Skip to content
Closed
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
801 commits
Select commit Hold shift + click to select a range
6365a30
[TIR] Fix reduce buffer allocation position (#17799)
wrongtest-intellif Apr 3, 2025
cdebac1
[Relax][PyTorch] Fix torch 2.6 compatibility issues (#17807)
mshr-h Apr 4, 2025
1bf3423
[Relax][PyTorch] Delete duplicate converter function `_to` (#17809)
mshr-h Apr 4, 2025
43adad7
[Relax][PyTorch] Add support for argsort, sort, topk ops (#17810)
tlopex Apr 4, 2025
9c97bba
[NFC] Fix `explict` typo (#17811)
c8ef Apr 7, 2025
6a3e5e4
[Relax][PyTorch] Support argsort, topk ops for ExportedProgram import…
tlopex Apr 7, 2025
88d9aa6
[Flashinfer] Added jit flow for sampling kernel (#17763)
annanyapr Apr 7, 2025
0d2eab2
[Cublas] Added support for bfloat16 while dispatching to cublas kerne…
annanyapr Apr 7, 2025
32a6f01
[Relax][PyTorch] Improve ExportedProgram frontend by supporting `unfl…
mshr-h Apr 8, 2025
52fe358
[Relax][PyTorch] Support dynamic shapes in ExportedProgram frontend (…
mshr-h Apr 9, 2025
bf61216
[Relax][PyTorch] Add Softplus Op Support for Exported Program and FX …
Deivanayaki-S Apr 9, 2025
03ba03e
[Relax][PyTorch] Cleanup tests for ExportedProgram frontend (#17822)
mshr-h Apr 10, 2025
5842bdb
[Relax][PyTorch] Add support for broadcast_to, narrow ops (#17820)
tlopex Apr 10, 2025
5ae3db2
[Relax][PyTorch] Add stack.default and sum.default to exported progra…
hugolatendresse Apr 11, 2025
1c60502
[release] Update version to 0.20.0 on main branch
ysh329 Apr 11, 2025
3319b41
[release] Update version to 0.21.dev0 on main branch
ysh329 Apr 11, 2025
4790798
[Install] Fix error during python/tvm installation (#17808)
vacu9708 Apr 14, 2025
820642b
[Relax] Fix Torch frontends to report all the missing ops (#17826)
Deivanayaki-S Apr 14, 2025
4e41b42
[Relax][PyTorch] Support narrow and broadcast_to ops for ExportedProg…
tlopex Apr 14, 2025
6bd55f0
[Relax][PyTorch] full.default, full_like.default, ones.default (#17832)
hugolatendresse Apr 14, 2025
f1ba5ed
[BugFix][TIR] Schedule support reverse-inline with reduction blocks (…
MasterJH5574 Apr 15, 2025
fcb8853
[Relax] Refactor missing op check into shared utility for Torch front…
Deivanayaki-S Apr 16, 2025
f081656
[Relax][PyTorch] Add support for norm op (#17841)
tlopex Apr 17, 2025
1bb7833
[Relax][PyTorch] Add Logaddexp op support for exported program (#17803)
AishwaryaElango Apr 17, 2025
899e121
fixing incorrect docstring in upsampling.py (#17845)
Cookiee235 Apr 17, 2025
601d570
[CI] Upgrade ubuntu runner image for GitHub CI (#17846)
mshr-h Apr 17, 2025
103e54b
[Relax][PyTorch] Add PReLU Op Support for Exported Program and FX gra…
Deivanayaki-S Apr 18, 2025
2abff88
[Relax][Pytorch] Add masked_fill op support in ExportedProgram (#17850)
kavin-sai-krishna Apr 18, 2025
aafb0db
[Relax][PyTorch] Add RSub Op Support for Exported Program and FX grap…
Deivanayaki-S Apr 18, 2025
982b46c
Fix docstring in batch_to_space_nd and bitpack (#17848)
Cookiee235 Apr 18, 2025
ba9f174
[3rdparty] Bump DLPack to v1.1 for float8/6/4 dtype supports (#17831)
MasterJH5574 Apr 18, 2025
c00f52a
[Relax][PyTorch] Add Stack Op Support for Exported Program (#17819)
Pratheesh-04-MCW Apr 18, 2025
bf6e101
[Relax][PyTorch] Add Pad Op Support for Exported Program and FX graph…
Deivanayaki-S Apr 18, 2025
4e36b9b
[Relax][PyTorch] Add mul_.Tensor, max.default, min.default and pow.Sc…
Deivanayaki-S Apr 18, 2025
f41c3b5
[Relax][PyTorch] Support `leaky_relu_.default` and `reshape_as.defaul…
mshr-h Apr 18, 2025
40a16db
Fix incorrect docstring in topi softmax (#17844)
Cookiee235 Apr 18, 2025
36f2502
Add op support for roll op (#17839)
Pratheesh-04-MCW Apr 18, 2025
b3d3a7a
[Relax][PyTorch] Add copy_ op support in fxGraph (#17858)
PhyG0 Apr 21, 2025
111ddf7
[Relax][PyTorch] Support eye op for ExportedProgram importer (#17864)
tlopex Apr 21, 2025
b5b0337
[Relax][PyTorch] support for index.Tensor (#17836)
hugolatendresse Apr 21, 2025
fe1b228
[Relax][Pytorch] Add support for bitwise_or op support (#17871)
kavin-sai-krishna Apr 21, 2025
2ca6ec8
[Relax][PyTorch] Sort.default (#17852)
hugolatendresse Apr 21, 2025
1ef153f
[Relax][PyTorch] Refactor norm op for ExportedProgram importer (#17857)
tlopex Apr 23, 2025
299ef81
[BugFix][Relax][Pytorch] Incorrect Handling of In-Place Ops in FX-Bas…
kavin-sai-krishna Apr 23, 2025
1437d5c
[Relax][Pytorch] Add support for ones_like, zero_, zeros, type_as, it…
kavin-sai-krishna Apr 24, 2025
1ab6f6c
[BugFix][Relax][Pytorch] Fix incorrect behaviour of % (mod) operator …
kavin-sai-krishna Apr 24, 2025
0149fcd
Add support for index_put_ op (#17865)
Pratheesh-04-MCW Apr 24, 2025
67297c4
[CI] Upgrade pytorch to 2.7.0, torchvision to 0.22.0, and vulkan sdk …
mshr-h Apr 25, 2025
c5675dd
[Relax][PyTorch] Support linspace op for ExportedProgram importer (#1…
tlopex Apr 25, 2025
2de5b1e
[Relax][PyTorch] Add torch.isin Op Support for Exported Program and F…
Deivanayaki-S Apr 26, 2025
87b37b4
Fix onnx expand op (#17900)
xinxilwl Apr 27, 2025
b1d1cdc
[Relax][PyTorch] Support torch.bfloat16 dtype in pytorch frontend (#1…
mshr-h Apr 27, 2025
47b95ca
[Relax][FRONTEND][Pytorch] Add fmod support (#17893)
kavin-sai-krishna Apr 27, 2025
ed477b0
Add op support for zeros_like and fill_ (#17896)
Pratheesh-04-MCW Apr 28, 2025
03eadc6
[CI] Install PyTorch 2.7 compatible with CUDA 11.8 (#17905)
mshr-h Apr 28, 2025
5ca9fa6
[CI] Update images to 20250428-080833-03eadc65 (#17891)
mshr-h Apr 28, 2025
502a133
[RPC] Fix Bug That Change Dict When Iterate The Keys
Johnson9009 Apr 29, 2025
49564f7
Add masked_fill_.scalar, logical_not.default in Exported Program fron…
Pratheesh-04-MCW Apr 30, 2025
0c0cdd8
Add op support for new_zeros op in Exported Program and fx graph fron…
Pratheesh-04-MCW Apr 30, 2025
fd7fecc
[Relax][PyTorch] Add support for eye op in fx graph (#17908)
tlopex Apr 30, 2025
85e3689
Fix off-by-one error in the type index range check within Object::IsI…
ConvolutedDog Apr 27, 2025
cbba0e3
Make slot end calculation more readable
ConvolutedDog Apr 29, 2025
4955f3b
[Relax][PyTorch] Add Pixel Shuffle Op Support for Exported Program an…
Deivanayaki-S May 1, 2025
c211e48
[Triton] Support latest `triton.compile` interface (#17913)
MasterJH5574 May 1, 2025
3f27aa8
[WebGPU][CodeGen] Override PrintVecElemLoad and Store for WebGPU (#17…
CharlieFRuan May 4, 2025
4ef582a
[Relax][PyTorch] Add support for linspace op in fx graph (#17915)
tlopex May 4, 2025
fa26a05
[Relax][PyTorch] Add Meshgrid Op Support for Exported Program and FX …
Deivanayaki-S May 6, 2025
95d1268
[REFACTOR] Introduce and modernize FFI system (#17920)
tqchen May 6, 2025
bcb68b1
[Relax][PyTorch] Add div.Tensor_mode and trunc Op Support for Exporte…
Deivanayaki-S May 7, 2025
16b2783
[REFACTOR][FFI] Cleanup PackedFunc related redirection (#17923)
tqchen May 7, 2025
da6d510
[FFI][FEAT] AutoDLPack for taking external tensor objects (#17927)
tqchen May 8, 2025
914590a
[Relax][PyTorch] Add tests for all the dtypes supported in the PyTorc…
mshr-h May 8, 2025
64cea4a
[Relax][PyTorch] Add MaxPool 1D and 3D Op Support for Exported Progra…
Deivanayaki-S May 8, 2025
d3e3d43
[Relax][PyTorch] CrossEntropyLoss (#17863)
hugolatendresse May 8, 2025
1f8103e
[REFACTOR][FFI] Cleanup container redirections (#17929)
tqchen May 8, 2025
bc896a5
[Relax][PyTorch] Add Adaptive AvgPool 1D and 3D Op Support for Export…
Deivanayaki-S May 9, 2025
7275cf0
[REFACTOR][FFI][RPC] Migrate RPC to use the latest FFI ABI (#17931)
tqchen May 9, 2025
9c77baf
[REFACTOR] Phase out legacy go ffi (#17940)
tqchen May 9, 2025
283fecb
[REFACTOR] Phase out legacy rust ffi (#17939)
tqchen May 10, 2025
b818183
[Relax][PyTorch] Add AvgPool 1D and 3D Op Support for Exported Progra…
Deivanayaki-S May 10, 2025
dcb5a3a
[Relax][PyTorch] Add UpSample Bicubic Op Support for Exported Program…
Deivanayaki-S May 10, 2025
f4704f2
[Relax][PyTorch] Add torch.outer Op Support for Exported Program and …
Deivanayaki-S May 10, 2025
8a6c9bf
[Relax][PyTorch] Add ReLU6 Op Support for Exported Program and FX gra…
Deivanayaki-S May 10, 2025
92ea12d
[FFI] Variant specialize for all ObjectRef (#17943)
tqchen May 10, 2025
8929d3f
[TOPI] Add shape validation to prevent negative dimensions in conv op…
jikechao May 11, 2025
d5cd945
[REFACTOR][FFI][Web] Upgrade Web Runtime to new FFI (#17946)
tqchen May 11, 2025
a54833d
[NODE] Fix structural equality for Array<Any> specialization (#17951)
tqchen May 12, 2025
46f41ce
Add registion for the operator asin and acos in llvm (#17945)
jikechao May 12, 2025
ae053e6
[FFI][LLVM] Fix compilation errors with clang20 (#17954)
cbalint13 May 12, 2025
beca091
[LLVM] Fix JIT unknown reloc issue for case of RISCV
cbalint13 May 12, 2025
70aa379
[Docker][CI] Reintroduce NNEF to CI images (#17955)
agoston-mc May 13, 2025
71f1ad0
[Fix][Relax] Fix dangling reference in GetTargetFunctions() (#17950)
Phoslight May 13, 2025
01268ac
[LLVM][Codegen] Enable SVE/VLA for RISCV targets
cbalint13 May 13, 2025
ab51a01
Update estimate_flops.cc
jikechao May 12, 2025
dc2e06a
Update estimate_flops.cc
jikechao May 13, 2025
561d085
Update estimate_flops.cc
jikechao May 13, 2025
b77f948
Fix sqrt/rsqrt Compatibility with Integer Data Types (#17953)
jikechao May 13, 2025
21a00f5
[Bugfix][Relax][Pytorch] Bugfix of conv_transpose1d and conv_transpos…
kavin-sai-krishna May 14, 2025
9aa5bf7
[FFI][REFACTOR] Update to distinguish as and cast (#17979)
tqchen May 15, 2025
f766c34
Fix g.costs (#17972)
thaisacs May 15, 2025
68dd534
Add registion for the operator asinh, acosh, atanh in llvm (#17969)
jikechao May 15, 2025
196b694
[FRONTEND][ONNX] Make bias input optional in LayerNormalization (#17980)
vacu9708 May 15, 2025
942d03c
[CI] Update images to `20250513-063354-70aa3797` (#17981)
mshr-h May 16, 2025
3db71bb
[FFI][JVM] Upgrade tvm4j to latest FFI (#17983)
tqchen May 18, 2025
9265328
[Relax][PyTorch] Re-enable test_subgraph_capture in dynamo test (#17925)
mshr-h May 20, 2025
ffa82b3
[Relax][Frontend]Fix: Output tensor with zero dimension after torch.u…
eaten-cake May 20, 2025
08f3365
Fix zero-extent loops in PerStoreFeature to prevent crashes (#17995)
jikechao May 21, 2025
2ab5192
[Vulkan] Add TIR unary trigonometric/hyperbolic intrinsic definitions…
fairywreath May 21, 2025
ada7c7c
[Relax][Frontend][ONNX] Fix: bitwise_not misclassified as binary (is …
eaten-cake May 22, 2025
2d964b4
[REFACTOR][FFI] Phase out legacy C API (#18010)
tqchen May 24, 2025
fbd9fe4
Fix division truncation in window size calculation for small dtypes i…
jikechao May 26, 2025
a2ab47b
Fix RuntimeError: parallel_for_dynamic (#18013)
thaisacs May 27, 2025
44fac51
[Relax][ONNX] Replace deprecated `mapping.TENSOR_TYPE_TO_NP_TYPE` usa…
mshr-h May 27, 2025
bcae402
Fix FLOP estimation for EvaluateNode by implementing VisitStmt_ handl…
jikechao May 28, 2025
29de9ab
Add op support for slice_scatter (#18019)
kavin-sai-krishna May 28, 2025
4289efa
[REFACTOR][PYTHON] Phase out tvm._ffi and Limited API support (#18020)
tqchen May 28, 2025
cac6ff8
[TOPI] Support integer type input for log10 (#18015)
jikechao May 29, 2025
19395ee
[REFACTOR][FFI] Cleanup PackedFunc redirections (#18022)
tqchen May 29, 2025
ac65e56
[FFI] More strict tuple constructor checking (#18023)
tqchen May 29, 2025
bf24cab
[Python] Fix library lookup path for pip installed packages (#18026)
MasterJH5574 May 31, 2025
731f133
[ARITH] Fix canonical simplify for LE with incorrect range assumption…
Jun 1, 2025
3ad4eb3
[CUDA] Fix thrust with latest FFI refactor (#18024)
MasterJH5574 Jun 1, 2025
55b8980
[TOPI] Fix index handling in expand_like operator for axis expansion …
jikechao Jun 1, 2025
eca92bd
Fix IR generation conflict in topi.nn.simplify by separating Tensor a…
jikechao Jun 2, 2025
1b6adcd
[ROCm] Fix ROCm build after FFI refactor (#18029)
MasterJH5574 Jun 2, 2025
8c9026d
[Codegen] Resolve issue #17965 where the same model produces differen…
vacu9708 Jun 2, 2025
0607484
[ARITH] Canonicalize mul-coefficient to rhs (#18031)
tqchen Jun 2, 2025
aa312a5
[BugFix][CUDA] Fix: Update settings for rerun on Increase FloatImm pr…
DerrickYLJ Jun 3, 2025
2dce84f
[Dtype] Low-precision Blackwell Datatype Support (#18027)
Kathryn-cat Jun 3, 2025
44eda85
[Metal] Fix `GetFunction` of metal runtime (#18034)
MasterJH5574 Jun 3, 2025
bb14b27
[CI] Further robustify is_last_build check (#18037)
tqchen Jun 4, 2025
b40b6fa
[REFACTOR][FFI] Update symbol name for library module (#18042)
tqchen Jun 5, 2025
76946b4
Resolving inconsistency between attention/attention_bias (#18038)
xinxilwl Jun 5, 2025
b2b7c62
[REFACTOR] Phase out the relax tuning_api (#18043)
tqchen Jun 6, 2025
aee96e6
[DTYPE] Fix dtype functions after dtype refactor (#18041)
tqchen Jun 6, 2025
fd9c091
[CUTLASS] Add GeMM kernels for Blackwell GPUs (#18033)
MasterJH5574 Jun 6, 2025
a00f8c5
[Backend] JIT compile FlashInfer kernel with FFI header (#18047)
MasterJH5574 Jun 6, 2025
74eae2f
[3rdparty] Phasing out FlashInfer AOT from 3rdparty (#18046)
MasterJH5574 Jun 6, 2025
61e7c8f
[Refactor] Rename `relax_vm` to `vm` (#18049)
MasterJH5574 Jun 6, 2025
9cb6705
[FFI] Enhance FFI Object exception safety during init (#18050)
tqchen Jun 8, 2025
fa46d7a
Add support for hamming_window op (#18036)
kavin-sai-krishna Jun 11, 2025
8cb23da
[DOCS] Update installation instruction based ffi refactor (#18056)
tqchen Jun 12, 2025
1cf31bc
[Pytest] Remove obsolete test suite entries (#18054)
mshr-h Jun 12, 2025
75f710f
[TIR] Phase out ProducerStore, ProducerRealize and Prefetch (#18057)
Jun 14, 2025
a187a79
[TEST] Move temp files into tempdir (#18058)
tqchen Jun 14, 2025
0f5bb20
[FFI][REFACTOR] Enhance reflection (#18059)
tqchen Jun 15, 2025
9a7c0dd
[FFI][REFACTOR] Update registry to have complete meta-data (#18062)
tqchen Jun 16, 2025
2ae2aca
Add Python functor support for TIR expressions and statements (#18060)
Jun 16, 2025
6c540e0
[CUTLASS] Fix CUTLASS kernel build on Hopper (#18064)
MasterJH5574 Jun 17, 2025
9dad95d
[TIR] Fix block access region detection for nested let bindings (#18069)
Jun 17, 2025
f09e61b
[TIR] Extend address_of to support Buffer objects (#18068)
Jun 17, 2025
43e7676
[Script] Remove deprecated attributes from Constant AST node (#18066)
Jun 17, 2025
3c7c515
[FFI] Introduce FFI reflection support in python (#18065)
tqchen Jun 17, 2025
e009599
[CI] Update windows to 2025 (#18071)
tqchen Jun 17, 2025
6de3f39
[NVSHMEM] Update NDArray allocation (#18073)
MasterJH5574 Jun 17, 2025
b5d352f
[TOPI][NN][Layer_Norm] Fix layer_norm error with reduce-only axes (#1…
eaten-cake Jun 18, 2025
2d63574
[ARITH] Add IsBound method to ConstIntBoundAnalyzer (#18067)
Jun 18, 2025
d988106
[FFI] Optimize atomic decref in Object (#18077)
tqchen Jun 18, 2025
10e6693
[REFACTOR] Phase out the RelaxExpr.checked_type in favor of struct_in…
tqchen Jun 18, 2025
a35d609
[FFI][REFACTOR] Stablize container ABI and implementation (#18076)
tqchen Jun 18, 2025
5a5ccd8
[REFACTOR] Phase out LegacyReprPrinter and improve CommonSubExprElim …
tqchen Jun 19, 2025
ce8875e
[FFI] Update typeinfo to speedup parent reflection (#18083)
tqchen Jun 19, 2025
c3281c0
[Script] Add support for merging block annotations (#18079)
Jun 20, 2025
6ff3efa
add support for softsign op (#18075)
logeshwaranmcw Jun 23, 2025
8327a8c
[Relax][ONNX] Update ReduceL1 to opset 18 (#18072)
vacu9708 Jun 23, 2025
a1f5a29
[Relax] Support InstanceNorm & [TOPI] Bugfix of InstanceNorm (#18039)
kavin-sai-krishna Jun 23, 2025
795fc32
[SCRIPT] Bump Python minimum version to 3.9 and update AST compatibil…
Jun 23, 2025
f6a406a
[Script] Enhance alloc buffer handling in nested frames (#18088)
Jun 23, 2025
23bcbc5
[KVCache] Per Layer Sliding Window (#17928)
joshua-j-hong Jun 23, 2025
437d00a
[Relax][ONNX] Update Reduce ops to support axes as input (#18090)
vacu9708 Jun 24, 2025
f77817b
TVM Patch for TileLang
Hzfengsy Jun 11, 2025
bcfd0af
[FFI] Provide Field Visit bridge so we can do gradual transition (#18…
tqchen Jun 25, 2025
a842004
[BugFix] Fix exception when tvm not built with llvm support (#18087)
ddwolf Jun 25, 2025
bf752dc
[Fix] Fix ExecBuilderDeclareFunction method name in exec_builder.py (…
ConvolutedDog Jun 25, 2025
a40f73f
[NVSHMEM] Extend CUDA backend to compile and link TIR modules with NV…
Kathryn-cat Jun 26, 2025
c5c733c
[FFI][REFACTOR] Migrate attrs to use new reflection (#18095)
tqchen Jun 27, 2025
f0e62eb
[REFACTOR] Transition VisitAttrs to new reflection mechanism in tir/i…
tqchen Jun 27, 2025
4d3cf08
[REFACTOR] Transition VisitAttrs to new reflection mechanism (#18098)
tqchen Jun 29, 2025
b6db2ec
[Runtime] CutensorMap support (#18097)
spectrometerHBH Jun 30, 2025
9eb8b30
Add support for bucketize (#18040)
kavin-sai-krishna Jun 30, 2025
910aeaf
[RELAX] Fix rotary embedding buffer size calculation (#18102)
harrywhoo Jul 1, 2025
83c0c4c
[Fix] Replace dmlc::Error with std::exception in VerifyGPUCode (#18103)
w1049 Jul 1, 2025
17113f8
[REFACTOR] Formalize namespace for all objects (#18101)
tqchen Jul 1, 2025
956b659
fix: guard tensormap with cuda version check (#18107)
spectrometerHBH Jul 2, 2025
1b9da40
[REFACTOR][FFI] Phase out old VisitAttrs mechanism (#18106)
tqchen Jul 3, 2025
6620fe2
Add LLVM Legalization for tir.erf (#18104)
Jul 3, 2025
ed4c92c
[FFI] Introduce GlobalDef for function registration (#18111)
tqchen Jul 4, 2025
51bf3d6
[FFI] Cleanup visit_attrs attribute after refactor (#18112)
tqchen Jul 4, 2025
2b6b9e4
[CMake] Refine C++/CUDA standard settings in CMakeLists.txt (#18113)
PragmaTwice Jul 4, 2025
9378575
[Fix][Serialization] Add support for NaN value serialization (#18115)
micricket Jul 4, 2025
a4dc883
[FFI] Replace `__attribute__` with C++ standard attributes (#18114)
PragmaTwice Jul 4, 2025
8e478f5
[FFI] Use fold expression to simplify for_each (#18116)
PragmaTwice Jul 5, 2025
351dacf
bump cutlass_fpA_intB_gemm (#18118)
yongwww Jul 6, 2025
49b139b
[FFI] Replace `Arg2Str` with a more powerful `for_each` (#18117)
PragmaTwice Jul 6, 2025
11f8606
Revert "[FFI] Replace `Arg2Str` with a more powerful `for_each`" (#18…
tqchen Jul 7, 2025
68bb125
[release] Update version to 0.21.0 on main branch
ysh329 Jul 2, 2025
045eb5b
[release] Update version to 0.22.dev0 on main branch
ysh329 Jul 2, 2025
351dd80
[Codegen] Update LLVM version requirement for `insertDeclare` (#18123)
MasterJH5574 Jul 7, 2025
d9f0838
[KVCache] Fix kernel dispatch based on attention kinds (#18122)
MasterJH5574 Jul 7, 2025
3427445
Update CMakeLists.txt to include Python include directory and clean u…
LeiWang1999 Jul 8, 2025
79368ce
[Relax][ONNX][Transform] Add mode choice, new mode, and warning for t…
vacu9708 Jul 8, 2025
c6969d7
[Refactor] Build cython with isolate environment (#18124)
LeiWang1999 Jul 8, 2025
d230129
phaseout ck dependency
LeiWang1999 Jul 8, 2025
2139f47
phaseout flashinfer
LeiWang1999 Jul 8, 2025
9249de3
phase out vta
LeiWang1999 Jul 8, 2025
39d113b
support T.address_of(B[i, j])
LeiWang1999 Jul 9, 2025
6fce4b7
Phase out StackVM runtime support (#18125)
Hzfengsy Jul 9, 2025
f7b57ec
Revert "[Refactor] Build cython with isolate environment" (#18127)
tqchen Jul 9, 2025
458b0ab
[Doc] Visualize the architecture using a UML sequence diagram (#18128)
vacu9708 Jul 10, 2025
acdc164
[Target] Support CUDA device function calls (#18055)
Hzfengsy Jul 11, 2025
c3d3596
Delete redundant imports (#18129)
mlbo Jul 11, 2025
27a314e
[MISC] Fix compilation warnings of unnecessary `std::move()` calls (#…
Hzfengsy Jul 11, 2025
99f5e66
[REFACTOR] Migrate TVM_FFI_REGISTER_GLOBAL to new reflection style (#…
tqchen Jul 13, 2025
6ecf8bd
[FFI][PYTHON] Improve the traceback generation in python (#18141)
tqchen Jul 13, 2025
a519517
[Fix][ONNX] Fix CumSum conversion when loading ONNX model (#18137)
ConvolutedDog Jul 13, 2025
a61e3ca
[Fix][ONNX] Fixed constant ROI handling in resize2d when loading onnx…
ConvolutedDog Jul 14, 2025
359e05e
[FRONTEND][ONNX] Extend axes for layer_norm when gamma/beta are multi…
ConvolutedDog Jul 14, 2025
9aa1dca
[Test] Use roi_list variable instead of hardcoded values in ROI tenso…
ConvolutedDog Jul 14, 2025
a3194e5
[CodeGen][CUDA] Add sinhf CUDA Math API for CodeGen (#18144)
ConvolutedDog Jul 14, 2025
11f0e42
[BugFix][NNAPI] Fix type mismatch and test_mean annotation (#18140)
Henryshsieh Jul 14, 2025
ea4369c
[TIR] Add `T.thread_return()` for early thread exit in CUDA kernels (…
Hzfengsy Jul 14, 2025
36fb02c
[FFI][REFACTOR] Modularize refelection (#18147)
tqchen Jul 15, 2025
b4a6b73
[FFI][REFACTOR] Phase out TVM_FFI_REGISTER_GLOBAL in favor of GlobalD…
tqchen Jul 15, 2025
7707496
[FFI] Log and throw in function dup registration (#18149)
tqchen Jul 16, 2025
3c72b8f
Fix CMakeLists.txt to remove unnecessary '-I' flag from Python build …
LeiWang1999 Jul 16, 2025
1a1d27c
[Misc] Fix Release Package Test Script (#18153)
ysh329 Jul 17, 2025
89f9573
[TIR] Decouple DeepEqual from StructuralEqual (#18151)
tqchen Jul 17, 2025
e05064f
[FFI] Structural equal and hash based on reflection (#18156)
tqchen Jul 19, 2025
5e12a5c
[Fix][Relax] Fix potential out-of-bounds access in `TupleRewriterNode…
Phoslight Jul 19, 2025
5aa4dfd
[BugFix] Fix NCCL build with GlobalDef registration (#18158)
MasterJH5574 Jul 21, 2025
d4e7bd3
[FFI][REFACTOR] Introduce TypeAttr in reflection (#18160)
tqchen Jul 22, 2025
ce08d9c
Merge branch 'main' of https://github.com/apache/tvm into upstream-dev
LeiWang1999 Jul 24, 2025
9611cc7
c api fix
LeiWang1999 Jul 24, 2025
493f937
[FFI] Remove unused Grid constant and add HANDLE_TO_REFERENCE conversion
LeiWang1999 Jul 24, 2025
9a00cd6
preserve unit loop for reindex scheduling.
LeiWang1999 Jan 20, 2024
fc29e7b
Add skip_simplify option to reindex method for improved index handling
LeiWang1999 Jul 28, 2025
5cc56c9
fix
LeiWang1999 Jul 28, 2025
763f196
Update LetFrameNode to allow mutable value and register reflection ac…
LeiWang1999 Jul 28, 2025
ab733d1
Refactor argument extraction in ExprEvaluator to streamline handling …
LeiWang1999 Jul 22, 2025
ccc68f5
Enhance error reporting in IndexMapInverseImpl by including index map…
LeiWang1999 May 29, 2025
555cc71
Remove redundant type check in Allocate constructor for improved clar…
LeiWang1999 May 21, 2025
d39953f
Change annotations type in Allocate constructor from Map<String, Obje…
LeiWang1999 Jul 29, 2025
9574805
Update minimum Python version requirement from 3.9 to 3.8 for compati…
LeiWang1999 Jul 29, 2025
a08b7c3
Revert "Update minimum Python version requirement from 3.9 to 3.8 for…
LeiWang1999 Jul 29, 2025
cb0fd6d
Refactor stride naming in Namer to use name_hint when defined, improv…
LeiWang1999 Aug 11, 2025
e11521e
Refactor MergeAnnotations function to accept Map<Any, Any> instead of…
LeiWang1999 Aug 12, 2025
e5558ac
Merge branch 'tilelang_main' of https://github.com/TileLang/tvm into …
LeiWang1999 Aug 12, 2025
5a433cc
phaseout legacy components
LeiWang1999 Aug 12, 2025
a64a592
Add support for 'tir.exp2' operation and register 'hip' target kind w…
Alex4210987 Aug 12, 2025
ec79331
Add tilelang assume attribute to support custom assumption
kurisu6912 Sep 5, 2025
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
The diff you're trying to view is too large. We only load the first 3000 changed files.
4 changes: 0 additions & 4 deletions .asf.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -58,17 +58,13 @@ github:
contexts:
- unity/pr-head
- arm/pr-head
- cortexm/pr-head
- cpu/pr-head
- docker/pr-head
- gpu/pr-head
- hexagon/pr-head
- i386/pr-head
- lint/pr-head
- minimal/pr-head
- riscv/pr-head
- wasm/pr-head
- cross-isa-minimal/pr-head

required_pull_request_reviews:
required_approving_review_count: 1
7 changes: 0 additions & 7 deletions .github/CODEOWNERSHIP
Original file line number Diff line number Diff line change
Expand Up @@ -132,13 +132,6 @@ src/runtime/** @vinx13 @tqchen @FronzenGene @liangfu @areusch @tmoreau89 @ajtul
include/tvm/runtime/** @vinx13 @tqchen @FronzenGene @liangfu @areusch @tmoreau89 @ajtulloch @masahi @kazum @ZihengJiang @junrushao1994
python/tvm/runtime/** @vinx13 @tqchen @FronzenGene @liangfu @areusch @tmoreau89 @ajtulloch @masahi @kazum @ZihengJiang @junrushao1994

# runtime/micro
src/runtime/micro/** @areusch @liangfu @tmoreau89 @manupa-arm
src/runtime/crt/** @areusch @liangfu @tmoreau89 @manupa-arm
include/tvm/runtime/crt/** @areusch @liangfu @tmoreau89 @manupa-arm
include/tvm/runtime/micro/** @areusch @liangfu @tmoreau89 @manupa-arm
python/tvm/micro/** @areusch @liangfu @tmoreau89 @manupa-arm

# relay
src/relay/** @jroesch @slyubomirsky @icemelon @MarisaKirisame @ZihengJiang @yzhliu @vinx13 @mbrookhart @jwfromm @zhiics @anijain2305 @wweic @eqy @junrushao1994
include/tvm/relay/** @jroesch @slyubomirsky @icemelon @MarisaKirisame @ZihengJiang @yzhliu @vinx13 @mbrookhart @jwfromm @zhiics @anijain2305 @wweic @eqy @junrushao1994
Expand Down
21 changes: 12 additions & 9 deletions .github/actions/setup/action.yml
Original file line number Diff line number Diff line change
@@ -1,35 +1,38 @@
runs:
using: "composite"
steps:
- uses: actions/cache@v1
- uses: actions/cache@v3
env:
CACHE_NUMBER: 0
CACHE_NUMBER: 2
with:
path: ~/conda_pkgs_dir
key: ${{ runner.os }}-conda-${{ env.CACHE_NUMBER }}-${{ hashFiles('conda/build-environment.yaml') }}
- uses: conda-incubator/setup-miniconda@v2
- uses: conda-incubator/setup-miniconda@v3
continue-on-error: true
id: conda1
with:
activate-environment: tvm-build
channel-priority: strict
environment-file: conda/build-environment.yaml
auto-activate-base: false
use-only-tar-bz2: true
python-version: 3.7
miniforge-version: latest
python-version: 3.9
condarc-file: conda/condarc
- uses: conda-incubator/setup-miniconda@v2
- uses: conda-incubator/setup-miniconda@v3
if: steps.conda1.outcome == 'failure'
with:
activate-environment: tvm-build
channel-priority: strict
environment-file: conda/build-environment.yaml
auto-activate-base: false
miniforge-version: latest
use-only-tar-bz2: true
python-version: 3.7
python-version: 3.9
condarc-file: conda/condarc
- name: Conda info
shell: pwsh
run: |
conda info
conda list
mamba info
mamba list
mamba info --envs
mamba list --name base
170 changes: 63 additions & 107 deletions .github/workflows/main.yml
Original file line number Diff line number Diff line change
Expand Up @@ -49,22 +49,22 @@ jobs:
run: >-
conda build --output-folder=conda/pkg conda/recipe &&
conda install tvm -c ./conda/pkg
- name: Build iOS RPC
run: |
IOS_VERSION="14.0"
CMAKE_FLAGS="-DCMAKE_BUILD_TYPE=Release \
-DCMAKE_SYSTEM_NAME=iOS \
-DCMAKE_SYSTEM_VERSION=${IOS_VERSION} \
-DCMAKE_OSX_SYSROOT=iphonesimulator \
-DCMAKE_OSX_ARCHITECTURES=x86_64 \
-DCMAKE_OSX_DEPLOYMENT_TARGET=14.0 \
-DCMAKE_BUILD_WITH_INSTALL_NAME_DIR=ON \
-DUSE_IOS_RPC=ON"

mkdir build-ios-simulator
cd build-ios-simulator
cmake .. ${CMAKE_FLAGS}
cmake --build . --target ios_rpc
# - name: Build iOS RPC
# run: |
# IOS_VERSION="14.0"
# CMAKE_FLAGS="-DCMAKE_BUILD_TYPE=Release \
# -DCMAKE_SYSTEM_NAME=iOS \
# -DCMAKE_SYSTEM_VERSION=${IOS_VERSION} \
# -DCMAKE_OSX_SYSROOT=iphonesimulator \
# -DCMAKE_OSX_ARCHITECTURES=x86_64 \
# -DCMAKE_OSX_DEPLOYMENT_TARGET=14.0 \
# -DCMAKE_BUILD_WITH_INSTALL_NAME_DIR=ON \
# -DUSE_IOS_RPC=ON"
#
# mkdir build-ios-simulator
# cd build-ios-simulator
# cmake .. ${CMAKE_FLAGS}
# cmake --build . --target ios_rpc
- name: Test
shell: bash -l {0}
run: >-
Expand All @@ -77,19 +77,21 @@ jobs:
- name: Minimal Metal Compile-and-Run
shell: bash -l {0}
run: >-
python -m pytest -v -s 'tests/python/codegen/test_target_codegen_metal.py'
python -m pytest -v -s 'tests/python/codegen/test_target_codegen_gpu_common.py'
python -m pytest -v -s 'tests/python/codegen/test_gpu_codegen_allreduce.py::test_allreduce_sum[dims0-metal]'
- name: Test iOS RPC
shell: bash -l {0}
run: >-
python -m pip install tornado psutil cloudpickle &&
export PYTHONPATH=tests/python/contrib:${PYTHONPATH} &&
export BUNDLE_ID=org.apache.tvmrpc &&
export BUNDLE_PATH=build-ios-simulator/apps/ios_rpc/ios_rpc/src/ios_rpc-build/Release-iphonesimulator/tvmrpc.app &&
python -m pytest -v tests/python/contrib/test_rpc_server_device.py
# - name: Test iOS RPC
# shell: bash -l {0}
# run: >-
# python -m pip install tornado psutil cloudpickle &&
# export PYTHONPATH=tests/python/contrib:${PYTHONPATH} &&
# export BUNDLE_ID=org.apache.tvmrpc &&
# export BUNDLE_PATH=build-ios-simulator/apps/ios_rpc/ios_rpc/src/ios_rpc-build/Release-iphonesimulator/tvmrpc.app &&
# python -m pytest -v tests/python/contrib/test_rpc_server_device.py

Windows:
if: ${{ github.repository == 'apache/tvm' }}
runs-on: windows-2019
runs-on: windows-latest
steps:
- uses: actions/checkout@v2
with:
Expand All @@ -106,85 +108,39 @@ jobs:
run: >-
python -m pytest -v tests/python/all-platform-minimal-test

# Disabled due to https://github.com/apache/tvm/issues/13950
# Windows-Static-Runtime:
# if: ${{ github.repository == 'apache/tvm' }}
# runs-on: windows-2019
# steps:
# - uses: actions/checkout@v2
# with:
# submodules: 'recursive'
# - name: Set up environment
# uses: ./.github/actions/setup
# - name: Build static TVM runtime
# shell: bash -l {0}
# run: |
# tests/scripts/task_config_build_static.sh build
# cd build
# cmake .. -A x64 -DCMAKE_CONFIGURATION_TYPES="Release"
# cmake --build . --config Release --target runtime

Linux-Static-Runtime:
if: ${{ github.repository == 'apache/tvm' }}
runs-on: Ubuntu-20.04
steps:
- uses: actions/checkout@v2
with:
submodules: 'recursive'
- name: Set up environment
uses: ./.github/actions/setup
- name: Build static TVM runtime
shell: bash -l {0}
run: |
tests/scripts/task_config_build_static.sh build
cd build
cmake ..
cmake --build . --config Release --target runtime

Android:
if: ${{ github.repository == 'apache/tvm' }}
runs-on: Ubuntu-20.04
steps:
- uses: actions/checkout@v2
with:
submodules: 'recursive'
- name: Set up environment
uses: ./.github/actions/setup
- name: Set up java
uses: actions/setup-java@v3
with:
distribution: 'zulu'
java-version: '11'
- name: Build TVM
shell: bash -l {0}
run: |
mkdir build
cd build
../tests/scripts/task_config_build_jvm.sh .
cmake ..
make
- name: Build TVM4J
run: |
make jvmpkg
- name: Build android_rpc
working-directory: apps/android_rpc
run: |
set -eux
export PATH="${ANDROID_NDK_LATEST_HOME}:$PATH"
gradle clean build
- name: Upload android_rpc APK
uses: actions/upload-artifact@v2
with:
name: android_rpc-debug.apk
path: ./apps/android_rpc/app/build/outputs/apk/debug/app-debug.apk
- name: Build android_deploy
working-directory: apps/android_deploy
run: |
set -eux
export PATH="${ANDROID_NDK_LATEST_HOME}:$PATH"
gradle clean build
- name: Upload android_deploy APK
uses: actions/upload-artifact@v2
with:
name: android_deploy-debug.apk
path: ./apps/android_deploy/app/build/outputs/apk/debug/app-debug.apk
# Android:
# if: ${{ github.repository == 'apache/tvm' }}
# runs-on: ubuntu-22.04
# steps:
# - uses: actions/checkout@v2
# with:
# submodules: 'recursive'
# - name: Set up environment
# uses: ./.github/actions/setup
# - name: Set up java
# uses: actions/setup-java@v3
# with:
# distribution: 'zulu'
# java-version: '11'
# - name: Build TVM
# shell: bash -l {0}
# run: |
# mkdir build
# cd build
# ../tests/scripts/task_config_build_jvm.sh .
# cmake ..
# make
# - name: Build TVM4J
# run: |
# make jvmpkg
# - name: Build android_rpc
# working-directory: apps/android_rpc
# run: |
# set -eux
# export PATH="${ANDROID_NDK_LATEST_HOME}:$PATH"
# gradle clean build
# - name: Upload android_rpc APK
# uses: actions/upload-artifact@v4
# with:
# name: android_rpc-debug.apk
# path: ./apps/android_rpc/app/build/outputs/apk/debug/app-debug.apk
2 changes: 1 addition & 1 deletion .github/workflows/nightly_docker_update.yml
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,7 @@ jobs:
pull-requests: write
statuses: write
if: github.repository == 'apache/tvm'
runs-on: ubuntu-20.04
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v2
- name: Open PR to update Docker images
Expand Down
2 changes: 1 addition & 1 deletion .github/workflows/ping_reviewers.yml
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@ concurrency:
jobs:
ping:
if: github.repository == 'apache/tvm'
runs-on: ubuntu-20.04
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v2
- name: Ping reviewers
Expand Down
2 changes: 1 addition & 1 deletion .github/workflows/tvmbot.yml
Original file line number Diff line number Diff line change
Expand Up @@ -21,7 +21,7 @@ jobs:
pull-requests: write
statuses: write
if: ${{ github.event.issue.pull_request && github.repository == 'apache/tvm' }}
runs-on: ubuntu-20.04
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v2
- name: Run tvm-bot
Expand Down
2 changes: 1 addition & 1 deletion .github/workflows/update_last_successful_branch.yml
Original file line number Diff line number Diff line change
Expand Up @@ -33,7 +33,7 @@ concurrency:
jobs:
update-last-successful-branch:
if: github.repository == 'apache/tvm'
runs-on: ubuntu-20.04
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v2
- name: Update last-successful branch
Expand Down
2 changes: 1 addition & 1 deletion .github/workflows/update_nightly_branch.yml
Original file line number Diff line number Diff line change
Expand Up @@ -31,7 +31,7 @@ concurrency:
jobs:
update-nightly-branch:
if: github.repository == 'apache/tvm'
runs-on: ubuntu-20.04
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v2
- name: Update nightly branch
Expand Down
2 changes: 1 addition & 1 deletion .github/workflows/upload_ci_resource.yml
Original file line number Diff line number Diff line change
Expand Up @@ -39,7 +39,7 @@ concurrency:
jobs:
upload-ci-resource:
if: github.repository == 'apache/tvm'
runs-on: ubuntu-20.04
runs-on: ubuntu-latest
steps:
- name: Download item and upload to S3
env:
Expand Down
3 changes: 0 additions & 3 deletions .gitignore
Original file line number Diff line number Diff line change
Expand Up @@ -263,9 +263,6 @@ src/runtime/hexagon/rpc/hexagon_rpc_stub.c
# Local tvm-site checkout
tvm-site/

# Generated docs files
gallery/how_to/work_with_microtvm/micro_tvmc.py

# Test sample data files
!tests/python/ci/sample_prs/*.json

Expand Down
12 changes: 6 additions & 6 deletions .gitmodules
Original file line number Diff line number Diff line change
Expand Up @@ -7,9 +7,6 @@
[submodule "3rdparty/rang"]
path = 3rdparty/rang
url = https://github.com/agauniyal/rang.git
[submodule "3rdparty/vta-hw"]
path = 3rdparty/vta-hw
url = https://github.com/apache/tvm-vta.git
[submodule "3rdparty/libbacktrace"]
path = 3rdparty/libbacktrace
url = https://github.com/tlc-pack/libbacktrace.git
Expand All @@ -28,6 +25,9 @@
[submodule "3rdparty/libflash_attn"]
path = 3rdparty/libflash_attn
url = https://github.com/tlc-pack/libflash_attn
[submodule "3rdparty/flashinfer"]
path = 3rdparty/flashinfer
url = https://github.com/flashinfer-ai/flashinfer.git
[submodule "3rdparty/zlib"]
path = 3rdparty/zlib
url = https://github.com/madler/zlib.git
[submodule "ffi/3rdparty/dlpack"]
path = ffi/3rdparty/dlpack
url = https://github.com/dmlc/dlpack.git
Loading