Merge commit '05b0998f511ffa699407465d48c7d5805f746ad2'

* commit '05b0998f511ffa699407465d48c7d5805f746ad2':
  dsputil: Fix error by not using redzone and register name
  swscale: GBRP output support

Conflicts:
	libswscale/output.c
	libswscale/swscale.c
	libswscale/swscale_internal.h
	libswscale/utils.c
	tests/ref/lavfi/pixdesc
	tests/ref/lavfi/pixfmts_copy
	tests/ref/lavfi/pixfmts_null
	tests/ref/lavfi/pixfmts_scale
	tests/ref/lavfi/pixfmts_vflip

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

5 files changed, 35 insertions, 29 deletions

diff --git a/libavcodec/x86/hpeldsp.asm b/libavcodec/x86/hpeldsp.asm
index ea366a65c5..72bc111baa 100644
--- a/libavcodec/x86/hpeldsp.asm
+++ b/libavcodec/x86/hpeldsp.asm
@@ -318,8 +318,8 @@ PUT_NO_RND_PIXELS8_Y2_EXACT
 ; avg_pixels8(uint8_t *block, const uint8_t *pixels, int line_size, int h)
 %macro AVG_PIXELS8 0
 cglobal avg_pixels8, 4,5
-    movsxdifnidn r2, edx
-    lea          r4, [r2+r2]
+    movsxdifnidn r2, r2d
+    lea          r4, [r2*2]
 .loop:
     mova         m0, [r0]
     mova         m1, [r0+r2]
@@ -349,7 +349,7 @@ AVG_PIXELS8
 ; avg_pixels8_x2(uint8_t *block, const uint8_t *pixels, int line_size, int h)
 %macro AVG_PIXELS8_X2 0
 cglobal avg_pixels8_x2, 4,5
-    movsxdifnidn r2, edx
+    movsxdifnidn r2, r2d
     lea          r4, [r2*2]
 .loop:
     mova         m0, [r1]
diff --git a/libavcodec/x86/mpeg4qpel.asm b/libavcodec/x86/mpeg4qpel.asm
index 053630356e..fe47bf302e 100644
--- a/libavcodec/x86/mpeg4qpel.asm
+++ b/libavcodec/x86/mpeg4qpel.asm
@@ -169,7 +169,7 @@ INIT_MMX 3dnow
 PUT_NO_RND_PIXELS16_l2
 
 %macro MPEG4_QPEL16_H_LOWPASS 1
-cglobal %1_mpeg4_qpel16_h_lowpass, 5, 5, 0, 8
+cglobal %1_mpeg4_qpel16_h_lowpass, 5, 5, 0, 16
     movsxdifnidn r2, r2d
     movsxdifnidn r3, r3d
     pxor         m7, m7
@@ -202,7 +202,7 @@ cglobal %1_mpeg4_qpel16_h_lowpass, 5, 5, 0, 8
     paddw        m6, [PW_ROUND]
     paddw        m0, m6
     psraw        m0, 5
-    mova    [rsp-8], m0
+    mova    [rsp+8], m0
     mova         m0, [r1+5]
     mova         m5, m0
     mova         m6, m0
@@ -226,7 +226,7 @@ cglobal %1_mpeg4_qpel16_h_lowpass, 5, 5, 0, 8
     paddw        m1, [PW_ROUND]
     paddw        m3, m1
     psraw        m3, 5
-    mova         m1, [rsp-8]
+    mova         m1, [rsp+8]
     packuswb     m1, m3
     OP_MOV     [r0], m1, m4
     mova         m1, [r1+9]
diff --git a/libswscale/output.c b/libswscale/output.c
index e49cb73f15..d1a3bee59e 100644
--- a/libswscale/output.c
+++ b/libswscale/output.c
@@ -1374,11 +1374,11 @@ YUV2RGBWRAPPERX(yuv2, rgb_full, rgb8_full,   AV_PIX_FMT_RGB8,  0)
 
 static void
 yuv2gbrp_full_X_c(SwsContext *c, const int16_t *lumFilter,
-                          const int16_t **lumSrc, int lumFilterSize,
-                          const int16_t *chrFilter, const int16_t **chrUSrc,
-                          const int16_t **chrVSrc, int chrFilterSize,
-                          const int16_t **alpSrc, uint8_t **dest,
-                          int dstW, int y)
+                  const int16_t **lumSrc, int lumFilterSize,
+                  const int16_t *chrFilter, const int16_t **chrUSrc,
+                  const int16_t **chrVSrc, int chrFilterSize,
+                  const int16_t **alpSrc, uint8_t **dest,
+                  int dstW, int y)
 {
     const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(c->dstFormat);
     int i;
@@ -1388,36 +1388,42 @@ yuv2gbrp_full_X_c(SwsContext *c, const int16_t *lumFilter,
 
     for (i = 0; i < dstW; i++) {
         int j;
-        int Y = 1<<9;
-        int U = (1<<9)-(128 << 19);
-        int V = (1<<9)-(128 << 19);
+        int Y = 1 << 9;
+        int U = (1 << 9) - (128 << 19);
+        int V = (1 << 9) - (128 << 19);
         int R, G, B, A;
 
-        for (j = 0; j < lumFilterSize; j++) {
+        for (j = 0; j < lumFilterSize; j++)
             Y += lumSrc[j][i] * lumFilter[j];
-        }
+
         for (j = 0; j < chrFilterSize; j++) {
             U += chrUSrc[j][i] * chrFilter[j];
             V += chrVSrc[j][i] * chrFilter[j];
         }
+
         Y >>= 10;
         U >>= 10;
         V >>= 10;
+
         if (hasAlpha) {
             A = 1 << 18;
-            for (j = 0; j < lumFilterSize; j++) {
+
+            for (j = 0; j < lumFilterSize; j++)
                 A += alpSrc[j][i] * lumFilter[j];
-            }
+
             A >>= 19;
+
             if (A & 0x100)
                 A = av_clip_uint8(A);
         }
+
         Y -= c->yuv2rgb_y_offset;
         Y *= c->yuv2rgb_y_coeff;
         Y += 1 << 21;
-        R = Y + V*c->yuv2rgb_v2r_coeff;
-        G = Y + V*c->yuv2rgb_v2g_coeff + U*c->yuv2rgb_u2g_coeff;
-        B = Y +                          U*c->yuv2rgb_u2b_coeff;
+        R = Y + V * c->yuv2rgb_v2r_coeff;
+        G = Y + V * c->yuv2rgb_v2g_coeff + U * c->yuv2rgb_u2g_coeff;
+        B = Y +                            U * c->yuv2rgb_u2b_coeff;
+
         if ((R | G | B) & 0xC0000000) {
             R = av_clip_uintp2(R, 30);
             G = av_clip_uintp2(G, 30);
diff --git a/libswscale/swscale_internal.h b/libswscale/swscale_internal.h
index 0a320c2da4..83d3a0049c 100644
--- a/libswscale/swscale_internal.h
+++ b/libswscale/swscale_internal.h
@@ -250,12 +250,12 @@ typedef void (*yuv2packedX_fn)(struct SwsContext *c, const int16_t *lumFilter,
  *                      or some output formats.
  */
 typedef void (*yuv2anyX_fn)(struct SwsContext *c, const int16_t *lumFilter,
-                               const int16_t **lumSrc, int lumFilterSize,
-                               const int16_t *chrFilter,
-                               const int16_t **chrUSrc,
-                               const int16_t **chrVSrc, int chrFilterSize,
-                               const int16_t **alpSrc, uint8_t **dest,
-                               int dstW, int y);
+                            const int16_t **lumSrc, int lumFilterSize,
+                            const int16_t *chrFilter,
+                            const int16_t **chrUSrc,
+                            const int16_t **chrVSrc, int chrFilterSize,
+                            const int16_t **alpSrc, uint8_t **dest,
+                            int dstW, int y);
 
 /* This struct should be aligned on at least a 32-byte boundary. */
 typedef struct SwsContext {
diff --git a/libswscale/utils.c b/libswscale/utils.c
index c10bd7ce73..9e73adce38 100644
--- a/libswscale/utils.c
+++ b/libswscale/utils.c
@@ -1022,8 +1022,8 @@ av_cold int sws_init_context(SwsContext *c, SwsFilter *srcFilter,
     if (isPlanarRGB(dstFormat)) {
         if (!(flags & SWS_FULL_CHR_H_INT)) {
             av_log(c, AV_LOG_DEBUG,
-                "%s output is not supported with half chroma resolution, switching to full\n",
-                av_get_pix_fmt_name(dstFormat));
+                   "%s output is not supported with half chroma resolution, switching to full\n",
+                   av_get_pix_fmt_name(dstFormat));
             flags   |= SWS_FULL_CHR_H_INT;
             c->flags = flags;
         }