Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
793 commits
Select commit Hold shift + click to select a range
54c3fb4
[Relax][PyTorch] Handle unknown output shapes for _sym_size_int (#18521)
guan404ming Nov 30, 2025
d6e6334
[Bugfix] Prevent segfault when instantiating abstract SearchStrategy …
Dayuxiaoshui Dec 1, 2025
ec0026e
[Relax][PyTorch] Fix index_put with broadcast indices (#18533)
guan404ming Dec 1, 2025
cd820b5
[ARITH] Fix InternalError: Check failed: (eval_vec_) is false (#18536)
cchung100m Dec 3, 2025
ed97234
Revert "[ARITH] Fix InternalError: Check failed: (eval_vec_) is false…
tlopex Dec 4, 2025
c71aefc
[Docs] Fix e2e_opt_model tutorial for GPU deployment (#18539)
tlopex Dec 4, 2025
001ed57
[Schedule] Fix LocalBuilder Check failed: (index_map_func.has_value()…
Asuka0630 Dec 4, 2025
141431c
[TIR][Schedule] Fix mma tensorize error (#18528)
Asuka0630 Dec 4, 2025
a747614
[Relax][PyTroch] Add NHWC layout support (#18548)
guan404ming Dec 5, 2025
4c2249d
[CI] Remove hardcoded user and repo values (#18549)
guan404ming Dec 5, 2025
5138efc
[Relax][PyTorch] Unify dtype used in conv2d tests (#18553)
guan404ming Dec 6, 2025
6cf49e6
[Relax][PyTorch] Enhance scale_factor handling in interpolation (#18550)
guan404ming Dec 6, 2025
0d9d178
[MISC] Fix duplicate `PresburgerSetNode` registration when `USE_MLIR=…
mshr-h Dec 6, 2025
dcecb86
[Docs] Improve static shape tuning parameter configuration (follow-up…
ConvolutedDog Dec 6, 2025
2b4a1e2
[Relax][Frontend] Introduce ModuleDict (#18551)
mshr-h Dec 7, 2025
26b107f
[Relax][PyTorch] Add support for masked_select (#18535)
guan404ming Dec 7, 2025
e78fbd8
[Attn] Fix calling FlashInfer attention plan function (#18557)
MasterJH5574 Dec 8, 2025
bddc091
[TIR][Schedule] Fix bug on bfloat16 conversion (#18556)
grf53 Dec 8, 2025
7271feb
[Relax][PyTorch] Add support for Custom Ops for ExportedProgram front…
mshr-h Dec 9, 2025
8218b18
[Relax] Add mod operator support (#18559)
guan404ming Dec 9, 2025
04f06b5
[Relax] Add edge padding mode (#18558)
guan404ming Dec 9, 2025
85a8770
[Relax] Enhance unique block name generation with numeric suffixes (#…
guan404ming Dec 10, 2025
6248b5d
[Relax][Torch] Fixed issues related to sum op when without dim and ke…
locnd182644 Dec 13, 2025
f2930d5
[LLVM][Codegen] Avoid segfault when `arith::GetVScaleValues` returns …
mshr-h Dec 16, 2025
d375f74
Fix ACOS precision issue for boundary values (x=±1.0) (#18582)
Dayuxiaoshui Dec 17, 2025
5989ef5
[BugFix][OpenCL] Guard QCOM perf hint behind USE_OPENCL_EXTN_QCOM to …
ping-ee Dec 17, 2025
44d973b
[Relax] Add layout inference support for repeat operator (#18579)
guan404ming Dec 18, 2025
1c209e2
[Relax] Clean up scatter_elements unknown dtype handling (#18577)
guan404ming Dec 20, 2025
f4e28d3
[Relax] Chore: Fix the DeprecationWarning: invalid escape sequence \ …
cchung100m Dec 21, 2025
69ccf8b
[Relax][Torch] AssertionError: Unsupported function types ['mean.defa…
locnd182644 Dec 22, 2025
ec0fed0
[Runtime][MatchShape] Type error: Cannot convert from type ' DLTensor…
locnd182644 Dec 24, 2025
8521c2f
[TIR][Schedule] FuseReductionEpilogue: Add Clipping pattern support (…
kimm240 Dec 25, 2025
7ed4d39
[MetaSchedule] Fix tune_tir crash with ScheduleError in RewriteParall…
Dayuxiaoshui Dec 25, 2025
5dc4e78
[Relax] Fix batch normalization computation logic (#18609)
Asuka0630 Dec 26, 2025
5015047
[Relax] Fix flaky test_conv2d gradient numeric test (#18620)
guan404ming Dec 27, 2025
cfadc1a
[CUDA][FFI] Extend kernel launch config to support Programmatic Depen…
w169q169 Dec 27, 2025
3765df7
[Relax][ONNX] Add edge padding mode (#18575)
guan404ming Dec 28, 2025
7db7a91
[Relax][PyTorch] Fix PyTorch Dynamo frontend for Darwin compatibility…
guan404ming Dec 28, 2025
41cac2e
[Relax] Add test case for op attributes in AST printer (#18618)
guan404ming Dec 28, 2025
45bef45
[CUDA] Fix cuModuleUnload crash during interpreter shutdown (#18624)
guan404ming Dec 29, 2025
c404868
[Relax]feat: Implement FRelaxInferLayout for tile operator (#18593)
Dayuxiaoshui Dec 30, 2025
0d017c1
[Relax] Replaced call_pure_packed with tensor_to_shape operator (#18616)
guan404ming Dec 30, 2025
5abc5c7
[Relax] Remove duplicated test case: test_if_branch_var_scope (#18613)
cchung100m Dec 31, 2025
d30e347
[Relax] Use weight shape instead of dim in Embedding.forward (#18621)
guan404ming Dec 31, 2025
ee09866
[Relax] Add gpu-generic fallback for unrecognized GPU targets (#18615)
guan404ming Dec 31, 2025
7edf17b
[Fix] Fix typo in file header comment (#18628)
Henryshsieh Dec 31, 2025
e3f5ac1
[Relax] Correct YaRN RoPE frequency scaling formula to align with the…
Liberatedwinner Dec 31, 2025
899556d
[Relax][Op][PyTorch] Supported Median operator (#18626)
locnd182644 Jan 2, 2026
a393b47
[Relax] Add FInferMixedPrecision and FRelaxInferLayout for conv trans…
guan404ming Jan 3, 2026
7b9d3d9
[Relax] Move GetUsedVars to analysis module (#18632)
guan404ming Jan 5, 2026
e46c061
[Relax][Onnx] Pass output_padding param in ConvTranspose (#18635)
locnd182644 Jan 5, 2026
304a74a
[Relax] Add FRelaxInferLayout and TMixedPrecisionPolicy for dynamic_s…
guan404ming Jan 5, 2026
a95b90c
[Relax] Add FRelaxInferLayout for flip operator (#18637)
guan404ming Jan 6, 2026
0c7adc2
[Relax] Add FRelaxInferLayout for scatter_elements operator (#18638)
guan404ming Jan 6, 2026
b50ee5f
[TIR][Schedule] Fix type checker to support subscripted generics in P…
khj809 Jan 6, 2026
06a7cda
[Relax][Op] Fixed incorrect output shape of Pool op when ceil_mode = …
locnd182644 Jan 6, 2026
b975db9
[Relax] Add FRelaxInferLayout for scatter_nd operator (#18643)
guan404ming Jan 7, 2026
b3b6024
[Relax] Add FRelaxInferLayout for gather_elements operator (#18642)
guan404ming Jan 7, 2026
fa905d2
[Compile] accelerate compilation speed using NVRTC (#18519)
Kathryn-cat Jan 8, 2026
8e40211
[ADRENO][TEXTURE] Texture based lowering (#18523)
srkreddy1238 Jan 9, 2026
99c1d92
[Relax] Remove obsolete TODO comments (#18649)
guan404ming Jan 11, 2026
fa51ea2
Put options before objects when compiling (#18656)
oraluben Jan 11, 2026
038b327
[Relax][Onnx][PReLU] Handle slope and axis argument with different sl…
locnd182644 Jan 12, 2026
eb4eb3e
[FFI] bump to latest version (#18654)
tqchen Jan 14, 2026
2b4b58d
[Relax][Onnx][Resize] Handle non-4D input tensors (#18666)
LudovicoYIN Jan 16, 2026
c866abc
[Relax] Fix wrong memory planning when only lower bound was provided …
mshr-h Jan 16, 2026
fbc7f1e
[Relax] Replace topi.take with relax.op.take (#18665)
guan404ming Jan 16, 2026
361d21b
[Relax][ONNX] add support for unique optional outputs (#18652)
guan404ming Jan 20, 2026
26c6b13
fix: MSVC pragma (#18674)
junrushao Jan 20, 2026
d8c973e
[RELAX][LAYOUT] Support for dynamic layout specification (#18675)
srkreddy1238 Jan 20, 2026
fed71ef
[Relax] Add native size operator (#18667)
guan404ming Jan 20, 2026
15ac9db
[Relax][Onnx] Support Local Response Normalization (LRN) (#18668)
locnd182644 Jan 20, 2026
4533d31
[Relax] Add FDataDependent operator attribute for LegalizeOps (#18664)
guan404ming Jan 21, 2026
78b5ed0
[Relax] Implement dynamic output trimming for NMS (#18676)
guan404ming Jan 22, 2026
7493164
[Web] Fix incorrect FFI export name in runtime.ts (#18686)
guan404ming Jan 23, 2026
74adf2d
[Web] Fix RPC argument parsing for new FFI string/bytes types (#18683)
guan404ming Jan 23, 2026
2004a8b
[NVRTC] Add NVSHMEM support to NVRTC compilation path (#18681)
Kathryn-cat Jan 24, 2026
75ce1b4
[release] Update version to 0.23.0 on main branch
ysh329 Jan 17, 2026
87f478e
[release] Update version to 0.24.dev0 on main branch
ysh329 Jan 17, 2026
66f7f37
[Web] Handle LocalSession init in WASM RPC server (#18687)
guan404ming Jan 27, 2026
ab25b49
[TIR] Fix InjectPTXLDG32 segfaults and skip non-CUDA targets (#18671)
LudovicoYIN Jan 27, 2026
877b448
[REFACTOR][TIR] Rename tir.Block to SBlock (#18689)
tqchen Jan 28, 2026
95f41d6
[Web][Version] Fix WebLLM vision model issues (#18680)
akaashrp Jan 28, 2026
c043e4f
[Relax] Add NN operator attributes include to TensorRT codegen (#18690)
guan404ming Jan 28, 2026
71ac3e8
[Relax][NN] Add batch_flatten operator (#18677)
guan404ming Jan 29, 2026
701a753
[Relax][Onnx] Support Multi Input Ops with Multidirectional Broadcast…
locnd182644 Jan 29, 2026
4fc4450
[FIX] Fix T.sblock due to concurrent merge (#18695)
tqchen Jan 29, 2026
b4dc8a5
[BugFix][TIR] Fix incorrect optimization when lowering floordiv and f…
guocju Jan 30, 2026
6be0db4
[Relax][Onnx][Resize] Fix ROI values when tensor ROI is Empty pass no…
locnd182644 Jan 30, 2026
4b1bd6d
[TIR] Further robustify floordiv/mod intrin lowering to prevent overf…
tqchen Jan 30, 2026
4747a92
[Python][Relax] Fix YaRN correction dim calculation (#18661)
akaashrp Jan 30, 2026
4b2b639
[TIR][Schedule]Generalize fuseReductionEpilogue to support arbitrary …
kimm240 Feb 2, 2026
c575973
[Relax][Onnx][BatchNorm] Pass momentum and training_mode into BatchNo…
locnd182644 Feb 2, 2026
8f40544
[LLVM] Fix insertDbgValueIntrinsic for Metal backend (#18706)
MasterJH5574 Feb 2, 2026
efdf2b2
[CI] Fix NameError in nightly docker update workflow (#18692)
guan404ming Feb 3, 2026
0079ff7
[REFACTOR][TEST] Replace CompareBeforeAfter for pytest compact (#18711)
tqchen Feb 4, 2026
d76c729
[REFACTOR][S-TIR] Initialize the s_tir module (#18712)
tqchen Feb 5, 2026
dcad737
[Relax][Torch] Avoid decomposition crash with sparse CSR buffers (#18…
LudovicoYIN Feb 5, 2026
739027d
[SPIRV] Fix forloop codegen in vulkan (#18715)
tqchen Feb 5, 2026
87c1e47
[REFACTOR] Migrate old tir.ir_builder to tvmscript or builder (#18716)
tqchen Feb 6, 2026
c6e8b02
[CI] Fix double newlines in nightly docker update (#18717)
guan404ming Feb 6, 2026
057aa10
[DOCS] Clarify trusted usage (#18718)
tqchen Feb 6, 2026
176c773
[ci][docker] Nightly Docker image update (#18710)
github-actions[bot] Feb 6, 2026
198df47
[REFACTOR][TEST] Migrate all codegen test to tvmscript (#18719)
tqchen Feb 7, 2026
49b6d7b
Remove cron schedule from nightly Docker update workflow (#18724)
tqchen Feb 7, 2026
3dce4ae
[REFACTOR][S-TIR] Lift transform passes to s_tir namespace (#18722)
tqchen Feb 7, 2026
f7948c5
[REFACTOR][S-TIR] Migrate tir/schedule to s_tir (#18728)
tqchen Feb 7, 2026
d23d1db
fix: correct typos in Python docstrings (#18727)
thecaptain789 Feb 7, 2026
9eac0e1
[Relax][PyTorch] Fix scalar parameter inputs in Dynamo (#18725)
mshr-h Feb 7, 2026
6f3f691
[Relax][PyTorch] Simplify tensor args conversion in Dynamo (#18726)
mshr-h Feb 8, 2026
17074cd
[DOC] Fix grammar (#18731)
ruslo Feb 8, 2026
73a188f
Add Windows-specific build notes to installation guide (#18705)
AshwiniBokka Feb 8, 2026
2e34d8b
[REFACTOR][S-TIR] Migrate meta_schedule into s_tir namespace (#18735)
tqchen Feb 8, 2026
adda179
[REFACTOR][S-TIR] Lift dlight into s_tir namespace (#18734)
tqchen Feb 9, 2026
c414e4c
[chore][TIR] reorganize src/tir/transforms to src/tir/transform (#18738)
tqchen Feb 9, 2026
061b348
fix: correct typo 'occuring' to 'occurring' (#18740)
thecaptain789 Feb 9, 2026
833212d
fix: skip dsymutil for static tvm_runtime on Apple platforms (#18744)
MasterJH5574 Feb 9, 2026
90f2732
fix: correct typos 'recieve' and 'occurence' (#18739)
thecaptain789 Feb 10, 2026
88a9e91
[Fix] Handle empty variable name in NameSupply::FreshName (#18742)
MasterJH5574 Feb 10, 2026
2ae5097
[CI] Remove i386 and hexagon from `.asf.yaml` (#18748)
mshr-h Feb 10, 2026
c814064
[REFACTOR][S-TIR] Move remaining data structures to s_tir (#18743)
tqchen Feb 10, 2026
d129767
[Relax] Support constant folding for call_tir with tuple outputs (#18…
guan404ming Feb 11, 2026
4578a61
[DOC] Fix the loop length in a loop tiling example (#18753)
ruslo Feb 11, 2026
6f18e4e
[CI] Remove i386 and Hexagon from CI pipeline (1) (#18737)
mshr-h Feb 11, 2026
262c6d2
[Relax] Migrate NN conv/pooling/grad attrs from Array<IntImm> to Arra…
guan404ming Feb 11, 2026
467b330
[REFACTOR][S-TIR] Cleanup items on block scope (#18754)
tqchen Feb 11, 2026
d55204c
[CI][TEST] Cleanup legacy tests and migrate unity tests to main one (…
tqchen Feb 11, 2026
44d3554
Fix Customize Optimization tutorial import error #18584 (#18697)
AshwiniBokka Feb 11, 2026
38a3f21
[CI] Move conda config files to tests/conda and remove unused conda b…
tlopex Feb 11, 2026
c0f5206
[RUNTIME][COMPACT] Fix 32bit compact in vm (#18756)
tqchen Feb 11, 2026
1a22b62
[CI] Further cleanup CI after merging unity to main test (#18757)
tqchen Feb 11, 2026
6117888
[Cleanup] Remove redundant python/pyproject.toml and gen_requirements…
tlopex Feb 11, 2026
1c69074
[CI] Remove i386 and Hexagon from CI pipeline (2) (#18749)
mshr-h Feb 11, 2026
c08a701
[REFATOR][TIR] Phase out AllocConst (#18761)
tqchen Feb 12, 2026
119b536
[DOC] Fix RST syntax (#18732)
ruslo Feb 12, 2026
1f65047
[Relax] Remove TODO comment for moving code in fuse_tir.cc (#18762)
guan404ming Feb 12, 2026
6901832
[chore] Cleanup stale dependencies (#18760)
tqchen Feb 12, 2026
00a88b1
[REFACTOR] Remove picojson dependency, replace with tvm::ffi::json AP…
MasterJH5574 Feb 12, 2026
6e08d90
[REFACTOR][TIR] Phaseout BufferRealize (#18763)
tqchen Feb 12, 2026
c932f2a
[Relax] Add size heuristic to skip folding large creation ops (#18764)
guan404ming Feb 12, 2026
1bd40fa
[Relax] Fix llama4_rope_with_position_map to support partial rotary f…
guan404ming Feb 12, 2026
021547f
[CI] Upgrade Python to 3.10 in CI (#18647)
mshr-h Feb 13, 2026
783a9bf
[REFACTOR][S-TIR] Migrate more transform to s_tir (#18771)
tqchen Feb 13, 2026
84d05e9
[LLVM][Codegen] Cast NaN to bool gives true (#18772)
locnd182644 Feb 13, 2026
82b01c9
[REFACTOR][S-TIR] More migrations to s-tir (#18776)
tqchen Feb 14, 2026
2a448ce
[CI] Update Pillow, pytest-rerunfailures, junitparser, xgboost, onnx …
mshr-h Feb 14, 2026
f8dbbda
[REFACTOR] Phase out dmlc dep (#18779)
tqchen Feb 15, 2026
0291afa
[CI] Remove unity from tvm-bot (#18780)
mshr-h Feb 15, 2026
09be548
[S-TIR][FIX] Remove redundant std::move() to itself (#18782)
cbalint13 Feb 15, 2026
ac70260
[CI] Update system cuda version 12.4->12.8 (#18783)
mshr-h Feb 15, 2026
d537195
[chore] cleanup unsed legacy backtrac code in logging (#18781)
tqchen Feb 15, 2026
c1b6e50
[REFACTOR][RUNTIME] Transition metadata into ffi (#18784)
tqchen Feb 16, 2026
7e36a1e
[DOC] Fix PYTHONPATH in "Install from Source" (#18770)
ruslo Feb 16, 2026
8e59a56
[CONTRIB] Cache the shape and dtype array in json access (#18786)
tqchen Feb 16, 2026
2030db3
[REFACTOR][TARGET] Phase out legacy target string in favor of json (#…
tqchen Feb 16, 2026
c6b3fea
[CI] Update images to `20260214-152058-2a448ce4` (#18775)
mshr-h Feb 17, 2026
4d4351d
[DOC] Fix various grammar and punctuation issues (#18790)
ruslo Feb 17, 2026
bcd716b
[DOC] Fix various RST issues (#18789)
ruslo Feb 17, 2026
0460d82
[REFACTOR][TARGET] Cleanup target config (#18788)
tqchen Feb 17, 2026
e49f50b
[RELAX] Improve well-formed checker error messages with function name…
guan404ming Feb 17, 2026
283fd19
[REFACTOR][TARGET] Further cleanup target python api (#18793)
tqchen Feb 18, 2026
8057a76
[CI] Update actions to latest and move to full commit SHA pinning (#1…
mshr-h Feb 18, 2026
993c785
[ONNX] Handle Gelu approximate attribute from Opset 20 (#18773)
Mr-Neutr0n Feb 18, 2026
04b351c
[DOC] Unify GitHub naming (#18794)
ruslo Feb 18, 2026
52e4547
[DOC] Unify CUDA naming (#18797)
ruslo Feb 19, 2026
72f5185
[CI] Refactor Dockerfiles and installation scripts (#18796)
mshr-h Feb 19, 2026
1ebd5e0
[DOC] Fix docstring, unify CMake, nvidia-docker deprecation (#18799)
ruslo Feb 19, 2026
d8a4e6d
[FFI] Bring up latest tvm-ffi (#18802)
tqchen Feb 20, 2026
9752557
support integer types in fast_tanh and fast_exp (#18768)
jikechao Feb 20, 2026
c35f868
[CI] Update images to `20260219-160550-72f51851` (#18800)
mshr-h Feb 20, 2026
f533d0b
[REFACTOR] Migrate CHECK macros to tvm-ffi ones (#18803)
tqchen Feb 21, 2026
7e2ebc9
[REFACTOR][TEST] Remove unused te imports from test files (#18804)
tqchen Feb 21, 2026
c0828bc
[REFACTOR][TEST] Migrate tir-transform tests from TE to TVMScript (#1…
tqchen Feb 21, 2026
aa2e609
[LINT] Modernize lint to use pre-commit hooks (#18807)
tqchen Feb 22, 2026
7ac12eb
[CI] Add GitHub Actions lint workflow (#18809)
tqchen Feb 22, 2026
55027b7
[DRAFT][DO NOT MERGE] Bump tvm-ffi to v0.1.9 (#18801)
yongwww Feb 22, 2026
33dcea1
[REFACTOR][LINT] Modernize ruff config (#18810)
tqchen Feb 23, 2026
2cbbcd5
[Refactor][Metal] Update ICHECK to TVM_FFI_ICHECK in Metal runtime (#…
ballinyouup Feb 23, 2026
d463395
[REFACTOR][RUNTIME] Phase out legacy contrib runtime backends (#18813)
tqchen Feb 23, 2026
097d05f
WebGPU: gate subgroup ops via target parser thread_warp_size
ksgr5566 Feb 24, 2026
6d46fa7
[Relax][PyTroch] Add randn.default and randn_like.default support (#1…
guan404ming Feb 24, 2026
4fad6fc
[Relax] Make ShapeType ndim parameter mandatory (#18814)
guan404ming Feb 24, 2026
8fe6c8a
[REFACTOR][S-TIR] Lift STIR-only attributes out of tir::attr namespac…
tqchen Feb 24, 2026
f8dc0fb
[REFACTOR][CI] Modernize Python dependency management with uv (#18812)
mshr-h Feb 25, 2026
f119bbd
merge main
ksgr5566 Feb 25, 2026
b1e3688
merge main
ksgr5566 Feb 25, 2026
07d011c
updates based on latest merge
ksgr5566 Feb 25, 2026
327315c
[REFACTOR][METAL] Update CHECK_LE to TVM_FFI_ICHECK_LE in Metal runti…
mshr-h Feb 25, 2026
8c259a3
[REFACTOR][CI] Further cleanup docker images (#18817)
mshr-h Feb 26, 2026
8cb946e
[TARGET] Specify correct `mcpu` for Metal target tags (#18822)
mshr-h Feb 26, 2026
24a4e6c
[LLVM] Fix build failures when building with llvm>=22 (#18818)
mshr-h Feb 26, 2026
7b771ce
fix: replace 6 bare except clauses with except Exception (#18821)
haosenwang1018 Feb 26, 2026
c7af8e4
[DOC] Update installation docs with missing dependencies (#18194) (#1…
AshwiniBokka Feb 26, 2026
e30208c
[REFACTOR] Phase out root Makefile (#18828)
tqchen Feb 26, 2026
2d1f1cd
[REFACTOR] Update CHECK and ICHECK_GE to TVM_FFI_ICHECK and TVM_FFI_I…
babusid Feb 26, 2026
e5f483c
[REFACTOR][NODE] Remove node redirect headers (#18829)
tqchen Feb 27, 2026
6653620
[ROCm] Fix some ROCm codegen bugs (#15518)
zhangxiao-stack Feb 27, 2026
c1d3243
[BugFix][TOPI] Fix resize accuracy issue with non-floor rounding (#18…
tqchen Feb 27, 2026
95a4b1a
[IR][TIR] Remove body from AssertStmt (#18832)
tqchen Feb 28, 2026
9a8320a
[LINT][PYTHON] Modernize annotations with ruff UP rules (#18830)
tqchen Feb 28, 2026
3298e94
Merge remote-tracking branch 'origin/main' into webgpu-subgroup-test
ksgr5566 Feb 28, 2026
b1139a9
Added tests for warp reduce - subgroup enabled webgpu
ksgr5566 Feb 28, 2026
f7ebbdb
[REFACTOR] Further cleanup node redirections (#18844)
tqchen Feb 28, 2026
ab4ddd2
[CI] Remove legacy lint scripts and Apache RAT (#18848)
tqchen Feb 28, 2026
c45f72d
[CMAKE] Remove unused Libbacktrace.cmake (#18849)
tqchen Feb 28, 2026
5319c9c
[REFACTOR] Remove unused mscclpp contrib module (#18852)
tqchen Feb 28, 2026
7d5c46e
[TIR][FEAT] Require DeclBuffer before use in verify_well_formed (#18843)
tqchen Feb 28, 2026
2fd4e11
[PYTHON] Fix PEP 563 compat and remove args_converter (#18847)
tqchen Feb 28, 2026
3e5e4cf
[TEST] Remove stale URL mappings from request_hook (#18850)
tqchen Feb 28, 2026
91e9e2d
[CI] Remove stale test scripts (i386, hexagon, mypy) (#18851)
tqchen Feb 28, 2026
ff29817
[LLVM] Bump minimum LLVM version to 15 (#18853)
tqchen Feb 28, 2026
2f43f77
[REFACTOR][CONTRIB] Remove MSC contrib module (#18845)
tqchen Feb 28, 2026
293b580
[Docs] Fix RPC tutorial to use set_input + invoke_stateful API (#18855)
igotyuandme320 Feb 28, 2026
ae1f8f7
[DOCKER] Refactor bash.sh: auto-detect rootless, add --shell, TVM_DEV…
tqchen Feb 28, 2026
61f8081
[TVMScript] Fix PEP 563 closure variable resolution (#18856)
tqchen Mar 1, 2026
173bb23
[Runtime][Builtin] Handle mismatched type on argument #0 when calling…
locnd182644 Mar 1, 2026
9ffffae
fix: Complete ICHECK update across codebase (#18858)
MasterJH5574 Mar 1, 2026
2bab8bc
fix: Use T.decl_buffer instead of T.Buffer for aliased buffers in Lon…
MasterJH5574 Mar 1, 2026
63f099a
[CI][Docs] Update Sphinx dependencies (#18860)
mshr-h Mar 1, 2026
e693452
fix: Complete CHECK update across contrib runtime (#18861)
MasterJH5574 Mar 1, 2026
e9697fe
Lint errors fixed
ksgr5566 Mar 1, 2026
d95827a
[WebGPU] Add unit tests for subgroup warp reduction and fix uint32 ca…
ksgr5566 Mar 1, 2026
611a815
[TIR][Refactor] Enhance error reporting with structured AssertStmt an…
tqchen Mar 2, 2026
6627296
[S-TIR][Test] Mark meta_schedule tuning tests as skip (#18863)
tqchen Mar 2, 2026
c481950
[Relax][Refactor] Phase out FewShotTuning (#18864)
tqchen Mar 2, 2026
521440e
[REFACTOR][TIR] Cleanup AttrStmt attributes (#18862)
tqchen Mar 2, 2026
75f8589
[CI] Update images to `20260301-134651-63f099ad` (#18827)
mshr-h Mar 3, 2026
397ac1b
[Webgpu] Segment fault bug fix
ksgr5566 Mar 4, 2026
89d6142
[WebGPU] Add missing JSON implementation includes to WASM unity build
ksgr5566 Mar 4, 2026
21e5225
[ADRENO] Revive and consolicate Adreno features (#18867)
srkreddy1238 Mar 4, 2026
0fba160
[REFACTOR][TIR] Introduce AllocBuffer and phase out Allocate+DeclBuff…
tqchen Mar 4, 2026
969fad3
[TIR] Add VisitBufferDef/VisitBufferUse to base StmtVisitor/StmtMutat…
tqchen Mar 5, 2026
079e4af
[REFACTOR][TIR] Rename LetStmt to Bind and flatten to sequential sema…
tqchen Mar 5, 2026
dd7e260
[chore] Update docker/README.md documentation and fix links (#18875)
zhengkezhou1 Mar 5, 2026
419a8c8
Batched GPU dispatch and object caching for WebGPU runtime (#18871)
mitiskuma Mar 6, 2026
14e41c6
[Relax][ONNX] Support dynamic repeats for Tile (#18878)
LudovicoYIN Mar 6, 2026
689d2b5
[REFACTOR][TIR] Remove body from AllocBuffer and DeclBuffer (#18876)
tqchen Mar 6, 2026
17d1a28
[FIX][Adreno] Replace AllocBuffer with Bind in texture alloc injectio…
tqchen Mar 6, 2026
c0a305d
[TARGET] Fix round-trip reconstruction of targets with canonicalizer-…
mshr-h Mar 7, 2026
9dfa116
[Metal] Batched command dispatch and staging buffer pool (#18877)
mitiskuma Mar 7, 2026
717c822
[FIX] Fix cumsum kernel sblock_alloc_buffer for non-sblock buffer (#1…
tqchen Mar 7, 2026
72de122
[TIR][REFACTOR] Revamp Common Subexpression Elimination (#18886)
tqchen Mar 7, 2026
f83cebb
[TVMScript] Normalize T.Bind to T.bind for statement builder conventi…
tqchen Mar 8, 2026
5d3b525
[TIR] Reject non-floating inputs for trig unary ops (#18879)
LudovicoYIN Mar 8, 2026
a89b9f2
[TOPI] Reject non-float inputs for inverse unary math ops (#18880)
LudovicoYIN Mar 8, 2026
87bd8af
[TVMScript] Remove T.Bind backward-compat alias (#18891)
tqchen Mar 9, 2026
728eb15
[Build] Fix version regex to anchor at line start in pyproject.toml (…
MasterJH5574 Mar 9, 2026
7a41af0
[FIX][LLVM] Use isCPUStringValid for mcpu validation instead of enume…
gnguralnick Mar 10, 2026
1499bda
[Relax][PyTorch] Fix crash on dynamic shapes with identity slice in E…
mshr-h Mar 10, 2026
9a3edc9
Merge remote-tracking branch 'origin' into webgpu-subgroup-test
ksgr5566 Mar 11, 2026
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.
37 changes: 28 additions & 9 deletions .asf.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -16,7 +16,7 @@
# under the License.

github:
description: "Open deep learning compiler stack for cpu, gpu and specialized accelerators"
description: "Open Machine Learning Compiler Framework"
homepage: https://tvm.apache.org/
labels:
- tvm
Expand All @@ -33,6 +33,12 @@ github:
- spirv
- machine-learning

features:
# Enable issue management
issues: true
# Enable projects for project management boards
projects: true

# Triage perm for collaborators(test run)
#
# The perm is given based on needs and not based on
Expand All @@ -45,26 +51,39 @@ github:
# participation, permission is given on a three month
# cycle. PMC may review and recycle slots when necessary.
collaborators:
- hpanda-naut
- denise-k
- janetsc
- naut-thomas
- tvm-bot # For automated feedback in PR review.

# See https://cwiki.apache.org/confluence/display/INFRA/Git+-+.asf.yaml+features#Git.asf.yamlfeatures-Branchprotection
protected_branches:
main:
required_status_checks:
contexts:
- unity/pr-head
- arm/pr-head
- cpu/pr-head
- docker/pr-head
- gpu/pr-head
- hexagon/pr-head
- i386/pr-head
- lint/pr-head
- wasm/pr-head

required_pull_request_reviews:
required_approving_review_count: 1

enabled_merge_buttons:
# enable squash button:
squash: true
# default commit message when merging with a squash commit
# can either be: DEFAULT | PR_TITLE | PR_TITLE_AND_COMMIT_DETAILS | PR_TITLE_AND_DESC
squash_commit_message: PR_TITLE_AND_DESC
# enable merge button:
merge: false
# default commit message when merging with a merge commit
# can either be: DEFAULT | PR_TITLE | PR_TITLE_AND_DESC
merge_commit_message: DEFAULT
# enable rebase button for rare use.
rebase: true

notifications:
commits: commits@tvm.apache.org
issues: discuss-archive@tvm.apache.org
pullrequests: discuss-archive@tvm.apache.org
jobs: discuss-archive@tvm.apache.org
discussions: discuss-archive@tvm.apache.org
2 changes: 1 addition & 1 deletion .clang-format
Original file line number Diff line number Diff line change
Expand Up @@ -4,5 +4,5 @@
# https://clang.llvm.org/docs/ClangFormat.html
BasedOnStyle: Google
DerivePointerAlignment: false
ColumnLimit: 100
ColumnLimit: 100
PointerAlignment: Left
4 changes: 2 additions & 2 deletions .github/CODEOWNERSHIP
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,7 @@
# specific language governing permissions and limitations
# under the License.

# Github code owners file
# GitHub code owners file
# This file is used as a convenient tool to map
# committers' areas of expertise and faciliate the review process.
#
Expand Down Expand Up @@ -66,7 +66,7 @@ web/** @tqchen @jroesch
docker/** @areusch @leandron @jroesch

# Conda
conda/** @tqchen @junrushao1994 @comaniac
tests/conda/** @tqchen @junrushao1994 @comaniac

# CMake
cmake/** @jroesch @tqchen @areusch @junrushao1994 @comaniac
Expand Down
39 changes: 21 additions & 18 deletions .github/actions/setup/action.yml
Original file line number Diff line number Diff line change
@@ -1,39 +1,42 @@
runs:
using: "composite"
steps:
- uses: actions/cache@v3
- uses: actions/cache@cdf6c1fa76f9f475f3d7449005a359c84ca0f306 # v5.0.3
env:
CACHE_NUMBER: 1
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@v3
key: ${{ runner.os }}-conda-${{ env.CACHE_NUMBER }}-${{ hashFiles('tests/conda/build-environment.yaml') }}
- uses: conda-incubator/setup-miniconda@fc2d68f6413eb2d87b895e92f8584b5b94a10167 # v3.3.0
continue-on-error: true
id: conda1
with:
activate-environment: tvm-build
channel-priority: strict
environment-file: conda/build-environment.yaml
environment-file: tests/conda/build-environment.yaml
auto-activate-base: false
conda-solver: classic
use-only-tar-bz2: true
python-version: 3.9
condarc-file: conda/condarc
- uses: conda-incubator/setup-miniconda@v3
miniforge-version: latest
python-version: "3.10"
condarc-file: tests/conda/condarc
- uses: conda-incubator/setup-miniconda@fc2d68f6413eb2d87b895e92f8584b5b94a10167 # v3.3.0
if: steps.conda1.outcome == 'failure'
with:
activate-environment: tvm-build
channel-priority: strict
environment-file: conda/build-environment.yaml
environment-file: tests/conda/build-environment.yaml
auto-activate-base: false
conda-solver: classic
miniforge-version: latest
use-only-tar-bz2: true
python-version: 3.9
condarc-file: conda/condarc
python-version: "3.10"
condarc-file: tests/conda/condarc
- name: Conda info
shell: pwsh
run: |
conda info
conda list
conda info --envs
conda list --name base
mamba info
mamba list
mamba info --envs
mamba list --name base
- name: Install tvm-ffi pip package
shell: bash -l {0}
run: |
pip install -v ./3rdparty/tvm-ffi
2 changes: 1 addition & 1 deletion .github/workflows/cc_bot.yml
Original file line number Diff line number Diff line change
Expand Up @@ -35,7 +35,7 @@ jobs:
if: github.repository == 'apache/tvm'
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v2
- uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
submodules: "recursive"
- name: Add cc'ed reviewers
Expand Down
42 changes: 42 additions & 0 deletions .github/workflows/lint.yml
Original file line number Diff line number Diff line change
@@ -0,0 +1,42 @@
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements. See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership. The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied. See the License for the
# specific language governing permissions and limitations
# under the License.

name: Lint

on:
push:
branches: [main]
pull_request:
branches: [main]

concurrency:
group: Lint-${{ github.event.pull_request.number || github.sha }}
cancel-in-progress: true

jobs:
lint:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
fetch-depth: 0
fetch-tags: true
- name: Set up uv
uses: astral-sh/setup-uv@b75a909f75acd358c2196fb9a5f1299a9a8868a4 # v6.7.0
- name: Set up Python environment
run: uv sync --group lint --no-install-project
- uses: pre-commit/action@2c7b3805fd2a0fd8c1884dcaebf91fc102a13ecd # v3.0.1
133 changes: 28 additions & 105 deletions .github/workflows/main.yml
Original file line number Diff line number Diff line change
Expand Up @@ -39,32 +39,26 @@ jobs:
if: ${{ github.repository == 'apache/tvm' }}
runs-on: macOS-latest
steps:
- uses: actions/checkout@v2
- uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
submodules: 'recursive'
- name: Set up environment
uses: ./.github/actions/setup
- name: Conda Build
- name: Install LLVM dependencies
shell: bash -l {0}
run: |
conda install -c conda-forge llvmdev cmake ninja zlib
- name: Build TVM wheel
shell: bash -l {0}
run: |
pip install scikit-build-core
export CMAKE_ARGS="-DUSE_LLVM=ON -DBUILD_TESTING=OFF"
pip wheel --no-deps -w dist . -v
- name: Install TVM from wheel
shell: bash -l {0}
run: >-
conda build --output-folder=conda/pkg conda/recipe &&
conda install tvm -c ./conda/pkg
- name: Build iOS RPC
run: |
IOS_VERSION="14.0"
CMAKE_FLAGS="-DCMAKE_BUILD_TYPE=Release \
-DCMAKE_SYSTEM_NAME=iOS \
-DCMAKE_SYSTEM_VERSION=${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"
pip install dist/*.whl

mkdir build-ios-simulator
cd build-ios-simulator
cmake .. ${CMAKE_FLAGS}
cmake --build . --target ios_rpc
- name: Test
shell: bash -l {0}
run: >-
Expand All @@ -80,102 +74,31 @@ jobs:
python -m pytest -v -s 'tests/python/codegen/test_target_codegen_metal.py'
python -m pytest -v -s 'tests/python/codegen/test_target_codegen_gpu_common.py'
python -m pytest -v -s 'tests/python/codegen/test_gpu_codegen_allreduce.py::test_allreduce_sum[dims0-metal]'
# - name: Test iOS RPC
# shell: bash -l {0}
# run: >-
# python -m pip install tornado psutil cloudpickle &&
# export PYTHONPATH=tests/python/contrib:${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

Windows:
if: ${{ github.repository == 'apache/tvm' }}
runs-on: windows-2019
runs-on: windows-latest
steps:
- uses: actions/checkout@v2
- uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
with:
submodules: 'recursive'
- name: Set up environment
uses: ./.github/actions/setup
- name: Conda Build
- name: Install LLVM dependencies
shell: cmd /C call {0}
run: >-
conda build --output-folder=conda/pkg conda/recipe &&
conda install tvm -c ./conda/pkg
run: |
conda install -c conda-forge llvmdev cmake ninja zlib libxml2-devel
- name: Install TVM
shell: cmd /C call {0}
run: |
pip install scikit-build-core
set CMAKE_ARGS=-DUSE_LLVM=ON -DBUILD_TESTING=OFF
pip install --no-deps . -v
- name: Install test dependencies
shell: cmd /C call {0}
run: |
pip install psutil cloudpickle ml_dtypes numpy packaging scipy tornado typing_extensions
- name: Test
shell: cmd /C call {0}
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@v4
with:
name: android_rpc-debug.apk
path: ./apps/android_rpc/app/build/outputs/apk/debug/app-debug.apk
6 changes: 2 additions & 4 deletions .github/workflows/nightly_docker_update.yml
Original file line number Diff line number Diff line change
@@ -1,8 +1,6 @@

name: Nightly Docker Update
on:
schedule:
- cron: "0 0 * * *"
workflow_dispatch:

concurrency:
Expand All @@ -20,9 +18,9 @@ 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
- uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
- name: Open PR to update Docker images
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
Expand Down
Loading