tExpanded profiling information - sphere - GPU-based 3D discrete element method algorithm with optional fluid coupling
(HTM) git clone git://src.adamsgaard.dk/sphere
(DIR) Log
(DIR) Files
(DIR) Refs
(DIR) LICENSE
---
(DIR) commit cc37e07feeb2da74ccd3d284296f2d75a9636500
(DIR) parent c8fda5738803686667567217cc927beb0c39a0fe
(HTM) Author: Anders Damsgaard <adc@geo.au.dk>
Date: Tue, 16 Oct 2012 12:26:51 +0200
Expanded profiling information
Diffstat:
M src/device.cu | 26 ++++++++++++++++++--------
1 file changed, 18 insertions(+), 8 deletions(-)
---
(DIR) diff --git a/src/device.cu b/src/device.cu
t@@ -713,15 +713,25 @@ __host__ void gpuMain(Float4* host_x,
// Report time spent on each kernel
if (PROFILING == 1) {
+ double t_sum = t_calcParticleCellID + t_thrustsort + t_reorderArrays
+ + t_topology + t_interact + t_summation + t_integrateWalls;
cout << "\nKernel profiling statistics:\n"
- << " - calcParticleCellID:\t" << t_calcParticleCellID/1000.0 << " s\n"
- << " - thrustsort:\t\t" << t_thrustsort/1000.0 << " s\n"
- << " - reorderArrays:\t" << t_reorderArrays/1000.0 << " s\n"
- << " - topology:\t\t" << t_topology/1000.0 << " s\n"
- << " - interact:\t\t" << t_interact/1000.0 << " s\n"
- << " - integrate:\t\t" << t_integrate/1000.0 << " s\n"
- << " - summation:\t\t" << t_summation/1000.0 << " s\n"
- << " - integrateWalls:\t" << t_integrateWalls/1000.0 << " s\n";
+ << " - calcParticleCellID:\t" << t_calcParticleCellID/1000.0 << " s"
+ << " (" << 100.0*t_calcParticleCellID/t_sum << " %)\n"
+ << " - thrustsort:\t\t" << t_thrustsort/1000.0 << " s"
+ << " (" << 100.0*t_thrustsort/t_sum << " %)\n"
+ << " - reorderArrays:\t" << t_reorderArrays/1000.0 << " s"
+ << " (" << 100.0*t_reorderArrays/t_sum << " %)\n"
+ << " - topology:\t\t" << t_topology/1000.0 << " s"
+ << " (" << 100.0*t_topology/t_sum << " %)\n"
+ << " - interact:\t\t" << t_interact/1000.0 << " s"
+ << " (" << 100.0*t_interact/t_sum << " %)\n"
+ << " - integrate:\t\t" << t_integrate/1000.0 << " s"
+ << " (" << 100.0*t_integrate/t_sum << " %)\n"
+ << " - summation:\t\t" << t_summation/1000.0 << " s"
+ << " (" << 100.0*t_summation/t_sum << " %)\n"
+ << " - integrateWalls:\t" << t_integrateWalls/1000.0 << " s"
+ << " (" << 100.0*t_integrateWalls/t_sum << " %)\n";
}