This is an automated email from the ASF dual-hosted git repository.
tqchen pushed a change to branch unity-staging
in repository https://gitbox.apache.org/repos/asf/tvm.git
discard 4570472596 [TVMScript][Unity] Improve PyLint Compatibility (#14276)
discard d6a695157f [Unity][ci] Use CPU-SMALL instances (#14256)
discard 20247dd8c0 [Unity] Introduce call_dps_packed (#14183)
discard 57732cc432 [Unity] Consider target context for Relay to Relax
conversion (#14269)
discard 6c2cfa8ef5 [Unity][Frontend] Import `tanh` and fix `layer_norm`
(#14247)
discard 5d1d033f4f [Unity][BYOC] Add conv2d and residual block patterns for
Relax cutlass BYOC (#14252)
discard 9bf6b9e410 [Unity] Allow user defined func attrs in emit_te (#14255)
discard 8306c9c48b [Unity][Op] Add repeat, tile, conv2d_transpose, avg_pool2d
(#14238)
discard db153b17ef [Unity][Op][Tweak] Improve `StructInfo` inference for
`shape_of` (#14243)
discard db42e205a5 [Unity][WEB] Improve ndarray cache (#14236)
discard 252abf4ae1 [Unity][WEB] Update text prompts for syntactical
correctness (#14237)
discard 0af0d79356 [Unity][TVMScript] Fix prim_func lost issue in
relax.emit_te (#14189)
discard 6b73d38c45 [Unity][TVMScript] Enable Context-Aware Parsing (#14234)
discard 29e41bc603 [Unity][Bugfix] Do not include `PrimFunc`s in the
dependency graph when checking for recursion (#14228)
discard 6ae4f25b3a [Unity][Transform] SimplifyNormInference (#14221)
discard 3492f4b1cc [Unity] Improve implementation of FuseOps (#14229)
discard e6f335cdd3 [Unity] ensure memory.alloc_tensor/storage roundtrippable
(#14226)
discard b9c7efb6ff [Unity][WEB] Simplify WebGPU Codegen per spec (#14225)
discard 9b87830963 [Unity][Transform] Memory plan across the IRModule (#14220)
discard 6861939059 [Unity][BYOC] Add dynamic shape support to CUTLASS matmul
(#14216)
discard 479529e763 [Unity][Frontend] from_fx keeps parameters in order (#14214)
discard 001ce2d580 [Unity][WEB] Improve webgpu codegen options to skip
readonly (#14213)
discard 2dd319d767 [Unity][Frontend] FX translator supports unwrapping unit
return tuple (#14212)
discard 193bb9d789 [Unity][Frontend] Attach imported model weights, deprecate
ImporterOutput (#14211)
discard 3ea14402aa [Unity] Introduce Default GPU Schedule Pass (#14182)
discard 92486a8a0d [Unity][Frontend] FX translator support torch.baddbmm
(#14202)
discard a62f0635c0 [Unity][TIR][Pass] ForceNarrowIndexToInt32 (#14203)
discard 497e9f6c7c [Unity][Fix] FX translating dtype (#14201)
discard 798bb66e15 [Unity][Frontend] FX translator returning weights with
`keep_params_as_input` (#14197)
discard bbe8edd758 [Unity][Frontend] FX translator supporting more ops (#14196)
discard 111495db36 [Unity][Op] Legalize `round`, `floor`, `ceil`, `sign`
(#14198)
discard 45487af1da [Unity][Op] Argmax and argmin (#14195)
discard c9f64c6415 [Unity][Op] Group normalization (#14194)
discard 27322e59a8 [Unity][Transform] LiftTransformParams handling multiple
functions (#14192)
discard ae321c51e7 [Unity][WEBGPU] Codegen improvements and WebRuntime (#14187)
discard 240621d736 [Unity][OP] Add an operator for fused multi head attention
(#14150)
discard e4eb1f6ee3 [Unity][Analysis] Restore Python bindings for var analyses
(#14180)
discard 2862549222 [Unity][Op] Full support of Relax op `power` (#14171)
discard 7cf4933a16 [Unity][BYOC] Add batch matmul support to Relax CUTLASS
BYOC (#14166)
discard fbc8721c31 [Unity][Analysis] Analysis for detecting recursion in Relax
(#14149)
discard 08e8cf88e3 [Unity] Add bind_constants option to FuseOpsByPattern
(#14151)
discard 17dff73ad1 [Unity][BYOC] Use Relax legalize + CPU build for reference
in tests (#14162)
discard 946cb26588 [Unity][Analysis] Checking function return struct info in
well-formed check (#14155)
discard 34ba550eb4 [Unity][Pass] Support Symbolic Shape Deduction during
BindParam (#14154)
discard 8743e4032b [Unity][Debugging] AST printer (#14152)
discard bb6eae0dfe [Unity][Pass] Enhance constant folding to fold relax ops by
evaluating them. (#14146)
discard e9b62b1af8 [Unity][Legalize] Fix Scalar Constant Legalization (#14127)
discard 267a92817c [Unity] Add callback to FuseOpsByPattern to check match
result is accepted (#14109)
discard c5be69e08b [Unity][BYOC] Assign group to unused bindings and ignroe
PrimFunc (#14139)
discard 6c15b4be3e [Unity][TVMScript] emit_te sugar (#14123)
discard b1b76a573a [Unity][BYOC] Add transposed matmul support to Relax
CUTLASS BYOC (#14128)
discard ec0ec24653 [Unity] Add Global info (#14132)
discard 5efb0b9311 [Unity][WEB] Relax vm on web runtime (#14131)
discard 3fa889dc2d [Unity][BlockBuilder] Add `name_hint` argument for `emit`
and `emit_output` (#14126)
discard b6224e3e99 [Unity][Fix] Fix bug in MergeCompositeFunctions (#14117)
discard 47f98f2da5 [Unity] Update tests again to adapt to latest TVMScript
syntax (#14115)
discard 0eabde5d89 [Unity][BYOC]Add relax backend pattern registry (#14106)
discard 561ac3801d [Unity] Remove attributes of relax.print, assert and unique
(#14101)
discard 093bf6d521 [Unity][Layout] Add layout transformation analysis for
PrimFunc (#14066)
discard aad7502ceb [Unity] Relax Recursive function (#14092)
discard 0d9f3a3970 [Unity] Lower `shape_of` to a builtin (#14093)
discard ff366ae4b8 [Unity] Fix typo in the comment (#14096)
discard ca18e9eacd [Unity][Relax] Set Shape Function to Be Host Function
(#14090)
discard 9cb42187b8 [Unity] Refactor Relax Build JIT UX (#14088)
discard d12123a6cf [Unity][Fix][Pass] FoldConstant with DCE in dataflow block
(#14087)
discard 30f004fe58 [Unity][Analysis] TIR pattern kind analysis for
multi-buffer write block (#14075)
discard a115e9ecf1 [Unity][Op] `log_softmax` and `cross_entropy_with_logits`
(#14083)
discard 255a2358fb [Unity][BYOC] Add DNNL backend (#14082)
discard db54a99138 [Unity][BYOC] Add CUTLASS backend (#14081)
discard 86490994c8 [Unity] Add testcases for `expr_args_converter` (#14080)
discard fd7c3cc197 [Unity][Pass] Canonicalize Bindings (#14079)
discard 9642340e8d [Unity][BYOC][Pass] RunCodegen and TensorRT (#14078)
discard 78ec7f1dea [Unity][Transform] Add LiftTransformParams pass (#14069)
discard a4b69ed1e2 [Unity][Frontend] Annotate number of non-static input of FX
function (#14067)
discard e5caf4085b [Unity][BYOC] Add pass to merge composite functions to
offload large subgraphs (#14062)
discard 2b8a9cccd9 [Unity][Pass] Remove Unused Function (#14061)
discard 7801803ee6 [Unity][Fix][Pass] Fix FuseOps for lack graph edges (#14058)
discard ec61975ba6 [Unity] Relax op: collapse sum (#14059)
discard 29f706f43c [Unity][BYOC] Add pattern-based partitioning pass (#14054)
discard 34c0248626 [Unity][VM] Add per-op profiling support (#14053)
discard 9336baac97 [Unity][TVMScript] Overload `__neg__` for relax expr
(#14045)
discard 50785f45b5 [Unity][Pass] FuseOps FuseTIR fixes (#14044)
discard ea105909ae [Unity] Statement rewriter for DataflowBlock (#14043)
discard e1b5400e1b [Unity] Relax dataflow pattern language (matching) (#14041)
discard 7fef9e3399 [Unity] Update tests to adapt to latest TVMScript syntax
(#14039)
discard f2c3180d73 [Unity] Disallow inline prim_func in relax IR (#14040)
discard bcb84f77eb [Unity][Pass] Block-level static memory planning (#14038)
discard f825714f5c [Unity] Initial PyTorch Frontend (#14037)
discard e5b0c6d879 [Unity][Op] Add ShapeExpr Tests for Reshape Op (#14035)
discard 4c3c347a08 [Unity][Pass] Operator legalization (#14029)
discard 966b874c84 [Unity][TVMScript] Move tir/relax import in script out of
__init__.py (#14033)
discard b6b1d272a9 [Unity][Pass] Wellformed Analysis (#14032)
discard 7ee0fe4fb6 [Unity][BlockBuilder] CallTE convert PrimValue args
(#14028)
discard 1e354c75c2 [Unity][Pass] Normalize Pass (#14031)
discard e3466b5478 [Unity] Relay -> Relax translator (#14026)
discard f420250abf [Unity][Pass][TuningAPI] Introduce TuningAPI and
MetaSchedule pass (#14014)
discard 4dd06e427e [Unity][Pass] BindParams pass, FoldConstant pass (#14016)
discard 01a172e7ec [Unity][VM] Supporting "compiled" exec mode. (#14015)
discard 89bf330f34 [Unity][Pass] LambdaLift pass (#14012)
discard 479a21cd31 [Unity][Pass] Operator Fusion Passes (#14001)
discard e6d1d2e2c0 [Unity] NestedMsg Support utility (#13995)
discard 388f9347fc [Unity] Relax op: manipulation (#13989)
discard 2affc5bb2b [Unity] Relax op: search (#13992)
discard 1c51d92e96 [Unity] Relax op: linear algebra (#13988)
discard 19c7925b57 [Unity] Relax op: creation (#13984)
discard 1fa77f2f29 [Unity] Relax op: neural networks (#13993)
discard ba81ffa577 [Unity] Relax op: statistical (#13991)
discard 35daa96508 [Unity] Relax op: arithmetic, comparison (#13983)
discard e28fe86403 [Unity] Relax op: image (#13994)
discard 41b811549d [Unity] Relax op: set (#13990)
discard 4c1b5b95a0 [Unity] Relax op: datatype (#13986)
discard 6e6f222b44 [Unity] Relax op: index (#13987)
discard 6a62f21fba [Unity][TVMScript] Use explicit `R.shape` in TVMScript
(#13979)
discard 166d60c281 [Unity] e2e Relax minimum build flow (#13961)
discard 66bd67d432 [Unity] Relax VM shape lowering pass (#13956)
discard 26c0653220 [Unity] Relax VM codegen (#13954)
discard 897cc206d9 [Unity] Relax TVMScript Printer (#13944)
discard 9fcc29ebfe [Unity] Relax TVMScript Parser. (#13932)
new 3c2ec306ec [Unity] Relax TVMScript Parser. (#13932)
new b37330d545 [Unity] Relax TVMScript Printer (#13944)
new d1fdb5ad39 [Unity] Relax VM codegen (#13954)
new 9c8ba6c936 [Unity] Relax VM shape lowering pass (#13956)
new bc38ebf5d8 [Unity] e2e Relax minimum build flow (#13961)
new caa797fb49 [Unity][TVMScript] Use explicit `R.shape` in TVMScript
(#13979)
new 595c367243 [Unity] Relax op: index (#13987)
new 925a8af703 [Unity] Relax op: datatype (#13986)
new bbdc478ed0 [Unity] Relax op: set (#13990)
new 5ff321a289 [Unity] Relax op: image (#13994)
new 71362cc20a [Unity] Relax op: arithmetic, comparison (#13983)
new e3b7761821 [Unity] Relax op: statistical (#13991)
new 153a438eb8 [Unity] Relax op: neural networks (#13993)
new 9dffef8c33 [Unity] Relax op: creation (#13984)
new 1434aa9ea4 [Unity] Relax op: linear algebra (#13988)
new b8d7d76049 [Unity] Relax op: search (#13992)
new ecf79209bc [Unity] Relax op: manipulation (#13989)
new 29ef5fd528 [Unity] NestedMsg Support utility (#13995)
new 0b9c28b033 [Unity][Pass] Operator Fusion Passes (#14001)
new 7bfe9080ec [Unity][Pass] LambdaLift pass (#14012)
new 13ecb22ed2 [Unity][VM] Supporting "compiled" exec mode. (#14015)
new 6df9918f65 [Unity][Pass] BindParams pass, FoldConstant pass (#14016)
new c00635d436 [Unity][Pass][TuningAPI] Introduce TuningAPI and
MetaSchedule pass (#14014)
new 883e58e148 [Unity] Relay -> Relax translator (#14026)
new 0ac137563a [Unity][Pass] Normalize Pass (#14031)
new 413a0dd59d [Unity][BlockBuilder] CallTE convert PrimValue args
(#14028)
new 557df54fd7 [Unity][Pass] Wellformed Analysis (#14032)
new 68ebd35704 [Unity][TVMScript] Move tir/relax import in script out of
__init__.py (#14033)
new 1753919c38 [Unity][Pass] Operator legalization (#14029)
new 2327c86e33 [Unity][Op] Add ShapeExpr Tests for Reshape Op (#14035)
new fd6969f025 [Unity] Initial PyTorch Frontend (#14037)
new 75908d7028 [Unity][Pass] Block-level static memory planning (#14038)
new 7de97c5ea3 [Unity] Disallow inline prim_func in relax IR (#14040)
new 114d0260af [Unity] Update tests to adapt to latest TVMScript syntax
(#14039)
new 7341d4dfdc [Unity] Relax dataflow pattern language (matching) (#14041)
new 32d0da5931 [Unity] Statement rewriter for DataflowBlock (#14043)
new 279621309b [Unity][Pass] FuseOps FuseTIR fixes (#14044)
new 784dfc5c99 [Unity][TVMScript] Overload `__neg__` for relax expr
(#14045)
new 150f5548e8 [Unity][VM] Add per-op profiling support (#14053)
new 9876e1861c [Unity][BYOC] Add pattern-based partitioning pass (#14054)
new c3d8ee992e [Unity] Relax op: collapse sum (#14059)
new 47768c583d [Unity][Fix][Pass] Fix FuseOps for lack graph edges (#14058)
new 1bacca41de [Unity][Pass] Remove Unused Function (#14061)
new 4223cef690 [Unity][BYOC] Add pass to merge composite functions to
offload large subgraphs (#14062)
new ac9f99ad7b [Unity][Frontend] Annotate number of non-static input of FX
function (#14067)
new 108cb65cc5 [Unity][Transform] Add LiftTransformParams pass (#14069)
new aa929d6871 [Unity][BYOC][Pass] RunCodegen and TensorRT (#14078)
new 1bc3583bf2 [Unity][Pass] Canonicalize Bindings (#14079)
new cb0750ec11 [Unity] Add testcases for `expr_args_converter` (#14080)
new 205bb258c1 [Unity][BYOC] Add CUTLASS backend (#14081)
new 49bebfe1c9 [Unity][BYOC] Add DNNL backend (#14082)
new 48dbd87a62 [Unity][Op] `log_softmax` and `cross_entropy_with_logits`
(#14083)
new 01fea31388 [Unity][Analysis] TIR pattern kind analysis for
multi-buffer write block (#14075)
new c07360d048 [Unity][Fix][Pass] FoldConstant with DCE in dataflow block
(#14087)
new edc82c1e79 [Unity] Refactor Relax Build JIT UX (#14088)
new 42ba80e87d [Unity][Relax] Set Shape Function to Be Host Function
(#14090)
new a8e8064732 [Unity] Fix typo in the comment (#14096)
new e26347ccbe [Unity] Lower `shape_of` to a builtin (#14093)
new 412023fb9d [Unity] Relax Recursive function (#14092)
new 110c045c81 [Unity][Layout] Add layout transformation analysis for
PrimFunc (#14066)
new a88c58a1a1 [Unity] Remove attributes of relax.print, assert and unique
(#14101)
new bd2f63a15b [Unity][BYOC]Add relax backend pattern registry (#14106)
new d6733d601b [Unity] Update tests again to adapt to latest TVMScript
syntax (#14115)
new 9f79a32c11 [Unity][Fix] Fix bug in MergeCompositeFunctions (#14117)
new be0c0faf44 [Unity][BlockBuilder] Add `name_hint` argument for `emit`
and `emit_output` (#14126)
new 3c11bd56e9 [Unity][WEB] Relax vm on web runtime (#14131)
new 7b92f9730f [Unity] Add Global info (#14132)
new 716d3b954b [Unity][BYOC] Add transposed matmul support to Relax
CUTLASS BYOC (#14128)
new 42fa65c622 [Unity][TVMScript] emit_te sugar (#14123)
new 809530aefa [Unity][BYOC] Assign group to unused bindings and ignroe
PrimFunc (#14139)
new b167176526 [Unity] Add callback to FuseOpsByPattern to check match
result is accepted (#14109)
new 83adf7ba63 [Unity][Legalize] Fix Scalar Constant Legalization (#14127)
new ce0cd47183 [Unity][Pass] Enhance constant folding to fold relax ops by
evaluating them. (#14146)
new 75a96711e4 [Unity][Debugging] AST printer (#14152)
new b353f9da25 [Unity][Pass] Support Symbolic Shape Deduction during
BindParam (#14154)
new 8eab261ac2 [Unity][Analysis] Checking function return struct info in
well-formed check (#14155)
new 9eea97edb6 [Unity][BYOC] Use Relax legalize + CPU build for reference
in tests (#14162)
new b2097291da [Unity] Add bind_constants option to FuseOpsByPattern
(#14151)
new 64f800cd85 [Unity][Analysis] Analysis for detecting recursion in Relax
(#14149)
new c2942f6261 [Unity][BYOC] Add batch matmul support to Relax CUTLASS
BYOC (#14166)
new b8efd1857b [Unity][Op] Full support of Relax op `power` (#14171)
new fc55d6ffce [Unity][Analysis] Restore Python bindings for var analyses
(#14180)
new 2b50663cc2 [Unity][OP] Add an operator for fused multi head attention
(#14150)
new 469eb3fa5c [Unity][WEBGPU] Codegen improvements and WebRuntime (#14187)
new 48278e2702 [Unity][Transform] LiftTransformParams handling multiple
functions (#14192)
new 5af7853aa5 [Unity][Op] Group normalization (#14194)
new b9dbafeab0 [Unity][Op] Argmax and argmin (#14195)
new 95e0e975c5 [Unity][Op] Legalize `round`, `floor`, `ceil`, `sign`
(#14198)
new 13c14936d0 [Unity][Frontend] FX translator supporting more ops (#14196)
new fdd9579115 [Unity][Frontend] FX translator returning weights with
`keep_params_as_input` (#14197)
new ce8e29ba43 [Unity][Fix] FX translating dtype (#14201)
new d949fb4581 [Unity][TIR][Pass] ForceNarrowIndexToInt32 (#14203)
new 1f24b02b17 [Unity][Frontend] FX translator support torch.baddbmm
(#14202)
new a7b5c5dea4 [Unity] Introduce Default GPU Schedule Pass (#14182)
new c92848936a [Unity][Frontend] Attach imported model weights, deprecate
ImporterOutput (#14211)
new 8e5333ae22 [Unity][Frontend] FX translator supports unwrapping unit
return tuple (#14212)
new 6f95d1fad4 [Unity][WEB] Improve webgpu codegen options to skip
readonly (#14213)
new e353e40686 [Unity][Frontend] from_fx keeps parameters in order (#14214)
new 4781c24f44 [Unity][BYOC] Add dynamic shape support to CUTLASS matmul
(#14216)
new b0387bf430 [Unity][Transform] Memory plan across the IRModule (#14220)
new 4e1f5c8af3 [Unity][WEB] Simplify WebGPU Codegen per spec (#14225)
new d1d97b8492 [Unity] ensure memory.alloc_tensor/storage roundtrippable
(#14226)
new ba09e2ebbd [Unity] Improve implementation of FuseOps (#14229)
new 4fb4bf60a4 [Unity][Transform] SimplifyNormInference (#14221)
new 83b31e478f [Unity][Bugfix] Do not include `PrimFunc`s in the
dependency graph when checking for recursion (#14228)
new ea315c7d1d [Unity][TVMScript] Enable Context-Aware Parsing (#14234)
new 5ceb308432 [Unity][TVMScript] Fix prim_func lost issue in
relax.emit_te (#14189)
new 1681d80cc3 [Unity][WEB] Update text prompts for syntactical
correctness (#14237)
new f6e4b99e4a [Unity][WEB] Improve ndarray cache (#14236)
new df06bd46f1 [Unity][Op][Tweak] Improve `StructInfo` inference for
`shape_of` (#14243)
new a933bc9e33 [Unity][Op] Add repeat, tile, conv2d_transpose, avg_pool2d
(#14238)
new 8eac72c2d8 [Unity] Allow user defined func attrs in emit_te (#14255)
new 269cb18465 [Unity][BYOC] Add conv2d and residual block patterns for
Relax cutlass BYOC (#14252)
new 89bfd45d98 [Unity][Frontend] Import `tanh` and fix `layer_norm`
(#14247)
new b73e550538 [Unity] Consider target context for Relay to Relax
conversion (#14269)
new fd90a3e948 [Unity] Introduce call_dps_packed (#14183)
new 74c6cb7f6d [Unity][ci] Use CPU-SMALL instances (#14256)
new 302de0c0ad [TVMScript][Unity] Improve PyLint Compatibility (#14276)
This update added new revisions after undoing existing revisions.
That is to say, some revisions that were in the old version of the
branch are not in the new version. This situation occurs
when a user --force pushes a change and generates a repository
containing something like this:
* -- * -- B -- O -- O -- O (4570472596)
\
N -- N -- N refs/heads/unity-staging (302de0c0ad)
You should already have received notification emails for all of the O
revisions, and so the following emails describe only the N revisions
from the common base, B.
Any revisions marked "omit" are not gone; other references still
refer to them. Any revisions marked "discard" are gone forever.
The 118 revisions listed above as "new" are entirely new to this
repository and will be described in separate emails. The revisions
listed as "add" were already present in the repository and have only
been added to this reference.
Summary of changes:
python/tvm/script/parser/tir/operation.py | 14 --------------
1 file changed, 14 deletions(-)
