Shuenhoy/vtk-m2
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Prevent compiler from optimizing out benchmark body.

Browse Source 
Since the result of the Reduce call was not used, several compilers
were omitting the call completely on release builds.
This commit is contained in:
Allison Vacanti 2017-09-19 14:22:42 -04:00
parent c4c05d8b72
commit de44c58bb2
1 changed files with 12 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
vtkm/benchmarking/BenchmarkDeviceAdapter.cxx
View File

@ -298,6 +298,9 @@ private:
typedef vtkm::cont::ArrayHandle<Value, StorageTag> ValueArrayHandle;
ValueArrayHandle InputHandle;
// We don't actually use this, but we need it to prevent sufficently
// smart compilers from optimizing the Reduce call out.
Value Result;
VTKM_CONT
BenchReduce()
@ -305,14 +308,21 @@ private:
Algorithm::Schedule(
FillTestValueKernel<Value>(InputHandle.PrepareForOutput(ARRAY_SIZE, DeviceAdapterTag())),
ARRAY_SIZE);
this->Result =
Algorithm::Reduce(this->InputHandle, vtkm::TypeTraits<Value>::ZeroInitialization());
}
VTKM_CONT
vtkm::Float64 operator()()
{
Timer timer;
Algorithm::Reduce(InputHandle, Value());
return timer.GetElapsedTime();
Value tmp = Algorithm::Reduce(InputHandle, vtkm::TypeTraits<Value>::ZeroInitialization());
vtkm::Float64 time = timer.GetElapsedTime();
if (tmp != this->Result)
{
this->Result = tmp;
}
return time;
}
VTKM_CONT