Commit Graph

Select branches

Hide Pull Requests

fix-extelemwise-in-combine-ops

gh-pages

jit-hook

keren/assert

keren/improve-hook

keren/insert-slice-other-nonzero

keren/perf-debug

keren/v100-perf-regression

master

phil/fused-attention-perf-fixup

phil/mma-v1-is-row-debug

phil/swizzle-bug-repro

port-fma

rocm

#10

#100

#1000

#1001

#1002

#1004

#1006

#1007

#1008

#101

#1010

#1012

#1013

#1013

#1014

#1018

#1019

#102

#1020

#1020

#1025

#1027

#1028

#1029

#103

#1030

#1033

#1034

#1036

#1037

#1038

#1039

#104

#1042

#1043

#1043

#105

#106

#107

#108

#109

#11

#11

#110

#111

#112

#114

#116

#118

#119

#120

#121

#123

#124

#125

#126

#127

#128

#129

#13

#131

#132

#133

#134

#135

#136

#137

#138

#139

#140

#141

#142

#143

#144

#145

#146

#147

#148

#149

#15

#151

#152

#158

#164

#165

#167

#168

#172

#173

#178

#179

#18

#180

#185

#186

#188

#19

#190

#192

#193

#195

#198

#199

#20

#200

#203

#204

#205

#207

#209

#212

#219

#22

#222

#224

#225

#228

#23

#23

#231

#24

#240

#249

#250

#251

#253

#255

#256

#257

#258

#259

#260

#261

#268

#271

#272

#273

#276

#279

#28

#280

#281

#282

#283

#285

#286

#287

#288

#291

#292

#293

#294

#295

#296

#297

#298

#299

#3

#300

#301

#302

#303

#304

#305

#306

#307

#308

#309

#311

#312

#313

#314

#317

#318

#320

#324

#326

#331

#336

#337

#338

#342

#344

#345

#346

#347

#348

#349

#35

#350

#351

#356

#357

#358

#361

#362

#367

#368

#372

#373

#374

#377

#379

#38

#380

#381

#382

#383

#386

#387

#388

#390

#391

#392

#393

#394

#395

#396

#397

#399

#40

#400

#401

#403

#406

#407

#408

#409

#41

#413

#414

#415

#417

#418

#420

#421

#422

#423

#424

#425

#426

#427

#428

#430

#431

#432

#436

#438

#439

#440

#442

#444

#445

#446

#447

#448

#449

#45

#450

#451

#453

#455

#456

#457

#458

#462

#463

#464

#467

#468

#469

#470

#471

#473

#474

#478

#481

#482

#483

#484

#485

#487

#488

#490

#491

#492

#493

#495

#499

#500

#501

#502

#503

#505

#507

#510

#513

#514

#515

#516

#519

#52

#520

#522

#523

#524

#526

#527

#528

#53

#531

#533

#534

#535

#538

#538

#539

#541

#545

#546

#547

#548

#549

#551

#552

#553

#555

#556

#557

#559

#560

#561

#562

#564

#565

#567

#569

#57

#570

#571

#572

#575

#575

#577

#578

#579

#58

#582

#587

#588

#59

#590

#595

#598

#60

#600

#601

#602

#604

#606

#607

#608

#61

#614

#614

#617

#62

#623

#63

#632

#636

#637

#644

#65

#650

#651

#652

#653

#654

#655

#657

#658

#66

#660

#661

#662

#663

#664

#665

#666

#667

#668

#669

#670

#671

#672

#678

#68

#682

#683

#684

#685

#689

#69

#691

#692

#693

#694

#696

#697

#699

#7

#70

#700

#701

#702

#703

#704

#706

#708

#709

#71

#710

#711

#712

#715

#716

#718

#722

#724

#726

#727

#728

#729

#73

#732

#733

#735

#736

#738

#739

#740

#742

#746

#747

#749

#75

#750

#751

#752

#753

#754

#755

#757

#758

#759

#764

#765

#766

#767

#769

#77

#774

#775

#776

#777

#78

#780

#782

#784

#785

#786

#788

#789

#790

#791

#792

#794

#796

#797

#798

#799

#80

#800

#801

#803

#804

#805

#809

#81

#812

#814

#815

#816

#817

#818

#819

#82

#820

#821

#822

#823

#825

#826

#827

#829

#83

#830

#831

#833

#834

#835

#836

#837

#838

#839

#840

#841

#842

#843

#844

#845

#847

#848

#849

#850

#851

#852

#853

#854

#856

#857

#858

#859

#86

#862

#863

#864

#867

#868

#869

#87

#872

#873

#874

#875

#876

#877

#878

#879

#88

#880

#881

#883

#885

#886

#887

#887

#888

#889

#89

#890

#890

#894

#896

#897

#898

#899

#90

#901

#902

#903

#904

#906

#907

#908

#909

#91

#910

#912

#913

#914

#915

#916

#917

#918

#92

#920

#921

#922

#923

#924

#925

#926

#927

#928

#929

#93

#930

#931

#933

#936

#937

#938

#939

#94

#941

#943

#944

#945

#946

#947

#947

#948

#95

#951

#952

#953

#956

#957

#958

#959

#96

#960

#961

#962

#963

#964

#966

#968

#969

#97

#970

#971

#972

#973

#975

#976

#977

#978

#979

#980

#982

#982

#983

#985

#987

#988

#990

#991

#993

#994

#995

#996

#997

#998

#999

isaac

legacy-backend

v0.1

v0.2.3

v0.4

v1.0

v1.1

v1.1.1

v1.1.2

Mono Color

• 8b82c8ed1a [GH-PAGES] Updated website Philippe Tillet 2022-02-09 03:39:04 +00:00
• 5c5dc17308 [GH-PAGES] Updated website Philippe Tillet 2022-02-09 03:23:53 +00:00
• e2e2cafecc [GH-PAGES] Updated website Philippe Tillet 2022-02-09 03:17:47 +00:00
• 1caa8f007a [GH-PAGES] Updated website Philippe Tillet 2022-02-09 01:09:10 +00:00
• 4941bc7001 [DOCS] Some more fixes (#455) Philippe Tillet 2022-02-08 16:53:56 -08:00
• 87b6bdfc6e Updated update-website.sh Philippe Tillet 2022-02-08 16:51:48 -08:00
• 989c163b13 [GH-PAGES] Updated website Philippe Tillet 2022-02-08 23:45:21 +00:00
• 95bb988ed0 [GH-PAGES] Updated website Philippe Tillet 2022-02-08 20:05:45 +00:00
• 5b4729ede6 [GH-PAGES] Updated website Philippe Tillet 2022-02-08 20:02:05 +00:00
• 2fdf0a4fe8 [DOCS] changed build command Philippe Tillet 2022-02-08 11:45:21 -08:00
• 077d6c8ff0 [DOCS] re-activated tutorials Philippe Tillet 2022-02-08 11:42:39 -08:00
• a47b2f5208 [GH-PAGES] Updated website Philippe Tillet 2022-02-08 19:40:59 +00:00
• 822ddcd14b [DOCS] Added versioning (#453) Philippe Tillet 2022-02-08 11:28:18 -08:00
• cf8c3ba438 [GH-PAGES] Updated website Philippe Tillet 2022-02-08 00:24:15 +00:00
• 0f03cfcfd3 [GH-PAGES] Updated website Philippe Tillet 2022-02-07 03:25:11 +00:00
• 7b48340ffd [CI] Some fixes for the build (#451) Philippe Tillet 2022-02-06 19:11:33 -08:00
• 5a8a544d10 [OPS][BLOCKSPARSE] Improved robustness, clarity and performance (#450) Philippe Tillet 2022-02-06 18:00:45 -08:00
• 69ff52ea1f [CODEGEN] removed buggy (and mostly useless) optimization in peephole pass (#449) Philippe Tillet 2022-02-05 21:37:23 -08:00
• 137bb67fad [LANG] Add fp16 to fp8 conversion (#444) TC 2022-02-02 20:42:09 -08:00
• 3b20170fa3 Merge pull request #448 from openai/v2.0 Philippe Tillet 2022-01-30 20:49:08 -08:00
• b0d6e2f322 [STYLE] run autopep Philippe Tillet 2022-01-30 20:27:44 -08:00
• 2922dc141c Merge branch 'master' into v2.0 Philippe Tillet 2022-01-30 20:25:01 -08:00
• 807d8a1945 [ALL] Merge master (#447) Philippe Tillet 2022-01-30 20:21:20 -08:00
• bef76b142a [BACKEND] float division is now approximate by default (#446) Philippe Tillet 2022-01-29 18:29:29 -08:00
• bd52e530a0 [OPS][BLOCKSPARSE] Fix padding issue in DSD LUT (#445) Philippe Tillet 2022-01-28 21:40:30 -08:00
• e68d6a7776 [BACKEND] Making the warp-level tile "more square" to increase data-reuse for tl.dot. (#442) daadaada 2022-01-28 01:59:54 +08:00
• 59d371c6eb [BACKEND] Added Int8 mma (#440) daadaada 2022-01-28 01:12:44 +08:00
• 3a23c1dd33 [BACKEND] minor, hotfix for gcc compilation (#439) Benjamin Lefaudeux 2022-01-23 17:24:02 -05:00
• ccf9abe0ba [FRONTEND][RANDOM] Improved backward compatibility of RNG (#438) Philippe Tillet 2022-01-21 18:05:55 -08:00
• 4c97d1ecd7 [FRONTEND] Bunch of fixes here and there (#436) Philippe Tillet 2022-01-20 10:55:59 -08:00
• e0c5709cc8 [FRONTEND] Fixed semantics bug on ptr to bool conversions (#432) Philippe Tillet 2022-01-17 18:00:03 -08:00
• 2a944ded53 [TESTS] Added bfloat16 tests (#430) daadaada 2022-01-14 15:38:32 +08:00
• 4c94359199 [FRONTEND] Alignment fix-up (#428) Philippe Tillet 2022-01-11 23:11:58 -08:00
• bbc78f6516 [FRONTEND][RANDOM] Make sure offset dtype is always uint32 before calling uint32_to_uniform_float (#427) Philippe Tillet 2022-01-11 11:08:49 -08:00
• bf32205edc [OPS][BLOCKSPARSE] Remove unnecessary loop and add cuda bool layout support (#425) Botao Yu 2022-01-12 03:07:16 +08:00
• 94a2e10fe5 [BACKEND] Add bf16 & tf32 mma supports (on A100) (#426) daadaada 2022-01-12 02:20:31 +08:00
• efdabe6073 [STYLE] check python with flake8 (#424) Madeleine Thompson 2022-01-07 15:28:36 -08:00
• a70acfec77 [STYLE] add isort and autopep8 config files and check on CI (#423) Madeleine Thompson 2022-01-07 13:11:34 -08:00
• 9801aa7b56 [DOCS] fix tutorials for v2.0 (#422) Madeleine Thompson 2022-01-07 12:34:38 -08:00
• 8bf551ae7a [STYLE] run autopep8 and isort (#421) Madeleine Thompson 2022-01-06 14:34:17 -08:00
• 6f7acad48f [CODEGEN] Avoid use of deprecated AST nodes (#418) Shantanu 2022-01-06 12:04:33 -08:00
• 120cda015e [FRONTEND] use unsigned integers to simplify RNG (#417) Madeleine Thompson 2022-01-06 10:49:09 -08:00
• 001fb757fe [OPS][BLOCKSPARSE] Added .contiguous() in blocksparse inputs when necessary (#420) Philippe Tillet 2022-01-06 12:56:22 -05:00
• 0ab9d67bad uint8, uint16, uint32, and uint64 in kernels (#413) Madeleine Thompson 2022-01-05 15:27:17 -08:00
• d8db0308cb [TEST] use numpy for reference results in test_core.py (#409) Madeleine Thompson 2022-01-04 13:07:29 -08:00
• 03f1256f60 [FRONTEND] Added volatile flag for load (#407) Philippe Tillet 2021-12-30 22:33:24 -08:00
• 3edc2633e9 [TUTORIALS] Fix 01-vector-add.py typo (#406) Noah Ziems 2021-12-29 18:09:34 -05:00
• 985798f101 add missing bfloat16 repr and improve assertions (#403) Madeleine Thompson 2021-12-23 17:01:17 -08:00
• d8fce83e7a [FRONTEND] Remade exception picklable Philippe Tillet 2021-12-21 22:14:06 -08:00
• a425f24d54 [FRONTEND] Better cache hook (#400) Philippe Tillet 2021-12-21 21:29:47 -08:00
• 2509124dd0 [DRIVER] Fixed some issue with how ptxas is used (#399) Philippe Tillet 2021-12-21 14:31:51 -08:00
• 39d4bfed83 [OPS] Add performance model for gemm/gemv (#397) daadaada 2021-12-22 01:56:10 +08:00
• 5cdb948c05 [FRONTEND] signed-integer math fixes and testing (#395) Madeleine Thompson 2021-12-21 09:46:05 -08:00
• 4a8953efa3 [FRONTEND] Replace the legacy print call in triton.cc with the SlotTracker-based one. (#396) daadaada 2021-12-19 10:03:22 +08:00
• fa62b4a8f6 [FRONTEND] better stringification (#394) Madeleine Thompson 2021-12-17 20:11:45 -08:00
• 4e93b41c52 [GENERAL] Some minor fixups (#393) Philippe Tillet 2021-12-17 18:06:21 -08:00
• e062812969 [CODEGEN] Disabled peephole for masked load + select -- masked_load doesn't work as expected when vectorized Philippe Tillet 2021-12-17 12:44:47 -08:00
• eb077fc993 [RUNTIME] fixed NVidia DLL names on Windows (#392) Victor 2021-12-16 22:09:52 -08:00
• e0b92c1380 [FRONTEND] Reverted from .random import *. There are still some namespace errors in the Triton frontend apparently Philippe Tillet 2021-12-16 18:37:51 -08:00
• 558555630f [FRONTEND] Added xor_sum Philippe Tillet 2021-12-16 17:55:35 -08:00
• 94d5c2e8b5 [ROCM] enable matmul(dot) and others (#391) Michael Melesse 2021-12-13 12:28:15 -08:00
• e575ae3443 [FRONTEND] Minor accumulated style and warning fixes (#388) Madeleine Thompson 2021-12-10 15:19:20 -08:00
• 9def2424ab [RUNTIME] Fix typo in IfExp Philippe Tillet 2021-12-09 15:14:06 -08:00
• e31b9b4e66 [RUNTIME] Better support for None (#387) Philippe Tillet 2021-12-09 13:21:22 -08:00
• 73b04d71b2 Fixes for building on Windows (#382) Victor 2021-12-07 14:10:58 -08:00
• 0ff1a26b70 fixed p2p tests failing when there are no supported p2p devices (#386) Victor 2021-12-06 18:14:03 -08:00
• f23bf55f15 [RUNTIME] release the gil on launch (#383) Philippe Tillet 2021-12-03 13:01:01 -08:00
• 8ec9f037bb [BACKEND/CODE_GEN] Fixed float32 matmul problem (#380) Philippe Tillet 2021-11-30 22:00:56 -08:00
• c86ad9c9ab [FRONTEND] Added default arguments to non-kernel @triton.jit'd function (#379) Philippe Tillet 2021-11-29 19:11:26 -08:00
• 1296eb877b [RUNTIME] Config hook v2.0 (#373) daadaada 2021-11-22 03:20:59 +08:00
• 5693b582ea [RUNTIME] Now using pybind11 to avoid memory leaks (#377) Philippe Tillet 2021-11-21 02:30:22 -08:00
• edd4b0c8b7 [CODEGEN] Fixed issue with jit function passed as constexpr Philippe Tillet 2021-11-16 09:53:34 -08:00
• 5b7ba3eb96 [CODEGEN] Reverted to old launch method (memory leak?) Philippe Tillet 2021-11-16 01:21:03 -08:00
• 791b953b21 [CODEGEN] Reverted to old way to query current stream Philippe Tillet 2021-11-16 00:17:27 -08:00
• b908095872 [VERSION] Bumped triton.__version__ to 2.0.0 Philippe Tillet 2021-11-12 15:10:04 -08:00
• 01cc3d4503 [RUNTIME] Restored do_not_specialize (#374) Philippe Tillet 2021-11-12 15:06:55 -08:00
• e66bf76354 [RUNTIME] Bunch of bugfixes (#372) Philippe Tillet 2021-11-12 00:55:00 -08:00
• f7ab96cfd7 [FRONTEND] Fixed some issues with constexpr Philippe Tillet 2021-11-05 09:26:33 -07:00
• 9a02dddf29 Fix sdd_lut (#368) daadaada 2021-11-09 00:25:05 +08:00
• 5d54352164 [FRONTEND] Significantly reduce kernel launch time (#367) Philippe Tillet 2021-11-04 13:25:24 -07:00
• 2acaa4d0dd [LANG] Added support for constexpr (#361) Philippe Tillet 2021-10-30 00:32:58 -07:00
• b7f0e87dc2 [DRIVER] Removed std::cout log message Philippe Tillet 2021-10-29 10:42:10 -07:00
• 770ea96cca [PACKAGING] Bumped dev version to 2.0.0 Philippe Tillet 2021-10-29 01:28:17 -07:00
• 969d6de8a2 [PACKAGING] Bumped dev version to 1.1.2 Philippe Tillet 2021-10-29 01:24:19 -07:00
• 2d6df9b518 [PACKAGING] Bumped dev version to 1.1.2 v1.1.2 Philippe Tillet 2021-10-29 01:24:19 -07:00
• 1b842f8e5e [CI] Now running integration tests on pull requests on branch v2.0 Philippe Tillet 2021-10-29 01:06:50 -07:00
• d3e584d4ba Revert "[DRIVER] Fixed CUDA 10.1 bug (#357)" (#358) Philippe Tillet 2021-10-26 15:04:49 -07:00
• d35014ba47 [DRIVER] Fixed CUDA 10.1 bug (#357) Philippe Tillet 2021-10-26 11:17:06 -07:00
• 5ce1b726dc [CODEGEN] Various bugfixes that make it possible to fuse RNG in a matmul epilogue (#356) Philippe Tillet 2021-10-24 02:30:46 -07:00
• 858dec8372 [CODEGEN] Add cache modifier to tl.load (#351) daadaada 2021-10-18 13:14:04 +08:00
• 90ded16c32 [DOCS] Added placeholder docstring for layernorm tutorial Philippe Tillet 2021-10-15 19:04:01 -07:00
• abbc554838 [VERSION] Bumped version to 1.1.1 (#350) v1.1.1 Philippe Tillet 2021-10-14 18:09:39 -07:00
• 9b32075062 [CODEGEN] Some compiler improvements (#349) Philippe Tillet 2021-10-13 17:49:39 -07:00
• c2e6b90ff1 [CODEGEN] Fixes masked load exception (#342) Stephen McGroarty 2021-10-13 21:31:52 +01:00
• bfacc191b3 [FRONTEND] Now cache re-compiles when language changes (#348) Philippe Tillet 2021-10-13 12:29:57 -07:00
• f5ad168686 [PYTHON] Fix up __version__ (#345) Shantanu 2021-10-13 00:09:00 -07:00
• c3c0ff0552 [LANGUAGE] Fixed issue with duplicates in large arrays of random uniform numbers (#338) Philippe Tillet 2021-10-10 15:22:34 -07:00
• 9e9d781912 [CODEGEN] Pipeline fixup (#336) daadaada 2021-10-10 16:47:11 +08:00
• d5f20dbce0 [IR] Fix error when building in debug mode (#331) daadaada 2021-10-09 12:40:20 +08:00
• d4baad426d [DOCS] Added layer norm example (#326) Philippe Tillet 2021-10-08 11:02:10 -07:00