/external/libgav1/libgav1/src/dsp/arm/ |
D | distance_weighted_blend_neon.cc | 217 inline uint16x4x4_t ComputeWeightedAverage8(const uint16x4x4_t pred0, in ComputeWeightedAverage8() 218 const uint16x4x4_t pred1, in ComputeWeightedAverage8() 235 uint16x4x4_t result; in ComputeWeightedAverage8() 261 inline uint16x4x4_t LoadU16x4_x4(uint16_t const* ptr) { in LoadU16x4_x4() 262 uint16x4x4_t x; in LoadU16x4_x4() 297 const uint16x4x4_t src0 = LoadU16x4_x4(pred_0); in DistanceWeightedBlend_NEON() 298 const uint16x4x4_t src1 = LoadU16x4_x4(pred_1); in DistanceWeightedBlend_NEON() 299 const uint16x4x4_t res = ComputeWeightedAverage8(src0, src1, weights); in DistanceWeightedBlend_NEON() 314 const uint16x4x4_t src0 = LoadU16x4_x4(pred_0 + x); in DistanceWeightedBlend_NEON() 315 const uint16x4x4_t src1 = LoadU16x4_x4(pred_1 + x); in DistanceWeightedBlend_NEON() [all …]
|
/external/llvm-project/llvm/test/CodeGen/ARM/ |
D | arm-vlddup.ll | 6 %struct.uint16x4x4_t = type { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> } 43 declare %struct.uint16x4x4_t @llvm.arm.neon.vld4dup.v4i16.p0i8(i8*, i32) 125 define %struct.uint16x4x4_t @test_vld4_dup_u16(i8* %src) { 127 %tmp = tail call %struct.uint16x4x4_t @llvm.arm.neon.vld4dup.v4i16.p0i8(i8* %src, i32 2) 128 ret %struct.uint16x4x4_t %tmp
|
D | arm-vld1.ll | 6 %struct.uint16x4x4_t = type { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> } 38 declare %struct.uint16x4x4_t @llvm.arm.neon.vld1x4.v4i16.p0i16(i16*) nounwind readonly 84 define %struct.uint16x4x4_t @test_vld1_u16_x4(i16* %a) nounwind { 85 %tmp = tail call %struct.uint16x4x4_t @llvm.arm.neon.vld1x4.v4i16.p0i16(i16* %a) 86 ret %struct.uint16x4x4_t %tmp
|
D | arm-vst1.ll | 6 ; %struct.uint16x4x4_t = type { <4 x i16>, <4 x i16>, <4 x i16>, <4 x i16> } 38 %struct.uint16x4x4_t = type { [4 x <4 x i16>] } 116 define void @test_vst1_u16_x4(i16* %a, %struct.uint16x4x4_t %b) nounwind { 118 %b0 = extractvalue %struct.uint16x4x4_t %b, 0, 0 119 %b1 = extractvalue %struct.uint16x4x4_t %b, 0, 1 120 %b2 = extractvalue %struct.uint16x4x4_t %b, 0, 2 121 %b3 = extractvalue %struct.uint16x4x4_t %b, 0, 3
|
/external/skqp/third_party/skcms/src/ |
D | Transform_inl.h | 745 uint16x4x4_t v = vld4_u16(rgba); in exec_ops() 785 uint16x4x4_t v = vld4_u16(rgba); in exec_ops() 824 uint16x4x4_t v = vld4_u16(rgba); in exec_ops() 1064 uint16x4x4_t v = {{ in exec_ops() 1107 uint16x4x4_t v = {{ in exec_ops() 1155 uint16x4x4_t v = {{ in exec_ops()
|
/external/llvm-project/clang/test/CodeGen/ |
D | arm-neon-vld.c | 533 uint16x4x4_t test_vld1_u16_x4(uint16_t const *a) { in test_vld1_u16_x4() 1938 void test_vld4_dup_u16(uint16x4x4_t *dest, const uint16_t *src) { in test_vld4_dup_u16()
|
D | arm-neon-vst.c | 876 void test_vst1_u16_x4(uint16_t *a, uint16x4x4_t b) { in test_vst1_u16_x4()
|
D | aarch64-neon-ldst-one.c | 3581 uint16x4x4_t test_vld4_lane_u16(uint16_t *a, uint16x4x4_t b) { in test_vld4_lane_u16() 6395 void test_vst4_lane_u16(uint16_t *a, uint16x4x4_t b) { in test_vst4_lane_u16()
|
D | arm_neon_intrinsics.c | 5993 uint16x4x4_t test_vld4_u16(uint16_t const * a) { in test_vld4_u16() 6405 uint16x4x4_t test_vld4_lane_u16(uint16_t const * a, uint16x4x4_t b) { in test_vld4_lane_u16() 18461 void test_vst4_u16(uint16_t * a, uint16x4x4_t b) { in test_vst4_u16() 19118 void test_vst4_lane_u16(uint16_t * a, uint16x4x4_t b) { in test_vst4_lane_u16()
|
D | aarch64-neon-intrinsics.c | 10475 uint16x4x4_t test_vld4_u16(uint16_t const *a) { in test_vld4_u16() 12786 void test_vst4_u16(uint16_t *a, uint16x4x4_t b) { in test_vst4_u16()
|
/external/clang/test/CodeGen/ |
D | aarch64-neon-ldst-one.c | 1593 uint16x4x4_t test_vld4_dup_u16(uint16_t *a) { in test_vld4_dup_u16() 4759 uint16x4x4_t test_vld4_lane_u16(uint16_t *a, uint16x4x4_t b) { in test_vld4_lane_u16() 7573 void test_vst4_lane_u16(uint16_t *a, uint16x4x4_t b) { in test_vst4_lane_u16()
|
D | arm_neon_intrinsics.c | 7334 uint16x4x4_t test_vld4_u16(uint16_t const * a) { in test_vld4_u16() 7536 uint16x4x4_t test_vld4_dup_u16(uint16_t const * a) { in test_vld4_dup_u16() 8159 uint16x4x4_t test_vld4_lane_u16(uint16_t const * a, uint16x4x4_t b) { in test_vld4_lane_u16() 20946 void test_vst4_u16(uint16_t * a, uint16x4x4_t b) { in test_vst4_u16() 21604 void test_vst4_lane_u16(uint16_t * a, uint16x4x4_t b) { in test_vst4_lane_u16()
|
D | aarch64-neon-intrinsics.c | 11303 uint16x4x4_t test_vld4_u16(uint16_t const *a) { in test_vld4_u16() 13614 void test_vst4_u16(uint16_t *a, uint16x4x4_t b) { in test_vst4_u16() 15245 uint16x4x4_t test_vld1_u16_x4(uint16_t const *a) { in test_vld1_u16_x4() 17505 void test_vst1_u16_x4(uint16_t *a, uint16x4x4_t b) { in test_vst1_u16_x4()
|
/external/skia/third_party/skcms/src/ |
D | Transform_inl.h | 932 uint16x4x4_t v = vld4_u16(rgba); in exec_ops() 983 uint16x4x4_t v = vld4_u16(rgba); in exec_ops() 1033 uint16x4x4_t v = vld4_u16(rgba); in exec_ops() 1356 uint16x4x4_t v = {{ in exec_ops() 1414 uint16x4x4_t v = {{ in exec_ops() 1477 uint16x4x4_t v = {{ in exec_ops()
|
/external/libjpeg-turbo/simd/arm/ |
D | jidctint-neon.c | 683 uint16x4x4_t cols_01_23_45_67 = { { in jsimd_idct_islow_pass2_regular() 786 uint16x4x4_t cols_01_23_45_67 = { { in jsimd_idct_islow_pass2_sparse()
|
/external/skia/include/private/ |
D | SkNx_neon.h | 311 uint16x4x4_t rgba = vld4_u16((const uint16_t*)ptr); in Load4() 324 uint16x4x4_t rgba = {{ in Store4()
|
/external/skqp/include/private/ |
D | SkNx_neon.h | 333 uint16x4x4_t rgba = vld4_u16((const uint16_t*)ptr); in Load4() 346 uint16x4x4_t rgba = {{ in Store4()
|
/external/skqp/src/opts/ |
D | SkRasterPipeline_opts.h | 246 uint16x4x4_t rgba; 261 if ( true ) { vst4_lane_u16(ptr + 0, (uint16x4x4_t{{r,g,b,a}}), 0); } 262 if (tail > 1) { vst4_lane_u16(ptr + 4, (uint16x4x4_t{{r,g,b,a}}), 1); } 263 if (tail > 2) { vst4_lane_u16(ptr + 8, (uint16x4x4_t{{r,g,b,a}}), 2); } 265 vst4_u16(ptr, (uint16x4x4_t{{r,g,b,a}}));
|
/external/neon_2_sse/ |
D | NEON_2_SSE.h | 302 typedef struct int16x4x4_t uint16x4x4_t; typedef 1383 _NEON2SSESTORAGE uint16x4x4_t vld4_u16(__transfersize(16) uint16_t const * ptr); // VLD4.16 {d0, d1… 1420 _NEON2SSESTORAGE uint16x4x4_t vld4_dup_u16(__transfersize(4) uint16_t const * ptr); // VLD4.16 {d0[… 1475 _NEON2SSESTORAGE uint16x4x4_t vld4_lane_u16(__transfersize(4) uint16_t const * ptr, uint16x4x4_t sr… 1540 _NEON2SSESTORAGE void vst4_u16(__transfersize(16) uint16_t * ptr, uint16x4x4_t val); // VST4.16 {d0… 1594 _NEON2SSESTORAGE void vst4_lane_u16(__transfersize(4) uint16_t * ptr, uint16x4x4_t val, __constrang… 10349 _NEON2SSESTORAGE uint16x4x4_t vld4_u16(__transfersize(16) uint16_t const * ptr); // VLD4.16 {d0, d1… 10350 _NEON2SSE_INLINE uint16x4x4_t vld4_u16(__transfersize(16) uint16_t const * ptr) // VLD4.16 {d0, d1,… 10352 uint16x4x4_t v; 10615 _NEON2SSESTORAGE uint16x4x4_t vld4_dup_u16(__transfersize(4) uint16_t const * ptr); // VLD4.16 {d0[… [all …]
|
/external/skia/src/opts/ |
D | SkRasterPipeline_opts.h | 267 uint16x4x4_t rgba; 283 if ( true ) { vst4_lane_u16(ptr + 0, (uint16x4x4_t{{r,g,b,a}}), 0); } 284 if (tail > 1) { vst4_lane_u16(ptr + 4, (uint16x4x4_t{{r,g,b,a}}), 1); } 285 if (tail > 2) { vst4_lane_u16(ptr + 8, (uint16x4x4_t{{r,g,b,a}}), 2); } 287 vst4_u16(ptr, (uint16x4x4_t{{r,g,b,a}}));
|