/external/libaom/libaom/aom_dsp/x86/ |
D | sse_sse4.c | 31 static INLINE void summary_32_sse4(const __m128i *sum32, __m128i *sum64) { in summary_32_sse4() argument 32 const __m128i sum0 = _mm_cvtepu32_epi64(*sum32); in summary_32_sse4() 33 const __m128i sum1 = _mm_cvtepu32_epi64(_mm_srli_si128(*sum32, 8)); in summary_32_sse4() 229 __m128i sum32 = _mm_setzero_si128(); in aom_highbd_sse_sse4_1() local 231 highbd_sse_w8_sse4_1(&sum32, a, b); in aom_highbd_sse_sse4_1() 232 highbd_sse_w8_sse4_1(&sum32, a + 8, b + 8); in aom_highbd_sse_sse4_1() 237 summary_32_sse4(&sum32, &sum); in aom_highbd_sse_sse4_1() 245 __m128i sum32 = _mm_setzero_si128(); in aom_highbd_sse_sse4_1() local 247 highbd_sse_w8_sse4_1(&sum32, a, b); in aom_highbd_sse_sse4_1() 248 highbd_sse_w8_sse4_1(&sum32, a + 8 * 1, b + 8 * 1); in aom_highbd_sse_sse4_1() [all …]
|
D | sse_avx2.c | 48 static INLINE void summary_32_avx2(const __m256i *sum32, __m256i *sum) { in summary_32_avx2() argument 50 _mm256_cvtepu32_epi64(_mm256_castsi256_si128(*sum32)); in summary_32_avx2() 52 _mm256_cvtepu32_epi64(_mm256_extracti128_si256(*sum32, 1)); in summary_32_avx2() 287 __m256i sum32 = _mm256_setzero_si256(); in aom_highbd_sse_avx2() local 289 highbd_sse_w16_avx2(&sum32, a, b); in aom_highbd_sse_avx2() 290 highbd_sse_w16_avx2(&sum32, a + 16, b + 16); in aom_highbd_sse_avx2() 295 summary_32_avx2(&sum32, &sum); in aom_highbd_sse_avx2() 303 __m256i sum32 = _mm256_setzero_si256(); in aom_highbd_sse_avx2() local 305 highbd_sse_w16_avx2(&sum32, a, b); in aom_highbd_sse_avx2() 306 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/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 | 177 int32_t sum32 = 0; in WebRtcIsacfix_MatrixProduct1Neon() local 181 sum32 += (WEBRTC_SPL_MUL_16_32_RSFT16(matrix0[matrix0_index], in WebRtcIsacfix_MatrixProduct1Neon() 186 matrix_product[matrix_prod_index] = sum32; in WebRtcIsacfix_MatrixProduct1Neon()
|
D | entropy_coding.c | 863 int32_t sum32 = 0; in WebRtcIsacfix_MatrixProduct1C() local 867 sum32 += (WEBRTC_SPL_MUL_16_32_RSFT16(matrix0[matrix0_index], in WebRtcIsacfix_MatrixProduct1C() 872 matrix_product[matrix_prod_index] = sum32; in WebRtcIsacfix_MatrixProduct1C() 899 int32_t sum32 = 0, sum32_2 = 0; in WebRtcIsacfix_MatrixProduct2C() local 903 sum32 += (WEBRTC_SPL_MUL_16_32_RSFT16(matrix0[matrix0_index], in WebRtcIsacfix_MatrixProduct2C() 910 matrix_product[matrix_prod_index] = sum32 >> 3; in WebRtcIsacfix_MatrixProduct2C()
|
/external/libaom/libaom/av1/encoder/x86/ |
D | pickrst_avx2.c | 507 __m256i sum32 = _mm256_setzero_si256(); in av1_lowbd_pixel_proj_error_avx2() local 533 sum32 = _mm256_add_epi32(sum32, err0); in av1_lowbd_pixel_proj_error_avx2() 546 _mm256_cvtepi32_epi64(_mm256_castsi256_si128(sum32)); in av1_lowbd_pixel_proj_error_avx2() 548 _mm256_cvtepi32_epi64(_mm256_extracti128_si256(sum32, 1)); in av1_lowbd_pixel_proj_error_avx2() 559 __m256i sum32 = _mm256_setzero_si256(); in av1_lowbd_pixel_proj_error_avx2() local 578 sum32 = _mm256_add_epi32(sum32, err0); in av1_lowbd_pixel_proj_error_avx2() 590 _mm256_cvtepi32_epi64(_mm256_castsi256_si128(sum32)); in av1_lowbd_pixel_proj_error_avx2() 592 _mm256_cvtepi32_epi64(_mm256_extracti128_si256(sum32, 1)); in av1_lowbd_pixel_proj_error_avx2() 597 __m256i sum32 = _mm256_setzero_si256(); in av1_lowbd_pixel_proj_error_avx2() local 604 sum32 = _mm256_add_epi32(sum32, err0); in av1_lowbd_pixel_proj_error_avx2() [all …]
|
D | pickrst_sse4.c | 516 __m128i sum32 = _mm_setzero_si128(); in av1_lowbd_pixel_proj_error_sse4_1() local 536 sum32 = _mm_add_epi32(sum32, err0); in av1_lowbd_pixel_proj_error_sse4_1() 548 const __m128i sum64_0 = _mm_cvtepi32_epi64(sum32); in av1_lowbd_pixel_proj_error_sse4_1() 549 const __m128i sum64_1 = _mm_cvtepi32_epi64(_mm_srli_si128(sum32, 8)); in av1_lowbd_pixel_proj_error_sse4_1() 560 __m128i sum32 = _mm_setzero_si128(); in av1_lowbd_pixel_proj_error_sse4_1() local 575 sum32 = _mm_add_epi32(sum32, err0); in av1_lowbd_pixel_proj_error_sse4_1() 586 const __m128i sum64_0 = _mm_cvtepi32_epi64(sum32); in av1_lowbd_pixel_proj_error_sse4_1() 587 const __m128i sum64_1 = _mm_cvtepi32_epi64(_mm_srli_si128(sum32, 8)); in av1_lowbd_pixel_proj_error_sse4_1() 592 __m128i sum32 = _mm_setzero_si128(); in av1_lowbd_pixel_proj_error_sse4_1() local 605 sum32 = _mm_add_epi32(sum32, err0); in av1_lowbd_pixel_proj_error_sse4_1() [all …]
|
/external/swiftshader/third_party/LLVM/test/CodeGen/Generic/ |
D | add-with-overflow-24.ll | 10 %sum32 = sext i24 %sum to i32 15 …all i32 (i8*, ...)* @printf( i8* getelementptr ([4 x i8]* @ok, i32 0, i32 0), i32 %sum32 ) nounwind 27 %sum32 = zext i24 %sum to i32 32 …all i32 (i8*, ...)* @printf( i8* getelementptr ([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 …all i32 (i8*, ...)* @printf( i8* getelementptr ([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/swiftshader/third_party/llvm-7.0/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/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]
|
/external/swiftshader/third_party/LLVM/test/CodeGen/X86/ |
D | code_placement.ll | 80 %ctg2.sum32 = add i64 %tmp18, 20 ; <i64> [#uses=1] 81 %52 = getelementptr i8* %rk26, i64 %ctg2.sum32 ; <i8*> [#uses=1]
|
D | lsr-loop-exit-cond.ll | 81 %ctg2.sum32 = add i64 %tmp18, 20 ; <i64> [#uses=1] 82 %52 = getelementptr i8* %rk26, i64 %ctg2.sum32 ; <i8*> [#uses=1]
|
/external/swiftshader/third_party/llvm-7.0/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 | 91 %ctg2.sum32 = add i64 %tmp18, 20 ; <i64> [#uses=1] 92 %52 = getelementptr i8, i8* %rk26, i64 %ctg2.sum32 ; <i8*> [#uses=1]
|
D | sad.ll | 1386 %sum32 = add <32 x i32> %abs, %h32 1387 …%h0 = shufflevector <32 x i32> %sum32, <32 x i32> undef, <32 x i32> <i32 8, i32 9, i32 10, i32 11,… 1388 %sum0 = add <32 x i32> %sum32, %h0
|
/external/swiftshader/third_party/LLVM/test/CodeGen/ARM/ |
D | lsr-on-unrolled-loops.ll | 455 %.sum32 = add i32 %cur_match_addr.0, %best_len.2 ; <i32> [#uses=1] 456 %38 = getelementptr inbounds i8* %3, i32 %.sum32 ; <i8*> [#uses=1]
|