mem: Add documentations for REP string insturctions

Signed-off-by: Joe Richey <joerichey@google.com>

src/mem/x86_64.rs

@@ -1,5 +1,15 @@
 use super::c_int;
 
+// On recent Intel processors, "rep movsb" and "rep stosb" have been enhanced to
+// select the best microarchitecural implemenetation based on length and
+// alignment. This allows for very fast memcmp/memset implementations.
+// See "Intel® 64 and IA-32 Architectures Optimization Reference Manual"
+//  - Enhanced REP MOVSB and STOSB operation (ERMSB) (section 3.7.7)
+//  - Fast Short REP MOV (FSRM)
+//
+// While the "rep cmpsb" instructions have not seen similar optimizations, they
+// are still faster than our simple fallback implementations for memcmp/bcmp.
+
 #[inline(always)]
 unsafe fn copy_forward(dest: *mut u8, src: *const u8, count: usize) {
     asm!(