arm_compute v19.11
diff --git a/documentation/direct__convolution1x1_8cl_source.xhtml b/documentation/direct__convolution1x1_8cl_source.xhtml
index 928a6fb..82654f0 100644
--- a/documentation/direct__convolution1x1_8cl_source.xhtml
+++ b/documentation/direct__convolution1x1_8cl_source.xhtml
@@ -40,7 +40,7 @@
<img alt="Compute Library" src="https://raw.githubusercontent.com/ARM-software/ComputeLibrary/gh-pages/ACL_logo.png" style="max-width: 100%;margin-top: 15px;margin-left: 10px"/>
<td style="padding-left: 0.5em;">
<div id="projectname">
-  <span id="projectnumber">19.08</span>
+  <span id="projectnumber">19.11</span>
</div>
</td>
</tr>
@@ -100,35 +100,35 @@
<div class="title">direct_convolution1x1.cl</div> </div>
</div><!--header-->
<div class="contents">
-<a href="direct__convolution1x1_8cl.xhtml">Go to the documentation of this file.</a><div class="fragment"><div class="line"><a name="l00001"></a><span class="lineno"> 1</span> <span class="comment">/*</span></div><div class="line"><a name="l00002"></a><span class="lineno"> 2</span> <span class="comment"> * Copyright (c) 2016-2018 ARM Limited.</span></div><div class="line"><a name="l00003"></a><span class="lineno"> 3</span> <span class="comment"> *</span></div><div class="line"><a name="l00004"></a><span class="lineno"> 4</span> <span class="comment"> * SPDX-License-Identifier: MIT</span></div><div class="line"><a name="l00005"></a><span class="lineno"> 5</span> <span class="comment"> *</span></div><div class="line"><a name="l00006"></a><span class="lineno"> 6</span> <span class="comment"> * Permission is hereby granted, free of charge, to any person obtaining a copy</span></div><div class="line"><a name="l00007"></a><span class="lineno"> 7</span> <span class="comment"> * of this software and associated documentation files (the "Software"), to</span></div><div class="line"><a name="l00008"></a><span class="lineno"> 8</span> <span class="comment"> * deal in the Software without restriction, including without limitation the</span></div><div class="line"><a name="l00009"></a><span class="lineno"> 9</span> <span class="comment"> * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or</span></div><div class="line"><a name="l00010"></a><span class="lineno"> 10</span> <span class="comment"> * sell copies of the Software, and to permit persons to whom the Software is</span></div><div class="line"><a name="l00011"></a><span class="lineno"> 11</span> <span class="comment"> * furnished to do so, subject to the following conditions:</span></div><div class="line"><a name="l00012"></a><span class="lineno"> 12</span> <span class="comment"> *</span></div><div class="line"><a name="l00013"></a><span class="lineno"> 13</span> <span class="comment"> * The above copyright notice and this permission notice shall be included in all</span></div><div class="line"><a name="l00014"></a><span class="lineno"> 14</span> <span class="comment"> * copies or substantial portions of the Software.</span></div><div class="line"><a name="l00015"></a><span class="lineno"> 15</span> <span class="comment"> *</span></div><div class="line"><a name="l00016"></a><span class="lineno"> 16</span> <span class="comment"> * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR</span></div><div class="line"><a name="l00017"></a><span class="lineno"> 17</span> <span class="comment"> * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,</span></div><div class="line"><a name="l00018"></a><span class="lineno"> 18</span> <span class="comment"> * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE</span></div><div class="line"><a name="l00019"></a><span class="lineno"> 19</span> <span class="comment"> * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER</span></div><div class="line"><a name="l00020"></a><span class="lineno"> 20</span> <span class="comment"> * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,</span></div><div class="line"><a name="l00021"></a><span class="lineno"> 21</span> <span class="comment"> * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE</span></div><div class="line"><a name="l00022"></a><span class="lineno"> 22</span> <span class="comment"> * SOFTWARE.</span></div><div class="line"><a name="l00023"></a><span class="lineno"> 23</span> <span class="comment"> */</span></div><div class="line"><a name="l00024"></a><span class="lineno"> 24</span> <span class="preprocessor">#include "<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml">helpers.h</a>"</span></div><div class="line"><a name="l00025"></a><span class="lineno"> 25</span> </div><div class="line"><a name="l00026"></a><span class="lineno"> 26</span> <span class="preprocessor">#undef CONVERT_SAT</span></div><div class="line"><a name="l00027"></a><span class="lineno"> 27</span> </div><div class="line"><a name="l00028"></a><span class="lineno"><a class="line" href="direct__convolution1x1_8cl.xhtml#aebbeb1f22eca3a3f4c3e019e8f419f39"> 28</a></span> <span class="preprocessor">#define ADD_OP(a, b) ((a) + (b))</span></div><div class="line"><a name="l00029"></a><span class="lineno"><a class="line" href="direct__convolution1x1_8cl.xhtml#ad3cc858846806e6b1d3694b9d0a2e6da"> 29</a></span> <span class="preprocessor">#define MUL_OP(a, b) ((a) * (b))</span></div><div class="line"><a name="l00030"></a><span class="lineno"><a class="line" href="direct__convolution1x1_8cl.xhtml#a1f15728672380ade7a238f5e783d54d2"> 30</a></span> <span class="preprocessor">#define CONVERT_SAT(a, b) ((a))</span></div><div class="line"><a name="l00031"></a><span class="lineno"> 31</span> </div><div class="line"><a name="l00032"></a><span class="lineno"> 32</span> <span class="preprocessor">#if defined(DATA_TYPE) && defined(DATA_SIZE) && defined(STRIDE_X) && defined(WEIGHTS_DEPTH)</span></div><div class="line"><a name="l00033"></a><span class="lineno"> 33</span> </div><div class="line"><a name="l00034"></a><span class="lineno"> 34</span> <span class="preprocessor">#if defined(DATA_LAYOUT_NHWC)</span></div><div class="line"><a name="l00035"></a><span class="lineno"> 35</span> </div><div class="line"><a name="l00036"></a><span class="lineno"> 36</span> <span class="preprocessor">#define PTR_TO_VALUE(PTR, DATA_TYPE) *((__global DATA_TYPE *)(PTR))</span></div><div class="line"><a name="l00037"></a><span class="lineno"> 37</span> </div><div class="line"><a name="l00076"></a><span class="lineno"> 76</span> __kernel <span class="keywordtype">void</span> direct_convolution1x1_nhwc(</div><div class="line"><a name="l00077"></a><span class="lineno"> 77</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>),</div><div class="line"><a name="l00078"></a><span class="lineno"> 78</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>),</div><div class="line"><a name="l00079"></a><span class="lineno"> 79</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>),</div><div class="line"><a name="l00080"></a><span class="lineno"> 80</span> #ifdef HAS_BIAS</div><div class="line"><a name="l00081"></a><span class="lineno"> 81</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(biases),</div><div class="line"><a name="l00082"></a><span class="lineno"> 82</span> #endif <span class="comment">/* defined(HAS_BIAS) */</span></div><div class="line"><a name="l00083"></a><span class="lineno"> 83</span>  <span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> weights_stride_w)</div><div class="line"><a name="l00084"></a><span class="lineno"> 84</span> {</div><div class="line"><a name="l00085"></a><span class="lineno"> 85</span>  <a class="code" href="struct_image.xhtml">Image</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a> = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#aebe814363556c244be043b13e7969197">CONVERT_TO_IMAGE_STRUCT</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>);</div><div class="line"><a name="l00086"></a><span class="lineno"> 86</span>  <a class="code" href="struct_tensor3_d.xhtml">Tensor3D</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a> = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a79e8e562daa6599317d2d1cd86ef1bf2">CONVERT_TO_TENSOR3D_STRUCT_NO_STEP</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>);</div><div class="line"><a name="l00087"></a><span class="lineno"> 87</span>  <a class="code" href="struct_tensor3_d.xhtml">Tensor3D</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a> = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a31c8c760f08fb1a331b16b7c204321dc">CONVERT_TO_TENSOR3D_STRUCT</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>);</div><div class="line"><a name="l00088"></a><span class="lineno"> 88</span> </div><div class="line"><a name="l00089"></a><span class="lineno"> 89</span> <span class="preprocessor">#ifdef HAS_BIAS</span></div><div class="line"><a name="l00090"></a><span class="lineno"> 90</span>  <a class="code" href="struct_vector.xhtml">Vector</a> biases = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a64d779f80eeb923e0ab2313433f7b40b">CONVERT_TO_VECTOR_STRUCT_NO_STEP</a>(biases);</div><div class="line"><a name="l00091"></a><span class="lineno"> 91</span> <span class="preprocessor">#endif </span><span class="comment">/* defined(HAS_BIAS) */</span><span class="preprocessor"></span></div><div class="line"><a name="l00092"></a><span class="lineno"> 92</span> </div><div class="line"><a name="l00093"></a><span class="lineno"> 93</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(DATA_TYPE_PROMOTED, 8)</div><div class="line"><a name="l00094"></a><span class="lineno"> 94</span>  values = 0;</div><div class="line"><a name="l00095"></a><span class="lineno"> 95</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> id0 = get_global_id(0);</div><div class="line"><a name="l00096"></a><span class="lineno"> 96</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> id1 = get_global_id(1);</div><div class="line"><a name="l00097"></a><span class="lineno"> 97</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> id2 = get_global_id(2);</div><div class="line"><a name="l00098"></a><span class="lineno"> 98</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>.ptr += id0 * weights_stride_w;</div><div class="line"><a name="l00099"></a><span class="lineno"> 99</span>  __global uchar *src_addr = (__global uchar *)<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a009469e4d9b8fce3b6d5e97d2077827d">offset</a>(&<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>, 0, 0) - src_stride_x * id0 + id2 * STRIDE_Y * (int)src_stride_z;</div><div class="line"><a name="l00100"></a><span class="lineno"> 100</span> </div><div class="line"><a name="l00101"></a><span class="lineno"> 101</span>  <span class="keywordflow">for</span>(<span class="keyword">volatile</span> <span class="keywordtype">int</span> d = 0; d < WEIGHTS_DEPTH; ++d)</div><div class="line"><a name="l00102"></a><span class="lineno"> 102</span>  {</div><div class="line"><a name="l00103"></a><span class="lineno"> 103</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> weight = *(__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>.ptr;</div><div class="line"><a name="l00104"></a><span class="lineno"> 104</span> #<span class="keywordflow">if</span> STRIDE_X == 1</div><div class="line"><a name="l00105"></a><span class="lineno"> 105</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8)</div><div class="line"><a name="l00106"></a><span class="lineno"> 106</span>  col0 = (<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8))(</div><div class="line"><a name="l00107"></a><span class="lineno"> 107</span>  PTR_TO_VALUE(src_addr + 0 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00108"></a><span class="lineno"> 108</span>  PTR_TO_VALUE(src_addr + 1 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00109"></a><span class="lineno"> 109</span>  PTR_TO_VALUE(src_addr + 2 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00110"></a><span class="lineno"> 110</span>  PTR_TO_VALUE(src_addr + 3 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00111"></a><span class="lineno"> 111</span>  PTR_TO_VALUE(src_addr + 4 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00112"></a><span class="lineno"> 112</span>  PTR_TO_VALUE(src_addr + 5 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00113"></a><span class="lineno"> 113</span>  PTR_TO_VALUE(src_addr + 6 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00114"></a><span class="lineno"> 114</span>  PTR_TO_VALUE(src_addr + 7 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>));</div><div class="line"><a name="l00115"></a><span class="lineno"> 115</span> <span class="preprocessor">#elif STRIDE_X == 2 </span><span class="comment">/* STRIDE_X == 1 */</span><span class="preprocessor"></span></div><div class="line"><a name="l00116"></a><span class="lineno"> 116</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8)</div><div class="line"><a name="l00117"></a><span class="lineno"> 117</span>  col0 = (<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8))(</div><div class="line"><a name="l00118"></a><span class="lineno"> 118</span>  PTR_TO_VALUE(src_addr + 0 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00119"></a><span class="lineno"> 119</span>  PTR_TO_VALUE(src_addr + 2 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00120"></a><span class="lineno"> 120</span>  PTR_TO_VALUE(src_addr + 4 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00121"></a><span class="lineno"> 121</span>  PTR_TO_VALUE(src_addr + 6 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00122"></a><span class="lineno"> 122</span>  PTR_TO_VALUE(src_addr + 8 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00123"></a><span class="lineno"> 123</span>  PTR_TO_VALUE(src_addr + 10 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00124"></a><span class="lineno"> 124</span>  PTR_TO_VALUE(src_addr + 12 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00125"></a><span class="lineno"> 125</span>  PTR_TO_VALUE(src_addr + 14 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>));</div><div class="line"><a name="l00126"></a><span class="lineno"> 126</span> <span class="preprocessor">#else </span><span class="comment">/* STRIDE_X not equals 1 or 2 */</span><span class="preprocessor"></span></div><div class="line"><a name="l00127"></a><span class="lineno"> 127</span> <span class="preprocessor">#error "STRIDE_X larger than 2 is not supported"</span></div><div class="line"><a name="l00128"></a><span class="lineno"> 128</span> <span class="preprocessor">#endif </span><span class="comment">/* STRIDE_X == 2 */</span><span class="preprocessor"></span></div><div class="line"><a name="l00129"></a><span class="lineno"> 129</span>  values = <a class="code" href="direct__convolution1x1_8cl.xhtml#aebbeb1f22eca3a3f4c3e019e8f419f39">ADD_OP</a>(values, <a class="code" href="direct__convolution1x1_8cl.xhtml#ad3cc858846806e6b1d3694b9d0a2e6da">MUL_OP</a>((<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8))weight, col0));</div><div class="line"><a name="l00130"></a><span class="lineno"> 130</span> </div><div class="line"><a name="l00131"></a><span class="lineno"> 131</span>  src_addr += src_stride_x;</div><div class="line"><a name="l00132"></a><span class="lineno"> 132</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>.ptr += weights_stride_x;</div><div class="line"><a name="l00133"></a><span class="lineno"> 133</span>  }</div><div class="line"><a name="l00134"></a><span class="lineno"> 134</span> </div><div class="line"><a name="l00135"></a><span class="lineno"> 135</span> <span class="preprocessor">#ifdef HAS_BIAS</span></div><div class="line"><a name="l00136"></a><span class="lineno"> 136</span>  values = <a class="code" href="direct__convolution1x1_8cl.xhtml#aebbeb1f22eca3a3f4c3e019e8f419f39">ADD_OP</a>(values, (<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(DATA_TYPE_PROMOTED, 8)) * ((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a7e4940407322d6f0ccb8b6b86b856019">vector_offset</a>(&biases, id0))));</div><div class="line"><a name="l00137"></a><span class="lineno"> 137</span> <span class="preprocessor">#endif </span><span class="comment">/* defined(HAS_BIAS) */</span><span class="preprocessor"></span></div><div class="line"><a name="l00138"></a><span class="lineno"> 138</span> </div><div class="line"><a name="l00139"></a><span class="lineno"> 139</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr) = values.s0;</div><div class="line"><a name="l00140"></a><span class="lineno"> 140</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 1 * dst_stride_y)) = values.s1;</div><div class="line"><a name="l00141"></a><span class="lineno"> 141</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 2 * dst_stride_y)) = values.s2;</div><div class="line"><a name="l00142"></a><span class="lineno"> 142</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 3 * dst_stride_y)) = values.s3;</div><div class="line"><a name="l00143"></a><span class="lineno"> 143</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 4 * dst_stride_y)) = values.s4;</div><div class="line"><a name="l00144"></a><span class="lineno"> 144</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 5 * dst_stride_y)) = values.s5;</div><div class="line"><a name="l00145"></a><span class="lineno"> 145</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 6 * dst_stride_y)) = values.s6;</div><div class="line"><a name="l00146"></a><span class="lineno"> 146</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 7 * dst_stride_y)) = values.s7;</div><div class="line"><a name="l00147"></a><span class="lineno"> 147</span> }</div><div class="line"><a name="l00148"></a><span class="lineno"> 148</span> <span class="preprocessor">#endif // defined(DATA_LAYOUT_NHWC)</span></div><div class="line"><a name="l00149"></a><span class="lineno"> 149</span> </div><div class="line"><a name="l00150"></a><span class="lineno"> 150</span> <span class="preprocessor">#if STRIDE_X == 3</span></div><div class="line"><a name="l00151"></a><span class="lineno"> 151</span> <span class="preprocessor">#define INPUT_PIXEL_STR(data_size) extract_input_stride3_##data_size</span></div><div class="line"><a name="l00152"></a><span class="lineno"> 152</span> <span class="preprocessor">#define INPUT_PIXEL(data_size) INPUT_PIXEL_STR(data_size)</span></div><div class="line"><a name="l00153"></a><span class="lineno"> 153</span> <span class="preprocessor">#elif STRIDE_X == 2</span></div><div class="line"><a name="l00154"></a><span class="lineno"> 154</span> <span class="preprocessor">#define INPUT_PIXEL(data_size) extract_input_stride2</span></div><div class="line"><a name="l00155"></a><span class="lineno"> 155</span> <span class="preprocessor">#elif STRIDE_X == 1</span></div><div class="line"><a name="l00156"></a><span class="lineno"> 156</span> <span class="preprocessor">#define INPUT_PIXEL(data_size) extract_input_stride1</span></div><div class="line"><a name="l00157"></a><span class="lineno"> 157</span> <span class="preprocessor">#else </span><span class="comment">/* STRIDE_X not equals 1, 2 or 3 */</span><span class="preprocessor"></span></div><div class="line"><a name="l00158"></a><span class="lineno"> 158</span> <span class="preprocessor">#error "Only support strides 1, 2 and 3"</span></div><div class="line"><a name="l00159"></a><span class="lineno"> 159</span> <span class="preprocessor">#endif </span><span class="comment">/* STRIDE_X == 3 */</span><span class="preprocessor"></span></div><div class="line"><a name="l00160"></a><span class="lineno"> 160</span> </div><div class="line"><a name="l00167"></a><span class="lineno"> 167</span> <span class="keyword">inline</span> <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8) extract_input_stride1(__global const <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *input_pixel)</div><div class="line"><a name="l00168"></a><span class="lineno"> 168</span> {</div><div class="line"><a name="l00169"></a><span class="lineno"> 169</span>  <span class="keywordflow">return</span> vload8(0, input_pixel);</div><div class="line"><a name="l00170"></a><span class="lineno"> 170</span> }</div><div class="line"><a name="l00171"></a><span class="lineno"> 171</span> </div><div class="line"><a name="l00178"></a><span class="lineno"> 178</span> <span class="keyword">inline</span> <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8) extract_input_stride2(__global const <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *input_pixel)</div><div class="line"><a name="l00179"></a><span class="lineno"> 179</span> {</div><div class="line"><a name="l00180"></a><span class="lineno"> 180</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 16)</div><div class="line"><a name="l00181"></a><span class="lineno"> 181</span>  temp = vload16(0, input_pixel);</div><div class="line"><a name="l00182"></a><span class="lineno"> 182</span>  return temp.s02468ace;</div><div class="line"><a name="l00183"></a><span class="lineno"> 183</span> }</div><div class="line"><a name="l00184"></a><span class="lineno"> 184</span> </div><div class="line"><a name="l00191"></a><span class="lineno"> 191</span> inline <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8) extract_input_stride3_32(__global const <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *input_pixel)</div><div class="line"><a name="l00192"></a><span class="lineno"> 192</span> {</div><div class="line"><a name="l00193"></a><span class="lineno"> 193</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 4)</div><div class="line"><a name="l00194"></a><span class="lineno"> 194</span>  temp1 = vload4(0, input_pixel);</div><div class="line"><a name="l00195"></a><span class="lineno"> 195</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 4)</div><div class="line"><a name="l00196"></a><span class="lineno"> 196</span>  temp2 = vload4(0, input_pixel + 6);</div><div class="line"><a name="l00197"></a><span class="lineno"> 197</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 4)</div><div class="line"><a name="l00198"></a><span class="lineno"> 198</span>  temp3 = vload4(0, input_pixel + 12);</div><div class="line"><a name="l00199"></a><span class="lineno"> 199</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 4)</div><div class="line"><a name="l00200"></a><span class="lineno"> 200</span>  temp4 = vload4(0, input_pixel + 18);</div><div class="line"><a name="l00201"></a><span class="lineno"> 201</span>  return (<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8))(temp1.s03, temp2.s03, temp3.s03, temp4.s03);</div><div class="line"><a name="l00202"></a><span class="lineno"> 202</span> }</div><div class="line"><a name="l00203"></a><span class="lineno"> 203</span> </div><div class="line"><a name="l00210"></a><span class="lineno"> 210</span> inline <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8) extract_input_stride3_16(__global const <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *input_pixel)</div><div class="line"><a name="l00211"></a><span class="lineno"> 211</span> {</div><div class="line"><a name="l00212"></a><span class="lineno"> 212</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8)</div><div class="line"><a name="l00213"></a><span class="lineno"> 213</span>  temp1 = vload8(0, input_pixel);</div><div class="line"><a name="l00214"></a><span class="lineno"> 214</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8)</div><div class="line"><a name="l00215"></a><span class="lineno"> 215</span>  temp2 = vload8(0, input_pixel + 8);</div><div class="line"><a name="l00216"></a><span class="lineno"> 216</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8)</div><div class="line"><a name="l00217"></a><span class="lineno"> 217</span>  temp3 = vload8(0, input_pixel + 16);</div><div class="line"><a name="l00218"></a><span class="lineno"> 218</span>  return (<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8))(temp1.s036, temp2.s147, temp3.s25);</div><div class="line"><a name="l00219"></a><span class="lineno"> 219</span> }</div><div class="line"><a name="l00220"></a><span class="lineno"> 220</span> </div><div class="line"><a name="l00227"></a><span class="lineno"> 227</span> inline <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8) extract_input_stride3_8(__global const <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *input_pixel)</div><div class="line"><a name="l00228"></a><span class="lineno"> 228</span> {</div><div class="line"><a name="l00229"></a><span class="lineno"> 229</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 16)</div><div class="line"><a name="l00230"></a><span class="lineno"> 230</span>  temp1 = vload16(0, input_pixel);</div><div class="line"><a name="l00231"></a><span class="lineno"> 231</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 16)</div><div class="line"><a name="l00232"></a><span class="lineno"> 232</span>  temp2 = vload16(0, input_pixel + 12);</div><div class="line"><a name="l00233"></a><span class="lineno"> 233</span>  return (<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8))(temp1.s0369, temp2.s0369);</div><div class="line"><a name="l00234"></a><span class="lineno"> 234</span> }</div><div class="line"><a name="l00235"></a><span class="lineno"> 235</span> </div><div class="line"><a name="l00274"></a><span class="lineno"> 274</span> __kernel <span class="keywordtype">void</span> direct_convolution1x1(</div><div class="line"><a name="l00275"></a><span class="lineno"> 275</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>),</div><div class="line"><a name="l00276"></a><span class="lineno"> 276</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>),</div><div class="line"><a name="l00277"></a><span class="lineno"> 277</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>),</div><div class="line"><a name="l00278"></a><span class="lineno"> 278</span> <span class="preprocessor">#ifdef HAS_BIAS</span></div><div class="line"><a name="l00279"></a><span class="lineno"> 279</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(biases),</div><div class="line"><a name="l00280"></a><span class="lineno"> 280</span> <span class="preprocessor">#endif </span><span class="comment">/* defined(HAS_BIAS) */</span><span class="preprocessor"></span></div><div class="line"><a name="l00281"></a><span class="lineno"> 281</span>  <span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> weights_stride_w)</div><div class="line"><a name="l00282"></a><span class="lineno"> 282</span> {</div><div class="line"><a name="l00283"></a><span class="lineno"> 283</span>  <a class="code" href="struct_image.xhtml">Image</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a> = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#aebe814363556c244be043b13e7969197">CONVERT_TO_IMAGE_STRUCT</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>);</div><div class="line"><a name="l00284"></a><span class="lineno"> 284</span>  <a class="code" href="struct_tensor3_d.xhtml">Tensor3D</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a> = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a79e8e562daa6599317d2d1cd86ef1bf2">CONVERT_TO_TENSOR3D_STRUCT_NO_STEP</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>);</div><div class="line"><a name="l00285"></a><span class="lineno"> 285</span>  <a class="code" href="struct_tensor3_d.xhtml">Tensor3D</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a> = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a31c8c760f08fb1a331b16b7c204321dc">CONVERT_TO_TENSOR3D_STRUCT</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>);</div><div class="line"><a name="l00286"></a><span class="lineno"> 286</span> </div><div class="line"><a name="l00287"></a><span class="lineno"> 287</span> <span class="preprocessor">#ifdef HAS_BIAS</span></div><div class="line"><a name="l00288"></a><span class="lineno"> 288</span>  <a class="code" href="struct_vector.xhtml">Vector</a> biases = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a64d779f80eeb923e0ab2313433f7b40b">CONVERT_TO_VECTOR_STRUCT_NO_STEP</a>(biases);</div><div class="line"><a name="l00289"></a><span class="lineno"> 289</span> <span class="preprocessor">#endif </span><span class="comment">/* defined(HAS_BIAS) */</span><span class="preprocessor"></span></div><div class="line"><a name="l00290"></a><span class="lineno"> 290</span> </div><div class="line"><a name="l00291"></a><span class="lineno"> 291</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(DATA_TYPE_PROMOTED, 8)</div><div class="line"><a name="l00292"></a><span class="lineno"> 292</span>  values = 0;</div><div class="line"><a name="l00293"></a><span class="lineno"> 293</span> </div><div class="line"><a name="l00294"></a><span class="lineno"> 294</span>  const uint z_index = get_global_id(2);</div><div class="line"><a name="l00295"></a><span class="lineno"> 295</span> </div><div class="line"><a name="l00296"></a><span class="lineno"> 296</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>.ptr += z_index * weights_stride_w;</div><div class="line"><a name="l00297"></a><span class="lineno"> 297</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a0886942393a3ba0dfefaa7516b159784">for</a>(volatile <span class="keywordtype">int</span> d = 0; d < WEIGHTS_DEPTH; ++d)</div><div class="line"><a name="l00298"></a><span class="lineno"> 298</span>  {</div><div class="line"><a name="l00299"></a><span class="lineno"> 299</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> weight = *(__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>.ptr;</div><div class="line"><a name="l00300"></a><span class="lineno"> 300</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8)</div><div class="line"><a name="l00301"></a><span class="lineno"> 301</span>  input_pixel = INPUT_PIXEL(DATA_SIZE)((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>.ptr);</div><div class="line"><a name="l00302"></a><span class="lineno"> 302</span>  values = <a class="code" href="direct__convolution1x1_8cl.xhtml#aebbeb1f22eca3a3f4c3e019e8f419f39">ADD_OP</a>(values, <a class="code" href="direct__convolution1x1_8cl.xhtml#ad3cc858846806e6b1d3694b9d0a2e6da">MUL_OP</a>((<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8))weight, input_pixel));</div><div class="line"><a name="l00303"></a><span class="lineno"> 303</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>.ptr += src_stride_z;</div><div class="line"><a name="l00304"></a><span class="lineno"> 304</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>.ptr += weights_stride_z;</div><div class="line"><a name="l00305"></a><span class="lineno"> 305</span>  }</div><div class="line"><a name="l00306"></a><span class="lineno"> 306</span> </div><div class="line"><a name="l00307"></a><span class="lineno"> 307</span> <span class="preprocessor">#ifdef HAS_BIAS</span></div><div class="line"><a name="l00308"></a><span class="lineno"> 308</span>  values = <a class="code" href="direct__convolution1x1_8cl.xhtml#aebbeb1f22eca3a3f4c3e019e8f419f39">ADD_OP</a>(values, (<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(DATA_TYPE_PROMOTED, 8)) * ((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a7e4940407322d6f0ccb8b6b86b856019">vector_offset</a>(&biases, z_index))));</div><div class="line"><a name="l00309"></a><span class="lineno"> 309</span> <span class="preprocessor">#endif </span><span class="comment">/* defined(HAS_BIAS) */</span><span class="preprocessor"></span></div><div class="line"><a name="l00310"></a><span class="lineno"> 310</span> </div><div class="line"><a name="l00311"></a><span class="lineno"> 311</span>  vstore8(<a class="code" href="direct__convolution1x1_8cl.xhtml#a1f15728672380ade7a238f5e783d54d2">CONVERT_SAT</a>(values, <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8)), 0, (__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr);</div><div class="line"><a name="l00312"></a><span class="lineno"> 312</span> }</div><div class="line"><a name="l00313"></a><span class="lineno"> 313</span> <span class="preprocessor">#endif // defined(DATA_TYPE) && defined(DATA_SIZE) && defined(STRIDE_X) && defined(WEIGHTS_DEPTH)</span></div><div class="line"><a name="l00314"></a><span class="lineno"> 314</span> </div><div class="line"><a name="l00315"></a><span class="lineno"> 315</span> <span class="preprocessor">#if defined(WEIGHTS_DEPTH)</span></div><div class="line"><a name="l00316"></a><span class="lineno"> 316</span> </div><div class="line"><a name="l00317"></a><span class="lineno"> 317</span> <span class="preprocessor">#define CONVOLUTION1x1_BIFROST(acc, src, weight_value) \</span></div><div class="line"><a name="l00318"></a><span class="lineno"> 318</span> <span class="preprocessor"> ({ \</span></div><div class="line"><a name="l00319"></a><span class="lineno"> 319</span> <span class="preprocessor"> acc.s0 = mad(src.s0, weight_value, acc.s0); \</span></div><div class="line"><a name="l00320"></a><span class="lineno"> 320</span> <span class="preprocessor"> acc.s1 = mad(src.s1, weight_value, acc.s1); \</span></div><div class="line"><a name="l00321"></a><span class="lineno"> 321</span> <span class="preprocessor"> acc.s2 = mad(src.s2, weight_value, acc.s2); \</span></div><div class="line"><a name="l00322"></a><span class="lineno"> 322</span> <span class="preprocessor"> acc.s3 = mad(src.s3, weight_value, acc.s3); \</span></div><div class="line"><a name="l00323"></a><span class="lineno"> 323</span> <span class="preprocessor"> })</span></div><div class="line"><a name="l00324"></a><span class="lineno"> 324</span> </div><div class="line"><a name="l00361"></a><span class="lineno"> 361</span> __kernel <span class="keywordtype">void</span> direct_convolution1x1_f32_bifrost(</div><div class="line"><a name="l00362"></a><span class="lineno"> 362</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>),</div><div class="line"><a name="l00363"></a><span class="lineno"> 363</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>),</div><div class="line"><a name="l00364"></a><span class="lineno"> 364</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>),</div><div class="line"><a name="l00365"></a><span class="lineno"> 365</span> #ifdef HAS_BIAS</div><div class="line"><a name="l00366"></a><span class="lineno"> 366</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(biases),</div><div class="line"><a name="l00367"></a><span class="lineno"> 367</span> #endif <span class="comment">/* defined(HAS_BIAS) */</span></div><div class="line"><a name="l00368"></a><span class="lineno"> 368</span>  <span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> weights_stride_w)</div><div class="line"><a name="l00369"></a><span class="lineno"> 369</span> {</div><div class="line"><a name="l00370"></a><span class="lineno"> 370</span>  <span class="comment">// Get the kernel index</span></div><div class="line"><a name="l00371"></a><span class="lineno"> 371</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> kernel_index = get_global_id(2);</div><div class="line"><a name="l00372"></a><span class="lineno"> 372</span> </div><div class="line"><a name="l00373"></a><span class="lineno"> 373</span>  <a class="code" href="struct_image.xhtml">Image</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a> = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#aebe814363556c244be043b13e7969197">CONVERT_TO_IMAGE_STRUCT</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>);</div><div class="line"><a name="l00374"></a><span class="lineno"> 374</span>  <a class="code" href="struct_tensor3_d.xhtml">Tensor3D</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a> = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a31c8c760f08fb1a331b16b7c204321dc">CONVERT_TO_TENSOR3D_STRUCT</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>);</div><div class="line"><a name="l00375"></a><span class="lineno"> 375</span> </div><div class="line"><a name="l00376"></a><span class="lineno"> 376</span>  float4 acc0 = 0.0f;</div><div class="line"><a name="l00377"></a><span class="lineno"> 377</span>  float4 acc1 = 0.0f;</div><div class="line"><a name="l00378"></a><span class="lineno"> 378</span>  float4 acc2 = 0.0f;</div><div class="line"><a name="l00379"></a><span class="lineno"> 379</span>  float4 acc3 = 0.0f;</div><div class="line"><a name="l00380"></a><span class="lineno"> 380</span> </div><div class="line"><a name="l00381"></a><span class="lineno"> 381</span>  __global uchar *weights_addr = (__global uchar *)(weights_ptr + weights_offset_first_element_in_bytes + kernel_index * weights_stride_w);</div><div class="line"><a name="l00382"></a><span class="lineno"> 382</span>  __global uchar *src_addr = (__global uchar *)<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a009469e4d9b8fce3b6d5e97d2077827d">offset</a>(&<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>, 0, 0);</div><div class="line"><a name="l00383"></a><span class="lineno"> 383</span> </div><div class="line"><a name="l00384"></a><span class="lineno"> 384</span>  <span class="keywordflow">for</span>(ushort d = 0; d < (ushort)WEIGHTS_DEPTH; ++d)</div><div class="line"><a name="l00385"></a><span class="lineno"> 385</span>  {</div><div class="line"><a name="l00386"></a><span class="lineno"> 386</span>  <span class="comment">// Load the weights</span></div><div class="line"><a name="l00387"></a><span class="lineno"> 387</span>  <span class="keywordtype">float</span> weight = *((__global <span class="keywordtype">float</span> *)weights_addr);</div><div class="line"><a name="l00388"></a><span class="lineno"> 388</span> </div><div class="line"><a name="l00389"></a><span class="lineno"> 389</span>  <span class="comment">// Load values from row0 of input tensor</span></div><div class="line"><a name="l00390"></a><span class="lineno"> 390</span>  float4 src0 = vload4(0, (__global <span class="keywordtype">float</span> *)(src_addr + 0 * src_stride_y));</div><div class="line"><a name="l00391"></a><span class="lineno"> 391</span>  float4 src1 = vload4(0, (__global <span class="keywordtype">float</span> *)(src_addr + 1 * src_stride_y));</div><div class="line"><a name="l00392"></a><span class="lineno"> 392</span>  float4 src2 = vload4(0, (__global <span class="keywordtype">float</span> *)(src_addr + 2 * src_stride_y));</div><div class="line"><a name="l00393"></a><span class="lineno"> 393</span>  float4 src3 = vload4(0, (__global <span class="keywordtype">float</span> *)(src_addr + 3 * src_stride_y));</div><div class="line"><a name="l00394"></a><span class="lineno"> 394</span> </div><div class="line"><a name="l00395"></a><span class="lineno"> 395</span>  CONVOLUTION1x1_BIFROST(acc0, src0, weight);</div><div class="line"><a name="l00396"></a><span class="lineno"> 396</span>  CONVOLUTION1x1_BIFROST(acc1, src1, weight);</div><div class="line"><a name="l00397"></a><span class="lineno"> 397</span>  CONVOLUTION1x1_BIFROST(acc2, src2, weight);</div><div class="line"><a name="l00398"></a><span class="lineno"> 398</span>  CONVOLUTION1x1_BIFROST(acc3, src3, weight);</div><div class="line"><a name="l00399"></a><span class="lineno"> 399</span> </div><div class="line"><a name="l00400"></a><span class="lineno"> 400</span>  src_addr += src_stride_z;</div><div class="line"><a name="l00401"></a><span class="lineno"> 401</span>  weights_addr += weights_stride_z;</div><div class="line"><a name="l00402"></a><span class="lineno"> 402</span>  }</div><div class="line"><a name="l00403"></a><span class="lineno"> 403</span> </div><div class="line"><a name="l00404"></a><span class="lineno"> 404</span> <span class="preprocessor">#ifdef HAS_BIAS</span></div><div class="line"><a name="l00405"></a><span class="lineno"> 405</span>  <a class="code" href="struct_vector.xhtml">Vector</a> biases = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a64d779f80eeb923e0ab2313433f7b40b">CONVERT_TO_VECTOR_STRUCT_NO_STEP</a>(biases);</div><div class="line"><a name="l00406"></a><span class="lineno"> 406</span> </div><div class="line"><a name="l00407"></a><span class="lineno"> 407</span>  <span class="keywordtype">float</span> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a> = (float) * ((__global <span class="keywordtype">float</span> *)(<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a7e4940407322d6f0ccb8b6b86b856019">vector_offset</a>(&biases, kernel_index)));</div><div class="line"><a name="l00408"></a><span class="lineno"> 408</span> </div><div class="line"><a name="l00409"></a><span class="lineno"> 409</span>  acc0.s0 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00410"></a><span class="lineno"> 410</span>  acc0.s1 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00411"></a><span class="lineno"> 411</span>  acc0.s2 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00412"></a><span class="lineno"> 412</span>  acc0.s3 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00413"></a><span class="lineno"> 413</span>  acc1.s0 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00414"></a><span class="lineno"> 414</span>  acc1.s1 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00415"></a><span class="lineno"> 415</span>  acc1.s2 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00416"></a><span class="lineno"> 416</span>  acc1.s3 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00417"></a><span class="lineno"> 417</span>  acc2.s0 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00418"></a><span class="lineno"> 418</span>  acc2.s1 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00419"></a><span class="lineno"> 419</span>  acc2.s2 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00420"></a><span class="lineno"> 420</span>  acc2.s3 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00421"></a><span class="lineno"> 421</span>  acc3.s0 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00422"></a><span class="lineno"> 422</span>  acc3.s1 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00423"></a><span class="lineno"> 423</span>  acc3.s2 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00424"></a><span class="lineno"> 424</span>  acc3.s3 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00425"></a><span class="lineno"> 425</span> <span class="preprocessor">#endif </span><span class="comment">/* defined(HAS_BIAS) */</span><span class="preprocessor"></span></div><div class="line"><a name="l00426"></a><span class="lineno"> 426</span> </div><div class="line"><a name="l00427"></a><span class="lineno"> 427</span>  vstore4(acc0, 0, (__global <span class="keywordtype">float</span> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 0 * dst_stride_y));</div><div class="line"><a name="l00428"></a><span class="lineno"> 428</span>  vstore4(acc1, 0, (__global <span class="keywordtype">float</span> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 1 * dst_stride_y));</div><div class="line"><a name="l00429"></a><span class="lineno"> 429</span>  vstore4(acc2, 0, (__global <span class="keywordtype">float</span> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 2 * dst_stride_y));</div><div class="line"><a name="l00430"></a><span class="lineno"> 430</span>  vstore4(acc3, 0, (__global <span class="keywordtype">float</span> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 3 * dst_stride_y));</div><div class="line"><a name="l00431"></a><span class="lineno"> 431</span> }</div><div class="line"><a name="l00432"></a><span class="lineno"> 432</span> <span class="preprocessor">#endif // defined(WEIGHTS_DEPTH)</span></div><div class="ttc" id="struct_vector_xhtml"><div class="ttname"><a href="struct_vector.xhtml">Vector</a></div><div class="ttdoc">Structure to hold Vector information.</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00159">helpers.h:159</a></div></div>
-<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a009469e4d9b8fce3b6d5e97d2077827d"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a009469e4d9b8fce3b6d5e97d2077827d">offset</a></div><div class="ttdeci">__global uchar * offset(const Image *img, int x, int y)</div><div class="ttdoc">Get the pointer position of a Image.</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00328">helpers.h:328</a></div></div>
-<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_aebe814363556c244be043b13e7969197"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#aebe814363556c244be043b13e7969197">CONVERT_TO_IMAGE_STRUCT</a></div><div class="ttdeci">#define CONVERT_TO_IMAGE_STRUCT(name)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00129">helpers.h:129</a></div></div>
+<a href="direct__convolution1x1_8cl.xhtml">Go to the documentation of this file.</a><div class="fragment"><div class="line"><a name="l00001"></a><span class="lineno"> 1</span> <span class="comment">/*</span></div><div class="line"><a name="l00002"></a><span class="lineno"> 2</span> <span class="comment"> * Copyright (c) 2016-2018 ARM Limited.</span></div><div class="line"><a name="l00003"></a><span class="lineno"> 3</span> <span class="comment"> *</span></div><div class="line"><a name="l00004"></a><span class="lineno"> 4</span> <span class="comment"> * SPDX-License-Identifier: MIT</span></div><div class="line"><a name="l00005"></a><span class="lineno"> 5</span> <span class="comment"> *</span></div><div class="line"><a name="l00006"></a><span class="lineno"> 6</span> <span class="comment"> * Permission is hereby granted, free of charge, to any person obtaining a copy</span></div><div class="line"><a name="l00007"></a><span class="lineno"> 7</span> <span class="comment"> * of this software and associated documentation files (the "Software"), to</span></div><div class="line"><a name="l00008"></a><span class="lineno"> 8</span> <span class="comment"> * deal in the Software without restriction, including without limitation the</span></div><div class="line"><a name="l00009"></a><span class="lineno"> 9</span> <span class="comment"> * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or</span></div><div class="line"><a name="l00010"></a><span class="lineno"> 10</span> <span class="comment"> * sell copies of the Software, and to permit persons to whom the Software is</span></div><div class="line"><a name="l00011"></a><span class="lineno"> 11</span> <span class="comment"> * furnished to do so, subject to the following conditions:</span></div><div class="line"><a name="l00012"></a><span class="lineno"> 12</span> <span class="comment"> *</span></div><div class="line"><a name="l00013"></a><span class="lineno"> 13</span> <span class="comment"> * The above copyright notice and this permission notice shall be included in all</span></div><div class="line"><a name="l00014"></a><span class="lineno"> 14</span> <span class="comment"> * copies or substantial portions of the Software.</span></div><div class="line"><a name="l00015"></a><span class="lineno"> 15</span> <span class="comment"> *</span></div><div class="line"><a name="l00016"></a><span class="lineno"> 16</span> <span class="comment"> * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR</span></div><div class="line"><a name="l00017"></a><span class="lineno"> 17</span> <span class="comment"> * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,</span></div><div class="line"><a name="l00018"></a><span class="lineno"> 18</span> <span class="comment"> * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE</span></div><div class="line"><a name="l00019"></a><span class="lineno"> 19</span> <span class="comment"> * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER</span></div><div class="line"><a name="l00020"></a><span class="lineno"> 20</span> <span class="comment"> * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,</span></div><div class="line"><a name="l00021"></a><span class="lineno"> 21</span> <span class="comment"> * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE</span></div><div class="line"><a name="l00022"></a><span class="lineno"> 22</span> <span class="comment"> * SOFTWARE.</span></div><div class="line"><a name="l00023"></a><span class="lineno"> 23</span> <span class="comment"> */</span></div><div class="line"><a name="l00024"></a><span class="lineno"> 24</span> <span class="preprocessor">#include "<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml">helpers.h</a>"</span></div><div class="line"><a name="l00025"></a><span class="lineno"> 25</span> </div><div class="line"><a name="l00026"></a><span class="lineno"> 26</span> <span class="preprocessor">#undef CONVERT_SAT</span></div><div class="line"><a name="l00027"></a><span class="lineno"> 27</span> </div><div class="line"><a name="l00028"></a><span class="lineno"><a class="line" href="direct__convolution1x1_8cl.xhtml#aebbeb1f22eca3a3f4c3e019e8f419f39"> 28</a></span> <span class="preprocessor">#define ADD_OP(a, b) ((a) + (b))</span></div><div class="line"><a name="l00029"></a><span class="lineno"><a class="line" href="direct__convolution1x1_8cl.xhtml#ad3cc858846806e6b1d3694b9d0a2e6da"> 29</a></span> <span class="preprocessor">#define MUL_OP(a, b) ((a) * (b))</span></div><div class="line"><a name="l00030"></a><span class="lineno"><a class="line" href="direct__convolution1x1_8cl.xhtml#a1f15728672380ade7a238f5e783d54d2"> 30</a></span> <span class="preprocessor">#define CONVERT_SAT(a, b) ((a))</span></div><div class="line"><a name="l00031"></a><span class="lineno"> 31</span> </div><div class="line"><a name="l00032"></a><span class="lineno"> 32</span> <span class="preprocessor">#if defined(DATA_TYPE) && defined(DATA_SIZE) && defined(STRIDE_X) && defined(WEIGHTS_DEPTH)</span></div><div class="line"><a name="l00033"></a><span class="lineno"> 33</span> </div><div class="line"><a name="l00034"></a><span class="lineno"> 34</span> <span class="preprocessor">#if defined(DATA_LAYOUT_NHWC)</span></div><div class="line"><a name="l00035"></a><span class="lineno"> 35</span> </div><div class="line"><a name="l00036"></a><span class="lineno"> 36</span> <span class="preprocessor">#define PTR_TO_VALUE(PTR, DATA_TYPE) *((__global DATA_TYPE *)(PTR))</span></div><div class="line"><a name="l00037"></a><span class="lineno"> 37</span> <span class="comment"></span></div><div class="line"><a name="l00038"></a><span class="lineno"> 38</span> <span class="comment">/** This kernel performs a direct convolution to convolve the low three dimensions of a tensor with data layout NHWC</span></div><div class="line"><a name="l00039"></a><span class="lineno"> 39</span> <span class="comment"> *</span></div><div class="line"><a name="l00040"></a><span class="lineno"> 40</span> <span class="comment"> * @note The data type must be passed at compile time using -DDATA_TYPE: e.g. -DDATA_TYPE=float</span></div><div class="line"><a name="l00041"></a><span class="lineno"> 41</span> <span class="comment"> * @note The data size must be passed at compile time using -DDATA_SIZE e.g. -DDATA_SIZE=32</span></div><div class="line"><a name="l00042"></a><span class="lineno"> 42</span> <span class="comment"> * @note The convolution stride x must be passed at compile time using -DSTRIDE_X e.g. -DSTRIDE_X=1</span></div><div class="line"><a name="l00043"></a><span class="lineno"> 43</span> <span class="comment"> * @note The third dimensions of the weights tensors must be passed at compile time using -DWEIGHTS_DEPTH</span></div><div class="line"><a name="l00044"></a><span class="lineno"> 44</span> <span class="comment"> * @note In case biases will be added to the convolution -DHAS_BIAS has to be passed to append the final matrix with 1 in each row.</span></div><div class="line"><a name="l00045"></a><span class="lineno"> 45</span> <span class="comment"> *</span></div><div class="line"><a name="l00046"></a><span class="lineno"> 46</span> <span class="comment"> * @param[in] src_ptr Pointer to the source tensor. Supported data types: F16/F32</span></div><div class="line"><a name="l00047"></a><span class="lineno"> 47</span> <span class="comment"> * @param[in] src_stride_x Stride of the source tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00048"></a><span class="lineno"> 48</span> <span class="comment"> * @param[in] src_step_x src_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00049"></a><span class="lineno"> 49</span> <span class="comment"> * @param[in] src_stride_y Stride of the source tensor in Y dimension (in bytes)</span></div><div class="line"><a name="l00050"></a><span class="lineno"> 50</span> <span class="comment"> * @param[in] src_step_y src_stride_y * number of elements along Y processed per workitem(in bytes)</span></div><div class="line"><a name="l00051"></a><span class="lineno"> 51</span> <span class="comment"> * @param[in] src_stride_z Stride of the source tensor in Z dimension (in bytes)</span></div><div class="line"><a name="l00052"></a><span class="lineno"> 52</span> <span class="comment"> * @param[in] src_step_z src_stride_z * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00053"></a><span class="lineno"> 53</span> <span class="comment"> * @param[in] src_offset_first_element_in_bytes The offset of the first element in the source tensor</span></div><div class="line"><a name="l00054"></a><span class="lineno"> 54</span> <span class="comment"> * @param[out] dst_ptr Pointer to the destination tensor. Supported data types: same as @p src_ptr</span></div><div class="line"><a name="l00055"></a><span class="lineno"> 55</span> <span class="comment"> * @param[in] dst_stride_x Stride of the destination tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00056"></a><span class="lineno"> 56</span> <span class="comment"> * @param[in] dst_step_x dst_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00057"></a><span class="lineno"> 57</span> <span class="comment"> * @param[in] dst_stride_y Stride of the destination tensor in Y dimension (in bytes)</span></div><div class="line"><a name="l00058"></a><span class="lineno"> 58</span> <span class="comment"> * @param[in] dst_step_y dst_stride_y * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00059"></a><span class="lineno"> 59</span> <span class="comment"> * @param[in] dst_stride_z Stride of the destination tensor in Z dimension (in bytes)</span></div><div class="line"><a name="l00060"></a><span class="lineno"> 60</span> <span class="comment"> * @param[in] dst_step_z dst_stride_z * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00061"></a><span class="lineno"> 61</span> <span class="comment"> * @param[in] dst_offset_first_element_in_bytes The offset of the first element in the destination tensor</span></div><div class="line"><a name="l00062"></a><span class="lineno"> 62</span> <span class="comment"> * @param[in] weights_ptr Pointer to the weights tensor. Supported data types: same as @p src_ptr</span></div><div class="line"><a name="l00063"></a><span class="lineno"> 63</span> <span class="comment"> * @param[in] weights_stride_x Stride of the weights tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00064"></a><span class="lineno"> 64</span> <span class="comment"> * @param[in] weights_step_x weights_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00065"></a><span class="lineno"> 65</span> <span class="comment"> * @param[in] weights_stride_y Stride of the weights tensor in Y dimension (in bytes)</span></div><div class="line"><a name="l00066"></a><span class="lineno"> 66</span> <span class="comment"> * @param[in] weights_step_y weights_stride_y * number of elements along y processed per workitem(in bytes)</span></div><div class="line"><a name="l00067"></a><span class="lineno"> 67</span> <span class="comment"> * @param[in] weights_stride_z Stride of the weights tensor in Z dimension (in bytes)</span></div><div class="line"><a name="l00068"></a><span class="lineno"> 68</span> <span class="comment"> * @param[in] weights_step_z weights_stride_z * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00069"></a><span class="lineno"> 69</span> <span class="comment"> * @param[in] weights_offset_first_element_in_bytes The offset of the first element in the weights tensor</span></div><div class="line"><a name="l00070"></a><span class="lineno"> 70</span> <span class="comment"> * @param[in] biases_ptr Pointer to the biases tensor. Same as @p src_ptr</span></div><div class="line"><a name="l00071"></a><span class="lineno"> 71</span> <span class="comment"> * @param[in] biases_stride_x Stride of the biases tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00072"></a><span class="lineno"> 72</span> <span class="comment"> * @param[in] biases_step_x biases_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00073"></a><span class="lineno"> 73</span> <span class="comment"> * @param[in] biases_offset_first_element_in_bytes The offset of the first element in the biases tensor</span></div><div class="line"><a name="l00074"></a><span class="lineno"> 74</span> <span class="comment"> * @param[in] weights_stride_w Stride of the weights tensor in the 4th dimension</span></div><div class="line"><a name="l00075"></a><span class="lineno"> 75</span> <span class="comment"> */</span></div><div class="line"><a name="l00076"></a><span class="lineno"> 76</span> __kernel <span class="keywordtype">void</span> direct_convolution1x1_nhwc(</div><div class="line"><a name="l00077"></a><span class="lineno"> 77</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>),</div><div class="line"><a name="l00078"></a><span class="lineno"> 78</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>),</div><div class="line"><a name="l00079"></a><span class="lineno"> 79</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>),</div><div class="line"><a name="l00080"></a><span class="lineno"> 80</span> #ifdef HAS_BIAS</div><div class="line"><a name="l00081"></a><span class="lineno"> 81</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(biases),</div><div class="line"><a name="l00082"></a><span class="lineno"> 82</span> #endif <span class="comment">/* defined(HAS_BIAS) */</span></div><div class="line"><a name="l00083"></a><span class="lineno"> 83</span>  <span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> weights_stride_w)</div><div class="line"><a name="l00084"></a><span class="lineno"> 84</span> {</div><div class="line"><a name="l00085"></a><span class="lineno"> 85</span>  <a class="code" href="struct_image.xhtml">Image</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a> = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#aebe814363556c244be043b13e7969197">CONVERT_TO_IMAGE_STRUCT</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>);</div><div class="line"><a name="l00086"></a><span class="lineno"> 86</span>  <a class="code" href="struct_tensor3_d.xhtml">Tensor3D</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a> = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a79e8e562daa6599317d2d1cd86ef1bf2">CONVERT_TO_TENSOR3D_STRUCT_NO_STEP</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>);</div><div class="line"><a name="l00087"></a><span class="lineno"> 87</span>  <a class="code" href="struct_tensor3_d.xhtml">Tensor3D</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a> = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a31c8c760f08fb1a331b16b7c204321dc">CONVERT_TO_TENSOR3D_STRUCT</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>);</div><div class="line"><a name="l00088"></a><span class="lineno"> 88</span> </div><div class="line"><a name="l00089"></a><span class="lineno"> 89</span> <span class="preprocessor">#ifdef HAS_BIAS</span></div><div class="line"><a name="l00090"></a><span class="lineno"> 90</span>  <a class="code" href="struct_vector.xhtml">Vector</a> biases = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a64d779f80eeb923e0ab2313433f7b40b">CONVERT_TO_VECTOR_STRUCT_NO_STEP</a>(biases);</div><div class="line"><a name="l00091"></a><span class="lineno"> 91</span> <span class="preprocessor">#endif </span><span class="comment">/* defined(HAS_BIAS) */</span><span class="preprocessor"></span></div><div class="line"><a name="l00092"></a><span class="lineno"> 92</span> </div><div class="line"><a name="l00093"></a><span class="lineno"> 93</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(DATA_TYPE_PROMOTED, 8)</div><div class="line"><a name="l00094"></a><span class="lineno"> 94</span>  values = 0;</div><div class="line"><a name="l00095"></a><span class="lineno"> 95</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> id0 = get_global_id(0);</div><div class="line"><a name="l00096"></a><span class="lineno"> 96</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> id1 = get_global_id(1);</div><div class="line"><a name="l00097"></a><span class="lineno"> 97</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> id2 = get_global_id(2);</div><div class="line"><a name="l00098"></a><span class="lineno"> 98</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>.ptr += id0 * weights_stride_w;</div><div class="line"><a name="l00099"></a><span class="lineno"> 99</span>  __global uchar *src_addr = (__global uchar *)<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a009469e4d9b8fce3b6d5e97d2077827d">offset</a>(&<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>, 0, 0) - src_stride_x * id0 + id2 * STRIDE_Y * (int)src_stride_z;</div><div class="line"><a name="l00100"></a><span class="lineno"> 100</span> </div><div class="line"><a name="l00101"></a><span class="lineno"> 101</span>  <span class="keywordflow">for</span>(<span class="keyword">volatile</span> <span class="keywordtype">int</span> d = 0; d < WEIGHTS_DEPTH; ++d)</div><div class="line"><a name="l00102"></a><span class="lineno"> 102</span>  {</div><div class="line"><a name="l00103"></a><span class="lineno"> 103</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> weight = *(__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>.ptr;</div><div class="line"><a name="l00104"></a><span class="lineno"> 104</span> #<span class="keywordflow">if</span> STRIDE_X == 1</div><div class="line"><a name="l00105"></a><span class="lineno"> 105</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8)</div><div class="line"><a name="l00106"></a><span class="lineno"> 106</span>  col0 = (<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8))(</div><div class="line"><a name="l00107"></a><span class="lineno"> 107</span>  PTR_TO_VALUE(src_addr + 0 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00108"></a><span class="lineno"> 108</span>  PTR_TO_VALUE(src_addr + 1 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00109"></a><span class="lineno"> 109</span>  PTR_TO_VALUE(src_addr + 2 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00110"></a><span class="lineno"> 110</span>  PTR_TO_VALUE(src_addr + 3 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00111"></a><span class="lineno"> 111</span>  PTR_TO_VALUE(src_addr + 4 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00112"></a><span class="lineno"> 112</span>  PTR_TO_VALUE(src_addr + 5 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00113"></a><span class="lineno"> 113</span>  PTR_TO_VALUE(src_addr + 6 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00114"></a><span class="lineno"> 114</span>  PTR_TO_VALUE(src_addr + 7 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>));</div><div class="line"><a name="l00115"></a><span class="lineno"> 115</span> <span class="preprocessor">#elif STRIDE_X == 2 </span><span class="comment">/* STRIDE_X == 1 */</span><span class="preprocessor"></span></div><div class="line"><a name="l00116"></a><span class="lineno"> 116</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8)</div><div class="line"><a name="l00117"></a><span class="lineno"> 117</span>  col0 = (<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8))(</div><div class="line"><a name="l00118"></a><span class="lineno"> 118</span>  PTR_TO_VALUE(src_addr + 0 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00119"></a><span class="lineno"> 119</span>  PTR_TO_VALUE(src_addr + 2 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00120"></a><span class="lineno"> 120</span>  PTR_TO_VALUE(src_addr + 4 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00121"></a><span class="lineno"> 121</span>  PTR_TO_VALUE(src_addr + 6 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00122"></a><span class="lineno"> 122</span>  PTR_TO_VALUE(src_addr + 8 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00123"></a><span class="lineno"> 123</span>  PTR_TO_VALUE(src_addr + 10 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00124"></a><span class="lineno"> 124</span>  PTR_TO_VALUE(src_addr + 12 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>),</div><div class="line"><a name="l00125"></a><span class="lineno"> 125</span>  PTR_TO_VALUE(src_addr + 14 * src_stride_y, <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>));</div><div class="line"><a name="l00126"></a><span class="lineno"> 126</span> <span class="preprocessor">#else </span><span class="comment">/* STRIDE_X not equals 1 or 2 */</span><span class="preprocessor"></span></div><div class="line"><a name="l00127"></a><span class="lineno"> 127</span> <span class="preprocessor">#error "STRIDE_X larger than 2 is not supported"</span></div><div class="line"><a name="l00128"></a><span class="lineno"> 128</span> <span class="preprocessor">#endif </span><span class="comment">/* STRIDE_X == 2 */</span><span class="preprocessor"></span></div><div class="line"><a name="l00129"></a><span class="lineno"> 129</span>  values = <a class="code" href="direct__convolution1x1_8cl.xhtml#aebbeb1f22eca3a3f4c3e019e8f419f39">ADD_OP</a>(values, <a class="code" href="direct__convolution1x1_8cl.xhtml#ad3cc858846806e6b1d3694b9d0a2e6da">MUL_OP</a>((<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8))weight, col0));</div><div class="line"><a name="l00130"></a><span class="lineno"> 130</span> </div><div class="line"><a name="l00131"></a><span class="lineno"> 131</span>  src_addr += src_stride_x;</div><div class="line"><a name="l00132"></a><span class="lineno"> 132</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>.ptr += weights_stride_x;</div><div class="line"><a name="l00133"></a><span class="lineno"> 133</span>  }</div><div class="line"><a name="l00134"></a><span class="lineno"> 134</span> </div><div class="line"><a name="l00135"></a><span class="lineno"> 135</span> <span class="preprocessor">#ifdef HAS_BIAS</span></div><div class="line"><a name="l00136"></a><span class="lineno"> 136</span>  values = <a class="code" href="direct__convolution1x1_8cl.xhtml#aebbeb1f22eca3a3f4c3e019e8f419f39">ADD_OP</a>(values, (<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(DATA_TYPE_PROMOTED, 8)) * ((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a7e4940407322d6f0ccb8b6b86b856019">vector_offset</a>(&biases, id0))));</div><div class="line"><a name="l00137"></a><span class="lineno"> 137</span> <span class="preprocessor">#endif </span><span class="comment">/* defined(HAS_BIAS) */</span><span class="preprocessor"></span></div><div class="line"><a name="l00138"></a><span class="lineno"> 138</span> </div><div class="line"><a name="l00139"></a><span class="lineno"> 139</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr) = values.s0;</div><div class="line"><a name="l00140"></a><span class="lineno"> 140</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 1 * dst_stride_y)) = values.s1;</div><div class="line"><a name="l00141"></a><span class="lineno"> 141</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 2 * dst_stride_y)) = values.s2;</div><div class="line"><a name="l00142"></a><span class="lineno"> 142</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 3 * dst_stride_y)) = values.s3;</div><div class="line"><a name="l00143"></a><span class="lineno"> 143</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 4 * dst_stride_y)) = values.s4;</div><div class="line"><a name="l00144"></a><span class="lineno"> 144</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 5 * dst_stride_y)) = values.s5;</div><div class="line"><a name="l00145"></a><span class="lineno"> 145</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 6 * dst_stride_y)) = values.s6;</div><div class="line"><a name="l00146"></a><span class="lineno"> 146</span>  *((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 7 * dst_stride_y)) = values.s7;</div><div class="line"><a name="l00147"></a><span class="lineno"> 147</span> }</div><div class="line"><a name="l00148"></a><span class="lineno"> 148</span> <span class="preprocessor">#endif // defined(DATA_LAYOUT_NHWC)</span></div><div class="line"><a name="l00149"></a><span class="lineno"> 149</span> </div><div class="line"><a name="l00150"></a><span class="lineno"> 150</span> <span class="preprocessor">#if STRIDE_X == 3</span></div><div class="line"><a name="l00151"></a><span class="lineno"> 151</span> <span class="preprocessor">#define INPUT_PIXEL_STR(data_size) extract_input_stride3_##data_size</span></div><div class="line"><a name="l00152"></a><span class="lineno"> 152</span> <span class="preprocessor">#define INPUT_PIXEL(data_size) INPUT_PIXEL_STR(data_size)</span></div><div class="line"><a name="l00153"></a><span class="lineno"> 153</span> <span class="preprocessor">#elif STRIDE_X == 2</span></div><div class="line"><a name="l00154"></a><span class="lineno"> 154</span> <span class="preprocessor">#define INPUT_PIXEL(data_size) extract_input_stride2</span></div><div class="line"><a name="l00155"></a><span class="lineno"> 155</span> <span class="preprocessor">#elif STRIDE_X == 1</span></div><div class="line"><a name="l00156"></a><span class="lineno"> 156</span> <span class="preprocessor">#define INPUT_PIXEL(data_size) extract_input_stride1</span></div><div class="line"><a name="l00157"></a><span class="lineno"> 157</span> <span class="preprocessor">#else </span><span class="comment">/* STRIDE_X not equals 1, 2 or 3 */</span><span class="preprocessor"></span></div><div class="line"><a name="l00158"></a><span class="lineno"> 158</span> <span class="preprocessor">#error "Only support strides 1, 2 and 3"</span></div><div class="line"><a name="l00159"></a><span class="lineno"> 159</span> <span class="preprocessor">#endif </span><span class="comment">/* STRIDE_X == 3 */</span><span class="preprocessor"></span></div><div class="line"><a name="l00160"></a><span class="lineno"> 160</span> <span class="comment"></span></div><div class="line"><a name="l00161"></a><span class="lineno"> 161</span> <span class="comment">/** Extracts a 1D horizontal vector from the input tensor with stride as 1.</span></div><div class="line"><a name="l00162"></a><span class="lineno"> 162</span> <span class="comment"> *</span></div><div class="line"><a name="l00163"></a><span class="lineno"> 163</span> <span class="comment"> * @param[in] input_pixel Pointer to the first pixel.</span></div><div class="line"><a name="l00164"></a><span class="lineno"> 164</span> <span class="comment"> *</span></div><div class="line"><a name="l00165"></a><span class="lineno"> 165</span> <span class="comment"> * @return extracted input values.</span></div><div class="line"><a name="l00166"></a><span class="lineno"> 166</span> <span class="comment"> */</span></div><div class="line"><a name="l00167"></a><span class="lineno"> 167</span> <span class="keyword">inline</span> <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8) extract_input_stride1(__global const <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *input_pixel)</div><div class="line"><a name="l00168"></a><span class="lineno"> 168</span> {</div><div class="line"><a name="l00169"></a><span class="lineno"> 169</span>  <span class="keywordflow">return</span> vload8(0, input_pixel);</div><div class="line"><a name="l00170"></a><span class="lineno"> 170</span> }</div><div class="line"><a name="l00171"></a><span class="lineno"> 171</span> <span class="comment"></span></div><div class="line"><a name="l00172"></a><span class="lineno"> 172</span> <span class="comment">/** Extracts a 1D horizontal vector from the input tensor with stride as 2.</span></div><div class="line"><a name="l00173"></a><span class="lineno"> 173</span> <span class="comment"> *</span></div><div class="line"><a name="l00174"></a><span class="lineno"> 174</span> <span class="comment"> * @param[in] input_pixel Pointer to the first pixel.</span></div><div class="line"><a name="l00175"></a><span class="lineno"> 175</span> <span class="comment"> *</span></div><div class="line"><a name="l00176"></a><span class="lineno"> 176</span> <span class="comment"> * @return extracted input values.</span></div><div class="line"><a name="l00177"></a><span class="lineno"> 177</span> <span class="comment"> */</span></div><div class="line"><a name="l00178"></a><span class="lineno"> 178</span> <span class="keyword">inline</span> <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8) extract_input_stride2(__global const <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *input_pixel)</div><div class="line"><a name="l00179"></a><span class="lineno"> 179</span> {</div><div class="line"><a name="l00180"></a><span class="lineno"> 180</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 16)</div><div class="line"><a name="l00181"></a><span class="lineno"> 181</span>  temp = vload16(0, input_pixel);</div><div class="line"><a name="l00182"></a><span class="lineno"> 182</span>  return temp.s02468ace;</div><div class="line"><a name="l00183"></a><span class="lineno"> 183</span> }</div><div class="line"><a name="l00184"></a><span class="lineno"> 184</span> <span class="comment"></span></div><div class="line"><a name="l00185"></a><span class="lineno"> 185</span> <span class="comment">/** Extracts a 1D horizontal vector from the input tensor with stride as 3 and 32-bit data size.</span></div><div class="line"><a name="l00186"></a><span class="lineno"> 186</span> <span class="comment"> *</span></div><div class="line"><a name="l00187"></a><span class="lineno"> 187</span> <span class="comment"> * @param[in] input_pixel Pointer to the first pixel.</span></div><div class="line"><a name="l00188"></a><span class="lineno"> 188</span> <span class="comment"> *</span></div><div class="line"><a name="l00189"></a><span class="lineno"> 189</span> <span class="comment"> * @return extracted input values.</span></div><div class="line"><a name="l00190"></a><span class="lineno"> 190</span> <span class="comment"> */</span></div><div class="line"><a name="l00191"></a><span class="lineno"> 191</span> inline <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8) extract_input_stride3_32(__global const <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *input_pixel)</div><div class="line"><a name="l00192"></a><span class="lineno"> 192</span> {</div><div class="line"><a name="l00193"></a><span class="lineno"> 193</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 4)</div><div class="line"><a name="l00194"></a><span class="lineno"> 194</span>  temp1 = vload4(0, input_pixel);</div><div class="line"><a name="l00195"></a><span class="lineno"> 195</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 4)</div><div class="line"><a name="l00196"></a><span class="lineno"> 196</span>  temp2 = vload4(0, input_pixel + 6);</div><div class="line"><a name="l00197"></a><span class="lineno"> 197</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 4)</div><div class="line"><a name="l00198"></a><span class="lineno"> 198</span>  temp3 = vload4(0, input_pixel + 12);</div><div class="line"><a name="l00199"></a><span class="lineno"> 199</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 4)</div><div class="line"><a name="l00200"></a><span class="lineno"> 200</span>  temp4 = vload4(0, input_pixel + 18);</div><div class="line"><a name="l00201"></a><span class="lineno"> 201</span>  return (<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8))(temp1.s03, temp2.s03, temp3.s03, temp4.s03);</div><div class="line"><a name="l00202"></a><span class="lineno"> 202</span> }</div><div class="line"><a name="l00203"></a><span class="lineno"> 203</span> <span class="comment"></span></div><div class="line"><a name="l00204"></a><span class="lineno"> 204</span> <span class="comment">/** Extracts a 1D horizontal vector from the input tensor with stride as 3 and 16-bit data size.</span></div><div class="line"><a name="l00205"></a><span class="lineno"> 205</span> <span class="comment"> *</span></div><div class="line"><a name="l00206"></a><span class="lineno"> 206</span> <span class="comment"> * @param[in] input_pixel Pointer to the first pixel.</span></div><div class="line"><a name="l00207"></a><span class="lineno"> 207</span> <span class="comment"> *</span></div><div class="line"><a name="l00208"></a><span class="lineno"> 208</span> <span class="comment"> * @return extracted input values.</span></div><div class="line"><a name="l00209"></a><span class="lineno"> 209</span> <span class="comment"> */</span></div><div class="line"><a name="l00210"></a><span class="lineno"> 210</span> inline <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8) extract_input_stride3_16(__global const <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *input_pixel)</div><div class="line"><a name="l00211"></a><span class="lineno"> 211</span> {</div><div class="line"><a name="l00212"></a><span class="lineno"> 212</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8)</div><div class="line"><a name="l00213"></a><span class="lineno"> 213</span>  temp1 = vload8(0, input_pixel);</div><div class="line"><a name="l00214"></a><span class="lineno"> 214</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8)</div><div class="line"><a name="l00215"></a><span class="lineno"> 215</span>  temp2 = vload8(0, input_pixel + 8);</div><div class="line"><a name="l00216"></a><span class="lineno"> 216</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8)</div><div class="line"><a name="l00217"></a><span class="lineno"> 217</span>  temp3 = vload8(0, input_pixel + 16);</div><div class="line"><a name="l00218"></a><span class="lineno"> 218</span>  return (<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8))(temp1.s036, temp2.s147, temp3.s25);</div><div class="line"><a name="l00219"></a><span class="lineno"> 219</span> }</div><div class="line"><a name="l00220"></a><span class="lineno"> 220</span> <span class="comment"></span></div><div class="line"><a name="l00221"></a><span class="lineno"> 221</span> <span class="comment">/** Extracts a 1D horizontal vector from the input tensor with stride as 3 and 8-bit data size.</span></div><div class="line"><a name="l00222"></a><span class="lineno"> 222</span> <span class="comment"> *</span></div><div class="line"><a name="l00223"></a><span class="lineno"> 223</span> <span class="comment"> * @param[in] input_pixel Pointer to the first pixel.</span></div><div class="line"><a name="l00224"></a><span class="lineno"> 224</span> <span class="comment"> *</span></div><div class="line"><a name="l00225"></a><span class="lineno"> 225</span> <span class="comment"> * @return extracted input values.</span></div><div class="line"><a name="l00226"></a><span class="lineno"> 226</span> <span class="comment"> */</span></div><div class="line"><a name="l00227"></a><span class="lineno"> 227</span> inline <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8) extract_input_stride3_8(__global const <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *input_pixel)</div><div class="line"><a name="l00228"></a><span class="lineno"> 228</span> {</div><div class="line"><a name="l00229"></a><span class="lineno"> 229</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 16)</div><div class="line"><a name="l00230"></a><span class="lineno"> 230</span>  temp1 = vload16(0, input_pixel);</div><div class="line"><a name="l00231"></a><span class="lineno"> 231</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 16)</div><div class="line"><a name="l00232"></a><span class="lineno"> 232</span>  temp2 = vload16(0, input_pixel + 12);</div><div class="line"><a name="l00233"></a><span class="lineno"> 233</span>  return (<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8))(temp1.s0369, temp2.s0369);</div><div class="line"><a name="l00234"></a><span class="lineno"> 234</span> }</div><div class="line"><a name="l00235"></a><span class="lineno"> 235</span> <span class="comment"></span></div><div class="line"><a name="l00236"></a><span class="lineno"> 236</span> <span class="comment">/** This kernel performs a direct convolution to convolve the low three dimensions.</span></div><div class="line"><a name="l00237"></a><span class="lineno"> 237</span> <span class="comment"> *</span></div><div class="line"><a name="l00238"></a><span class="lineno"> 238</span> <span class="comment"> * @note The data type must be passed at compile time using -DDATA_TYPE: e.g. -DDATA_TYPE=float</span></div><div class="line"><a name="l00239"></a><span class="lineno"> 239</span> <span class="comment"> * @note The data size must be passed at compile time using -DDATA_SIZE e.g. -DDATA_SIZE=32</span></div><div class="line"><a name="l00240"></a><span class="lineno"> 240</span> <span class="comment"> * @note The convolution stride x must be passed at compile time using -DSTRIDE_X e.g. -DSTRIDE_X=1</span></div><div class="line"><a name="l00241"></a><span class="lineno"> 241</span> <span class="comment"> * @note The third dimensions of the weights tensors must be passed at compile time using -DWEIGHTS_DEPTH</span></div><div class="line"><a name="l00242"></a><span class="lineno"> 242</span> <span class="comment"> * @note In case biases will be added to the convolution -DHAS_BIAS has to be passed to append the final matrix with 1 in each row.</span></div><div class="line"><a name="l00243"></a><span class="lineno"> 243</span> <span class="comment"> *</span></div><div class="line"><a name="l00244"></a><span class="lineno"> 244</span> <span class="comment"> * @param[in] src_ptr Pointer to the source tensor. Supported data types: F16/F32</span></div><div class="line"><a name="l00245"></a><span class="lineno"> 245</span> <span class="comment"> * @param[in] src_stride_x Stride of the source tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00246"></a><span class="lineno"> 246</span> <span class="comment"> * @param[in] src_step_x src_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00247"></a><span class="lineno"> 247</span> <span class="comment"> * @param[in] src_stride_y Stride of the source tensor in Y dimension (in bytes)</span></div><div class="line"><a name="l00248"></a><span class="lineno"> 248</span> <span class="comment"> * @param[in] src_step_y src_stride_y * number of elements along Y processed per workitem(in bytes)</span></div><div class="line"><a name="l00249"></a><span class="lineno"> 249</span> <span class="comment"> * @param[in] src_stride_z Stride of the source tensor in Z dimension (in bytes)</span></div><div class="line"><a name="l00250"></a><span class="lineno"> 250</span> <span class="comment"> * @param[in] src_step_z src_stride_z * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00251"></a><span class="lineno"> 251</span> <span class="comment"> * @param[in] src_offset_first_element_in_bytes The offset of the first element in the source tensor</span></div><div class="line"><a name="l00252"></a><span class="lineno"> 252</span> <span class="comment"> * @param[out] dst_ptr Pointer to the destination tensor. Supported data types: same as @p src_ptr</span></div><div class="line"><a name="l00253"></a><span class="lineno"> 253</span> <span class="comment"> * @param[in] dst_stride_x Stride of the destination tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00254"></a><span class="lineno"> 254</span> <span class="comment"> * @param[in] dst_step_x dst_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00255"></a><span class="lineno"> 255</span> <span class="comment"> * @param[in] dst_stride_y Stride of the destination tensor in Y dimension (in bytes)</span></div><div class="line"><a name="l00256"></a><span class="lineno"> 256</span> <span class="comment"> * @param[in] dst_step_y dst_stride_y * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00257"></a><span class="lineno"> 257</span> <span class="comment"> * @param[in] dst_stride_z Stride of the destination tensor in Z dimension (in bytes)</span></div><div class="line"><a name="l00258"></a><span class="lineno"> 258</span> <span class="comment"> * @param[in] dst_step_z dst_stride_z * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00259"></a><span class="lineno"> 259</span> <span class="comment"> * @param[in] dst_offset_first_element_in_bytes The offset of the first element in the destination tensor</span></div><div class="line"><a name="l00260"></a><span class="lineno"> 260</span> <span class="comment"> * @param[in] weights_ptr Pointer to the weights tensor. Supported data types: same as @p src_ptr</span></div><div class="line"><a name="l00261"></a><span class="lineno"> 261</span> <span class="comment"> * @param[in] weights_stride_x Stride of the weights tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00262"></a><span class="lineno"> 262</span> <span class="comment"> * @param[in] weights_step_x weights_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00263"></a><span class="lineno"> 263</span> <span class="comment"> * @param[in] weights_stride_y Stride of the weights tensor in Y dimension (in bytes)</span></div><div class="line"><a name="l00264"></a><span class="lineno"> 264</span> <span class="comment"> * @param[in] weights_step_y weights_stride_y * number of elements along y processed per workitem(in bytes)</span></div><div class="line"><a name="l00265"></a><span class="lineno"> 265</span> <span class="comment"> * @param[in] weights_stride_z Stride of the weights tensor in Z dimension (in bytes)</span></div><div class="line"><a name="l00266"></a><span class="lineno"> 266</span> <span class="comment"> * @param[in] weights_step_z weights_stride_z * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00267"></a><span class="lineno"> 267</span> <span class="comment"> * @param[in] weights_offset_first_element_in_bytes The offset of the first element in the weights tensor</span></div><div class="line"><a name="l00268"></a><span class="lineno"> 268</span> <span class="comment"> * @param[in] biases_ptr Pointer to the biases tensor. Same as @p src_ptr</span></div><div class="line"><a name="l00269"></a><span class="lineno"> 269</span> <span class="comment"> * @param[in] biases_stride_x Stride of the biases tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00270"></a><span class="lineno"> 270</span> <span class="comment"> * @param[in] biases_step_x biases_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00271"></a><span class="lineno"> 271</span> <span class="comment"> * @param[in] biases_offset_first_element_in_bytes The offset of the first element in the biases tensor</span></div><div class="line"><a name="l00272"></a><span class="lineno"> 272</span> <span class="comment"> * @param[in] weights_stride_w Stride of the weights tensor in the 4th dimension</span></div><div class="line"><a name="l00273"></a><span class="lineno"> 273</span> <span class="comment"> */</span></div><div class="line"><a name="l00274"></a><span class="lineno"> 274</span> __kernel <span class="keywordtype">void</span> direct_convolution1x1(</div><div class="line"><a name="l00275"></a><span class="lineno"> 275</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>),</div><div class="line"><a name="l00276"></a><span class="lineno"> 276</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>),</div><div class="line"><a name="l00277"></a><span class="lineno"> 277</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>),</div><div class="line"><a name="l00278"></a><span class="lineno"> 278</span> <span class="preprocessor">#ifdef HAS_BIAS</span></div><div class="line"><a name="l00279"></a><span class="lineno"> 279</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(biases),</div><div class="line"><a name="l00280"></a><span class="lineno"> 280</span> <span class="preprocessor">#endif </span><span class="comment">/* defined(HAS_BIAS) */</span><span class="preprocessor"></span></div><div class="line"><a name="l00281"></a><span class="lineno"> 281</span>  <span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> weights_stride_w)</div><div class="line"><a name="l00282"></a><span class="lineno"> 282</span> {</div><div class="line"><a name="l00283"></a><span class="lineno"> 283</span>  <a class="code" href="struct_image.xhtml">Image</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a> = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#aebe814363556c244be043b13e7969197">CONVERT_TO_IMAGE_STRUCT</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>);</div><div class="line"><a name="l00284"></a><span class="lineno"> 284</span>  <a class="code" href="struct_tensor3_d.xhtml">Tensor3D</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a> = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a79e8e562daa6599317d2d1cd86ef1bf2">CONVERT_TO_TENSOR3D_STRUCT_NO_STEP</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>);</div><div class="line"><a name="l00285"></a><span class="lineno"> 285</span>  <a class="code" href="struct_tensor3_d.xhtml">Tensor3D</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a> = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a31c8c760f08fb1a331b16b7c204321dc">CONVERT_TO_TENSOR3D_STRUCT</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>);</div><div class="line"><a name="l00286"></a><span class="lineno"> 286</span> </div><div class="line"><a name="l00287"></a><span class="lineno"> 287</span> <span class="preprocessor">#ifdef HAS_BIAS</span></div><div class="line"><a name="l00288"></a><span class="lineno"> 288</span>  <a class="code" href="struct_vector.xhtml">Vector</a> biases = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a64d779f80eeb923e0ab2313433f7b40b">CONVERT_TO_VECTOR_STRUCT_NO_STEP</a>(biases);</div><div class="line"><a name="l00289"></a><span class="lineno"> 289</span> <span class="preprocessor">#endif </span><span class="comment">/* defined(HAS_BIAS) */</span><span class="preprocessor"></span></div><div class="line"><a name="l00290"></a><span class="lineno"> 290</span> </div><div class="line"><a name="l00291"></a><span class="lineno"> 291</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(DATA_TYPE_PROMOTED, 8)</div><div class="line"><a name="l00292"></a><span class="lineno"> 292</span>  values = 0;</div><div class="line"><a name="l00293"></a><span class="lineno"> 293</span> </div><div class="line"><a name="l00294"></a><span class="lineno"> 294</span>  const uint z_index = get_global_id(2);</div><div class="line"><a name="l00295"></a><span class="lineno"> 295</span> </div><div class="line"><a name="l00296"></a><span class="lineno"> 296</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>.ptr += z_index * weights_stride_w;</div><div class="line"><a name="l00297"></a><span class="lineno"> 297</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a0886942393a3ba0dfefaa7516b159784">for</a>(volatile <span class="keywordtype">int</span> d = 0; d < WEIGHTS_DEPTH; ++d)</div><div class="line"><a name="l00298"></a><span class="lineno"> 298</span>  {</div><div class="line"><a name="l00299"></a><span class="lineno"> 299</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> weight = *(__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>.ptr;</div><div class="line"><a name="l00300"></a><span class="lineno"> 300</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8)</div><div class="line"><a name="l00301"></a><span class="lineno"> 301</span>  input_pixel = INPUT_PIXEL(DATA_SIZE)((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>.ptr);</div><div class="line"><a name="l00302"></a><span class="lineno"> 302</span>  values = <a class="code" href="direct__convolution1x1_8cl.xhtml#aebbeb1f22eca3a3f4c3e019e8f419f39">ADD_OP</a>(values, <a class="code" href="direct__convolution1x1_8cl.xhtml#ad3cc858846806e6b1d3694b9d0a2e6da">MUL_OP</a>((<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8))weight, input_pixel));</div><div class="line"><a name="l00303"></a><span class="lineno"> 303</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>.ptr += src_stride_z;</div><div class="line"><a name="l00304"></a><span class="lineno"> 304</span>  <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>.ptr += weights_stride_z;</div><div class="line"><a name="l00305"></a><span class="lineno"> 305</span>  }</div><div class="line"><a name="l00306"></a><span class="lineno"> 306</span> </div><div class="line"><a name="l00307"></a><span class="lineno"> 307</span> <span class="preprocessor">#ifdef HAS_BIAS</span></div><div class="line"><a name="l00308"></a><span class="lineno"> 308</span>  values = <a class="code" href="direct__convolution1x1_8cl.xhtml#aebbeb1f22eca3a3f4c3e019e8f419f39">ADD_OP</a>(values, (<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(DATA_TYPE_PROMOTED, 8)) * ((__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a7e4940407322d6f0ccb8b6b86b856019">vector_offset</a>(&biases, z_index))));</div><div class="line"><a name="l00309"></a><span class="lineno"> 309</span> <span class="preprocessor">#endif </span><span class="comment">/* defined(HAS_BIAS) */</span><span class="preprocessor"></span></div><div class="line"><a name="l00310"></a><span class="lineno"> 310</span> </div><div class="line"><a name="l00311"></a><span class="lineno"> 311</span>  vstore8(<a class="code" href="direct__convolution1x1_8cl.xhtml#a1f15728672380ade7a238f5e783d54d2">CONVERT_SAT</a>(values, <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>, 8)), 0, (__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr);</div><div class="line"><a name="l00312"></a><span class="lineno"> 312</span> }</div><div class="line"><a name="l00313"></a><span class="lineno"> 313</span> <span class="preprocessor">#endif // defined(DATA_TYPE) && defined(DATA_SIZE) && defined(STRIDE_X) && defined(WEIGHTS_DEPTH)</span></div><div class="line"><a name="l00314"></a><span class="lineno"> 314</span> </div><div class="line"><a name="l00315"></a><span class="lineno"> 315</span> <span class="preprocessor">#if defined(WEIGHTS_DEPTH)</span></div><div class="line"><a name="l00316"></a><span class="lineno"> 316</span> </div><div class="line"><a name="l00317"></a><span class="lineno"> 317</span> <span class="preprocessor">#define CONVOLUTION1x1_BIFROST(acc, src, weight_value) \</span></div><div class="line"><a name="l00318"></a><span class="lineno"> 318</span> <span class="preprocessor"> ({ \</span></div><div class="line"><a name="l00319"></a><span class="lineno"> 319</span> <span class="preprocessor"> acc.s0 = mad(src.s0, weight_value, acc.s0); \</span></div><div class="line"><a name="l00320"></a><span class="lineno"> 320</span> <span class="preprocessor"> acc.s1 = mad(src.s1, weight_value, acc.s1); \</span></div><div class="line"><a name="l00321"></a><span class="lineno"> 321</span> <span class="preprocessor"> acc.s2 = mad(src.s2, weight_value, acc.s2); \</span></div><div class="line"><a name="l00322"></a><span class="lineno"> 322</span> <span class="preprocessor"> acc.s3 = mad(src.s3, weight_value, acc.s3); \</span></div><div class="line"><a name="l00323"></a><span class="lineno"> 323</span> <span class="preprocessor"> })</span></div><div class="line"><a name="l00324"></a><span class="lineno"> 324</span> <span class="comment"></span></div><div class="line"><a name="l00325"></a><span class="lineno"> 325</span> <span class="comment">/** An optimized direct convolution 1x1 OpenCL kernel for Bifrost architectures when the data type is F32</span></div><div class="line"><a name="l00326"></a><span class="lineno"> 326</span> <span class="comment"> *</span></div><div class="line"><a name="l00327"></a><span class="lineno"> 327</span> <span class="comment"> * @note This OpenCL kernel works only with stride_x and stride_y equal to 1</span></div><div class="line"><a name="l00328"></a><span class="lineno"> 328</span> <span class="comment"> * @note The third dimensions of the weights tensors must be passed at compile time using -DWEIGHTS_DEPTH</span></div><div class="line"><a name="l00329"></a><span class="lineno"> 329</span> <span class="comment"> * @note In case biases, -DHAS_BIAS must to be passed at compile</span></div><div class="line"><a name="l00330"></a><span class="lineno"> 330</span> <span class="comment"> *</span></div><div class="line"><a name="l00331"></a><span class="lineno"> 331</span> <span class="comment"> * @param[in] src_ptr Pointer to the source tensor. Supported data types: F32</span></div><div class="line"><a name="l00332"></a><span class="lineno"> 332</span> <span class="comment"> * @param[in] src_stride_x Stride of the source tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00333"></a><span class="lineno"> 333</span> <span class="comment"> * @param[in] src_step_x src_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00334"></a><span class="lineno"> 334</span> <span class="comment"> * @param[in] src_stride_y Stride of the source tensor in Y dimension (in bytes)</span></div><div class="line"><a name="l00335"></a><span class="lineno"> 335</span> <span class="comment"> * @param[in] src_step_y src_stride_y * number of elements along Y processed per workitem(in bytes)</span></div><div class="line"><a name="l00336"></a><span class="lineno"> 336</span> <span class="comment"> * @param[in] src_stride_z Stride of the source tensor in Z dimension (in bytes)</span></div><div class="line"><a name="l00337"></a><span class="lineno"> 337</span> <span class="comment"> * @param[in] src_step_z src_stride_z * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00338"></a><span class="lineno"> 338</span> <span class="comment"> * @param[in] src_offset_first_element_in_bytes The offset of the first element in the source tensor</span></div><div class="line"><a name="l00339"></a><span class="lineno"> 339</span> <span class="comment"> * @param[out] dst_ptr Pointer to the destination tensor. Supported data types: same as @p src_ptr</span></div><div class="line"><a name="l00340"></a><span class="lineno"> 340</span> <span class="comment"> * @param[in] dst_stride_x Stride of the destination tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00341"></a><span class="lineno"> 341</span> <span class="comment"> * @param[in] dst_step_x dst_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00342"></a><span class="lineno"> 342</span> <span class="comment"> * @param[in] dst_stride_y Stride of the destination tensor in Y dimension (in bytes)</span></div><div class="line"><a name="l00343"></a><span class="lineno"> 343</span> <span class="comment"> * @param[in] dst_step_y dst_stride_y * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00344"></a><span class="lineno"> 344</span> <span class="comment"> * @param[in] dst_stride_z Stride of the destination tensor in Z dimension (in bytes)</span></div><div class="line"><a name="l00345"></a><span class="lineno"> 345</span> <span class="comment"> * @param[in] dst_step_z dst_stride_z * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00346"></a><span class="lineno"> 346</span> <span class="comment"> * @param[in] dst_offset_first_element_in_bytes The offset of the first element in the destination tensor</span></div><div class="line"><a name="l00347"></a><span class="lineno"> 347</span> <span class="comment"> * @param[in] weights_ptr Pointer to the weights tensor. Supported data types: same as @p src_ptr</span></div><div class="line"><a name="l00348"></a><span class="lineno"> 348</span> <span class="comment"> * @param[in] weights_stride_x Stride of the weights tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00349"></a><span class="lineno"> 349</span> <span class="comment"> * @param[in] weights_step_x weights_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00350"></a><span class="lineno"> 350</span> <span class="comment"> * @param[in] weights_stride_y Stride of the weights tensor in Y dimension (in bytes)</span></div><div class="line"><a name="l00351"></a><span class="lineno"> 351</span> <span class="comment"> * @param[in] weights_step_y weights_stride_y * number of elements along y processed per workitem(in bytes)</span></div><div class="line"><a name="l00352"></a><span class="lineno"> 352</span> <span class="comment"> * @param[in] weights_stride_z Stride of the weights tensor in Z dimension (in bytes)</span></div><div class="line"><a name="l00353"></a><span class="lineno"> 353</span> <span class="comment"> * @param[in] weights_step_z weights_stride_z * number of elements along Z processed per workitem(in bytes)</span></div><div class="line"><a name="l00354"></a><span class="lineno"> 354</span> <span class="comment"> * @param[in] weights_offset_first_element_in_bytes The offset of the first element in the weights tensor</span></div><div class="line"><a name="l00355"></a><span class="lineno"> 355</span> <span class="comment"> * @param[in] biases_ptr Pointer to the biases tensor. Same as @p src_ptr</span></div><div class="line"><a name="l00356"></a><span class="lineno"> 356</span> <span class="comment"> * @param[in] biases_stride_x Stride of the biases tensor in X dimension (in bytes)</span></div><div class="line"><a name="l00357"></a><span class="lineno"> 357</span> <span class="comment"> * @param[in] biases_step_x biases_stride_x * number of elements along X processed per workitem(in bytes)</span></div><div class="line"><a name="l00358"></a><span class="lineno"> 358</span> <span class="comment"> * @param[in] biases_offset_first_element_in_bytes The offset of the first element in the biases tensor</span></div><div class="line"><a name="l00359"></a><span class="lineno"> 359</span> <span class="comment"> * @param[in] weights_stride_w Stride of the weights tensor in the 4th dimension</span></div><div class="line"><a name="l00360"></a><span class="lineno"> 360</span> <span class="comment"> */</span></div><div class="line"><a name="l00361"></a><span class="lineno"> 361</span> __kernel <span class="keywordtype">void</span> direct_convolution1x1_f32_bifrost(</div><div class="line"><a name="l00362"></a><span class="lineno"> 362</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>),</div><div class="line"><a name="l00363"></a><span class="lineno"> 363</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>),</div><div class="line"><a name="l00364"></a><span class="lineno"> 364</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">weights</a>),</div><div class="line"><a name="l00365"></a><span class="lineno"> 365</span> #ifdef HAS_BIAS</div><div class="line"><a name="l00366"></a><span class="lineno"> 366</span>  <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a>(biases),</div><div class="line"><a name="l00367"></a><span class="lineno"> 367</span> #endif <span class="comment">/* defined(HAS_BIAS) */</span></div><div class="line"><a name="l00368"></a><span class="lineno"> 368</span>  <span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> weights_stride_w)</div><div class="line"><a name="l00369"></a><span class="lineno"> 369</span> {</div><div class="line"><a name="l00370"></a><span class="lineno"> 370</span>  <span class="comment">// Get the kernel index</span></div><div class="line"><a name="l00371"></a><span class="lineno"> 371</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> kernel_index = get_global_id(2);</div><div class="line"><a name="l00372"></a><span class="lineno"> 372</span> </div><div class="line"><a name="l00373"></a><span class="lineno"> 373</span>  <a class="code" href="struct_image.xhtml">Image</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a> = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#aebe814363556c244be043b13e7969197">CONVERT_TO_IMAGE_STRUCT</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>);</div><div class="line"><a name="l00374"></a><span class="lineno"> 374</span>  <a class="code" href="struct_tensor3_d.xhtml">Tensor3D</a> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a> = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a31c8c760f08fb1a331b16b7c204321dc">CONVERT_TO_TENSOR3D_STRUCT</a>(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>);</div><div class="line"><a name="l00375"></a><span class="lineno"> 375</span> </div><div class="line"><a name="l00376"></a><span class="lineno"> 376</span>  float4 acc0 = 0.0f;</div><div class="line"><a name="l00377"></a><span class="lineno"> 377</span>  float4 acc1 = 0.0f;</div><div class="line"><a name="l00378"></a><span class="lineno"> 378</span>  float4 acc2 = 0.0f;</div><div class="line"><a name="l00379"></a><span class="lineno"> 379</span>  float4 acc3 = 0.0f;</div><div class="line"><a name="l00380"></a><span class="lineno"> 380</span> </div><div class="line"><a name="l00381"></a><span class="lineno"> 381</span>  __global uchar *weights_addr = (__global uchar *)(weights_ptr + weights_offset_first_element_in_bytes + kernel_index * weights_stride_w);</div><div class="line"><a name="l00382"></a><span class="lineno"> 382</span>  __global uchar *src_addr = (__global uchar *)<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a009469e4d9b8fce3b6d5e97d2077827d">offset</a>(&<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">src</a>, 0, 0);</div><div class="line"><a name="l00383"></a><span class="lineno"> 383</span> </div><div class="line"><a name="l00384"></a><span class="lineno"> 384</span>  <span class="keywordflow">for</span>(ushort d = 0; d < (ushort)WEIGHTS_DEPTH; ++d)</div><div class="line"><a name="l00385"></a><span class="lineno"> 385</span>  {</div><div class="line"><a name="l00386"></a><span class="lineno"> 386</span>  <span class="comment">// Load the weights</span></div><div class="line"><a name="l00387"></a><span class="lineno"> 387</span>  <span class="keywordtype">float</span> weight = *((__global <span class="keywordtype">float</span> *)weights_addr);</div><div class="line"><a name="l00388"></a><span class="lineno"> 388</span> </div><div class="line"><a name="l00389"></a><span class="lineno"> 389</span>  <span class="comment">// Load values from row0 of input tensor</span></div><div class="line"><a name="l00390"></a><span class="lineno"> 390</span>  float4 src0 = vload4(0, (__global <span class="keywordtype">float</span> *)(src_addr + 0 * src_stride_y));</div><div class="line"><a name="l00391"></a><span class="lineno"> 391</span>  float4 src1 = vload4(0, (__global <span class="keywordtype">float</span> *)(src_addr + 1 * src_stride_y));</div><div class="line"><a name="l00392"></a><span class="lineno"> 392</span>  float4 src2 = vload4(0, (__global <span class="keywordtype">float</span> *)(src_addr + 2 * src_stride_y));</div><div class="line"><a name="l00393"></a><span class="lineno"> 393</span>  float4 src3 = vload4(0, (__global <span class="keywordtype">float</span> *)(src_addr + 3 * src_stride_y));</div><div class="line"><a name="l00394"></a><span class="lineno"> 394</span> </div><div class="line"><a name="l00395"></a><span class="lineno"> 395</span>  CONVOLUTION1x1_BIFROST(acc0, src0, weight);</div><div class="line"><a name="l00396"></a><span class="lineno"> 396</span>  CONVOLUTION1x1_BIFROST(acc1, src1, weight);</div><div class="line"><a name="l00397"></a><span class="lineno"> 397</span>  CONVOLUTION1x1_BIFROST(acc2, src2, weight);</div><div class="line"><a name="l00398"></a><span class="lineno"> 398</span>  CONVOLUTION1x1_BIFROST(acc3, src3, weight);</div><div class="line"><a name="l00399"></a><span class="lineno"> 399</span> </div><div class="line"><a name="l00400"></a><span class="lineno"> 400</span>  src_addr += src_stride_z;</div><div class="line"><a name="l00401"></a><span class="lineno"> 401</span>  weights_addr += weights_stride_z;</div><div class="line"><a name="l00402"></a><span class="lineno"> 402</span>  }</div><div class="line"><a name="l00403"></a><span class="lineno"> 403</span> </div><div class="line"><a name="l00404"></a><span class="lineno"> 404</span> <span class="preprocessor">#ifdef HAS_BIAS</span></div><div class="line"><a name="l00405"></a><span class="lineno"> 405</span>  <a class="code" href="struct_vector.xhtml">Vector</a> biases = <a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a64d779f80eeb923e0ab2313433f7b40b">CONVERT_TO_VECTOR_STRUCT_NO_STEP</a>(biases);</div><div class="line"><a name="l00406"></a><span class="lineno"> 406</span> </div><div class="line"><a name="l00407"></a><span class="lineno"> 407</span>  <span class="keywordtype">float</span> <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a> = (float) * ((__global <span class="keywordtype">float</span> *)(<a class="code" href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a7e4940407322d6f0ccb8b6b86b856019">vector_offset</a>(&biases, kernel_index)));</div><div class="line"><a name="l00408"></a><span class="lineno"> 408</span> </div><div class="line"><a name="l00409"></a><span class="lineno"> 409</span>  acc0.s0 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00410"></a><span class="lineno"> 410</span>  acc0.s1 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00411"></a><span class="lineno"> 411</span>  acc0.s2 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00412"></a><span class="lineno"> 412</span>  acc0.s3 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00413"></a><span class="lineno"> 413</span>  acc1.s0 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00414"></a><span class="lineno"> 414</span>  acc1.s1 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00415"></a><span class="lineno"> 415</span>  acc1.s2 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00416"></a><span class="lineno"> 416</span>  acc1.s3 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00417"></a><span class="lineno"> 417</span>  acc2.s0 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00418"></a><span class="lineno"> 418</span>  acc2.s1 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00419"></a><span class="lineno"> 419</span>  acc2.s2 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00420"></a><span class="lineno"> 420</span>  acc2.s3 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00421"></a><span class="lineno"> 421</span>  acc3.s0 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00422"></a><span class="lineno"> 422</span>  acc3.s1 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00423"></a><span class="lineno"> 423</span>  acc3.s2 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00424"></a><span class="lineno"> 424</span>  acc3.s3 += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>;</div><div class="line"><a name="l00425"></a><span class="lineno"> 425</span> <span class="preprocessor">#endif </span><span class="comment">/* defined(HAS_BIAS) */</span><span class="preprocessor"></span></div><div class="line"><a name="l00426"></a><span class="lineno"> 426</span> </div><div class="line"><a name="l00427"></a><span class="lineno"> 427</span>  vstore4(acc0, 0, (__global <span class="keywordtype">float</span> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 0 * dst_stride_y));</div><div class="line"><a name="l00428"></a><span class="lineno"> 428</span>  vstore4(acc1, 0, (__global <span class="keywordtype">float</span> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 1 * dst_stride_y));</div><div class="line"><a name="l00429"></a><span class="lineno"> 429</span>  vstore4(acc2, 0, (__global <span class="keywordtype">float</span> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 2 * dst_stride_y));</div><div class="line"><a name="l00430"></a><span class="lineno"> 430</span>  vstore4(acc3, 0, (__global <span class="keywordtype">float</span> *)(<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">dst</a>.ptr + 3 * dst_stride_y));</div><div class="line"><a name="l00431"></a><span class="lineno"> 431</span> }</div><div class="line"><a name="l00432"></a><span class="lineno"> 432</span> <span class="preprocessor">#endif // defined(WEIGHTS_DEPTH)</span></div><div class="ttc" id="struct_vector_xhtml"><div class="ttname"><a href="struct_vector.xhtml">Vector</a></div><div class="ttdoc">Structure to hold Vector information.</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00341">helpers.h:341</a></div></div>
+<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a009469e4d9b8fce3b6d5e97d2077827d"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a009469e4d9b8fce3b6d5e97d2077827d">offset</a></div><div class="ttdeci">__global uchar * offset(const Image *img, int x, int y)</div><div class="ttdoc">Get the pointer position of a Image.</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00510">helpers.h:510</a></div></div>
+<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_aebe814363556c244be043b13e7969197"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#aebe814363556c244be043b13e7969197">CONVERT_TO_IMAGE_STRUCT</a></div><div class="ttdeci">#define CONVERT_TO_IMAGE_STRUCT(name)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00311">helpers.h:311</a></div></div>
<div class="ttc" id="convolution3x3_8cl_xhtml_afb8c72ce35c4a1f4a2588d6573e54aa1"><div class="ttname"><a href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a></div><div class="ttdeci">#define DATA_TYPE</div><div class="ttdef"><b>Definition:</b> <a href="convolution3x3_8cl_source.xhtml#l00027">convolution3x3.cl:27</a></div></div>
<div class="ttc" id="direct__convolution1x1_8cl_xhtml_a1f15728672380ade7a238f5e783d54d2"><div class="ttname"><a href="direct__convolution1x1_8cl.xhtml#a1f15728672380ade7a238f5e783d54d2">CONVERT_SAT</a></div><div class="ttdeci">#define CONVERT_SAT(a, b)</div><div class="ttdef"><b>Definition:</b> <a href="direct__convolution1x1_8cl_source.xhtml#l00030">direct_convolution1x1.cl:30</a></div></div>
<div class="ttc" id="namespacearm__compute_1_1test_1_1validation_xhtml_a0886942393a3ba0dfefaa7516b159784"><div class="ttname"><a href="namespacearm__compute_1_1test_1_1validation.xhtml#a0886942393a3ba0dfefaa7516b159784">arm_compute::test::validation::for</a></div><div class="ttdeci">for(size_t k=0;k< _target.size();++k)</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_2_unstack_8cpp_source.xhtml#l00091">Unstack.cpp:91</a></div></div>
-<div class="ttc" id="struct_tensor3_d_xhtml"><div class="ttname"><a href="struct_tensor3_d.xhtml">Tensor3D</a></div><div class="ttdoc">Structure to hold 3D tensor information.</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00176">helpers.h:176</a></div></div>
-<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a79e8e562daa6599317d2d1cd86ef1bf2"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a79e8e562daa6599317d2d1cd86ef1bf2">CONVERT_TO_TENSOR3D_STRUCT_NO_STEP</a></div><div class="ttdeci">#define CONVERT_TO_TENSOR3D_STRUCT_NO_STEP(name)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00148">helpers.h:148</a></div></div>
+<div class="ttc" id="struct_tensor3_d_xhtml"><div class="ttname"><a href="struct_tensor3_d.xhtml">Tensor3D</a></div><div class="ttdoc">Structure to hold 3D tensor information.</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00358">helpers.h:358</a></div></div>
+<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a79e8e562daa6599317d2d1cd86ef1bf2"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a79e8e562daa6599317d2d1cd86ef1bf2">CONVERT_TO_TENSOR3D_STRUCT_NO_STEP</a></div><div class="ttdeci">#define CONVERT_TO_TENSOR3D_STRUCT_NO_STEP(name)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00330">helpers.h:330</a></div></div>
<div class="ttc" id="direct__convolution1x1_8cl_xhtml_aebbeb1f22eca3a3f4c3e019e8f419f39"><div class="ttname"><a href="direct__convolution1x1_8cl.xhtml#aebbeb1f22eca3a3f4c3e019e8f419f39">ADD_OP</a></div><div class="ttdeci">#define ADD_OP(a, b)</div><div class="ttdef"><b>Definition:</b> <a href="direct__convolution1x1_8cl_source.xhtml#l00028">direct_convolution1x1.cl:28</a></div></div>
-<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a40a6eb9f2a7712f08d6bb8ff6c9e6ca7"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a></div><div class="ttdeci">#define VECTOR_DECLARATION(name)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00087">helpers.h:87</a></div></div>
+<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a40a6eb9f2a7712f08d6bb8ff6c9e6ca7"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a40a6eb9f2a7712f08d6bb8ff6c9e6ca7">VECTOR_DECLARATION</a></div><div class="ttdeci">#define VECTOR_DECLARATION(name)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00269">helpers.h:269</a></div></div>
<div class="ttc" id="namespacearm__compute_1_1test_1_1validation_xhtml_adbf67dcee294e673cf796f1ed8aeb6a4"><div class="ttname"><a href="namespacearm__compute_1_1test_1_1validation.xhtml#adbf67dcee294e673cf796f1ed8aeb6a4">arm_compute::test::validation::dst</a></div><div class="ttdeci">CLTensor dst</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_2_absolute_difference_8cpp_source.xhtml#l00102">AbsoluteDifference.cpp:102</a></div></div>
<div class="ttc" id="namespacearm__compute_1_1test_1_1validation_xhtml_a3a77be8aebd8e00522b32061d46ccdbd"><div class="ttname"><a href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">arm_compute::test::validation::bias</a></div><div class="ttdeci">CLTensor bias</div><div class="ttdef"><b>Definition:</b> <a href="validation_2_c_l_2_convolution_layer_8cpp_source.xhtml#l00181">ConvolutionLayer.cpp:181</a></div></div>
-<div class="ttc" id="struct_image_xhtml"><div class="ttname"><a href="struct_image.xhtml">Image</a></div><div class="ttdoc">Structure to hold Image information.</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00167">helpers.h:167</a></div></div>
+<div class="ttc" id="struct_image_xhtml"><div class="ttname"><a href="struct_image.xhtml">Image</a></div><div class="ttdoc">Structure to hold Image information.</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00349">helpers.h:349</a></div></div>
<div class="ttc" id="direct__convolution1x1_8cl_xhtml_ad3cc858846806e6b1d3694b9d0a2e6da"><div class="ttname"><a href="direct__convolution1x1_8cl.xhtml#ad3cc858846806e6b1d3694b9d0a2e6da">MUL_OP</a></div><div class="ttdeci">#define MUL_OP(a, b)</div><div class="ttdef"><b>Definition:</b> <a href="direct__convolution1x1_8cl_source.xhtml#l00029">direct_convolution1x1.cl:29</a></div></div>
-<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a31c8c760f08fb1a331b16b7c204321dc"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a31c8c760f08fb1a331b16b7c204321dc">CONVERT_TO_TENSOR3D_STRUCT</a></div><div class="ttdeci">#define CONVERT_TO_TENSOR3D_STRUCT(name)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00144">helpers.h:144</a></div></div>
-<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a7e4940407322d6f0ccb8b6b86b856019"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a7e4940407322d6f0ccb8b6b86b856019">vector_offset</a></div><div class="ttdeci">__global const uchar * vector_offset(const Vector *vec, int x)</div><div class="ttdoc">Get the pointer position of a Vector.</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00317">helpers.h:317</a></div></div>
+<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a31c8c760f08fb1a331b16b7c204321dc"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a31c8c760f08fb1a331b16b7c204321dc">CONVERT_TO_TENSOR3D_STRUCT</a></div><div class="ttdeci">#define CONVERT_TO_TENSOR3D_STRUCT(name)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00326">helpers.h:326</a></div></div>
+<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a7e4940407322d6f0ccb8b6b86b856019"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a7e4940407322d6f0ccb8b6b86b856019">vector_offset</a></div><div class="ttdeci">__global const uchar * vector_offset(const Vector *vec, int x)</div><div class="ttdoc">Get the pointer position of a Vector.</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00499">helpers.h:499</a></div></div>
<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml">helpers.h</a></div></div>
<div class="ttc" id="namespacearm__compute_1_1test_1_1validation_xhtml_a64a08a9fec5aeee8650e7182b6d171d0"><div class="ttname"><a href="namespacearm__compute_1_1test_1_1validation.xhtml#a64a08a9fec5aeee8650e7182b6d171d0">arm_compute::test::validation::weights</a></div><div class="ttdeci">CLTensor weights</div><div class="ttdef"><b>Definition:</b> <a href="validation_2_c_l_2_convolution_layer_8cpp_source.xhtml#l00180">ConvolutionLayer.cpp:180</a></div></div>
-<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a6b83038822d1ae7ab619b684ed3b7fc0"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a></div><div class="ttdeci">#define TENSOR3D_DECLARATION(name)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00101">helpers.h:101</a></div></div>
-<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a64d779f80eeb923e0ab2313433f7b40b"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a64d779f80eeb923e0ab2313433f7b40b">CONVERT_TO_VECTOR_STRUCT_NO_STEP</a></div><div class="ttdeci">#define CONVERT_TO_VECTOR_STRUCT_NO_STEP(name)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00126">helpers.h:126</a></div></div>
+<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a6b83038822d1ae7ab619b684ed3b7fc0"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a6b83038822d1ae7ab619b684ed3b7fc0">TENSOR3D_DECLARATION</a></div><div class="ttdeci">#define TENSOR3D_DECLARATION(name)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00283">helpers.h:283</a></div></div>
+<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a64d779f80eeb923e0ab2313433f7b40b"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a64d779f80eeb923e0ab2313433f7b40b">CONVERT_TO_VECTOR_STRUCT_NO_STEP</a></div><div class="ttdeci">#define CONVERT_TO_VECTOR_STRUCT_NO_STEP(name)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00308">helpers.h:308</a></div></div>
<div class="ttc" id="namespacearm__compute_1_1test_1_1validation_xhtml_a989ab3e96426615bb98e04e0235088ca"><div class="ttname"><a href="namespacearm__compute_1_1test_1_1validation.xhtml#a989ab3e96426615bb98e04e0235088ca">arm_compute::test::validation::src</a></div><div class="ttdeci">cast configure & src</div><div class="ttdef"><b>Definition:</b> <a href="_c_l_2_cast_8cpp_source.xhtml#l00169">Cast.cpp:169</a></div></div>
-<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a36f754c05b6fddf6df0d8d0a74f8159f"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a></div><div class="ttdeci">#define VEC_DATA_TYPE(type, size)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00073">helpers.h:73</a></div></div>
+<div class="ttc" id="src_2core_2_c_l_2cl__kernels_2_helpers_8h_xhtml_a36f754c05b6fddf6df0d8d0a74f8159f"><div class="ttname"><a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a></div><div class="ttdeci">#define VEC_DATA_TYPE(type, size)</div><div class="ttdef"><b>Definition:</b> <a href="src_2core_2_c_l_2cl__kernels_2_helpers_8h_source.xhtml#l00255">helpers.h:255</a></div></div>
</div><!-- fragment --></div><!-- contents -->
</div><!-- doc-content -->
<!-- start footer part -->
<div id="nav-path" class="navpath"><!-- id is needed for treeview function! -->
<ul>
<li class="navelem"><a class="el" href="dir_68267d1309a1af8e8297ef4c3efbcdba.xhtml">src</a></li><li class="navelem"><a class="el" href="dir_aebb8dcc11953d78e620bbef0b9e2183.xhtml">core</a></li><li class="navelem"><a class="el" href="dir_8c278f79c760e5c5fbd911f9870614c1.xhtml">CL</a></li><li class="navelem"><a class="el" href="dir_25885286e9dad4fa105b7b25a8031bbf.xhtml">cl_kernels</a></li><li class="navelem"><a class="el" href="direct__convolution1x1_8cl.xhtml">direct_convolution1x1.cl</a></li>
- <li class="footer">Generated on Mon Sep 2 2019 11:47:21 for Compute Library by
+ <li class="footer">Generated on Thu Nov 28 2019 16:52:57 for Compute Library by
<a href="http://www.doxygen.org/index.html">
<img class="footer" src="doxygen.png" alt="doxygen"/></a> 1.8.15 </li>
</ul>