[llvm] [X86] Vectorize non-power-of-two integer division (PR #215076)

[Simon Pilgrim via llvm-commits]

================
@@ -443,6 +443,19 @@ define <8 x i32> @test_divv_8i32(<8 x i32> %a, <8 x i32> %b) nounwind {
   ret <8 x i32> %res
 }
 
+
+define <7 x i32> @test_divv_7i32(<7 x i32> %a, <7 x i32> %b) nounwind {
----------------
RKSimon wrote:

now use the update script

https://github.com/llvm/llvm-project/pull/215076