Update gm.print_readable to include Annotation by SherlockNoMad · Pull Request #165397 · pytorch/pytorch

SherlockNoMad · 2025-10-14T03:26:51Z

Stack from ghstack (oldest at bottom):

-> Update gm.print_readable to include Annotation #165397

Sample output

[rank0]:        # Annotation: {'compile_with_inductor': 'flex_attention'} File: /data/users/bahuang/pytorch/torch/nn/attention/flex_attention.py:1490 in flex_attention, code: out, lse, max_scores = flex_attention_hop(
[rank0]:        score_mod_2 = self.score_mod_2
[rank0]:        mask_fn_2 = self.mask_fn_2
[rank0]:        flex_attention_1 = torch.ops.higher_order.flex_attention(xq_5, xk_5, xv_3, score_mod_2, (2048, 2048, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_indices, 128, 128, mask_fn_2), 0.25, {'PRESCALE_QK': False, 'ROWS_GUARANTEED_SAFE': False, 'BLOCKS_ARE_CONTIGUOUS': False, 'WRITE_DQ': True, 'OUTPUT_LOGSUMEXP': True, 'OUTPUT_MAX': False}, (), (g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___mask_mod___closure___0_cell_contents,));  xq_5 = xk_5 = xv_3 = score_mod_2 = mask_fn_2 = None
[rank0]:        out_2: "bf16[8, 4, 2048, 16]" = flex_attention_1[0];  flex_attention_1 = None

cc @ezyang @EikanWang @jgong5 @wenzhe-nrv @voznesenskym @penguinwu @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @jiayisunx @ipiszy @chenyang78 @kadeng @muchulee8 @amjames @chauhang @aakhundov @coconutruben @Lucaskabela

[ghstack-poisoned]

pytorch-bot · 2025-10-14T03:26:54Z

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/165397

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

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

✅ No Failures

As of commit 559618d with merge base 2ce894b ():
💚 Looks good so far! There are no failures yet. 💚

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

ghstack-source-id: 19e3411 Pull Request resolved: #165397

Sample output ``` [rank0]: # Annotation: {'compile_with_inductor': 'flex_attention'} File: /data/users/bahuang/pytorch/torch/nn/attention/flex_attention.py:1490 in flex_attention, code: out, lse, max_scores = flex_attention_hop( [rank0]: score_mod_2 = self.score_mod_2 [rank0]: mask_fn_2 = self.mask_fn_2 [rank0]: flex_attention_1 = torch.ops.higher_order.flex_attention(xq_5, xk_5, xv_3, score_mod_2, (2048, 2048, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_indices, 128, 128, mask_fn_2), 0.25, {'PRESCALE_QK': False, 'ROWS_GUARANTEED_SAFE': False, 'BLOCKS_ARE_CONTIGUOUS': False, 'WRITE_DQ': True, 'OUTPUT_LOGSUMEXP': True, 'OUTPUT_MAX': False}, (), (g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___mask_mod___closure___0_cell_contents,)); xq_5 = xk_5 = xv_3 = score_mod_2 = mask_fn_2 = None [rank0]: out_2: "bf16[8, 4, 2048, 16]" = flex_attention_1[0]; flex_attention_1 = None ``` cc ezyang EikanWang jgong5 wenzhe-nrv [ghstack-poisoned]

ghstack-source-id: bf7628b Pull Request resolved: #165397

Sample output ``` [rank0]: # Annotation: {'compile_with_inductor': 'flex_attention'} File: /data/users/bahuang/pytorch/torch/nn/attention/flex_attention.py:1490 in flex_attention, code: out, lse, max_scores = flex_attention_hop( [rank0]: score_mod_2 = self.score_mod_2 [rank0]: mask_fn_2 = self.mask_fn_2 [rank0]: flex_attention_1 = torch.ops.higher_order.flex_attention(xq_5, xk_5, xv_3, score_mod_2, (2048, 2048, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_indices, 128, 128, mask_fn_2), 0.25, {'PRESCALE_QK': False, 'ROWS_GUARANTEED_SAFE': False, 'BLOCKS_ARE_CONTIGUOUS': False, 'WRITE_DQ': True, 'OUTPUT_LOGSUMEXP': True, 'OUTPUT_MAX': False}, (), (g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___mask_mod___closure___0_cell_contents,)); xq_5 = xk_5 = xv_3 = score_mod_2 = mask_fn_2 = None [rank0]: out_2: "bf16[8, 4, 2048, 16]" = flex_attention_1[0]; flex_attention_1 = None ``` cc ezyang EikanWang jgong5 wenzhe-nrv voznesenskym penguinwu Guobing-Chen XiaobingSuper zhuhaozhe blzheng jiayisunx chenyang78 kadeng chauhang amjames Lucaskabela [ghstack-poisoned]

ghstack-source-id: b7a8741 Pull Request resolved: #165397

Sample output ``` [rank0]: # Annotation: {'compile_with_inductor': 'flex_attention'} File: /data/users/bahuang/pytorch/torch/nn/attention/flex_attention.py:1490 in flex_attention, code: out, lse, max_scores = flex_attention_hop( [rank0]: score_mod_2 = self.score_mod_2 [rank0]: mask_fn_2 = self.mask_fn_2 [rank0]: flex_attention_1 = torch.ops.higher_order.flex_attention(xq_5, xk_5, xv_3, score_mod_2, (2048, 2048, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_indices, 128, 128, mask_fn_2), 0.25, {'PRESCALE_QK': False, 'ROWS_GUARANTEED_SAFE': False, 'BLOCKS_ARE_CONTIGUOUS': False, 'WRITE_DQ': True, 'OUTPUT_LOGSUMEXP': True, 'OUTPUT_MAX': False}, (), (g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___mask_mod___closure___0_cell_contents,)); xq_5 = xk_5 = xv_3 = score_mod_2 = mask_fn_2 = None [rank0]: out_2: "bf16[8, 4, 2048, 16]" = flex_attention_1[0]; flex_attention_1 = None ``` cc ezyang EikanWang jgong5 wenzhe-nrv voznesenskym penguinwu Guobing-Chen XiaobingSuper zhuhaozhe blzheng jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben Lucaskabela [ghstack-poisoned]

ghstack-source-id: bd647e4 Pull Request resolved: #165397

Sample output ``` [rank0]: # Annotation: {'compile_with_inductor': 'flex_attention'} File: /data/users/bahuang/pytorch/torch/nn/attention/flex_attention.py:1490 in flex_attention, code: out, lse, max_scores = flex_attention_hop( [rank0]: score_mod_2 = self.score_mod_2 [rank0]: mask_fn_2 = self.mask_fn_2 [rank0]: flex_attention_1 = torch.ops.higher_order.flex_attention(xq_5, xk_5, xv_3, score_mod_2, (2048, 2048, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_indices, 128, 128, mask_fn_2), 0.25, {'PRESCALE_QK': False, 'ROWS_GUARANTEED_SAFE': False, 'BLOCKS_ARE_CONTIGUOUS': False, 'WRITE_DQ': True, 'OUTPUT_LOGSUMEXP': True, 'OUTPUT_MAX': False}, (), (g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___mask_mod___closure___0_cell_contents,)); xq_5 = xk_5 = xv_3 = score_mod_2 = mask_fn_2 = None [rank0]: out_2: "bf16[8, 4, 2048, 16]" = flex_attention_1[0]; flex_attention_1 = None ``` cc ezyang EikanWang jgong5 wenzhe-nrv voznesenskym penguinwu Guobing-Chen XiaobingSuper zhuhaozhe blzheng jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben Lucaskabela [ghstack-poisoned]

ghstack-source-id: 5676b78 Pull Request resolved: #165397

Sample output ``` [rank0]: # Annotation: {'compile_with_inductor': 'flex_attention'} File: /data/users/bahuang/pytorch/torch/nn/attention/flex_attention.py:1490 in flex_attention, code: out, lse, max_scores = flex_attention_hop( [rank0]: score_mod_2 = self.score_mod_2 [rank0]: mask_fn_2 = self.mask_fn_2 [rank0]: flex_attention_1 = torch.ops.higher_order.flex_attention(xq_5, xk_5, xv_3, score_mod_2, (2048, 2048, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_indices, 128, 128, mask_fn_2), 0.25, {'PRESCALE_QK': False, 'ROWS_GUARANTEED_SAFE': False, 'BLOCKS_ARE_CONTIGUOUS': False, 'WRITE_DQ': True, 'OUTPUT_LOGSUMEXP': True, 'OUTPUT_MAX': False}, (), (g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___mask_mod___closure___0_cell_contents,)); xq_5 = xk_5 = xv_3 = score_mod_2 = mask_fn_2 = None [rank0]: out_2: "bf16[8, 4, 2048, 16]" = flex_attention_1[0]; flex_attention_1 = None ``` cc ezyang EikanWang jgong5 wenzhe-nrv voznesenskym penguinwu Guobing-Chen XiaobingSuper zhuhaozhe blzheng jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben Lucaskabela [ghstack-poisoned]

ghstack-source-id: 796d1d4 Pull Request resolved: #165397

Sample output ``` [rank0]: # Annotation: {'compile_with_inductor': 'flex_attention'} File: /data/users/bahuang/pytorch/torch/nn/attention/flex_attention.py:1490 in flex_attention, code: out, lse, max_scores = flex_attention_hop( [rank0]: score_mod_2 = self.score_mod_2 [rank0]: mask_fn_2 = self.mask_fn_2 [rank0]: flex_attention_1 = torch.ops.higher_order.flex_attention(xq_5, xk_5, xv_3, score_mod_2, (2048, 2048, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_indices, 128, 128, mask_fn_2), 0.25, {'PRESCALE_QK': False, 'ROWS_GUARANTEED_SAFE': False, 'BLOCKS_ARE_CONTIGUOUS': False, 'WRITE_DQ': True, 'OUTPUT_LOGSUMEXP': True, 'OUTPUT_MAX': False}, (), (g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___mask_mod___closure___0_cell_contents,)); xq_5 = xk_5 = xv_3 = score_mod_2 = mask_fn_2 = None [rank0]: out_2: "bf16[8, 4, 2048, 16]" = flex_attention_1[0]; flex_attention_1 = None ``` cc ezyang EikanWang jgong5 wenzhe-nrv voznesenskym penguinwu Guobing-Chen XiaobingSuper zhuhaozhe blzheng jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben Lucaskabela [ghstack-poisoned]

ghstack-source-id: e03d1b2 Pull Request resolved: #165397

Sample output ``` [rank0]: # Annotation: {'compile_with_inductor': 'flex_attention'} File: /data/users/bahuang/pytorch/torch/nn/attention/flex_attention.py:1490 in flex_attention, code: out, lse, max_scores = flex_attention_hop( [rank0]: score_mod_2 = self.score_mod_2 [rank0]: mask_fn_2 = self.mask_fn_2 [rank0]: flex_attention_1 = torch.ops.higher_order.flex_attention(xq_5, xk_5, xv_3, score_mod_2, (2048, 2048, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_indices, 128, 128, mask_fn_2), 0.25, {'PRESCALE_QK': False, 'ROWS_GUARANTEED_SAFE': False, 'BLOCKS_ARE_CONTIGUOUS': False, 'WRITE_DQ': True, 'OUTPUT_LOGSUMEXP': True, 'OUTPUT_MAX': False}, (), (g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___mask_mod___closure___0_cell_contents,)); xq_5 = xk_5 = xv_3 = score_mod_2 = mask_fn_2 = None [rank0]: out_2: "bf16[8, 4, 2048, 16]" = flex_attention_1[0]; flex_attention_1 = None ``` cc ezyang EikanWang jgong5 wenzhe-nrv voznesenskym penguinwu Guobing-Chen XiaobingSuper zhuhaozhe blzheng jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben Lucaskabela [ghstack-poisoned]

ghstack-source-id: 096334e Pull Request resolved: #165397

Sample output ``` [rank0]: # Annotation: {'compile_with_inductor': 'flex_attention'} File: /data/users/bahuang/pytorch/torch/nn/attention/flex_attention.py:1490 in flex_attention, code: out, lse, max_scores = flex_attention_hop( [rank0]: score_mod_2 = self.score_mod_2 [rank0]: mask_fn_2 = self.mask_fn_2 [rank0]: flex_attention_1 = torch.ops.higher_order.flex_attention(xq_5, xk_5, xv_3, score_mod_2, (2048, 2048, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_indices, 128, 128, mask_fn_2), 0.25, {'PRESCALE_QK': False, 'ROWS_GUARANTEED_SAFE': False, 'BLOCKS_ARE_CONTIGUOUS': False, 'WRITE_DQ': True, 'OUTPUT_LOGSUMEXP': True, 'OUTPUT_MAX': False}, (), (g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___mask_mod___closure___0_cell_contents,)); xq_5 = xk_5 = xv_3 = score_mod_2 = mask_fn_2 = None [rank0]: out_2: "bf16[8, 4, 2048, 16]" = flex_attention_1[0]; flex_attention_1 = None ``` cc ezyang EikanWang jgong5 wenzhe-nrv voznesenskym penguinwu Guobing-Chen XiaobingSuper zhuhaozhe blzheng jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben Lucaskabela [ghstack-poisoned]

ghstack-source-id: cdc348a Pull Request resolved: #165397

Sample output ``` [rank0]: # Annotation: {'compile_with_inductor': 'flex_attention'} File: /data/users/bahuang/pytorch/torch/nn/attention/flex_attention.py:1490 in flex_attention, code: out, lse, max_scores = flex_attention_hop( [rank0]: score_mod_2 = self.score_mod_2 [rank0]: mask_fn_2 = self.mask_fn_2 [rank0]: flex_attention_1 = torch.ops.higher_order.flex_attention(xq_5, xk_5, xv_3, score_mod_2, (2048, 2048, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_indices, 128, 128, mask_fn_2), 0.25, {'PRESCALE_QK': False, 'ROWS_GUARANTEED_SAFE': False, 'BLOCKS_ARE_CONTIGUOUS': False, 'WRITE_DQ': True, 'OUTPUT_LOGSUMEXP': True, 'OUTPUT_MAX': False}, (), (g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___mask_mod___closure___0_cell_contents,)); xq_5 = xk_5 = xv_3 = score_mod_2 = mask_fn_2 = None [rank0]: out_2: "bf16[8, 4, 2048, 16]" = flex_attention_1[0]; flex_attention_1 = None ``` cc ezyang EikanWang jgong5 wenzhe-nrv voznesenskym penguinwu Guobing-Chen XiaobingSuper zhuhaozhe blzheng jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben Lucaskabela [ghstack-poisoned]

malfet · 2025-10-17T22:34:07Z

@pytorchbot revert -m "I don't know how/why, but it breaks windows tests, see https://hud.pytorch.org/hud/pytorch/pytorch/2e22b1a61ea20a54448edf34a5d22fbe8391d626/1?per_page=50&name_filter=win&mergeEphemeralLF=true" -c nosignal

pytorchmergebot · 2025-10-17T22:35:46Z

@pytorchbot successfully started a revert job. Check the current status here.
Questions? Feedback? Please reach out to the PyTorch DevX Team

This reverts commit 7a65770. Reverted #165397 on behalf of https://github.com/malfet due to I don't know how/why, but it breaks windows tests, see https://hud.pytorch.org/hud/pytorch/pytorch/2e22b1a61ea20a54448edf34a5d22fbe8391d626/1?per_page=50&name_filter=win&mergeEphemeralLF=true ([comment](#165397 (comment)))

pytorchmergebot · 2025-10-17T22:35:54Z

@SherlockNoMad your PR has been successfully reverted.

Sample output ``` [rank0]: # Annotation: {'compile_with_inductor': 'flex_attention'} File: /data/users/bahuang/pytorch/torch/nn/attention/flex_attention.py:1490 in flex_attention, code: out, lse, max_scores = flex_attention_hop( [rank0]: score_mod_2 = self.score_mod_2 [rank0]: mask_fn_2 = self.mask_fn_2 [rank0]: flex_attention_1 = torch.ops.higher_order.flex_attention(xq_5, xk_5, xv_3, score_mod_2, (2048, 2048, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_indices, 128, 128, mask_fn_2), 0.25, {'PRESCALE_QK': False, 'ROWS_GUARANTEED_SAFE': False, 'BLOCKS_ARE_CONTIGUOUS': False, 'WRITE_DQ': True, 'OUTPUT_LOGSUMEXP': True, 'OUTPUT_MAX': False}, (), (g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___mask_mod___closure___0_cell_contents,)); xq_5 = xk_5 = xv_3 = score_mod_2 = mask_fn_2 = None [rank0]: out_2: "bf16[8, 4, 2048, 16]" = flex_attention_1[0]; flex_attention_1 = None ``` Pull Request resolved: pytorch#165397 Approved by: https://github.com/yushangdi, https://github.com/anijain2305

This reverts commit 7a65770. Reverted pytorch#165397 on behalf of https://github.com/malfet due to I don't know how/why, but it breaks windows tests, see https://hud.pytorch.org/hud/pytorch/pytorch/2e22b1a61ea20a54448edf34a5d22fbe8391d626/1?per_page=50&name_filter=win&mergeEphemeralLF=true ([comment](pytorch#165397 (comment)))

Sample output ``` [rank0]: # Annotation: {'compile_with_inductor': 'flex_attention'} File: /data/users/bahuang/pytorch/torch/nn/attention/flex_attention.py:1490 in flex_attention, code: out, lse, max_scores = flex_attention_hop( [rank0]: score_mod_2 = self.score_mod_2 [rank0]: mask_fn_2 = self.mask_fn_2 [rank0]: flex_attention_1 = torch.ops.higher_order.flex_attention(xq_5, xk_5, xv_3, score_mod_2, (2048, 2048, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_indices, 128, 128, mask_fn_2), 0.25, {'PRESCALE_QK': False, 'ROWS_GUARANTEED_SAFE': False, 'BLOCKS_ARE_CONTIGUOUS': False, 'WRITE_DQ': True, 'OUTPUT_LOGSUMEXP': True, 'OUTPUT_MAX': False}, (), (g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___mask_mod___closure___0_cell_contents,)); xq_5 = xk_5 = xv_3 = score_mod_2 = mask_fn_2 = None [rank0]: out_2: "bf16[8, 4, 2048, 16]" = flex_attention_1[0]; flex_attention_1 = None ``` Pull Request resolved: pytorch#165397 Approved by: https://github.com/yushangdi, https://github.com/anijain2305

This reverts commit 7a65770. Reverted pytorch#165397 on behalf of https://github.com/malfet due to I don't know how/why, but it breaks windows tests, see https://hud.pytorch.org/hud/pytorch/pytorch/2e22b1a61ea20a54448edf34a5d22fbe8391d626/1?per_page=50&name_filter=win&mergeEphemeralLF=true ([comment](pytorch#165397 (comment)))

Sample output ``` [rank0]: # Annotation: {'compile_with_inductor': 'flex_attention'} File: /data/users/bahuang/pytorch/torch/nn/attention/flex_attention.py:1490 in flex_attention, code: out, lse, max_scores = flex_attention_hop( [rank0]: score_mod_2 = self.score_mod_2 [rank0]: mask_fn_2 = self.mask_fn_2 [rank0]: flex_attention_1 = torch.ops.higher_order.flex_attention(xq_5, xk_5, xv_3, score_mod_2, (2048, 2048, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_indices, 128, 128, mask_fn_2), 0.25, {'PRESCALE_QK': False, 'ROWS_GUARANTEED_SAFE': False, 'BLOCKS_ARE_CONTIGUOUS': False, 'WRITE_DQ': True, 'OUTPUT_LOGSUMEXP': True, 'OUTPUT_MAX': False}, (), (g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___mask_mod___closure___0_cell_contents,)); xq_5 = xk_5 = xv_3 = score_mod_2 = mask_fn_2 = None [rank0]: out_2: "bf16[8, 4, 2048, 16]" = flex_attention_1[0]; flex_attention_1 = None ``` cc ezyang EikanWang jgong5 wenzhe-nrv voznesenskym penguinwu Guobing-Chen XiaobingSuper zhuhaozhe blzheng jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben Lucaskabela [ghstack-poisoned]

ghstack-source-id: d9191e9 Pull Request resolved: #165397

Sample output ``` [rank0]: # Annotation: {'compile_with_inductor': 'flex_attention'} File: /data/users/bahuang/pytorch/torch/nn/attention/flex_attention.py:1490 in flex_attention, code: out, lse, max_scores = flex_attention_hop( [rank0]: score_mod_2 = self.score_mod_2 [rank0]: mask_fn_2 = self.mask_fn_2 [rank0]: flex_attention_1 = torch.ops.higher_order.flex_attention(xq_5, xk_5, xv_3, score_mod_2, (2048, 2048, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_indices, 128, 128, mask_fn_2), 0.25, {'PRESCALE_QK': False, 'ROWS_GUARANTEED_SAFE': False, 'BLOCKS_ARE_CONTIGUOUS': False, 'WRITE_DQ': True, 'OUTPUT_LOGSUMEXP': True, 'OUTPUT_MAX': False}, (), (g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___mask_mod___closure___0_cell_contents,)); xq_5 = xk_5 = xv_3 = score_mod_2 = mask_fn_2 = None [rank0]: out_2: "bf16[8, 4, 2048, 16]" = flex_attention_1[0]; flex_attention_1 = None ``` cc ezyang EikanWang jgong5 wenzhe-nrv voznesenskym penguinwu Guobing-Chen XiaobingSuper zhuhaozhe blzheng jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben Lucaskabela [ghstack-poisoned]

ghstack-source-id: b5f027a Pull Request resolved: #165397

Sample output ``` [rank0]: # Annotation: {'compile_with_inductor': 'flex_attention'} File: /data/users/bahuang/pytorch/torch/nn/attention/flex_attention.py:1490 in flex_attention, code: out, lse, max_scores = flex_attention_hop( [rank0]: score_mod_2 = self.score_mod_2 [rank0]: mask_fn_2 = self.mask_fn_2 [rank0]: flex_attention_1 = torch.ops.higher_order.flex_attention(xq_5, xk_5, xv_3, score_mod_2, (2048, 2048, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_kv_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___q_indices, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_num_blocks, g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___full_q_indices, 128, 128, mask_fn_2), 0.25, {'PRESCALE_QK': False, 'ROWS_GUARANTEED_SAFE': False, 'BLOCKS_ARE_CONTIGUOUS': False, 'WRITE_DQ': True, 'OUTPUT_LOGSUMEXP': True, 'OUTPUT_MAX': False}, (), (g____import_torchtitan_dot_models_dot_attention___flex_attention_block_masks___block_causal___none___mask_mod___closure___0_cell_contents,)); xq_5 = xk_5 = xv_3 = score_mod_2 = mask_fn_2 = None [rank0]: out_2: "bf16[8, 4, 2048, 16]" = flex_attention_1[0]; flex_attention_1 = None ``` cc ezyang EikanWang jgong5 wenzhe-nrv voznesenskym penguinwu Guobing-Chen XiaobingSuper zhuhaozhe blzheng jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben Lucaskabela [ghstack-poisoned]

ghstack-source-id: 160b58c Pull Request resolved: #165397

SherlockNoMad · 2025-10-28T13:47:06Z

@pytorchbot merge

pytorchmergebot · 2025-10-28T13:49:02Z

Merge started

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

Learn more about merging in the wiki.

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

Advanced Debugging

Check the merge workflow status
here

Update gm.print_readable to include Annotation

c1da78d

[ghstack-poisoned]

pytorch-bot bot added the release notes: fx release notes category label Oct 14, 2025

SherlockNoMad added a commit that referenced this pull request Oct 14, 2025

Update gm.print_readable to include Annotation

a9a408d

ghstack-source-id: 19e3411 Pull Request resolved: #165397

facebook-github-bot added the fx label Oct 14, 2025

SherlockNoMad requested review from anijain2305 and yushangdi October 14, 2025 03:27

SherlockNoMad added the topic: not user facing topic category label Oct 14, 2025

yushangdi approved these changes Oct 14, 2025

View reviewed changes

anijain2305 approved these changes Oct 14, 2025

View reviewed changes

pytorch-bot bot added ciflow/inductor module: dynamo labels Oct 14, 2025

SherlockNoMad added a commit that referenced this pull request Oct 14, 2025

Update gm.print_readable to include Annotation

71e2328

ghstack-source-id: bf7628b Pull Request resolved: #165397

SherlockNoMad added a commit that referenced this pull request Oct 14, 2025

Update gm.print_readable to include Annotation

8c2c72d

ghstack-source-id: b7a8741 Pull Request resolved: #165397

pytorch-bot bot added the module: inductor label Oct 14, 2025

SherlockNoMad added a commit that referenced this pull request Oct 14, 2025

Update gm.print_readable to include Annotation

4bee05a

ghstack-source-id: bd647e4 Pull Request resolved: #165397

SherlockNoMad added a commit that referenced this pull request Oct 15, 2025

Update gm.print_readable to include Annotation

076c662

ghstack-source-id: 5676b78 Pull Request resolved: #165397

SherlockNoMad added a commit that referenced this pull request Oct 15, 2025

ghstack Update gm.print_readable to include Annotation

2c9ca56

ghstack-source-id: 796d1d4 Pull Request resolved: #165397

SherlockNoMad added a commit that referenced this pull request Oct 15, 2025

ghstack Update gm.print_readable to include Annotation

21774ab

ghstack-source-id: e03d1b2 Pull Request resolved: #165397

SherlockNoMad added a commit that referenced this pull request Oct 15, 2025

ghstack Update gm.print_readable to include Annotation

5b6d1b7

ghstack-source-id: 096334e Pull Request resolved: #165397

SherlockNoMad added a commit that referenced this pull request Oct 15, 2025

ghstack Update gm.print_readable to include Annotation

1cb24a5

ghstack-source-id: cdc348a Pull Request resolved: #165397

pytorchmergebot added Reverted ci-no-td Do not run TD on this PR labels Oct 17, 2025

pytorchmergebot reopened this Oct 17, 2025

pytorch-auto-revert bot mentioned this pull request Oct 17, 2025

[DO NOT CLOSE] Autorevert actions shadow mode stream #163650

Open

SherlockNoMad added a commit that referenced this pull request Oct 27, 2025

ghstack Update gm.print_readable to include Annotation

414f38e

ghstack-source-id: d9191e9 Pull Request resolved: #165397

SherlockNoMad closed this Oct 27, 2025

SherlockNoMad reopened this Oct 27, 2025

SherlockNoMad added a commit that referenced this pull request Oct 27, 2025

ghstack Update gm.print_readable to include Annotation

420bf63

ghstack-source-id: b5f027a Pull Request resolved: #165397

SherlockNoMad added a commit that referenced this pull request Oct 28, 2025

ghstack Update gm.print_readable to include Annotation

a49daf5

ghstack-source-id: 160b58c Pull Request resolved: #165397

mlazos approved these changes Oct 28, 2025

View reviewed changes

pytorchmergebot added the merging label Oct 28, 2025

pytorchmergebot closed this in 34d6ef7 Oct 28, 2025

pytorchmergebot removed the merging label Oct 28, 2025

This was referenced Oct 29, 2025

Fix space in FX graph stacktrace summary for better backward compat #166590

Closed

[CI] Fix debug_str() to be compatible with latest PyTorch nightly pytorch/helion#1050

Merged

github-actions bot deleted the gh/SherlockNoMad/14/head branch November 29, 2025 02:18

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Update gm.print_readable to include Annotation#165397

Update gm.print_readable to include Annotation#165397
SherlockNoMad wants to merge 15 commits intogh/SherlockNoMad/14/basefrom
gh/SherlockNoMad/14/head

SherlockNoMad commented Oct 14, 2025 •

edited by pytorch-bot bot

Loading

Uh oh!

pytorch-bot bot commented Oct 14, 2025 •

edited

Loading

Uh oh!

malfet commented Oct 17, 2025

Uh oh!

pytorchmergebot commented Oct 17, 2025

Uh oh!

pytorchmergebot commented Oct 17, 2025

Uh oh!

SherlockNoMad commented Oct 28, 2025

Uh oh!

pytorchmergebot commented Oct 28, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants

Conversation

SherlockNoMad commented Oct 14, 2025 • edited by pytorch-bot bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pytorch-bot bot commented Oct 14, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/165397

✅ No Failures

Uh oh!

malfet commented Oct 17, 2025

Uh oh!

pytorchmergebot commented Oct 17, 2025

Uh oh!

pytorchmergebot commented Oct 17, 2025

Uh oh!

SherlockNoMad commented Oct 28, 2025

Uh oh!

pytorchmergebot commented Oct 28, 2025

Merge started

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants

SherlockNoMad commented Oct 14, 2025 •

edited by pytorch-bot bot

Loading

pytorch-bot bot commented Oct 14, 2025 •

edited

Loading