[llvm] [CodeGen][PreISelIntrinsicLowering] Add VP-based lowering for memcpy/memmove/memset (PR #165585)

[David Del Río via llvm-commits]

================
@@ -373,6 +465,173 @@ tryInsertCastToCommonAddrSpace(IRBuilderBase &B, Value *Addr1, Value *Addr2,
   return {ResAddr1, ResAddr2};
 }
 
+// Lower memmove to IR using VP intrinsics. memmove is required to correctly
+// copy overlapping memory regions; therefore, it has to check the relative
+// positions of the source and destination pointers and choose the copy
+// direction accordingly.
+//
+// The code below is an IR rendition of this C function using RISCV intrinsics:
+//
+// void* memmove(void* dst, const void* src, size_t n) {
+//    uint8_t *d = (uint8_t *)dest;
+//    const uint8_t *s = (const uint8_t *)src;
+//    if (d < s || d >= s + n) {
----------------
dadra-oc wrote:

That's is true, I already fixed the comment.

https://github.com/llvm/llvm-project/pull/165585