[all-commits] [llvm/llvm-project] 9fa6b3: [mlir][bufferization] Improve aliasing OpOperand/O...

[Matthias Springer via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 9fa6b3504bed69b34b7c47c511a17eec08151ead
      https://github.com/llvm/llvm-project/commit/9fa6b3504bed69b34b7c47c511a17eec08151ead
  Author: Matthias Springer <springerm at google.com>
  Date:   2023-02-09 (Thu, 09 Feb 2023)

  Changed paths:
    M mlir/include/mlir/Dialect/Bufferization/IR/BufferizableOpInterface.h
    M mlir/include/mlir/Dialect/Bufferization/IR/BufferizableOpInterface.td
    M mlir/include/mlir/Dialect/Bufferization/IR/DstBufferizableOpInterfaceImpl.h
    M mlir/lib/Dialect/Arith/Transforms/BufferizableOpInterfaceImpl.cpp
    M mlir/lib/Dialect/Bufferization/IR/BufferizableOpInterface.cpp
    M mlir/lib/Dialect/Bufferization/Transforms/FuncBufferizableOpInterfaceImpl.cpp
    M mlir/lib/Dialect/Bufferization/Transforms/OneShotAnalysis.cpp
    M mlir/lib/Dialect/SCF/Transforms/BufferizableOpInterfaceImpl.cpp
    M mlir/lib/Dialect/Shape/Transforms/BufferizableOpInterfaceImpl.cpp
    M mlir/lib/Dialect/SparseTensor/Transforms/BufferizableOpInterfaceImpl.cpp
    M mlir/lib/Dialect/Tensor/Transforms/BufferizableOpInterfaceImpl.cpp
    M mlir/lib/Dialect/Vector/Transforms/BufferizableOpInterfaceImpl.cpp

  Log Message:
  -----------
  [mlir][bufferization] Improve aliasing OpOperand/OpResult property

`getAliasingOpOperands`/`getAliasingOpResults` now encodes OpOperand/OpResult, buffer relation and a degree of certainty. E.g.:
```
// aliasingOpOperands(%r) = {(%t, EQUIV, DEFINITE)}
// aliasingOpResults(%t) = {(%r, EQUIV, DEFINITE)}
%r = tensor.insert %f into %t[%idx] : tensor<?xf32>

// aliasingOpOperands(%r) = {(%t0, EQUIV, MAYBE), (%t1, EQUIV, MAYBE)}
// aliasingOpResults(%t0) = {(%r, EQUIV, MAYBE)}
// aliasingOpResults(%t1) = {(%r, EQUIV, MAYBE)}
%r = arith.select %c, %t0, %t1 : tensor<?xf32>
```

`BufferizableOpInterface::bufferRelation` is removed, as it is now part of `getAliasingOpOperands`/`getAliasingOpResults`.

This change allows for better analysis, in particular wrt. equivalence. This allows additional optimizations and better error checking (which is sometimes overly conservative). Examples:

* EmptyTensorElimination can eliminate `tensor.empty` inside `scf.if` blocks. This requires a modeling of equivalence: It is not a per-OpResult property anymore. Instead, it can be specified for each OpOperand and OpResult. This is important because `tensor.empty` may be eliminated only if all values on the SSA use-def chain to the final consumer (`tensor.insert_slice`) are equivalent.
* The detection of "returning allocs from a block" can be improved. (Addresses a TODO in `assertNoAllocsReturned`.) This allows us to bufferize IR such as "yielding a `tensor.extract_slice` result from an `scf.if` branch", which currently fails to bufferize because the alloc detection is too conservative.
* Better bufferization of loops. Aliases of the iter_arg can be yielded (even if they are not equivalent) without having to realloc and copy the entire buffer on each iteration.

The above-mentioned examples are not yet implemented with this change. This change just improves the BufferizableOpInterface, its implementations and related helper functions, so that better aliasing information is available for each op.

Differential Revision: https://reviews.llvm.org/D142129