mirror of
https://github.com/NVIDIA/cuda-samples.git
synced 2024-11-25 12:29:17 +08:00
101 lines
2.7 KiB
XML
101 lines
2.7 KiB
XML
<?xml version="1.0" encoding="UTF-8"?>
|
|
<!DOCTYPE entry SYSTEM "SamplesInfo.dtd">
|
|
<entry>
|
|
<name>shfl_scan</name>
|
|
<cflags>
|
|
<flag>--std=c++11</flag>
|
|
<flag>-O3</flag>
|
|
</cflags>
|
|
<cuda_api_list>
|
|
<toolkit>cudaMemset</toolkit>
|
|
<toolkit>cudaFree</toolkit>
|
|
<toolkit>cudaEventRecord</toolkit>
|
|
<toolkit>cudaMallocHost</toolkit>
|
|
<toolkit>cudaEventCreate</toolkit>
|
|
<toolkit>cudaEventElapsedTime</toolkit>
|
|
<toolkit>cudaEventSynchronize</toolkit>
|
|
<toolkit>cudaFreeHost</toolkit>
|
|
<toolkit>cudaMalloc</toolkit>
|
|
<toolkit>cudaMemcpy</toolkit>
|
|
<toolkit>cudaGetDeviceProperties</toolkit>
|
|
<toolkit>cudaGetDevice</toolkit>
|
|
</cuda_api_list>
|
|
<description><![CDATA[This example demonstrates how to use the shuffle intrinsic __shfl_up_sync to perform a scan operation across a thread block. ]]></description>
|
|
<devicecompilation>whole</devicecompilation>
|
|
<includepaths>
|
|
<path>./</path>
|
|
<path>../</path>
|
|
<path>../../../Common</path>
|
|
</includepaths>
|
|
<keyconcepts>
|
|
<concept level="advanced">Data-Parallel Algorithms</concept>
|
|
<concept level="advanced">Performance Strategies</concept>
|
|
</keyconcepts>
|
|
<keywords>
|
|
<keyword>GPGPU</keyword>
|
|
<keyword>CPP11</keyword>
|
|
<keyword>CUDA</keyword>
|
|
<keyword>scan</keyword>
|
|
<keyword>parallel prefix sum</keyword>
|
|
<keyword>Data-Parallel Algorithms</keyword>
|
|
</keywords>
|
|
<libraries>
|
|
</libraries>
|
|
<librarypaths>
|
|
</librarypaths>
|
|
<nsight_eclipse>true</nsight_eclipse>
|
|
<primary_file>shfl_scan.cu</primary_file>
|
|
<required_dependencies>
|
|
<dependency>CPP11</dependency>
|
|
</required_dependencies>
|
|
<scopes>
|
|
<scope>1:CUDA Advanced Topics</scope>
|
|
<scope>1:Data-Parallel Algorithms</scope>
|
|
<scope>1:Performance Strategies</scope>
|
|
</scopes>
|
|
<sm-arch>sm35</sm-arch>
|
|
<sm-arch>sm37</sm-arch>
|
|
<sm-arch>sm50</sm-arch>
|
|
<sm-arch>sm52</sm-arch>
|
|
<sm-arch>sm53</sm-arch>
|
|
<sm-arch>sm60</sm-arch>
|
|
<sm-arch>sm61</sm-arch>
|
|
<sm-arch>sm70</sm-arch>
|
|
<sm-arch>sm72</sm-arch>
|
|
<sm-arch>sm75</sm-arch>
|
|
<sm-arch>sm80</sm-arch>
|
|
<sm-arch>sm86</sm-arch>
|
|
<sm-arch>sm87</sm-arch>
|
|
<supported_envs>
|
|
<env>
|
|
<arch>x86_64</arch>
|
|
<platform>linux</platform>
|
|
</env>
|
|
<env>
|
|
<platform>windows7</platform>
|
|
</env>
|
|
<env>
|
|
<arch>x86_64</arch>
|
|
<platform>macosx</platform>
|
|
</env>
|
|
<env>
|
|
<arch>arm</arch>
|
|
</env>
|
|
<env>
|
|
<arch>aarch64</arch>
|
|
</env>
|
|
<env>
|
|
<arch>sbsa</arch>
|
|
</env>
|
|
<env>
|
|
<arch>ppc64le</arch>
|
|
<platform>linux</platform>
|
|
</env>
|
|
</supported_envs>
|
|
<supported_sm_architectures>
|
|
<from>3.5</from>
|
|
</supported_sm_architectures>
|
|
<title>CUDA Parallel Prefix Sum with Shuffle Intrinsics (SHFL_Scan)</title>
|
|
<type>exe</type>
|
|
</entry>
|