Re: optimalisation for strlcpy (lib/string.c)

From: Folkert van Heusden
Date: Sun Dec 10 2006 - 17:04:41 EST


> > by this:
> > size_t strlcpy(char *dest, const char *src, size_t size)
> > {
> > char *tmp = dest;
> >
> > for(;;)
> > {
> > *dest = *src;
> > if (!*src)
> > break;
> >
> > if (--size == 0)
> > break;
> >
> > dest++;
> > src++;
> > }
> >
> > *dest = 0x00;
> >
> > return dest - tmp;
> > }
> >
> > I've tested the speed difference with this:
> > http://www.vanheusden.com/misc/kernel-strlcpy-opt-test.c
> > and the speed difference is quite a bit on a P4: 28% faster.
>
> The two do not do exactly the same. The first one handles 'size == 0'
> (should not happen?) safely while the other does not.

Ok, small change:
diff -uNrBbd lib/string.c string-new.c
--- lib/string.c 2006-11-04 02:33:58.000000000 +0100
+++ string-new.c 2006-12-10 23:02:26.000000000 +0100
@@ -121,14 +122,27 @@
*/
size_t strlcpy(char *dest, const char *src, size_t size)
{
- size_t ret = strlen(src);
+ char *tmp = dest;

- if (size) {
- size_t len = (ret >= size) ? size - 1 : ret;
- memcpy(dest, src, len);
- dest[len] = '\0';
+ if (likeley(size > 0))
+ {
+ for(;;)
+ {
+ *dest = *src;
+ if (unlikely(!*src))
+ break;
+
+ if (unlikely(--size == 0))
+ break;
+
+ dest++;
+ src++;
}
- return ret;
+ }
+
+ *dest = 0x00;
+
+ return dest - tmp;
}
EXPORT_SYMBOL(strlcpy);
#endif


Folkert van Heusden

--
Ever wonder what is out there? Any alien races? Then please support
the seti@home project: setiathome.ssl.berkeley.edu
----------------------------------------------------------------------
Phone: +31-6-41278122, PGP-key: 1F28D8AE, www.vanheusden.com
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/