## Summary
- Make `PrimExpr` a typed C++ view over `Expr` values whose
`ExprNode::ty` is `PrimType`, instead of using a separate runtime node
class as the proof of primitive-ness.
- Use the shared `ir::Call` node for Relax, TIRX, and primitive-valued
calls, while keeping primitive-only APIs explicit at their semantic
boundaries.
- Keep Python on the general `Expr` surface for primitive-typed values
so `isinstance` behavior does not imply a nominal primitive-expression
subclass.
## Design Rationale
The main advantage of this change is that common expression nodes such
as `Call` can be unified without specializing each one to `PrimType`. A
single `ir::Call` can represent a Relax tensor call, a Relax scalar
call, or a primitive-valued intrinsic call; the result type stored in
`ExprNode::ty` determines whether that particular value can be viewed as
`PrimExpr`.
This keeps the IR node hierarchy focused on expression structure rather
than result-type categories. Nodes that are intrinsically primitive,
such as integer and floating-point literals or TIRX primitive operators,
still have strongly typed C++ APIs and data structures. General nodes
whose result type may vary, such as `Call`, remain general `Expr` nodes
and are narrowed to `PrimExpr` only where primitive-only semantics are
required.
The PR also keeps the compatibility surface practical: C++
primitive-only APIs continue to accept `PrimExpr`, Python exposes a
compatibility predicate for checking the primitive typed category, and
visitors/printers use one natural `Call` path rather than duplicating
Relax and primitive call handling. Missing expression types are
represented explicitly with `Type::Missing()` so constructors can leave
type inference to later analysis without relying on nullable `Type`
values.
This PR lets Relax expressions directly take `PrimExpr` values without
requiring the explicit `PrimValue` wrapper, continuing the Relax IR
unification work by removing Relax-specific leaf/base expression layers.
Summary:
- Remove `LeafExpr` / `LeafExprNode` and use direct expression-node
checks where needed.
- Converge Relax expression typing onto the shared IR `Expr` base.
- Remove the `PrimValue` node wrapper while keeping `relax.prim_value` /
`R.prim_value` as conversion helpers that return existing `PrimExpr`
values unchanged.
- Register direct `PrimExpr` handling through exact concrete node
dispatch, aligned with the `tirx` expression visitor list and excluding
arith iter-map intermediate nodes.
- Inline the private Python primitive conversion helper into public
`relax.prim_value`.
- Handle direct `PrimExpr` values in frontend scalar paths without
assuming a `.value` field on non-immediate expressions.
## Summary
- unify Relax's former StructInfo surface into the Type vocabulary and
Expr.ty storage path
- remove leftover DependentTypeNode and legacy OpNode::op_type storage
- keep base Type nullable while concrete Relax/DTensor type refs are
non-nullable
- clean stale StructInfo/TensorStructInfo/sinfo vocabulary in
Python/docs and distributed-op macros
- address Gemini follow-ups for parser annotations, BlockBuilder
docstring, and Adreno TensorType cast audit
This PR batches several post-bringup TIRx follow-ups, rebased onto
current `main`.
### Changes
- **op-dispatch**: per-call exec scope via `Tx.<scope>.op`; remove
`ExecScopeStmt`
- **namespaces**: split TIRx op namespaces; remove tile-primitive kind
attrs
- **codegen**: support explicit CUDA launch bounds
- **gemm-async**: support contiguous-axis (K-major) operand slicing
- **backend reorg**: move in-tree GPU backends out of core into
`src/backend/<target>/` and `python/tvm/backend/<target>/`
(codegen/runtime/op), with the corresponding `CMakeLists.txt` /
`cmake/modules` and include-path updates
### Testing
- Builds with `USE_CUDA=ON` / `USE_LLVM=ON`
- The TIRx Python test suite (`tests/python/tirx/`) passes locally
Replace TVM's `Diagnostic` / `DiagnosticContext` machinery with the
tvm-ffi
`visit_error_context` mechanism. Validators throw an `ffi::Error` seeded
with the
offending node; leaf pass executors (`ModulePass` / relax `Function` /
`DataflowBlock`) catch and rethrow `EnrichPassErrorWithContext`, which
appends the
failing pass name and a TVMScript-rendered, underlined source location.
`relax.analysis.well_formed` now throws on the first violation; a new
`check_well_formed` returns a bool, and all C++/Python/test callers are
routed
accordingly. `include/tvm/ir/diagnostic.h` and `src/ir/diagnostic.cc`
are deleted.
The enrichment renders with `num_context_lines=10` so a small function
shows in
full with no skipped-lines marker, while a large module stays bounded.
The TVMScript parser diagnostics
(`python/tvm/script/parser/core/diagnostics.py`)
stay self-contained pure-Python with no `DiagnosticContext` dependency,
and
restore multi-line source rendering: a diagnostic whose offending AST
node spans
multiple source lines now renders every spanned line with its gutter
line number
and an underline covering the span. `tvm.error.DiagnosticError` (used by
the
TVMScript parser) is retained.
A rendered end-to-end enriched-error example is posted as a comment
below.
## Summary
Python callers should reach the canonical tvm-ffi structural helpers
directly instead of going through a TVM-side redirect layer. This makes
the public tvm.ir bindings exact aliases of the tvm_ffi APIs and exposes
get_first_structural_mismatch from tvm.ir.
Main changes:
- Import structural_equal, get_first_structural_mismatch, and
structural_hash directly from tvm_ffi
- Remove the pure wrappers from tvm.ir.base while keeping
assert_structural_equal's TVM-specific formatting
- Update mismatch tests and add identity coverage for the direct
bindings
## Summary
Lifts 10 host-toolchain / CLI / process / utility modules from
`python/tvm/contrib/` to a new `python/tvm/support/` package, and
deletes two dead contrib shims.
`tvm.support` is the home for Python helpers that integrate TVM with
external CLIs and host-side tools — compilers, archivers, subprocess
pools, and build-info queries. These are load-bearing internal pieces
that TVM's compile/link/run paths depend on. `tvm.contrib` is reserved
for optional vendor SDK integrations and experimental features. The
distinction is documented in the `tvm.support` package docstring.
Moved (one commit each):
- `tvm.contrib.cc` → `tvm.support.cc`
- `tvm.contrib.nvcc` → `tvm.support.nvcc`
- `tvm.contrib.rocm` → `tvm.support.rocm`
- `tvm.contrib.ndk` → `tvm.support.ndk`
- `tvm.contrib.xcode` → `tvm.support.xcode`
- `tvm.contrib.clang` → `tvm.support.clang`
- `tvm.contrib.emcc` → `tvm.support.emcc`
- `tvm.contrib.popen_pool` → `tvm.support.popen_pool`
- `tvm.contrib.utils` → `tvm.support.utils`
- `tvm.contrib.tar` → `tvm.support.tar`
Deleted:
- `tvm.contrib.spirv` — single `optimize()` wrapping `spirv-opt`; zero
importers.
- `tvm.contrib.rpc` — self-deprecation shim with "removed in 0.5"
banner; honoring it.
Package conversion:
- `python/tvm/support.py` → `python/tvm/support/__init__.py` with
inclusion-rule docstring.
- `libinfo()` extracted into `python/tvm/support/libinfo.py`.
- `FrontendTestModule` dropped (audit confirmed zero callers outside its
own definition).
## Compatibility
Hard break — no `tvm.contrib.<mod>` re-export shims. All callers updated
in this PR.
C++-side FFI registry keys (`tvm.contrib.nvcc.*`, etc.) are unchanged —
only the Python module path moves. Renaming the FFI keys is a separate
follow-up.
## Summary
- `SubroutineMixin._get_subroutine()` used `structural_hash` as the sole
cache key without `structural_equal` verification. If two different
`arg_sinfo` values produced the same 64-bit hash (collision), the cache
would return a previously compiled function with mismatched parameter
shapes, leading to silently incorrect compiled output.
- Changed the cache to store a list of `(arg_sinfo, result)` pairs per
hash bucket and verify with `structural_equal` on lookup, consistent
with the pattern in `block_builder.cc`.
- Added a security advisory document and regression test.
## Root Cause
The subroutine cache (`cls._gvar`) was keyed by
`(structural_hash(arg_sinfo), is_dataflow)`. A hash match was treated as
proof of structural equality, skipping the necessary `structural_equal`
check. This is a hash-only lookup anti-pattern — hash determines the
bucket, but equality must confirm the match.
For comparison, `block_builder.cc` correctly uses `StructuralHash` +
`StructuralEqual` together as the hash and equality functions for
`std::unordered_map`.
## Test plan
- [ ] Existing test `test_linear` passes (no regression)
- [ ] New test `test_different_shapes_produce_distinct_subroutines`
passes — verifies that the same Module class with different input shapes
generates distinct subroutines
🤖 Generated with [Claude Code](https://claude.com/claude-code)
---------
Co-authored-by: 79475432@qq.com <dem0@kali.kali>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
DPL (`tvm.relax.dpl`) is heavily used across the TVM stack — operator
fusion, CUTLASS/cuBLAS/cuDNN backend dispatch, and user-defined graph
transforms all rely on it. Since there is no doc explaining how to use
it, this pr adds deep-dive documentation
(`docs/deep_dive/relax/dpl.rst`) covering DPL's pattern construction,
matching, rewriting APIs, and integration with `FuseOpsByPattern`
backend dispatch passes.
This PR adds the API reference documentation for `tvm.s_tir.analysis`.
`tvm.s_tir.analysis` functions use Var in their type annotations, which
exists in both `tvm.tirx` and `tvm.relax`. The existing disambiguator
uses common module prefix to pick the right one, but `tvm.s_tir` shares
no prefix with either. The new `tvm_module_type_preference` mapping
tells the disambiguator to prefer `tvm.tirx` types for `tvm.s_tir.*`
modules.
This PR is a follow-up of #18965
- Fix incorrect variable names in Relax dataflow code example (`lv0` →
`lv`, `b` → `n`) in
`docs/deep_dive/relax/learning.rst`
- Fix `func.time_evaluator(func.entry_name, ...)` to
`func.time_evaluator("add_one", ...)`
in `docs/how_to/tutorials/cross_compilation_and_rpc.py`, since
`entry_name` is a class
constant `"main"` but the compiled function is named `"add_one"`
- Fix typo `tvfm.testing` → `tvm.testing` in
`docs/how_to/dev/pytest_target_parametrization.rst`
- Add missing `tvm.relax.frontend.tflite` automodule entry to
`docs/reference/api/python/relax/frontend.rst`
- Fix incorrect function names (`lnumpy_matmul`→`lnumpy_linear`,
`lnumpy_relu`→`lnumpy_relu0`) and undefined variables (`lv0`→`lv`,
`b`→`n`) in Relax learning tutorial
- Add missing `I`, `T`, `R` imports in Relax and TensorIR learning
tutorials
- Update `pass_infra.rst` to match current source: fix `PassInfoNode`
field order and add `traceable`, correct `PassContextNode`
array types (`Expr`→`String`), remove obsolete `StringImm` cast in
`SequentialNode`, and add `traceable` param to `Create*Pass`
signatures
- Replace stale `PrintIRBefore`/`PrintAfter` TODOs with
already-implemented instruments (`PrintBeforeAll`, `PrintAfterAll`,
`PassPrintingInstrument`, `DumpIR`)
- Add missing `tvm.relax.op.vision` and `tvm.relax.op.vm` to API
reference
- Add `PythonDomain.find_obj` patch to resolve ambiguous
cross-references for classes that exist in multiple TVM namespaces (e.g.
`StringImm` in both `tvm.relax` and `tvm.tirx`). This is a general
solution that reuses the existing `tvm_class_name_rewrite_map` and also
benefits `Var`, `Call`, etc.
This PR brings up the tirx namespace. We have been spliting out the
original tir namespace to include high-level component s_tir and this PR
updates the remaining low-level part as tirx namespace
This PR removes legacy runtime contrib backends that have no existing
compiler backend,
no active development. They can always be brought back in future in case
we find there is a need
This PR initalizes the s_tir for scheduable TensorIR. The change mainly
starts from python side, the we will gradually move towards the c++ side
in followup PRs. The python main change:
tir.Schedule => s_tir.Schedule
This PR renames tir.Block to SBlock. This clearly indicate the
scheduable property of the block and is a prereq for followup stir
passes refactor.
Main changes:
- Data structure change from Block to SBlock
- Syntax change from T.block to T.sblock
This PR Updates the NDArray => Tensor.
Both tensor and ndarray are commonly used terms.
Because the term Tensor is getting more common in the context of ML,
we do the rename to stay more aligned with torch.Tensor and DLTensor.
[REFACTOR] Phase out getattr based attribute handling
This PR phases out getattar based attribute handling as they are slower
and introduces extra code path.
This does mean that if an Object is not explicitly registered
in python side, we will no longer be able to access the field by name.
Likely this is also desirable as we would like to enable faster use that
updates the python end and do not rely on these behavior.
This PR renames the filenames/namespaces of `relax_vm`
to `vm`.
Previously, both VMs of relay and relax exist, and to avoid the
name conflicts, we added the prefix `relax_` to relax VM.
With the Relay runtime being phased out, we can now rename
`relax_vm` to `vm` for conciseness.
* [REFACTOR] Phase out te.schedule python components
This PR phases out te.schedule python components.
te.compute is kept around for future usages.
tir.Schedule is a more modern version of the scheduling that we can use onwards.
Doing so also helps us to cleanup the testcases that relies on
explicit full build and execution. As we move future unit testcases
towards structural equality based unit tests.
* Simplify CI to focus on UT
The main rationale is that we should only have very few target
dependent UT in tests/python/codegen and possible
a new category in future for op-level integration if needed.
* Re-enable wasm
* fix lint
* remove hybrid,sparse autodoc and remove tests
---------
Co-authored-by: Siyuan Feng <hzfengsy@sjtu.edu.cn>
This PR starts the step 0 to phase out relay from the current
development main branch. This PR focuses on the python
components of relay, autotvm, auto_scheduler. To make the change
manageable, we will also do followup steps on te.Schedule and
c++ components in followup PRs.
To continue support community members who depends on
legacy flows, the [v0.19.0](https://github.com/apache/tvm/tree/v0.19.0)
branch will continue contain these components.
As noted in [discussion on phasing out legacy components](https://discuss.tvm.apache.org/t/phasing-out-legacy-components/17703/30),
this would help us to do two purposes:
- By removing outdated or redundant elements, we can significantly
reduce complexity and improve maintainability.
- Unify our focus: Concentrating our efforts on the new unity flow
will allow for more efficient development and innovation.
It is also a good opportunity for us to revisit and reduce CI time.
The past relay legacy flow contains a lot of end to end tests that
requires hardware resources to run and causing long CI time.
Moving onwards, we can focus more on unit-tests that focuses
on structural equality and runs within seconds, while be mindful
about tests that requires hardware resources (by restricting them
to specific folders and CI nightly in some cases).
---
Co-authored-by: Siyuan Feng <hzfengsy@sjtu.edu.cn>
This PR phases out VTA from the current development main branch.
The particular component will remain available in past releases
and is not actively maintained as of now.
* [Doc] Update Architecture Overview
Update and reorganize architecture documentation
This commit updates the architecture documentation by removing outdated files
and reorganizing the content. It also updates related sections in the deep dive
and developer tutorial.
* lint
* lint
* [Docs] Introduce Relax API and move legacy part to standalone page
As the TVM project evolves, the Unity strategy has been the recommended
way to use Apache TVM applications. Hence, we are pushing documentation
for the Relax API to the forefront and moving the legacy part to a
standalone page, which may be removed in the future.
* update for ci
* update for ci
Tianqi Chen