[all-commits] [llvm/llvm-project] 7248f4: [mlir][bufferization] Add "bottom-up from terminat...

[Matthias Springer via All-commits]

  Branch: refs/heads/users/matthias-springer/bufferization_heuristic
  Home:   https://github.com/llvm/llvm-project
  Commit: 7248f4e2294c3c717c83a9483c0e56a38e450f14
      https://github.com/llvm/llvm-project/commit/7248f4e2294c3c717c83a9483c0e56a38e450f14
  Author: Matthias Springer <springerm at google.com>
  Date:   2024-03-05 (Tue, 05 Mar 2024)

  Changed paths:
    M mlir/include/mlir/Dialect/Bufferization/IR/BufferizableOpInterface.h
    M mlir/include/mlir/Dialect/Bufferization/Transforms/OneShotAnalysis.h
    M mlir/include/mlir/Dialect/Bufferization/Transforms/Passes.td
    M mlir/lib/Dialect/Bufferization/Transforms/Bufferize.cpp
    M mlir/lib/Dialect/Bufferization/Transforms/OneShotAnalysis.cpp
    M mlir/test/Dialect/Arith/one-shot-bufferize.mlir
    M mlir/test/Dialect/Bufferization/Transforms/one-shot-bufferize-allow-return-allocs.mlir
    A mlir/test/Dialect/Bufferization/Transforms/one-shot-bufferize-analysis-bottom-up-from-terminators.mlir
    M mlir/test/Dialect/Bufferization/Transforms/one-shot-bufferize-partial.mlir
    M mlir/test/Dialect/Bufferization/Transforms/one-shot-bufferize.mlir
    M mlir/test/Dialect/Bufferization/Transforms/one-shot-module-bufferize-allow-return-allocs.mlir
    M mlir/test/Dialect/Bufferization/Transforms/one-shot-module-bufferize-analysis.mlir
    M mlir/test/Dialect/Bufferization/Transforms/one-shot-module-bufferize.mlir
    M mlir/test/Dialect/Linalg/one-shot-bufferize.mlir
    M mlir/test/Dialect/SCF/one-shot-bufferize-analysis.mlir
    M mlir/test/Dialect/SCF/one-shot-bufferize.mlir
    M mlir/test/Dialect/Tensor/one-shot-bufferize.mlir

  Log Message:
  -----------
  [mlir][bufferization] Add "bottom-up from terminators" analysis heuristic

One-Shot Bufferize currently does not support loops where a yielded value bufferizes to a buffer that is different from the buffer of the region iter_arg. In such a case, the bufferization fails with an error such as:
```
Yield operand #0 is not equivalent to the corresponding iter bbArg
    scf.yield %0 : tensor<5xf32>
```

One common reason for non-equivalent buffers is that an op on the path from the region iter_arg to the terminator bufferizes out-of-place. Ops that are analyzed earlier are more likely to bufferize in-place.

This commit adds a new heuristic that gives preference to ops that are reachable on the reverse SSA use-def chain from a region terminator and are within the parent region of the terminator. This is expected to work better than the existing heuristics for loops where an iter_arg is written to multiple times within a loop, but only one write is fed into the terminator.

Current users of One-Shot Bufferize are not affected by this change. "Bottom-up" is still the default heuristic. Users can switch to the new heuristic manually.

This commit also turns the "fuzzer" pass option into a heuristic, cleaning up the code a bit.



To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications