BLOCK_SIZE3_xpu_float32 - AssertionError: False is not true : Output/Grad with NaN
FAILED [21.7808s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_function_composition_float16_xpu_float16 - AssertionError: False is not true : Out Compiled error 192.1906166074955 is greater than ref error 0.00012206174377040455 by more than 10.0X.
FAILED [12.4455s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod0_BLOCK_SIZE2_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.022920401532767025 is greater than ref error 0.0001265045794619076 by more than 1.1X.
FAILED [19.2662s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod8_BLOCK_SIZE_128_xpu_float32 - AssertionError: False is not true : Output/Grad with NaN
FAILED [12.6453s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod0_BLOCK_SIZE3_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.02013400521825362 is greater than ref error 0.0001265045794619076 by more than 1.1X.
FAILED [23.6467s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_function_composition_float32_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [12.5331s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod0_BLOCK_SIZE_128_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.0219257595992431 is greater than ref error 0.0001265045794619076 by more than 1.1X.
FAILED [25.8879s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_bfloat16_score_mod3_head_dims1_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.6343299112631939 is greater than ref error 0.0010671715135685913 by more than 10.0X.
FAILED [43.2249s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float32_score_mod8_head_dims1_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [19.6406s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_float32_score_mod8_BLOCK_SIZE_64_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [12.7514s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod0_BLOCK_SIZE_64_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.023306068858280925 is greater than ref error 0.0001265045794619076 by more than 1.1X.
FAILED [26.2719s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float16_score_mod1_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.0395s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_bfloat16_score_mod3_head_dims2_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.7190885848176549 is greater than ref error 0.0011583906671148718 by more than 10.0X.
FAILED [12.5315s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod1_BLOCK_SIZE2_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 1.4751425748652758 is greater than ref error 0.001172120550190836 by more than 10.0X.
FAILED [13.0567s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod1_BLOCK_SIZE3_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 1.5297057608906213 is greater than ref error 0.001172120550190836 by more than 10.0X.
FAILED [21.5254s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float16_score_mod1_head_dims2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [25.3473s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_head_dependent_mask_mod_float16_score_mod0_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.025121028490459676 is greater than ref error 1.7555860772478018e-05 by more than 1.1X.
FAILED [12.7467s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_different_block_size_bfloat16_score_mod1_BLOCK_SIZE_128_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 1.4334253063157454 is greater than ref error 0.001172120550190836 by more than 10.0X.
FAILED [19.9853s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_head_dependent_mask_mod_float16_score_mod0_head_dims2_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.024808963156055564 is greater than ref error 1.5102663991191184e-05 by more than 1.1X.
FAILED [22.4077s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_captured_buffers_all_dims_bfloat16_xpu_bfloat16 - AssertionError: Tensor-likes are not close!
FAILED [26.1745s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_bfloat16_score_mod4_head_dims1_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.025557158162491513 is greater than ref error 0.0001258655349551468 by more than 1.1X.
FAILED [26.6985s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_head_dependent_mask_mod_float16_score_mod6_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.6192s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_captured_buffers_all_dims_float16_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.024645010443882655 is greater than ref error 2.367869422316025e-05 by more than 1.1X.
FAILED [25.5430s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float16_score_mod2_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error 5.492086589746054 is greater than ref error 0.0037202815924252675 by more than 10.0X.
FAILED [21.9644s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_bfloat16_score_mod4_head_dims2_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.025062369559358005 is greater than ref error 0.00013171543179015056 by more than 1.1X.
FAILED [27.1005s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float32_score_mod1_head_dims1_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [20.4176s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_head_dependent_mask_mod_float16_score_mod6_head_dims2_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.6289872100059668 is greater than ref error 0.00012413905046315384 by more than 10.0X.
FAILED [25.3608s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_captured_buffers_all_dims_float32_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [21.7112s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float16_score_mod2_head_dims2_xpu_float16 - AssertionError: False is not true : Out Compiled error 6.739088681148166 is greater than ref error 0.0032400867448312 by more than 10.0X.
FAILED [25.7077s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_head_dependent_mask_mod_float16_score_mod1_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.098983053679648 is greater than ref error 0.000172681043506695 by more than 1.1X.
FAILED [27.4285s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float32_score_mod1_head_dims2_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [21.8978s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_captured_buffers_bfloat16_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.021520371246672804 is greater than ref error 0.00011360576823631229 by more than 1.1X.
FAILED [19.7837s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_head_dependent_mask_mod_float16_score_mod1_head_dims2_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.08210842750006775 is greater than ref error 0.00010185541020514117 by more than 1.1X.
FAILED [21.9875s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_captured_buffers_float16_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.02359548336395011 is greater than ref error 1.4289653339368973e-05 by more than 1.1X.
FAILED [26.1206s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_head_dependent_mask_mod_float16_score_mod7_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.17558069150956618 is greater than ref error 8.595348035070197e-05 by more than 1.1X.
FAILED [16.7985s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims0_score_mod0_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.02390652157218249 is greater than ref error 1.6609302868459983e-05 by more than 1.1X.
FAILED [25.9744s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_bfloat16_score_mod5_head_dims1_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.4185012161383613 is greater than ref error 0.0010402551972226812 by more than 10.0X.
FAILED [25.3047s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float16_score_mod3_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.5530824678694444 is greater than ref error 0.00011629662867563564 by more than 10.0X.
FAILED [16.2479s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims0_score_mod1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [16.9137s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims1_score_mod3_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.7776300817458832 is greater than ref error 0.00012424099404598739 by more than 10.0X.
FAILED [20.2597s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_head_dependent_mask_mod_float16_score_mod7_head_dims2_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.17101892750682363 is greater than ref error 4.6945631518156974e-05 by more than 1.1X.
FAILED [25.0558s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_head_dependent_mask_mod_float16_score_mod2_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.46134618534228927 is greater than ref error 0.0006624688682835495 by more than 1.1X.
FAILED [16.3262s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims0_score_mod2_xpu_float16 - AssertionError: False is not true : Out Compiled error 4.818831938703672 is greater than ref error 0.0031813524860268374 by more than 10.0X.
FAILED [16.9546s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims1_score_mod4_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.0243348382709467 is greater than ref error 1.67722984166606e-05 by more than 1.1X.
FAILED [21.3494s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float16_score_mod3_head_dims2_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.7440478585455992 is greater than ref error 0.00017160034579432443 by more than 10.0X.
FAILED [19.8425s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_head_dependent_mask_mod_float16_score_mod2_head_dims2_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.43407400537078566 is greater than ref error 0.0007391163106520281 by more than 1.1X.
FAILED [27.1471s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float32_score_mod2_head_dims1_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [16.6517s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims0_score_mod3_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.6609154886840163 is greater than ref error 0.00014103705082866444 by more than 10.0X.
FAILED [17.0429s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims1_score_mod5_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.5991013986537083 is greater than ref error 0.00011102855127838042 by more than 10.0X.
FAILED [21.6804s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_captured_scale_float16_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.026325557060507716 is greater than ref error 1.6445839539482343e-05 by more than 1.1X.
FAILED [16.4624s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims0_score_mod4_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.023710104295798524 is greater than ref error 1.6609302868459983e-05 by more than 1.1X.
FAILED [16.7280s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims1_score_mod6_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.627642956267664 is greater than ref error 0.00012424099404598739 by more than 10.0X.
FAILED [25.6525s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_head_dependent_mask_mod_float16_score_mod8_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.031033920828232094 is greater than ref error 2.6001188679320198e-05 by more than 1.1X.
FAILED [27.4309s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float32_score_mod2_head_dims2_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [18.3109s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_decode_at_different_input_position_float16_score_mod0_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [16.8190s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims0_score_mod5_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.6424811446508363 is greater than ref error 0.00012054365745074735 by more than 10.0X.
FAILED [17.1027s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims1_score_mod7_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [20.3439s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_head_dependent_mask_mod_float16_score_mod8_head_dims2_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.03206493094891834 is greater than ref error 2.0875937204112364e-05 by more than 1.1X.
FAILED [25.8829s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float16_score_mod4_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.025588897878382484 is greater than ref error 1.7247964328523697e-05 by more than 1.1X.
FAILED [25.7457s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_head_dependent_mask_mod_float16_score_mod3_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [17.7697s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_decode_at_different_input_position_float16_score_mod1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [16.8338s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims0_score_mod6_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.6626185148355337 is greater than ref error 0.00014103705082866444 by more than 10.0X.
FAILED [16.8842s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims1_score_mod8_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.035428704401387325 is greater than ref error 2.4044672512836237e-05 by more than 1.1X.
FAILED [19.9882s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_head_dependent_mask_mod_float16_score_mod3_head_dims2_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.6795835568390842 is greater than ref error 0.00012413905046315384 by more than 10.0X.
FAILED [16.8079s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims0_score_mod7_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [17.7042s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_decode_at_different_input_position_float16_score_mod2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.1625s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float16_score_mod4_head_dims2_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.025822489362625524 is greater than ref error 1.5890373837252596e-05 by more than 1.1X.
FAILED [16.2743s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims2_score_mod0_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.024463706708427048 is greater than ref error 1.6787429579252684e-05 by more than 1.1X.
FAILED [16.8373s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims0_score_mod8_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.03424245497180618 is greater than ref error 2.3193092961609886e-05 by more than 1.1X.
FAILED [18.1038s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_decode_at_different_input_position_float16_score_mod3_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [16.3760s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims2_score_mod1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [27.1431s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float32_score_mod3_head_dims1_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [16.5209s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims1_score_mod0_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.023199830590769233 is greater than ref error 1.67722984166606e-05 by more than 1.1X.
FAILED [16.2866s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims2_score_mod2_xpu_float16 - AssertionError: False is not true : Out Compiled error 6.023112115953538 is greater than ref error 0.0031925557251229197 by more than 10.0X.
FAILED [18.1276s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_decode_at_different_input_position_float16_score_mod4_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [16.0754s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims1_score_mod1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [25.2011s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_head_dependent_mask_mod_float16_score_mod4_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.02497018198498044 is greater than ref error 1.7555860772478018e-05 by more than 1.1X.
FAILED [16.5763s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims2_score_mod3_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.7967673013743024 is greater than ref error 0.0001394019202507444 by more than 10.0X.
FAILED [17.6166s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_decode_at_different_input_position_float16_score_mod5_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [27.0057s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float32_score_mod3_head_dims2_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [25.6481s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float16_score_mod5_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.45481485165528585 is greater than ref error 0.00013737981183021674 by more than 10.0X.
FAILED [16.6763s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims1_score_mod2_xpu_float16 - AssertionError: False is not true : Out Compiled error 5.029972097025138 is greater than ref error 0.0029070296637228044 by more than 10.0X.
FAILED [16.3078s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims2_score_mod4_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.024752318740717767 is greater than ref error 1.6787429579252684e-05 by more than 1.1X.
FAILED [20.4470s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_head_dependent_mask_mod_float16_score_mod4_head_dims2_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.025309705784576666 is greater than ref error 1.5102663991191184e-05 by more than 1.1X.
FAILED [18.0396s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_decode_at_different_input_position_float16_score_mod6_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.6322s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims3_score_mod1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [21.4068s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float16_score_mod5_head_dims2_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.5025366555932953 is greater than ref error 0.00012944134382701804 by more than 10.0X.
FAILED [12.7381s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims0_score_mod7_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [16.2456s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims2_score_mod5_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.5902718276338978 is greater than ref error 0.0001223289516978452 by more than 10.0X.
FAILED [17.6641s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_decode_at_different_input_position_float16_score_mod7_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.3774s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims3_score_mod2_xpu_float16 - AssertionError: False is not true : Out Compiled error 5.188318813432179 is greater than ref error 0.002872170318853992 by more than 10.0X.
FAILED [12.7807s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims0_score_mod8_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.03504584984586534 is greater than ref error 2.2680977162406786e-05 by more than 1.1X.
FAILED [16.4678s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims2_score_mod6_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.7831641120981203 is greater than ref error 0.0001394019202507444 by more than 10.0X.
FAILED [12.8026s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims3_score_mod3_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.7729968269185292 is greater than ref error 0.000138954427370671 by more than 10.0X.
FAILED [17.7879s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_decode_at_different_input_position_float16_score_mod8_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.6567s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims1_score_mod0_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.026116440232452257 is greater than ref error 1.585065978764196e-05 by more than 1.1X.
FAILED [12.4283s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims3_score_mod4_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [16.5478s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims2_score_mod7_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.8855s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims1_score_mod1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [25.5052s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_head_dependent_mask_mod_float16_score_mod5_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.24949484663250196 is greater than ref error 0.00014237915235942812 by more than 1.1X.
FAILED [12.5188s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims3_score_mod5_xpu_float16 - AssertionError: False is not true : Out Compiled error 144.45843399721832 is greater than ref error 0.00012144301599904188 by more than 10.0X.
FAILED [12.6462s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims1_score_mod2_xpu_float16 - AssertionError: False is not true : Out Compiled error 6.607383510218547 is greater than ref error 0.003530131125706193 by more than 10.0X.
FAILED [16.4567s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims2_score_mod8_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.035383635449389925 is greater than ref error 2.3569248036532012e-05 by more than 1.1X.
FAILED [12.6253s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims3_score_mod6_xpu_float16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.000138954427370671 by more than 10.0X.
FAILED [20.0791s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_head_dependent_mask_mod_float16_score_mod5_head_dims2_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.09163521762118332 is greater than ref error 0.00013616321417785657 by more than 1.1X.
FAILED [26.1258s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float16_score_mod6_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.6851326655737466 is greater than ref error 0.00011629662867563564 by more than 10.0X.
FAILED [12.7833s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims1_score_mod3_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.7845603662506676 is greater than ref error 0.0001618725054925074 by more than 10.0X.
FAILED [12.9083s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims3_score_mod0_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.024093355900472953 is greater than ref error 1.5769351158345875e-05 by more than 1.1X.
FAILED [13.2526s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims3_score_mod7_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.6276s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims2_score_mod4_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.7547s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims1_score_mod4_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [21.6846s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float16_score_mod6_head_dims2_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.6670430367690585 is greater than ref error 0.00017160034579432443 by more than 10.0X.
FAILED [12.0570s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims2_score_mod5_xpu_float16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.00012372573162020928 by more than 10.0X.
FAILED [12.9955s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims1_batch_dims3_score_mod8_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [24.7312s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_multiple_score_mod_calls2_xpu - AssertionError: tensor(False, device='xpu:0') is not true
FAILED [12.6426s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims1_score_mod5_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.5969904450482881 is greater than ref error 0.0001238447305206121 by more than 10.0X.
FAILED [21.3810s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod0_bfloat16_head_dims0_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.02349398860523384 is greater than ref error 0.00012812386276237143 by more than 1.1X.
FAILED [12.7663s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims2_score_mod6_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.7334980401133784 is greater than ref error 0.0001459718312766157 by more than 10.0X.
FAILED [12.6897s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims0_score_mod0_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.02771624074399766 is greater than ref error 1.63666756416536e-05 by more than 1.1X.
FAILED [0.0000s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_larger_block_mask_bug_float16_xpu_float16
FAILED [12.6410s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims1_score_mod6_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.7508236288529645 is greater than ref error 0.0001618725054925074 by more than 10.0X.
FAILED [12.8194s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims0_score_mod1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [13.4410s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims2_score_mod7_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [21.9070s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod0_bfloat16_head_dims1_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.02196628602555796 is greater than ref error 0.0001425539293124946 by more than 1.1X.
FAILED [12.8641s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims1_score_mod7_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.9781s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims0_score_mod2_xpu_float16 - AssertionError: False is not true : Out Compiled error 6.811600594769826 is greater than ref error 0.003486427031954009 by more than 10.0X.
FAILED [12.8023s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims2_score_mod8_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [35.9124s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_multiple_score_mod_calls_paged_attention2_xpu - AssertionError: Tensor-likes are not close!
FAILED [12.7747s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims1_score_mod8_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.036155639184592116 is greater than ref error 2.2318057059312528e-05 by more than 1.1X.
FAILED [12.7030s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims0_score_mod3_xpu_float16 - AssertionError: False is not true : Out Compiled error 1.0573896821933033 is greater than ref error 0.00015953196383595788 by more than 10.0X.
FAILED [12.6717s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims3_score_mod0_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.02584956773109671 is greater than ref error 1.5297417934263673e-05 by more than 1.1X.
FAILED [22.1197s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod0_float16_head_dims0_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.022595704326584892 is greater than ref error 1.6351976513031915e-05 by more than 1.1X.
FAILED [27.3367s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod3_bfloat16_head_dims1_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.0010255322811853424 by more than 10.0X.
FAILED [12.6247s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims2_score_mod0_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.025911989461761084 is greater than ref error 1.5467311717054773e-05 by more than 1.1X.
FAILED [26.3264s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float16_score_mod7_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.6397s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims3_score_mod1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.8642s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims0_score_mod4_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.027403354068551808 is greater than ref error 1.63666756416536e-05 by more than 1.1X.
FAILED [12.6716s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims2_score_mod1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.0392s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod0_float16_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.02474408703791608 is greater than ref error 1.7563100968313013e-05 by more than 1.1X.
FAILED [21.9053s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod3_float16_head_dims0_xpu_float16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.0001557088130255083 by more than 10.0X.
FAILED [35.1696s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_multiple_score_mod_calls_paged_attention_xpu - AssertionError: Tensor-likes are not close!
FAILED [12.9340s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims3_score_mod2_xpu_float16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.0029230504787345693 by more than 10.0X.
FAILED [12.7322s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims0_score_mod5_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.737158653483907 is greater than ref error 0.00011927616603150369 by more than 10.0X.
FAILED [13.2012s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims2_score_mod2_xpu_float16 - AssertionError: False is not true : Out Compiled error 6.323486819928371 is greater than ref error 0.0030942026734117195 by more than 10.0X.
FAILED [22.0976s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float16_score_mod7_head_dims2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.9695s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims3_score_mod3_xpu_float16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.0001311401376602405 by more than 10.0X.
FAILED [12.9833s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims0_score_mod6_xpu_float16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.00015953196383595788 by more than 10.0X.
FAILED [28.1145s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float32_score_mod5_head_dims1_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [12.6838s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims2_score_mod3_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.7067373045275112 is greater than ref error 0.0001459718312766157 by more than 10.0X.
FAILED [21.9894s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod3_float16_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.00013497051758062462 by more than 10.0X.
FAILED [24.4588s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_multiple_score_mod_calls_xpu - AssertionError: Tensor-likes are not close!
FAILED [13.0731s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims3_score_mod4_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [23.8202s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod1_float32_head_dims1_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [12.3881s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims3_score_mod5_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.5127457185191859 is greater than ref error 0.00010665879545798779 by more than 10.0X.
FAILED [21.6066s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod5_bfloat16_head_dims1_xpu_bfloat16 - AssertionError: False is not true : Output/Grad with NaN
FAILED [27.6703s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float32_score_mod5_head_dims2_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [23.7659s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod3_float32_head_dims0_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [22.3624s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_njt_causal_bfloat16_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.4904571935059152 is greater than ref error 0.0010116035068031565 by more than 10.0X.
FAILED [13.3811s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims3_score_mod6_xpu_float16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.0001311401376602405 by more than 10.0X.
FAILED [22.1350s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod2_bfloat16_head_dims0_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 4.254082485780825 is greater than ref error 0.02999668672602221 by more than 10.0X.
FAILED [21.7713s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod5_float16_head_dims0_xpu_float16 - AssertionError: False is not true : Out Compiled error 80.39537559594007 is greater than ref error 0.00012206174377040455 by more than 10.0X.
FAILED [12.7170s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims3_score_mod7_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [26.8496s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float16_score_mod8_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.03446156663991619 is greater than ref error 2.3163466462807564e-05 by more than 1.1X.
FAILED [23.4803s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod3_float32_head_dims1_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [21.7640s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_njt_causal_float16_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [12.7497s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_kv_batch_broadcast_float16_head_dims2_batch_dims3_score_mod8_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [21.9006s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod2_bfloat16_head_dims1_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 5.001472991811788 is greater than ref error 0.03275536218314387 by more than 10.0X.
FAILED [22.0213s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod5_float16_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error 96.47421038631978 is greater than ref error 0.00011563770138277958 by more than 10.0X.
FAILED [22.0675s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float16_score_mod8_head_dims2_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.03635488117537754 is greater than ref error 2.3216174978922013e-05 by more than 1.1X.
FAILED [22.2996s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod4_bfloat16_head_dims0_xpu_bfloat16 - AssertionError: Tensor-likes are not close!
FAILED [25.5885s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_njt_causal_float32_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [0.0051s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_divisible_multi_token_offset_mask_with_captured_buffer_xpu - AssertionError: Torch not compiled with CUDA enabled
FAILED [24.3189s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod6_float32_head_dims1_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [22.1918s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod1_bfloat16_head_dims0_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 1.3454201379526725 is greater than ref error 0.001172120550190836 by more than 10.0X.
FAILED [21.5563s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod2_float16_head_dims0_xpu_float16 - AssertionError: False is not true : Out Compiled error 4.608919981825865 is greater than ref error 0.003364109592890685 by more than 10.0X.
FAILED [13.0384s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_divisible_multi_token_offset_mask_xpu - AssertionError: False is not true : Out Compiled error 0.01742401653837484 is greater than ref error 2.1662135672508252e-05 by more than 1.1X.
FAILED [23.7120s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod5_float32_head_dims0_xpu_float32 - AssertionError: False is not true : Output/Grad with NaN
FAILED [21.9249s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod8_float16_head_dims0_xpu_float16 - AssertionError: False is not true : Output/Grad with NaN
FAILED [22.0595s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod4_bfloat16_head_dims1_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.02419618689054293 is greater than ref error 0.0001425539293124946 by more than 1.1X.
FAILED [27.4824s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float32_score_mod6_head_dims1_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [14.6977s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_divisible_offset_mask_with_captured_buffer_xpu - AssertionError: False is not true : Out Compiled error 0.02740695336831437 is greater than ref error 3.3055930559450684e-05 by more than 1.1X.
FAILED [21.7974s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod1_bfloat16_head_dims1_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 1.3432760857473318 is greater than ref error 0.0014959765980922524 by more than 10.0X.
FAILED [21.9330s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod7_bfloat16_head_dims0_xpu_bfloat16 - AssertionError: Tensor-likes are not close!
FAILED [21.6215s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod2_float16_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error 5.018416021201188 is greater than ref error 0.003289282095979479 by more than 10.0X.
FAILED [24.2098s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod5_float32_head_dims1_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [22.2712s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod8_float16_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.2958s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod4_float16_head_dims0_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.022907685410628344 is greater than ref error 1.6351976513031915e-05 by more than 1.1X.
FAILED [12.8317s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_divisible_offset_mask_xpu - AssertionError: False is not true : Out Compiled error 0.02819649339394347 is greater than ref error 2.2602800331637366e-05 by more than 1.1X.
FAILED [27.6237s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float32_score_mod6_head_dims2_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [21.8485s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod1_float16_head_dims0_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.1042s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod7_bfloat16_head_dims1_xpu_bfloat16 - AssertionError: Tensor-likes are not close!
FAILED [23.6489s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod2_float32_head_dims0_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [22.0291s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod6_bfloat16_head_dims0_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.001368046840070739 by more than 10.0X.
FAILED [21.8695s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod4_float16_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.023507280658360365 is greater than ref error 1.7563100968313013e-05 by more than 1.1X.
FAILED [22.3723s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_padded_dense_causal_float16_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [23.8718s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod8_float32_head_dims0_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [21.5638s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod1_float16_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [21.4950s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod7_float16_head_dims0_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [23.2519s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod2_float32_head_dims1_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [22.4605s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod6_bfloat16_head_dims1_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.0010255322811853424 by more than 10.0X.
FAILED [23.5210s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod8_float32_head_dims1_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [21.8928s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod7_float16_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [23.3839s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod1_float32_head_dims0_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [0.0000s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod3_bfloat16_head_dims0_xpu_bfloat16
FAILED [21.9834s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod6_float16_head_dims0_xpu_float16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.0001557088130255083 by more than 10.0X.
FAILED [15.3144s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod0_head_dims1_page_size_64_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.07147235726210886 is greater than ref error 4.8129315944429405e-05 by more than 1.1X.
FAILED [23.6589s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod7_float32_head_dims0_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [28.4128s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float32_score_mod7_head_dims1_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [15.0197s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod0_head_dims1_page_size_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.07781057013521864 is greater than ref error 4.8129315944429405e-05 by more than 1.1X.
FAILED [22.1574s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod6_float16_head_dims1_xpu_float16 - AssertionError: False is not true : Out Compiled error inf is greater than ref error 0.00013497051758062462 by more than 10.0X.
FAILED [17.8157s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod0_head_dims2_page_size_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.07007061990968647 is greater than ref error 4.412599562491552e-05 by more than 1.1X.
FAILED [14.8334s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod0_head_dims1_page_size_256_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.0742037702240147 is greater than ref error 4.8129315944429405e-05 by more than 1.1X.
FAILED [23.5888s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod7_float32_head_dims1_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [17.4112s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod0_head_dims2_page_size_256_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.07263498201996264 is greater than ref error 4.412599562491552e-05 by more than 1.1X.
FAILED [27.6218s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_builtin_score_mods_float32_score_mod7_head_dims2_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [23.8060s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod6_float32_head_dims0_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [21.7051s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod8_bfloat16_head_dims0_xpu_bfloat16 - AssertionError: Tensor-likes are not close!
FAILED [17.2422s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod0_head_dims2_page_size_64_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.07250755464703379 is greater than ref error 4.412599562491552e-05 by more than 1.1X.
FAILED [14.6573s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod2_head_dims1_page_size_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.44953105228828105 is greater than ref error 0.0005200395794881008 by more than 1.1X.
FAILED [18.1107s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod3_head_dims2_page_size_256_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.7145678406337481 is greater than ref error 0.00017160034579432443 by more than 10.0X.
FAILED [22.2511s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod5_bfloat16_head_dims0_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 224.2104682694457 is greater than ref error 0.0010757839871797922 by more than 10.0X.
FAILED [15.4817s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod5_head_dims1_page_size_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.1474031503108404 is greater than ref error 0.00014058486647187713 by more than 1.1X.
FAILED [14.7595s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod2_head_dims1_page_size_256_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.43685490567196716 is greater than ref error 0.0005200395794881008 by more than 1.1X.
FAILED [15.3753s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod4_head_dims1_page_size_256_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.0738311233305766 is greater than ref error 4.8129315944429405e-05 by more than 1.1X.
FAILED [17.6531s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod3_head_dims2_page_size_64_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.7151272099541472 is greater than ref error 0.00017160034579432443 by more than 10.0X.
FAILED [22.0065s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_equal_head_dims_score_mod8_bfloat16_head_dims1_xpu_bfloat16 - AssertionError: Tensor-likes are not close!
FAILED [15.4709s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod5_head_dims1_page_size_256_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.1421535605266243 is greater than ref error 0.00014058486647187713 by more than 1.1X.
FAILED [16.0361s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod2_head_dims1_page_size_64_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.4304789993638367 is greater than ref error 0.0005200395794881008 by more than 1.1X.
FAILED [15.5344s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod4_head_dims1_page_size_64_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.0738311233305766 is greater than ref error 4.8129315944429405e-05 by more than 1.1X.
FAILED [18.3832s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod5_head_dims2_page_size_256_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.11260299778270352 is greater than ref error 0.0001464789912952207 by more than 1.1X.
FAILED [15.1302s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod5_head_dims1_page_size_64_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.13848516494166535 is greater than ref error 0.00014058486647187713 by more than 1.1X.
FAILED [17.4222s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_sparse_mulitple_block_size_xpu - AssertionError: Tensor-likes are not close!
FAILED [15.3814s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod3_head_dims1_page_size_128_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [17.5660s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod2_head_dims2_page_size_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.37529537909042343 is greater than ref error 0.0006157928519068745 by more than 1.1X.
FAILED [17.5439s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod4_head_dims2_page_size_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.07560655014273629 is greater than ref error 4.412599562491552e-05 by more than 1.1X.
FAILED [15.9781s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod6_head_dims1_page_size_128_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [17.8274s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod5_head_dims2_page_size_64_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.1152499169076956 is greater than ref error 0.0001464789912952207 by more than 1.1X.
FAILED [17.7173s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod5_head_dims2_page_size_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.11345573834894418 is greater than ref error 0.0001464789912952207 by more than 1.1X.
FAILED [15.1511s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod3_head_dims1_page_size_256_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [14.9494s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod1_head_dims1_page_size_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.16848931468775225 is greater than ref error 0.00017543317174728882 by more than 1.1X.
FAILED [17.5835s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod2_head_dims2_page_size_256_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.43600237824136245 is greater than ref error 0.0006157928519068745 by more than 1.1X.
FAILED [17.9336s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod4_head_dims2_page_size_256_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.07355692348767026 is greater than ref error 4.412599562491552e-05 by more than 1.1X.
FAILED [15.5914s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod6_head_dims1_page_size_256_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [15.1812s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod3_head_dims1_page_size_64_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [15.7151s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod7_head_dims1_page_size_256_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.09990616703893715 is greater than ref error 7.414601025017982e-05 by more than 1.1X.
FAILED [17.9468s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod6_head_dims2_page_size_64_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.748235569302409 is greater than ref error 0.00017160034579432443 by more than 10.0X.
FAILED [16.2860s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod4_head_dims1_page_size_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.0757171856936616 is greater than ref error 4.8129315944429405e-05 by more than 1.1X.
FAILED [17.4237s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod2_head_dims2_page_size_64_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.42627043915822027 is greater than ref error 0.0006157928519068745 by more than 1.1X.
FAILED [17.6271s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod4_head_dims2_page_size_64_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.07480259867280706 is greater than ref error 4.412599562491552e-05 by more than 1.1X.
FAILED [15.6181s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod6_head_dims1_page_size_64_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [15.5769s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod7_head_dims1_page_size_64_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.10340989579323201 is greater than ref error 7.414601025017982e-05 by more than 1.1X.
FAILED [17.6894s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod3_head_dims2_page_size_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.697545856325263 is greater than ref error 0.00017160034579432443 by more than 10.0X.
FAILED [17.5572s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod7_head_dims2_page_size_64_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.21543665403300435 is greater than ref error 6.228557640524872e-05 by more than 1.1X.
FAILED [15.3715s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod8_head_dims1_page_size_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.07598256732057232 is greater than ref error 4.8129315944429405e-05 by more than 1.1X.
FAILED [17.8070s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod8_head_dims2_page_size_256_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.07590426366435538 is greater than ref error 4.412599562491552e-05 by more than 1.1X.
FAILED [17.8468s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod6_head_dims2_page_size_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.7761378302235187 is greater than ref error 0.00017160034579432443 by more than 10.0X.
FAILED [17.7042s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod7_head_dims2_page_size_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.2173200802580238 is greater than ref error 6.228557640524872e-05 by more than 1.1X.
FAILED [21.7202s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_silu_on_score_float16_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.022891767878119064 is greater than ref error 1.5799023128344582e-05 by more than 1.1X.
FAILED [15.4272s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod8_head_dims1_page_size_256_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.0742037702240147 is greater than ref error 4.8129315944429405e-05 by more than 1.1X.
FAILED [17.6612s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod8_head_dims2_page_size_64_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.06767064243402594 is greater than ref error 4.412599562491552e-05 by more than 1.1X.
FAILED [17.5019s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod6_head_dims2_page_size_256_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.7082585449203123 is greater than ref error 0.00017160034579432443 by more than 10.0X.
FAILED [17.5892s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod7_head_dims2_page_size_256_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.269451050209244 is greater than ref error 6.228557640524872e-05 by more than 1.1X.
FAILED [15.3666s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod8_head_dims1_page_size_64_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.07781057013521864 is greater than ref error 4.8129315944429405e-05 by more than 1.1X.
FAILED [22.3889s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_skip_odd_keys_bfloat16_xpu_bfloat16 - AssertionError: False is not true : Out Compiled error 0.03365975271645137 is greater than ref error 0.00017587732186179306 by more than 1.1X.
FAILED [15.6681s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod7_head_dims1_page_size_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.10103700527098847 is greater than ref error 7.414601025017982e-05 by more than 1.1X.
FAILED [17.8508s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s0_v_s0_head_dims2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [18.2706s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s0_v_s1_head_dims2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.0883s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_recompile_changed_score_mod_float16_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.022714596319718325 is greater than ref error 1.6445839539482343e-05 by more than 1.1X.
FAILED [45.2676s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_skip_odd_keys_float32_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [17.7755s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod8_head_dims2_page_size_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.07091968792768943 is greater than ref error 4.412599562491552e-05 by more than 1.1X.
FAILED [18.2968s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s0_v_s2_head_dims2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [21.8747s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_skip_odd_keys_float16_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.030990117757616974 is greater than ref error 2.19345386642078e-05 by more than 1.1X.
FAILED [18.3815s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s0_v_s3_head_dims2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.6112s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_seq_masking_float16_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.1125s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s1_v_s0_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [18.5238s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s1_v_s1_head_dims2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [18.3102s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s1_v_s0_head_dims2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.1524s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s0_v_s1_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.3747s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s0_v_s2_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.4519s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s0_v_s0_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.0979s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s0_v_s3_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.1568s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s1_v_s1_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [18.7193s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s1_v_s3_head_dims2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.2548s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s1_v_s2_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.2422s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s2_v_s0_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [18.3930s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s2_v_s1_head_dims2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [18.0990s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s1_v_s2_head_dims2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [18.6329s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s2_v_s0_head_dims2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.2708s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s2_v_s1_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.1669s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s1_v_s3_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [18.0054s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s2_v_s2_head_dims2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [18.4187s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s2_v_s3_head_dims2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.7557s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s2_v_s2_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.0910s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s3_v_s0_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [18.9143s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s3_v_s0_head_dims2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.4828s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s2_v_s3_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [18.1068s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s3_v_s1_head_dims2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [22.4480s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s3_v_s1_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [17.4488s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_windowed_partial_block_vs_sdpa_paged_attention_xpu - AssertionError: False is not true : Out Compiled error 0.01617082569718646 is greater than ref error 1.1149073756663964e-05 by more than 1.1X.
FAILED [22.1030s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s3_v_s3_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [17.4610s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s3_v_s2_head_dims2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [21.9756s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s3_v_s2_head_dims1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [18.1195s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_strided_inputs_float16_k_s3_v_s3_head_dims2_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [17.8239s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_windowed_full_mask_vs_sdpa_paged_attention_xpu - AssertionError: False is not true : Out Compiled error 0.016937407239053952 is greater than ref error 1.1149073756663964e-05 by more than 1.1X.
FAILED [12.6098s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_windowed_partial_block_vs_sdpa_xpu - AssertionError: False is not true : Out Compiled error inf is greater than ref error 1.1149073756663964e-05 by more than 1.1X.
FAILED [14.8055s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_windowed_no_mask_vs_sdpa_xpu - AssertionError: False is not true : Out Compiled error 0.01679543891224409 is greater than ref error 1.1149073756663964e-05 by more than 1.1X.
FAILED [12.8428s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_windowed_full_mask_vs_sdpa_xpu - AssertionError: False is not true : Out Compiled error inf is greater than ref error 1.1149073756663964e-05 by more than 1.1X.
FAILED [18.2713s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_windowed_no_mask_vs_sdpa_paged_attention_xpu - AssertionError: False is not true : Out Compiled error 0.016292724061776606 is greater than ref error 1.1149073756663964e-05 by more than 1.1X.
FAILED [22.7015s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_load_from_bias_seq_batch_float16_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.022083244846520883 is greater than ref error 2.179386190180216e-05 by more than 1.1X.
FAILED [22.6727s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_load_from_bias_head_seq_batch_float16_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.021532936153954317 is greater than ref error 2.1765403862680297e-05 by more than 1.1X.
FAILED [22.0469s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_logsumexp_correctness_bfloat16_score_mod0_xpu_bfloat16 - AssertionError: Tensor-likes are not close!
FAILED [22.8384s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_load_from_bias_seq_only_float16_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.024423766432413323 is greater than ref error 2.3206530392965934e-05 by more than 1.1X.
FAILED [22.4466s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_logsumexp_correctness_bfloat16_score_mod1_xpu_bfloat16 - AssertionError: Tensor-likes are not close!
FAILED [22.2402s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_logsumexp_correctness_float16_score_mod0_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [24.4628s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_logsumexp_correctness_float32_score_mod0_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [24.1998s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_logsumexp_correctness_float32_score_mod1_xpu_float32 - AssertionError: Tensor-likes are not close!
FAILED [22.7321s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_logsumexp_correctness_float16_score_mod1_xpu_float16 - AssertionError: Tensor-likes are not close!
FAILED [14.6984s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod1_head_dims1_page_size_256_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.18105732518233758 is greater than ref error 0.00017543317174728882 by more than 1.1X.
FAILED [15.0346s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod1_head_dims1_page_size_64_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.18105732518233758 is greater than ref error 0.00017543317174728882 by more than 1.1X.
FAILED [25.8828s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_max_autotune_xpu - AssertionError: Tensor-likes are not close!
FAILED [27.3095s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_max_autotune_with_captured_xpu - AssertionError: False is not true : Out Compiled error 0.026626568199411275 is greater than ref error 2.369540654010762e-05 by more than 1.1X.
FAILED [17.2776s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod1_head_dims2_page_size_128_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.17310399416131325 is greater than ref error 0.00018149984157827815 by more than 1.1X.
FAILED [17.0753s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod1_head_dims2_page_size_256_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.1789110207797917 is greater than ref error 0.00018149984157827815 by more than 1.1X.
FAILED [16.8655s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_paged_attention_page_size_float16_score_mod1_head_dims2_page_size_64_xpu_float16 - AssertionError: False is not true : Out Compiled error 0.1734951345912008 is greater than ref error 0.00018149984157827815 by more than 1.1X.

Most cases passed if this commit reverted: 9facf00.

FAILED [0.0000s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_larger_block_mask_bug_float16_xpu_float16
FAILED [0.0049s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_divisible_multi_token_offset_mask_with_captured_buffer_xpu - AssertionError: Torch not compiled with CUDA enabled
=== 2 failed, 570 passed, 2 xfailed in 1727.57s (0:28:47) ===

@chengjunlu
Copy link
Contributor

Thanks @ESI-SYD help to narrow down the regression commit.

I have created an PR. Please help to review the changes #3999

@chengjunlu chengjunlu linked a pull request Apr 24, 2025 that will close this issue
@chengjunlu
Copy link
Contributor

FAILED [0.0000s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_larger_block_mask_bug_float16_xpu_float16
FAILED [0.0049s] test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_divisible_multi_token_offset_mask_with_captured_buffer_xpu - AssertionError: Torch not compiled with CUDA enabled
=== 2 failed, 570 passed, 2 xfailed in 1727.57s (0:28:47) ===

These two failure cases passed on the latest Triton main.
@hoshibara Please try the latest Triton main branch.

@hoshibara
Copy link
Author

These two failure cases passed on the latest Triton main. @hoshibara Please try the latest Triton main branch.

OK, I'll check if latest triton commit can solve all flex decoding UTs.

@hoshibara
Copy link
Author

=========================== short test summary info ============================
FAILED [0.0049s] ../hoshibara-pytorch/test/inductor/test_flex_decoding.py::TestFlexDecodingXPU::test_non_divisible_multi_token_offset_mask_with_captured_buffer_xpu
============ 1 failed, 571 passed, 2 xfailed in 13456.67s (3:44:16) ============
Done

Nearly all cases are passing. The last remaining one is an issue with UT enablement, and I will fix it in the next commit.

@hoshibara
Copy link
Author

There is a regression case. Maybe caused by updating driver version.

python test/inductor/test_flex_decoding.py TestFlexDecodingXPU.test_larger_block_mask_bug_float16_xpu_float16

Env:

[conda] torch                     2.8.0a0+gitffdb463           dev_0    <develop>
[conda] triton                    3.3.0+git5271aa4e          pypi_0    pypi

@chengjunlu
Copy link
Contributor

There is a regression case. Maybe caused by updating driver version.

python test/inductor/test_flex_decoding.py TestFlexDecodingXPU.test_larger_block_mask_bug_float16_xpu_float16

Env:

[conda] torch                     2.8.0a0+gitffdb463           dev_0    <develop>
[conda] triton                    3.3.0+git5271aa4e          pypi_0    pypi

The test passed on internal IGC release intel-igc-opencl-devel_2.8.0+18783.

Let's re-verify it with the next agama release.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging a pull request may close this issue.

4 participants