This patch improves strncpy performance by using strnlen/memcpy rather than a byte loop. Performance

on bench-strncpy is 1.9-2.1x faster on average. I tried several variations, and using a tailcall and
calling memset conditionally gave the best overall results.

2 files changed, 10 insertions, 54 deletions

diff --git a/ChangeLog b/ChangeLog
index d013561..5f40e4e 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -1,5 +1,10 @@
 2015-08-05  Wilco Dijkstra  <wdijkstr@arm.com>
 
+	* string/strncpy.c (strncpy):
+	Improve performance by using __strnlen/memcpy.
+
+2015-08-05  Wilco Dijkstra  <wdijkstr@arm.com>
+
 	* sysdeps/aarch64/fpu/feenablxcpt.c (feenableexcept):
 	Optimize to avoid an unnecessary FPCR read.
 
diff --git a/string/strncpy.c b/string/strncpy.c
index 37af5aa..d464bbb 100644
--- a/string/strncpy.c
+++ b/string/strncpy.c
@@ -16,68 +16,19 @@
    <http://www.gnu.org/licenses/>.  */
 
 #include <string.h>
-#include <memcopy.h>
 
 #undef strncpy
 
 #ifndef STRNCPY
-#define STRNCPY strncpy
+ #define STRNCPY strncpy
 #endif
 
 char *
 STRNCPY (char *s1, const char *s2, size_t n)
 {
-  char c;
-  char *s = s1;
-
-  --s1;
-
-  if (n >= 4)
-    {
-      size_t n4 = n >> 2;
-
-      for (;;)
-	{
-	  c = *s2++;
-	  *++s1 = c;
-	  if (c == '\0')
-	    break;
-	  c = *s2++;
-	  *++s1 = c;
-	  if (c == '\0')
-	    break;
-	  c = *s2++;
-	  *++s1 = c;
-	  if (c == '\0')
-	    break;
-	  c = *s2++;
-	  *++s1 = c;
-	  if (c == '\0')
-	    break;
-	  if (--n4 == 0)
-	    goto last_chars;
-	}
-      s1++;
-      n = n - (s1 - s);
-      memset (s1, '\0', n);
-      return s;
-    }
-
- last_chars:
-  n &= 3;
-  if (n == 0)
-    return s;
-
-  do
-    {
-      c = *s2++;
-      *++s1 = c;
-      if (--n == 0)
-	return s;
-    }
-  while (c != '\0');
-
-  memset (s1 + 1, '\0', n);
-  return s;
+  size_t size = __strnlen (s2, n);
+  if (size != n)
+    memset (s1 + size, '\0', n - size);
+  return memcpy (s1, s2, size);
 }
 libc_hidden_builtin_def (strncpy)