Merge remote-tracking branch 'qatar/master'

* qatar/master:
  SBR DSP: fix SSE code to not use SSE2 instructions.
  cpu: initialize mask to -1, so that by default, optimizations are used.
  error_resilience: initialize s->block_index[].
  svq3: protect against negative quantizers.
  Don't use ff_cropTbl[] for IDCT.
  swscale: make filterPos 32bit.
  FATE: add CPUFLAGS variable, mapping to -cpuflags avconv option.
  avconv: add -cpuflags option for setting supported cpuflags.
  cpu: add av_set_cpu_flags_mask().
  libx264: Allow overriding the sliced threads option
  avconv: fix counting encoded video size.

Conflicts:
	doc/APIchanges
	doc/fate.texi
	doc/ffmpeg.texi
	ffmpeg.c
	libavcodec/h264idct_template.c
	libavcodec/svq3.c
	libavutil/avutil.h
	libavutil/cpu.c
	libavutil/cpu.h
	libswscale/swscale.c
	tests/Makefile
	tests/fate-run.sh
	tests/regression-funcs.sh

Merged-by: Michael Niedermayer <michaelni@gmx.at>

1 files changed, 26 insertions, 36 deletions

diff --git a/libavcodec/vc1dsp.c b/libavcodec/vc1dsp.c
index 91578a74b8..da7593feca 100644
--- a/libavcodec/vc1dsp.c
+++ b/libavcodec/vc1dsp.c
@@ -139,8 +139,6 @@ static void vc1_h_s_overlap_c(DCTELEM *left, DCTELEM *right)
  * @see 8.6
  */
 static av_always_inline int vc1_filter_line(uint8_t* src, int stride, int pq){
-    uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
-
     int a0 = (2*(src[-2*stride] - src[ 1*stride]) - 5*(src[-1*stride] - src[ 0*stride]) + 4) >> 3;
     int a0_sign = a0 >> 31;        /* Store sign */
     a0 = (a0 ^ a0_sign) - a0_sign; /* a0 = FFABS(a0); */
@@ -163,8 +161,8 @@ static av_always_inline int vc1_filter_line(uint8_t* src, int stride, int pq){
                 else{
                     d = FFMIN(d, clip);
                     d = (d ^ d_sign) - d_sign;          /* Restore sign */
-                    src[-1*stride] = cm[src[-1*stride] - d];
-                    src[ 0*stride] = cm[src[ 0*stride] + d];
+                    src[-1*stride] = av_clip_uint8(src[-1*stride] - d);
+                    src[ 0*stride] = av_clip_uint8(src[ 0*stride] + d);
                 }
                 return 1;
             }
@@ -234,19 +232,17 @@ static void vc1_inv_trans_8x8_dc_c(uint8_t *dest, int linesize, DCTELEM *block)
 {
     int i;
     int dc = block[0];
-    const uint8_t *cm;
     dc = (3 * dc +  1) >> 1;
     dc = (3 * dc + 16) >> 5;
-    cm = ff_cropTbl + MAX_NEG_CROP + dc;
     for(i = 0; i < 8; i++){
-        dest[0] = cm[dest[0]];
-        dest[1] = cm[dest[1]];
-        dest[2] = cm[dest[2]];
-        dest[3] = cm[dest[3]];
-        dest[4] = cm[dest[4]];
-        dest[5] = cm[dest[5]];
-        dest[6] = cm[dest[6]];
-        dest[7] = cm[dest[7]];
+        dest[0] = av_clip_uint8(dest[0] + dc);
+        dest[1] = av_clip_uint8(dest[1] + dc);
+        dest[2] = av_clip_uint8(dest[2] + dc);
+        dest[3] = av_clip_uint8(dest[3] + dc);
+        dest[4] = av_clip_uint8(dest[4] + dc);
+        dest[5] = av_clip_uint8(dest[5] + dc);
+        dest[6] = av_clip_uint8(dest[6] + dc);
+        dest[7] = av_clip_uint8(dest[7] + dc);
         dest += linesize;
     }
 }
@@ -326,19 +322,17 @@ static void vc1_inv_trans_8x4_dc_c(uint8_t *dest, int linesize, DCTELEM *block)
 {
     int i;
     int dc = block[0];
-    const uint8_t *cm;
     dc = ( 3 * dc +  1) >> 1;
     dc = (17 * dc + 64) >> 7;
-    cm = ff_cropTbl + MAX_NEG_CROP + dc;
     for(i = 0; i < 4; i++){
-        dest[0] = cm[dest[0]];
-        dest[1] = cm[dest[1]];
-        dest[2] = cm[dest[2]];
-        dest[3] = cm[dest[3]];
-        dest[4] = cm[dest[4]];
-        dest[5] = cm[dest[5]];
-        dest[6] = cm[dest[6]];
-        dest[7] = cm[dest[7]];
+        dest[0] = av_clip_uint8(dest[0] + dc);
+        dest[1] = av_clip_uint8(dest[1] + dc);
+        dest[2] = av_clip_uint8(dest[2] + dc);
+        dest[3] = av_clip_uint8(dest[3] + dc);
+        dest[4] = av_clip_uint8(dest[4] + dc);
+        dest[5] = av_clip_uint8(dest[5] + dc);
+        dest[6] = av_clip_uint8(dest[6] + dc);
+        dest[7] = av_clip_uint8(dest[7] + dc);
         dest += linesize;
     }
 }
@@ -403,15 +397,13 @@ static void vc1_inv_trans_4x8_dc_c(uint8_t *dest, int linesize, DCTELEM *block)
 {
     int i;
     int dc = block[0];
-    const uint8_t *cm;
     dc = (17 * dc +  4) >> 3;
     dc = (12 * dc + 64) >> 7;
-    cm = ff_cropTbl + MAX_NEG_CROP + dc;
     for(i = 0; i < 8; i++){
-        dest[0] = cm[dest[0]];
-        dest[1] = cm[dest[1]];
-        dest[2] = cm[dest[2]];
-        dest[3] = cm[dest[3]];
+        dest[0] = av_clip_uint8(dest[0] + dc);
+        dest[1] = av_clip_uint8(dest[1] + dc);
+        dest[2] = av_clip_uint8(dest[2] + dc);
+        dest[3] = av_clip_uint8(dest[3] + dc);
         dest += linesize;
     }
 }
@@ -476,15 +468,13 @@ static void vc1_inv_trans_4x4_dc_c(uint8_t *dest, int linesize, DCTELEM *block)
 {
     int i;
     int dc = block[0];
-    const uint8_t *cm;
     dc = (17 * dc +  4) >> 3;
     dc = (17 * dc + 64) >> 7;
-    cm = ff_cropTbl + MAX_NEG_CROP + dc;
     for(i = 0; i < 4; i++){
-        dest[0] = cm[dest[0]];
-        dest[1] = cm[dest[1]];
-        dest[2] = cm[dest[2]];
-        dest[3] = cm[dest[3]];
+        dest[0] = av_clip_uint8(dest[0] + dc);
+        dest[1] = av_clip_uint8(dest[1] + dc);
+        dest[2] = av_clip_uint8(dest[2] + dc);
+        dest[3] = av_clip_uint8(dest[3] + dc);
         dest += linesize;
     }
 }