[PATCH] D105020: [SLP]Improve graph reordering.

[Peter Hawkins via Phabricator via llvm-commits]

phawkins added a comment.

I bisected a miscompilation in XLA to this change.

Repro:

- grab the three files in this gist: https://gist.github.com/hawkinsp/93de2fcb1a4d13ca01c826288bde4b9b
- build clang at this revision

At the this commit

  clang driver.cc module_0000.primitive_computation_broadcast_in_dim.3.ir-no-opt.ll -o a.out
  ./a.out buffer-assignment

(no optimization)

and

  clang -O3 -march=haswell driver.cc module_0000.primitive_computation_broadcast_in_dim.3.ir-no-opt.ll -o a.out
  ./a.out buffer-assignment

produce different outputs: the first 4 values differ.

At the previous revision, both produce identical outputs.

A quick inspection of the optimized IR seems to show it reading memory out of bounds. Disabling SLP vectorization also fixes the problem.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D105020/new/

https://reviews.llvm.org/D105020