| /external/libaom/libaom/aom_dsp/x86/ |
| D | sse_sse4.c | 32 static INLINE void summary_32_sse4(const __m128i *sum32, __m128i *sum64) { in summary_32_sse4() argument
33 const __m128i sum0 = _mm_cvtepu32_epi64(*sum32); in summary_32_sse4()
34 const __m128i sum1 = _mm_cvtepu32_epi64(_mm_srli_si128(*sum32, 8)); in summary_32_sse4()
231 __m128i sum32 = _mm_setzero_si128(); in aom_highbd_sse_sse4_1() local
233 highbd_sse_w8_sse4_1(&sum32, a, b); in aom_highbd_sse_sse4_1()
234 highbd_sse_w8_sse4_1(&sum32, a + 8, b + 8); in aom_highbd_sse_sse4_1()
239 summary_32_sse4(&sum32, &sum); in aom_highbd_sse_sse4_1()
247 __m128i sum32 = _mm_setzero_si128(); in aom_highbd_sse_sse4_1() local
249 highbd_sse_w8_sse4_1(&sum32, a, b); in aom_highbd_sse_sse4_1()
250 highbd_sse_w8_sse4_1(&sum32, a + 8 * 1, b + 8 * 1); in aom_highbd_sse_sse4_1()
[all …]
|
| D | sse_avx2.c | 49 static INLINE void summary_32_avx2(const __m256i *sum32, __m256i *sum) { in summary_32_avx2() argument
51 _mm256_cvtepu32_epi64(_mm256_castsi256_si128(*sum32)); in summary_32_avx2()
53 _mm256_cvtepu32_epi64(_mm256_extracti128_si256(*sum32, 1)); in summary_32_avx2()
290 __m256i sum32 = _mm256_setzero_si256(); in aom_highbd_sse_avx2() local
292 highbd_sse_w16_avx2(&sum32, a, b); in aom_highbd_sse_avx2()
293 highbd_sse_w16_avx2(&sum32, a + 16, b + 16); in aom_highbd_sse_avx2()
298 summary_32_avx2(&sum32, &sum); in aom_highbd_sse_avx2()
306 __m256i sum32 = _mm256_setzero_si256(); in aom_highbd_sse_avx2() local
308 highbd_sse_w16_avx2(&sum32, a, b); in aom_highbd_sse_avx2()
309 highbd_sse_w16_avx2(&sum32, a + 16 * 1, b + 16 * 1); in aom_highbd_sse_avx2()
[all …]
|
| D | highbd_intrapred_sse2.c | 824 uint32_t sum32 = _mm_cvtsi128_si32(sum); in aom_highbd_dc_predictor_4x8_sse2() local
825 sum32 >>= 16; in aom_highbd_dc_predictor_4x8_sse2()
826 sum32 += 6; in aom_highbd_dc_predictor_4x8_sse2()
827 sum32 /= 12; in aom_highbd_dc_predictor_4x8_sse2()
828 const __m128i row = _mm_set1_epi16((uint16_t)sum32); in aom_highbd_dc_predictor_4x8_sse2()
845 uint32_t sum32 = _mm_cvtsi128_si32(sum); in aom_highbd_dc_predictor_8x4_sse2() local
846 sum32 >>= 16; in aom_highbd_dc_predictor_8x4_sse2()
847 sum32 += 6; in aom_highbd_dc_predictor_8x4_sse2()
848 sum32 /= 12; in aom_highbd_dc_predictor_8x4_sse2()
849 const __m128i row = _mm_set1_epi16((uint16_t)sum32); in aom_highbd_dc_predictor_8x4_sse2()
[all …]
|
| /external/webrtc/modules/audio_coding/codecs/isac/fix/source/ |
| D | entropy_coding_mips.c | 30 int32_t r0, r1, r2, sum32; in WebRtcIsacfix_MatrixProduct1MIPS() local
90 [sum32] "=&r" (sum32), [r1] "=&r" (r1),[r2] "=&r" (r2) in WebRtcIsacfix_MatrixProduct1MIPS()
100 int32_t r0, r1, r2, sum32; in WebRtcIsacfix_MatrixProduct1MIPS() local
160 [sum32] "=&r" (sum32), [r1] "=&r" (r1),[r2] "=&r" (r2) in WebRtcIsacfix_MatrixProduct1MIPS()
186 int32_t r0, r1, r2, r3, r4, sum32, sum32_2; in WebRtcIsacfix_MatrixProduct2MIPS() local
241 [n] "=&r" (n), [sum32] "=&r" (sum32), [sum32_2] "=&r" (sum32_2), in WebRtcIsacfix_MatrixProduct2MIPS()
|
| D | entropy_coding_neon.c | 176 int32_t sum32 = 0; in WebRtcIsacfix_MatrixProduct1Neon() local
180 sum32 += (WEBRTC_SPL_MUL_16_32_RSFT16(matrix0[matrix0_index], in WebRtcIsacfix_MatrixProduct1Neon()
185 matrix_product[matrix_prod_index] = sum32; in WebRtcIsacfix_MatrixProduct1Neon()
|
| D | entropy_coding.c | 884 int32_t sum32 = 0; in WebRtcIsacfix_MatrixProduct1C() local
888 sum32 += WEBRTC_SPL_MUL_16_32_RSFT16( in WebRtcIsacfix_MatrixProduct1C()
893 matrix_product[matrix_prod_index] = sum32; in WebRtcIsacfix_MatrixProduct1C()
920 int32_t sum32 = 0, sum32_2 = 0; in WebRtcIsacfix_MatrixProduct2C() local
924 sum32 += (WEBRTC_SPL_MUL_16_32_RSFT16(matrix0[matrix0_index], in WebRtcIsacfix_MatrixProduct2C()
931 matrix_product[matrix_prod_index] = sum32 >> 3; in WebRtcIsacfix_MatrixProduct2C()
|
| /external/libaom/libaom/av1/encoder/x86/ |
| D | pickrst_sse4.c | 518 __m128i sum32 = _mm_setzero_si128(); in av1_lowbd_pixel_proj_error_sse4_1() local
538 sum32 = _mm_add_epi32(sum32, err0); in av1_lowbd_pixel_proj_error_sse4_1()
550 const __m128i sum64_0 = _mm_cvtepi32_epi64(sum32); in av1_lowbd_pixel_proj_error_sse4_1()
551 const __m128i sum64_1 = _mm_cvtepi32_epi64(_mm_srli_si128(sum32, 8)); in av1_lowbd_pixel_proj_error_sse4_1()
562 __m128i sum32 = _mm_setzero_si128(); in av1_lowbd_pixel_proj_error_sse4_1() local
577 sum32 = _mm_add_epi32(sum32, err0); in av1_lowbd_pixel_proj_error_sse4_1()
588 const __m128i sum64_0 = _mm_cvtepi32_epi64(sum32); in av1_lowbd_pixel_proj_error_sse4_1()
589 const __m128i sum64_1 = _mm_cvtepi32_epi64(_mm_srli_si128(sum32, 8)); in av1_lowbd_pixel_proj_error_sse4_1()
594 __m128i sum32 = _mm_setzero_si128(); in av1_lowbd_pixel_proj_error_sse4_1() local
607 sum32 = _mm_add_epi32(sum32, err0); in av1_lowbd_pixel_proj_error_sse4_1()
[all …]
|
| D | pickrst_avx2.c | 515 __m256i sum32 = _mm256_setzero_si256(); in av1_lowbd_pixel_proj_error_avx2() local
541 sum32 = _mm256_add_epi32(sum32, err0); in av1_lowbd_pixel_proj_error_avx2()
554 _mm256_cvtepi32_epi64(_mm256_castsi256_si128(sum32)); in av1_lowbd_pixel_proj_error_avx2()
556 _mm256_cvtepi32_epi64(_mm256_extracti128_si256(sum32, 1)); in av1_lowbd_pixel_proj_error_avx2()
567 __m256i sum32 = _mm256_setzero_si256(); in av1_lowbd_pixel_proj_error_avx2() local
586 sum32 = _mm256_add_epi32(sum32, err0); in av1_lowbd_pixel_proj_error_avx2()
598 _mm256_cvtepi32_epi64(_mm256_castsi256_si128(sum32)); in av1_lowbd_pixel_proj_error_avx2()
600 _mm256_cvtepi32_epi64(_mm256_extracti128_si256(sum32, 1)); in av1_lowbd_pixel_proj_error_avx2()
605 __m256i sum32 = _mm256_setzero_si256(); in av1_lowbd_pixel_proj_error_avx2() local
612 sum32 = _mm256_add_epi32(sum32, err0); in av1_lowbd_pixel_proj_error_avx2()
[all …]
|
| /external/llvm-project/llvm/test/CodeGen/Generic/ |
| D | add-with-overflow-24.ll | 10 %sum32 = sext i24 %sum to i32
15 …i8*, ...) @printf( i8* getelementptr ([4 x i8], [4 x i8]* @ok, i32 0, i32 0), i32 %sum32 ) nounwind
27 %sum32 = zext i24 %sum to i32
32 …i8*, ...) @printf( i8* getelementptr ([4 x i8], [4 x i8]* @ok, i32 0, i32 0), i32 %sum32 ) nounwind
|
| D | add-with-overflow-128.ll | 12 %sum32 = trunc i128 %sum to i32
17 …i8*, ...) @printf( i8* getelementptr ([4 x i8], [4 x i8]* @ok, i32 0, i32 0), i32 %sum32 ) nounwind
|
| /external/llvm/test/CodeGen/Generic/ |
| D | add-with-overflow-24.ll | 10 %sum32 = sext i24 %sum to i32
15 …i8*, ...) @printf( i8* getelementptr ([4 x i8], [4 x i8]* @ok, i32 0, i32 0), i32 %sum32 ) nounwind
27 %sum32 = zext i24 %sum to i32
32 …i8*, ...) @printf( i8* getelementptr ([4 x i8], [4 x i8]* @ok, i32 0, i32 0), i32 %sum32 ) nounwind
|
| D | add-with-overflow-128.ll | 12 %sum32 = trunc i128 %sum to i32
17 …i8*, ...) @printf( i8* getelementptr ([4 x i8], [4 x i8]* @ok, i32 0, i32 0), i32 %sum32 ) nounwind
|
| /external/flac/src/libFLAC/ |
| D | lpc_intrin_vsx.c | 69 vector float sum32 = { 0.0f, 0.0f, 0.0f, 0.0f}; in FLAC__lpc_compute_autocorrelation_intrin_power8_vsx_lag_16() local
131 sum32 += d2 * d; in FLAC__lpc_compute_autocorrelation_intrin_power8_vsx_lag_16()
151 sum1 += vec_perm(sum31, sum32, (vector unsigned char)vperm3); in FLAC__lpc_compute_autocorrelation_intrin_power8_vsx_lag_16()
152 sum2 += vec_perm(sum32, sum33, (vector unsigned char)vperm3); in FLAC__lpc_compute_autocorrelation_intrin_power8_vsx_lag_16()
201 vector float sum32 = { 0.0f, 0.0f, 0.0f, 0.0f}; in FLAC__lpc_compute_autocorrelation_intrin_power8_vsx_lag_12() local
258 sum32 += d2 * d; in FLAC__lpc_compute_autocorrelation_intrin_power8_vsx_lag_12()
274 sum1 += vec_perm(sum31, sum32, (vector unsigned char)vperm3); in FLAC__lpc_compute_autocorrelation_intrin_power8_vsx_lag_12()
275 sum2 += vec_perm(sum32, sum30, (vector unsigned char)vperm3); in FLAC__lpc_compute_autocorrelation_intrin_power8_vsx_lag_12()
515 vector float sum32 = { 0.0f, 0.0f, 0.0f, 0.0f}; in FLAC__lpc_compute_autocorrelation_intrin_power9_vsx_lag_16() local
577 sum32 += d2 * d; in FLAC__lpc_compute_autocorrelation_intrin_power9_vsx_lag_16()
[all …]
|
| /external/llvm-project/llvm/test/CodeGen/AArch64/ |
| D | arm64_32-pointer-extend.ll | 46 %sum32 = trunc i64 %sum to i32
47 %res = inttoptr i32 %sum32 to i8*
|
| /external/llvm-project/llvm/test/CodeGen/X86/ |
| D | code_placement.ll | 83 %ctg2.sum32 = add i64 %tmp18, 20 ; <i64> [#uses=1]
84 %52 = getelementptr i8, i8* %rk26, i64 %ctg2.sum32 ; <i8*> [#uses=1]
|
| D | lsr-loop-exit-cond.ll | 246 %ctg2.sum32 = add i64 %tmp18, 20 ; <i64> [#uses=1]
247 %52 = getelementptr i8, i8* %rk26, i64 %ctg2.sum32 ; <i8*> [#uses=1]
|
| D | sad.ll | 855 %sum32 = add <32 x i32> %abs, %h32
856 …%h0 = shufflevector <32 x i32> %sum32, <32 x i32> undef, <32 x i32> <i32 8, i32 9, i32 10, i32 11,…
857 %sum0 = add <32 x i32> %sum32, %h0
964 %sum32 = add <64 x i32> %sum64, %h32
965 …%h0 = shufflevector <64 x i32> %sum32, <64 x i32> undef, <64 x i32> <i32 8, i32 9, i32 10, i32 11,…
966 %sum0 = add <64 x i32> %sum32, %h0
|
| /external/llvm/test/CodeGen/X86/ |
| D | code_placement.ll | 80 %ctg2.sum32 = add i64 %tmp18, 20 ; <i64> [#uses=1]
81 %52 = getelementptr i8, i8* %rk26, i64 %ctg2.sum32 ; <i8*> [#uses=1]
|
| D | lsr-loop-exit-cond.ll | 89 %ctg2.sum32 = add i64 %tmp18, 20 ; <i64> [#uses=1]
90 %52 = getelementptr i8, i8* %rk26, i64 %ctg2.sum32 ; <i8*> [#uses=1]
|