Home
last modified time | relevance | path

Searched refs:_mm_add_epi64 (Results 1 – 25 of 216) sorted by relevance

123456789

/external/libavc/encoder/x86/
Dime_distortion_metrics_sse42.c131 sad_val = _mm_add_epi64(res_r0, res_r1); in ime_compute_sad_16x16_sse42()
132 sad_val = _mm_add_epi64(sad_val, res_r2); in ime_compute_sad_16x16_sse42()
133 sad_val = _mm_add_epi64(sad_val, res_r3); in ime_compute_sad_16x16_sse42()
154 sad_val = _mm_add_epi64(sad_val, res_r0); in ime_compute_sad_16x16_sse42()
155 sad_val = _mm_add_epi64(sad_val, res_r1); in ime_compute_sad_16x16_sse42()
156 sad_val = _mm_add_epi64(sad_val, res_r2); in ime_compute_sad_16x16_sse42()
157 sad_val = _mm_add_epi64(sad_val, res_r3); in ime_compute_sad_16x16_sse42()
177 sad_val = _mm_add_epi64(sad_val, res_r0); in ime_compute_sad_16x16_sse42()
178 sad_val = _mm_add_epi64(sad_val, res_r1); in ime_compute_sad_16x16_sse42()
179 sad_val = _mm_add_epi64(sad_val, res_r2); in ime_compute_sad_16x16_sse42()
[all …]
/external/rust/crates/quiche/deps/boringssl/src/crypto/poly1305/
Dpoly1305_vec.c267 T0 = _mm_add_epi64(T0, T5); in poly1305_blocks()
268 T1 = _mm_add_epi64(T1, T6); in poly1305_blocks()
271 T0 = _mm_add_epi64(T0, T5); in poly1305_blocks()
272 T1 = _mm_add_epi64(T1, T6); in poly1305_blocks()
275 T0 = _mm_add_epi64(T0, T5); in poly1305_blocks()
276 T1 = _mm_add_epi64(T1, T6); in poly1305_blocks()
279 T0 = _mm_add_epi64(T0, T5); in poly1305_blocks()
280 T1 = _mm_add_epi64(T1, T6); in poly1305_blocks()
283 T2 = _mm_add_epi64(T2, T5); in poly1305_blocks()
284 T3 = _mm_add_epi64(T3, T6); in poly1305_blocks()
[all …]
/external/boringssl/src/crypto/poly1305/
Dpoly1305_vec.c271 T0 = _mm_add_epi64(T0, T5); in poly1305_blocks()
272 T1 = _mm_add_epi64(T1, T6); in poly1305_blocks()
275 T0 = _mm_add_epi64(T0, T5); in poly1305_blocks()
276 T1 = _mm_add_epi64(T1, T6); in poly1305_blocks()
279 T0 = _mm_add_epi64(T0, T5); in poly1305_blocks()
280 T1 = _mm_add_epi64(T1, T6); in poly1305_blocks()
283 T0 = _mm_add_epi64(T0, T5); in poly1305_blocks()
284 T1 = _mm_add_epi64(T1, T6); in poly1305_blocks()
287 T2 = _mm_add_epi64(T2, T5); in poly1305_blocks()
288 T3 = _mm_add_epi64(T3, T6); in poly1305_blocks()
[all …]
/external/rust/crates/ring/crypto/poly1305/
Dpoly1305_vec.c275 T0 = _mm_add_epi64(T0, T5); in poly1305_blocks()
276 T1 = _mm_add_epi64(T1, T6); in poly1305_blocks()
279 T0 = _mm_add_epi64(T0, T5); in poly1305_blocks()
280 T1 = _mm_add_epi64(T1, T6); in poly1305_blocks()
283 T0 = _mm_add_epi64(T0, T5); in poly1305_blocks()
284 T1 = _mm_add_epi64(T1, T6); in poly1305_blocks()
287 T0 = _mm_add_epi64(T0, T5); in poly1305_blocks()
288 T1 = _mm_add_epi64(T1, T6); in poly1305_blocks()
291 T2 = _mm_add_epi64(T2, T5); in poly1305_blocks()
292 T3 = _mm_add_epi64(T3, T6); in poly1305_blocks()
[all …]
/external/libvpx/libvpx/vp9/common/x86/
Dvp9_highbd_iht16x16_add_sse4.c50 s0[0] = _mm_add_epi64(t00[0], t11[0]); in highbd_iadst_butterfly_sse4_1()
51 s0[1] = _mm_add_epi64(t00[1], t11[1]); in highbd_iadst_butterfly_sse4_1()
75 x0[0] = _mm_add_epi64(s0[0], s8[0]); in highbd_iadst16_4col_sse4_1()
76 x0[1] = _mm_add_epi64(s0[1], s8[1]); in highbd_iadst16_4col_sse4_1()
77 x1[0] = _mm_add_epi64(s1[0], s9[0]); in highbd_iadst16_4col_sse4_1()
78 x1[1] = _mm_add_epi64(s1[1], s9[1]); in highbd_iadst16_4col_sse4_1()
79 x2[0] = _mm_add_epi64(s2[0], s10[0]); in highbd_iadst16_4col_sse4_1()
80 x2[1] = _mm_add_epi64(s2[1], s10[1]); in highbd_iadst16_4col_sse4_1()
81 x3[0] = _mm_add_epi64(s3[0], s11[0]); in highbd_iadst16_4col_sse4_1()
82 x3[1] = _mm_add_epi64(s3[1], s11[1]); in highbd_iadst16_4col_sse4_1()
[all …]
Dvp9_highbd_iht4x4_add_sse4.c50 t0[0] = _mm_add_epi64(s0[0], s3[0]); in highbd_iadst4_sse4_1()
51 t0[1] = _mm_add_epi64(s0[1], s3[1]); in highbd_iadst4_sse4_1()
52 t0[0] = _mm_add_epi64(t0[0], s5[0]); in highbd_iadst4_sse4_1()
53 t0[1] = _mm_add_epi64(t0[1], s5[1]); in highbd_iadst4_sse4_1()
64 s0[0] = _mm_add_epi64(t0[0], s2[0]); in highbd_iadst4_sse4_1()
65 s0[1] = _mm_add_epi64(t0[1], s2[1]); in highbd_iadst4_sse4_1()
66 s1[0] = _mm_add_epi64(t1[0], s2[0]); in highbd_iadst4_sse4_1()
67 s1[1] = _mm_add_epi64(t1[1], s2[1]); in highbd_iadst4_sse4_1()
68 s3[0] = _mm_add_epi64(t0[0], t1[0]); in highbd_iadst4_sse4_1()
69 s3[1] = _mm_add_epi64(t0[1], t1[1]); in highbd_iadst4_sse4_1()
Dvp9_highbd_iht8x8_add_sse4.c50 s0[0] = _mm_add_epi64(t00[0], t11[0]); in highbd_iadst_butterfly_sse4_1()
51 s0[1] = _mm_add_epi64(t00[1], t11[1]); in highbd_iadst_butterfly_sse4_1()
65 x0[0] = _mm_add_epi64(s0[0], s4[0]); in highbd_iadst8_sse4_1()
66 x0[1] = _mm_add_epi64(s0[1], s4[1]); in highbd_iadst8_sse4_1()
67 x1[0] = _mm_add_epi64(s1[0], s5[0]); in highbd_iadst8_sse4_1()
68 x1[1] = _mm_add_epi64(s1[1], s5[1]); in highbd_iadst8_sse4_1()
76 x2[0] = _mm_add_epi64(s2[0], s6[0]); in highbd_iadst8_sse4_1()
77 x2[1] = _mm_add_epi64(s2[1], s6[1]); in highbd_iadst8_sse4_1()
78 x3[0] = _mm_add_epi64(s3[0], s7[0]); in highbd_iadst8_sse4_1()
79 x3[1] = _mm_add_epi64(s3[1], s7[1]); in highbd_iadst8_sse4_1()
[all …]
/external/XNNPACK/src/qs8-requantization/
Dq31-sse4.c67 const __m128i x_product_even = _mm_add_epi64(_mm_mul_epi32(x, vmultiplier), vq31rounding); in xnn_qs8_requantize_q31__sse4()
68 const __m128i y_product_even = _mm_add_epi64(_mm_mul_epi32(y, vmultiplier), vq31rounding); in xnn_qs8_requantize_q31__sse4()
69 const __m128i z_product_even = _mm_add_epi64(_mm_mul_epi32(z, vmultiplier), vq31rounding); in xnn_qs8_requantize_q31__sse4()
70 const __m128i w_product_even = _mm_add_epi64(_mm_mul_epi32(w, vmultiplier), vq31rounding); in xnn_qs8_requantize_q31__sse4()
72 const __m128i x_product_odd = _mm_add_epi64(_mm_mul_epi32(x_rev, vmultiplier), vq31rounding); in xnn_qs8_requantize_q31__sse4()
73 const __m128i y_product_odd = _mm_add_epi64(_mm_mul_epi32(y_rev, vmultiplier), vq31rounding); in xnn_qs8_requantize_q31__sse4()
74 const __m128i z_product_odd = _mm_add_epi64(_mm_mul_epi32(z_rev, vmultiplier), vq31rounding); in xnn_qs8_requantize_q31__sse4()
75 const __m128i w_product_odd = _mm_add_epi64(_mm_mul_epi32(w_rev, vmultiplier), vq31rounding); in xnn_qs8_requantize_q31__sse4()
78 const __m128i x_q31product_odd = _mm_add_epi64(x_product_odd, x_product_odd); in xnn_qs8_requantize_q31__sse4()
80 const __m128i y_q31product_odd = _mm_add_epi64(y_product_odd, y_product_odd); in xnn_qs8_requantize_q31__sse4()
[all …]
Dprecise-sse4.c74 const __m128i x_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(x_absmul02, vrounding), vshiftlo); in xnn_qs8_requantize_precise__sse4()
75 const __m128i x_abs_scaled13 = _mm_srl_epi32(_mm_add_epi64(x_absmul13, vrounding), vshifthi); in xnn_qs8_requantize_precise__sse4()
76 const __m128i y_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(y_absmul02, vrounding), vshiftlo); in xnn_qs8_requantize_precise__sse4()
77 const __m128i y_abs_scaled13 = _mm_srl_epi32(_mm_add_epi64(y_absmul13, vrounding), vshifthi); in xnn_qs8_requantize_precise__sse4()
78 const __m128i z_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(z_absmul02, vrounding), vshiftlo); in xnn_qs8_requantize_precise__sse4()
79 const __m128i z_abs_scaled13 = _mm_srl_epi32(_mm_add_epi64(z_absmul13, vrounding), vshifthi); in xnn_qs8_requantize_precise__sse4()
80 const __m128i w_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(w_absmul02, vrounding), vshiftlo); in xnn_qs8_requantize_precise__sse4()
81 const __m128i w_abs_scaled13 = _mm_srl_epi32(_mm_add_epi64(w_absmul13, vrounding), vshifthi); in xnn_qs8_requantize_precise__sse4()
Dprecise-ssse3.c73 const __m128i x_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(x_absmul02, vrounding), vshift); in xnn_qs8_requantize_precise__ssse3()
74 const __m128i x_abs_scaled13 = _mm_srl_epi64(_mm_add_epi64(x_absmul13, vrounding), vshift); in xnn_qs8_requantize_precise__ssse3()
75 const __m128i y_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(y_absmul02, vrounding), vshift); in xnn_qs8_requantize_precise__ssse3()
76 const __m128i y_abs_scaled13 = _mm_srl_epi64(_mm_add_epi64(y_absmul13, vrounding), vshift); in xnn_qs8_requantize_precise__ssse3()
77 const __m128i z_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(z_absmul02, vrounding), vshift); in xnn_qs8_requantize_precise__ssse3()
78 const __m128i z_abs_scaled13 = _mm_srl_epi64(_mm_add_epi64(z_absmul13, vrounding), vshift); in xnn_qs8_requantize_precise__ssse3()
79 const __m128i w_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(w_absmul02, vrounding), vshift); in xnn_qs8_requantize_precise__ssse3()
80 const __m128i w_abs_scaled13 = _mm_srl_epi64(_mm_add_epi64(w_absmul13, vrounding), vshift); in xnn_qs8_requantize_precise__ssse3()
Dprecise-sse2.c78 const __m128i x_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(x_absmul02, vrounding), vshift); in xnn_qs8_requantize_precise__sse2()
79 const __m128i x_abs_scaled13 = _mm_srl_epi64(_mm_add_epi64(x_absmul13, vrounding), vshift); in xnn_qs8_requantize_precise__sse2()
80 const __m128i y_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(y_absmul02, vrounding), vshift); in xnn_qs8_requantize_precise__sse2()
81 const __m128i y_abs_scaled13 = _mm_srl_epi64(_mm_add_epi64(y_absmul13, vrounding), vshift); in xnn_qs8_requantize_precise__sse2()
82 const __m128i z_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(z_absmul02, vrounding), vshift); in xnn_qs8_requantize_precise__sse2()
83 const __m128i z_abs_scaled13 = _mm_srl_epi64(_mm_add_epi64(z_absmul13, vrounding), vshift); in xnn_qs8_requantize_precise__sse2()
84 const __m128i w_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(w_absmul02, vrounding), vshift); in xnn_qs8_requantize_precise__sse2()
85 const __m128i w_abs_scaled13 = _mm_srl_epi64(_mm_add_epi64(w_absmul13, vrounding), vshift); in xnn_qs8_requantize_precise__sse2()
/external/XNNPACK/src/qu8-requantization/
Dq31-sse4.c67 const __m128i x_product_even = _mm_add_epi64(_mm_mul_epi32(x, vmultiplier), vq31rounding); in xnn_qu8_requantize_q31__sse4()
68 const __m128i y_product_even = _mm_add_epi64(_mm_mul_epi32(y, vmultiplier), vq31rounding); in xnn_qu8_requantize_q31__sse4()
69 const __m128i z_product_even = _mm_add_epi64(_mm_mul_epi32(z, vmultiplier), vq31rounding); in xnn_qu8_requantize_q31__sse4()
70 const __m128i w_product_even = _mm_add_epi64(_mm_mul_epi32(w, vmultiplier), vq31rounding); in xnn_qu8_requantize_q31__sse4()
72 const __m128i x_product_odd = _mm_add_epi64(_mm_mul_epi32(x_rev, vmultiplier), vq31rounding); in xnn_qu8_requantize_q31__sse4()
73 const __m128i y_product_odd = _mm_add_epi64(_mm_mul_epi32(y_rev, vmultiplier), vq31rounding); in xnn_qu8_requantize_q31__sse4()
74 const __m128i z_product_odd = _mm_add_epi64(_mm_mul_epi32(z_rev, vmultiplier), vq31rounding); in xnn_qu8_requantize_q31__sse4()
75 const __m128i w_product_odd = _mm_add_epi64(_mm_mul_epi32(w_rev, vmultiplier), vq31rounding); in xnn_qu8_requantize_q31__sse4()
78 const __m128i x_q31product_odd = _mm_add_epi64(x_product_odd, x_product_odd); in xnn_qu8_requantize_q31__sse4()
80 const __m128i y_q31product_odd = _mm_add_epi64(y_product_odd, y_product_odd); in xnn_qu8_requantize_q31__sse4()
[all …]
Dprecise-sse4.c74 const __m128i x_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(x_absmul02, vrounding), vshiftlo); in xnn_qu8_requantize_precise__sse4()
75 const __m128i x_abs_scaled13 = _mm_srl_epi32(_mm_add_epi64(x_absmul13, vrounding), vshifthi); in xnn_qu8_requantize_precise__sse4()
76 const __m128i y_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(y_absmul02, vrounding), vshiftlo); in xnn_qu8_requantize_precise__sse4()
77 const __m128i y_abs_scaled13 = _mm_srl_epi32(_mm_add_epi64(y_absmul13, vrounding), vshifthi); in xnn_qu8_requantize_precise__sse4()
78 const __m128i z_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(z_absmul02, vrounding), vshiftlo); in xnn_qu8_requantize_precise__sse4()
79 const __m128i z_abs_scaled13 = _mm_srl_epi32(_mm_add_epi64(z_absmul13, vrounding), vshifthi); in xnn_qu8_requantize_precise__sse4()
80 const __m128i w_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(w_absmul02, vrounding), vshiftlo); in xnn_qu8_requantize_precise__sse4()
81 const __m128i w_abs_scaled13 = _mm_srl_epi32(_mm_add_epi64(w_absmul13, vrounding), vshifthi); in xnn_qu8_requantize_precise__sse4()
Dprecise-ssse3.c73 const __m128i x_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(x_absmul02, vrounding), vshift); in xnn_qu8_requantize_precise__ssse3()
74 const __m128i x_abs_scaled13 = _mm_srl_epi64(_mm_add_epi64(x_absmul13, vrounding), vshift); in xnn_qu8_requantize_precise__ssse3()
75 const __m128i y_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(y_absmul02, vrounding), vshift); in xnn_qu8_requantize_precise__ssse3()
76 const __m128i y_abs_scaled13 = _mm_srl_epi64(_mm_add_epi64(y_absmul13, vrounding), vshift); in xnn_qu8_requantize_precise__ssse3()
77 const __m128i z_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(z_absmul02, vrounding), vshift); in xnn_qu8_requantize_precise__ssse3()
78 const __m128i z_abs_scaled13 = _mm_srl_epi64(_mm_add_epi64(z_absmul13, vrounding), vshift); in xnn_qu8_requantize_precise__ssse3()
79 const __m128i w_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(w_absmul02, vrounding), vshift); in xnn_qu8_requantize_precise__ssse3()
80 const __m128i w_abs_scaled13 = _mm_srl_epi64(_mm_add_epi64(w_absmul13, vrounding), vshift); in xnn_qu8_requantize_precise__ssse3()
Dprecise-sse2.c78 const __m128i x_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(x_absmul02, vrounding), vshift); in xnn_qu8_requantize_precise__sse2()
79 const __m128i x_abs_scaled13 = _mm_srl_epi64(_mm_add_epi64(x_absmul13, vrounding), vshift); in xnn_qu8_requantize_precise__sse2()
80 const __m128i y_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(y_absmul02, vrounding), vshift); in xnn_qu8_requantize_precise__sse2()
81 const __m128i y_abs_scaled13 = _mm_srl_epi64(_mm_add_epi64(y_absmul13, vrounding), vshift); in xnn_qu8_requantize_precise__sse2()
82 const __m128i z_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(z_absmul02, vrounding), vshift); in xnn_qu8_requantize_precise__sse2()
83 const __m128i z_abs_scaled13 = _mm_srl_epi64(_mm_add_epi64(z_absmul13, vrounding), vshift); in xnn_qu8_requantize_precise__sse2()
84 const __m128i w_abs_scaled02 = _mm_srl_epi64(_mm_add_epi64(w_absmul02, vrounding), vshift); in xnn_qu8_requantize_precise__sse2()
85 const __m128i w_abs_scaled13 = _mm_srl_epi64(_mm_add_epi64(w_absmul13, vrounding), vshift); in xnn_qu8_requantize_precise__sse2()
/external/flac/src/libFLAC/
Dlpc_intrin_sse41.c97 xmm7 = _mm_add_epi64(xmm7, xmm6); in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41()
104 xmm7 = _mm_add_epi64(xmm7, xmm6); in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41()
111 xmm7 = _mm_add_epi64(xmm7, xmm6); in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41()
118 xmm7 = _mm_add_epi64(xmm7, xmm6); in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41()
125 xmm7 = _mm_add_epi64(xmm7, xmm6); in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41()
127 xmm7 = _mm_add_epi64(xmm7, _mm_srli_si128(xmm7, 8)); in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41()
157 xmm7 = _mm_add_epi64(xmm7, xmm6); in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41()
164 xmm7 = _mm_add_epi64(xmm7, xmm6); in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41()
171 xmm7 = _mm_add_epi64(xmm7, xmm6); in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41()
178 xmm7 = _mm_add_epi64(xmm7, xmm6); in FLAC__lpc_compute_residual_from_qlp_coefficients_wide_intrin_sse41()
[all …]
/external/python/cpython3/Modules/_blake2/impl/
Dblake2b-round.h36 : (-(c) == 63) ? _mm_xor_si128(_mm_srli_epi64((x), -(c)), _mm_add_epi64((x), (x))) \
48 row1l = _mm_add_epi64(_mm_add_epi64(row1l, b0), row2l); \
49 row1h = _mm_add_epi64(_mm_add_epi64(row1h, b1), row2h); \
57 row3l = _mm_add_epi64(row3l, row4l); \
58 row3h = _mm_add_epi64(row3h, row4h); \
67 row1l = _mm_add_epi64(_mm_add_epi64(row1l, b0), row2l); \
68 row1h = _mm_add_epi64(_mm_add_epi64(row1h, b1), row2h); \
76 row3l = _mm_add_epi64(row3l, row4l); \
77 row3h = _mm_add_epi64(row3h, row4h); \
/external/XNNPACK/src/qs8-gavgpool/gen/
D7x-minmax-sse41-c24-acc2.c147 const __m128i vabsout02 = _mm_srl_epi64(_mm_add_epi64(vabsprod02, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c24_acc2()
148 const __m128i vabsout13 = _mm_srl_epi64(_mm_add_epi64(vabsprod13, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c24_acc2()
149 const __m128i vabsout46 = _mm_srl_epi64(_mm_add_epi64(vabsprod46, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c24_acc2()
150 const __m128i vabsout57 = _mm_srl_epi64(_mm_add_epi64(vabsprod57, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c24_acc2()
151 const __m128i vabsout8A = _mm_srl_epi64(_mm_add_epi64(vabsprod8A, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c24_acc2()
152 const __m128i vabsout9B = _mm_srl_epi64(_mm_add_epi64(vabsprod9B, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c24_acc2()
153 const __m128i vabsoutCE = _mm_srl_epi64(_mm_add_epi64(vabsprodCE, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c24_acc2()
154 const __m128i vabsoutDF = _mm_srl_epi64(_mm_add_epi64(vabsprodDF, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c24_acc2()
155 const __m128i vabsoutGI = _mm_srl_epi64(_mm_add_epi64(vabsprodGI, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c24_acc2()
156 const __m128i vabsoutHJ = _mm_srl_epi64(_mm_add_epi64(vabsprodHJ, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c24_acc2()
[all …]
D7x-minmax-sse41-c16-acc2.c124 const __m128i vabsout02 = _mm_srl_epi64(_mm_add_epi64(vabsprod02, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c16_acc2()
125 const __m128i vabsout13 = _mm_srl_epi64(_mm_add_epi64(vabsprod13, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c16_acc2()
126 const __m128i vabsout46 = _mm_srl_epi64(_mm_add_epi64(vabsprod46, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c16_acc2()
127 const __m128i vabsout57 = _mm_srl_epi64(_mm_add_epi64(vabsprod57, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c16_acc2()
128 const __m128i vabsout8A = _mm_srl_epi64(_mm_add_epi64(vabsprod8A, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c16_acc2()
129 const __m128i vabsout9B = _mm_srl_epi64(_mm_add_epi64(vabsprod9B, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c16_acc2()
130 const __m128i vabsoutCE = _mm_srl_epi64(_mm_add_epi64(vabsprodCE, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c16_acc2()
131 const __m128i vabsoutDF = _mm_srl_epi64(_mm_add_epi64(vabsprodDF, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c16_acc2()
201 const __m128i vabsout02 = _mm_srl_epi64(_mm_add_epi64(vabsprod02, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c16_acc2()
202 const __m128i vabsout13 = _mm_srl_epi64(_mm_add_epi64(vabsprod13, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse41_c16_acc2()
[all …]
D7x-minmax-ssse3-c24-acc2.c171 const __m128i vabsout02 = _mm_srl_epi64(_mm_add_epi64(vabsprod02, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__ssse3_c24_acc2()
172 const __m128i vabsout13 = _mm_srl_epi64(_mm_add_epi64(vabsprod13, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__ssse3_c24_acc2()
173 const __m128i vabsout46 = _mm_srl_epi64(_mm_add_epi64(vabsprod46, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__ssse3_c24_acc2()
174 const __m128i vabsout57 = _mm_srl_epi64(_mm_add_epi64(vabsprod57, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__ssse3_c24_acc2()
175 const __m128i vabsout8A = _mm_srl_epi64(_mm_add_epi64(vabsprod8A, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__ssse3_c24_acc2()
176 const __m128i vabsout9B = _mm_srl_epi64(_mm_add_epi64(vabsprod9B, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__ssse3_c24_acc2()
177 const __m128i vabsoutCE = _mm_srl_epi64(_mm_add_epi64(vabsprodCE, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__ssse3_c24_acc2()
178 const __m128i vabsoutDF = _mm_srl_epi64(_mm_add_epi64(vabsprodDF, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__ssse3_c24_acc2()
179 const __m128i vabsoutGI = _mm_srl_epi64(_mm_add_epi64(vabsprodGI, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__ssse3_c24_acc2()
180 const __m128i vabsoutHJ = _mm_srl_epi64(_mm_add_epi64(vabsprodHJ, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__ssse3_c24_acc2()
[all …]
D7x-minmax-sse2-c24-acc2.c178 const __m128i vabsout02 = _mm_srl_epi64(_mm_add_epi64(vabsprod02, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse2_c24_acc2()
179 const __m128i vabsout13 = _mm_srl_epi64(_mm_add_epi64(vabsprod13, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse2_c24_acc2()
180 const __m128i vabsout46 = _mm_srl_epi64(_mm_add_epi64(vabsprod46, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse2_c24_acc2()
181 const __m128i vabsout57 = _mm_srl_epi64(_mm_add_epi64(vabsprod57, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse2_c24_acc2()
182 const __m128i vabsout8A = _mm_srl_epi64(_mm_add_epi64(vabsprod8A, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse2_c24_acc2()
183 const __m128i vabsout9B = _mm_srl_epi64(_mm_add_epi64(vabsprod9B, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse2_c24_acc2()
184 const __m128i vabsoutCE = _mm_srl_epi64(_mm_add_epi64(vabsprodCE, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse2_c24_acc2()
185 const __m128i vabsoutDF = _mm_srl_epi64(_mm_add_epi64(vabsprodDF, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse2_c24_acc2()
186 const __m128i vabsoutGI = _mm_srl_epi64(_mm_add_epi64(vabsprodGI, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse2_c24_acc2()
187 const __m128i vabsoutHJ = _mm_srl_epi64(_mm_add_epi64(vabsprodHJ, vrounding), vshift); in xnn_qs8_gavgpool_minmax_ukernel_7x__sse2_c24_acc2()
[all …]
/external/pdfium/third_party/libopenjpeg20/
Dmct.c248 lo = _mm_add_epi64(lo, mulround); in opj_mct_encode_real()
249 hi = _mm_add_epi64(hi, mulround); in opj_mct_encode_real()
258 lo = _mm_add_epi64(lo, mulround); in opj_mct_encode_real()
259 hi = _mm_add_epi64(hi, mulround); in opj_mct_encode_real()
268 lo = _mm_add_epi64(lo, mulround); in opj_mct_encode_real()
269 hi = _mm_add_epi64(hi, mulround); in opj_mct_encode_real()
283 lo = _mm_add_epi64(lo, mulround); in opj_mct_encode_real()
284 hi = _mm_add_epi64(hi, mulround); in opj_mct_encode_real()
293 lo = _mm_add_epi64(lo, mulround); in opj_mct_encode_real()
294 hi = _mm_add_epi64(hi, mulround); in opj_mct_encode_real()
[all …]
/external/libopus/silk/fixed/x86/
Dvector_ops_FIX_sse4_1.c71 acc1 = _mm_add_epi64( acc1, xmm_tempa ); in silk_inner_prod16_aligned_64_sse4_1()
72 acc2 = _mm_add_epi64( acc2, inVec1_76543210 ); in silk_inner_prod16_aligned_64_sse4_1()
75 acc1 = _mm_add_epi64( acc1, acc2 ); in silk_inner_prod16_aligned_64_sse4_1()
79 acc1 = _mm_add_epi64( acc1, acc2 ); in silk_inner_prod16_aligned_64_sse4_1()
/external/webp/src/dsp/
Drescaler_sse2.c155 const __m128i E1 = _mm_add_epi64(D1, rounder); in RescalerImportRowShrink_SSE2()
156 const __m128i E2 = _mm_add_epi64(D2, rounder); in RescalerImportRowShrink_SSE2()
206 const __m128i C0 = _mm_add_epi64(B0, rounder); in ProcessRow_SSE2()
207 const __m128i C1 = _mm_add_epi64(B1, rounder); in ProcessRow_SSE2()
208 const __m128i C2 = _mm_add_epi64(B2, rounder); in ProcessRow_SSE2()
209 const __m128i C3 = _mm_add_epi64(B3, rounder); in ProcessRow_SSE2()
261 const __m128i C0 = _mm_add_epi64(A0, B0); in RescalerExportRowExpand_SSE2()
262 const __m128i C1 = _mm_add_epi64(A1, B1); in RescalerExportRowExpand_SSE2()
263 const __m128i C2 = _mm_add_epi64(A2, B2); in RescalerExportRowExpand_SSE2()
264 const __m128i C3 = _mm_add_epi64(A3, B3); in RescalerExportRowExpand_SSE2()
[all …]
/external/XNNPACK/src/qs8-gemm/gen/
D4x4c2-minmax-xop-ld128.c202 const __m128i vprod0x02 = _mm_add_epi64(_mm_mul_epi32(vacc0x0123, vmultiplier), vrounding); in xnn_qs8_gemm_minmax_ukernel_4x4c2__xop_ld128()
203 const __m128i vprod1x02 = _mm_add_epi64(_mm_mul_epi32(vacc1x0123, vmultiplier), vrounding); in xnn_qs8_gemm_minmax_ukernel_4x4c2__xop_ld128()
204 const __m128i vprod2x02 = _mm_add_epi64(_mm_mul_epi32(vacc2x0123, vmultiplier), vrounding); in xnn_qs8_gemm_minmax_ukernel_4x4c2__xop_ld128()
205 const __m128i vprod3x02 = _mm_add_epi64(_mm_mul_epi32(vacc3x0123, vmultiplier), vrounding); in xnn_qs8_gemm_minmax_ukernel_4x4c2__xop_ld128()
207 const __m128i vprod0x13 = _mm_add_epi64(_mm_mul_epi32(vacc0x1133, vmultiplier), vrounding); in xnn_qs8_gemm_minmax_ukernel_4x4c2__xop_ld128()
208 const __m128i vprod1x13 = _mm_add_epi64(_mm_mul_epi32(vacc1x1133, vmultiplier), vrounding); in xnn_qs8_gemm_minmax_ukernel_4x4c2__xop_ld128()
209 const __m128i vprod2x13 = _mm_add_epi64(_mm_mul_epi32(vacc2x1133, vmultiplier), vrounding); in xnn_qs8_gemm_minmax_ukernel_4x4c2__xop_ld128()
210 const __m128i vprod3x13 = _mm_add_epi64(_mm_mul_epi32(vacc3x1133, vmultiplier), vrounding); in xnn_qs8_gemm_minmax_ukernel_4x4c2__xop_ld128()
213 const __m128i vq31prod0x13 = _mm_add_epi64(vprod0x13, vprod0x13); in xnn_qs8_gemm_minmax_ukernel_4x4c2__xop_ld128()
215 const __m128i vq31prod1x13 = _mm_add_epi64(vprod1x13, vprod1x13); in xnn_qs8_gemm_minmax_ukernel_4x4c2__xop_ld128()
[all …]

123456789