1 /* 2 3 Copyright (c) 2009, 2010, 2011, 2013 STMicroelectronics 4 Written by Christophe Lyon 5 6 Permission is hereby granted, free of charge, to any person obtaining a copy 7 of this software and associated documentation files (the "Software"), to deal 8 in the Software without restriction, including without limitation the rights 9 to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 10 copies of the Software, and to permit persons to whom the Software is 11 furnished to do so, subject to the following conditions: 12 13 The above copyright notice and this permission notice shall be included in 14 all copies or substantial portions of the Software. 15 16 THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 17 IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 18 FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 19 AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 20 LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, 21 OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN 22 THE SOFTWARE. 23 24 */ 25 26 #if defined(__arm__) || defined(__aarch64__) 27 #include <arm_neon.h> 28 #else 29 #include "stm-arm-neon.h" 30 #endif 31 32 #include "stm-arm-neon-ref.h" 33 34 #define TEST_MSG "VLD1_LANE/VLD1_LANEQ" 35 void exec_vld1_lane (void) 36 { 37 /* Fill vector_src with 0xAA, then load 1 lane */ 38 #define TEST_VLD1_LANE(Q, T1, T2, W, N, L) \ 39 memset (VECT_VAR(buffer_src, T1, W, N), 0xAA, W/8*N); \ 40 VECT_VAR(vector_src, T1, W, N) = \ 41 vld1##Q##_##T2##W(VECT_VAR(buffer_src, T1, W, N)); \ 42 VECT_VAR(vector, T1, W, N) = \ 43 vld1##Q##_lane_##T2##W(VECT_VAR(buffer, T1, W, N), \ 44 VECT_VAR(vector_src, T1, W, N), L); \ 45 vst1##Q##_##T2##W(VECT_VAR(result, T1, W, N), VECT_VAR(vector, T1, W, N)) 46 47 /* With ARM RVCT, we need to declare variables before any executable 48 statement */ 49 DECL_VARIABLE_ALL_VARIANTS(vector); 50 #if defined(__ARM_FP16_FORMAT_IEEE) && ( ((__ARM_FP & 0x2) != 0) || ((__ARM_NEON_FP16_INTRINSICS & 1) != 0) ) 51 DECL_VARIABLE(vector, float, 16, 4); 52 DECL_VARIABLE(vector, float, 16, 8); 53 #endif 54 DECL_VARIABLE_ALL_VARIANTS(vector_src); 55 #if defined(__ARM_FP16_FORMAT_IEEE) && ( ((__ARM_FP & 0x2) != 0) || ((__ARM_NEON_FP16_INTRINSICS & 1) != 0) ) 56 DECL_VARIABLE(vector_src, float, 16, 4); 57 DECL_VARIABLE(vector_src, float, 16, 8); 58 #endif 59 60 ARRAY(buffer_src, int, 8, 8); 61 ARRAY(buffer_src, int, 16, 4); 62 ARRAY(buffer_src, int, 32, 2); 63 ARRAY(buffer_src, int, 64, 1); 64 ARRAY(buffer_src, uint, 8, 8); 65 ARRAY(buffer_src, uint, 16, 4); 66 ARRAY(buffer_src, uint, 32, 2); 67 ARRAY(buffer_src, uint, 64, 1); 68 ARRAY(buffer_src, poly, 8, 8); 69 ARRAY(buffer_src, poly, 16, 4); 70 ARRAY(buffer_src, float, 32, 2); 71 #if defined(__ARM_FP16_FORMAT_IEEE) && ( ((__ARM_FP & 0x2) != 0) || ((__ARM_NEON_FP16_INTRINSICS & 1) != 0) ) 72 ARRAY(buffer_src, float, 16, 4); 73 #endif 74 75 ARRAY(buffer_src, int, 8, 16); 76 ARRAY(buffer_src, int, 16, 8); 77 ARRAY(buffer_src, int, 32, 4); 78 ARRAY(buffer_src, int, 64, 2); 79 ARRAY(buffer_src, uint, 8, 16); 80 ARRAY(buffer_src, uint, 16, 8); 81 ARRAY(buffer_src, uint, 32, 4); 82 ARRAY(buffer_src, uint, 64, 2); 83 ARRAY(buffer_src, poly, 8, 16); 84 ARRAY(buffer_src, poly, 16, 8); 85 ARRAY(buffer_src, float, 32, 4); 86 #if defined(__ARM_FP16_FORMAT_IEEE) && ( ((__ARM_FP & 0x2) != 0) || ((__ARM_NEON_FP16_INTRINSICS & 1) != 0) ) 87 ARRAY(buffer_src, float, 16, 8); 88 #endif 89 90 clean_results (); 91 92 /* Choose lane arbitrarily */ 93 TEST_VLD1_LANE(, int, s, 8, 8, 6); 94 TEST_VLD1_LANE(, int, s, 16, 4, 3); 95 TEST_VLD1_LANE(, int, s, 32, 2, 1); 96 TEST_VLD1_LANE(, int, s, 64, 1, 0); 97 TEST_VLD1_LANE(, uint, u, 8, 8, 7); 98 TEST_VLD1_LANE(, uint, u, 16, 4, 3); 99 TEST_VLD1_LANE(, uint, u, 32, 2, 1); 100 TEST_VLD1_LANE(, uint, u, 64, 1, 0); 101 TEST_VLD1_LANE(, poly, p, 8, 8, 7); 102 TEST_VLD1_LANE(, poly, p, 16, 4, 3); 103 TEST_VLD1_LANE(, float, f, 32, 2, 1); 104 #if defined(__ARM_FP16_FORMAT_IEEE) && ( ((__ARM_FP & 0x2) != 0) || ((__ARM_NEON_FP16_INTRINSICS & 1) != 0) ) 105 TEST_VLD1_LANE(, float, f, 16, 4, 2); 106 #endif 107 108 TEST_VLD1_LANE(q, int, s, 8, 16, 15); 109 TEST_VLD1_LANE(q, int, s, 16, 8, 5); 110 TEST_VLD1_LANE(q, int, s, 32, 4, 2); 111 TEST_VLD1_LANE(q, int, s, 64, 2, 1); 112 TEST_VLD1_LANE(q, uint, u, 8, 16, 12); 113 TEST_VLD1_LANE(q, uint, u, 16, 8, 6); 114 TEST_VLD1_LANE(q, uint, u, 32, 4, 2); 115 TEST_VLD1_LANE(q, uint, u, 64, 2, 0); 116 TEST_VLD1_LANE(q, poly, p, 8, 16, 12); 117 TEST_VLD1_LANE(q, poly, p, 16, 8, 6); 118 TEST_VLD1_LANE(q, float, f, 32, 4, 2); 119 #if defined(__ARM_FP16_FORMAT_IEEE) && ( ((__ARM_FP & 0x2) != 0) || ((__ARM_NEON_FP16_INTRINSICS & 1) != 0) ) 120 TEST_VLD1_LANE(q, float, f, 16, 8, 5); 121 #endif 122 123 #ifndef __CC_ARM 124 /* Check runtime assertions. With RVCT, the check is performed at 125 compile-time */ 126 // TEST_VLD1_LANE(, int, s, 64, 1, 1); 127 #endif 128 129 dump_results_hex (TEST_MSG); 130 } 131
