[Mlir-commits] [mlir] [MLIR][Linalg] Transform pass to optimize and lower vector contract operation to FMA (PR #121748)

Rolf Morel llvmlistbot at llvm.org
Mon Jan 6 09:19:47 PST 2025


================
@@ -1824,6 +1824,15 @@ void populateConstantFoldLinalgOperations(RewritePatternSet &patterns,
 /// suffices for achieving the sum.
 void populateFoldAddIntoDestPatterns(RewritePatternSet &patterns);
 
+
+/// Pattern to hoists the vector transfer reads/writes outside the reduction and
----------------
rolfmorel wrote:

nit: "hoists the" -> "hoist"

https://github.com/llvm/llvm-project/pull/121748


More information about the Mlir-commits mailing list