[llvm] [PreISelIntrinsicLowering] Lower bounded memcpy/memmove to masked loa… (PR #212710)

[Simon Pilgrim via llvm-commits]

https://github.com/RKSimon commented:

why do you rely on the AC and not allow just general value tracking?
```ll
define void @memcpy_bounded(ptr %dst, ptr %src, i64 %size) {
  %size.mod = and i64 %size, 63
  tail call void @llvm.memcpy.p0.p0.i64(ptr %dst, ptr %src, i64 %size.mod, i1 false)
  ret void
}
```

https://github.com/llvm/llvm-project/pull/212710