| /dports/security/libdecaf/libdecaf-1.0.0/src/p448/arch_arm_32/ |
| H A D | f_impl.c | 118 accum2 = accum3; in gf_mul()
141 accum3 += accum2; in gf_mul()
189 accum2 = accum3; in gf_mul()
477 accum3 = -accum3; in gf_sqr()
493 tmp = -accum3; accum3 = tmp-accum2; accum2 = tmp; in gf_sqr()
539 accum3 = -accum3; in gf_sqr()
556 tmp = -accum3; accum3 = tmp-accum2; accum2 = tmp; in gf_sqr()
608 accum3 = -accum3; in gf_sqr()
626 tmp = -accum3; accum3 = tmp-accum2; accum2 = tmp; in gf_sqr()
676 accum3 = -accum3; in gf_sqr()
[all …]
|
| /dports/emulators/mednafen/mednafen/src/sound/ |
| H A D | OwlResampler_sse.inc | 4 __m128 accum0, accum1, accum2, accum3;
9 tmp3 = accum0 = accum1 = accum2 = accum3 = _mm_setzero_ps();
16 accum3 = _mm_add_ps(accum3, tmp3);
34 accum3 = _mm_add_ps(accum3, tmp3);
39 sum = _mm_add_ps(_mm_add_ps(accum0, accum1), _mm_add_ps(accum2, accum3));
|
| H A D | SwiftResampler_neon.inc | 5 register int32x4_t accum0, accum1, accum2, accum3;
13 accum0 = accum1 = accum2 = accum3 = vdupq_n_s32(0);
41 accum3 = vmlal_s16(accum3, ctmp3, wtmp3);
44 …register int32x4_t tmp = vhaddq_s32(vhaddq_s32(accum0, accum1), vhaddq_s32(accum2, accum3)); // >>…
|
| /dports/www/qt5-webengine/qtwebengine-everywhere-src-5.15.2/src/3rdparty/chromium/skia/ext/ |
| H A D | convolver_SSE2.cc | 172 __m128i accum3 = _mm_setzero_si128(); in Convolve4RowsHorizontally_SSE2() local
210 ITERATION(src_data[3] + start, accum3); in Convolve4RowsHorizontally_SSE2()
235 ITERATION(src_data[3] + start, accum3); in Convolve4RowsHorizontally_SSE2()
247 accum3 = _mm_srai_epi32(accum3, ConvolutionFilter1D::kShiftBits); in Convolve4RowsHorizontally_SSE2()
248 accum3 = _mm_packs_epi32(accum3, zero); in Convolve4RowsHorizontally_SSE2()
249 accum3 = _mm_packus_epi16(accum3, zero); in Convolve4RowsHorizontally_SSE2()
278 __m128i accum0, accum1, accum2, accum3, coeff16; in ConvolveVertically_SSE2() local
287 accum3 = _mm_setzero_si128(); in ConvolveVertically_SSE2()
326 accum3 = _mm_add_epi32(accum3, t); in ConvolveVertically_SSE2()
333 accum3 = _mm_srai_epi32(accum3, ConvolutionFilter1D::kShiftBits); in ConvolveVertically_SSE2()
[all …]
|
| H A D | convolver_neon.cc | 108 int32x4_t accum3 = vdupq_n_s32(0); in Convolve4RowsHorizontally_Neon() local
134 accum3 += iteration(src_data[3] + start); in Convolve4RowsHorizontally_Neon()
149 accum3 += AccumRemainder(src_data[3] + remainder_offset, filter_values, in Convolve4RowsHorizontally_Neon()
161 uint8x8_t res3 = pack_result(accum3); in Convolve4RowsHorizontally_Neon()
198 int32x4_t accum3 = vdupq_n_s32(0); in ConvolveVertically_Neon() local
215 accum3 = in ConvolveVertically_Neon()
216 vmlal_n_s16(accum3, vget_high_s16(src16_23), filter_values[filter_y]); in ConvolveVertically_Neon()
224 int16x4_t accum16_3 = vqshrn_n_s32(accum3, ConvolutionFilter1D::kShiftBits); in ConvolveVertically_Neon()
|
| /dports/www/firefox-legacy/firefox-52.8.0esr/gfx/2d/ |
| H A D | convolverSSE2.cpp | 192 __m128i accum3 = _mm_setzero_si128(); in ConvolveHorizontally4_SSE2() local
230 ITERATION(src_data[3] + start, accum3); in ConvolveHorizontally4_SSE2()
255 ITERATION(src_data[3] + start, accum3); in ConvolveHorizontally4_SSE2()
267 accum3 = _mm_srai_epi32(accum3, ConvolutionFilter1D::kShiftBits); in ConvolveHorizontally4_SSE2()
268 accum3 = _mm_packs_epi32(accum3, zero); in ConvolveHorizontally4_SSE2()
269 accum3 = _mm_packus_epi16(accum3, zero); in ConvolveHorizontally4_SSE2()
298 __m128i accum0, accum1, accum2, accum3, coeff16; in ConvolveVertically_SSE2_impl() local
307 accum3 = _mm_setzero_si128(); in ConvolveVertically_SSE2_impl()
346 accum3 = _mm_add_epi32(accum3, t); in ConvolveVertically_SSE2_impl()
353 accum3 = _mm_srai_epi32(accum3, ConvolutionFilter1D::kShiftBits); in ConvolveVertically_SSE2_impl()
[all …]
|
| H A D | convolverLS3.cpp | 379 _mm_xor(accum3, accum3, accum3) in ConvolveHorizontally4_LS3()
507 _mm_psraw(accum3, accum3, sra) in ConvolveHorizontally4_LS3()
508 _mm_packsswh(accum3, accum3, zero, t) in ConvolveHorizontally4_LS3()
509 _mm_packushb(accum3, accum3, zero, t) in ConvolveHorizontally4_LS3()
574 _mm_xor(accum3, accum3, accum3) in ConvolveVertically_LS3_impl()
645 _mm_paddw(accum3, accum3, t) in ConvolveVertically_LS3_impl()
663 _mm_psraw(accum3, accum3, sra) in ConvolveVertically_LS3_impl()
668 _mm_packsswh(accum2, accum2, accum3, t) in ConvolveVertically_LS3_impl()
|
| /dports/www/chromium-legacy/chromium-88.0.4324.182/skia/ext/ |
| H A D | convolver_SSE2.cc | 172 __m128i accum3 = _mm_setzero_si128(); in Convolve4RowsHorizontally_SSE2() local
210 ITERATION(src_data[3] + start, accum3); in Convolve4RowsHorizontally_SSE2()
235 ITERATION(src_data[3] + start, accum3); in Convolve4RowsHorizontally_SSE2()
247 accum3 = _mm_srai_epi32(accum3, ConvolutionFilter1D::kShiftBits); in Convolve4RowsHorizontally_SSE2()
248 accum3 = _mm_packs_epi32(accum3, zero); in Convolve4RowsHorizontally_SSE2()
249 accum3 = _mm_packus_epi16(accum3, zero); in Convolve4RowsHorizontally_SSE2()
278 __m128i accum0, accum1, accum2, accum3, coeff16; in ConvolveVertically_SSE2() local
287 accum3 = _mm_setzero_si128(); in ConvolveVertically_SSE2()
326 accum3 = _mm_add_epi32(accum3, t); in ConvolveVertically_SSE2()
333 accum3 = _mm_srai_epi32(accum3, ConvolutionFilter1D::kShiftBits); in ConvolveVertically_SSE2()
[all …]
|
| H A D | convolver_neon.cc | 108 int32x4_t accum3 = vdupq_n_s32(0); in Convolve4RowsHorizontally_Neon() local
134 accum3 += iteration(src_data[3] + start); in Convolve4RowsHorizontally_Neon()
149 accum3 += AccumRemainder(src_data[3] + remainder_offset, filter_values, in Convolve4RowsHorizontally_Neon()
161 uint8x8_t res3 = pack_result(accum3); in Convolve4RowsHorizontally_Neon()
198 int32x4_t accum3 = vdupq_n_s32(0); in ConvolveVertically_Neon() local
215 accum3 = in ConvolveVertically_Neon()
216 vmlal_n_s16(accum3, vget_high_s16(src16_23), filter_values[filter_y]); in ConvolveVertically_Neon()
224 int16x4_t accum16_3 = vqshrn_n_s32(accum3, ConvolutionFilter1D::kShiftBits); in ConvolveVertically_Neon()
|
| /dports/www/firefox-legacy/firefox-52.8.0esr/gfx/skia/skia/src/opts/ |
| H A D | SkBitmapFilter_opts_SSE2.cpp | 202 __m128i accum3 = _mm_setzero_si128(); in convolve4RowsHorizontally_SSE2() local
240 ITERATION(src_data[3] + start, accum3); in convolve4RowsHorizontally_SSE2()
265 ITERATION(src_data[3] + start, accum3); in convolve4RowsHorizontally_SSE2()
277 accum3 = _mm_srai_epi32(accum3, SkConvolutionFilter1D::kShiftBits); in convolve4RowsHorizontally_SSE2()
278 accum3 = _mm_packs_epi32(accum3, zero); in convolve4RowsHorizontally_SSE2()
279 accum3 = _mm_packus_epi16(accum3, zero); in convolve4RowsHorizontally_SSE2()
311 __m128i accum0, accum1, accum2, accum3, coeff16; in convolveVertically_SSE2() local
320 accum3 = _mm_setzero_si128(); in convolveVertically_SSE2()
359 accum3 = _mm_add_epi32(accum3, t); in convolveVertically_SSE2()
366 accum3 = _mm_srai_epi32(accum3, SkConvolutionFilter1D::kShiftBits); in convolveVertically_SSE2()
[all …]
|
| H A D | SkBitmapProcState_arm_neon.cpp | 198 int32x4_t accum0, accum1, accum2, accum3; in convolveVertically_neon() local
205 accum0 = accum1 = accum2 = accum3 = vdupq_n_s32(0); in convolveVertically_neon()
228 accum3 += vmull_s16(src16_3, coeff16); in convolveVertically_neon()
235 accum3 = vshrq_n_s32(accum3, SkConvolutionFilter1D::kShiftBits); in convolveVertically_neon()
241 int16x8_t accum16_1 = vcombine_s16(vqmovn_s32(accum2), vqmovn_s32(accum3)); in convolveVertically_neon()
397 int32x4_t accum3 = vdupq_n_s32(0); in convolve4RowsHorizontally_neon() local
432 ITERATION(srcData[3] + start, accum3); in convolve4RowsHorizontally_neon()
455 ITERATION(srcData[3] + start, accum3); in convolve4RowsHorizontally_neon()
469 PACK_RESULT(accum3, res3); in convolve4RowsHorizontally_neon()
|
| /dports/security/libdecaf/libdecaf-1.0.0/_aux/attic/p521/arch_x86_64_r12/ |
| H A D | f_impl.c | 313 __uint128_t accum0 = 0, accum3 = 0, accum6 = 0; in gf_mulw() local
317 accum3 += widemulu(b, a[1]); in gf_mulw()
320 c[1] = accum3 & mask; accum3 >>= 58; in gf_mulw()
324 accum3 += widemulu(b, a[5]); in gf_mulw()
327 c[5] = accum3 & mask; accum3 >>= 58; in gf_mulw()
331 accum3 += widemulu(b, a[9]); in gf_mulw()
334 c[9] = accum3 & mask; accum3 >>= 58; in gf_mulw()
341 accum3 += c[2]; in gf_mulw()
342 c[2] = accum3 & mask; in gf_mulw()
343 c[6] += accum3 >> 58; in gf_mulw()
|
| /dports/www/firefox-esr/firefox-91.8.0/gfx/skia/skia/src/opts/ |
| H A D | SkBitmapFilter_opts.h | 152 __m128i accum3 = _mm_setzero_si128(); in convolve_4_rows_horizontally() local
215 accum3 = _mm_srai_epi32(accum3, SkConvolutionFilter1D::kShiftBits); in convolve_4_rows_horizontally()
216 accum3 = _mm_packs_epi32(accum3, zero); in convolve_4_rows_horizontally()
217 accum3 = _mm_packus_epi16(accum3, zero); in convolve_4_rows_horizontally()
254 __m128i accum3 = _mm_setzero_si128(); in ConvolveVertically() local
293 accum3 = _mm_add_epi32(accum3, t); in ConvolveVertically()
300 accum3 = _mm_srai_epi32(accum3, SkConvolutionFilter1D::kShiftBits); in ConvolveVertically()
517 int32x4_t accum3 = vdupq_n_s32(0);
582 PACK_RESULT(accum3, res3);
617 int32x4_t accum3 = vdupq_n_s32(0);
[all …]
|
| /dports/lang/spidermonkey78/firefox-78.9.0/gfx/skia/skia/src/opts/ |
| H A D | SkBitmapFilter_opts.h | 152 __m128i accum3 = _mm_setzero_si128(); in convolve_4_rows_horizontally() local
215 accum3 = _mm_srai_epi32(accum3, SkConvolutionFilter1D::kShiftBits); in convolve_4_rows_horizontally()
216 accum3 = _mm_packs_epi32(accum3, zero); in convolve_4_rows_horizontally()
217 accum3 = _mm_packus_epi16(accum3, zero); in convolve_4_rows_horizontally()
254 __m128i accum3 = _mm_setzero_si128(); in ConvolveVertically() local
293 accum3 = _mm_add_epi32(accum3, t); in ConvolveVertically()
300 accum3 = _mm_srai_epi32(accum3, SkConvolutionFilter1D::kShiftBits); in ConvolveVertically()
517 int32x4_t accum3 = vdupq_n_s32(0);
582 PACK_RESULT(accum3, res3);
617 int32x4_t accum3 = vdupq_n_s32(0);
[all …]
|
| /dports/www/firefox/firefox-99.0/gfx/skia/skia/src/opts/ |
| H A D | SkBitmapFilter_opts.h | 152 __m128i accum3 = _mm_setzero_si128(); in convolve_4_rows_horizontally() local
215 accum3 = _mm_srai_epi32(accum3, SkConvolutionFilter1D::kShiftBits); in convolve_4_rows_horizontally()
216 accum3 = _mm_packs_epi32(accum3, zero); in convolve_4_rows_horizontally()
217 accum3 = _mm_packus_epi16(accum3, zero); in convolve_4_rows_horizontally()
254 __m128i accum3 = _mm_setzero_si128(); in ConvolveVertically() local
293 accum3 = _mm_add_epi32(accum3, t); in ConvolveVertically()
300 accum3 = _mm_srai_epi32(accum3, SkConvolutionFilter1D::kShiftBits); in ConvolveVertically()
517 int32x4_t accum3 = vdupq_n_s32(0);
582 PACK_RESULT(accum3, res3);
617 int32x4_t accum3 = vdupq_n_s32(0);
[all …]
|
| /dports/mail/thunderbird/thunderbird-91.8.0/gfx/skia/skia/src/opts/ |
| H A D | SkBitmapFilter_opts.h | 152 __m128i accum3 = _mm_setzero_si128(); in convolve_4_rows_horizontally() local
215 accum3 = _mm_srai_epi32(accum3, SkConvolutionFilter1D::kShiftBits); in convolve_4_rows_horizontally()
216 accum3 = _mm_packs_epi32(accum3, zero); in convolve_4_rows_horizontally()
217 accum3 = _mm_packus_epi16(accum3, zero); in convolve_4_rows_horizontally()
254 __m128i accum3 = _mm_setzero_si128(); in ConvolveVertically() local
293 accum3 = _mm_add_epi32(accum3, t); in ConvolveVertically()
300 accum3 = _mm_srai_epi32(accum3, SkConvolutionFilter1D::kShiftBits); in ConvolveVertically()
517 int32x4_t accum3 = vdupq_n_s32(0);
582 PACK_RESULT(accum3, res3);
617 int32x4_t accum3 = vdupq_n_s32(0);
[all …]
|
| /dports/lang/spidermonkey60/firefox-60.9.0/gfx/skia/skia/src/opts/ |
| H A D | SkBitmapFilter_opts.h | 152 __m128i accum3 = _mm_setzero_si128(); in convolve_4_rows_horizontally() local
215 accum3 = _mm_srai_epi32(accum3, SkConvolutionFilter1D::kShiftBits); in convolve_4_rows_horizontally()
216 accum3 = _mm_packs_epi32(accum3, zero); in convolve_4_rows_horizontally()
217 accum3 = _mm_packus_epi16(accum3, zero); in convolve_4_rows_horizontally()
254 __m128i accum3 = _mm_setzero_si128(); in ConvolveVertically() local
293 accum3 = _mm_add_epi32(accum3, t); in ConvolveVertically()
300 accum3 = _mm_srai_epi32(accum3, SkConvolutionFilter1D::kShiftBits); in ConvolveVertically()
517 int32x4_t accum3 = vdupq_n_s32(0);
582 PACK_RESULT(accum3, res3);
617 int32x4_t accum3 = vdupq_n_s32(0);
[all …]
|
| /dports/math/apache-commons-math/commons-math3-3.6.1-src/src/main/java/org/apache/commons/math3/stat/descriptive/moment/ |
| H A D | Kurtosis.java | 178 double accum3 = 0.0; in evaluate() local
180 accum3 += FastMath.pow(values[i] - mean, 4.0); in evaluate()
182 accum3 /= FastMath.pow(stdDev, 4.0d); in evaluate()
193 kurt = (coefficientOne * accum3) - termTwo; in evaluate()
|
| H A D | Skewness.java | 185 double accum3 = 0.0; in evaluate() local
188 accum3 += d * d * d; in evaluate()
190 accum3 /= variance * FastMath.sqrt(variance); in evaluate()
196 skew = (n0 / ((n0 - 1) * (n0 - 2))) * accum3; in evaluate()
|
| /dports/security/libdecaf/libdecaf-1.0.0/_aux/attic/p521/arch_ref64/ |
| H A D | f_impl.c | 151 __uint128_t accum0 = 0, accum3 = 0, accum6 = 0; in gf_mulw() local
157 accum3 += widemul(b, a[i+3]); in gf_mulw()
160 c[i+3] = accum3 & mask; accum3 >>= 58; in gf_mulw()
172 accum3 += c[6]; in gf_mulw()
173 c[6] = accum3 & mask; in gf_mulw()
174 c[7] += accum3 >> 58; in gf_mulw()
|
| /dports/multimedia/handbrake/zimg-release-3.0.1/src/zimg/resize/x86/ |
| H A D | resize_impl_avx.cpp | 215 __m256 accum3 = _mm256_setzero_ps(); in resize_line_v_f32_avx_xiter() local
247 accum3 = x; in resize_line_v_f32_avx_xiter()
257 accum3 = _mm256_add_ps(accum3, x); in resize_line_v_f32_avx_xiter()
261 accum2 = (N >= 5) ? _mm256_add_ps(accum2, accum3) : accum2; in resize_line_v_f32_avx_xiter()
|
| /dports/graphics/sekrit-twc-zimg/zimg-release-3.0.3/src/zimg/resize/x86/ |
| H A D | resize_impl_avx.cpp | 215 __m256 accum3 = _mm256_setzero_ps(); in resize_line_v_f32_avx_xiter() local
247 accum3 = x; in resize_line_v_f32_avx_xiter()
257 accum3 = _mm256_add_ps(accum3, x); in resize_line_v_f32_avx_xiter()
261 accum2 = (N >= 5) ? _mm256_add_ps(accum2, accum3) : accum2; in resize_line_v_f32_avx_xiter()
|
| /dports/multimedia/kvazaar/kvazaar-2.1.0/src/strategies/avx2/ |
| H A D | picture-avx2.c | 1132 __m256 accum3 = _mm256_add_ps (accum, accum2); in pixel_var_avx2_largebuf() local
1133 __m256 accum4 = _mm256_permute_ps (accum3, _MM_SHUFFLE(1, 0, 3, 2)); in pixel_var_avx2_largebuf()
1134 __m256 accum5 = _mm256_add_ps (accum3, accum4); in pixel_var_avx2_largebuf()
1227 __m256i accum3 = _mm256_add_epi64 (accum, accum2); in pixel_var_avx2() local
1228 __m256i accum4 = _mm256_permute4x64_epi64(accum3, _MM_SHUFFLE(2, 3, 1, 0)); in pixel_var_avx2()
1229 __m256i v_tot = _mm256_add_epi64 (accum3, accum4); in pixel_var_avx2()
|
| /dports/science/rdkit/rdkit-Release_2021_03_5/Code/GraphMol/ |
| H A D | AddHs.cpp | 456 auto accum3 = angle13 + angle23; in setTerminalAtomCoords() local
457 if (accum1 <= accum2 && accum1 <= accum3) { in setTerminalAtomCoords()
459 } else if (accum2 <= accum1 && accum2 <= accum3) { in setTerminalAtomCoords()
|
| /dports/games/arx-libertatis/arx-libertatis-1.1.2/src/ai/ |
| H A D | Paths.cpp | 1497 PHYSVERT * source, * target, * accum1, * accum2, * accum3, * accum4; in RK4Integrate() local
1539 accum3 = &m_TempSys[3][kk]; in RK4Integrate()
1543 Vec3f dv = accum1->force + ((accum2->force + accum3->force) * 2.f) + accum4->force; in RK4Integrate()
1546 Vec3f dp = accum1->velocity + ((accum2->velocity + accum3->velocity) * 2.f) in RK4Integrate()
|