diff --git a/libc/src/string/CMakeLists.txt b/libc/src/string/CMakeLists.txt
--- a/libc/src/string/CMakeLists.txt
+++ b/libc/src/string/CMakeLists.txt
@@ -56,7 +56,7 @@
     SRCS ${ADD_IMPL_SRCS}
     HDRS ${ADD_IMPL_HDRS}
     DEPENDS ${ADD_IMPL_DEPENDS}
-    COMPILE_OPTIONS ${ADD_IMPL_COMPILE_OPTIONS} ${flags}
+    COMPILE_OPTIONS ${ADD_IMPL_COMPILE_OPTIONS} ${flags} -O2
   )
   get_fq_target_name(${impl_name} fq_target_name)
   set_target_properties(${fq_target_name} PROPERTIES REQUIRE_CPU_FEATURES "${ADD_IMPL_REQUIRE}")
@@ -76,7 +76,7 @@
 
 function(add_memcpy memcpy_name)
   add_implementation(memcpy ${memcpy_name}
-    SRCS ${LIBC_SOURCE_DIR}/src/string/${LIBC_STRING_TARGET_ARCH}/memcpy.cpp
+    SRCS ${LIBC_SOURCE_DIR}/src/string/memcpy.cpp
     HDRS ${LIBC_SOURCE_DIR}/src/string/memcpy.h
     DEPENDS
       .memory_utils.memory_utils
@@ -87,7 +87,11 @@
   )
 endfunction()
 
-add_memcpy(memcpy MARCH native)
+if(${LIBC_STRING_TARGET_ARCH} STREQUAL "x86")
+  add_memcpy(memcpy MARCH native)
+else()
+  add_memcpy(memcpy)
+endif()
 
 # ------------------------------------------------------------------------------
 # memset
@@ -106,7 +110,11 @@
   )
 endfunction()
 
-add_memset(memset MARCH native)
+if(${LIBC_STRING_TARGET_ARCH} STREQUAL "x86")
+  add_memset(memset MARCH native)
+else()
+  add_memset(memset)
+endif()
 
 # ------------------------------------------------------------------------------
 # bzero
@@ -126,7 +134,11 @@
   )
 endfunction()
 
-add_bzero(bzero MARCH native)
+if(${LIBC_STRING_TARGET_ARCH} STREQUAL "x86")
+  add_bzero(bzero MARCH native)
+else()
+  add_bzero(bzero)
+endif()
 
 # ------------------------------------------------------------------------------
 # Add all other relevant implementations for the native target.
diff --git a/libc/src/string/x86/memcpy.cpp b/libc/src/string/memcpy.cpp
rename from libc/src/string/x86/memcpy.cpp
rename to libc/src/string/memcpy.cpp
--- a/libc/src/string/x86/memcpy.cpp
+++ b/libc/src/string/memcpy.cpp
@@ -12,15 +12,6 @@
 
 namespace __llvm_libc {
 
-static void CopyRepMovsb(char *__restrict dst, const char *__restrict src,
-                         size_t count) {
-  // FIXME: Add MSVC support with
-  // #include <intrin.h>
-  // __movsb(reinterpret_cast<unsigned char *>(dst),
-  //         reinterpret_cast<const unsigned char *>(src), count);
-  asm volatile("rep movsb" : "+D"(dst), "+S"(src), "+c"(count) : : "memory");
-}
-
 #if defined(__AVX__)
 #define BEST_SIZE 64
 #else
@@ -45,8 +36,8 @@
 //   implementation parameters.
 // - As compilers and processors get better, the generated code is improved
 //   with little change on the code side.
-static void memcpy_x86(char *__restrict dst, const char *__restrict src,
-                       size_t count) {
+static void memcpy_impl(char *__restrict dst, const char *__restrict src,
+                        size_t count) {
   if (count == 0)
     return;
   if (count == 1)
@@ -75,19 +66,13 @@
   if (count < 256)
     return CopyBlockOverlap<128>(dst, src, count);
 #endif
-  // kRepMovsBSize == -1 : Only CopyAligned is used.
-  // kRepMovsBSize ==  0 : Only RepMovsb is used.
-  // else CopyAligned is used to to kRepMovsBSize and then RepMovsb.
-  constexpr size_t kRepMovsBSize = -1;
-  if (count <= kRepMovsBSize)
-    return CopyAlignedBlocks<BEST_SIZE>(dst, src, count);
-  return CopyRepMovsb(dst, src, count);
+  return CopyAlignedBlocks<BEST_SIZE>(dst, src, count);
 }
 
 void *LLVM_LIBC_ENTRYPOINT(memcpy)(void *__restrict dst,
                                    const void *__restrict src, size_t size) {
-  memcpy_x86(reinterpret_cast<char *>(dst), reinterpret_cast<const char *>(src),
-             size);
+  memcpy_impl(reinterpret_cast<char *>(dst),
+              reinterpret_cast<const char *>(src), size);
   return dst;
 }