[llvm] [RISCV] Tune flag for fast vrgather.vv (PR #124664)

[Petr Penzin via llvm-commits]

================
@@ -4228,7 +4237,7 @@ static SDValue lowerBUILD_VECTOR(SDValue Op, SelectionDAG &DAG,
   // Cap the cost at a value linear to the number of elements in the vector.
   // The default lowering is to use the stack.  The vector store + scalar loads
   // is linear in VL.  However, at high lmuls vslide1down and vslidedown end up
-  // being (at least) linear in LMUL.  As a result, using the vslidedown
+  // being (at least) linear in LMUL.  As a resultdedown
----------------
ppenzin wrote:

Indeed, good catch. Should be addressed.

https://github.com/llvm/llvm-project/pull/124664