X-Git-Url: https://git.sesse.net/?a=blobdiff_plain;f=libavcodec%2Frectangle.h;h=73e8b0abd18846ea0a91062d2d6c3a3e3a7b291d;hb=a0e1c3517a656dd32293f054a339e0ac73328138;hp=cfd0f0696dfcc881e565fbe64c8fbb8b0737d8a9;hpb=bad5537e2c2caeb5deb1ff9d771ea01058b8010c;p=ffmpeg

diff --git a/libavcodec/rectangle.h b/libavcodec/rectangle.h
index cfd0f0696df..73e8b0abd18 100644
--- a/libavcodec/rectangle.h
+++ b/libavcodec/rectangle.h
@@ -2,25 +2,25 @@
  * rectangle filling function
  * Copyright (c) 2003 Michael Niedermayer <michaelni@gmx.at>
  *
- * This file is part of FFmpeg.
+ * This file is part of Libav.
  *
- * FFmpeg is free software; you can redistribute it and/or
+ * Libav is free software; you can redistribute it and/or
  * modify it under the terms of the GNU Lesser General Public
  * License as published by the Free Software Foundation; either
  * version 2.1 of the License, or (at your option) any later version.
  *
- * FFmpeg is distributed in the hope that it will be useful,
+ * Libav is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  * Lesser General Public License for more details.
  *
  * You should have received a copy of the GNU Lesser General Public
- * License along with FFmpeg; if not, write to the Free Software
+ * License along with Libav; if not, write to the Free Software
  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
  */
 
 /**
- * @file libavcodec/rectangle.h
+ * @file
  * useful rectangle filling function
  * @author Michael Niedermayer <michaelni@gmx.at>
  */
@@ -31,23 +31,21 @@
 #include <assert.h>
 #include "config.h"
 #include "libavutil/common.h"
-#include "dsputil.h"
 
 /**
  * fill a rectangle.
  * @param h height of the rectangle, should be a constant
  * @param w width of the rectangle, should be a constant
- * @param size the size of val (1 or 4), should be a constant
+ * @param size the size of val (1, 2 or 4), should be a constant
  */
 static av_always_inline void fill_rectangle(void *vp, int w, int h, int stride, uint32_t val, int size){
     uint8_t *p= (uint8_t*)vp;
-    assert(size==1 || size==4);
+    assert(size==1 || size==2 || size==4);
     assert(w<=4);
 
     w      *= size;
     stride *= size;
 
-    assert((((long)vp)&(FFMIN(w, STRIDE_ALIGN)-1)) == 0);
     assert((stride&(w-1))==0);
     if(w==2){
         const uint16_t v= size==4 ? val : val*0x0101;
@@ -58,7 +56,7 @@ static av_always_inline void fill_rectangle(void *vp, int w, int h, int stride,
         *(uint16_t*)(p + 2*stride)= v;
         *(uint16_t*)(p + 3*stride)= v;
     }else if(w==4){
-        const uint32_t v= size==4 ? val : val*0x01010101;
+        const uint32_t v= size==4 ? val : size==2 ? val*0x00010001 : val*0x01010101;
         *(uint32_t*)(p + 0*stride)= v;
         if(h==1) return;
         *(uint32_t*)(p + 1*stride)= v;
@@ -68,7 +66,7 @@ static av_always_inline void fill_rectangle(void *vp, int w, int h, int stride,
     }else if(w==8){
     //gcc can't optimize 64bit math on x86_32
 #if HAVE_FAST_64BIT
-        const uint64_t v= val*0x0100000001ULL;
+        const uint64_t v=  size==2 ? val*0x0001000100010001ULL : val*0x0100000001ULL;
         *(uint64_t*)(p + 0*stride)= v;
         if(h==1) return;
         *(uint64_t*)(p + 1*stride)= v;
@@ -87,16 +85,17 @@ static av_always_inline void fill_rectangle(void *vp, int w, int h, int stride,
         *(uint64_t*)(p + 0+3*stride)= v;
         *(uint64_t*)(p + 8+3*stride)= v;
 #else
-        *(uint32_t*)(p + 0+0*stride)= val;
-        *(uint32_t*)(p + 4+0*stride)= val;
+        const uint32_t v= size==2 ? val*0x00010001 : val;
+        *(uint32_t*)(p + 0+0*stride)= v;
+        *(uint32_t*)(p + 4+0*stride)= v;
         if(h==1) return;
-        *(uint32_t*)(p + 0+1*stride)= val;
-        *(uint32_t*)(p + 4+1*stride)= val;
+        *(uint32_t*)(p + 0+1*stride)= v;
+        *(uint32_t*)(p + 4+1*stride)= v;
         if(h==2) return;
-        *(uint32_t*)(p + 0+2*stride)= val;
-        *(uint32_t*)(p + 4+2*stride)= val;
-        *(uint32_t*)(p + 0+3*stride)= val;
-        *(uint32_t*)(p + 4+3*stride)= val;
+        *(uint32_t*)(p + 0+2*stride)= v;
+        *(uint32_t*)(p + 4+2*stride)= v;
+        *(uint32_t*)(p + 0+3*stride)= v;
+        *(uint32_t*)(p + 4+3*stride)= v;
     }else if(w==16){
         *(uint32_t*)(p + 0+0*stride)= val;
         *(uint32_t*)(p + 4+0*stride)= val;