Merge pull request #264 from ComputationalRadiationPhysics/develop

merge develop to master

Author: BenjaminW3

README.md

@@ -50,7 +50,6 @@ Accelerator Back-ends
 |OpenMP 4.0+ (CPU)|OpenMP 4.0+|Host CPU (multi core)|parallel (undefined)|parallel (preemptive multitasking)|
 | std::thread | std::thread |Host CPU (multi core)|sequential|parallel (preemptive multitasking)|
 |CUDA 7.0+|CUDA 7.0+|NVIDIA GPUs SM 2.0+|parallel (undefined)|parallel (lock-step within warps)|
-|TBB 2.2+ blocks|TBB 2.2+|Host CPU (multi core)|parallel (preemptive multitasking)|sequential (only 1 thread per block)|
 
 
 Supported Compilers
@@ -66,7 +65,6 @@ This library uses C++11 (or newer when available).
 |OpenMP 4.0+ (CPU)|:white_check_mark:|:white_check_mark:|:white_check_mark:|:x:|:x:|:x:|:x:|
 | std::thread |:white_check_mark:|:white_check_mark:|:white_check_mark:|:white_check_mark:|:white_check_mark:|:white_check_mark:|:white_check_mark:|
 |CUDA 7.0+|:white_check_mark: (nvcc 7.0+)|:x:|:x:|:x:|:x:|:white_check_mark: (native)|:x:|
-|TBB 2.2+|:white_check_mark:|:white_check_mark:|:white_check_mark:|:white_check_mark:|:white_check_mark:|:white_check_mark:|:grey_question:|
 
 
 Dependencies
@@ -81,8 +79,6 @@ When an accelerator back-end using *CUDA* is enabled, version *7.0* of the *CUDA
 
 When an accelerator back-end using *OpenMP* is enabled, the compiler and the platform have to support the corresponding minimum *OpenMP* version.
 
-When an accelerator back-end using *TBB* is enabled, the compiler and the platform have to support the corresponding minimum *TBB* version.
-
 
 Usage
 -----

cmake/dev.cmake

@@ -1,5 +1,5 @@
 #
-# Copyright 2014-2015 Benjamin Worpitz
+# Copyright 2014-2016 Benjamin Worpitz
 #
 # This file is part of alpaka.
 #
@@ -24,8 +24,9 @@
 
 #MSVC
 IF(MSVC)
-    # Force to always compile with W4
+    # Force to always compile with W4 and WX
     LIST(APPEND ALPAKA_DEV_COMPILE_OPTIONS "/W4")
+    LIST(APPEND ALPAKA_DEV_COMPILE_OPTIONS "/WX")
     # Improve debugging.
     IF(CMAKE_BUILD_TYPE MATCHES "Debug")
         LIST(APPEND ALPAKA_DEV_COMPILE_OPTIONS "-d2Zi+")
@@ -34,11 +35,15 @@ ELSE()
     # GNU
     IF(CMAKE_COMPILER_IS_GNUCXX)
         LIST(APPEND ALPAKA_DEV_COMPILE_OPTIONS "-Wall")
-        LIST(APPEND ALPAKA_DEV_COMPILE_OPTIONS "-pedantic")
         LIST(APPEND ALPAKA_DEV_COMPILE_OPTIONS "-Wextra")
+        LIST(APPEND ALPAKA_DEV_COMPILE_OPTIONS "-pedantic")
+        LIST(APPEND ALPAKA_DEV_COMPILE_OPTIONS "-Werror")
     # Clang or AppleClang
     ELSEIF(CMAKE_CXX_COMPILER_ID MATCHES "Clang")
         LIST(APPEND ALPAKA_DEV_COMPILE_OPTIONS "-Wall")
+        LIST(APPEND ALPAKA_DEV_COMPILE_OPTIONS "-Wextra")
+        LIST(APPEND ALPAKA_DEV_COMPILE_OPTIONS "-pedantic")
+        LIST(APPEND ALPAKA_DEV_COMPILE_OPTIONS "-Werror")
     # ICC
     ELSEIF(${CMAKE_CXX_COMPILER_ID} STREQUAL "Intel")
         LIST(APPEND ALPAKA_DEV_COMPILE_OPTIONS "-Wall")

doc/doxygen/Doxyfile

@@ -1,4 +1,4 @@
-# Doxyfile 1.8.10
+# Doxyfile 1.8.11
 
 # This file describes the settings to be used by the documentation system
 # doxygen (www.doxygen.org) for a project.
@@ -749,6 +749,12 @@ WARN_IF_DOC_ERROR      = YES
 
 WARN_NO_PARAMDOC       = YES
 
+# If the WARN_AS_ERROR tag is set to YES then doxygen will immediately stop when
+# a warning is encountered.
+# The default value is: NO.
+
+WARN_AS_ERROR          = NO
+
 # The WARN_FORMAT tag determines the format of the warning messages that doxygen
 # can produce. The string should contain the $file, $line, and $text tags, which
 # will be replaced by the file and line number from which the warning originated
@@ -777,9 +783,6 @@ WARN_LOGFILE           =
 
 INPUT                  = ../../include/ \
                          ../../README.md \
-                         ../Distinction.md \
-                         ../Abstraction.md \
-                         ../Threading.md \
                          ../Rationale.md
 
 # This tag can be used to specify the character encoding of the source files
@@ -802,8 +805,8 @@ INPUT_ENCODING         = UTF-8
 # If left blank the following patterns are tested:*.c, *.cc, *.cxx, *.cpp,
 # *.c++, *.java, *.ii, *.ixx, *.ipp, *.i++, *.inl, *.idl, *.ddl, *.odl, *.h,
 # *.hh, *.hxx, *.hpp, *.h++, *.cs, *.d, *.php, *.php4, *.php5, *.phtml, *.inc,
-# *.m, *.markdown, *.md, *.mm, *.dox, *.py, *.f90, *.f, *.for, *.tcl, *.vhd,
-# *.vhdl, *.ucf, *.qsf, *.as and *.js.
+# *.m, *.markdown, *.md, *.mm, *.dox, *.py, *.pyw, *.f90, *.f, *.for, *.tcl,
+# *.vhd, *.vhdl, *.ucf, *.qsf, *.as and *.js.
 
 FILE_PATTERNS          = *.c \
                          *.cc \
@@ -932,6 +935,10 @@ IMAGE_PATH             =
 # Note that the filter must not add or remove lines; it is applied before the
 # code is scanned, but not when the output code is generated. If lines are added
 # or removed, the anchors will not be placed correctly.
+#
+# Note that for custom extensions or not directly supported extensions you also
+# need to set EXTENSION_MAPPING for the extension otherwise the files are not
+# properly processed by doxygen.
 
 INPUT_FILTER           = 
 
@@ -941,6 +948,10 @@ INPUT_FILTER           =
 # (like *.cpp=my_cpp_filter). See INPUT_FILTER for further information on how
 # filters are used. If the FILTER_PATTERNS tag is empty or if none of the
 # patterns match the file name, INPUT_FILTER is applied.
+#
+# Note that for custom extensions or not directly supported extensions you also
+# need to set EXTENSION_MAPPING for the extension otherwise the files are not
+# properly processed by doxygen.
 
 FILTER_PATTERNS        = 
 
@@ -1058,7 +1069,7 @@ VERBATIM_HEADERS       = YES
 # rich C++ code for which doxygen's built-in parser lacks the necessary type
 # information.
 # Note: The availability of this option depends on whether or not doxygen was
-# compiled with the --with-libclang option.
+# generated with the -Duse-libclang=ON option for CMake.
 # The default value is: NO.
 
 CLANG_ASSISTED_PARSING = NO
@@ -1802,6 +1813,14 @@ LATEX_SOURCE_CODE      = NO
 
 LATEX_BIB_STYLE        = plain
 
+# If the LATEX_TIMESTAMP tag is set to YES then the footer of each generated
+# page will contain the date and time when the page was generated. Setting this
+# to NO can help when comparing the output of multiple runs.
+# The default value is: NO.
+# This tag requires that the tag GENERATE_LATEX is set to YES.
+
+LATEX_TIMESTAMP        = NO
+
 #---------------------------------------------------------------------------
 # Configuration options related to the RTF output
 #---------------------------------------------------------------------------

example/bufferCopy/src/bufferCopy.cpp

@@ -169,8 +169,8 @@ auto main()
     /***************************************************************************
      * Get the first device
      **************************************************************************/
-    DevAcc devAcc(alpaka::pltf::getDevByIdx<PltfAcc>(0u));
-    DevHost devHost(alpaka::pltf::getDevByIdx<PltfHost>(0u));
+    DevAcc const devAcc(alpaka::pltf::getDevByIdx<PltfAcc>(0u));
+    DevHost const devHost(alpaka::pltf::getDevByIdx<PltfHost>(0u));
 
     /***************************************************************************
      * Create sync stream

example/helloWorld/src/helloWorld.cpp

@@ -117,8 +117,8 @@ auto main()
      * can also retrieve all devices in a vector (getDevs()).
      * In this example the first devices is choosen.
      */
-    DevAcc devAcc(alpaka::pltf::getDevByIdx<PltfAcc>(0u));
-    DevHost devHost(alpaka::pltf::getDevByIdx<PltfHost>(0u));
+    DevAcc const devAcc(alpaka::pltf::getDevByIdx<PltfAcc>(0u));
+    DevHost const devHost(alpaka::pltf::getDevByIdx<PltfHost>(0u));
 
     /**
      * Create a stream to the accelerator device

example/vectorAdd/src/main.cpp

@@ -99,11 +99,11 @@ auto main()
     VectorAddKernel kernel;
 
     // Get the host device.
-    auto devHost(
+    auto const devHost(
         alpaka::pltf::getDevByIdx<PltfHost>(0u));
 
     // Select a device to execute on.
-    auto devAcc(
+    auto const devAcc(
         alpaka::pltf::getDevByIdx<PltfAcc>(0));
 
     // Get a stream on this device.

include/alpaka/idx/Traits.hpp

@@ -34,6 +34,9 @@
 #include <alpaka/workdiv/Traits.hpp>        // workdiv::getWorkDiv
 
 #include <boost/config.hpp>                 // BOOST_NO_CXX14_RETURN_TYPE_DEDUCTION
+#if !BOOST_ARCH_CUDA_DEVICE
+    #include <boost/core/ignore_unused.hpp> // boost::ignore_unused
+#endif
 
 #include <utility>                          // std::forward
 
@@ -228,6 +231,9 @@ namespace alpaka
             TThreadElemExtent const & threadElemExtent)
         -> Vec<dim::Dim<TIdxWorkDiv>, size::Size<TIdxWorkDiv>>
         {
+#if !BOOST_ARCH_CUDA_DEVICE
+            boost::ignore_unused(idxWorkDiv);
+#endif
             return gridThreadIdx * threadElemExtent;
         }
         //-----------------------------------------------------------------------------

include/alpaka/mem/buf/BufCpu.hpp

@@ -37,6 +37,10 @@
 
 #include <alpaka/meta/DependentFalseType.hpp>   // meta::DependentFalseType
 
+#if !BOOST_ARCH_CUDA_DEVICE
+    #include <boost/core/ignore_unused.hpp>     // boost::ignore_unused
+#endif
+
 #include <cassert>                              // assert
 #include <memory>                               // std::shared_ptr
 
@@ -638,6 +642,7 @@ namespace alpaka
 #if defined(ALPAKA_ACC_GPU_CUDA_ENABLED) && BOOST_LANG_CUDA
                         return bufImpl.m_bPinned;
 #else
+                        boost::ignore_unused(bufImpl);
                         return false;
 #endif
                     }

include/alpaka/mem/view/Traits.hpp

@@ -29,7 +29,6 @@
 #include <alpaka/stream/Traits.hpp>     // stream::enqueue
 
 #include <alpaka/vec/Vec.hpp>           // Vec
-
 #include <alpaka/meta/Fold.hpp>         // meta::foldr
 #include <alpaka/core/Common.hpp>       // ALPAKA_FN_HOST
 
@@ -566,13 +565,7 @@ namespace alpaka
                     rowPrefix,
                     rowSuffix);
             }
-        }
-    }
 
-    namespace mem
-    {
-        namespace view
-        {
             namespace detail
             {
                 //#############################################################################

include/alpaka/mem/view/ViewPlainPtr.hpp

@@ -57,15 +57,8 @@ namespace alpaka
                         m_pMem(pMem),
                         m_dev(dev),
                         m_extentElements(extent::getExtentVecEnd<TDim>(extent)),
-                        m_pitchBytes(Vec<TDim, TSize>::all(0))
-                {
-                    // Calculate the pitches by hand.
-                    m_pitchBytes[TDim::value - 1u] = extent[TDim::value - 1u] * sizeof(TElem);
-                    for(TSize i = TDim::value - 1u; i > static_cast<TSize>(0u); --i)
-                    {
-                        m_pitchBytes[i-1] = extent[i-1] * m_pitchBytes[i];
-                    }
-                }
+                        m_pitchBytes(calculatePitchesFromExtents(m_extentElements))
+                {}
 
                 //-----------------------------------------------------------------------------
                 //! Constructor
@@ -116,11 +109,31 @@ namespace alpaka
                 ALPAKA_NO_HOST_ACC_WARNING
                 ALPAKA_FN_HOST_ACC ~ViewPlainPtr() = default;
 
+            private:
+                //-----------------------------------------------------------------------------
+                //! Calculate the pitches purely from the extents.
+                //-----------------------------------------------------------------------------
+                ALPAKA_NO_HOST_ACC_WARNING
+                template<
+                    typename TExtent>
+                ALPAKA_FN_HOST_ACC static auto calculatePitchesFromExtents(
+                    TExtent const & extent)
+                -> Vec<TDim, TSize>
+                {
+                    Vec<TDim, TSize> pitchBytes(Vec<TDim, TSize>::all(0));
+                    pitchBytes[TDim::value - 1u] = extent[TDim::value - 1u] * sizeof(TElem);
+                    for(TSize i = TDim::value - 1u; i > static_cast<TSize>(0u); --i)
+                    {
+                        pitchBytes[i-1] = extent[i-1] * pitchBytes[i];
+                    }
+                    return pitchBytes;
+                }
+
             public:
-                TElem * m_pMem;
-                TDev m_dev;
-                Vec<TDim, TSize> m_extentElements;
-                Vec<TDim, TSize> m_pitchBytes;
+                TElem * const m_pMem;
+                TDev const m_dev;
+                Vec<TDim, TSize> const m_extentElements;
+                Vec<TDim, TSize> const m_pitchBytes;
             };
         }
     }