#include <stdint.h>

typedef struct DSPContext {
    void (*h263_h_loop_filter)(uint8_t *src, int stride, int qscale);
} DSPContext;


static inline void transpose4x4(uint8_t *dst, uint8_t *src, int dst_stride, int src_stride){
    asm volatile( //FIXME could save 1 instruction if done as 8x4 ...
        "movd  %4, %%mm0		\n\t"
        "movd  %5, %%mm1		\n\t"
        "movd  %6, %%mm2		\n\t"
        "movd  %7, %%mm3		\n\t"
        "punpcklbw %%mm1, %%mm0		\n\t"
        "punpcklbw %%mm3, %%mm2		\n\t"
        "movq %%mm0, %%mm1		\n\t"
        "punpcklwd %%mm2, %%mm0		\n\t"
        "punpckhwd %%mm2, %%mm1		\n\t"
        "movd  %%mm0, %0		\n\t"
        "punpckhdq %%mm0, %%mm0		\n\t"
        "movd  %%mm0, %1		\n\t"
        "movd  %%mm1, %2		\n\t"
        "punpckhdq %%mm1, %%mm1		\n\t"
        "movd  %%mm1, %3		\n\t"
        
        : "=m" (*(uint32_t*)(dst + 0*dst_stride)),
          "=m" (*(uint32_t*)(dst + 1*dst_stride)),
          "=m" (*(uint32_t*)(dst + 2*dst_stride)),
          "=m" (*(uint32_t*)(dst + 3*dst_stride))
        :  "m" (*(uint32_t*)(src + 0*src_stride)),
           "m" (*(uint32_t*)(src + 1*src_stride)),
           "m" (*(uint32_t*)(src + 2*src_stride)),
           "m" (*(uint32_t*)(src + 3*src_stride))
    );
}

static void h263_h_loop_filter_mmx(uint8_t *src, int stride, int qscale){
    uint64_t temp[4] __attribute__ ((aligned(8)));
    uint8_t *btemp= (uint8_t*)temp;
    
    src -= 2;

    transpose4x4(btemp  , src           , 8, stride);

    asm volatile(
        "movq %%mm5, %%mm1		\n\t"
        "movq %%mm4, %%mm0		\n\t"
        "punpcklbw %%mm3, %%mm5		\n\t"
        "punpcklbw %%mm6, %%mm4		\n\t"
        "punpckhbw %%mm3, %%mm1		\n\t"
        "punpckhbw %%mm6, %%mm0		\n\t"
        "movq %%mm5, %%mm3		\n\t"
        "movq %%mm1, %%mm6		\n\t"
        "punpcklwd %%mm4, %%mm5		\n\t"
        "punpcklwd %%mm0, %%mm1		\n\t"
        "punpckhwd %%mm4, %%mm3		\n\t"
        "punpckhwd %%mm0, %%mm6		\n\t"
        "movd %%mm5, %0			\n\t"
        "punpckhdq %%mm5, %%mm5		\n\t"
        "movd %%mm5, %1			\n\t"
        "movd %%mm3, %2			\n\t"
        "punpckhdq %%mm3, %%mm3		\n\t"
        "movd %%mm3, %3			\n\t"
        "movd %%mm1, %4			\n\t"
        "punpckhdq %%mm1, %%mm1		\n\t"
        "movd %%mm1, %5			\n\t"
        "movd %%mm6, %6			\n\t"
        "punpckhdq %%mm6, %%mm6		\n\t"
        "movd %%mm6, %7			\n\t"
        : "=m" (*(uint32_t*)(src + 0*stride)),
          "=m" (*(uint32_t*)(src + 1*stride)),
          "=m" (*(uint32_t*)(src + 2*stride)),
          "=m" (*(uint32_t*)(src + 3*stride)),
          "=m" (*(uint32_t*)(src + 4*stride)),
          "=m" (*(uint32_t*)(src + 5*stride)),
          "=m" (*(uint32_t*)(src + 6*stride)),
          "=m" (*(uint32_t*)(src + 7*stride))
    );
}

void dsputil_init_mmx(DSPContext* c)
{
        c->h263_h_loop_filter= h263_h_loop_filter_mmx;
}