arm_compute v19.11
diff --git a/documentation/namespacearm__gemm.xhtml b/documentation/namespacearm__gemm.xhtml
index 882b885..46c2de2 100644
--- a/documentation/namespacearm__gemm.xhtml
+++ b/documentation/namespacearm__gemm.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>
@@ -157,42 +157,48 @@
</table><table class="memberdecls">
<tr class="heading"><td colspan="2"><h2 class="groupheader"><a name="func-members"></a>
Functions</h2></td></tr>
+<tr class="memitem:a7450234a5b6959d08a3ddac191b3fe1f"><td class="memTemplParams" colspan="2">template<typename T > </td></tr>
+<tr class="memitem:a7450234a5b6959d08a3ddac191b3fe1f"><td class="memTemplItemLeft" align="right" valign="top">void </td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#a7450234a5b6959d08a3ddac191b3fe1f">bias_adder</a> (T *out, unsigned int stride, const T *bias, unsigned int rows, unsigned int cols)</td></tr>
+<tr class="separator:a7450234a5b6959d08a3ddac191b3fe1f"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:ad5675a7992a4e9eb9b2bc67e5715146d"><td class="memTemplParams" colspan="2">template<bool DoBias, typename T > </td></tr>
+<tr class="memitem:ad5675a7992a4e9eb9b2bc67e5715146d"><td class="memTemplItemLeft" align="right" valign="top">void </td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#ad5675a7992a4e9eb9b2bc67e5715146d">activator</a> (T *out, unsigned int stride, const T *bias, Activation act, unsigned int rows, unsigned int cols)</td></tr>
+<tr class="separator:ad5675a7992a4e9eb9b2bc67e5715146d"><td class="memSeparator" colspan="2"> </td></tr>
<tr class="memitem:a60b1162856ab084a3588fc2d5fdec539"><td class="memTemplParams" colspan="2">template<> </td></tr>
<tr class="memitem:a60b1162856ab084a3588fc2d5fdec539"><td class="memTemplItemLeft" align="right" valign="top">const <a class="el" href="structarm__gemm_1_1_gemm_implementation.xhtml">GemmImplementation</a>< float, float > * </td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#a60b1162856ab084a3588fc2d5fdec539">gemm_implementation_list< float, float ></a> ()</td></tr>
<tr class="separator:a60b1162856ab084a3588fc2d5fdec539"><td class="memSeparator" colspan="2"> </td></tr>
-<tr class="memitem:ab98c2935f5bb60c9a4add6c3e9de53e4"><td class="memItemLeft" align="right" valign="top">template UniqueGemmCommon< float, float > </td><td class="memItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#ab98c2935f5bb60c9a4add6c3e9de53e4">gemm< float, float, Nothing ></a> (const GemmArgs< float > &args, const Nothing &)</td></tr>
-<tr class="separator:ab98c2935f5bb60c9a4add6c3e9de53e4"><td class="memSeparator" colspan="2"> </td></tr>
-<tr class="memitem:abd013457d4280a12045a79c201bf25e0"><td class="memItemLeft" align="right" valign="top">template KernelDescription </td><td class="memItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#abd013457d4280a12045a79c201bf25e0">get_gemm_method< float, float, Nothing ></a> (const GemmArgs< float > &args, const Nothing &)</td></tr>
-<tr class="separator:abd013457d4280a12045a79c201bf25e0"><td class="memSeparator" colspan="2"> </td></tr>
-<tr class="memitem:a9194360a5ba6924c83f0269ade8608f5"><td class="memItemLeft" align="right" valign="top">template std::vector< KernelDescription > </td><td class="memItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#a9194360a5ba6924c83f0269ade8608f5">get_compatible_kernels< float, float, Nothing ></a> (const GemmArgs< float > &args, const Nothing &)</td></tr>
-<tr class="separator:a9194360a5ba6924c83f0269ade8608f5"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:a4fb5776f35967c7173f432cb096de217"><td class="memItemLeft" align="right" valign="top">template UniqueGemmCommon< float, float > </td><td class="memItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#a4fb5776f35967c7173f432cb096de217">gemm< float, float, Nothing ></a> (const GemmArgs &args, const Nothing &)</td></tr>
+<tr class="separator:a4fb5776f35967c7173f432cb096de217"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:aeaa8a3f397520dca43d1dabb3774e2c9"><td class="memItemLeft" align="right" valign="top">template KernelDescription </td><td class="memItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#aeaa8a3f397520dca43d1dabb3774e2c9">get_gemm_method< float, float, Nothing ></a> (const GemmArgs &args, const Nothing &)</td></tr>
+<tr class="separator:aeaa8a3f397520dca43d1dabb3774e2c9"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:a4ac39db987b3fbd8e8a6e971961b41f4"><td class="memItemLeft" align="right" valign="top">template std::vector< KernelDescription > </td><td class="memItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#a4ac39db987b3fbd8e8a6e971961b41f4">get_compatible_kernels< float, float, Nothing ></a> (const GemmArgs &args, const Nothing &)</td></tr>
+<tr class="separator:a4ac39db987b3fbd8e8a6e971961b41f4"><td class="memSeparator" colspan="2"> </td></tr>
<tr class="memitem:aafbf874fc03cb6d5e6d79e93f5dc1d76"><td class="memTemplParams" colspan="2">template<typename Top , typename Tret , class OutputStage = Nothing> </td></tr>
<tr class="memitem:aafbf874fc03cb6d5e6d79e93f5dc1d76"><td class="memTemplItemLeft" align="right" valign="top">const <a class="el" href="structarm__gemm_1_1_gemm_implementation.xhtml">GemmImplementation</a>< Top, Tret, OutputStage > * </td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#aafbf874fc03cb6d5e6d79e93f5dc1d76">gemm_implementation_list</a> ()</td></tr>
<tr class="separator:aafbf874fc03cb6d5e6d79e93f5dc1d76"><td class="memSeparator" colspan="2"> </td></tr>
-<tr class="memitem:a7af0b0e9e111e3ec37682c89ee803d82"><td class="memTemplParams" colspan="2">template<typename Top , typename Tret , class OutputStage > </td></tr>
-<tr class="memitem:a7af0b0e9e111e3ec37682c89ee803d82"><td class="memTemplItemLeft" align="right" valign="top">bool </td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#a7af0b0e9e111e3ec37682c89ee803d82">find_implementation</a> (const GemmArgs< Tret > &args, const OutputStage &os, const <a class="el" href="structarm__gemm_1_1_gemm_implementation.xhtml">GemmImplementation</a>< Top, Tret, OutputStage > *&impl)</td></tr>
-<tr class="separator:a7af0b0e9e111e3ec37682c89ee803d82"><td class="memSeparator" colspan="2"> </td></tr>
-<tr class="memitem:a8eb5d997d45efbbdaa4374f45e541d86"><td class="memTemplParams" colspan="2">template<typename Top , typename Tret , class OutputStage > </td></tr>
-<tr class="memitem:a8eb5d997d45efbbdaa4374f45e541d86"><td class="memTemplItemLeft" align="right" valign="top">std::vector< KernelDescription > </td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#a8eb5d997d45efbbdaa4374f45e541d86">get_compatible_kernels</a> (const GemmArgs< Tret > &args, const OutputStage &os)</td></tr>
-<tr class="separator:a8eb5d997d45efbbdaa4374f45e541d86"><td class="memSeparator" colspan="2"> </td></tr>
-<tr class="memitem:a3cf8a3a8f4c31d116db5013daa6d83d9"><td class="memTemplParams" colspan="2">template<typename Top , typename Tret , class OutputStage > </td></tr>
-<tr class="memitem:a3cf8a3a8f4c31d116db5013daa6d83d9"><td class="memTemplItemLeft" align="right" valign="top">UniqueGemmCommon< Top, Tret > </td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#a3cf8a3a8f4c31d116db5013daa6d83d9">gemm</a> (const GemmArgs< Tret > &args, const OutputStage &os)</td></tr>
-<tr class="separator:a3cf8a3a8f4c31d116db5013daa6d83d9"><td class="memSeparator" colspan="2"> </td></tr>
-<tr class="memitem:a777476ec2693309861bef88dbef2d23f"><td class="memTemplParams" colspan="2">template<typename Top , typename Tret , class OutputStage > </td></tr>
-<tr class="memitem:a777476ec2693309861bef88dbef2d23f"><td class="memTemplItemLeft" align="right" valign="top">KernelDescription </td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#a777476ec2693309861bef88dbef2d23f">get_gemm_method</a> (const GemmArgs< Tret > &args, const OutputStage &os)</td></tr>
-<tr class="separator:a777476ec2693309861bef88dbef2d23f"><td class="memSeparator" colspan="2"> </td></tr>
-<tr class="memitem:a4c160021f8ceaa70922f23cafd6b3c99"><td class="memTemplParams" colspan="2">template<unsigned int twidth, unsigned int height, bool sve = false, typename Tin , typename Tout > </td></tr>
-<tr class="memitem:a4c160021f8ceaa70922f23cafd6b3c99"><td class="memTemplItemLeft" align="right" valign="top">void </td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#a4c160021f8ceaa70922f23cafd6b3c99">MergeResults</a> (Tout *out, const Tin *in, int ldc, int y0, int ymax, int x0, int xmax, const Tout alpha, const Tout beta)</td></tr>
-<tr class="separator:a4c160021f8ceaa70922f23cafd6b3c99"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:adfe29ef1042864157b1fd4c72cb21fa0"><td class="memTemplParams" colspan="2">template<typename Top , typename Tret , class OutputStage > </td></tr>
+<tr class="memitem:adfe29ef1042864157b1fd4c72cb21fa0"><td class="memTemplItemLeft" align="right" valign="top">bool </td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#adfe29ef1042864157b1fd4c72cb21fa0">find_implementation</a> (const GemmArgs &args, const OutputStage &os, const <a class="el" href="structarm__gemm_1_1_gemm_implementation.xhtml">GemmImplementation</a>< Top, Tret, OutputStage > *&impl)</td></tr>
+<tr class="separator:adfe29ef1042864157b1fd4c72cb21fa0"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:abce3ed84b7a8f5c6e45809ca1089b1cf"><td class="memTemplParams" colspan="2">template<typename Top , typename Tret , class OutputStage > </td></tr>
+<tr class="memitem:abce3ed84b7a8f5c6e45809ca1089b1cf"><td class="memTemplItemLeft" align="right" valign="top">std::vector< KernelDescription > </td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#abce3ed84b7a8f5c6e45809ca1089b1cf">get_compatible_kernels</a> (const GemmArgs &args, const OutputStage &os)</td></tr>
+<tr class="separator:abce3ed84b7a8f5c6e45809ca1089b1cf"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:a64b2f658fb34f894449e0279747a6390"><td class="memTemplParams" colspan="2">template<typename Top , typename Tret , class OutputStage > </td></tr>
+<tr class="memitem:a64b2f658fb34f894449e0279747a6390"><td class="memTemplItemLeft" align="right" valign="top">UniqueGemmCommon< Top, Tret > </td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#a64b2f658fb34f894449e0279747a6390">gemm</a> (const GemmArgs &args, const OutputStage &os)</td></tr>
+<tr class="separator:a64b2f658fb34f894449e0279747a6390"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:ad35cfe7c388edeac570fb9e01180dda6"><td class="memTemplParams" colspan="2">template<typename Top , typename Tret , class OutputStage > </td></tr>
+<tr class="memitem:ad35cfe7c388edeac570fb9e01180dda6"><td class="memTemplItemLeft" align="right" valign="top">KernelDescription </td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#ad35cfe7c388edeac570fb9e01180dda6">get_gemm_method</a> (const GemmArgs &args, const OutputStage &os)</td></tr>
+<tr class="separator:ad35cfe7c388edeac570fb9e01180dda6"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:aa2ed1c3595c8114064e8170b00df37c5"><td class="memTemplParams" colspan="2">template<unsigned int twidth, unsigned int height, bool sve = false, typename Tin , typename Tout > </td></tr>
+<tr class="memitem:aa2ed1c3595c8114064e8170b00df37c5"><td class="memTemplItemLeft" align="right" valign="top">void </td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#aa2ed1c3595c8114064e8170b00df37c5">MergeResults</a> (Tout *out, const Tin *in, int ldc, int y0, int ymax, int x0, int xmax, const Tout *bias, Activation act, bool append)</td></tr>
+<tr class="separator:aa2ed1c3595c8114064e8170b00df37c5"><td class="memSeparator" colspan="2"> </td></tr>
<tr class="memitem:a5367d66fa8e4648db3f690e9cfa7d498"><td class="memTemplParams" colspan="2">template<typename Tin , typename Tout > </td></tr>
<tr class="memitem:a5367d66fa8e4648db3f690e9cfa7d498"><td class="memTemplItemLeft" align="right" valign="top">void </td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#a5367d66fa8e4648db3f690e9cfa7d498">requantize_block_32</a> (const ARequantizeLayer32 &qp, unsigned int width, unsigned int height, const Tin *input, unsigned int in_stride, Tout *output, unsigned int out_stride, const int32_t *row_bias, const int32_t *col_bias)</td></tr>
<tr class="separator:a5367d66fa8e4648db3f690e9cfa7d498"><td class="memSeparator" colspan="2"> </td></tr>
<tr class="memitem:a79d7971bf26ac0d4df7662a47d65d068"><td class="memTemplParams" colspan="2">template<typename T > </td></tr>
<tr class="memitem:a79d7971bf26ac0d4df7662a47d65d068"><td class="memTemplItemLeft" align="right" valign="top">void </td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#a79d7971bf26ac0d4df7662a47d65d068">compute_row_sums</a> (const ARequantizeLayer32 &qp, unsigned int width, unsigned int height, const T *input, unsigned int in_stride, int32_t *row_bias)</td></tr>
<tr class="separator:a79d7971bf26ac0d4df7662a47d65d068"><td class="memSeparator" colspan="2"> </td></tr>
-<tr class="memitem:ac6ac0fe47fd694fe6d6dbba1ce6e9aac"><td class="memTemplParams" colspan="2">template<typename T > </td></tr>
-<tr class="memitem:ac6ac0fe47fd694fe6d6dbba1ce6e9aac"><td class="memTemplItemLeft" align="right" valign="top">void </td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#ac6ac0fe47fd694fe6d6dbba1ce6e9aac">compute_col_sums</a> (const ARequantizeLayer32 &qp, unsigned int width, unsigned int height, const T *input, unsigned int in_stride, int32_t *col_bias, unsigned int depth, unsigned int first_col)</td></tr>
-<tr class="separator:ac6ac0fe47fd694fe6d6dbba1ce6e9aac"><td class="memSeparator" colspan="2"> </td></tr>
+<tr class="memitem:af0b4b7759cb0829cb2a54e73b1555a01"><td class="memTemplParams" colspan="2">template<typename T > </td></tr>
+<tr class="memitem:af0b4b7759cb0829cb2a54e73b1555a01"><td class="memTemplItemLeft" align="right" valign="top">void </td><td class="memTemplItemRight" valign="bottom"><a class="el" href="namespacearm__gemm.xhtml#af0b4b7759cb0829cb2a54e73b1555a01">compute_col_sums</a> (const ARequantizeLayer32 &qp, unsigned int width, unsigned int height, const T *input, unsigned int in_stride, int32_t *col_bias, unsigned int depth, unsigned int multi, unsigned int first_col)</td></tr>
+<tr class="separator:af0b4b7759cb0829cb2a54e73b1555a01"><td class="memSeparator" colspan="2"> </td></tr>
</table><table class="memberdecls">
<tr class="heading"><td colspan="2"><h2 class="groupheader"><a name="var-members"></a>
Variables</h2></td></tr>
@@ -233,8 +239,135 @@
</div>
</div>
<h2 class="groupheader">Function Documentation</h2>
-<a id="ac6ac0fe47fd694fe6d6dbba1ce6e9aac"></a>
-<h2 class="memtitle"><span class="permalink"><a href="#ac6ac0fe47fd694fe6d6dbba1ce6e9aac">◆ </a></span>compute_col_sums()</h2>
+<a id="ad5675a7992a4e9eb9b2bc67e5715146d"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#ad5675a7992a4e9eb9b2bc67e5715146d">◆ </a></span>activator()</h2>
+
+<div class="memitem">
+<div class="memproto">
+<table class="mlabels">
+ <tr>
+ <td class="mlabels-left">
+ <table class="memname">
+ <tr>
+ <td class="memname">void arm_gemm::activator </td>
+ <td>(</td>
+ <td class="paramtype">T * </td>
+ <td class="paramname"><em>out</em>, </td>
+ </tr>
+ <tr>
+ <td class="paramkey"></td>
+ <td></td>
+ <td class="paramtype">unsigned int </td>
+ <td class="paramname"><em>stride</em>, </td>
+ </tr>
+ <tr>
+ <td class="paramkey"></td>
+ <td></td>
+ <td class="paramtype">const T * </td>
+ <td class="paramname"><em>bias</em>, </td>
+ </tr>
+ <tr>
+ <td class="paramkey"></td>
+ <td></td>
+ <td class="paramtype">Activation </td>
+ <td class="paramname"><em>act</em>, </td>
+ </tr>
+ <tr>
+ <td class="paramkey"></td>
+ <td></td>
+ <td class="paramtype">unsigned int </td>
+ <td class="paramname"><em>rows</em>, </td>
+ </tr>
+ <tr>
+ <td class="paramkey"></td>
+ <td></td>
+ <td class="paramtype">unsigned int </td>
+ <td class="paramname"><em>cols</em> </td>
+ </tr>
+ <tr>
+ <td></td>
+ <td>)</td>
+ <td></td><td></td>
+ </tr>
+ </table>
+ </td>
+ <td class="mlabels-right">
+<span class="mlabels"><span class="mlabel">inline</span></span> </td>
+ </tr>
+</table>
+</div><div class="memdoc">
+
+<p class="definition">Definition at line <a class="el" href="bias__adder_8hpp_source.xhtml#l00040">40</a> of file <a class="el" href="bias__adder_8hpp_source.xhtml">bias_adder.hpp</a>.</p>
+<div class="fragment"><div class="line"><a name="l00040"></a><span class="lineno"> 40</span>  {</div><div class="line"><a name="l00041"></a><span class="lineno"> 41</span>  <span class="keywordflow">if</span> (act.type == Activation::Type::None) {</div><div class="line"><a name="l00042"></a><span class="lineno"> 42</span>  <span class="keywordflow">if</span> (DoBias) {</div><div class="line"><a name="l00043"></a><span class="lineno"> 43</span>  <a class="code" href="namespacearm__gemm.xhtml#a7450234a5b6959d08a3ddac191b3fe1f">bias_adder</a>(out, stride, <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>, rows, cols);</div><div class="line"><a name="l00044"></a><span class="lineno"> 44</span>  }</div><div class="line"><a name="l00045"></a><span class="lineno"> 45</span>  <span class="keywordflow">return</span>;</div><div class="line"><a name="l00046"></a><span class="lineno"> 46</span>  }</div><div class="line"><a name="l00047"></a><span class="lineno"> 47</span> </div><div class="line"><a name="l00048"></a><span class="lineno"> 48</span>  <span class="keywordflow">if</span> (act.type == Activation::Type::ReLU) {</div><div class="line"><a name="l00049"></a><span class="lineno"> 49</span>  <span class="keywordflow">for</span> (<span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> row=0; row<rows; row++) {</div><div class="line"><a name="l00050"></a><span class="lineno"> 50</span>  <span class="keywordflow">for</span> (<span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> col=0; col<cols; col++) {</div><div class="line"><a name="l00051"></a><span class="lineno"> 51</span>  T &v = out[row * stride + col];</div><div class="line"><a name="l00052"></a><span class="lineno"> 52</span>  <span class="keywordflow">if</span> (DoBias) {</div><div class="line"><a name="l00053"></a><span class="lineno"> 53</span>  v += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>[col];</div><div class="line"><a name="l00054"></a><span class="lineno"> 54</span>  }</div><div class="line"><a name="l00055"></a><span class="lineno"> 55</span>  v = std::max(static_cast<T>(0), v);</div><div class="line"><a name="l00056"></a><span class="lineno"> 56</span>  }</div><div class="line"><a name="l00057"></a><span class="lineno"> 57</span>  }</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> </div><div class="line"><a name="l00060"></a><span class="lineno"> 60</span>  <span class="keywordflow">if</span> (act.type == Activation::Type::BoundedReLU) {</div><div class="line"><a name="l00061"></a><span class="lineno"> 61</span>  <span class="keyword">const</span> T max = static_cast<T>(act.param1);</div><div class="line"><a name="l00062"></a><span class="lineno"> 62</span> </div><div class="line"><a name="l00063"></a><span class="lineno"> 63</span>  <span class="keywordflow">for</span> (<span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> row=0; row<rows; row++) {</div><div class="line"><a name="l00064"></a><span class="lineno"> 64</span>  <span class="keywordflow">for</span> (<span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> col=0; col<cols; col++) {</div><div class="line"><a name="l00065"></a><span class="lineno"> 65</span>  T &v = out[row * stride + col];</div><div class="line"><a name="l00066"></a><span class="lineno"> 66</span>  <span class="keywordflow">if</span> (DoBias) {</div><div class="line"><a name="l00067"></a><span class="lineno"> 67</span>  v += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>[col];</div><div class="line"><a name="l00068"></a><span class="lineno"> 68</span>  }</div><div class="line"><a name="l00069"></a><span class="lineno"> 69</span>  v = std::max(static_cast<T>(0), std::min(v, max));</div><div class="line"><a name="l00070"></a><span class="lineno"> 70</span>  }</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>  }</div><div class="line"><a name="l00073"></a><span class="lineno"> 73</span> }</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="namespacearm__gemm_xhtml_a7450234a5b6959d08a3ddac191b3fe1f"><div class="ttname"><a href="namespacearm__gemm.xhtml#a7450234a5b6959d08a3ddac191b3fe1f">arm_gemm::bias_adder</a></div><div class="ttdeci">void bias_adder(T *out, unsigned int stride, const T *bias, unsigned int rows, unsigned int cols)</div><div class="ttdef"><b>Definition:</b> <a href="bias__adder_8hpp_source.xhtml#l00031">bias_adder.hpp:31</a></div></div>
+</div><!-- fragment -->
+<p class="reference">References <a class="el" href="validation_2_c_l_2_convolution_layer_8cpp_source.xhtml#l00181">arm_compute::test::validation::bias</a>, and <a class="el" href="bias__adder_8hpp_source.xhtml#l00031">bias_adder()</a>.</p>
+
+</div>
+</div>
+<a id="a7450234a5b6959d08a3ddac191b3fe1f"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#a7450234a5b6959d08a3ddac191b3fe1f">◆ </a></span>bias_adder()</h2>
+
+<div class="memitem">
+<div class="memproto">
+<table class="mlabels">
+ <tr>
+ <td class="mlabels-left">
+ <table class="memname">
+ <tr>
+ <td class="memname">void arm_gemm::bias_adder </td>
+ <td>(</td>
+ <td class="paramtype">T * </td>
+ <td class="paramname"><em>out</em>, </td>
+ </tr>
+ <tr>
+ <td class="paramkey"></td>
+ <td></td>
+ <td class="paramtype">unsigned int </td>
+ <td class="paramname"><em>stride</em>, </td>
+ </tr>
+ <tr>
+ <td class="paramkey"></td>
+ <td></td>
+ <td class="paramtype">const T * </td>
+ <td class="paramname"><em>bias</em>, </td>
+ </tr>
+ <tr>
+ <td class="paramkey"></td>
+ <td></td>
+ <td class="paramtype">unsigned int </td>
+ <td class="paramname"><em>rows</em>, </td>
+ </tr>
+ <tr>
+ <td class="paramkey"></td>
+ <td></td>
+ <td class="paramtype">unsigned int </td>
+ <td class="paramname"><em>cols</em> </td>
+ </tr>
+ <tr>
+ <td></td>
+ <td>)</td>
+ <td></td><td></td>
+ </tr>
+ </table>
+ </td>
+ <td class="mlabels-right">
+<span class="mlabels"><span class="mlabel">inline</span></span> </td>
+ </tr>
+</table>
+</div><div class="memdoc">
+
+<p class="definition">Definition at line <a class="el" href="bias__adder_8hpp_source.xhtml#l00031">31</a> of file <a class="el" href="bias__adder_8hpp_source.xhtml">bias_adder.hpp</a>.</p>
+<div class="fragment"><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="keywordflow">for</span> (<span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> row=0; row<rows; row++) {</div><div class="line"><a name="l00033"></a><span class="lineno"> 33</span>  <span class="keywordflow">for</span> (<span class="keywordtype">unsigned</span> <span class="keywordtype">int</span> col=0; col<cols; col++) {</div><div class="line"><a name="l00034"></a><span class="lineno"> 34</span>  out[row * stride + col] += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>[col];</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>  }</div><div class="line"><a name="l00037"></a><span class="lineno"> 37</span> }</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><!-- fragment -->
+<p class="reference">References <a class="el" href="validation_2_c_l_2_convolution_layer_8cpp_source.xhtml#l00181">arm_compute::test::validation::bias</a>.</p>
+
+<p class="reference">Referenced by <a class="el" href="bias__adder_8hpp_source.xhtml#l00040">activator()</a>.</p>
+
+</div>
+</div>
+<a id="af0b4b7759cb0829cb2a54e73b1555a01"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#af0b4b7759cb0829cb2a54e73b1555a01">◆ </a></span>compute_col_sums()</h2>
<div class="memitem">
<div class="memproto">
@@ -285,6 +418,12 @@
<td class="paramkey"></td>
<td></td>
<td class="paramtype">unsigned int </td>
+ <td class="paramname"><em>multi</em>, </td>
+ </tr>
+ <tr>
+ <td class="paramkey"></td>
+ <td></td>
+ <td class="paramtype">unsigned int </td>
<td class="paramname"><em>first_col</em> </td>
</tr>
<tr>
@@ -295,7 +434,7 @@
</table>
</div><div class="memdoc">
-<p class="reference">Referenced by <a class="el" href="gemm__hybrid__quantized_8hpp_source.xhtml#l00263">GemmHybridQuantized< strategy, To, Tr >::pretranspose_B_array()</a>.</p>
+<p class="reference">Referenced by <a class="el" href="gemm__hybrid__quantized_8hpp_source.xhtml#l00259">GemmHybridQuantized< strategy, To, Tr >::pretranspose_B_array()</a>.</p>
</div>
</div>
@@ -351,8 +490,8 @@
</div>
</div>
-<a id="a7af0b0e9e111e3ec37682c89ee803d82"></a>
-<h2 class="memtitle"><span class="permalink"><a href="#a7af0b0e9e111e3ec37682c89ee803d82">◆ </a></span>find_implementation()</h2>
+<a id="adfe29ef1042864157b1fd4c72cb21fa0"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#adfe29ef1042864157b1fd4c72cb21fa0">◆ </a></span>find_implementation()</h2>
<div class="memitem">
<div class="memproto">
@@ -360,7 +499,7 @@
<tr>
<td class="memname">bool arm_gemm::find_implementation </td>
<td>(</td>
- <td class="paramtype">const GemmArgs< Tret > & </td>
+ <td class="paramtype">const GemmArgs & </td>
<td class="paramname"><em>args</em>, </td>
</tr>
<tr>
@@ -384,15 +523,16 @@
</div><div class="memdoc">
<p class="definition">Definition at line <a class="el" href="gemm__implementation_8hpp_source.xhtml#l00119">119</a> of file <a class="el" href="gemm__implementation_8hpp_source.xhtml">gemm_implementation.hpp</a>.</p>
-<div class="fragment"><div class="line"><a name="l00119"></a><span class="lineno"> 119</span>  {</div><div class="line"><a name="l00120"></a><span class="lineno"> 120</span>  <span class="keyword">auto</span> gemms = gemm_implementation_list<Top, Tret, OutputStage>();</div><div class="line"><a name="l00121"></a><span class="lineno"> 121</span>  <span class="keyword">const</span> GemmConfig *cfg = args._cfg;</div><div class="line"><a name="l00122"></a><span class="lineno"> 122</span> </div><div class="line"><a name="l00123"></a><span class="lineno"> 123</span>  <span class="keyword">const</span> GemmImplementation<Top, Tret, OutputStage> *saved_impl = <span class="keyword">nullptr</span>;</div><div class="line"><a name="l00124"></a><span class="lineno"> 124</span> </div><div class="line"><a name="l00125"></a><span class="lineno"> 125</span>  <span class="keywordflow">for</span> (<span class="keyword">const</span> GemmImplementation<Top, Tret, OutputStage> *i = gemms; i->method != GemmMethod::DEFAULT; i++) {</div><div class="line"><a name="l00126"></a><span class="lineno"> 126</span>  <span class="comment">/* Skip if this implementation doesn't support these args. */</span></div><div class="line"><a name="l00127"></a><span class="lineno"> 127</span>  <span class="keywordflow">if</span> (!i->do_is_supported(args, os)) {</div><div class="line"><a name="l00128"></a><span class="lineno"> 128</span>  <span class="keywordflow">continue</span>;</div><div class="line"><a name="l00129"></a><span class="lineno"> 129</span>  }</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>  <span class="comment">/* Skip if a specific method is requested and this is a different one. */</span></div><div class="line"><a name="l00132"></a><span class="lineno"> 132</span>  <span class="keywordflow">if</span> (cfg && cfg->method != GemmMethod::DEFAULT && i->method != cfg->method) {</div><div class="line"><a name="l00133"></a><span class="lineno"> 133</span>  <span class="keywordflow">continue</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> </div><div class="line"><a name="l00136"></a><span class="lineno"> 136</span>  <span class="comment">/* Skip if a filter is to be applied and it doesn't match. */</span></div><div class="line"><a name="l00137"></a><span class="lineno"> 137</span>  <span class="keywordflow">if</span> (cfg && cfg->filter != <span class="stringliteral">""</span> && !strstr(i->name, cfg->filter.c_str())) {</div><div class="line"><a name="l00138"></a><span class="lineno"> 138</span>  <span class="keywordflow">continue</span>;</div><div class="line"><a name="l00139"></a><span class="lineno"> 139</span>  }</div><div class="line"><a name="l00140"></a><span class="lineno"> 140</span> </div><div class="line"><a name="l00141"></a><span class="lineno"> 141</span>  <span class="comment">/* At this point, if we don't have a saved implementation, save this</span></div><div class="line"><a name="l00142"></a><span class="lineno"> 142</span> <span class="comment"> * one. This is so that we always return something if a filter</span></div><div class="line"><a name="l00143"></a><span class="lineno"> 143</span> <span class="comment"> * matches, even if it doesn't recommend itself.</span></div><div class="line"><a name="l00144"></a><span class="lineno"> 144</span> <span class="comment"> */</span></div><div class="line"><a name="l00145"></a><span class="lineno"> 145</span>  <span class="keywordflow">if</span> (saved_impl == <span class="keyword">nullptr</span>) {</div><div class="line"><a name="l00146"></a><span class="lineno"> 146</span>  saved_impl=i;</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> </div><div class="line"><a name="l00149"></a><span class="lineno"> 149</span>  <span class="comment">/* Check that this method recommends itself. */</span></div><div class="line"><a name="l00150"></a><span class="lineno"> 150</span>  <span class="keywordflow">if</span> (!i->do_is_recommended(args, os)) {</div><div class="line"><a name="l00151"></a><span class="lineno"> 151</span>  <span class="keywordflow">continue</span>;</div><div class="line"><a name="l00152"></a><span class="lineno"> 152</span>  }</div><div class="line"><a name="l00153"></a><span class="lineno"> 153</span> </div><div class="line"><a name="l00154"></a><span class="lineno"> 154</span>  impl=i;</div><div class="line"><a name="l00155"></a><span class="lineno"> 155</span> </div><div class="line"><a name="l00156"></a><span class="lineno"> 156</span>  <span class="keywordflow">return</span> <span class="keyword">true</span>;</div><div class="line"><a name="l00157"></a><span class="lineno"> 157</span>  }</div><div class="line"><a name="l00158"></a><span class="lineno"> 158</span> </div><div class="line"><a name="l00159"></a><span class="lineno"> 159</span>  <span class="comment">/* We didn't find an option matching the filters that recommended</span></div><div class="line"><a name="l00160"></a><span class="lineno"> 160</span> <span class="comment"> * itself. But if we found something earlier that matched the filters</span></div><div class="line"><a name="l00161"></a><span class="lineno"> 161</span> <span class="comment"> * but wasn't recommended, return it here. */</span></div><div class="line"><a name="l00162"></a><span class="lineno"> 162</span>  <span class="keywordflow">if</span> (saved_impl != <span class="keyword">nullptr</span>) {</div><div class="line"><a name="l00163"></a><span class="lineno"> 163</span>  impl = saved_impl;</div><div class="line"><a name="l00164"></a><span class="lineno"> 164</span>  <span class="keywordflow">return</span> <span class="keyword">true</span>;</div><div class="line"><a name="l00165"></a><span class="lineno"> 165</span>  }</div><div class="line"><a name="l00166"></a><span class="lineno"> 166</span> </div><div class="line"><a name="l00167"></a><span class="lineno"> 167</span>  <span class="keywordflow">return</span> <span class="keyword">false</span>;</div><div class="line"><a name="l00168"></a><span class="lineno"> 168</span> }</div></div><!-- fragment -->
-<p class="reference">References <a class="el" href="gemm__implementation_8hpp_source.xhtml#l00037">GemmImplementation< Top, Tret, OutputStage >::method</a>.</p>
+<div class="fragment"><div class="line"><a name="l00119"></a><span class="lineno"> 119</span>  {</div><div class="line"><a name="l00120"></a><span class="lineno"> 120</span>  <span class="keyword">auto</span> gemms = gemm_implementation_list<Top, Tret, OutputStage>();</div><div class="line"><a name="l00121"></a><span class="lineno"> 121</span>  <span class="keyword">const</span> GemmConfig *cfg = <a class="code" href="namespace_gemm_tuner.xhtml#a8187411843a6284ffb964ef3fb9fcab3">args</a>._cfg;</div><div class="line"><a name="l00122"></a><span class="lineno"> 122</span> </div><div class="line"><a name="l00123"></a><span class="lineno"> 123</span>  <span class="keyword">const</span> GemmImplementation<Top, Tret, OutputStage> *saved_impl = <span class="keyword">nullptr</span>;</div><div class="line"><a name="l00124"></a><span class="lineno"> 124</span> </div><div class="line"><a name="l00125"></a><span class="lineno"> 125</span>  <span class="keywordflow">for</span> (<span class="keyword">const</span> GemmImplementation<Top, Tret, OutputStage> *i = gemms; i->method != GemmMethod::DEFAULT; i++) {</div><div class="line"><a name="l00126"></a><span class="lineno"> 126</span>  <span class="comment">/* Skip if this implementation doesn't support these args. */</span></div><div class="line"><a name="l00127"></a><span class="lineno"> 127</span>  <span class="keywordflow">if</span> (!i->do_is_supported(<a class="code" href="namespace_gemm_tuner.xhtml#a8187411843a6284ffb964ef3fb9fcab3">args</a>, os)) {</div><div class="line"><a name="l00128"></a><span class="lineno"> 128</span>  <span class="keywordflow">continue</span>;</div><div class="line"><a name="l00129"></a><span class="lineno"> 129</span>  }</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>  <span class="comment">/* Skip if a specific method is requested and this is a different one. */</span></div><div class="line"><a name="l00132"></a><span class="lineno"> 132</span>  <span class="keywordflow">if</span> (cfg && cfg->method != GemmMethod::DEFAULT && i->method != cfg->method) {</div><div class="line"><a name="l00133"></a><span class="lineno"> 133</span>  <span class="keywordflow">continue</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> </div><div class="line"><a name="l00136"></a><span class="lineno"> 136</span>  <span class="comment">/* Skip if a filter is to be applied and it doesn't match. */</span></div><div class="line"><a name="l00137"></a><span class="lineno"> 137</span>  <span class="keywordflow">if</span> (cfg && cfg->filter != <span class="stringliteral">""</span> && !strstr(i->name, cfg->filter.c_str())) {</div><div class="line"><a name="l00138"></a><span class="lineno"> 138</span>  <span class="keywordflow">continue</span>;</div><div class="line"><a name="l00139"></a><span class="lineno"> 139</span>  }</div><div class="line"><a name="l00140"></a><span class="lineno"> 140</span> </div><div class="line"><a name="l00141"></a><span class="lineno"> 141</span>  <span class="comment">/* At this point, if we don't have a saved implementation, save this</span></div><div class="line"><a name="l00142"></a><span class="lineno"> 142</span> <span class="comment"> * one. This is so that we always return something if a filter</span></div><div class="line"><a name="l00143"></a><span class="lineno"> 143</span> <span class="comment"> * matches, even if it doesn't recommend itself.</span></div><div class="line"><a name="l00144"></a><span class="lineno"> 144</span> <span class="comment"> */</span></div><div class="line"><a name="l00145"></a><span class="lineno"> 145</span>  <span class="keywordflow">if</span> (saved_impl == <span class="keyword">nullptr</span>) {</div><div class="line"><a name="l00146"></a><span class="lineno"> 146</span>  saved_impl=i;</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> </div><div class="line"><a name="l00149"></a><span class="lineno"> 149</span>  <span class="comment">/* Check that this method recommends itself. */</span></div><div class="line"><a name="l00150"></a><span class="lineno"> 150</span>  <span class="keywordflow">if</span> (!i->do_is_recommended(<a class="code" href="namespace_gemm_tuner.xhtml#a8187411843a6284ffb964ef3fb9fcab3">args</a>, os)) {</div><div class="line"><a name="l00151"></a><span class="lineno"> 151</span>  <span class="keywordflow">continue</span>;</div><div class="line"><a name="l00152"></a><span class="lineno"> 152</span>  }</div><div class="line"><a name="l00153"></a><span class="lineno"> 153</span> </div><div class="line"><a name="l00154"></a><span class="lineno"> 154</span>  impl=i;</div><div class="line"><a name="l00155"></a><span class="lineno"> 155</span> </div><div class="line"><a name="l00156"></a><span class="lineno"> 156</span>  <span class="keywordflow">return</span> <span class="keyword">true</span>;</div><div class="line"><a name="l00157"></a><span class="lineno"> 157</span>  }</div><div class="line"><a name="l00158"></a><span class="lineno"> 158</span> </div><div class="line"><a name="l00159"></a><span class="lineno"> 159</span>  <span class="comment">/* We didn't find an option matching the filters that recommended</span></div><div class="line"><a name="l00160"></a><span class="lineno"> 160</span> <span class="comment"> * itself. But if we found something earlier that matched the filters</span></div><div class="line"><a name="l00161"></a><span class="lineno"> 161</span> <span class="comment"> * but wasn't recommended, return it here. */</span></div><div class="line"><a name="l00162"></a><span class="lineno"> 162</span>  <span class="keywordflow">if</span> (saved_impl != <span class="keyword">nullptr</span>) {</div><div class="line"><a name="l00163"></a><span class="lineno"> 163</span>  impl = saved_impl;</div><div class="line"><a name="l00164"></a><span class="lineno"> 164</span>  <span class="keywordflow">return</span> <span class="keyword">true</span>;</div><div class="line"><a name="l00165"></a><span class="lineno"> 165</span>  }</div><div class="line"><a name="l00166"></a><span class="lineno"> 166</span> </div><div class="line"><a name="l00167"></a><span class="lineno"> 167</span>  <span class="keywordflow">return</span> <span class="keyword">false</span>;</div><div class="line"><a name="l00168"></a><span class="lineno"> 168</span> }</div><div class="ttc" id="namespace_gemm_tuner_xhtml_a8187411843a6284ffb964ef3fb9fcab3"><div class="ttname"><a href="namespace_gemm_tuner.xhtml#a8187411843a6284ffb964ef3fb9fcab3">GemmTuner.args</a></div><div class="ttdeci">args</div><div class="ttdef"><b>Definition:</b> <a href="_gemm_tuner_8py_source.xhtml#l00555">GemmTuner.py:555</a></div></div>
+</div><!-- fragment -->
+<p class="reference">References <a class="el" href="_gemm_tuner_8py_source.xhtml#l00555">GemmTuner::args</a>, and <a class="el" href="gemm__implementation_8hpp_source.xhtml#l00037">GemmImplementation< Top, Tret, OutputStage >::method</a>.</p>
<p class="reference">Referenced by <a class="el" href="gemm__implementation_8hpp_source.xhtml#l00171">get_compatible_kernels()</a>.</p>
</div>
</div>
-<a id="a3cf8a3a8f4c31d116db5013daa6d83d9"></a>
-<h2 class="memtitle"><span class="permalink"><a href="#a3cf8a3a8f4c31d116db5013daa6d83d9">◆ </a></span>gemm()</h2>
+<a id="a64b2f658fb34f894449e0279747a6390"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#a64b2f658fb34f894449e0279747a6390">◆ </a></span>gemm()</h2>
<div class="memitem">
<div class="memproto">
@@ -400,7 +540,7 @@
<tr>
<td class="memname">UniqueGemmCommon<Top, Tret> arm_gemm::gemm </td>
<td>(</td>
- <td class="paramtype">const GemmArgs< Tret > & </td>
+ <td class="paramtype">const GemmArgs & </td>
<td class="paramname"><em>args</em>, </td>
</tr>
<tr>
@@ -418,21 +558,22 @@
</div><div class="memdoc">
<p class="definition">Definition at line <a class="el" href="gemm__implementation_8hpp_source.xhtml#l00193">193</a> of file <a class="el" href="gemm__implementation_8hpp_source.xhtml">gemm_implementation.hpp</a>.</p>
-<div class="fragment"><div class="line"><a name="l00193"></a><span class="lineno"> 193</span>  {</div><div class="line"><a name="l00194"></a><span class="lineno"> 194</span>  <span class="keyword">const</span> GemmImplementation<Top, Tret, OutputStage> *impl;</div><div class="line"><a name="l00195"></a><span class="lineno"> 195</span> </div><div class="line"><a name="l00196"></a><span class="lineno"> 196</span>  <span class="keywordflow">if</span> (find_implementation<Top, Tret, OutputStage>(args, os, impl)) {</div><div class="line"><a name="l00197"></a><span class="lineno"> 197</span>  <span class="keywordflow">return</span> UniqueGemmCommon<Top, Tret>(impl->do_instantiate(args, os));</div><div class="line"><a name="l00198"></a><span class="lineno"> 198</span>  }</div><div class="line"><a name="l00199"></a><span class="lineno"> 199</span> </div><div class="line"><a name="l00200"></a><span class="lineno"> 200</span>  <span class="keywordflow">return</span> UniqueGemmCommon<Top, Tret>(<span class="keyword">nullptr</span>);</div><div class="line"><a name="l00201"></a><span class="lineno"> 201</span> }</div></div><!-- fragment -->
-<p class="reference">References <a class="el" href="gemm__implementation_8hpp_source.xhtml#l00059">GemmImplementation< Top, Tret, OutputStage >::do_instantiate()</a>.</p>
+<div class="fragment"><div class="line"><a name="l00193"></a><span class="lineno"> 193</span>  {</div><div class="line"><a name="l00194"></a><span class="lineno"> 194</span>  <span class="keyword">const</span> GemmImplementation<Top, Tret, OutputStage> *impl;</div><div class="line"><a name="l00195"></a><span class="lineno"> 195</span> </div><div class="line"><a name="l00196"></a><span class="lineno"> 196</span>  <span class="keywordflow">if</span> (find_implementation<Top, Tret, OutputStage>(<a class="code" href="namespace_gemm_tuner.xhtml#a8187411843a6284ffb964ef3fb9fcab3">args</a>, os, impl)) {</div><div class="line"><a name="l00197"></a><span class="lineno"> 197</span>  <span class="keywordflow">return</span> UniqueGemmCommon<Top, Tret>(impl->do_instantiate(<a class="code" href="namespace_gemm_tuner.xhtml#a8187411843a6284ffb964ef3fb9fcab3">args</a>, os));</div><div class="line"><a name="l00198"></a><span class="lineno"> 198</span>  }</div><div class="line"><a name="l00199"></a><span class="lineno"> 199</span> </div><div class="line"><a name="l00200"></a><span class="lineno"> 200</span>  <span class="keywordflow">return</span> UniqueGemmCommon<Top, Tret>(<span class="keyword">nullptr</span>);</div><div class="line"><a name="l00201"></a><span class="lineno"> 201</span> }</div><div class="ttc" id="namespace_gemm_tuner_xhtml_a8187411843a6284ffb964ef3fb9fcab3"><div class="ttname"><a href="namespace_gemm_tuner.xhtml#a8187411843a6284ffb964ef3fb9fcab3">GemmTuner.args</a></div><div class="ttdeci">args</div><div class="ttdef"><b>Definition:</b> <a href="_gemm_tuner_8py_source.xhtml#l00555">GemmTuner.py:555</a></div></div>
+</div><!-- fragment -->
+<p class="reference">References <a class="el" href="_gemm_tuner_8py_source.xhtml#l00555">GemmTuner::args</a>, and <a class="el" href="gemm__implementation_8hpp_source.xhtml#l00059">GemmImplementation< Top, Tret, OutputStage >::do_instantiate()</a>.</p>
</div>
</div>
-<a id="ab98c2935f5bb60c9a4add6c3e9de53e4"></a>
-<h2 class="memtitle"><span class="permalink"><a href="#ab98c2935f5bb60c9a4add6c3e9de53e4">◆ </a></span>gemm< float, float, Nothing >()</h2>
+<a id="a4fb5776f35967c7173f432cb096de217"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#a4fb5776f35967c7173f432cb096de217">◆ </a></span>gemm< float, float, Nothing >()</h2>
<div class="memitem">
<div class="memproto">
<table class="memname">
<tr>
- <td class="memname">template UniqueGemmCommon<float, float> <a class="el" href="namespacearm__gemm.xhtml#a3cf8a3a8f4c31d116db5013daa6d83d9">arm_gemm::gemm</a>< float, float, Nothing > </td>
+ <td class="memname">template UniqueGemmCommon<float, float> <a class="el" href="namespacearm__gemm.xhtml#a64b2f658fb34f894449e0279747a6390">arm_gemm::gemm</a>< float, float, Nothing > </td>
<td>(</td>
- <td class="paramtype">const GemmArgs< float > & </td>
+ <td class="paramtype">const GemmArgs & </td>
<td class="paramname"><em>args</em>, </td>
</tr>
<tr>
@@ -483,12 +624,12 @@
</table>
</div><div class="memdoc">
-<p class="definition">Definition at line <a class="el" href="gemm__fp32_8cpp_source.xhtml#l00168">168</a> of file <a class="el" href="gemm__fp32_8cpp_source.xhtml">gemm_fp32.cpp</a>.</p>
-<div class="fragment"><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> gemm_fp32_methods;</div><div class="line"><a name="l00170"></a><span class="lineno"> 170</span> }</div></div><!-- fragment -->
+<p class="definition">Definition at line <a class="el" href="gemm__fp32_8cpp_source.xhtml#l00160">160</a> of file <a class="el" href="gemm__fp32_8cpp_source.xhtml">gemm_fp32.cpp</a>.</p>
+<div class="fragment"><div class="line"><a name="l00160"></a><span class="lineno"> 160</span>  {</div><div class="line"><a name="l00161"></a><span class="lineno"> 161</span>  <span class="keywordflow">return</span> gemm_fp32_methods;</div><div class="line"><a name="l00162"></a><span class="lineno"> 162</span> }</div></div><!-- fragment -->
</div>
</div>
-<a id="a8eb5d997d45efbbdaa4374f45e541d86"></a>
-<h2 class="memtitle"><span class="permalink"><a href="#a8eb5d997d45efbbdaa4374f45e541d86">◆ </a></span>get_compatible_kernels()</h2>
+<a id="abce3ed84b7a8f5c6e45809ca1089b1cf"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#abce3ed84b7a8f5c6e45809ca1089b1cf">◆ </a></span>get_compatible_kernels()</h2>
<div class="memitem">
<div class="memproto">
@@ -496,7 +637,7 @@
<tr>
<td class="memname">std::vector<KernelDescription> arm_gemm::get_compatible_kernels </td>
<td>(</td>
- <td class="paramtype">const GemmArgs< Tret > & </td>
+ <td class="paramtype">const GemmArgs & </td>
<td class="paramname"><em>args</em>, </td>
</tr>
<tr>
@@ -514,22 +655,23 @@
</div><div class="memdoc">
<p class="definition">Definition at line <a class="el" href="gemm__implementation_8hpp_source.xhtml#l00171">171</a> of file <a class="el" href="gemm__implementation_8hpp_source.xhtml">gemm_implementation.hpp</a>.</p>
-<div class="fragment"><div class="line"><a name="l00171"></a><span class="lineno"> 171</span>  {</div><div class="line"><a name="l00172"></a><span class="lineno"> 172</span>  std::vector<KernelDescription> res;</div><div class="line"><a name="l00173"></a><span class="lineno"> 173</span> </div><div class="line"><a name="l00174"></a><span class="lineno"> 174</span>  <span class="comment">/* Find out what the default implementation in so we can set the flag accordingly later. */</span></div><div class="line"><a name="l00175"></a><span class="lineno"> 175</span>  <span class="keyword">const</span> GemmImplementation<Top, Tret, OutputStage> *default_impl;</div><div class="line"><a name="l00176"></a><span class="lineno"> 176</span>  <a class="code" href="namespacearm__gemm.xhtml#a7af0b0e9e111e3ec37682c89ee803d82">find_implementation</a>(args, os, default_impl);</div><div class="line"><a name="l00177"></a><span class="lineno"> 177</span> </div><div class="line"><a name="l00178"></a><span class="lineno"> 178</span>  <span class="keyword">auto</span> gemms = gemm_implementation_list<Top, Tret, OutputStage>();</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>  <span class="keywordflow">for</span> (<span class="keyword">const</span> GemmImplementation<Top, Tret, OutputStage> *i = gemms; i->method != GemmMethod::DEFAULT; i++) {</div><div class="line"><a name="l00181"></a><span class="lineno"> 181</span>  <span class="comment">/* Check that this implementation supports the presented problem. */</span></div><div class="line"><a name="l00182"></a><span class="lineno"> 182</span>  <span class="keywordflow">if</span> (!i->do_is_supported(args, os)) {</div><div class="line"><a name="l00183"></a><span class="lineno"> 183</span>  <span class="keywordflow">continue</span>;</div><div class="line"><a name="l00184"></a><span class="lineno"> 184</span>  }</div><div class="line"><a name="l00185"></a><span class="lineno"> 185</span> </div><div class="line"><a name="l00186"></a><span class="lineno"> 186</span>  res.push_back(KernelDescription(i->method, i->name, i==default_impl));</div><div class="line"><a name="l00187"></a><span class="lineno"> 187</span>  }</div><div class="line"><a name="l00188"></a><span class="lineno"> 188</span> </div><div class="line"><a name="l00189"></a><span class="lineno"> 189</span>  <span class="keywordflow">return</span> res;</div><div class="line"><a name="l00190"></a><span class="lineno"> 190</span> }</div><div class="ttc" id="namespacearm__gemm_xhtml_a7af0b0e9e111e3ec37682c89ee803d82"><div class="ttname"><a href="namespacearm__gemm.xhtml#a7af0b0e9e111e3ec37682c89ee803d82">arm_gemm::find_implementation</a></div><div class="ttdeci">bool find_implementation(const GemmArgs< Tret > &args, const OutputStage &os, const GemmImplementation< Top, Tret, OutputStage > *&impl)</div><div class="ttdef"><b>Definition:</b> <a href="gemm__implementation_8hpp_source.xhtml#l00119">gemm_implementation.hpp:119</a></div></div>
+<div class="fragment"><div class="line"><a name="l00171"></a><span class="lineno"> 171</span>  {</div><div class="line"><a name="l00172"></a><span class="lineno"> 172</span>  std::vector<KernelDescription> res;</div><div class="line"><a name="l00173"></a><span class="lineno"> 173</span> </div><div class="line"><a name="l00174"></a><span class="lineno"> 174</span>  <span class="comment">/* Find out what the default implementation in so we can set the flag accordingly later. */</span></div><div class="line"><a name="l00175"></a><span class="lineno"> 175</span>  <span class="keyword">const</span> GemmImplementation<Top, Tret, OutputStage> *default_impl;</div><div class="line"><a name="l00176"></a><span class="lineno"> 176</span>  <a class="code" href="namespacearm__gemm.xhtml#adfe29ef1042864157b1fd4c72cb21fa0">find_implementation</a>(<a class="code" href="namespace_gemm_tuner.xhtml#a8187411843a6284ffb964ef3fb9fcab3">args</a>, os, default_impl);</div><div class="line"><a name="l00177"></a><span class="lineno"> 177</span> </div><div class="line"><a name="l00178"></a><span class="lineno"> 178</span>  <span class="keyword">auto</span> gemms = gemm_implementation_list<Top, Tret, OutputStage>();</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>  <span class="keywordflow">for</span> (<span class="keyword">const</span> GemmImplementation<Top, Tret, OutputStage> *i = gemms; i->method != GemmMethod::DEFAULT; i++) {</div><div class="line"><a name="l00181"></a><span class="lineno"> 181</span>  <span class="comment">/* Check that this implementation supports the presented problem. */</span></div><div class="line"><a name="l00182"></a><span class="lineno"> 182</span>  <span class="keywordflow">if</span> (!i->do_is_supported(<a class="code" href="namespace_gemm_tuner.xhtml#a8187411843a6284ffb964ef3fb9fcab3">args</a>, os)) {</div><div class="line"><a name="l00183"></a><span class="lineno"> 183</span>  <span class="keywordflow">continue</span>;</div><div class="line"><a name="l00184"></a><span class="lineno"> 184</span>  }</div><div class="line"><a name="l00185"></a><span class="lineno"> 185</span> </div><div class="line"><a name="l00186"></a><span class="lineno"> 186</span>  res.push_back(KernelDescription(i->method, i->name, i==default_impl));</div><div class="line"><a name="l00187"></a><span class="lineno"> 187</span>  }</div><div class="line"><a name="l00188"></a><span class="lineno"> 188</span> </div><div class="line"><a name="l00189"></a><span class="lineno"> 189</span>  <span class="keywordflow">return</span> res;</div><div class="line"><a name="l00190"></a><span class="lineno"> 190</span> }</div><div class="ttc" id="namespace_gemm_tuner_xhtml_a8187411843a6284ffb964ef3fb9fcab3"><div class="ttname"><a href="namespace_gemm_tuner.xhtml#a8187411843a6284ffb964ef3fb9fcab3">GemmTuner.args</a></div><div class="ttdeci">args</div><div class="ttdef"><b>Definition:</b> <a href="_gemm_tuner_8py_source.xhtml#l00555">GemmTuner.py:555</a></div></div>
+<div class="ttc" id="namespacearm__gemm_xhtml_adfe29ef1042864157b1fd4c72cb21fa0"><div class="ttname"><a href="namespacearm__gemm.xhtml#adfe29ef1042864157b1fd4c72cb21fa0">arm_gemm::find_implementation</a></div><div class="ttdeci">bool find_implementation(const GemmArgs &args, const OutputStage &os, const GemmImplementation< Top, Tret, OutputStage > *&impl)</div><div class="ttdef"><b>Definition:</b> <a href="gemm__implementation_8hpp_source.xhtml#l00119">gemm_implementation.hpp:119</a></div></div>
</div><!-- fragment -->
-<p class="reference">References <a class="el" href="gemm__implementation_8hpp_source.xhtml#l00119">find_implementation()</a>, and <a class="el" href="gemm__implementation_8hpp_source.xhtml#l00037">GemmImplementation< Top, Tret, OutputStage >::method</a>.</p>
+<p class="reference">References <a class="el" href="_gemm_tuner_8py_source.xhtml#l00555">GemmTuner::args</a>, <a class="el" href="gemm__implementation_8hpp_source.xhtml#l00119">find_implementation()</a>, and <a class="el" href="gemm__implementation_8hpp_source.xhtml#l00037">GemmImplementation< Top, Tret, OutputStage >::method</a>.</p>
</div>
</div>
-<a id="a9194360a5ba6924c83f0269ade8608f5"></a>
-<h2 class="memtitle"><span class="permalink"><a href="#a9194360a5ba6924c83f0269ade8608f5">◆ </a></span>get_compatible_kernels< float, float, Nothing >()</h2>
+<a id="a4ac39db987b3fbd8e8a6e971961b41f4"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#a4ac39db987b3fbd8e8a6e971961b41f4">◆ </a></span>get_compatible_kernels< float, float, Nothing >()</h2>
<div class="memitem">
<div class="memproto">
<table class="memname">
<tr>
- <td class="memname">template std::vector<KernelDescription> <a class="el" href="namespacearm__gemm.xhtml#a8eb5d997d45efbbdaa4374f45e541d86">arm_gemm::get_compatible_kernels</a>< float, float, Nothing > </td>
+ <td class="memname">template std::vector<KernelDescription> <a class="el" href="namespacearm__gemm.xhtml#abce3ed84b7a8f5c6e45809ca1089b1cf">arm_gemm::get_compatible_kernels</a>< float, float, Nothing > </td>
<td>(</td>
- <td class="paramtype">const GemmArgs< float > & </td>
+ <td class="paramtype">const GemmArgs & </td>
<td class="paramname"><em>args</em>, </td>
</tr>
<tr>
@@ -548,8 +690,8 @@
</div>
</div>
-<a id="a777476ec2693309861bef88dbef2d23f"></a>
-<h2 class="memtitle"><span class="permalink"><a href="#a777476ec2693309861bef88dbef2d23f">◆ </a></span>get_gemm_method()</h2>
+<a id="ad35cfe7c388edeac570fb9e01180dda6"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#ad35cfe7c388edeac570fb9e01180dda6">◆ </a></span>get_gemm_method()</h2>
<div class="memitem">
<div class="memproto">
@@ -557,7 +699,7 @@
<tr>
<td class="memname">KernelDescription arm_gemm::get_gemm_method </td>
<td>(</td>
- <td class="paramtype">const GemmArgs< Tret > & </td>
+ <td class="paramtype">const GemmArgs & </td>
<td class="paramname"><em>args</em>, </td>
</tr>
<tr>
@@ -575,21 +717,22 @@
</div><div class="memdoc">
<p class="definition">Definition at line <a class="el" href="gemm__implementation_8hpp_source.xhtml#l00204">204</a> of file <a class="el" href="gemm__implementation_8hpp_source.xhtml">gemm_implementation.hpp</a>.</p>
-<div class="fragment"><div class="line"><a name="l00204"></a><span class="lineno"> 204</span>  {</div><div class="line"><a name="l00205"></a><span class="lineno"> 205</span>  <span class="keyword">const</span> GemmImplementation<Top, Tret, OutputStage> *impl;</div><div class="line"><a name="l00206"></a><span class="lineno"> 206</span> </div><div class="line"><a name="l00207"></a><span class="lineno"> 207</span>  <span class="keywordflow">if</span> (find_implementation<Top, Tret>(args, os, impl)) {</div><div class="line"><a name="l00208"></a><span class="lineno"> 208</span>  <span class="keywordflow">return</span> KernelDescription(impl->method, impl->name);</div><div class="line"><a name="l00209"></a><span class="lineno"> 209</span>  }</div><div class="line"><a name="l00210"></a><span class="lineno"> 210</span> </div><div class="line"><a name="l00211"></a><span class="lineno"> 211</span>  <span class="comment">/* This shouldn't happen - there should always be at least one valid implementation. */</span></div><div class="line"><a name="l00212"></a><span class="lineno"> 212</span>  <span class="keywordflow">return</span> KernelDescription();</div><div class="line"><a name="l00213"></a><span class="lineno"> 213</span> }</div></div><!-- fragment -->
-<p class="reference">References <a class="el" href="gemm__implementation_8hpp_source.xhtml#l00037">GemmImplementation< Top, Tret, OutputStage >::method</a>, and <a class="el" href="gemm__implementation_8hpp_source.xhtml#l00038">GemmImplementation< Top, Tret, OutputStage >::name</a>.</p>
+<div class="fragment"><div class="line"><a name="l00204"></a><span class="lineno"> 204</span>  {</div><div class="line"><a name="l00205"></a><span class="lineno"> 205</span>  <span class="keyword">const</span> GemmImplementation<Top, Tret, OutputStage> *impl;</div><div class="line"><a name="l00206"></a><span class="lineno"> 206</span> </div><div class="line"><a name="l00207"></a><span class="lineno"> 207</span>  <span class="keywordflow">if</span> (find_implementation<Top, Tret>(<a class="code" href="namespace_gemm_tuner.xhtml#a8187411843a6284ffb964ef3fb9fcab3">args</a>, os, impl)) {</div><div class="line"><a name="l00208"></a><span class="lineno"> 208</span>  <span class="keywordflow">return</span> KernelDescription(impl->method, impl->name);</div><div class="line"><a name="l00209"></a><span class="lineno"> 209</span>  }</div><div class="line"><a name="l00210"></a><span class="lineno"> 210</span> </div><div class="line"><a name="l00211"></a><span class="lineno"> 211</span>  <span class="comment">/* This shouldn't happen - there should always be at least one valid implementation. */</span></div><div class="line"><a name="l00212"></a><span class="lineno"> 212</span>  <span class="keywordflow">return</span> KernelDescription();</div><div class="line"><a name="l00213"></a><span class="lineno"> 213</span> }</div><div class="ttc" id="namespace_gemm_tuner_xhtml_a8187411843a6284ffb964ef3fb9fcab3"><div class="ttname"><a href="namespace_gemm_tuner.xhtml#a8187411843a6284ffb964ef3fb9fcab3">GemmTuner.args</a></div><div class="ttdeci">args</div><div class="ttdef"><b>Definition:</b> <a href="_gemm_tuner_8py_source.xhtml#l00555">GemmTuner.py:555</a></div></div>
+</div><!-- fragment -->
+<p class="reference">References <a class="el" href="_gemm_tuner_8py_source.xhtml#l00555">GemmTuner::args</a>, <a class="el" href="gemm__implementation_8hpp_source.xhtml#l00037">GemmImplementation< Top, Tret, OutputStage >::method</a>, and <a class="el" href="gemm__implementation_8hpp_source.xhtml#l00038">GemmImplementation< Top, Tret, OutputStage >::name</a>.</p>
</div>
</div>
-<a id="abd013457d4280a12045a79c201bf25e0"></a>
-<h2 class="memtitle"><span class="permalink"><a href="#abd013457d4280a12045a79c201bf25e0">◆ </a></span>get_gemm_method< float, float, Nothing >()</h2>
+<a id="aeaa8a3f397520dca43d1dabb3774e2c9"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#aeaa8a3f397520dca43d1dabb3774e2c9">◆ </a></span>get_gemm_method< float, float, Nothing >()</h2>
<div class="memitem">
<div class="memproto">
<table class="memname">
<tr>
- <td class="memname">template KernelDescription <a class="el" href="namespacearm__gemm.xhtml#a777476ec2693309861bef88dbef2d23f">arm_gemm::get_gemm_method</a>< float, float, Nothing > </td>
+ <td class="memname">template KernelDescription <a class="el" href="namespacearm__gemm.xhtml#ad35cfe7c388edeac570fb9e01180dda6">arm_gemm::get_gemm_method</a>< float, float, Nothing > </td>
<td>(</td>
- <td class="paramtype">const GemmArgs< float > & </td>
+ <td class="paramtype">const GemmArgs & </td>
<td class="paramname"><em>args</em>, </td>
</tr>
<tr>
@@ -608,17 +751,14 @@
</div>
</div>
-<a id="a4c160021f8ceaa70922f23cafd6b3c99"></a>
-<h2 class="memtitle"><span class="permalink"><a href="#a4c160021f8ceaa70922f23cafd6b3c99">◆ </a></span>MergeResults()</h2>
+<a id="aa2ed1c3595c8114064e8170b00df37c5"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#aa2ed1c3595c8114064e8170b00df37c5">◆ </a></span>MergeResults()</h2>
<div class="memitem">
<div class="memproto">
-<table class="mlabels">
- <tr>
- <td class="mlabels-left">
<table class="memname">
<tr>
- <td class="memname">void arm_gemm::MergeResults </td>
+ <td class="memname">void MergeResults </td>
<td>(</td>
<td class="paramtype">Tout * </td>
<td class="paramname"><em>out</em>, </td>
@@ -662,14 +802,20 @@
<tr>
<td class="paramkey"></td>
<td></td>
- <td class="paramtype">const Tout </td>
- <td class="paramname"><em>alpha</em>, </td>
+ <td class="paramtype">const Tout * </td>
+ <td class="paramname"><em>bias</em>, </td>
</tr>
<tr>
<td class="paramkey"></td>
<td></td>
- <td class="paramtype">const Tout </td>
- <td class="paramname"><em>beta</em> </td>
+ <td class="paramtype">Activation </td>
+ <td class="paramname"><em>act</em>, </td>
+ </tr>
+ <tr>
+ <td class="paramkey"></td>
+ <td></td>
+ <td class="paramtype">bool </td>
+ <td class="paramname"><em>append</em> </td>
</tr>
<tr>
<td></td>
@@ -677,17 +823,12 @@
<td></td><td></td>
</tr>
</table>
- </td>
- <td class="mlabels-right">
-<span class="mlabels"><span class="mlabel">inline</span></span> </td>
- </tr>
-</table>
</div><div class="memdoc">
-<p class="definition">Definition at line <a class="el" href="mergeresults_8hpp_source.xhtml#l00036">36</a> of file <a class="el" href="mergeresults_8hpp_source.xhtml">mergeresults.hpp</a>.</p>
-<div class="fragment"><div class="line"><a name="l00036"></a><span class="lineno"> 36</span>  {</div><div class="line"><a name="l00037"></a><span class="lineno"> 37</span>  <span class="comment">// For SVE cases, multiply the width up by the vector length.</span></div><div class="line"><a name="l00038"></a><span class="lineno"> 38</span>  <span class="comment">// Use the *input* type to determine this, since this will be what the kernel operated on.</span></div><div class="line"><a name="l00039"></a><span class="lineno"> 39</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> width = twidth * (sve ? get_vector_length<Tin>() : 1);</div><div class="line"><a name="l00040"></a><span class="lineno"> 40</span> </div><div class="line"><a name="l00041"></a><span class="lineno"> 41</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> full_y_blocks = (ymax - y0) / height;</div><div class="line"><a name="l00042"></a><span class="lineno"> 42</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> y_remainder = (ymax - y0) % height;</div><div class="line"><a name="l00043"></a><span class="lineno"> 43</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> y_blocks = full_y_blocks + (y_remainder ? 1 : 0);</div><div class="line"><a name="l00044"></a><span class="lineno"> 44</span> </div><div class="line"><a name="l00045"></a><span class="lineno"> 45</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> full_x_blocks = (xmax - x0) / width;</div><div class="line"><a name="l00046"></a><span class="lineno"> 46</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> x_remainder = (xmax - x0) % width;</div><div class="line"><a name="l00047"></a><span class="lineno"> 47</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> x_blocks = full_x_blocks + (x_remainder ? 1 : 0);</div><div class="line"><a name="l00048"></a><span class="lineno"> 48</span> </div><div class="line"><a name="l00049"></a><span class="lineno"> 49</span>  <span class="keywordflow">for</span> (<span class="keywordtype">int</span> y_block = 0; y_block < y_blocks; y_block++) {</div><div class="line"><a name="l00050"></a><span class="lineno"> 50</span>  <span class="keywordtype">int</span> ybase = y0 + (y_block * height);</div><div class="line"><a name="l00051"></a><span class="lineno"> 51</span> </div><div class="line"><a name="l00052"></a><span class="lineno"> 52</span>  <span class="keywordtype">int</span> fill_rows = (y_block < full_y_blocks) ? height : y_remainder;</div><div class="line"><a name="l00053"></a><span class="lineno"> 53</span> </div><div class="line"><a name="l00054"></a><span class="lineno"> 54</span>  <span class="keywordflow">for</span> (<span class="keywordtype">int</span> x_block = 0; x_block < x_blocks; x_block++) {</div><div class="line"><a name="l00055"></a><span class="lineno"> 55</span>  <span class="keywordtype">int</span> xbase = x0 + (x_block * width);</div><div class="line"><a name="l00056"></a><span class="lineno"> 56</span> </div><div class="line"><a name="l00057"></a><span class="lineno"> 57</span>  <span class="keywordtype">int</span> fill_cols = (x_block < full_x_blocks) ? width : x_remainder;</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>  <span class="keywordflow">for</span> (<span class="keywordtype">int</span> row=0; row < fill_rows; row++) {</div><div class="line"><a name="l00060"></a><span class="lineno"> 60</span>  <span class="keywordflow">for</span> (<span class="keywordtype">int</span> col=0; col < fill_cols; col++) {</div><div class="line"><a name="l00061"></a><span class="lineno"> 61</span>  Tout &p = out[(ybase + row) * ldc + xbase + col];</div><div class="line"><a name="l00062"></a><span class="lineno"> 62</span> </div><div class="line"><a name="l00063"></a><span class="lineno"> 63</span>  <span class="comment">// Special case for beta==0 - don't read the input;</span></div><div class="line"><a name="l00064"></a><span class="lineno"> 64</span>  <span class="comment">// (0 * x == 0) is not always true for FP types.</span></div><div class="line"><a name="l00065"></a><span class="lineno"> 65</span>  <span class="keywordflow">if</span> (beta == static_cast<Tout>(0)) {</div><div class="line"><a name="l00066"></a><span class="lineno"> 66</span>  p = (<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3161c2c93c655dd30953372064ec627b">alpha</a> * in[row * width + col]);</div><div class="line"><a name="l00067"></a><span class="lineno"> 67</span>  } <span class="keywordflow">else</span> {</div><div class="line"><a name="l00068"></a><span class="lineno"> 68</span>  p = (p * beta) + (<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3161c2c93c655dd30953372064ec627b">alpha</a> * in[row * width + col]);</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>  }</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> </div><div class="line"><a name="l00073"></a><span class="lineno"> 73</span>  in += (width * height);</div><div class="line"><a name="l00074"></a><span class="lineno"> 74</span>  }</div><div class="line"><a name="l00075"></a><span class="lineno"> 75</span>  }</div><div class="line"><a name="l00076"></a><span class="lineno"> 76</span> }</div><div class="ttc" id="namespacearm__compute_1_1test_1_1validation_xhtml_a3161c2c93c655dd30953372064ec627b"><div class="ttname"><a href="namespacearm__compute_1_1test_1_1validation.xhtml#a3161c2c93c655dd30953372064ec627b">arm_compute::test::validation::alpha</a></div><div class="ttdeci">const float alpha</div><div class="ttdef"><b>Definition:</b> <a href="_n_e_o_n_2_accumulate_8cpp_source.xhtml#l00103">Accumulate.cpp:103</a></div></div>
+<p class="definition">Definition at line <a class="el" href="mergeresults_8cpp_source.xhtml#l00038">38</a> of file <a class="el" href="mergeresults_8cpp_source.xhtml">mergeresults.cpp</a>.</p>
+<div class="fragment"><div class="line"><a name="l00038"></a><span class="lineno"> 38</span>  {</div><div class="line"><a name="l00039"></a><span class="lineno"> 39</span>  <span class="comment">// For SVE cases, multiply the width up by the vector length.</span></div><div class="line"><a name="l00040"></a><span class="lineno"> 40</span>  <span class="comment">// Use the *input* type to determine this, since this will be what the kernel operated on.</span></div><div class="line"><a name="l00041"></a><span class="lineno"> 41</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> width = twidth * (sve ? get_vector_length<Tin>() : 1);</div><div class="line"><a name="l00042"></a><span class="lineno"> 42</span> </div><div class="line"><a name="l00043"></a><span class="lineno"> 43</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> full_y_blocks = (ymax - y0) / height;</div><div class="line"><a name="l00044"></a><span class="lineno"> 44</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> y_remainder = (ymax - y0) % height;</div><div class="line"><a name="l00045"></a><span class="lineno"> 45</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> y_blocks = full_y_blocks + (y_remainder ? 1 : 0);</div><div class="line"><a name="l00046"></a><span class="lineno"> 46</span> </div><div class="line"><a name="l00047"></a><span class="lineno"> 47</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> full_x_blocks = (xmax - x0) / width;</div><div class="line"><a name="l00048"></a><span class="lineno"> 48</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> x_remainder = (xmax - x0) % width;</div><div class="line"><a name="l00049"></a><span class="lineno"> 49</span>  <span class="keyword">const</span> <span class="keywordtype">int</span> x_blocks = full_x_blocks + (x_remainder ? 1 : 0);</div><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>  <span class="keywordflow">for</span> (<span class="keywordtype">int</span> y_block = 0; y_block < y_blocks; y_block++) {</div><div class="line"><a name="l00052"></a><span class="lineno"> 52</span>  <span class="keywordtype">int</span> ybase = y0 + (y_block * height);</div><div class="line"><a name="l00053"></a><span class="lineno"> 53</span> </div><div class="line"><a name="l00054"></a><span class="lineno"> 54</span>  <span class="keywordtype">int</span> fill_rows = (y_block < full_y_blocks) ? height : y_remainder;</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>  <span class="keywordflow">for</span> (<span class="keywordtype">int</span> x_block = 0; x_block < x_blocks; x_block++) {</div><div class="line"><a name="l00057"></a><span class="lineno"> 57</span>  <span class="keywordtype">int</span> xbase = x0 + (x_block * width);</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>  <span class="keywordtype">int</span> fill_cols = (x_block < full_x_blocks) ? width : x_remainder;</div><div class="line"><a name="l00060"></a><span class="lineno"> 60</span> </div><div class="line"><a name="l00061"></a><span class="lineno"> 61</span>  <span class="keywordflow">for</span> (<span class="keywordtype">int</span> row=0; row < fill_rows; row++) {</div><div class="line"><a name="l00062"></a><span class="lineno"> 62</span>  <span class="keywordflow">for</span> (<span class="keywordtype">int</span> col=0; col < fill_cols; col++) {</div><div class="line"><a name="l00063"></a><span class="lineno"> 63</span>  Tout &r = out[(ybase + row) * ldc + xbase + col];</div><div class="line"><a name="l00064"></a><span class="lineno"> 64</span>  Tout v = in[row * width + col];</div><div class="line"><a name="l00065"></a><span class="lineno"> 65</span> </div><div class="line"><a name="l00066"></a><span class="lineno"> 66</span>  <span class="keywordflow">if</span> (append) {</div><div class="line"><a name="l00067"></a><span class="lineno"> 67</span>  v += r;</div><div class="line"><a name="l00068"></a><span class="lineno"> 68</span>  }</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>  <span class="keywordflow">if</span> (<a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>) {</div><div class="line"><a name="l00071"></a><span class="lineno"> 71</span>  v += <a class="code" href="namespacearm__compute_1_1test_1_1validation.xhtml#a3a77be8aebd8e00522b32061d46ccdbd">bias</a>[xbase + col];</div><div class="line"><a name="l00072"></a><span class="lineno"> 72</span>  }</div><div class="line"><a name="l00073"></a><span class="lineno"> 73</span> </div><div class="line"><a name="l00074"></a><span class="lineno"> 74</span>  <span class="keywordflow">switch</span>(act.type) {</div><div class="line"><a name="l00075"></a><span class="lineno"> 75</span>  <span class="keywordflow">default</span>:</div><div class="line"><a name="l00076"></a><span class="lineno"> 76</span>  <span class="keywordflow">case</span> Activation::Type::None:</div><div class="line"><a name="l00077"></a><span class="lineno"> 77</span>  <span class="keywordflow">break</span>;</div><div class="line"><a name="l00078"></a><span class="lineno"> 78</span> </div><div class="line"><a name="l00079"></a><span class="lineno"> 79</span>  <span class="keywordflow">case</span> Activation::Type::ReLU:</div><div class="line"><a name="l00080"></a><span class="lineno"> 80</span>  v = std::max(v, static_cast<Tout>(0));</div><div class="line"><a name="l00081"></a><span class="lineno"> 81</span>  <span class="keywordflow">break</span>;</div><div class="line"><a name="l00082"></a><span class="lineno"> 82</span> </div><div class="line"><a name="l00083"></a><span class="lineno"> 83</span>  <span class="keywordflow">case</span> Activation::Type::BoundedReLU:</div><div class="line"><a name="l00084"></a><span class="lineno"> 84</span>  v = std::max(std::min(v, static_cast<Tout>(act.param1)), static_cast<Tout>(0));</div><div class="line"><a name="l00085"></a><span class="lineno"> 85</span>  <span class="keywordflow">break</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> </div><div class="line"><a name="l00088"></a><span class="lineno"> 88</span>  r = v;</div><div class="line"><a name="l00089"></a><span class="lineno"> 89</span>  }</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> </div><div class="line"><a name="l00092"></a><span class="lineno"> 92</span>  in += (width * height);</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>  }</div><div class="line"><a name="l00095"></a><span class="lineno"> 95</span> }</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><!-- fragment -->
-<p class="reference">References <a class="el" href="_n_e_o_n_2_accumulate_8cpp_source.xhtml#l00103">arm_compute::test::validation::alpha</a>.</p>
+<p class="reference">References <a class="el" href="validation_2_c_l_2_convolution_layer_8cpp_source.xhtml#l00181">arm_compute::test::validation::bias</a>.</p>
<p class="reference">Referenced by <a class="el" href="std__transforms__sve_8hpp_source.xhtml#l00066">StdTransformsSVE< TOperand, TResult, height, width_vectors, block, mmla >::Merge()</a>.</p>
@@ -786,7 +927,7 @@
<div id="nav-path" class="navpath"><!-- id is needed for treeview function! -->
<ul>
<li class="navelem"><a class="el" href="namespacearm__gemm.xhtml">arm_gemm</a></li>
- <li class="footer">Generated on Mon Sep 2 2019 11:47:42 for Compute Library by
+ <li class="footer">Generated on Thu Nov 28 2019 16:53:21 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>