[Mlir-commits] [mlir] [mlir][x86vector] AVX512-BF16 Dot op (PR #124800)

[Adam Siemieniuk]

================
@@ -0,0 +1,25 @@
+// RUN: mlir-opt %s -convert-vector-to-scf -convert-scf-to-cf -convert-vector-to-llvm="enable-x86vector" -convert-to-llvm -reconcile-unrealized-casts | \
+// RUN: mlir-translate --mlir-to-llvmir | \
+// RUN: %lli --entry-function=entry --mattr="avx512bf16" --dlopen=%mlir_c_runner_utils | \
----------------
adam-smnk wrote:

Collapsed extensions into `.avx512.` within MLIR.
For LLVM intrinsic generation, `AVX512_IntrOp` now optionally accepts `extension` string that's used for LLVM intrinsic enum generation. 

https://github.com/llvm/llvm-project/pull/124800