Commit bdc1e3e3 authored by Ronald S. Bultje's avatar Ronald S. Bultje Committed by Michael Niedermayer

vp9/x86: intra prediction sse2/32bit support.

Signed-off-by: 's avatarMichael Niedermayer <michaelni@gmx.at>
parent b6e17112
...@@ -40,7 +40,7 @@ DECLARE_ALIGNED(8, const uint64_t, ff_pw_53) = 0x0035003500350035ULL; ...@@ -40,7 +40,7 @@ DECLARE_ALIGNED(8, const uint64_t, ff_pw_53) = 0x0035003500350035ULL;
DECLARE_ALIGNED(16, const xmm_reg, ff_pw_64) = { 0x0040004000400040ULL, 0x0040004000400040ULL }; DECLARE_ALIGNED(16, const xmm_reg, ff_pw_64) = { 0x0040004000400040ULL, 0x0040004000400040ULL };
DECLARE_ALIGNED(8, const uint64_t, ff_pw_96) = 0x0060006000600060ULL; DECLARE_ALIGNED(8, const uint64_t, ff_pw_96) = 0x0060006000600060ULL;
DECLARE_ALIGNED(8, const uint64_t, ff_pw_128) = 0x0080008000800080ULL; DECLARE_ALIGNED(8, const uint64_t, ff_pw_128) = 0x0080008000800080ULL;
DECLARE_ALIGNED(8, const uint64_t, ff_pw_255) = 0x00ff00ff00ff00ffULL; DECLARE_ALIGNED(8, const xmm_reg, ff_pw_255) = { 0x00ff00ff00ff00ffULL, 0x00ff00ff00ff00ffULL };
DECLARE_ALIGNED(32, const ymm_reg, ff_pw_256) = { 0x0100010001000100ULL, 0x0100010001000100ULL, DECLARE_ALIGNED(32, const ymm_reg, ff_pw_256) = { 0x0100010001000100ULL, 0x0100010001000100ULL,
0x0100010001000100ULL, 0x0100010001000100ULL }; 0x0100010001000100ULL, 0x0100010001000100ULL };
DECLARE_ALIGNED(16, const xmm_reg, ff_pw_512) = { 0x0200020002000200ULL, 0x0200020002000200ULL }; DECLARE_ALIGNED(16, const xmm_reg, ff_pw_512) = { 0x0200020002000200ULL, 0x0200020002000200ULL };
......
...@@ -42,7 +42,7 @@ extern const uint64_t ff_pw_53; ...@@ -42,7 +42,7 @@ extern const uint64_t ff_pw_53;
extern const xmm_reg ff_pw_64; extern const xmm_reg ff_pw_64;
extern const uint64_t ff_pw_96; extern const uint64_t ff_pw_96;
extern const uint64_t ff_pw_128; extern const uint64_t ff_pw_128;
extern const uint64_t ff_pw_255; extern const xmm_reg ff_pw_255;
extern const xmm_reg ff_pw_512; extern const xmm_reg ff_pw_512;
extern const xmm_reg ff_pw_1024; extern const xmm_reg ff_pw_1024;
extern const xmm_reg ff_pw_2048; extern const xmm_reg ff_pw_2048;
......
...@@ -243,40 +243,58 @@ lpf_funcs(88, 16, avx); ...@@ -243,40 +243,58 @@ lpf_funcs(88, 16, avx);
void ff_vp9_ipred_##type##_##size##x##size##_##opt(uint8_t *dst, ptrdiff_t stride, \ void ff_vp9_ipred_##type##_##size##x##size##_##opt(uint8_t *dst, ptrdiff_t stride, \
const uint8_t *l, const uint8_t *a) const uint8_t *l, const uint8_t *a)
#define ipred_funcs(type, opt) \ ipred_func(8, v, mmx);
ipred_func(4, type, opt); \
ipred_func(8, type, opt); \
ipred_func(16, type, opt); \
ipred_func(32, type, opt)
ipred_funcs(dc, ssse3); #define ipred_dc_funcs(size, opt) \
ipred_funcs(dc_left, ssse3); ipred_func(size, dc, opt); \
ipred_funcs(dc_top, ssse3); ipred_func(size, dc_left, opt); \
ipred_func(size, dc_top, opt)
#undef ipred_funcs ipred_dc_funcs(4, mmxext);
ipred_dc_funcs(8, mmxext);
ipred_func(8, v, mmx); #define ipred_dir_tm_funcs(size, opt) \
ipred_func(16, v, sse2); ipred_func(size, tm, opt); \
ipred_func(32, v, sse2); ipred_func(size, dl, opt); \
ipred_func(size, dr, opt); \
#define ipred_func_set(size, type, opt1, opt2) \ ipred_func(size, hd, opt); \
ipred_func(size, type, opt1); \ ipred_func(size, hu, opt); \
ipred_func(size, type, opt2) ipred_func(size, vl, opt); \
ipred_func(size, vr, opt)
#define ipred_funcs(type, opt1, opt2) \
ipred_func(4, type, opt1); \ ipred_dir_tm_funcs(4, mmxext);
ipred_func_set(8, type, opt1, opt2); \
ipred_func_set(16, type, opt1, opt2); \ ipred_func(16, v, sse);
ipred_func_set(32, type, opt1, opt2) ipred_func(32, v, sse);
ipred_funcs(h, ssse3, avx); ipred_dc_funcs(16, sse2);
ipred_funcs(tm, ssse3, avx); ipred_dc_funcs(32, sse2);
ipred_funcs(dl, ssse3, avx);
ipred_funcs(dr, ssse3, avx); #define ipred_dir_tm_h_funcs(size, opt) \
ipred_funcs(hu, ssse3, avx); ipred_dir_tm_funcs(size, opt); \
ipred_funcs(hd, ssse3, avx); ipred_func(size, h, opt)
ipred_funcs(vl, ssse3, avx);
ipred_funcs(vr, ssse3, avx); ipred_dir_tm_h_funcs(8, sse2);
ipred_dir_tm_h_funcs(16, sse2);
ipred_dir_tm_h_funcs(32, sse2);
ipred_func(4, h, sse2);
#define ipred_all_funcs(size, opt) \
ipred_dc_funcs(size, opt); \
ipred_dir_tm_h_funcs(size, opt)
// FIXME hd/vl_4x4_ssse3 does not exist
ipred_all_funcs(4, ssse3);
ipred_all_funcs(8, ssse3);
ipred_all_funcs(16, ssse3);
ipred_all_funcs(32, ssse3);
ipred_dir_tm_h_funcs(8, avx);
ipred_dir_tm_h_funcs(16, avx);
ipred_dir_tm_h_funcs(32, avx);
ipred_func(32, v, avx);
ipred_func(32, dc, avx2); ipred_func(32, dc, avx2);
ipred_func(32, dc_left, avx2); ipred_func(32, dc_left, avx2);
...@@ -285,9 +303,14 @@ ipred_func(32, v, avx2); ...@@ -285,9 +303,14 @@ ipred_func(32, v, avx2);
ipred_func(32, h, avx2); ipred_func(32, h, avx2);
ipred_func(32, tm, avx2); ipred_func(32, tm, avx2);
#undef ipred_funcs ipred_dc_funcs(32, avx2);
#undef ipred_func_set ipred_func(32, h, avx2);
ipred_func(32, tm, avx2);
#undef ipred_func #undef ipred_func
#undef ipred_dir_tm_h_funcs
#undef ipred_dir_tm_funcs
#undef ipred_dc_funcs
#endif /* HAVE_YASM */ #endif /* HAVE_YASM */
...@@ -340,23 +363,32 @@ av_cold void ff_vp9dsp_init_x86(VP9DSPContext *dsp) ...@@ -340,23 +363,32 @@ av_cold void ff_vp9dsp_init_x86(VP9DSPContext *dsp)
} \ } \
} while (0) } while (0)
#define init_ipred(tx, sz, opt) do { \ #define init_ipred(sz, opt, t, e) \
dsp->intra_pred[tx][HOR_PRED] = ff_vp9_ipred_h_##sz##x##sz##_##opt; \ dsp->intra_pred[TX_##sz##X##sz][e##_PRED] = ff_vp9_ipred_##t##_##sz##x##sz##_##opt
dsp->intra_pred[tx][DIAG_DOWN_LEFT_PRED] = ff_vp9_ipred_dl_##sz##x##sz##_##opt; \
dsp->intra_pred[tx][DIAG_DOWN_RIGHT_PRED] = ff_vp9_ipred_dr_##sz##x##sz##_##opt; \ #define ff_vp9_ipred_hd_4x4_ssse3 ff_vp9_ipred_hd_4x4_mmxext
dsp->intra_pred[tx][HOR_DOWN_PRED] = ff_vp9_ipred_hd_##sz##x##sz##_##opt; \ #define ff_vp9_ipred_vl_4x4_ssse3 ff_vp9_ipred_vl_4x4_mmxext
dsp->intra_pred[tx][VERT_LEFT_PRED] = ff_vp9_ipred_vl_##sz##x##sz##_##opt; \ #define init_dir_tm_ipred(sz, opt) do { \
dsp->intra_pred[tx][HOR_UP_PRED] = ff_vp9_ipred_hu_##sz##x##sz##_##opt; \ init_ipred(sz, opt, dl, DIAG_DOWN_LEFT); \
if (ARCH_X86_64 || tx != TX_32X32) { \ init_ipred(sz, opt, dr, DIAG_DOWN_RIGHT); \
dsp->intra_pred[tx][VERT_RIGHT_PRED] = ff_vp9_ipred_vr_##sz##x##sz##_##opt; \ init_ipred(sz, opt, hd, HOR_DOWN); \
dsp->intra_pred[tx][TM_VP8_PRED] = ff_vp9_ipred_tm_##sz##x##sz##_##opt; \ init_ipred(sz, opt, vl, VERT_LEFT); \
} \ init_ipred(sz, opt, hu, HOR_UP); \
init_ipred(sz, opt, tm, TM_VP8); \
init_ipred(sz, opt, vr, VERT_RIGHT); \
} while (0)
#define init_dir_tm_h_ipred(sz, opt) do { \
init_dir_tm_ipred(sz, opt); \
init_ipred(sz, opt, h, HOR); \
} while (0)
#define init_dc_ipred(sz, opt) do { \
init_ipred(sz, opt, dc, DC); \
init_ipred(sz, opt, dc_left, LEFT_DC); \
init_ipred(sz, opt, dc_top, TOP_DC); \
} while (0) } while (0)
#define init_dc_ipred(tx, sz, opt) do { \ #define init_all_ipred(sz, opt) do { \
init_ipred(tx, sz, opt); \ init_dc_ipred(sz, opt); \
dsp->intra_pred[tx][DC_PRED] = ff_vp9_ipred_dc_##sz##x##sz##_##opt; \ init_dir_tm_h_ipred(sz, opt); \
dsp->intra_pred[tx][LEFT_DC_PRED] = ff_vp9_ipred_dc_left_##sz##x##sz##_##opt; \
dsp->intra_pred[tx][TOP_DC_PRED] = ff_vp9_ipred_dc_top_##sz##x##sz##_##opt; \
} while (0) } while (0)
if (EXTERNAL_MMX(cpu_flags)) { if (EXTERNAL_MMX(cpu_flags)) {
...@@ -366,7 +398,7 @@ av_cold void ff_vp9dsp_init_x86(VP9DSPContext *dsp) ...@@ -366,7 +398,7 @@ av_cold void ff_vp9dsp_init_x86(VP9DSPContext *dsp)
dsp->itxfm_add[4 /* lossless */][ADST_DCT] = dsp->itxfm_add[4 /* lossless */][ADST_DCT] =
dsp->itxfm_add[4 /* lossless */][DCT_ADST] = dsp->itxfm_add[4 /* lossless */][DCT_ADST] =
dsp->itxfm_add[4 /* lossless */][ADST_ADST] = ff_vp9_iwht_iwht_4x4_add_mmx; dsp->itxfm_add[4 /* lossless */][ADST_ADST] = ff_vp9_iwht_iwht_4x4_add_mmx;
dsp->intra_pred[TX_8X8][VERT_PRED] = ff_vp9_ipred_v_8x8_mmx; init_ipred(8, mmx, v, VERT);
} }
if (EXTERNAL_MMXEXT(cpu_flags)) { if (EXTERNAL_MMXEXT(cpu_flags)) {
...@@ -375,12 +407,17 @@ av_cold void ff_vp9dsp_init_x86(VP9DSPContext *dsp) ...@@ -375,12 +407,17 @@ av_cold void ff_vp9dsp_init_x86(VP9DSPContext *dsp)
init_fpel(4, 1, 4, avg, mmxext); init_fpel(4, 1, 4, avg, mmxext);
init_fpel(3, 1, 8, avg, mmxext); init_fpel(3, 1, 8, avg, mmxext);
dsp->itxfm_add[TX_4X4][DCT_DCT] = ff_vp9_idct_idct_4x4_add_mmxext; dsp->itxfm_add[TX_4X4][DCT_DCT] = ff_vp9_idct_idct_4x4_add_mmxext;
init_dc_ipred(4, mmxext);
init_dc_ipred(8, mmxext);
init_dir_tm_ipred(4, mmxext);
} }
if (EXTERNAL_SSE(cpu_flags)) { if (EXTERNAL_SSE(cpu_flags)) {
init_fpel(2, 0, 16, put, sse); init_fpel(2, 0, 16, put, sse);
init_fpel(1, 0, 32, put, sse); init_fpel(1, 0, 32, put, sse);
init_fpel(0, 0, 64, put, sse); init_fpel(0, 0, 64, put, sse);
init_ipred(16, sse, v, VERT);
init_ipred(32, sse, v, VERT);
} }
if (EXTERNAL_SSE2(cpu_flags)) { if (EXTERNAL_SSE2(cpu_flags)) {
...@@ -405,8 +442,12 @@ av_cold void ff_vp9dsp_init_x86(VP9DSPContext *dsp) ...@@ -405,8 +442,12 @@ av_cold void ff_vp9dsp_init_x86(VP9DSPContext *dsp)
dsp->itxfm_add[TX_32X32][ADST_DCT] = dsp->itxfm_add[TX_32X32][ADST_DCT] =
dsp->itxfm_add[TX_32X32][DCT_ADST] = dsp->itxfm_add[TX_32X32][DCT_ADST] =
dsp->itxfm_add[TX_32X32][DCT_DCT] = ff_vp9_idct_idct_32x32_add_sse2; dsp->itxfm_add[TX_32X32][DCT_DCT] = ff_vp9_idct_idct_32x32_add_sse2;
dsp->intra_pred[TX_16X16][VERT_PRED] = ff_vp9_ipred_v_16x16_sse2; init_dc_ipred(16, sse2);
dsp->intra_pred[TX_32X32][VERT_PRED] = ff_vp9_ipred_v_32x32_sse2; init_dc_ipred(32, sse2);
init_dir_tm_h_ipred(8, sse2);
init_dir_tm_h_ipred(16, sse2);
init_dir_tm_h_ipred(32, sse2);
init_ipred(4, sse2, h, HOR);
} }
if (EXTERNAL_SSSE3(cpu_flags)) { if (EXTERNAL_SSSE3(cpu_flags)) {
...@@ -429,10 +470,10 @@ av_cold void ff_vp9dsp_init_x86(VP9DSPContext *dsp) ...@@ -429,10 +470,10 @@ av_cold void ff_vp9dsp_init_x86(VP9DSPContext *dsp)
dsp->itxfm_add[TX_32X32][DCT_ADST] = dsp->itxfm_add[TX_32X32][DCT_ADST] =
dsp->itxfm_add[TX_32X32][DCT_DCT] = ff_vp9_idct_idct_32x32_add_ssse3; dsp->itxfm_add[TX_32X32][DCT_DCT] = ff_vp9_idct_idct_32x32_add_ssse3;
init_lpf(ssse3); init_lpf(ssse3);
init_dc_ipred(TX_4X4, 4, ssse3); init_all_ipred(4, ssse3);
init_dc_ipred(TX_8X8, 8, ssse3); init_all_ipred(8, ssse3);
init_dc_ipred(TX_16X16, 16, ssse3); init_all_ipred(16, ssse3);
init_dc_ipred(TX_32X32, 32, ssse3); init_all_ipred(32, ssse3);
} }
if (EXTERNAL_AVX(cpu_flags)) { if (EXTERNAL_AVX(cpu_flags)) {
...@@ -451,9 +492,10 @@ av_cold void ff_vp9dsp_init_x86(VP9DSPContext *dsp) ...@@ -451,9 +492,10 @@ av_cold void ff_vp9dsp_init_x86(VP9DSPContext *dsp)
init_fpel(1, 0, 32, put, avx); init_fpel(1, 0, 32, put, avx);
init_fpel(0, 0, 64, put, avx); init_fpel(0, 0, 64, put, avx);
init_lpf(avx); init_lpf(avx);
init_ipred(TX_8X8, 8, avx); init_dir_tm_h_ipred(8, avx);
init_ipred(TX_16X16, 16, avx); init_dir_tm_h_ipred(16, avx);
init_ipred(TX_32X32, 32, avx); init_dir_tm_h_ipred(32, avx);
init_ipred(32, avx, v, VERT);
} }
if (EXTERNAL_AVX2(cpu_flags)) { if (EXTERNAL_AVX2(cpu_flags)) {
...@@ -465,12 +507,9 @@ av_cold void ff_vp9dsp_init_x86(VP9DSPContext *dsp) ...@@ -465,12 +507,9 @@ av_cold void ff_vp9dsp_init_x86(VP9DSPContext *dsp)
init_subpel3_32_64(1, avg, avx2); init_subpel3_32_64(1, avg, avx2);
#endif #endif
} }
dsp->intra_pred[TX_32X32][DC_PRED] = ff_vp9_ipred_dc_32x32_avx2; init_dc_ipred(32, avx2);
dsp->intra_pred[TX_32X32][LEFT_DC_PRED] = ff_vp9_ipred_dc_left_32x32_avx2; init_ipred(32, avx2, h, HOR);
dsp->intra_pred[TX_32X32][TOP_DC_PRED] = ff_vp9_ipred_dc_top_32x32_avx2; init_ipred(32, avx2, tm, TM_VP8);
dsp->intra_pred[TX_32X32][VERT_PRED] = ff_vp9_ipred_v_32x32_avx2;
dsp->intra_pred[TX_32X32][HOR_PRED] = ff_vp9_ipred_h_32x32_avx2;
dsp->intra_pred[TX_32X32][TM_VP8_PRED] = ff_vp9_ipred_tm_32x32_avx2;
} }
#undef init_fpel #undef init_fpel
......
This diff is collapsed.
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment