arm_compute v17.12
diff --git a/documentation/gemv_8cl.xhtml b/documentation/gemv_8cl.xhtml
index a2ccd8c..557e4a0 100644
--- a/documentation/gemv_8cl.xhtml
+++ b/documentation/gemv_8cl.xhtml
@@ -4,7 +4,7 @@
<head>
<meta http-equiv="Content-Type" content="text/xhtml;charset=UTF-8"/>
<meta http-equiv="X-UA-Compatible" content="IE=9"/>
-<meta name="generator" content="Doxygen 1.8.6"/>
+<meta name="generator" content="Doxygen 1.8.11"/>
<meta name="robots" content="NOINDEX, NOFOLLOW" /> <!-- Prevent indexing by search engines -->
<title>Compute Library: src/core/CL/cl_kernels/gemv.cl File Reference</title>
<link href="tabs.css" rel="stylesheet" type="text/css"/>
@@ -12,22 +12,24 @@
<script type="text/javascript" src="dynsections.js"></script>
<link href="navtree.css" rel="stylesheet" type="text/css"/>
<script type="text/javascript" src="resize.js"></script>
+<script type="text/javascript" src="navtreedata.js"></script>
<script type="text/javascript" src="navtree.js"></script>
<script type="text/javascript">
$(document).ready(initResizable);
$(window).load(resizeHeight);
</script>
<link href="search/search.css" rel="stylesheet" type="text/css"/>
+<script type="text/javascript" src="search/searchdata.js"></script>
<script type="text/javascript" src="search/search.js"></script>
<script type="text/javascript">
- $(document).ready(function() { searchBox.OnSelectItem(0); });
+ $(document).ready(function() { init_search(); });
</script>
<script type="text/x-mathjax-config">
MathJax.Hub.Config({
extensions: ["tex2jax.js"],
jax: ["input/TeX","output/HTML-CSS"],
});
-</script><script src="http://cdn.mathjax.org/mathjax/latest/MathJax.js"></script>
+</script><script type="text/javascript" src="http://cdn.mathjax.org/mathjax/latest/MathJax.js"></script>
<link href="doxygen.css" rel="stylesheet" type="text/css" />
</head>
<body>
@@ -38,7 +40,7 @@
<tr style="height: 56px;">
<td style="padding-left: 0.5em;">
<div id="projectname">Compute Library
-  <span id="projectnumber">17.10</span>
+  <span id="projectnumber">17.12</span>
</div>
</td>
</tr>
@@ -46,7 +48,7 @@
</table>
</div>
<!-- end header part -->
-<!-- Generated by Doxygen 1.8.6 -->
+<!-- Generated by Doxygen 1.8.11 -->
<script type="text/javascript">
var searchBox = new SearchBox("searchBox", "search",false,'Search');
</script>
@@ -101,7 +103,7 @@
onmouseover="return searchBox.OnSearchSelectShow()"
onmouseout="return searchBox.OnSearchSelectHide()"
onkeydown="return searchBox.OnSearchSelectKey(event)">
-<a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(0)"><span class="SelectionMark"> </span>All</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(1)"><span class="SelectionMark"> </span>Data Structures</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(2)"><span class="SelectionMark"> </span>Namespaces</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(3)"><span class="SelectionMark"> </span>Files</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(4)"><span class="SelectionMark"> </span>Functions</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(5)"><span class="SelectionMark"> </span>Variables</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(6)"><span class="SelectionMark"> </span>Typedefs</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(7)"><span class="SelectionMark"> </span>Enumerations</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(8)"><span class="SelectionMark"> </span>Enumerator</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(9)"><span class="SelectionMark"> </span>Friends</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(10)"><span class="SelectionMark"> </span>Macros</a><a class="SelectItem" href="javascript:void(0)" onclick="searchBox.OnSelectItem(11)"><span class="SelectionMark"> </span>Pages</a></div>
+</div>
<!-- iframe showing the search results (closed by default) -->
<div id="MSearchResultsWindow">
@@ -117,14 +119,14 @@
<div class="title">gemv.cl File Reference</div> </div>
</div><!--header-->
<div class="contents">
-<div class="textblock"><code>#include "<a class="el" href="helpers_8h_source.xhtml">helpers.h</a>"</code><br/>
+<div class="textblock"><code>#include "<a class="el" href="helpers_8h_source.xhtml">helpers.h</a>"</code><br />
</div>
<p><a href="gemv_8cl_source.xhtml">Go to the source code of this file.</a></p>
<table class="memberdecls">
<tr class="heading"><td colspan="2"><h2 class="groupheader"><a name="func-members"></a>
Functions</h2></td></tr>
<tr class="memitem:a40788c7e9d25ce124c68ffd5dad44656"><td class="memItemLeft" align="right" valign="top">__kernel void </td><td class="memItemRight" valign="bottom"><a class="el" href="gemv_8cl.xhtml#a40788c7e9d25ce124c68ffd5dad44656">gemm_mv</a> (__global uchar *src_ptr, uint src_stride_x, uint src_step_x, uint src_stride_y, uint src_step_y, uint src_stride_z, uint src_step_z, uint src_offset_first_element_in_bytes, __global uchar *weights_ptr, uint weights_stride_x, uint weights_step_x, uint weights_stride_y, uint weights_step_y, uint weights_offset_first_element_in_bytes, __global uchar *dst_ptr, uint dst_stride_x, uint dst_step_x, uint dst_offset_first_element_in_bytes)</td></tr>
-<tr class="memdesc:a40788c7e9d25ce124c68ffd5dad44656"><td class="mdescLeft"> </td><td class="mdescRight">This kernel applies dot product to each plane on the input tensor and the corrispective column of the reshaped weight tensor. <a href="#a40788c7e9d25ce124c68ffd5dad44656">More...</a><br/></td></tr>
+<tr class="memdesc:a40788c7e9d25ce124c68ffd5dad44656"><td class="mdescLeft"> </td><td class="mdescRight">This kernel applies dot product to each plane on the input tensor and the corrispective column of the reshaped weight tensor. <a href="#a40788c7e9d25ce124c68ffd5dad44656">More...</a><br /></td></tr>
<tr class="separator:a40788c7e9d25ce124c68ffd5dad44656"><td class="memSeparator" colspan="2"> </td></tr>
</table>
<h2 class="groupheader">Function Documentation</h2>
@@ -260,7 +262,7 @@
<tr><td class="paramdir">[in]</td><td class="paramname">src_stride_z</td><td>Stride of the source tensor in Z dimension (in bytes) </td></tr>
<tr><td class="paramdir">[in]</td><td class="paramname">src_step_z</td><td>src_stride_z * number of elements along Y processed per workitem(in bytes) </td></tr>
<tr><td class="paramdir">[in]</td><td class="paramname">src_offset_first_element_in_bytes</td><td>The offset of the first element in the source tensor </td></tr>
- <tr><td class="paramdir">[out]</td><td class="paramname">weights_ptr</td><td>Pointer to the weights tensor. Same as <code>src_ptr</code> </td></tr>
+ <tr><td class="paramdir">[in]</td><td class="paramname">weights_ptr</td><td>Pointer to the weights tensor. Same as <code>src_ptr</code> </td></tr>
<tr><td class="paramdir">[in]</td><td class="paramname">weights_stride_x</td><td>Stride of the weights tensor in X dimension (in bytes) </td></tr>
<tr><td class="paramdir">[in]</td><td class="paramname">weights_step_x</td><td>weights_stride_x * number of elements along X processed per workitem(in bytes) </td></tr>
<tr><td class="paramdir">[in]</td><td class="paramname">weights_stride_y</td><td>Stride of the weights tensor in Y dimension (in bytes) </td></tr>
@@ -276,75 +278,13 @@
<p>Definition at line <a class="el" href="gemv_8cl_source.xhtml#l00049">49</a> of file <a class="el" href="gemv_8cl_source.xhtml">gemv.cl</a>.</p>
-<p>References <a class="el" href="helpers_8h_source.xhtml#l00111">CONVERT_TO_TENSOR3D_STRUCT</a>, <a class="el" href="convolution3x3_8cl_source.xhtml#l00027">DATA_TYPE</a>, <a class="el" href="helpers_8h_source.xhtml#l00295">offset()</a>, <a class="el" href="helpers_8h_source.xhtml#l00145">Tensor3D::ptr</a>, and <a class="el" href="fixed__point_8h_source.xhtml#l00093">VEC_DATA_TYPE</a>.</p>
-<div class="fragment"><div class="line"><a name="l00050"></a><span class="lineno"> 50</span> {</div>
-<div class="line"><a name="l00051"></a><span class="lineno"> 51</span>  <a class="code" href="struct_tensor3_d.xhtml">Tensor3D</a> src = <a class="code" href="helpers_8h.xhtml#a31c8c760f08fb1a331b16b7c204321dc">CONVERT_TO_TENSOR3D_STRUCT</a>(src);</div>
-<div class="line"><a name="l00052"></a><span class="lineno"> 52</span> </div>
-<div class="line"><a name="l00053"></a><span class="lineno"> 53</span>  <span class="keywordtype">int</span> y = get_global_id(1) * 4;</div>
-<div class="line"><a name="l00054"></a><span class="lineno"> 54</span>  <span class="keywordtype">int</span> z = get_global_id(2);</div>
-<div class="line"><a name="l00055"></a><span class="lineno"> 55</span> </div>
-<div class="line"><a name="l00056"></a><span class="lineno"> 56</span>  __global uchar *current_weights = weights_ptr + weights_offset_first_element_in_bytes + z * weights_stride_y;</div>
-<div class="line"><a name="l00057"></a><span class="lineno"> 57</span>  __global uchar *input_ptr = src.<a class="code" href="struct_tensor3_d.xhtml#acf52c23cbd7424606c10a606524e3e32">ptr</a>;</div>
-<div class="line"><a name="l00058"></a><span class="lineno"> 58</span> </div>
-<div class="line"><a name="l00059"></a><span class="lineno"> 59</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> acc0 = (<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>)0;</div>
-<div class="line"><a name="l00060"></a><span class="lineno"> 60</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> acc1 = (<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>)0;</div>
-<div class="line"><a name="l00061"></a><span class="lineno"> 61</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> acc2 = (<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>)0;</div>
-<div class="line"><a name="l00062"></a><span class="lineno"> 62</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> acc3 = (<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>)0;</div>
-<div class="line"><a name="l00063"></a><span class="lineno"> 63</span> </div>
-<div class="line"><a name="l00064"></a><span class="lineno"> 64</span>  <span class="comment">// This kernel handle 4 rows in per thread so that it can reuse the weights</span></div>
-<div class="line"><a name="l00065"></a><span class="lineno"> 65</span>  <span class="keywordflow">for</span>(<span class="keywordtype">int</span> i = 0; i < SRC_WIDTH; i += 4)</div>
-<div class="line"><a name="l00066"></a><span class="lineno"> 66</span>  {</div>
-<div class="line"><a name="l00067"></a><span class="lineno"> 67</span>  <a class="code" href="fixed__point_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="l00068"></a><span class="lineno"> 68</span>  weights = vload4(0, (__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(current_weights + i * weights_stride_x));</div>
-<div class="line"><a name="l00069"></a><span class="lineno"> 69</span> </div>
-<div class="line"><a name="l00070"></a><span class="lineno"> 70</span>  int4 <a class="code" href="helpers_8h.xhtml#a009469e4d9b8fce3b6d5e97d2077827d">offset</a> = (int4)i * (int4)src_stride_x + (int4)(0, 1, 2, 3) * (int4)src_stride_y;</div>
-<div class="line"><a name="l00071"></a><span class="lineno"> 71</span> </div>
-<div class="line"><a name="l00072"></a><span class="lineno"> 72</span>  <a class="code" href="fixed__point_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="l00073"></a><span class="lineno"> 73</span>  tmp0 = vload4(0, (__global DATA_TYPE *)(input_ptr + <a class="code" href="helpers_8h.xhtml#a009469e4d9b8fce3b6d5e97d2077827d">offset</a>.s0));</div>
-<div class="line"><a name="l00074"></a><span class="lineno"> 74</span>  <a class="code" href="fixed__point_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(DATA_TYPE, 4)</div>
-<div class="line"><a name="l00075"></a><span class="lineno"> 75</span>  tmp1 = vload4(0, (__global DATA_TYPE *)(input_ptr + <a class="code" href="helpers_8h.xhtml#a009469e4d9b8fce3b6d5e97d2077827d">offset</a>.s1));</div>
-<div class="line"><a name="l00076"></a><span class="lineno"> 76</span>  <a class="code" href="fixed__point_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(DATA_TYPE, 4)</div>
-<div class="line"><a name="l00077"></a><span class="lineno"> 77</span>  tmp2 = vload4(0, (__global DATA_TYPE *)(input_ptr + <a class="code" href="helpers_8h.xhtml#a009469e4d9b8fce3b6d5e97d2077827d">offset</a>.s2));</div>
-<div class="line"><a name="l00078"></a><span class="lineno"> 78</span>  <a class="code" href="fixed__point_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(DATA_TYPE, 4)</div>
-<div class="line"><a name="l00079"></a><span class="lineno"> 79</span>  tmp3 = vload4(0, (__global DATA_TYPE *)(input_ptr + <a class="code" href="helpers_8h.xhtml#a009469e4d9b8fce3b6d5e97d2077827d">offset</a>.s3));</div>
-<div class="line"><a name="l00080"></a><span class="lineno"> 80</span> </div>
-<div class="line"><a name="l00081"></a><span class="lineno"> 81</span>  acc0 += dot(weights, tmp0);</div>
-<div class="line"><a name="l00082"></a><span class="lineno"> 82</span>  acc1 += dot(weights, tmp1);</div>
-<div class="line"><a name="l00083"></a><span class="lineno"> 83</span>  acc2 += dot(weights, tmp2);</div>
-<div class="line"><a name="l00084"></a><span class="lineno"> 84</span>  acc3 += dot(weights, tmp3);</div>
-<div class="line"><a name="l00085"></a><span class="lineno"> 85</span>  }</div>
-<div class="line"><a name="l00086"></a><span class="lineno"> 86</span> </div>
-<div class="line"><a name="l00087"></a><span class="lineno"> 87</span>  __global uchar *output_ptr = dst_ptr + dst_offset_first_element_in_bytes + (y + z * SRC_HEIGHT) * dst_stride_x;</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="keywordtype">int</span> rows_left = SRC_HEIGHT - (y + 4);</div>
-<div class="line"><a name="l00090"></a><span class="lineno"> 90</span> </div>
-<div class="line"><a name="l00091"></a><span class="lineno"> 91</span>  <span class="comment">// This if check is used to handle the last few rows when it can't be divided by the four</span></div>
-<div class="line"><a name="l00092"></a><span class="lineno"> 92</span>  if(rows_left >= 0)</div>
-<div class="line"><a name="l00093"></a><span class="lineno"> 93</span>  {</div>
-<div class="line"><a name="l00094"></a><span class="lineno"> 94</span>  <a class="code" href="fixed__point_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(DATA_TYPE, 4)</div>
-<div class="line"><a name="l00095"></a><span class="lineno"> 95</span>  out = (<a class="code" href="fixed__point_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(DATA_TYPE, 4))(acc0, acc1, acc2, acc3);</div>
-<div class="line"><a name="l00096"></a><span class="lineno"> 96</span>  vstore4(out, 0, (__global DATA_TYPE *)output_ptr);</div>
-<div class="line"><a name="l00097"></a><span class="lineno"> 97</span>  }</div>
-<div class="line"><a name="l00098"></a><span class="lineno"> 98</span>  else</div>
-<div class="line"><a name="l00099"></a><span class="lineno"> 99</span>  {</div>
-<div class="line"><a name="l00100"></a><span class="lineno"> 100</span>  <span class="keywordflow">switch</span>(rows_left)</div>
-<div class="line"><a name="l00101"></a><span class="lineno"> 101</span>  {</div>
-<div class="line"><a name="l00102"></a><span class="lineno"> 102</span>  <span class="keywordflow">case</span> -1: <span class="comment">// three rows left; one is padding</span></div>
-<div class="line"><a name="l00103"></a><span class="lineno"> 103</span>  *((__global DATA_TYPE *)(output_ptr + 2 * dst_stride_x)) = acc2;</div>
-<div class="line"><a name="l00104"></a><span class="lineno"> 104</span>  <span class="keywordflow">case</span> -2: <span class="comment">// two rows left; two are padding</span></div>
-<div class="line"><a name="l00105"></a><span class="lineno"> 105</span>  *((__global DATA_TYPE *)(output_ptr + 1 * dst_stride_x)) = acc1;</div>
-<div class="line"><a name="l00106"></a><span class="lineno"> 106</span>  <span class="keywordflow">case</span> -3: <span class="comment">// one row left; three are padding</span></div>
-<div class="line"><a name="l00107"></a><span class="lineno"> 107</span>  *((__global DATA_TYPE *)(output_ptr + 0 * dst_stride_x)) = acc0;</div>
-<div class="line"><a name="l00108"></a><span class="lineno"> 108</span>  <span class="keywordflow">break</span>;</div>
-<div class="line"><a name="l00109"></a><span class="lineno"> 109</span>  }</div>
-<div class="line"><a name="l00110"></a><span class="lineno"> 110</span>  }</div>
-<div class="line"><a name="l00111"></a><span class="lineno"> 111</span> }</div>
-<div class="ttc" id="helpers_8h_xhtml_a31c8c760f08fb1a331b16b7c204321dc"><div class="ttname"><a href="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="helpers_8h_source.xhtml#l00111">helpers.h:111</a></div></div>
+<p>References <a class="el" href="helpers_8h_source.xhtml#l00117">CONVERT_TO_TENSOR3D_STRUCT</a>, <a class="el" href="convolution3x3_8cl_source.xhtml#l00027">DATA_TYPE</a>, <a class="el" href="helpers_8h_source.xhtml#l00301">offset()</a>, <a class="el" href="helpers_8h_source.xhtml#l00151">Tensor3D::ptr</a>, and <a class="el" href="fixed__point_8h_source.xhtml#l00093">VEC_DATA_TYPE</a>.</p>
+<div class="fragment"><div class="line"><a name="l00050"></a><span class="lineno"> 50</span> {</div><div class="line"><a name="l00051"></a><span class="lineno"> 51</span>  <a class="code" href="struct_tensor3_d.xhtml">Tensor3D</a> src = <a class="code" href="helpers_8h.xhtml#a31c8c760f08fb1a331b16b7c204321dc">CONVERT_TO_TENSOR3D_STRUCT</a>(src);</div><div class="line"><a name="l00052"></a><span class="lineno"> 52</span> </div><div class="line"><a name="l00053"></a><span class="lineno"> 53</span>  <span class="keywordtype">int</span> y = get_global_id(1) * 4;</div><div class="line"><a name="l00054"></a><span class="lineno"> 54</span>  <span class="keywordtype">int</span> z = get_global_id(2);</div><div class="line"><a name="l00055"></a><span class="lineno"> 55</span> </div><div class="line"><a name="l00056"></a><span class="lineno"> 56</span>  __global uchar *current_weights = weights_ptr + weights_offset_first_element_in_bytes + z * weights_stride_y;</div><div class="line"><a name="l00057"></a><span class="lineno"> 57</span>  __global uchar *input_ptr = src.<a class="code" href="struct_tensor3_d.xhtml#acf52c23cbd7424606c10a606524e3e32">ptr</a>;</div><div class="line"><a name="l00058"></a><span class="lineno"> 58</span> </div><div class="line"><a name="l00059"></a><span class="lineno"> 59</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> acc0 = (<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>)0;</div><div class="line"><a name="l00060"></a><span class="lineno"> 60</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> acc1 = (<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>)0;</div><div class="line"><a name="l00061"></a><span class="lineno"> 61</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> acc2 = (<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>)0;</div><div class="line"><a name="l00062"></a><span class="lineno"> 62</span>  <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> acc3 = (<a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a>)0;</div><div class="line"><a name="l00063"></a><span class="lineno"> 63</span> </div><div class="line"><a name="l00064"></a><span class="lineno"> 64</span>  <span class="comment">// This kernel handle 4 rows in per thread so that it can reuse the weights</span></div><div class="line"><a name="l00065"></a><span class="lineno"> 65</span>  <span class="keywordflow">for</span>(<span class="keywordtype">int</span> i = 0; i < SRC_WIDTH; i += 4)</div><div class="line"><a name="l00066"></a><span class="lineno"> 66</span>  {</div><div class="line"><a name="l00067"></a><span class="lineno"> 67</span>  <a class="code" href="fixed__point_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="l00068"></a><span class="lineno"> 68</span>  weights = vload4(0, (__global <a class="code" href="convolution3x3_8cl.xhtml#afb8c72ce35c4a1f4a2588d6573e54aa1">DATA_TYPE</a> *)(current_weights + i * weights_stride_x));</div><div class="line"><a name="l00069"></a><span class="lineno"> 69</span> </div><div class="line"><a name="l00070"></a><span class="lineno"> 70</span>  int4 <a class="code" href="helpers_8h.xhtml#a009469e4d9b8fce3b6d5e97d2077827d">offset</a> = (int4)i * (int4)src_stride_x + (int4)(0, 1, 2, 3) * (int4)src_stride_y;</div><div class="line"><a name="l00071"></a><span class="lineno"> 71</span> </div><div class="line"><a name="l00072"></a><span class="lineno"> 72</span>  <a class="code" href="fixed__point_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="l00073"></a><span class="lineno"> 73</span>  tmp0 = vload4(0, (__global DATA_TYPE *)(input_ptr + <a class="code" href="helpers_8h.xhtml#a009469e4d9b8fce3b6d5e97d2077827d">offset</a>.s0));</div><div class="line"><a name="l00074"></a><span class="lineno"> 74</span>  <a class="code" href="fixed__point_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(DATA_TYPE, 4)</div><div class="line"><a name="l00075"></a><span class="lineno"> 75</span>  tmp1 = vload4(0, (__global DATA_TYPE *)(input_ptr + <a class="code" href="helpers_8h.xhtml#a009469e4d9b8fce3b6d5e97d2077827d">offset</a>.s1));</div><div class="line"><a name="l00076"></a><span class="lineno"> 76</span>  <a class="code" href="fixed__point_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(DATA_TYPE, 4)</div><div class="line"><a name="l00077"></a><span class="lineno"> 77</span>  tmp2 = vload4(0, (__global DATA_TYPE *)(input_ptr + <a class="code" href="helpers_8h.xhtml#a009469e4d9b8fce3b6d5e97d2077827d">offset</a>.s2));</div><div class="line"><a name="l00078"></a><span class="lineno"> 78</span>  <a class="code" href="fixed__point_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(DATA_TYPE, 4)</div><div class="line"><a name="l00079"></a><span class="lineno"> 79</span>  tmp3 = vload4(0, (__global DATA_TYPE *)(input_ptr + <a class="code" href="helpers_8h.xhtml#a009469e4d9b8fce3b6d5e97d2077827d">offset</a>.s3));</div><div class="line"><a name="l00080"></a><span class="lineno"> 80</span> </div><div class="line"><a name="l00081"></a><span class="lineno"> 81</span>  acc0 += dot(weights, tmp0);</div><div class="line"><a name="l00082"></a><span class="lineno"> 82</span>  acc1 += dot(weights, tmp1);</div><div class="line"><a name="l00083"></a><span class="lineno"> 83</span>  acc2 += dot(weights, tmp2);</div><div class="line"><a name="l00084"></a><span class="lineno"> 84</span>  acc3 += dot(weights, tmp3);</div><div class="line"><a name="l00085"></a><span class="lineno"> 85</span>  }</div><div class="line"><a name="l00086"></a><span class="lineno"> 86</span> </div><div class="line"><a name="l00087"></a><span class="lineno"> 87</span>  __global uchar *output_ptr = dst_ptr + dst_offset_first_element_in_bytes + (y + z * SRC_HEIGHT) * dst_stride_x;</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="keywordtype">int</span> rows_left = SRC_HEIGHT - (y + 4);</div><div class="line"><a name="l00090"></a><span class="lineno"> 90</span> </div><div class="line"><a name="l00091"></a><span class="lineno"> 91</span>  <span class="comment">// This if check is used to handle the last few rows when it can't be divided by the four</span></div><div class="line"><a name="l00092"></a><span class="lineno"> 92</span>  if(rows_left >= 0)</div><div class="line"><a name="l00093"></a><span class="lineno"> 93</span>  {</div><div class="line"><a name="l00094"></a><span class="lineno"> 94</span>  <a class="code" href="fixed__point_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(DATA_TYPE, 4)</div><div class="line"><a name="l00095"></a><span class="lineno"> 95</span>  out = (<a class="code" href="fixed__point_8h.xhtml#a36f754c05b6fddf6df0d8d0a74f8159f">VEC_DATA_TYPE</a>(DATA_TYPE, 4))(acc0, acc1, acc2, acc3);</div><div class="line"><a name="l00096"></a><span class="lineno"> 96</span>  vstore4(out, 0, (__global DATA_TYPE *)output_ptr);</div><div class="line"><a name="l00097"></a><span class="lineno"> 97</span>  }</div><div class="line"><a name="l00098"></a><span class="lineno"> 98</span>  else</div><div class="line"><a name="l00099"></a><span class="lineno"> 99</span>  {</div><div class="line"><a name="l00100"></a><span class="lineno"> 100</span>  <span class="keywordflow">switch</span>(rows_left)</div><div class="line"><a name="l00101"></a><span class="lineno"> 101</span>  {</div><div class="line"><a name="l00102"></a><span class="lineno"> 102</span>  <span class="keywordflow">case</span> -1: <span class="comment">// three rows left; one is padding</span></div><div class="line"><a name="l00103"></a><span class="lineno"> 103</span>  *((__global DATA_TYPE *)(output_ptr + 2 * dst_stride_x)) = acc2;</div><div class="line"><a name="l00104"></a><span class="lineno"> 104</span>  <span class="keywordflow">case</span> -2: <span class="comment">// two rows left; two are padding</span></div><div class="line"><a name="l00105"></a><span class="lineno"> 105</span>  *((__global DATA_TYPE *)(output_ptr + 1 * dst_stride_x)) = acc1;</div><div class="line"><a name="l00106"></a><span class="lineno"> 106</span>  <span class="keywordflow">case</span> -3: <span class="comment">// one row left; three are padding</span></div><div class="line"><a name="l00107"></a><span class="lineno"> 107</span>  *((__global DATA_TYPE *)(output_ptr + 0 * dst_stride_x)) = acc0;</div><div class="line"><a name="l00108"></a><span class="lineno"> 108</span>  <span class="keywordflow">break</span>;</div><div class="line"><a name="l00109"></a><span class="lineno"> 109</span>  }</div><div class="line"><a name="l00110"></a><span class="lineno"> 110</span>  }</div><div class="line"><a name="l00111"></a><span class="lineno"> 111</span> }</div><div class="ttc" id="helpers_8h_xhtml_a31c8c760f08fb1a331b16b7c204321dc"><div class="ttname"><a href="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="helpers_8h_source.xhtml#l00117">helpers.h:117</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="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="helpers_8h_source.xhtml#l00143">helpers.h:143</a></div></div>
-<div class="ttc" id="helpers_8h_xhtml_a009469e4d9b8fce3b6d5e97d2077827d"><div class="ttname"><a href="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="helpers_8h_source.xhtml#l00295">helpers.h:295</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="helpers_8h_source.xhtml#l00149">helpers.h:149</a></div></div>
+<div class="ttc" id="helpers_8h_xhtml_a009469e4d9b8fce3b6d5e97d2077827d"><div class="ttname"><a href="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="helpers_8h_source.xhtml#l00301">helpers.h:301</a></div></div>
<div class="ttc" id="fixed__point_8h_xhtml_a36f754c05b6fddf6df0d8d0a74f8159f"><div class="ttname"><a href="fixed__point_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="fixed__point_8h_source.xhtml#l00093">fixed_point.h:93</a></div></div>
-<div class="ttc" id="struct_tensor3_d_xhtml_acf52c23cbd7424606c10a606524e3e32"><div class="ttname"><a href="struct_tensor3_d.xhtml#acf52c23cbd7424606c10a606524e3e32">Tensor3D::ptr</a></div><div class="ttdeci">__global uchar * ptr</div><div class="ttdoc">Pointer to the starting postion of the buffer. </div><div class="ttdef"><b>Definition:</b> <a href="helpers_8h_source.xhtml#l00145">helpers.h:145</a></div></div>
+<div class="ttc" id="struct_tensor3_d_xhtml_acf52c23cbd7424606c10a606524e3e32"><div class="ttname"><a href="struct_tensor3_d.xhtml#acf52c23cbd7424606c10a606524e3e32">Tensor3D::ptr</a></div><div class="ttdeci">__global uchar * ptr</div><div class="ttdoc">Pointer to the starting postion of the buffer. </div><div class="ttdef"><b>Definition:</b> <a href="helpers_8h_source.xhtml#l00151">helpers.h:151</a></div></div>
</div><!-- fragment -->
</div>
</div>
@@ -353,10 +293,10 @@
<!-- start footer part -->
<div id="nav-path" class="navpath"><!-- id is needed for treeview function! -->
<ul>
- <li class="navelem"><a class="el" href="dir_55b5a5006e943fb664ff8cff3cfe7768.xhtml">src</a></li><li class="navelem"><a class="el" href="dir_f6c3ae42d7e27145d0a3c3e38ca1c000.xhtml">core</a></li><li class="navelem"><a class="el" href="dir_1b7ae9123c88e650426d50d95c19d414.xhtml">CL</a></li><li class="navelem"><a class="el" href="dir_b7c3d80b0dadb2139bc73b7161751c12.xhtml">cl_kernels</a></li><li class="navelem"><a class="el" href="gemv_8cl.xhtml">gemv.cl</a></li>
- <li class="footer">Generated on Thu Oct 12 2017 14:26:35 for Compute Library by
+ <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="gemv_8cl.xhtml">gemv.cl</a></li>
+ <li class="footer">Generated on Thu Dec 14 2017 23:48:34 for Compute Library by
<a href="http://www.doxygen.org/index.html">
- <img class="footer" src="doxygen.png" alt="doxygen"/></a> 1.8.6 </li>
+ <img class="footer" src="doxygen.png" alt="doxygen"/></a> 1.8.11 </li>
</ul>
</div>
</body>