[all-commits] [llvm/llvm-project] 65aab9: [mlir][gpu] Generate multiple rank-specializations...

[Adam Paszke via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 65aab9e7222025f57c4bfc253d48c7b2ea8581da
      https://github.com/llvm/llvm-project/commit/65aab9e7222025f57c4bfc253d48c7b2ea8581da
  Author: Adam Paszke <apaszke at google.com>
  Date:   2023-12-01 (Fri, 01 Dec 2023)

  Changed paths:
    M mlir/lib/ExecutionEngine/CudaRuntimeWrappers.cpp

  Log Message:
  -----------
  [mlir][gpu] Generate multiple rank-specializations for tensor map cre… (#74082)

…ation

The previous code was technically incorrect in that the type indicated
that the memref only has 1 dimension, while the code below was happily
dereferencing the size array out of bounds. Now, if the compiler doesn't
get too smart about optimizations, this code *might even work*. But, if
the compiler realizes that the array has 1 element it might starrt doing
silly things. This generates a specialization per each supported rank,
making sure we don't do any UB.