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>
* [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
This adds a short lint to ensure that all files have a single trailing newline and no trailing whitespaces. This PR is in two commits, one to add the check and another to fix currently offending files in the repo. See https://github.com/apache/tvm/pull/13058/commits/ba2c2e235e2a16d62fdeed959044c65012d9f942 for just the significant changes. Auto-corrections applied with
```
pre-commit run --all-files
```
* RelayViz interface and terminal ast-dump.
This PR follows https://github.com/apache/tvm/pull/8668, with splitting
out interfaces class and terminal ast-dump implementation.
This visualizer is aimed for quick look-then-fix, so the interface is
simple. Despite that, customization is still possbile through
implementing interfaces defined in `interface.py` or overriding existent
implementations inside a renderer module, like `terminal.py`.
A tutorial is also provided in this PR.
A graphviz renderer will also be contributed after this PR.
* lint and typo
* Remove compile_engine.h for real
* Fix format
* RM compile_engine.cc
* Swap compile engine with TECompiler
* Cleanup on compile engine py leftovers
* [WIP] Exposing legacy compile engine capabilities through TE Compiler
* Swap usages for depreciated compile engine with TE compiler
* Track and replace usages of compile engine refactor them to TE compiler
* [Docs] Log helper mod
* Remove depreciated function for lookup compile engine cachce
* Fix typos
* Debug misc cleanups
* Register global pass for using te compiler for auto scheduler
* Fix tests using the legacy compile engine
* Fix broken autotuner tests and minor cleanups
* Swap compile engine with te_compiler in rst config
* PR nits
* Fix failed test
Co-authored-by: Jared Roesch <roeschinc@gmail.com>
* Documentation Refactor - Stage 1
RFC: https://github.com/apache/tvm-rfcs/blob/main/rfcs/0027-formalize-documentation-organization.md
Tracking Issue: https://github.com/apache/tvm/issues/8987
Stage 1 of the documentation refactor reorganizes the docs structure,
moving files (without content changes) and adding new scaffolding to
generate the proper document tree.
It does not address naming, style, content, links, or other existing
content in documents that were moved. State 2 will address fixing these
issues with existing content.
Major changes include but are not limited to:
* Dividing the existing tutorials into two sections:
* Tutorials
* How Tos
* Moving all of the existing tutorials out of the `/tutorial`
directory and into the more general `/gallery` directory.
* Breaking up how-tos into individual sections for more
flexibility and more consistent rendering.
* Moving content into new classifications:
* `/docs/arch` for architecture guides
* `/docs/reference` for API guides and other reference material
* `/docs/topic` for topic specific guides such as microTVM and VTA
* Restructuring `/docs/dev`
* Adding a table of contents to the doc index
* Adding instructions on how to install using third-party tlcpack
* Documentation Refactor - Stage 2
RFC: https://github.com/apache/tvm-rfcs/blob/main/rfcs/0027-formalize-documentation-organization.md
Tracking Issue: https://github.com/apache/tvm/issues/8987
Stage 2 of the documentation refactor fixes naming and links
in the documentation to be consistent with the overall structure.
Major changes include:
* an update to how to contribute to docs.
* several updated index pages with title changes to match
the organization style and bring consistency to the sections
* expanded descriptions of some page collections
* fixed links
* Documentation Refactor - Stage 3
RFC: https://github.com/apache/tvm-rfcs/blob/main/rfcs/0027-formalize-documentation-organization.md
Tracking Issue: https://github.com/apache/tvm/issues/8987
Stage 3 of the documentation refactor adjusts CI for the new structure.
The CI build script takes into account the new gallery format. It
also prevents deleting existing documents, and takes advantage of the
`_staging` and `_build` directories to clean out previous builds.
Shushi Hong