[Mlir-commits] [mlir] [mlir][Vector] Add a rewrite pattern for better low-precision ext(bit… (PR #65774)

[Oleksandr Alex Zinenko]

================
@@ -301,6 +302,25 @@ void populateVectorNarrowTypeEmulationPatterns(
     arith::NarrowTypeEmulationConverter &typeConverter,
     RewritePatternSet &patterns);
 
+/// Rewrite vector ext(maybe_broadcast(bitcast)) to use a more efficient
+/// sequence of vector operations comprising shuffles, shifts and bitwise
+/// logical ops. The rewrite uses an intermediate bitwidth equal to the licm of
----------------
ftynse wrote:

Nit: licm->lcm. 
Same in the commit description.

(Did you teach autocorrect about loop transformation names? :laughing: )

https://github.com/llvm/llvm-project/pull/65774