summaryrefslogtreecommitdiffstats
path: root/src/boost/libs/compute/perf/perf_thrust_partition.cu
diff options
context:
space:
mode:
authorDaniel Baumann <daniel.baumann@progress-linux.org>2024-04-27 18:24:20 +0000
committerDaniel Baumann <daniel.baumann@progress-linux.org>2024-04-27 18:24:20 +0000
commit483eb2f56657e8e7f419ab1a4fab8dce9ade8609 (patch)
treee5d88d25d870d5dedacb6bbdbe2a966086a0a5cf /src/boost/libs/compute/perf/perf_thrust_partition.cu
parentInitial commit. (diff)
downloadceph-upstream.tar.xz
ceph-upstream.zip
Adding upstream version 14.2.21.upstream/14.2.21upstream
Signed-off-by: Daniel Baumann <daniel.baumann@progress-linux.org>
Diffstat (limited to 'src/boost/libs/compute/perf/perf_thrust_partition.cu')
-rw-r--r--src/boost/libs/compute/perf/perf_thrust_partition.cu60
1 files changed, 60 insertions, 0 deletions
diff --git a/src/boost/libs/compute/perf/perf_thrust_partition.cu b/src/boost/libs/compute/perf/perf_thrust_partition.cu
new file mode 100644
index 00000000..5c89014c
--- /dev/null
+++ b/src/boost/libs/compute/perf/perf_thrust_partition.cu
@@ -0,0 +1,60 @@
+//---------------------------------------------------------------------------//
+// Copyright (c) 2013-2014 Kyle Lutz <kyle.r.lutz@gmail.com>
+//
+// Distributed under the Boost Software License, Version 1.0
+// See accompanying file LICENSE_1_0.txt or copy at
+// http://www.boost.org/LICENSE_1_0.txt
+//
+// See http://boostorg.github.com/compute for more information.
+//---------------------------------------------------------------------------//
+
+#include <algorithm>
+#include <cstdlib>
+#include <iostream>
+
+#include <thrust/copy.h>
+#include <thrust/device_vector.h>
+#include <thrust/generate.h>
+#include <thrust/host_vector.h>
+#include <thrust/partition.h>
+
+#include "perf.hpp"
+
+int rand_int()
+{
+ return static_cast<int>((rand() / double(RAND_MAX)) * 25.0);
+}
+
+struct less_than_ten : public thrust::unary_function<bool, int>
+{
+ __device__ bool operator()(int x) const
+ {
+ return x < 10;
+ }
+};
+
+int main(int argc, char *argv[])
+{
+ perf_parse_args(argc, argv);
+
+ std::cout << "size: " << PERF_N << std::endl;
+ thrust::host_vector<int> h_vec(PERF_N);
+ std::generate(h_vec.begin(), h_vec.end(), rand_int);
+
+ thrust::device_vector<int> d_vec(PERF_N);
+
+ perf_timer t;
+ for(size_t trial = 0; trial < PERF_TRIALS; trial++){
+ d_vec = h_vec;
+
+ t.start();
+ thrust::partition(
+ d_vec.begin(), d_vec.end(), less_than_ten()
+ );
+ cudaDeviceSynchronize();
+ t.stop();
+ }
+ std::cout << "time: " << t.min_time() / 1e6 << " ms" << std::endl;
+
+ return 0;
+}