/external/libvpx/libvpx/vpx_dsp/arm/ |
D | transpose_neon.h | 49 static INLINE uint8x16x2_t vpx_vtrnq_u64_to_u8(uint32x4_t a0, uint32x4_t a1) { in vpx_vtrnq_u64_to_u8() 50 uint8x16x2_t b0; in vpx_vtrnq_u64_to_u8() 535 const uint8x16x2_t b0 = in transpose_u8_8x8() 537 const uint8x16x2_t b1 = in transpose_u8_8x8() 815 const uint8x16x2_t b0 = vtrnq_u8(i0, i1); in transpose_u8_16x8() 816 const uint8x16x2_t b1 = vtrnq_u8(i2, i3); in transpose_u8_16x8() 817 const uint8x16x2_t b2 = vtrnq_u8(i4, i5); in transpose_u8_16x8() 818 const uint8x16x2_t b3 = vtrnq_u8(i6, i7); in transpose_u8_16x8() 944 const uint8x16x2_t b0 = vtrnq_u8(a0, a1); in transpose_u8_8x16() 945 const uint8x16x2_t b1 = vtrnq_u8(a2, a3); in transpose_u8_8x16() [all …]
|
D | intrapred_neon.c | 195 const uint8x16x2_t r = vld2q_u8(ref); in dc_sum_32() 206 uint8x16x2_t dc_dup; in dc_store_32x32() 217 const uint8x16x2_t a = vld2q_u8(above); in vpx_dc_predictor_32x32_neon() 218 const uint8x16x2_t l = vld2q_u8(left); in vpx_dc_predictor_32x32_neon()
|
/external/libjpeg-turbo/simd/arm/ |
D | jdsample-neon.c | 107 uint8x16x2_t output_pixels; in jsimd_h2v1_fancy_upsample_neon() 284 uint8x16x2_t output_pixels0 = { { in jsimd_h2v2_fancy_upsample_neon() 288 uint8x16x2_t output_pixels1 = { { in jsimd_h2v2_fancy_upsample_neon() 504 uint8x16x2_t output_pixels = { { samples, samples } }; in jsimd_h2v1_upsample_neon() 559 uint8x16x2_t output_pixels = { { samples, samples } }; in jsimd_h2v2_upsample_neon()
|
D | jidctfst-neon.c | 440 uint8x16x2_t cols_0145 = vtrnq_u8(vreinterpretq_u8_u32(cols_0415.val[0]), in jsimd_idct_ifast_neon() 442 uint8x16x2_t cols_2367 = vtrnq_u8(vreinterpretq_u8_u32(cols_2637.val[0]), in jsimd_idct_ifast_neon()
|
/external/libgav1/libgav1/src/dsp/arm/ |
D | common_neon.h | 613 const uint8x16x2_t b0 = in Transpose8x8() 615 const uint8x16x2_t b1 = in Transpose8x8() 639 const uint8x16x2_t a0 = in Transpose8x8() 641 const uint8x16x2_t a1 = in Transpose8x8() 768 const uint8x16x2_t b0 = vtrnq_u8(a[0], a[1]); in Transpose8x16() 769 const uint8x16x2_t b1 = vtrnq_u8(a[2], a[3]); in Transpose8x16() 770 const uint8x16x2_t b2 = vtrnq_u8(a[4], a[5]); in Transpose8x16() 771 const uint8x16x2_t b3 = vtrnq_u8(a[6], a[7]); in Transpose8x16()
|
D | loop_filter_neon.cc | 998 const uint8x16x2_t in01 = vtrnq_u8(input_0, input_1); in Vertical14_NEON() 999 const uint8x16x2_t in23 = vtrnq_u8(input_2, input_3); in Vertical14_NEON() 1111 const uint8x16x2_t out01 = vtrnq_u8(vreinterpretq_u8_u16(out02.val[0]), in Vertical14_NEON() 1113 const uint8x16x2_t out23 = vtrnq_u8(vreinterpretq_u8_u16(out02.val[1]), in Vertical14_NEON()
|
/external/clang/test/CodeGen/ |
D | aarch64-neon-tbl.c | 560 uint8x8_t test_vqtbl2_u8(uint8x16x2_t a, uint8x8_t b) { in test_vqtbl2_u8() 694 uint8x16_t test_vqtbl2q_u8(uint8x16x2_t a, uint8x16_t b) { in test_vqtbl2q_u8() 870 uint8x8_t test_vqtbx2_u8(uint8x8_t a, uint8x16x2_t b, uint8x8_t c) { in test_vqtbx2_u8() 949 uint8x16_t test_vqtbx2q_u8(uint8x16_t a, uint8x16x2_t b, uint8x16_t c) { in test_vqtbx2q_u8()
|
D | aarch64-neon-perm.c | 1218 uint8x16x2_t test_vuzpq_u8(uint8x16_t a, uint8x16_t b) { in test_vuzpq_u8() 1682 uint8x16x2_t test_vzipq_u8(uint8x16_t a, uint8x16_t b) { in test_vzipq_u8() 2146 uint8x16x2_t test_vtrnq_u8(uint8x16_t a, uint8x16_t b) { in test_vtrnq_u8()
|
D | aarch64-neon-ldst-one.c | 317 uint8x16x2_t test_vld2q_dup_u8(uint8_t *a) { in test_vld2q_dup_u8() 2177 uint8x16x2_t test_vld2q_lane_u8(uint8_t const * ptr, uint8x16x2_t src) { in test_vld2q_lane_u8() 5601 void test_vst2q_lane_u8(uint8_t *a, uint8x16x2_t b) { in test_vst2q_lane_u8()
|
/external/llvm-project/clang/test/CodeGen/ |
D | aarch64-neon-tbl.c | 560 uint8x8_t test_vqtbl2_u8(uint8x16x2_t a, uint8x8_t b) { in test_vqtbl2_u8() 694 uint8x16_t test_vqtbl2q_u8(uint8x16x2_t a, uint8x16_t b) { in test_vqtbl2q_u8() 870 uint8x8_t test_vqtbx2_u8(uint8x8_t a, uint8x16x2_t b, uint8x8_t c) { in test_vqtbx2_u8() 949 uint8x16_t test_vqtbx2q_u8(uint8x16_t a, uint8x16x2_t b, uint8x16_t c) { in test_vqtbx2q_u8()
|
D | arm-neon-vld.c | 1373 uint8x16x2_t test_vld1q_u8_x2(uint8_t const *a) { in test_vld1q_u8_x2() 2168 void test_vld2q_dup_u8(uint8x16x2_t *dest, const uint8_t *src) { in test_vld2q_dup_u8()
|
D | aarch64-neon-perm.c | 1161 uint8x16x2_t test_vuzpq_u8(uint8x16_t a, uint8x16_t b) { in test_vuzpq_u8() 1545 uint8x16x2_t test_vzipq_u8(uint8x16_t a, uint8x16_t b) { in test_vzipq_u8() 1929 uint8x16x2_t test_vtrnq_u8(uint8x16_t a, uint8x16_t b) { in test_vtrnq_u8()
|
D | arm-neon-vst.c | 2255 void test_vst1q_u8_x2(uint8_t *a, uint8x16x2_t b) { in test_vst1q_u8_x2()
|
D | aarch64-neon-ldst-one.c | 999 uint8x16x2_t test_vld2q_lane_u8(uint8_t const * ptr, uint8x16x2_t src) { in test_vld2q_lane_u8() 4423 void test_vst2q_lane_u8(uint8_t *a, uint8x16x2_t b) { in test_vst2q_lane_u8()
|
/external/llvm-project/llvm/test/CodeGen/ARM/ |
D | arm-vlddup.ll | 28 %struct.uint8x16x2_t = type { <16 x i8>, <16 x i8> } 47 declare %struct.uint8x16x2_t @llvm.arm.neon.vld2dup.v16i8.p0i8(i8*, i32) 176 define %struct.uint8x16x2_t @test_vld2q_dup_u8(i8* %src) { 178 %tmp = tail call %struct.uint8x16x2_t @llvm.arm.neon.vld2dup.v16i8.p0i8(i8* %src, i32 1) 179 ret %struct.uint8x16x2_t %tmp
|
D | arm-vld1.ll | 32 %struct.uint8x16x2_t = type { <16 x i8>, <16 x i8> } 64 declare %struct.uint8x16x2_t @llvm.arm.neon.vld1x2.v16i8.p0i8(i8*) nounwind readonly 223 define %struct.uint8x16x2_t @test_vld1q_u8_x2(i8* %a) nounwind { 224 %tmp = tail call %struct.uint8x16x2_t @llvm.arm.neon.vld1x2.v16i8.p0i8(i8* %a) 225 ret %struct.uint8x16x2_t %tmp
|
D | arm-vst1.ll | 32 ; %struct.uint8x16x2_t = type { <16 x i8>, <16 x i8> } 57 %struct.uint8x16x2_t = type { [2 x <16 x i8>] } 332 define void @test_vst1q_u8_x2(i8* %a, %struct.uint8x16x2_t %b) nounwind { 334 %b0 = extractvalue %struct.uint8x16x2_t %b, 0, 0 335 %b1 = extractvalue %struct.uint8x16x2_t %b, 0, 1
|
/external/libhevc/encoder/arm/ |
D | ihevce_itrans_recon_neon.c | 144 uint8x16x2_t a6; in ihevce_itrans_recon_dc_8x8_chroma_neon() 199 uint8x16x2_t a4; in ihevce_itrans_recon_dc_16x16_chroma_neon()
|
D | ihevce_scale_by_2_neon.c | 79 uint8x16x2_t src = vld2q_u8(pu1_src_tmp); in ihevce_horz_scale_neon_w16()
|
/external/gemmlowp/internal/ |
D | pack_neon.h | 61 uint8x16x2_t src_lines_intertwined_2x[2 * kCells]; in Pack() 68 uint8x16x2_t src_lines_intertwined_4x[2 * kCells]; in Pack()
|
/external/libaom/libaom/av1/common/arm/ |
D | transpose_neon.h | 34 const uint8x16x2_t b0 = in transpose_u8_8x8() 36 const uint8x16x2_t b1 = in transpose_u8_8x8()
|
/external/libaom/libaom/aom_dsp/simd/ |
D | v128_intrinsics_arm.h | 420 uint8x16x2_t r = vzipq_u8(vreinterpretq_u8_s64(y), vreinterpretq_u8_s64(x)); in v128_ziplo_8() 430 uint8x16x2_t r = vzipq_u8(vreinterpretq_u8_s64(y), vreinterpretq_u8_s64(x)); in v128_ziphi_8() 504 uint8x16x2_t r = vuzpq_u8(vreinterpretq_u8_s64(y), vreinterpretq_u8_s64(x)); in v128_unziplo_8() 514 uint8x16x2_t r = vuzpq_u8(vreinterpretq_u8_s64(y), vreinterpretq_u8_s64(x)); in v128_unziphi_8()
|
/external/libvpx/libvpx/vp9/encoder/arm/neon/ |
D | vp9_frame_scale_neon.c | 39 const uint8x16x2_t s = vld2q_u8(src); in scale_plane_2_to_1_phase_0() 118 const uint8x16x2_t s0 = vld2q_u8(src0); in scale_plane_2_to_1_bilinear() 121 const uint8x16x2_t s1 = vld2q_u8(src1); in scale_plane_2_to_1_bilinear()
|
/external/webp/src/dsp/ |
D | dec_neon.c | 95 const uint8x16x2_t row01 = vtrnq_u8(vreinterpretq_u8_u32(in.val[0]), in Load4x16_NEON() 97 const uint8x16x2_t row23 = vtrnq_u8(vreinterpretq_u8_u32(in.val[2]), in Load4x16_NEON() 188 const uint8x16x2_t row01 = vtrnq_u8(row0, row1); // u00 u10 u02 u12 ... in Load8x8x2T_NEON() 190 const uint8x16x2_t row23 = vtrnq_u8(row2, row3); // u20 u30 u22 u32 ... in Load8x8x2T_NEON() 192 const uint8x16x2_t row45 = vtrnq_u8(row4, row5); // ... in Load8x8x2T_NEON() 193 const uint8x16x2_t row67 = vtrnq_u8(row6, row7); // ... in Load8x8x2T_NEON()
|
/external/libvpx/libvpx/vp8/common/arm/neon/ |
D | vp8_loopfilter_neon.c | 328 uint8x16x2_t q2tmp8, q2tmp9, q2tmp10, q2tmp11; in vp8_loop_filter_vertical_edge_y_neon() 440 uint8x16x2_t q2tmp8, q2tmp9, q2tmp10, q2tmp11; in vp8_loop_filter_vertical_edge_uv_neon()
|