This generates better code on some non-x86 architectures.
---
libswscale/swscale.c | 49 ++++++++++++++++++-------------------------------
1 files changed, 18 insertions(+), 31 deletions(-)
diff --git a/libswscale/swscale.c b/libswscale/swscale.c
index 7495f2a..9945c01 100644
--- a/libswscale/swscale.c
+++ b/libswscale/swscale.c
@@ -434,6 +434,12 @@ static inline void yuv2nv12X_c(SwsContext *c, const
int16_t *lumFilter,
Y2>>=19;\
U >>=19;\
V >>=19;\
+ if ((Y1|Y2|U|V)&256) {\
+ Y1 = av_clip_uint8(Y1); \
+ Y2 = av_clip_uint8(Y2); \
+ U = av_clip_uint8(U); \
+ V = av_clip_uint8(V); \
+ }\
if (alpha) {\
A1 = 1<<18;\
A2 = 1<<18;\
@@ -443,20 +449,10 @@ static inline void yuv2nv12X_c(SwsContext *c, const
int16_t *lumFilter,
}\
A1>>=19;\
A2>>=19;\
- }\
- if ((Y1|Y2|U|V)&256) {\
- if (Y1>255) Y1=255; \
- else if (Y1<0)Y1=0; \
- if (Y2>255) Y2=255; \
- else if (Y2<0)Y2=0; \
- if (U>255) U=255; \
- else if (U<0) U=0; \
- if (V>255) V=255; \
- else if (V<0) V=0; \
- }\
- if (alpha && ((A1|A2)&256)) {\
- A1=av_clip_uint8(A1);\
- A2=av_clip_uint8(A2);\
+ if ((A1|A2)&256) {\
+ A1 = av_clip_uint8(A1); \
+ A2 = av_clip_uint8(A2); \
+ }\
}
#define YSCALE_YUV_2_RGBX_FULL_C(rnd,alpha) \
@@ -493,12 +489,9 @@ static inline void yuv2nv12X_c(SwsContext *c, const
int16_t *lumFilter,
G= Y + V*c->yuv2rgb_v2g_coeff + U*c->yuv2rgb_u2g_coeff;\
B= Y + U*c->yuv2rgb_u2b_coeff;\
if ((R|G|B)&(0xC0000000)) {\
- if (R>=(256<<22)) R=(256<<22)-1; \
- else if (R<0)R=0; \
- if (G>=(256<<22)) G=(256<<22)-1; \
- else if (G<0)G=0; \
- if (B>=(256<<22)) B=(256<<22)-1; \
- else if (B<0)B=0; \
+ R = av_clip_uintp2(R, 22); \
+ G = av_clip_uintp2(G, 22); \
+ B = av_clip_uintp2(B, 22); \
}
#define YSCALE_YUV_2_GRAY16_C \
@@ -506,8 +499,6 @@ static inline void yuv2nv12X_c(SwsContext *c, const int16_t
*lumFilter,
int j;\
int Y1 = 1<<18;\
int Y2 = 1<<18;\
- int U = 1<<18;\
- int V = 1<<18;\
\
const int i2= 2*i;\
\
@@ -517,11 +508,9 @@ static inline void yuv2nv12X_c(SwsContext *c, const
int16_t *lumFilter,
}\
Y1>>=11;\
Y2>>=11;\
- if ((Y1|Y2|U|V)&65536) {\
- if (Y1>65535) Y1=65535; \
- else if (Y1<0)Y1=0; \
- if (Y2>65535) Y2=65535; \
- else if (Y2<0)Y2=0; \
+ if ((Y1|Y2)&65536) {\
+ Y1 = av_clip_uint16(Y1); \
+ Y2 = av_clip_uint16(Y2); \
}
#define YSCALE_YUV_2_RGBX_C(type,alpha) \
@@ -635,10 +624,8 @@ static inline void yuv2nv12X_c(SwsContext *c, const
int16_t *lumFilter,
Y1>>=19;\
Y2>>=19;\
if ((Y1|Y2)&256) {\
- if (Y1>255) Y1=255;\
- else if (Y1<0)Y1=0;\
- if (Y2>255) Y2=255;\
- else if (Y2<0)Y2=0;\
+ Y1 = av_clip_uint8(Y1); \
+ Y2 = av_clip_uint8(Y2); \
}\
acc+= acc + g[Y1+d128[(i+0)&7]];\
acc+= acc + g[Y2+d128[(i+1)&7]];\
--
1.7.4.4
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel
