int blocksize)
{
int i;
- vector float m, a, s0, s1;
+ vector float m, a;
vector bool int t0, t1;
const vector unsigned int v_31 = //XXX
vec_add(vec_add(vec_splat_u32(15),vec_splat_u32(15)),vec_splat_u32(1));
t0 = vec_cmple(m, (vector float)vec_splat_u32(0));
t1 = vec_cmple(a, (vector float)vec_splat_u32(0));
a = vec_xor(a, (vector float) vec_sl((vector unsigned int)t0, v_31));
- s0 = vec_and(a, t1);
- s1 = vec_andc(a, t1);
- a = vec_add(m, s0);
- m = vec_sub(m, s1);
+ t0 = (vector bool int)vec_and(a, t1);
+ t1 = (vector bool int)vec_andc(a, t1);
+ a = vec_add(m, (vector float)t0);
+ m = vec_sub(m, (vector float)t1);
vec_ste(a, 0, ang+i);
vec_ste(m, 0, mag+i);
}
POWERPC_PERF_STOP_COUNT(altivec_avg_pixels8_xy2_num, 1);
#endif /* ALTIVEC_USE_REFERENCE_C_CODE */
}
+
+void dsputil_init_altivec(DSPContext* c, AVCodecContext *avctx)
+{
+ c->pix_abs[0][1] = sad16_x2_altivec;
+ c->pix_abs[0][2] = sad16_y2_altivec;
+ c->pix_abs[0][3] = sad16_xy2_altivec;
+ c->pix_abs[0][0] = sad16_altivec;
+ c->pix_abs[1][0] = sad8_altivec;
+ c->sad[0]= sad16_altivec;
+ c->sad[1]= sad8_altivec;
+ c->pix_norm1 = pix_norm1_altivec;
+ c->sse[1]= sse8_altivec;
+ c->sse[0]= sse16_altivec;
+ c->pix_sum = pix_sum_altivec;
+ c->diff_pixels = diff_pixels_altivec;
+ c->get_pixels = get_pixels_altivec;
+// next one disabled as it's untested.
+#if 0
+ c->add_bytes= add_bytes_altivec;
+#endif /* 0 */
+ c->put_pixels_tab[0][0] = put_pixels16_altivec;
+ /* the two functions do the same thing, so use the same code */
+ c->put_no_rnd_pixels_tab[0][0] = put_pixels16_altivec;
+ c->avg_pixels_tab[0][0] = avg_pixels16_altivec;
+ c->avg_pixels_tab[1][0] = avg_pixels8_altivec;
+ c->avg_pixels_tab[1][3] = avg_pixels8_xy2_altivec;
+ c->put_pixels_tab[1][3] = put_pixels8_xy2_altivec;
+ c->put_no_rnd_pixels_tab[1][3] = put_no_rnd_pixels8_xy2_altivec;
+ c->put_pixels_tab[0][3] = put_pixels16_xy2_altivec;
+ c->put_no_rnd_pixels_tab[0][3] = put_no_rnd_pixels16_xy2_altivec;
+
+ c->hadamard8_diff[0] = hadamard8_diff16_altivec;
+ c->hadamard8_diff[1] = hadamard8_diff8x8_altivec;
+ c->vorbis_inverse_coupling = vorbis_inverse_coupling_altivec;
+}
#ifdef HAVE_ALTIVEC
-extern int sad16_x2_altivec(void *v, uint8_t *pix1, uint8_t *pix2, int line_size, int h);
-extern int sad16_y2_altivec(void *v, uint8_t *pix1, uint8_t *pix2, int line_size, int h);
-extern int sad16_xy2_altivec(void *v, uint8_t *pix1, uint8_t *pix2, int line_size, int h);
-extern int sad16_altivec(void *v, uint8_t *pix1, uint8_t *pix2, int line_size, int h);
-extern int sad8_altivec(void *v, uint8_t *pix1, uint8_t *pix2, int line_size, int h);
-extern int pix_norm1_altivec(uint8_t *pix, int line_size);
-extern int sse8_altivec(void *v, uint8_t *pix1, uint8_t *pix2, int line_size, int h);
-extern int sse16_altivec(void *v, uint8_t *pix1, uint8_t *pix2, int line_size, int h);
-extern int pix_sum_altivec(uint8_t * pix, int line_size);
-extern void diff_pixels_altivec(DCTELEM* block, const uint8_t* s1, const uint8_t* s2, int stride);
-extern void get_pixels_altivec(DCTELEM* block, const uint8_t * pixels, int line_size);
-
-extern void add_bytes_altivec(uint8_t *dst, uint8_t *src, int w);
-extern void put_pixels_clamped_altivec(const DCTELEM *block, uint8_t *restrict pixels, int line_size);
-extern void put_pixels16_altivec(uint8_t *block, const uint8_t *pixels, int line_size, int h);
-extern void avg_pixels16_altivec(uint8_t *block, const uint8_t *pixels, int line_size, int h);
-extern void avg_pixels8_altivec(uint8_t * block, const uint8_t * pixels, int line_size, int h);
-extern void put_pixels8_xy2_altivec(uint8_t *block, const uint8_t *pixels, int line_size, int h);
-extern void put_no_rnd_pixels8_xy2_altivec(uint8_t *block, const uint8_t *pixels, int line_size, int h);
-extern void put_pixels16_xy2_altivec(uint8_t * block, const uint8_t * pixels, int line_size, int h);
-extern void put_no_rnd_pixels16_xy2_altivec(uint8_t * block, const uint8_t * pixels, int line_size, int h);
-extern int hadamard8_diff8x8_altivec(/*MpegEncContext*/ void *s, uint8_t *dst, uint8_t *src, int stride, int h);
-extern int hadamard8_diff16_altivec(/*MpegEncContext*/ void *s, uint8_t *dst, uint8_t *src, int stride, int h);
-extern void avg_pixels8_xy2_altivec(uint8_t *block, const uint8_t *pixels, int line_size, int h);
-
-extern void gmc1_altivec(uint8_t *dst, uint8_t *src, int stride, int h, int x16, int y16, int rounder);
-
extern int has_altivec(void);
// used to build registers permutation vectors (vcprm)
#include "dsputil_altivec.h"
extern void fdct_altivec(int16_t *block);
+extern void gmc1_altivec(uint8_t *dst, uint8_t *src, int stride, int h,
+ int x16, int y16, int rounder);
extern void idct_put_altivec(uint8_t *dest, int line_size, int16_t *block);
extern void idct_add_altivec(uint8_t *dest, int line_size, int16_t *block);
extern void ff_snow_horizontal_compose97i_altivec(DWTELEM *b, int width);
int src_stride, slice_buffer * sb,
int add, uint8_t * dst8);
+void dsputil_init_altivec(DSPContext* c, AVCodecContext *avctx);
void dsputil_h264_init_ppc(DSPContext* c, AVCodecContext *avctx);
void vc1dsp_init_altivec(DSPContext* c, AVCodecContext *avctx);
{
// Common optimizations whether Altivec is available or not
- switch (check_dcbzl_effect()) {
- case 32:
- c->clear_blocks = clear_blocks_dcbz32_ppc;
- break;
- case 128:
- c->clear_blocks = clear_blocks_dcbz128_ppc;
- break;
- default:
- break;
- }
+ switch (check_dcbzl_effect()) {
+ case 32:
+ c->clear_blocks = clear_blocks_dcbz32_ppc;
+ break;
+ case 128:
+ c->clear_blocks = clear_blocks_dcbz128_ppc;
+ break;
+ default:
+ break;
+ }
#ifdef HAVE_ALTIVEC
- dsputil_h264_init_ppc(c, avctx);
+ dsputil_h264_init_ppc(c, avctx);
if (has_altivec()) {
mm_flags |= MM_ALTIVEC;
- // Altivec specific optimisations
- c->pix_abs[0][1] = sad16_x2_altivec;
- c->pix_abs[0][2] = sad16_y2_altivec;
- c->pix_abs[0][3] = sad16_xy2_altivec;
- c->pix_abs[0][0] = sad16_altivec;
- c->pix_abs[1][0] = sad8_altivec;
- c->sad[0]= sad16_altivec;
- c->sad[1]= sad8_altivec;
- c->pix_norm1 = pix_norm1_altivec;
- c->sse[1]= sse8_altivec;
- c->sse[0]= sse16_altivec;
- c->pix_sum = pix_sum_altivec;
- c->diff_pixels = diff_pixels_altivec;
- c->get_pixels = get_pixels_altivec;
-// next one disabled as it's untested.
-#if 0
- c->add_bytes= add_bytes_altivec;
-#endif /* 0 */
- c->put_pixels_tab[0][0] = put_pixels16_altivec;
- /* the two functions do the same thing, so use the same code */
- c->put_no_rnd_pixels_tab[0][0] = put_pixels16_altivec;
- c->avg_pixels_tab[0][0] = avg_pixels16_altivec;
- c->avg_pixels_tab[1][0] = avg_pixels8_altivec;
- c->avg_pixels_tab[1][3] = avg_pixels8_xy2_altivec;
- c->put_pixels_tab[1][3] = put_pixels8_xy2_altivec;
- c->put_no_rnd_pixels_tab[1][3] = put_no_rnd_pixels8_xy2_altivec;
- c->put_pixels_tab[0][3] = put_pixels16_xy2_altivec;
- c->put_no_rnd_pixels_tab[0][3] = put_no_rnd_pixels16_xy2_altivec;
+ vc1dsp_init_altivec(c, avctx);
+ dsputil_init_altivec(c, avctx);
c->gmc1 = gmc1_altivec;
- c->hadamard8_diff[0] = hadamard8_diff16_altivec;
- c->hadamard8_diff[1] = hadamard8_diff8x8_altivec;
-
c->horizontal_compose97i = ff_snow_horizontal_compose97i_altivec;
c->vertical_compose97i = ff_snow_vertical_compose97i_altivec;
c->inner_add_yblock = ff_snow_inner_add_yblock_altivec;
- vc1dsp_init_altivec(c, avctx);
#ifdef CONFIG_ENCODERS
if (avctx->dct_algo == FF_DCT_AUTO ||
avctx->dct_algo == FF_DCT_ALTIVEC)
}
#endif //CONFIG_ENCODERS
- if (avctx->lowres==0)
- {
+ if (avctx->lowres==0)
+ {
if ((avctx->idct_algo == FF_IDCT_AUTO) ||
(avctx->idct_algo == FF_IDCT_ALTIVEC))
{
c->idct_put = idct_put_altivec;
c->idct_add = idct_add_altivec;
-#ifndef ALTIVEC_USE_REFERENCE_C_CODE
+ #ifndef ALTIVEC_USE_REFERENCE_C_CODE
c->idct_permutation_type = FF_TRANSPOSE_IDCT_PERM;
-#else /* ALTIVEC_USE_REFERENCE_C_CODE */
+ #else /* ALTIVEC_USE_REFERENCE_C_CODE */
c->idct_permutation_type = FF_NO_IDCT_PERM;
-#endif /* ALTIVEC_USE_REFERENCE_C_CODE */
+ #endif /* ALTIVEC_USE_REFERENCE_C_CODE */
+ }
}
- }
#ifdef POWERPC_PERFORMANCE_REPORT
{