simplify hpel_filter_c

[x264] / common / pixel.h

diff --git a/common/pixel.h b/common/pixel.h
index 63f60fcff1f0de334be32e247afb475fc2b8ba0e..d533620c01ccd1bc66bc5726eeceafad250f55ee 100644 (file)
--- a/common/pixel.h
+++ b/common/pixel.h
@@ -24,10 +24,11 @@
 #ifndef _PIXEL_H
 #define _PIXEL_H 1
 
-typedef int  (*x264_pixel_sad_t) ( uint8_t *, int, uint8_t *, int );
-typedef int  (*x264_pixel_satd_t)( uint8_t *, int, uint8_t *, int );
-typedef void (*x264_pixel_avg_t) ( uint8_t *, int, uint8_t *, int );
-typedef void (*x264_pixel_avg_weight_t) ( uint8_t *, int, uint8_t *, int, int );
+// SSD assumes all args aligned
+// other cmp functions assume first arg aligned
+typedef int  (*x264_pixel_cmp_t) ( uint8_t *, int, uint8_t *, int );
+typedef void (*x264_pixel_cmp_x3_t) ( uint8_t *, uint8_t *, uint8_t *, uint8_t *, int, int[3] );
+typedef void (*x264_pixel_cmp_x4_t) ( uint8_t *, uint8_t *, uint8_t *, uint8_t *, uint8_t *, int, int[4] );
 
 enum
 {
@@ -54,7 +55,7 @@ static const struct {
     {  4,  4 }
 };
 
-static const int x264_size2pixel[5][5] = {
+static const uint8_t x264_size2pixel[5][5] = {
     { 0, },
     { 0, PIXEL_4x4, PIXEL_8x4, 0, 0 },
     { 0, PIXEL_4x8, PIXEL_8x8, 0, PIXEL_16x8 },
@@ -64,12 +65,41 @@ static const int x264_size2pixel[5][5] = {
 
 typedef struct
 {
-    x264_pixel_sad_t  sad[7];
-    x264_pixel_satd_t satd[7];
-    x264_pixel_avg_t  avg[10];
-    x264_pixel_avg_weight_t avg_weight[10];
+    x264_pixel_cmp_t  sad[7];
+    x264_pixel_cmp_t  ssd[7];
+    x264_pixel_cmp_t satd[7];
+    x264_pixel_cmp_t ssim[7];
+    x264_pixel_cmp_t sa8d[4];
+    x264_pixel_cmp_t mbcmp[7]; /* either satd or sad for subpel refine and mode decision */
+    x264_pixel_cmp_t fpelcmp[7]; /* either satd or sad for fullpel motion search */
+    x264_pixel_cmp_x3_t fpelcmp_x3[7];
+    x264_pixel_cmp_x4_t fpelcmp_x4[7];
+
+    void (*ssim_4x4x2_core)( const uint8_t *pix1, int stride1,
+                             const uint8_t *pix2, int stride2, int sums[2][4] );
+    float (*ssim_end4)( int sum0[5][4], int sum1[5][4], int width );
+
+    /* multiple parallel calls to cmp. */
+    x264_pixel_cmp_x3_t sad_x3[7];
+    x264_pixel_cmp_x4_t sad_x4[7];
+    x264_pixel_cmp_x3_t satd_x3[7];
+    x264_pixel_cmp_x4_t satd_x4[7];
+
+    /* abs-diff-sum for successive elimination.
+     * may round width up to a multiple of 16. */
+    int (*ads[7])( int enc_dc[4], uint16_t *sums, int delta,
+                   uint16_t *cost_mvx, int16_t *mvs, int width, int thresh );
+
+    /* calculate satd of V, H, and DC modes.
+     * may be NULL, in which case just use pred+satd instead. */
+    void (*intra_satd_x3_16x16)( uint8_t *fenc, uint8_t *fdec, int res[3] );
+    void (*intra_satd_x3_8x8c)( uint8_t *fenc, uint8_t *fdec, int res[3] );
+    void (*intra_satd_x3_4x4)( uint8_t *fenc, uint8_t *fdec, int res[3] );
+    void (*intra_sa8d_x3_8x8)( uint8_t *fenc, uint8_t edge[33], int res[3] );
 } x264_pixel_function_t;
 
 void x264_pixel_init( int cpu, x264_pixel_function_t *pixf );
+int64_t x264_pixel_ssd_wxh( x264_pixel_function_t *pf, uint8_t *pix1, int i_pix1, uint8_t *pix2, int i_pix2, int i_width, int i_height );
+float x264_pixel_ssim_wxh( x264_pixel_function_t *pf, uint8_t *pix1, int i_pix1, uint8_t *pix2, int i_pix2, int i_width, int i_height );
 
 #endif