Commits

Branch selector

 main

User selector

laithsakka

Datepicker

All time

Commit History

Commits on Nov 5, 2024

• use torch.sym_sum instead of incremental sum in _cat_meta (#139653)

  Show description for 6ad52db
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  6ad52db

  Copy full SHA for 6ad52db

• increase add_loop benchmark and refresh all results! (#139703)

  Show description for de4216b
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  de4216b

  Copy full SHA for de4216b

Commits on Nov 4, 2024

• Classify miss-inplaced tensors in logs. (#139240)

  Show description for 3f248a5
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  3f248a5

  Copy full SHA for 3f248a5

Commits on Oct 31, 2024

• Don't uselessly recompute axiom dict every static eval call (#138967)

  Show description for 6a1c451
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  6a1c451

  Copy full SHA for 6a1c451

• Don't uselessly recompute axiom dict every static eval call (#138967)

  Show description for 24b695a
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  24b695a

  Copy full SHA for 24b695a

Commits on Oct 28, 2024

• Expliclty avoid recording when should_record_events is false in record_shapeenv_event (#138965)

  Show description for 475ba1d
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  475ba1d

  Copy full SHA for 475ba1d

• In Inductor, be willing to generate deferred runtime asserts when unbacked (#138804)

  Show description for c056dc4
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  c056dc4

  Copy full SHA for c056dc4

Commits on Oct 26, 2024

• Several enhancements for check_results.py (#137925)

  Show description for 705f5b3
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  705f5b3

  Copy full SHA for 705f5b3

• In Inductor, be willing to generate deferred runtime asserts when unbacked (#138804)

  Show description for f1a677c
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  f1a677c

  Copy full SHA for f1a677c

• Enable failing diffs on update_hint_regression and sum_floordiv_regression and autograd benchmarks regression (#137548)

  Show description for 10e2840
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  10e2840

  Copy full SHA for 10e2840

Commits on Oct 23, 2024

• Introduce torch.sym_add, variadic add (#138660)

  Show description for ed313a5
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  ed313a5

  Copy full SHA for ed313a5

• Generate slice.Tensor view operations instead of as_strided when split is used in the original program. (#137225)

  Show description for 72ea7ba
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  72ea7ba

  Copy full SHA for 72ea7ba

• Remove parallel_and and parallel_or (#138135)

  Show description for 662d07e
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  662d07e

  Copy full SHA for 662d07e

Commits on Oct 17, 2024

• include bw_compiler in strobelight profile (#138060)

  Show description for 7e16c9d
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  7e16c9d

  Copy full SHA for 7e16c9d

• Increase default COMPILE_STROBELIGHT_MAX_STACK_LENGTH to 500 (#138006)

  Show description for ba10259
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  ba10259

  Copy full SHA for ba10259

Commits on Oct 15, 2024

• enable auto functionalize v2 by default (#136685)

  Show description for 60cf72e
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  60cf72e

  Copy full SHA for 60cf72e

• Do not compute base in export mode (#137760)

  Show description for 05b6200
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  05b6200

  Copy full SHA for 05b6200

Commits on Oct 11, 2024

• bump up add_loop_inductor_gpu expected instruction count. (#137672)

  Show description for a06d49a
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  a06d49a

  Copy full SHA for a06d49a

Commits on Oct 10, 2024

• Avoid generating as_strided for alaising views in auto_functionalize_v2 (#137149)

  Show description for 1aa130e
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  1aa130e

  Copy full SHA for 1aa130e

• Enable failing diffs for regressions on basic_modules_ListOfLinears benchmarks (#137541)

  Show description for f394fb5
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  f394fb5

  Copy full SHA for f394fb5

• Generate new expected results file when there is failures in diff time benchmarks (#137551)

  Show description for 3610467
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  3610467

  Copy full SHA for 3610467

Commits on Oct 4, 2024

• Proper handling of arguments passed by in kwargs inside zip_schema (#137311)

  Show description for 7c1d939
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  7c1d939

  Copy full SHA for 7c1d939

• Enable regression test for add loop benchmarks (#136573)

  Show description for 8b9cbf2
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  8b9cbf2

  Copy full SHA for 8b9cbf2

Commits on Sep 30, 2024

• Enable tracing through auot_functionalized_v2 in compiled autograd (#136806)

  Show description for dfe1d45
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  dfe1d45

  Copy full SHA for dfe1d45

Commits on Sep 29, 2024

• Enable failing diffs on regression (#136551)

  Show description for e205193
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  e205193

  Copy full SHA for e205193

Commits on Sep 27, 2024

• Log total miss inplaced bytes (#136684)

  Show description for 73f038c
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  73f038c

  Copy full SHA for 73f038c

Commits on Sep 21, 2024

• Remove duplicate line (#136383)

  Show description for 0b91e7e
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  0b91e7e

  Copy full SHA for 0b91e7e

Commits on Sep 19, 2024

• add basic_modules_ListOfLinears_inductor_gpu_force_shape_pad (#136175)

  Show description for b71802f
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  b71802f

  Copy full SHA for b71802f

Commits on Sep 16, 2024

• use csv extention for test report in order for it to be uploaded to s3 (#136128)

  Show description for 071da87
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  071da87

  Copy full SHA for 071da87

Commits on Sep 14, 2024

• Only keep ListOfLinears module in basic_modules_benchmarks and add gpu version. (#135730)

  Show description for b82122b
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  b82122b

  Copy full SHA for b82122b

• Disable garbage collection during compile_time_instructions count in benchmark base by default. (#135768)

  Show description for 44dd218
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  44dd218

  Copy full SHA for 44dd218

Commits on Sep 13, 2024

• Reduce default iterations to 5 . (#135773)

  Show description for 46935c8
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  46935c8

  Copy full SHA for 46935c8

• Only measure compile time instruction count for sum_floordiv benchmark (#135785)

  Show description for 4f407c1
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  4f407c1

  Copy full SHA for 4f407c1

• Add gpu and gpu_dynamic versions of add_loop (#135809)

  Show description for 2e461e5
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  2e461e5

  Copy full SHA for 2e461e5

• Remove cycle dependency by localizing the import. (#135926)

  Show description for ba6e0f3
  

  laithsakka

  authored and

  pytorchmergebot

  committed
  ba6e0f3

  Copy full SHA for ba6e0f3

Pagination

Previous

Next