Commit a6f3ee4d authored by Tomáš Oberhuber's avatar Tomáš Oberhuber
Browse files

Merge branch 'nina' into 'develop' 

Nina

See merge request !15
parents e3c60cbf 1a721bad

Doxyfile

+17 −2
Original line number Diff line number Diff line
@@ -889,7 +889,22 @@ EXCLUDE_SYMLINKS = NO 
# Note that the wildcards are matched against the file with absolute path, so to

# exclude all test directories for example use the pattern */test/*



EXCLUDE_PATTERNS       =

EXCLUDE_PATTERNS       = */Communicators/* \

                         */Config/* \

                         */Debugging/* \

                         */Devices/* \

                         */DistributedContainers/* \

                         */Exceptions/* \

                         */Experimental/* \

                         */Functions/* \

                         */Images/* \

                         */Matrices/* \

                         */Meshes/* \

                         */Operators/* \

                         */Pointers/* \

                         */Problems/* \

                         */Solvers/* 





# The EXCLUDE_SYMBOLS tag can be used to specify one or more symbol names

# (namespaces, classes, functions, etc.) that should be excluded from the
@@ -906,7 +921,7 @@ EXCLUDE_SYMBOLS = 
# that contain example code fragments that are included (see the \include

# command).



EXAMPLE_PATH           =

EXAMPLE_PATH           = src/Examples



# If the value of the EXAMPLE_PATH tag contains directories, you can use the

# EXAMPLE_PATTERNS tag to specify one or more wildcard pattern (like *.cpp and

generate-doc

+2 −1
Original line number Diff line number Diff line
@@ -15,6 +15,7 @@ done 
doxygen



if [[ "$INSTALL" == "yes" ]]; then

   cp -r Documentation/* ${PREFIX}/share/doc

   mkdir -p ${PREFIX}/share/doc/tnl

   cp -r Documentation/* ${PREFIX}/share/doc/tnl

fi

src/Benchmarks/BLAS/tnl-benchmark-blas.h

+5 −5
Original line number Diff line number Diff line
@@ -43,7 +43,7 @@ runBlasBenchmarks( Benchmark & benchmark, 
                           metadata );

   for( std::size_t size = minSize; size <= maxSize; size *= 2 ) {

      benchmark.setMetadataColumns( Benchmark::MetadataColumns({

         {"size", String( size ) },

         {"size", convertToString( size ) },

      } ));

      benchmarkArrayOperations< Real >( benchmark, loops, size );

   }
@@ -53,7 +53,7 @@ runBlasBenchmarks( Benchmark & benchmark, 
                           metadata );

   for( std::size_t size = minSize; size <= maxSize; size *= sizeStepFactor ) {

      benchmark.setMetadataColumns( Benchmark::MetadataColumns({

         { "size", String( size ) },

         { "size", convertToString( size ) },

      } ));

      benchmarkVectorOperations< Real >( benchmark, loops, size );

   }
@@ -63,9 +63,9 @@ runBlasBenchmarks( Benchmark & benchmark, 
                           metadata );

   for( std::size_t size = minSize; size <= maxSize; size *= 2 ) {

      benchmark.setMetadataColumns( Benchmark::MetadataColumns({

         {"rows", String( size ) },

         {"columns", String( size ) },

         {"elements per row", String( elementsPerRow ) },

         {"rows", convertToString( size ) },

         {"columns", convertToString( size ) },

         {"elements per row", convertToString( elementsPerRow ) },

      } ));

      benchmarkSpmvSynthetic< Real >( benchmark, loops, size, elementsPerRow );

   }

src/Benchmarks/Benchmarks.h

+13 −13
Original line number Diff line number Diff line
@@ -504,25 +504,25 @@ Benchmark::MetadataMap getHardwareMetadata() 
       { "system release", Devices::SystemInfo::getSystemRelease() },

       { "start time", Devices::SystemInfo::getCurrentTime() },

#ifdef HAVE_MPI

       { "number of MPI processes", (Communicators::MpiCommunicator::IsInitialized())

       { "number of MPI processes", String( (Communicators::MpiCommunicator::IsInitialized())

                                       ? Communicators::MpiCommunicator::GetSize( Communicators::MpiCommunicator::AllGroup )

                                       : 1 },

                                       : 1 ) },

#endif

       { "OpenMP enabled", Devices::Host::isOMPEnabled() },

       { "OpenMP threads", Devices::Host::getMaxThreadsCount() },

       { "OpenMP enabled", String( Devices::Host::isOMPEnabled() ) },

       { "OpenMP threads", String( Devices::Host::getMaxThreadsCount() ) },

       { "CPU model name", Devices::SystemInfo::getCPUModelName( cpu_id ) },

       { "CPU cores", Devices::SystemInfo::getNumberOfCores( cpu_id ) },

       { "CPU threads per core", Devices::SystemInfo::getNumberOfThreads( cpu_id ) / Devices::SystemInfo::getNumberOfCores( cpu_id ) },

       { "CPU max frequency (MHz)", Devices::SystemInfo::getCPUMaxFrequency( cpu_id ) / 1e3 },

       { "CPU cores", String( Devices::SystemInfo::getNumberOfCores( cpu_id ) ) },

       { "CPU threads per core", String( Devices::SystemInfo::getNumberOfThreads( cpu_id ) / Devices::SystemInfo::getNumberOfCores( cpu_id ) ) },

       { "CPU max frequency (MHz)", String( Devices::SystemInfo::getCPUMaxFrequency( cpu_id ) / 1e3 ) },

       { "CPU cache sizes (L1d, L1i, L2, L3) (kiB)", cacheInfo },

#ifdef HAVE_CUDA

       { "GPU name", Devices::CudaDeviceInfo::getDeviceName( activeGPU ) },

       { "GPU architecture", deviceArch },

       { "GPU CUDA cores", Devices::CudaDeviceInfo::getCudaCores( activeGPU ) },

       { "GPU clock rate (MHz)", (double) Devices::CudaDeviceInfo::getClockRate( activeGPU ) / 1e3 },

       { "GPU global memory (GB)", (double) Devices::CudaDeviceInfo::getGlobalMemory( activeGPU ) / 1e9 },

       { "GPU memory clock rate (MHz)", (double) Devices::CudaDeviceInfo::getMemoryClockRate( activeGPU ) / 1e3 },

       { "GPU memory ECC enabled", Devices::CudaDeviceInfo::getECCEnabled( activeGPU ) },

       { "GPU architecture", String( deviceArch ) },

       { "GPU CUDA cores", String( Devices::CudaDeviceInfo::getCudaCores( activeGPU ) ) },

       { "GPU clock rate (MHz)", String( (double) Devices::CudaDeviceInfo::getClockRate( activeGPU ) / 1e3 ) },

       { "GPU global memory (GB)", String( (double) Devices::CudaDeviceInfo::getGlobalMemory( activeGPU ) / 1e9 ) },

       { "GPU memory clock rate (MHz)", String( (double) Devices::CudaDeviceInfo::getMemoryClockRate( activeGPU ) / 1e3 ) },

       { "GPU memory ECC enabled", String( Devices::CudaDeviceInfo::getECCEnabled( activeGPU ) ) },

#endif

   };

src/Benchmarks/DistSpMV/tnl-benchmark-distributed-spmv.h

+3 −3
Original line number Diff line number Diff line
@@ -177,11 +177,11 @@ struct SpmvBenchmark 
      benchmark.setMetadataColumns( Benchmark::MetadataColumns({

         // TODO: strip the device

//         {"matrix type", matrix.getType()},

         {"rows", matrix.getRows()},

         {"columns", matrix.getColumns()},

         {"rows", String( matrix.getRows() ) },

         {"columns", String( matrix.getColumns() ) },

         // FIXME: getMaxRowLengths() returns 0 for matrices loaded from file

//         {"max elements per row", matrix.getMaxRowLength()},

         {"max elements per row", maxRowLength},

         {"max elements per row", String( maxRowLength ) },

      } ));



      const bool reorder = parameters.getParameter< bool >( "reorder-dofs" );