cuda-samples/Samples/3_CUDA_Features/warpAggregatedAtomicsCG/NsightEclipse.xml

<?xml version="1.0" encoding="UTF-8"?> 
<!DOCTYPE entry SYSTEM "SamplesInfo.dtd">
<entry>
  <name>warpAggregatedAtomicsCG</name>
  <cflags>
    <flag>--std=c++11</flag>
  </cflags>
  <cuda_api_list>
    <toolkit>cudaMemcpy</toolkit>
    <toolkit>cudaFree</toolkit>
    <toolkit>cudaDeviceGetAttribute</toolkit>
    <toolkit>cudaMemset</toolkit>
    <toolkit>cudaMalloc</toolkit>
  </cuda_api_list>
  <description><![CDATA[This sample demonstrates how using Cooperative Groups (CG) to perform warp aggregated atomics to single and multiple counters, a useful technique to improve performance when many threads atomically add to a single or multiple counters.]]></description>
  <includepaths>
    <path>./</path>
    <path>../</path>
    <path>../../../Common</path>
  </includepaths>
  <keyconcepts>
    <concept level="basic">Cooperative Groups</concept>
    <concept level="basic">Atomic Intrinsics</concept>
  </keyconcepts>
  <keywords>
    <keyword>GPGPU</keyword>
    <keyword>Cooperative Groups</keyword>
    <keyword>Atomic</keyword>
    <keyword>CPP11</keyword>
  </keywords>
  <libraries>
  </libraries>
  <librarypaths>
  </librarypaths>
  <nsight_eclipse>true</nsight_eclipse>
  <primary_file>warpAggregatedAtomicsCG.cu</primary_file>
  <scopes>
    <scope>1:CUDA Advanced Topics</scope>
  </scopes>
  <supported_envs>
    <env>
      <arch>x86_64</arch>
      <platform>linux</platform>
    </env>
    <env>
      <arch>ppc64le</arch>
      <platform>linux</platform>
    </env>
    <env>
      <arch>x86_64</arch>
      <platform>macosx</platform>
    </env>
    <env>
      <platform>windows7</platform>
    </env>
    <env>
      <arch>arm</arch>
    </env>
    <env>
      <arch>aarch64</arch>
    </env>
    <env>
      <arch>sbsa</arch>
    </env>
  </supported_envs>
  <supported_sm_architectures>
    <from>3.5</from>
  </supported_sm_architectures>
  <title>Warp Aggregated Atomics using Cooperative Groups</title>
</entry>