OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-07 12:21:27 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity
88,878 Commits 8,304 Branches 150 Tags 1.5 GiB
87b002b6fb
Commit Graph

53 Commits

Author SHA1 Message Date
PyTorch MergeBot
eb152ab1dd Revert "Inductor logging + analysis of torch.profile (#149697)" 
This reverts commit 060838c231.

Reverted https://github.com/pytorch/pytorch/pull/149697 on behalf of https://github.com/clee2000 due to broke a bunch of tests internally D76299454, probably also broke rocm inductor/test_analysis.py::TestAnalysisCUDA::test_augment_trace_against_flop_counter_maxat0_cuda_float16 [GH job link](https://github.com/pytorch/pytorch/actions/runs/15545277599/job/43766911025) [HUD commit link](060838c231) ([comment](https://github.com/pytorch/pytorch/pull/149697#issuecomment-2959747153))
 2025-06-10 15:38:40 +00:00
Gabriel Ferns
060838c231 Inductor logging + analysis of torch.profile (#149697) 
Prereqs:
 - https://github.com/pytorch/pytorch/pull/152708

Features:
1. Adds inductor's estimate of flops and bandwidth to the json trace events that perfetto uses.
1. Only use the tflops estimation from triton if we don't have the info from the datasheet because Triton's estimates are inaccurate. I have a backlog item to fix triton flops estimation upstream. New `DeviceInfo` class, and new function `get_device_tflops`.
1. New helpers `countable_fx` and `count_flops_fx` helps get the flops of an `fx.Node`.
1. Extends Triton `torch.profiler` logging to `DebugAutotuner`.
1. New script `profile_analysis.py`: `--augment_trace` adds perf estimates to any perfetto json trace, `--analyze` creates a summary table of these perf estimates, and `--diff` will compare two traces side by side:
```python
Device(NVIDIA H100, 0):
 Kernel Name                              | resnet Kernel Count | resnet FLOPS       | resnet bw gbps        | resnet Dur (ms)    | resnet Achieved FLOPS % | resnet Achieved Bandwidth % | newresnet Kernel Count | newresnet FLOPS    | newresnet bw gbps     | newresnet Dur (ms) | newresnet Achieved FLOPS % | newresnet Achieved Bandwidth %
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
 triton_poi_fused__native_batch_norm_legi | 24                  | 0                  | 0.11395268248131513   | 2.5919166666666666 | 0                       | 0.003401572611382541        | 24                     | 0                  | 0.11395268248131513   | 2.5919166666666666 | 0                          | 0.003401572611382541
 sm90_xmma_fprop_implicit_gemm_f32f32_tf3 | 142                 | 16932673552.422373 | 0.2585007824198784    | 12.441619718309857 | 0.08683422334575583     | 0.007716441266265022        | 142                    | 16932673552.422373 | 0.2585007824198784    | 12.441619718309857 | 0.08683422334575583        | 0.007716441266265022
 triton_red_fused__native_batch_norm_legi | 39                  | 0                  | 0.13990024992108846   | 5.752589743589743  | 0                       | 0.004176126863316074        | 39                     | 0                  | 0.13990024992108846   | 5.752589743589743  | 0                          | 0.004176126863316074
 triton_poi_fused__native_batch_norm_legi | 25                  | 0                  | 0.31824055917536503   | 2.5291999999999994 | 0                       | 0.009499718184339253        | 25                     | 0                  | 0.31824055917536503   | 2.5291999999999994 | 0                          | 0.009499718184339253
 void cutlass::Kernel2<cutlass_80_tensoro | 98                  | 16211056473.596165 | 0.42972434051025826   | 7.130408163265306  | 0.08313362294151874     | 0.012827592254037562        | 98                     | 16211056473.596165 | 0.42972434051025826   | 7.130408163265306  | 0.08313362294151874        | 0.012827592254037562
 triton_red_fused__native_batch_norm_legi | 73                  | 0                  | 0.3225381327611705    | 9.987068493150682  | 0                       | 0.009628003963020014        | 73                     | 0                  | 0.3225381327611705    | 9.987068493150682  | 0                          | 0.009628003963020014
 triton_poi_fused__native_batch_norm_legi | 15                  | 0                  | 1.4491211346487216    | 4.439333333333333  | 0                       | 0.043257347302946926        | 15                     | 0                  | 1.4491211346487216    | 4.439333333333333  | 0                          | 0.043257347302946926
 void cutlass::Kernel2<cutlass_80_tensoro | 186                 | 14501701145.337954 | 0.2667131401910989    | 7.873865591397849  | 0.07436769818122027     | 0.007961586274361157        | 186                    | 14501701145.337954 | 0.2667131401910989    | 7.873865591397849  | 0.07436769818122027        | 0.007961586274361157
 triton_poi_fused__native_batch_norm_legi | 33                  | 0                  | 1.4924556538193923    | 4.3101515151515155 | 0                       | 0.044550915039384846        | 33                     | 0                  | 1.4924556538193923    | 4.3101515151515155 | 0                          | 0.044550915039384846
 triton_red_fused__native_batch_norm_legi | 29                  | 0                  | 0.25562590522631107   | 6.296275862068965  | 0                       | 0.007630624036606301        | 29                     | 0                  | 0.25562590522631107   | 6.296275862068965  | 0                          | 0.007630624036606301
 triton_poi_fused__native_batch_norm_legi | 13                  | 0                  | 0.5870562174192726    | 2.7397692307692307 | 0                       | 0.01752406619162008         | 13                     | 0                  | 0.5870562174192726    | 2.7397692307692307 | 0                          | 0.01752406619162008
 triton_poi_fused__native_batch_norm_legi | 34                  | 0                  | 0.41409928846284      | 2.853588235294117  | 0                       | 0.012361172789935523        | 34                     | 0                  | 0.41409928846284      | 2.853588235294117  | 0                          | 0.012361172789935523
 triton_per_fused__native_batch_norm_legi | 34                  | 0                  | 0.11705315007018151   | 3.460647058823529  | 0                       | 0.0034941238826919864       | 34                     | 0                  | 0.11705315007018151   | 3.460647058823529  | 0                          | 0.0034941238826919864
 triton_poi_fused__native_batch_norm_legi | 16                  | 0                  | 0.17207853197124584   | 2.3459375000000002 | 0                       | 0.005136672596156592        | 16                     | 0                  | 0.17207853197124584   | 2.3459375000000002 | 0                          | 0.005136672596156592
 triton_per_fused__native_batch_norm_legi | 30                  | 0                  | 0.2639714322022256    | 6.131199999999999  | 0                       | 0.007879744244842555        | 30                     | 0                  | 0.2639714322022256    | 6.131199999999999  | 0                          | 0.007879744244842555
 sm90_xmma_fprop_implicit_gemm_f32f32_tf3 | 100                 | 11875430356.891787 | 0.19494470869421385   | 16.36534           | 0.06089964285585531     | 0.005819245035648175        | 100                    | 11875430356.891787 | 0.19494470869421385   | 16.36534           | 0.06089964285585531        | 0.005819245035648175
 triton_poi_fused__native_batch_norm_legi | 8                   | 0                  | 0.9854096626224687    | 3.2757500000000004 | 0                       | 0.029415213809625928        | 8                      | 0                  | 0.9854096626224687    | 3.2757500000000004 | 0                          | 0.029415213809625928
 void cublasLt::splitKreduce_kernel<32, 1 | 56                  | 34377923395.147064 | 0.8310300045762317    | 3.4199999999999986 | 0.17629704305203628     | 0.024806865808245714        | 56                     | 34377923395.147064 | 0.8310300045762317    | 3.4199999999999986 | 0.17629704305203628        | 0.024806865808245714
 triton_poi_fused__native_batch_norm_legi | 23                  | 0                  | 0.9944002965861103    | 3.2431304347826084 | 0                       | 0.02968359094286896         | 23                     | 0                  | 0.9944002965861103    | 3.2431304347826084 | 0                          | 0.02968359094286896
 triton_per_fused__native_batch_norm_legi | 10                  | 0                  | 0.1826801058931057    | 4.428800000000001  | 0                       | 0.00545313748934644         | 10                     | 0                  | 0.1826801058931057    | 4.428800000000001  | 0                          | 0.00545313748934644
 triton_poi_fused__native_batch_norm_legi | 10                  | 0                  | 0.3168973585366449    | 2.5471999999999997 | 0                       | 0.009459622642884923        | 10                     | 0                  | 0.3168973585366449    | 2.5471999999999997 | 0                          | 0.009459622642884923
 triton_poi_fused__native_batch_norm_legi | 34                  | 0                  | 1.1463614897015777    | 4.124323529411764  | 0                       | 0.03421974596124114         | 34                     | 0                  | 1.1463614897015777    | 4.124323529411764  | 0                          | 0.03421974596124114
 void cask_plugin_cudnn::xmma_cudnn::init | 44                  | 44045510816.64277  | 2.0661232850348643    | 3.6887499999999993 | 0.22587441444432194     | 0.06167532194133924         | 44                     | 44045510816.64277  | 2.0661232850348643    | 3.6887499999999993 | 0.22587441444432194        | 0.06167532194133924
 sm90_xmma_fprop_implicit_gemm_f32f32_tf3 | 95                  | 7876855400.165316  | 0.4694941555946739    | 18.224315789473682 | 0.04039413025725802     | 0.014014750913273854        | 95                     | 7876855400.165316  | 0.4694941555946739    | 18.224315789473682 | 0.04039413025725802        | 0.014014750913273854
 triton_per_fused__native_batch_norm_legi | 41                  | 0                  | 0.06825669875995298   | 3.0384146341463416 | 0                       | 0.002037513395819492        | 41                     | 0                  | 0.06825669875995298   | 3.0384146341463416 | 0                          | 0.002037513395819492
 triton_poi_fused__native_batch_norm_legi | 23                  | 0                  | 0.08808154712430301   | 2.3275652173913044 | 0                       | 0.0026292999141582997       | 23                     | 0                  | 0.08808154712430301   | 2.3275652173913044 | 0                          | 0.0026292999141582997
 triton_per_fused__native_batch_norm_legi | 40                  | 0                  | 0.18179321034952417   | 4.556825           | 0                       | 0.005426662995508183        | 40                     | 0                  | 0.18179321034952417   | 4.556825           | 0                          | 0.005426662995508183
 triton_poi_fused__native_batch_norm_legi | 15                  | 0                  | 0.5887415155454232    | 2.783866666666667  | 0                       | 0.017574373598370836        | 15                     | 0                  | 0.5887415155454232    | 2.783866666666667  | 0                          | 0.017574373598370836
 void cutlass::Kernel2<cutlass_80_tensoro | 38                  | 14242013806.264643 | 0.256592404353939     | 7.217631578947369  | 0.0730359682372546      | 0.007659474756834           | 38                     | 14242013806.264643 | 0.256592404353939     | 7.217631578947369  | 0.0730359682372546         | 0.007659474756834
 triton_poi_fused__native_batch_norm_legi | 21                  | 0                  | 0.5842860973430516    | 2.7779047619047623 | 0                       | 0.017441376040091088        | 21                     | 0                  | 0.5842860973430516    | 2.7779047619047623 | 0                          | 0.017441376040091088
 triton_per_fused__native_batch_norm_legi | 16                  | 0                  | 0.11509365173486417   | 3.5959375000000002 | 0                       | 0.0034356313950705724       | 16                     | 0                  | 0.11509365173486417   | 3.5959375000000002 | 0                          | 0.0034356313950705724
 triton_poi_fused__native_batch_norm_legi | 14                  | 0                  | 0.1704672000243914    | 2.4044285714285714 | 0                       | 0.00508857313505646         | 14                     | 0                  | 0.1704672000243914    | 2.4044285714285714 | 0                          | 0.00508857313505646
 triton_poi_fused__native_batch_norm_legi | 58                  | 0                  | 2.307520779930795     | 8.190706896551722  | 0                       | 0.06888121731136704         | 58                     | 0                  | 2.307520779930795     | 8.190706896551722  | 0                          | 0.06888121731136704
 triton_per_fused__native_batch_norm_legi | 29                  | 0                  | 0.037243248971881276  | 3.0277586206896556 | 0                       | 0.001111738775280038        | 29                     | 0                  | 0.037243248971881276  | 3.0277586206896556 | 0                          | 0.001111738775280038
 triton_poi_fused__native_batch_norm_legi | 20                  | 0                  | 0.04741699795428918   | 2.2911500000000005 | 0                       | 0.0014154327747549007       | 20                     | 0                  | 0.04741699795428918   | 2.2911500000000005 | 0                          | 0.0014154327747549007
 triton_per_fused__native_batch_norm_legi | 25                  | 0                  | 0.13357016893727824   | 3.37536            | 0                       | 0.003987169222008305        | 25                     | 0                  | 0.13357016893727824   | 3.37536            | 0                          | 0.003987169222008305
 triton_poi_fused__native_batch_norm_legi | 13                  | 0                  | 0.3089862268300253    | 2.8111538461538457 | 0                       | 0.009223469457612694        | 13                     | 0                  | 0.3089862268300253    | 2.8111538461538457 | 0                          | 0.009223469457612694
 triton_poi_fused__native_batch_norm_legi | 17                  | 0                  | 0.3129385387909844    | 2.673              | 0                       | 0.009341448919133863        | 17                     | 0                  | 0.3129385387909844    | 2.673              | 0                          | 0.009341448919133863
 triton_per_fused__native_batch_norm_legi | 19                  | 0                  | 0.2215568162533158    | 3.8837368421052636 | 0                       | 0.0066136363060691275       | 19                     | 0                  | 0.2215568162533158    | 3.8837368421052636 | 0                          | 0.0066136363060691275
 std::enable_if<!(false), void>::type int | 23                  | 504916805.19297093 | 1.0118296096314707    | 8.113913043478261  | 0.0025893169497075447   | 0.030203868944223014        | 23                     | 504916805.19297093 | 1.0118296096314707    | 8.113913043478261  | 0.0025893169497075447      | 0.030203868944223014
 triton_poi_fused_add_copy__38            | 56                  | 0                  | 0                     | 2.132482142857143  | 0                       | 0                           | 56                     | 0                  | 0                     | 2.132482142857143  | 0                          | 0
 triton_poi_fused_convolution_0           | 18                  | 0                  | 0.43458610794936897   | 2.773333333333334  | 0                       | 0.012972719640279667        | 18                     | 0                  | 0.43458610794936897   | 2.773333333333334  | 0                          | 0.012972719640279667
 triton_poi_fused_convolution_1           | 17                  | 0                  | 0.028816312469162712  | 2.6145882352941174 | 0                       | 0.0008601884319153051       | 17                     | 0                  | 0.028816312469162712  | 2.6145882352941174 | 0                          | 0.0008601884319153051
 void convolve_common_engine_float_NHWC<f | 44                  | 8641868995.31118   | 0.024730540008465626  | 25.87327272727273  | 0.04431727689903169     | 0.0007382250748795709       | 44                     | 8641868995.31118   | 0.024730540008465626  | 25.87327272727273  | 0.04431727689903169        | 0.0007382250748795709
 triton_per_fused__native_batch_norm_legi | 12                  | 0                  | 0.6809930918986744    | 4.82675            | 0                       | 0.020328151996975356        | 12                     | 0                  | 0.6809930918986744    | 4.82675            | 0                          | 0.020328151996975356
 triton_per_fused__native_batch_norm_legi | 14                  | 0                  | 0.02883030597936608   | 2.6651428571428575 | 0                       | 0.0008606061486377935       | 14                     | 0                  | 0.02883030597936608   | 2.6651428571428575 | 0                          | 0.0008606061486377935
 triton_per_fused__native_batch_norm_legi | 16                  | 0                  | 0.0014658988233201874 | 2.098              | 0                       | 4.375817383045335e-05       | 16                     | 0                  | 0.0014658988233201874 | 2.098              | 0                          | 4.375817383045335e-05
 triton_poi_fused__native_batch_norm_legi | 13                  | 0                  | 0.9926297180284697    | 3.2367692307692306 | 0                       | 0.02963073785159611         | 13                     | 0                  | 0.9926297180284697    | 3.2367692307692306 | 0                          | 0.02963073785159611
 triton_poi_fused__native_batch_norm_legi | 9                   | 0                  | 1.3008817095666507    | 3.0863333333333336 | 0                       | 0.03883228983781048         | 9                      | 0                  | 1.3008817095666507    | 3.0863333333333336 | 0                          | 0.03883228983781048
 void at::native::(anonymous namespace):: | 98                  | 0                  | 0.09174335613709389   | 4.408520408163265  | 0                       | 0.0027386076458833994       | 98                     | 0                  | 0.09174335613709389   | 4.408520408163265  | 0                          | 0.0027386076458833994
 void at::native::vectorized_elementwise_ | 7                   | 0                  | 0                     | 1.7278571428571428 | 0                       | 0                           | 7                      | 0                  | 0                     | 1.7278571428571428 | 0                          | 0
```

Pull Request resolved: https://github.com/pytorch/pytorch/pull/149697
Approved by: https://github.com/eellison, https://github.com/shunting314
 2025-06-09 21:43:21 +00:00
PyTorch MergeBot
5e03433443 Revert "Inductor logging + analysis of torch.profile (#149697)" 
This reverts commit e5afbe3124.

Reverted https://github.com/pytorch/pytorch/pull/149697 on behalf of https://github.com/malfet due to Broke rocm, see 642687af29/1 ([comment](https://github.com/pytorch/pytorch/pull/149697#issuecomment-2942415600))
 2025-06-05 01:38:13 +00:00
Gabriel Ferns
e5afbe3124 Inductor logging + analysis of torch.profile (#149697) 
Prereqs:
 - https://github.com/pytorch/pytorch/pull/152708

Features:
1. Adds inductor's estimate of flops and bandwidth to the json trace events that perfetto uses.
1. Only use the tflops estimation from triton if we don't have the info from the datasheet because Triton's estimates are inaccurate. I have a backlog item to fix triton flops estimation upstream. New `DeviceInfo` class, and new function `get_device_tflops`.
1. New helpers `countable_fx` and `count_flops_fx` helps get the flops of an `fx.Node`.
1. Extends Triton `torch.profiler` logging to `DebugAutotuner`.
1. New script `profile_analysis.py`: `--augment_trace` adds perf estimates to any perfetto json trace, `--analyze` creates a summary table of these perf estimates, and `--diff` will compare two traces side by side:
```python
Device(NVIDIA H100, 0):
 Kernel Name                              | resnet Kernel Count | resnet FLOPS       | resnet bw gbps        | resnet Dur (ms)    | resnet Achieved FLOPS % | resnet Achieved Bandwidth % | newresnet Kernel Count | newresnet FLOPS    | newresnet bw gbps     | newresnet Dur (ms) | newresnet Achieved FLOPS % | newresnet Achieved Bandwidth %
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
 triton_poi_fused__native_batch_norm_legi | 24                  | 0                  | 0.11395268248131513   | 2.5919166666666666 | 0                       | 0.003401572611382541        | 24                     | 0                  | 0.11395268248131513   | 2.5919166666666666 | 0                          | 0.003401572611382541
 sm90_xmma_fprop_implicit_gemm_f32f32_tf3 | 142                 | 16932673552.422373 | 0.2585007824198784    | 12.441619718309857 | 0.08683422334575583     | 0.007716441266265022        | 142                    | 16932673552.422373 | 0.2585007824198784    | 12.441619718309857 | 0.08683422334575583        | 0.007716441266265022
 triton_red_fused__native_batch_norm_legi | 39                  | 0                  | 0.13990024992108846   | 5.752589743589743  | 0                       | 0.004176126863316074        | 39                     | 0                  | 0.13990024992108846   | 5.752589743589743  | 0                          | 0.004176126863316074
 triton_poi_fused__native_batch_norm_legi | 25                  | 0                  | 0.31824055917536503   | 2.5291999999999994 | 0                       | 0.009499718184339253        | 25                     | 0                  | 0.31824055917536503   | 2.5291999999999994 | 0                          | 0.009499718184339253
 void cutlass::Kernel2<cutlass_80_tensoro | 98                  | 16211056473.596165 | 0.42972434051025826   | 7.130408163265306  | 0.08313362294151874     | 0.012827592254037562        | 98                     | 16211056473.596165 | 0.42972434051025826   | 7.130408163265306  | 0.08313362294151874        | 0.012827592254037562
 triton_red_fused__native_batch_norm_legi | 73                  | 0                  | 0.3225381327611705    | 9.987068493150682  | 0                       | 0.009628003963020014        | 73                     | 0                  | 0.3225381327611705    | 9.987068493150682  | 0                          | 0.009628003963020014
 triton_poi_fused__native_batch_norm_legi | 15                  | 0                  | 1.4491211346487216    | 4.439333333333333  | 0                       | 0.043257347302946926        | 15                     | 0                  | 1.4491211346487216    | 4.439333333333333  | 0                          | 0.043257347302946926
 void cutlass::Kernel2<cutlass_80_tensoro | 186                 | 14501701145.337954 | 0.2667131401910989    | 7.873865591397849  | 0.07436769818122027     | 0.007961586274361157        | 186                    | 14501701145.337954 | 0.2667131401910989    | 7.873865591397849  | 0.07436769818122027        | 0.007961586274361157
 triton_poi_fused__native_batch_norm_legi | 33                  | 0                  | 1.4924556538193923    | 4.3101515151515155 | 0                       | 0.044550915039384846        | 33                     | 0                  | 1.4924556538193923    | 4.3101515151515155 | 0                          | 0.044550915039384846
 triton_red_fused__native_batch_norm_legi | 29                  | 0                  | 0.25562590522631107   | 6.296275862068965  | 0                       | 0.007630624036606301        | 29                     | 0                  | 0.25562590522631107   | 6.296275862068965  | 0                          | 0.007630624036606301
 triton_poi_fused__native_batch_norm_legi | 13                  | 0                  | 0.5870562174192726    | 2.7397692307692307 | 0                       | 0.01752406619162008         | 13                     | 0                  | 0.5870562174192726    | 2.7397692307692307 | 0                          | 0.01752406619162008
 triton_poi_fused__native_batch_norm_legi | 34                  | 0                  | 0.41409928846284      | 2.853588235294117  | 0                       | 0.012361172789935523        | 34                     | 0                  | 0.41409928846284      | 2.853588235294117  | 0                          | 0.012361172789935523
 triton_per_fused__native_batch_norm_legi | 34                  | 0                  | 0.11705315007018151   | 3.460647058823529  | 0                       | 0.0034941238826919864       | 34                     | 0                  | 0.11705315007018151   | 3.460647058823529  | 0                          | 0.0034941238826919864
 triton_poi_fused__native_batch_norm_legi | 16                  | 0                  | 0.17207853197124584   | 2.3459375000000002 | 0                       | 0.005136672596156592        | 16                     | 0                  | 0.17207853197124584   | 2.3459375000000002 | 0                          | 0.005136672596156592
 triton_per_fused__native_batch_norm_legi | 30                  | 0                  | 0.2639714322022256    | 6.131199999999999  | 0                       | 0.007879744244842555        | 30                     | 0                  | 0.2639714322022256    | 6.131199999999999  | 0                          | 0.007879744244842555
 sm90_xmma_fprop_implicit_gemm_f32f32_tf3 | 100                 | 11875430356.891787 | 0.19494470869421385   | 16.36534           | 0.06089964285585531     | 0.005819245035648175        | 100                    | 11875430356.891787 | 0.19494470869421385   | 16.36534           | 0.06089964285585531        | 0.005819245035648175
 triton_poi_fused__native_batch_norm_legi | 8                   | 0                  | 0.9854096626224687    | 3.2757500000000004 | 0                       | 0.029415213809625928        | 8                      | 0                  | 0.9854096626224687    | 3.2757500000000004 | 0                          | 0.029415213809625928
 void cublasLt::splitKreduce_kernel<32, 1 | 56                  | 34377923395.147064 | 0.8310300045762317    | 3.4199999999999986 | 0.17629704305203628     | 0.024806865808245714        | 56                     | 34377923395.147064 | 0.8310300045762317    | 3.4199999999999986 | 0.17629704305203628        | 0.024806865808245714
 triton_poi_fused__native_batch_norm_legi | 23                  | 0                  | 0.9944002965861103    | 3.2431304347826084 | 0                       | 0.02968359094286896         | 23                     | 0                  | 0.9944002965861103    | 3.2431304347826084 | 0                          | 0.02968359094286896
 triton_per_fused__native_batch_norm_legi | 10                  | 0                  | 0.1826801058931057    | 4.428800000000001  | 0                       | 0.00545313748934644         | 10                     | 0                  | 0.1826801058931057    | 4.428800000000001  | 0                          | 0.00545313748934644
 triton_poi_fused__native_batch_norm_legi | 10                  | 0                  | 0.3168973585366449    | 2.5471999999999997 | 0                       | 0.009459622642884923        | 10                     | 0                  | 0.3168973585366449    | 2.5471999999999997 | 0                          | 0.009459622642884923
 triton_poi_fused__native_batch_norm_legi | 34                  | 0                  | 1.1463614897015777    | 4.124323529411764  | 0                       | 0.03421974596124114         | 34                     | 0                  | 1.1463614897015777    | 4.124323529411764  | 0                          | 0.03421974596124114
 void cask_plugin_cudnn::xmma_cudnn::init | 44                  | 44045510816.64277  | 2.0661232850348643    | 3.6887499999999993 | 0.22587441444432194     | 0.06167532194133924         | 44                     | 44045510816.64277  | 2.0661232850348643    | 3.6887499999999993 | 0.22587441444432194        | 0.06167532194133924
 sm90_xmma_fprop_implicit_gemm_f32f32_tf3 | 95                  | 7876855400.165316  | 0.4694941555946739    | 18.224315789473682 | 0.04039413025725802     | 0.014014750913273854        | 95                     | 7876855400.165316  | 0.4694941555946739    | 18.224315789473682 | 0.04039413025725802        | 0.014014750913273854
 triton_per_fused__native_batch_norm_legi | 41                  | 0                  | 0.06825669875995298   | 3.0384146341463416 | 0                       | 0.002037513395819492        | 41                     | 0                  | 0.06825669875995298   | 3.0384146341463416 | 0                          | 0.002037513395819492
 triton_poi_fused__native_batch_norm_legi | 23                  | 0                  | 0.08808154712430301   | 2.3275652173913044 | 0                       | 0.0026292999141582997       | 23                     | 0                  | 0.08808154712430301   | 2.3275652173913044 | 0                          | 0.0026292999141582997
 triton_per_fused__native_batch_norm_legi | 40                  | 0                  | 0.18179321034952417   | 4.556825           | 0                       | 0.005426662995508183        | 40                     | 0                  | 0.18179321034952417   | 4.556825           | 0                          | 0.005426662995508183
 triton_poi_fused__native_batch_norm_legi | 15                  | 0                  | 0.5887415155454232    | 2.783866666666667  | 0                       | 0.017574373598370836        | 15                     | 0                  | 0.5887415155454232    | 2.783866666666667  | 0                          | 0.017574373598370836
 void cutlass::Kernel2<cutlass_80_tensoro | 38                  | 14242013806.264643 | 0.256592404353939     | 7.217631578947369  | 0.0730359682372546      | 0.007659474756834           | 38                     | 14242013806.264643 | 0.256592404353939     | 7.217631578947369  | 0.0730359682372546         | 0.007659474756834
 triton_poi_fused__native_batch_norm_legi | 21                  | 0                  | 0.5842860973430516    | 2.7779047619047623 | 0                       | 0.017441376040091088        | 21                     | 0                  | 0.5842860973430516    | 2.7779047619047623 | 0                          | 0.017441376040091088
 triton_per_fused__native_batch_norm_legi | 16                  | 0                  | 0.11509365173486417   | 3.5959375000000002 | 0                       | 0.0034356313950705724       | 16                     | 0                  | 0.11509365173486417   | 3.5959375000000002 | 0                          | 0.0034356313950705724
 triton_poi_fused__native_batch_norm_legi | 14                  | 0                  | 0.1704672000243914    | 2.4044285714285714 | 0                       | 0.00508857313505646         | 14                     | 0                  | 0.1704672000243914    | 2.4044285714285714 | 0                          | 0.00508857313505646
 triton_poi_fused__native_batch_norm_legi | 58                  | 0                  | 2.307520779930795     | 8.190706896551722  | 0                       | 0.06888121731136704         | 58                     | 0                  | 2.307520779930795     | 8.190706896551722  | 0                          | 0.06888121731136704
 triton_per_fused__native_batch_norm_legi | 29                  | 0                  | 0.037243248971881276  | 3.0277586206896556 | 0                       | 0.001111738775280038        | 29                     | 0                  | 0.037243248971881276  | 3.0277586206896556 | 0                          | 0.001111738775280038
 triton_poi_fused__native_batch_norm_legi | 20                  | 0                  | 0.04741699795428918   | 2.2911500000000005 | 0                       | 0.0014154327747549007       | 20                     | 0                  | 0.04741699795428918   | 2.2911500000000005 | 0                          | 0.0014154327747549007
 triton_per_fused__native_batch_norm_legi | 25                  | 0                  | 0.13357016893727824   | 3.37536            | 0                       | 0.003987169222008305        | 25                     | 0                  | 0.13357016893727824   | 3.37536            | 0                          | 0.003987169222008305
 triton_poi_fused__native_batch_norm_legi | 13                  | 0                  | 0.3089862268300253    | 2.8111538461538457 | 0                       | 0.009223469457612694        | 13                     | 0                  | 0.3089862268300253    | 2.8111538461538457 | 0                          | 0.009223469457612694
 triton_poi_fused__native_batch_norm_legi | 17                  | 0                  | 0.3129385387909844    | 2.673              | 0                       | 0.009341448919133863        | 17                     | 0                  | 0.3129385387909844    | 2.673              | 0                          | 0.009341448919133863
 triton_per_fused__native_batch_norm_legi | 19                  | 0                  | 0.2215568162533158    | 3.8837368421052636 | 0                       | 0.0066136363060691275       | 19                     | 0                  | 0.2215568162533158    | 3.8837368421052636 | 0                          | 0.0066136363060691275
 std::enable_if<!(false), void>::type int | 23                  | 504916805.19297093 | 1.0118296096314707    | 8.113913043478261  | 0.0025893169497075447   | 0.030203868944223014        | 23                     | 504916805.19297093 | 1.0118296096314707    | 8.113913043478261  | 0.0025893169497075447      | 0.030203868944223014
 triton_poi_fused_add_copy__38            | 56                  | 0                  | 0                     | 2.132482142857143  | 0                       | 0                           | 56                     | 0                  | 0                     | 2.132482142857143  | 0                          | 0
 triton_poi_fused_convolution_0           | 18                  | 0                  | 0.43458610794936897   | 2.773333333333334  | 0                       | 0.012972719640279667        | 18                     | 0                  | 0.43458610794936897   | 2.773333333333334  | 0                          | 0.012972719640279667
 triton_poi_fused_convolution_1           | 17                  | 0                  | 0.028816312469162712  | 2.6145882352941174 | 0                       | 0.0008601884319153051       | 17                     | 0                  | 0.028816312469162712  | 2.6145882352941174 | 0                          | 0.0008601884319153051
 void convolve_common_engine_float_NHWC<f | 44                  | 8641868995.31118   | 0.024730540008465626  | 25.87327272727273  | 0.04431727689903169     | 0.0007382250748795709       | 44                     | 8641868995.31118   | 0.024730540008465626  | 25.87327272727273  | 0.04431727689903169        | 0.0007382250748795709
 triton_per_fused__native_batch_norm_legi | 12                  | 0                  | 0.6809930918986744    | 4.82675            | 0                       | 0.020328151996975356        | 12                     | 0                  | 0.6809930918986744    | 4.82675            | 0                          | 0.020328151996975356
 triton_per_fused__native_batch_norm_legi | 14                  | 0                  | 0.02883030597936608   | 2.6651428571428575 | 0                       | 0.0008606061486377935       | 14                     | 0                  | 0.02883030597936608   | 2.6651428571428575 | 0                          | 0.0008606061486377935
 triton_per_fused__native_batch_norm_legi | 16                  | 0                  | 0.0014658988233201874 | 2.098              | 0                       | 4.375817383045335e-05       | 16                     | 0                  | 0.0014658988233201874 | 2.098              | 0                          | 4.375817383045335e-05
 triton_poi_fused__native_batch_norm_legi | 13                  | 0                  | 0.9926297180284697    | 3.2367692307692306 | 0                       | 0.02963073785159611         | 13                     | 0                  | 0.9926297180284697    | 3.2367692307692306 | 0                          | 0.02963073785159611
 triton_poi_fused__native_batch_norm_legi | 9                   | 0                  | 1.3008817095666507    | 3.0863333333333336 | 0                       | 0.03883228983781048         | 9                      | 0                  | 1.3008817095666507    | 3.0863333333333336 | 0                          | 0.03883228983781048
 void at::native::(anonymous namespace):: | 98                  | 0                  | 0.09174335613709389   | 4.408520408163265  | 0                       | 0.0027386076458833994       | 98                     | 0                  | 0.09174335613709389   | 4.408520408163265  | 0                          | 0.0027386076458833994
 void at::native::vectorized_elementwise_ | 7                   | 0                  | 0                     | 1.7278571428571428 | 0                       | 0                           | 7                      | 0                  | 0                     | 1.7278571428571428 | 0                          | 0
```

Pull Request resolved: https://github.com/pytorch/pytorch/pull/149697
Approved by: https://github.com/eellison, https://github.com/shunting314
 2025-06-04 20:03:46 +00:00
David Berard
43496e9b90 [NJT] fix flop counter for SDPA & test (#147032) 
Fixes 3 issues:
1. The test wasn't actually testing SDPA: both were checking cuda, and the inputs to SDPA were not transposed.
2. FlopCounterMode has been renamed _FlopCounterMode (and a wrapper named FlopCounterMode has been added)
3. offsets_to_list also needs to ignore the actual offset values if offsets is a meta tensor.

Differential Revision: [D69558785](https://our.internmc.facebook.com/intern/diff/D69558785)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/147032
Approved by: https://github.com/jbschlosser
 2025-02-13 07:14:58 +00:00
Aaron Orenstein
2f9d378f7b PEP585 update - torch/utils (#145201) 
See #145101 for details.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/145201
Approved by: https://github.com/bobrenjc93
 2025-01-21 21:04:10 +00:00
Luca Wehrstedt
a0d2c09115 Add flop formula for _scaled_mm (#144973) 
This will make it work correctly with the partitioner's AutoAC
Pull Request resolved: https://github.com/pytorch/pytorch/pull/144973
Approved by: https://github.com/jeffdaily
 2025-01-17 09:38:30 +00:00
PyTorch MergeBot
6559374494 Revert "Add flop formula for _scaled_mm (#144872)" 
This reverts commit f31452268b.

Reverted https://github.com/pytorch/pytorch/pull/144872 on behalf of https://github.com/lw due to Breaks ROCm jobs on main ([comment](https://github.com/pytorch/pytorch/pull/144872#issuecomment-2595994134))
 2025-01-16 15:16:18 +00:00
Luca Wehrstedt
f31452268b Add flop formula for _scaled_mm (#144872) 
This will make it work correctly with the partitioner's AutoAC
Pull Request resolved: https://github.com/pytorch/pytorch/pull/144872
Approved by: https://github.com/vkuzo
 2025-01-16 13:57:54 +00:00
Aaron Orenstein
45ef3309e3 [BE] typing for decorators (#144161) 
Summary:
Untyped decorators strip annotations from the decorated items.

- _compile
- _inductor/fx_passes/post_grad
- _inductor/lowering
- _library/custom_ops
- _meta_registrations
- _ops
- _refs/nn/functional
- ao/quantization/quantizer/xnnpack_quantizer_utils
- distributed/_composable/contract
- fx/experimental/graph_gradual_typechecker
- fx/experimental/migrate_gradual_types/constraint_generator
- optim/optimizer
- signal/windows/windows
- testing/_internal/common_device_type
- torch/_inductor/decomposition
- utils/flop_counter

Test Plan: unit tests

Differential Revision: D62302684

Pull Request resolved: https://github.com/pytorch/pytorch/pull/144161
Approved by: https://github.com/Skylion007, https://github.com/albanD
 2025-01-04 16:40:09 +00:00
Florian (Feuermagier)
4fa72168ea FlopCounterMode: Decompose ops for inference mode (#138508) 
Fixes #126268

I've basically followed @ezyang suggestion (I think) to use `func.decompose(...)`. Since `__torch_dispatch__` won't be called a second time for the same op, I've added a second `TorchDispatchMode` (`_DecomposedCounterMode`) that simpy dispatches to the parent flop counter. Using `self` as the inner context manager is not possible, since the second call to `__enter__` would re-initialize the counter's tracking state.

Let me know if there's something wrong with this implementation, since I'm quite unsure how the decomposition thing actually works :D

Pull Request resolved: https://github.com/pytorch/pytorch/pull/138508
Approved by: https://github.com/ezyang

Co-authored-by: Edward Z. Yang <ezyang@meta.com>
 2024-11-25 16:53:10 +00:00
PyTorch MergeBot
727f1a6da9 Revert "FlopCounterMode: Decompose ops for inference mode (#138508)" 
This reverts commit f915409c26.

Reverted https://github.com/pytorch/pytorch/pull/138508 on behalf of https://github.com/jamesjwu due to Failing internal jobs ([comment](https://github.com/pytorch/pytorch/pull/138508#issuecomment-2484310587))
 2024-11-18 22:59:36 +00:00
Florian (Feuermagier)
f915409c26 FlopCounterMode: Decompose ops for inference mode (#138508) 
Fixes #126268

I've basically followed @ezyang suggestion (I think) to use `func.decompose(...)`. Since `__torch_dispatch__` won't be called a second time for the same op, I've added a second `TorchDispatchMode` (`_DecomposedCounterMode`) that simpy dispatches to the parent flop counter. Using `self` as the inner context manager is not possible, since the second call to `__enter__` would re-initialize the counter's tracking state.

Let me know if there's something wrong with this implementation, since I'm quite unsure how the decomposition thing actually works :D

Pull Request resolved: https://github.com/pytorch/pytorch/pull/138508
Approved by: https://github.com/ezyang
 2024-11-09 03:13:53 +00:00
Tom Ritchford
c0582fd0f8 Remove unused Python variables in torch/[b-z]* (#136963) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/136963
Approved by: https://github.com/ezyang
 2024-10-19 16:45:22 +00:00
Aaron Gokaslan
afaa5fcecb [BE][Ez]: FURB142,FURB92 misc preview fixes (#133880) 
Fixes some miscellaneous code quality issues with some refurb rules that have not been enabled yet.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/133880
Approved by: https://github.com/soulitzer, https://github.com/malfet
 2024-08-21 13:54:51 +00:00
David Berard
1962f9475f [NJT][flop counter] attention: if offsets are fake, use max seqlen (#132356) 
The flop counter is used by the partitioner, in which case the tensors passed in can be fake.

The flop computations for nested attention use the offsets to determine the actual amount of compute that will be done. But when the offsets are fake, we end up with unbacked symints (from `(offsets[1:] - offsets[:-1]).to_list()`). If we find that the offsets are fake or functional tensors, then use the max sequence length instead.

Repro: https://gist.github.com/davidberard98/903fb3e586edb6d1d466786e1a610eba

Differential Revision: [D60597463](https://our.internmc.facebook.com/intern/diff/D60597463)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/132356
Approved by: https://github.com/soulitzer
 2024-08-02 20:42:29 +00:00
Michael Lazos
93979e7063 Skip frame if torch dispatch mode enabled (#131828) 
Fixes https://github.com/pytorch/pytorch/issues/105929

We now skip frames if a dispatch mode is enabled.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/131828
Approved by: https://github.com/bdhirsh, https://github.com/anijain2305
 2024-08-01 19:06:20 +00:00
PyTorch MergeBot
5ced63a005 Revert "[BE] typing for decorators - utils/flop_counter (#131580)" 
This reverts commit 81c26ba5ae.

Reverted https://github.com/pytorch/pytorch/pull/131580 on behalf of https://github.com/clee2000 due to breaking lint internally D60265575 ([comment](https://github.com/pytorch/pytorch/pull/131572#issuecomment-2254328359))
 2024-07-28 03:29:31 +00:00
rzou
a3cdbd8189 [FlopCounterMode] Fix register_flop_formula (#131777) 
Previously, FlopCounterMode would ignore any custom ops registered
through `register_flop_formula`. The problem was:
- register_flop_formula(target) requires target to be an OpOverloadPacket.
- register_flop_formula used register_decomposition to populate its registry
- register_decomposition decomposes the OpOverloadPacket into OpOverload before
  putting it into the registry
- FlopCounterMode ignores OpOverloads in its registry (it assumes the
  registry is a dictionary mapping OpOverloadPacket to flop formula).

register_decomposition is too heavy of a hammer, plus this isn't a
decomposition, so I changed the registration mechanism.

Test Plan:
- new tests
Pull Request resolved: https://github.com/pytorch/pytorch/pull/131777
Approved by: https://github.com/Chillee
 2024-07-26 18:44:50 +00:00
Aaron Orenstein
81c26ba5ae [BE] typing for decorators - utils/flop_counter (#131580) 
See #131429
Pull Request resolved: https://github.com/pytorch/pytorch/pull/131580
Approved by: https://github.com/oulgen, https://github.com/zou3519
ghstack dependencies: #131568, #131569, #131570, #131571, #131572, #131573, #131574, #131575, #131576, #131577, #131578, #131579
 2024-07-26 04:59:58 +00:00
Aaron Orenstein
5a0068cc69 [BE] mypy: disallow untyped decorators (#131428) 
Untyped decorators strip the types from their decorated function so even if the underlying function is fully typed then callers to it don't get any benefit from type annotations.

Step 1 - Enable the error and override in all the offending files.

#131429

Pull Request resolved: https://github.com/pytorch/pytorch/pull/131428
Approved by: https://github.com/justinchuby, https://github.com/oulgen
 2024-07-23 21:50:55 +00:00
eqy
f845a7a91a [cuDNN][SDPA] Remove TORCH_CUDNN_SDPA_ENABLED=1, enable cuDNN SDPA by default on H100 and 2nd on other archs >= sm80 (#125343) 
Looks like one of the first failures seen is `test_causal_variants_compile_causal_variant_CausalVariant_LOWER_RIGHT_shape0_cuda` when `test_causal_variants_causal_variant_CausalVariant_LOWER_RIGHT_shape0_cuda` passes.

What seems interesting here is that the `torch.compile` version fails while the eager version passes. Not sure what the difference would be here...

Nevertheless, is there a recommended mechanism to skip cuDNN SDPA as a backend for this test? CC @drisspg

Pull Request resolved: https://github.com/pytorch/pytorch/pull/125343
Approved by: https://github.com/Skylion007
 2024-06-30 19:22:16 +00:00
PyTorch MergeBot
999eec8dea Revert "[cuDNN][SDPA] Remove TORCH_CUDNN_SDPA_ENABLED=1, enable cuDNN SDPA by default on H100 and 2nd on other archs >= sm80 (#125343)" 
This reverts commit b7e7a4cb01.

Reverted https://github.com/pytorch/pytorch/pull/125343 on behalf of https://github.com/huydhn due to Sorry for reverting your change but it seems to break some test_transformer running on internal A100 and V100 ([comment](https://github.com/pytorch/pytorch/pull/125343#issuecomment-2196202003))
 2024-06-28 06:03:54 +00:00
Eddie Yan
b7e7a4cb01 [cuDNN][SDPA] Remove TORCH_CUDNN_SDPA_ENABLED=1, enable cuDNN SDPA by default on H100 and 2nd on other archs >= sm80 (#125343) 
Looks like one of the first failures seen is `test_causal_variants_compile_causal_variant_CausalVariant_LOWER_RIGHT_shape0_cuda` when `test_causal_variants_causal_variant_CausalVariant_LOWER_RIGHT_shape0_cuda` passes.

What seems interesting here is that the `torch.compile` version fails while the eager version passes. Not sure what the difference would be here...

Nevertheless, is there a recommended mechanism to skip cuDNN SDPA as a backend for this test? CC @drisspg

Pull Request resolved: https://github.com/pytorch/pytorch/pull/125343
Approved by: https://github.com/Skylion007
 2024-06-26 00:49:18 +00:00
PyTorch MergeBot
817ce6835b Revert "[cuDNN][SDPA] Remove TORCH_CUDNN_SDPA_ENABLED=1, enable cuDNN SDPA by default on H100 and 2nd on other archs >= sm80 (#125343)" 
This reverts commit 4c971932e8.

Reverted https://github.com/pytorch/pytorch/pull/125343 on behalf of https://github.com/facebook-github-bot due to Diff reverted internally ([comment](https://github.com/pytorch/pytorch/pull/125343#issuecomment-2163690162))
 2024-06-12 18:47:52 +00:00
eqy
4c971932e8 [cuDNN][SDPA] Remove TORCH_CUDNN_SDPA_ENABLED=1, enable cuDNN SDPA by default on H100 and 2nd on other archs >= sm80 (#125343) 
Looks like one of the first failures seen is `test_causal_variants_compile_causal_variant_CausalVariant_LOWER_RIGHT_shape0_cuda` when `test_causal_variants_causal_variant_CausalVariant_LOWER_RIGHT_shape0_cuda` passes.

What seems interesting here is that the `torch.compile` version fails while the eager version passes. Not sure what the difference would be here...

Nevertheless, is there a recommended mechanism to skip cuDNN SDPA as a backend for this test? CC @drisspg
Pull Request resolved: https://github.com/pytorch/pytorch/pull/125343
Approved by: https://github.com/Skylion007
 2024-06-09 06:53:34 +00:00
Aaron Orenstein
8db9dfa2d7 Flip default value for mypy disallow_untyped_defs [9/11] (#127846) 
See #127836 for details.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/127846
Approved by: https://github.com/ezyang
ghstack dependencies: #127842, #127843, #127844, #127845
 2024-06-08 18:50:06 +00:00
David Berard
9e85d3d830 Add "accurate" FlopCounter implementations for NestedTensor SDPA kernels (#125776) 
This adds implementations for:
* _flash_attention_forward
* _efficient_attention_forward
* _flash_attention_backward
* _efficient_attention_backward

These flop counts are implemented as follows:
* Unbind the batch elements
* Calculate flops individually for each element in the batch
* Sum the final result

This means that we are accessing the concrete sequence lengths (which could be slow, and may trigger a GPU/CPU sync); but, the FLOP numbers will vary with the sparsity of the NestedTensor - more accurate than if we just assumed we padded everything.

Differential Revision: [D57120139](https://our.internmc.facebook.com/intern/diff/D57120139)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/125776
Approved by: https://github.com/Chillee
 2024-05-10 19:49:37 +00:00
albanD
76a26a885d Add module tracker (#125352) 
This does a few things that were originally a few PRs but I am on a new machine and don't have ghstack.
If it is too problematic to review, I can re-split, just let me know.
This does:
- Cleanup context manager use in test_flop_counter
- Remove need for mod argument in FlopCounterMode, warning about it
- Re-implement a Module tracker from scratch using global forward Module use and multi_grad_hook (we cannot use global backward Module hook because they don't look for nested Tensor and they're custom Function based instead of multi_grad_hook).
- Update FlopCouterMode to use the new ModuleTracker. All the existing test suite passes as-is (only changes there are new tests and refactoring mentioned above)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/125352
Approved by: https://github.com/mikaylagawarecki
 2024-05-04 18:33:35 +00:00
Manish Rajpal
2bbfb70831 ignore unsupported module from flop counter (#125346) 
Summary:
Torchscript modules do not support forward hooks and thus can't work with flop_counter context manager for hierarchical output by passing a module to FlopCounterMode on construction.

Currently any module that includes a script module causes an exception to be thrown so adding a try/catch to ignore any script modules for forward hooks.

Test Plan: CI Signals

Differential Revision: D56850661

Pull Request resolved: https://github.com/pytorch/pytorch/pull/125346
Approved by: https://github.com/842974287
 2024-05-02 20:30:52 +00:00
Manish Rajpal
da44d2f7fb split out flop counting its own method (#125061) 
Summary: Modularizing code for reuse by splitting __torch_dispatch__ to move flop counting to its own method.

Test Plan: unit tests

Differential Revision: D56644523

Pull Request resolved: https://github.com/pytorch/pytorch/pull/125061
Approved by: https://github.com/842974287
 2024-04-29 14:13:44 +00:00
chilli
84580f76d9 fix flop counter issue with out parameters (#123768) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/123768
Approved by: https://github.com/zou3519
 2024-04-11 09:39:53 +00:00
chilli
8bfc87ce74 fixed flop counter formula for conv transposed backwards pass (#119874) 
Fixes #119806
Pull Request resolved: https://github.com/pytorch/pytorch/pull/119874
Approved by: https://github.com/zou3519
ghstack dependencies: #119521
 2024-02-16 02:43:49 +00:00
Aaron Gokaslan
3fe437b24b [BE]: Update flake8 to v6.1.0 and fix lints (#116591) 
Updates flake8 to v6.1.0 and fixes a few lints using sed and some ruff tooling.
- Replace `assert(0)` with `raise AssertionError()`
- Remove extraneous parenthesis i.e.
  - `assert(a == b)` -> `assert a == b`
  - `if(x > y or y < z):`->`if x > y or y < z:`
  - And `return('...')` -> `return '...'`

Co-authored-by: Nikita Shulga <2453524+malfet@users.noreply.github.com>

Pull Request resolved: https://github.com/pytorch/pytorch/pull/116591
Approved by: https://github.com/albanD, https://github.com/malfet
 2024-01-03 06:04:44 +00:00
Aaron Gokaslan
ee5d981249 [BE]: Enable RUFF PERF402 and apply fixes (#115505) 
* Enable PERF402. Makes code more efficient and succinct by removing useless list copies that could be accomplished either via a list constructor or extend call. All test cases have noqa added since performance is not as sensitive in that folder.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/115505
Approved by: https://github.com/malfet
 2023-12-20 18:01:24 +00:00
ChanBong
5e10dd2c78 fix docstring issues in torch.utils (#113335) 
Fixes #112634

Fixes all the issues listed except in `torch/utils/_pytree.py` as the file no longer exists.

### Error counts

|File | Count Before | Count now|
|---- | ---- | ---- |
|`torch/utils/collect_env.py` | 39 | 25|
|`torch/utils/cpp_extension.py` | 51 | 13|
|`torch/utils/flop_counter.py` | 25 | 8|
|`torch/utils/_foreach_utils.py.py` | 2 | 0|
|`torch/utils/_python_dispatch.py.py` | 26 | 25|
|`torch/utils/backend_registration.py` | 15 | 4|
|`torch/utils/checkpoint.py` | 29 | 21|

Pull Request resolved: https://github.com/pytorch/pytorch/pull/113335
Approved by: https://github.com/ezyang
 2023-11-13 19:37:25 +00:00
chilli
74adb4cccc Updated flop counter to accept pytree inputs/outputs (#111990) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/111990
Approved by: https://github.com/ezyang
 2023-10-26 01:25:27 +00:00
chilli
ada65508d2 Add option to flop counter formula registration to get raw values (#110591) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110591
Approved by: https://github.com/awgu
ghstack dependencies: #110501, #110504
 2023-10-05 21:14:41 +00:00
chilli
e686341f64 Consider that ops can be fused into cat in the min-cut partitioner (#110501) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/110501
Approved by: https://github.com/eellison
 2023-10-05 01:34:57 +00:00
Justin Chu
4cc1745b13 [BE] f-stringify torch/ and scripts (#105538) 
This PR is a follow up on the pyupgrade series to convert more strings to use f-strings using `flynt`.

- https://docs.python.org/3/reference/lexical_analysis.html#f-strings
- https://pypi.org/project/flynt/

Command used:

```
flynt torch/ -ll 120
flynt scripts/ -ll 120
flynt tools/ -ll 120
```

and excluded `collect_env.py`

Pull Request resolved: https://github.com/pytorch/pytorch/pull/105538
Approved by: https://github.com/ezyang, https://github.com/malfet
 2023-07-21 19:35:24 +00:00
Nikita Shulga
5837e95d30 [Reland] Update mypy to 1.4.1 (#105227) 
This PR re-lands
- [Typing] Fix PEP 484 Violation (#105022)
- Update mypy to 1.4.1 (#91983)

That were reverted due to the conflict with internal source repo.

Mostly fixes for PEP-484 violation (i.e. when default arg is set to None, but type is not annotated as optional)
Plus few real fixes:
  - Add missing `_get_upgraders_entry_map` to `torch/_C/__init__.pyi`
  - Add missing return statement to `torch._export. deserialize_graph`
  - Fix error message in `torch.ao.ns.fx.weight_utils.get_lstm_mod_weights`
  - Add assert it `torch/optim/optimizer.py` that Optional list is not None
TODO (in followup PR):
  - Fix erroneous `isinstance` check in `torch/ao/quantization/_pt2e/qat_utils.py`

Unrelated, to bypass CI failures due to the gcc9 dependency update in Ubuntu-18.04:
- Add hack to squash older libstdc++ from conda environment in favor one from OS to `.ci/docker/install_conda.sh`
- Update bazel cuda builds to focal, as with libstdc++-6.0.32 bazel builds loose the ability to catch exceptions (probably because they link with cupti statically, but I could not found where it is done)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/105227
Approved by: https://github.com/atalman, https://github.com/albanD, https://github.com/Skylion007
 2023-07-15 20:30:20 +00:00
PyTorch MergeBot
15fd1ea118 Revert "[Reland] Update mypy to 1.4.1 (#105227)" 
This reverts commit c9c4f8efc3.

Reverted https://github.com/pytorch/pytorch/pull/105227 on behalf of https://github.com/atalman due to trying to mitigate ci sev #105248 ([comment](https://github.com/pytorch/pytorch/pull/105227#issuecomment-1636510935))
 2023-07-14 22:28:35 +00:00
Nikita Shulga
c9c4f8efc3 [Reland] Update mypy to 1.4.1 (#105227) 
This PR re-lands
- [Typing] Fix PEP 484 Violation (#105022)
- Update mypy to 1.4.1 (#91983)

That were reverted due to the conflict with internal source repo.

Mostly fixes for PEP-484 violation (i.e. when default arg is set to None, but type is not annotated as optional)
Plus few real fixes:
  - Add missing `_get_upgraders_entry_map` to `torch/_C/__init__.pyi`
  - Add missing return statement to `torch._export. deserialize_graph`
  - Fix error message in `torch.ao.ns.fx.weight_utils.get_lstm_mod_weights`
  - Add assert it `torch/optim/optimizer.py` that Optional list is not None
TODO (in followup PR):
  - Fix erroneous `isinstance` check in `torch/ao/quantization/_pt2e/qat_utils.py`
Pull Request resolved: https://github.com/pytorch/pytorch/pull/105227
Approved by: https://github.com/atalman, https://github.com/albanD, https://github.com/Skylion007
 2023-07-14 20:45:12 +00:00
PyTorch MergeBot
3c5a494d7a Revert "Update mypy to 1.4.1 (#91983)" 
This reverts commit 634659e262.

Reverted https://github.com/pytorch/pytorch/pull/91983 on behalf of https://github.com/malfet due to It's dependent change was reverted, so reverting this one as well, to keep CI clean ([comment](https://github.com/pytorch/pytorch/pull/91983#issuecomment-1636059709))
 2023-07-14 15:59:16 +00:00
Nikita Shulga
634659e262 Update mypy to 1.4.1 (#91983) 
Mostly fixes for PEP-484 violation (i.e. when default arg is set to None, but type is not annotated as optional)
Plus few real fixes:
  - Add missing `_get_upgraders_entry_map` to `torch/_C/__init__.pyi`
  - Add missing return statement to `torch._export. deserialize_graph`
  - Fix error message in `torch.ao.ns.fx.weight_utils.get_lstm_mod_weights`
  -
TODO (in followup PR):
  - Fix erroneous `isinstance` check in `torch/ao/quantization/_pt2e/qat_utils.py`
Pull Request resolved: https://github.com/pytorch/pytorch/pull/91983
Approved by: https://github.com/kit1980, https://github.com/ZainRizvi, https://github.com/huydhn, https://github.com/thiagocrepaldi, https://github.com/aaronenyeshi
 2023-07-13 16:30:36 +00:00
Andrew Gu
974525c053 De-register forward hooks upon exiting flop counter context (#103744) 
This PR fixes https://github.com/pytorch/pytorch/issues/103684.
- Instead of registering forward hooks in `__init__()`, do it upon `__enter__()`.
- De-register those forward hooks upon `__exit__()`.
- Achieve this by saving an additional mapping `_module_to_forward_hook_handles: Dict[nn.Module, _ForwardHookHandles]`. Only the values in the mapping (i.e. not the keys) are useful for this change. (A `List[_ForwardHookHandles]` would suffice.)
- The unit test accesses private attributes `_forward_hooks` and `_forward_pre_hooks` :/

Note that this PR is technically not backward compatible since it does not register the hooks upon `__init__()`, which means that you will not get the flops counting without the context manager.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/103744
Approved by: https://github.com/Chillee
 2023-06-20 19:34:02 +00:00
Horace He
547bef11ee tweak heuristic for sdpa selection based off of *data* (and a decision tree) (#99644) 
High level approach:
1. I generated a bunch of data comparing FlashAttention and Cutlass implementations (https://pastebin.com/pe0j3YeK)
2. I trained a decision tree using standard train/val split methodology and hyperparameter sweeps (https://pastebin.com/fjYX1HjR).
2a. I did a bunch of feature augmentation to capture interactions between features.

The heuristic I ended up with is:
```
use_flash = seq_len / (num_heads * batch_size) > 6
```

TL;DR: On my dataset, where FlashAttention and Cutlass differ by more than 10%, the existing heuristic achieves 69% accuracy.  My new heuristic achieves 94% accuracy.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/99644
Approved by: https://github.com/ngimel, https://github.com/drisspg
 2023-04-21 23:28:44 +00:00
Horace He
c75dd7c413 grab bag of changes (#98572) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/98572
Approved by: https://github.com/shunting314, https://github.com/mlazos
 2023-04-07 20:02:59 +00:00
Horace He
44c9ecad8d fix flop formulas for sdpa (#96690) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/96690
Approved by: https://github.com/drisspg
 2023-03-16 04:55:56 +00:00
Horace He
2a08a62777 Add extra metadata (as comments) to Inductor generated code (#96581) 
New output
<img width="942" alt="image" src="https://user-images.githubusercontent.com/6355099/224794006-a993a2a8-d6ff-49da-8891-7b2373030a3d.png">

Pull Request resolved: https://github.com/pytorch/pytorch/pull/96581
Approved by: https://github.com/ngimel, https://github.com/shunting314, https://github.com/voznesenskym
 2023-03-14 03:59:59 +00:00