/external/llvm-project/llvm/test/CodeGen/ARM/ |
D | arm-vlddup-update.ll | 4 %struct.uint32x2x2_t = type { <2 x i32>, <2 x i32> } 8 declare %struct.uint32x2x2_t @llvm.arm.neon.vld2dup.v2i32.p0i8(i8*, i32) 15 define i8* @test_vld2_dup_update(%struct.uint32x2x2_t* %dest, i8* %src) { 17 %tmp = tail call %struct.uint32x2x2_t @llvm.arm.neon.vld2dup.v2i32.p0i8(i8* %src, i32 4) 18 store %struct.uint32x2x2_t %tmp, %struct.uint32x2x2_t* %dest, align 8
|
D | arm-vlddup.ll | 8 %struct.uint32x2x2_t = type { <2 x i32>, <2 x i32> } 34 declare %struct.uint32x2x2_t @llvm.arm.neon.vld2dup.v2i32.p0i8(i8*, i32) 69 define %struct.uint32x2x2_t @test_vld2_dup_u32(i8* %src) { 71 %tmp = tail call %struct.uint32x2x2_t @llvm.arm.neon.vld2dup.v2i32.p0i8(i8* %src, i32 4) 72 ret %struct.uint32x2x2_t %tmp
|
D | arm-vld1.ll | 8 %struct.uint32x2x2_t = type { <2 x i32>, <2 x i32> } 40 declare %struct.uint32x2x2_t @llvm.arm.neon.vld1x2.v2i32.p0i32(i32*) nounwind readonly 91 define %struct.uint32x2x2_t @test_vld1_u32_x2(i32* %a) nounwind { 92 %tmp = tail call %struct.uint32x2x2_t @llvm.arm.neon.vld1x2.v2i32.p0i32(i32* %a) 93 ret %struct.uint32x2x2_t %tmp
|
D | arm-vst1.ll | 8 ; %struct.uint32x2x2_t = type { <2 x i32>, <2 x i32> } 39 %struct.uint32x2x2_t = type { [2 x <2 x i32>] } 128 define void @test_vst1_u32_x2(i32* %a, %struct.uint32x2x2_t %b) nounwind { 130 %b0 = extractvalue %struct.uint32x2x2_t %b, 0, 0 131 %b1 = extractvalue %struct.uint32x2x2_t %b, 0, 1
|
/external/libaom/libaom/av1/common/arm/ |
D | transpose_neon.h | 118 const uint32x2x2_t c0 = vtrn_u32(vreinterpret_u32_u16(b0.val[0]), in transpose_u8_4x4() 151 const uint32x2x2_t b0 = in transpose_u8_4x8() 153 const uint32x2x2_t b1 = in transpose_u8_4x8() 155 const uint32x2x2_t b2 = in transpose_u8_4x8() 157 const uint32x2x2_t b3 = in transpose_u8_4x8() 228 uint32x2x2_t c0 = vtrn_u32(vreinterpret_u32_u16(b0.val[0]), in transpose_u16_4x8() 230 uint32x2x2_t c1 = vtrn_u32(vreinterpret_u32_u16(b0.val[1]), in transpose_u16_4x8() 232 uint32x2x2_t c2 = vtrn_u32(vreinterpret_u32_u16(b2.val[0]), in transpose_u16_4x8() 234 uint32x2x2_t c3 = vtrn_u32(vreinterpret_u32_u16(b2.val[1]), in transpose_u16_4x8()
|
/external/libvpx/libvpx/vp8/common/arm/neon/ |
D | loopfiltersimpleverticaledge_neon.c | 106 const uint32x2x2_t r04_u32 = in read_4x8() 108 const uint32x2x2_t r15_u32 = in read_4x8() 110 const uint32x2x2_t r26_u32 = in read_4x8() 112 const uint32x2x2_t r37_u32 = in read_4x8()
|
D | sixtappredict_neon.c | 50 const uint32x2x2_t a_shuf = vzip_u32(vreinterpret_u32_u8(vget_low_u8(a)), in filter_add_accumulate() 52 const uint32x2x2_t b_shuf = vzip_u32(vreinterpret_u32_u8(vget_low_u8(b)), in filter_add_accumulate() 61 const uint32x2x2_t a_shuf = vzip_u32(vreinterpret_u32_u8(vget_low_u8(a)), in filter_sub_accumulate() 63 const uint32x2x2_t b_shuf = vzip_u32(vreinterpret_u32_u8(vget_low_u8(b)), in filter_sub_accumulate() 169 uint32x2x2_t s01_f3_q, s23_f3_q, s01_f5_q, s23_f5_q; in vp8_sixtap_predict4x4_neon()
|
/external/XNNPACK/src/x32-zip/ |
D | xm-neon.c | 74 const uint32x2x2_t vxy = vzip_u32(vx, vy); in xnn_x32_zip_xm_ukernel__neon() 75 const uint32x2x2_t vzw = vzip_u32(vz, vw); in xnn_x32_zip_xm_ukernel__neon()
|
D | x2-neon.c | 34 uint32x2x2_t vxy; in xnn_x32_zip_x2_ukernel__neon()
|
/external/libaom/libaom/aom_dsp/arm/ |
D | loopfilter_neon.c | 25 uint32x2x2_t p0q0_p1q1; in lpf_mask() 57 uint32x2x2_t p0q0_p1q1; in lpf_mask2() 120 uint32x2x2_t p0q0_p1q1; in lpf_mask3_chroma() 623 uint32x2x2_t p6q6_p2q2, p5q5_p1q1, p4q4_p0q0, pxqx_p3q3; in aom_lpf_vertical_14_neon() 700 uint32x2x2_t p2q2_p1q1, p3q3_p0q0; in aom_lpf_vertical_8_neon() 743 uint32x2x2_t p2q2_p1q1, pxqy_p0q0; in aom_lpf_vertical_6_neon() 786 uint32x2x2_t p1q0_p0q1, p1q1_p0q0, p1p0_q1q0; in aom_lpf_vertical_4_neon()
|
/external/libhevc/encoder/arm/ |
D | ihevce_coarse_layer_sad_neon.c | 171 uint32x2x2_t l = vtrn_u32(a[0], a[1]); in hme_store_4x4_sads_high_speed_neon() 172 uint32x2x2_t m = vtrn_u32(a[2], a[3]); in hme_store_4x4_sads_high_speed_neon() 322 uint32x2x2_t l = vtrn_u32(a[0], a[1]); in hme_store_4x4_sads_high_quality_neon() 323 uint32x2x2_t m = vtrn_u32(a[2], a[3]); in hme_store_4x4_sads_high_quality_neon()
|
/external/libgav1/libgav1/src/dsp/arm/ |
D | common_neon.h | 430 const uint32x2x2_t c = vtrn_u32(a_32, b_32); in Interleave32() 484 const uint32x2x2_t d = in Transpose4x4() 489 const uint32x2x2_t e = in Transpose4x4() 507 const uint32x2x2_t d = in Transpose4x4()
|
D | inverse_transform_neon.cc | 195 uint32x2x2_t c0 = vtrn_u32(vreinterpret_u32_u16(b0.val[0]), in Transpose4x8To8x4() 197 uint32x2x2_t c1 = vtrn_u32(vreinterpret_u32_u16(b0.val[1]), in Transpose4x8To8x4() 199 uint32x2x2_t c2 = vtrn_u32(vreinterpret_u32_u16(b2.val[0]), in Transpose4x8To8x4() 201 uint32x2x2_t c3 = vtrn_u32(vreinterpret_u32_u16(b2.val[1]), in Transpose4x8To8x4()
|
/external/libvpx/libvpx/vpx_dsp/arm/ |
D | vpx_convolve_avg_neon.c | 29 uint32x2x2_t s01; in vpx_convolve_avg_neon()
|
D | transpose_neon.h | 82 const uint32x2x2_t c0 = vtrn_u32(vreinterpret_u32_u16(b0.val[0]), in transpose_u8_4x4() 210 const uint32x2x2_t b0 = in transpose_u8_4x8() 212 const uint32x2x2_t b1 = in transpose_u8_4x8() 214 const uint32x2x2_t b2 = in transpose_u8_4x8() 216 const uint32x2x2_t b3 = in transpose_u8_4x8()
|
/external/llvm-project/clang/test/CodeGen/ |
D | arm-neon-vld.c | 553 uint32x2x2_t test_vld1_u32_x2(uint32_t const *a) { in test_vld1_u32_x2() 1559 void test_vld2_dup_u32(uint32x2x2_t *dest, const uint32_t *src) { in test_vld2_dup_u32()
|
D | aarch64-neon-perm.c | 1013 uint32x2x2_t test_vuzp_u32(uint32x2_t a, uint32x2_t b) { in test_vuzp_u32() 1397 uint32x2x2_t test_vzip_u32(uint32x2_t a, uint32x2_t b) { in test_vzip_u32() 1781 uint32x2x2_t test_vtrn_u32(uint32x2_t a, uint32x2_t b) { in test_vtrn_u32()
|
D | arm-neon-vst.c | 905 void test_vst1_u32_x2(uint32_t *a, uint32x2x2_t b) { in test_vst1_u32_x2()
|
D | aarch64-neon-ldst-one.c | 1563 uint32x2x2_t test_vld2_lane_u32(uint32_t *a, uint32x2x2_t b) { in test_vld2_lane_u32() 4808 void test_vst2_lane_u32(uint32_t *a, uint32x2x2_t b) { in test_vst2_lane_u32()
|
/external/clang/test/CodeGen/ |
D | aarch64-neon-perm.c | 1041 uint32x2x2_t test_vuzp_u32(uint32x2_t a, uint32x2_t b) { in test_vuzp_u32() 1505 uint32x2x2_t test_vzip_u32(uint32x2_t a, uint32x2_t b) { in test_vzip_u32() 1969 uint32x2x2_t test_vtrn_u32(uint32x2_t a, uint32x2_t b) { in test_vtrn_u32()
|
D | aarch64-neon-ldst-one.c | 599 uint32x2x2_t test_vld2_dup_u32(uint32_t *a) { in test_vld2_dup_u32() 2741 uint32x2x2_t test_vld2_lane_u32(uint32_t *a, uint32x2x2_t b) { in test_vld2_lane_u32() 5986 void test_vst2_lane_u32(uint32_t *a, uint32x2x2_t b) { in test_vst2_lane_u32()
|
D | arm_neon_intrinsics.c | 5020 uint32x2x2_t test_vld2_u32(uint32_t const * a) { in test_vld2_u32() 5216 uint32x2x2_t test_vld2_dup_u32(uint32_t const * a) { in test_vld2_dup_u32() 5717 uint32x2x2_t test_vld2_lane_u32(uint32_t const * a, uint32x2x2_t b) { in test_vld2_lane_u32() 18749 void test_vst2_u32(uint32_t * a, uint32x2x2_t b) { in test_vst2_u32() 19229 void test_vst2_lane_u32(uint32_t * a, uint32x2x2_t b) { in test_vst2_lane_u32() 22863 uint32x2x2_t test_vtrn_u32(uint32x2_t a, uint32x2_t b) { in test_vtrn_u32() 23403 uint32x2x2_t test_vuzp_u32(uint32x2_t a, uint32x2_t b) { in test_vuzp_u32() 23758 uint32x2x2_t test_vzip_u32(uint32x2_t a, uint32x2_t b) { in test_vzip_u32()
|
/external/llvm/test/CodeGen/AArch64/ |
D | neon-perm.ll | 8 %struct.uint32x2x2_t = type { [2 x <2 x i32>] } 2538 define %struct.uint32x2x2_t @test_vuzp_u32(<2 x i32> %a, <2 x i32> %b) { 2545 %.fca.0.0.insert = insertvalue %struct.uint32x2x2_t undef, <2 x i32> %vuzp.i, 0, 0 2546 %.fca.0.1.insert = insertvalue %struct.uint32x2x2_t %.fca.0.0.insert, <2 x i32> %vuzp1.i, 0, 1 2547 ret %struct.uint32x2x2_t %.fca.0.1.insert 2754 define %struct.uint32x2x2_t @test_vzip_u32(<2 x i32> %a, <2 x i32> %b) { 2761 %.fca.0.0.insert = insertvalue %struct.uint32x2x2_t undef, <2 x i32> %vzip.i, 0, 0 2762 %.fca.0.1.insert = insertvalue %struct.uint32x2x2_t %.fca.0.0.insert, <2 x i32> %vzip1.i, 0, 1 2763 ret %struct.uint32x2x2_t %.fca.0.1.insert 2970 define %struct.uint32x2x2_t @test_vtrn_u32(<2 x i32> %a, <2 x i32> %b) { [all …]
|
/external/llvm-project/llvm/test/CodeGen/AArch64/ |
D | neon-perm.ll | 8 %struct.uint32x2x2_t = type { [2 x <2 x i32>] } 2538 define %struct.uint32x2x2_t @test_vuzp_u32(<2 x i32> %a, <2 x i32> %b) { 2545 %.fca.0.0.insert = insertvalue %struct.uint32x2x2_t undef, <2 x i32> %vuzp.i, 0, 0 2546 %.fca.0.1.insert = insertvalue %struct.uint32x2x2_t %.fca.0.0.insert, <2 x i32> %vuzp1.i, 0, 1 2547 ret %struct.uint32x2x2_t %.fca.0.1.insert 2754 define %struct.uint32x2x2_t @test_vzip_u32(<2 x i32> %a, <2 x i32> %b) { 2761 %.fca.0.0.insert = insertvalue %struct.uint32x2x2_t undef, <2 x i32> %vzip.i, 0, 0 2762 %.fca.0.1.insert = insertvalue %struct.uint32x2x2_t %.fca.0.0.insert, <2 x i32> %vzip1.i, 0, 1 2763 ret %struct.uint32x2x2_t %.fca.0.1.insert 2970 define %struct.uint32x2x2_t @test_vtrn_u32(<2 x i32> %a, <2 x i32> %b) { [all …]
|
/external/neon_2_sse/ |
D | NEON_2_SSE.h | 242 typedef struct int32x2x2_t uint32x2x2_t; typedef 1340 _NEON2SSESTORAGE uint32x2x2_t vld2_u32(__transfersize(4) uint32_t const * ptr); // VLD2.32 {d0, d1}… 1397 _NEON2SSESTORAGE uint32x2x2_t vld2_dup_u32(__transfersize(2) uint32_t const * ptr); // VLD2.32 {d0[… 1442 _NEON2SSESTORAGE uint32x2x2_t vld2_lane_u32(__transfersize(2) uint32_t const * ptr, uint32x2x2_t sr… 1497 _NEON2SSESTORAGE void vst2_u32(__transfersize(4) uint32_t * ptr, uint32x2x2_t val); // VST2.32 {d0,… 1561 _NEON2SSESTORAGE void vst2_lane_u32(__transfersize(2) uint32_t * ptr, uint32x2x2_t val, __constrang… 2223 _NEON2SSESTORAGE uint32x2x2_t vtrn_u32(uint32x2_t a, uint32x2_t b); // VTRN.32 d0,d0 2242 _NEON2SSESTORAGE uint32x2x2_t vzip_u32(uint32x2_t a, uint32x2_t b); // VZIP.32 d0,d0 2261 _NEON2SSESTORAGE uint32x2x2_t vuzp_u32(uint32x2_t a, uint32x2_t b); // VUZP.32 d0,d0 9866 _NEON2SSESTORAGE uint32x2x2_t vld2_u32(__transfersize(4) uint32_t const * ptr); // VLD2.32 {d0, d1}… [all …]
|