dsputil: add bswap16_buf()

There are several places where a buffer is byte-swapped in 16-bit units.
This allows them to share code which can be optimised for various
architectures.

Signed-off-by: Mans Rullgard <mans@mansr.com>

libavcodec/dsputil.c

@@ -222,6 +222,12 @@ static void bswap_buf(uint32_t *dst, const uint32_t *src, int w){
     }
 }
 
+static void bswap16_buf(uint16_t *dst, const uint16_t *src, int len)
+{
+    while (len--)
+        *dst++ = av_bswap16(*src++);
+}
+
 static int sse4_c(void *v, uint8_t * pix1, uint8_t * pix2, int line_size, int h)
 {
     int s, i;
@@ -4324,6 +4330,7 @@ av_cold void dsputil_init(DSPContext* c, AVCodecContext *avctx)
     c->add_hfyu_left_prediction  = add_hfyu_left_prediction_c;
     c->add_hfyu_left_prediction_bgr32 = add_hfyu_left_prediction_bgr32_c;
     c->bswap_buf= bswap_buf;
+    c->bswap16_buf = bswap16_buf;
 #if CONFIG_PNG_DECODER
     c->add_png_paeth_prediction= ff_add_png_paeth_prediction;
 #endif

libavcodec/dsputil.h

@@ -369,6 +369,7 @@ typedef struct DSPContext {
     /* this might write to dst[w] */
     void (*add_png_paeth_prediction)(uint8_t *dst, uint8_t *src, uint8_t *top, int w, int bpp);
     void (*bswap_buf)(uint32_t *dst, const uint32_t *src, int w);
+    void (*bswap16_buf)(uint16_t *dst, const uint16_t *src, int len);
 
     void (*h263_v_loop_filter)(uint8_t *src, int stride, int qscale);
     void (*h263_h_loop_filter)(uint8_t *src, int stride, int qscale);