Altivec version of h264_(h|v)_loop_filter_luma

patch by Graham Booker % perian A cod3r P com% with some minor fixes by me.
historic of the patch: http://trac.perian.org/ticket/113
Original thread:
Date: May 11, 2007 9:45 PM
Subject: [FFmpeg-devel] [PATCH] Altivec version of-altivec h264_h-v_loop_filter_luma

Originally committed as revision 9264 to svn://svn.ffmpeg.org/ffmpeg/trunk

libavcodec/ppc/dsputil_altivec.h

@@ -101,6 +101,17 @@ do { \
     h = vec_mergel (D2, H2); \
 } while (0)
 
+
+/** \brief loads unaligned vector \a *src with offset \a offset
+    and returns it */
+static inline vector unsigned char unaligned_load(int offset, uint8_t *src)
+{
+    register vector unsigned char first = vec_ld(offset, src);
+    register vector unsigned char second = vec_ld(offset+15, src);
+    register vector unsigned char mask = vec_lvsl(offset, src);
+    return vec_perm(first, second, mask);
+}
+
 #endif /* HAVE_ALTIVEC */
 
 #endif /* _DSPUTIL_ALTIVEC_ */