Refactored Makefile to include defines both to NVCC and g++.

aschi87 · Feb 12, 2016 · e8ab337 · e8ab337
1 parent e25e548
commit e8ab337
Show file tree

Hide file tree

Showing 3 changed files with 18 additions and 16 deletions.
diff --git a/Makefile b/Makefile
@@ -58,8 +58,10 @@ CXX = mpic++
 
 SOURCEDIR:= Source
 INCLUDEPATH:= $(addprefix $(SOURCEDIR)/, Common/Include Math CNTK ActionsLib ComputationNetworkLib SGDLib SequenceTrainingLib CNTK/BrainScript Readers/ReaderLib)
-CPPFLAGS:= -D_POSIX_SOURCE -D_XOPEN_SOURCE=600 -D__USE_XOPEN2K
-CXXFLAGS:= -msse3 -std=c++0x -std=c++11 -fopenmp -fpermissive -fPIC -Werror -fcheck-new
+# COMMON_FLAGS include settings that are passed both to NVCC and C++ compilers.
+COMMON_FLAGS:= -D_POSIX_SOURCE -D_XOPEN_SOURCE=600 -D__USE_XOPEN2K -std=c++11
+CPPFLAGS:= 
+CXXFLAGS:= -msse3 -std=c++0x -fopenmp -fpermissive -fPIC -Werror -fcheck-new
 LIBPATH:=
 LIBS:=
 LDFLAGS:=
@@ -78,7 +80,7 @@ SRC:=
 all : buildall
 
 # Set up basic nvcc options and add CUDA targets from above
-CUFLAGS = -std=c++11 -D_POSIX_SOURCE -D_XOPEN_SOURCE=600 -D__USE_XOPEN2K -m 64
+CUFLAGS = -m 64
 
 ifdef CUDA_PATH
   ifndef GDK_PATH
@@ -110,26 +112,26 @@ ifdef CUDA_PATH
     INCLUDEPATH += $(CUDNN_PATH)/cuda/include
     LIBPATH += $(CUDNN_PATH)/cuda/lib64
     LIBS += -lcudnn
-    CPPFLAGS +=-DUSE_CUDNN
+    COMMON_FLAGS +=-DUSE_CUDNN
   endif
 else
   DEVICE = cpu
 
-  CPPFLAGS +=-DCPUONLY
+  COMMON_FLAGS +=-DCPUONLY
 endif
 
 ifeq ("$(MATHLIB)","acml")
   INCLUDEPATH += $(ACML_PATH)/include
   LIBPATH += $(ACML_PATH)/lib
   LIBS += -lacml_mp -liomp5 -lm -lpthread
-  CPPFLAGS += -DUSE_ACML
+  COMMON_FLAGS += -DUSE_ACML
 endif
 
 ifeq ("$(MATHLIB)","mkl")
   INCLUDEPATH += $(MKL_PATH)/mkl/include
   LIBPATH += $(MKL_PATH)/compiler/lib/intel64 $(MKL_PATH)/mkl/lib/intel64 $(MKL_PATH)/compiler/lib/mic $(MKL_PATH)/mkl/lib/mic
   LIBS += -lmkl_intel_lp64 -lmkl_intel_thread -lmkl_core -lm -liomp5 -lpthread
-  CPPFLAGS += -DUSE_MKL
+  COMMON_FLAGS += -DUSE_MKL
 endif
 
 
@@ -161,7 +163,7 @@ ifeq ("$(BUILDTYPE)","debug")
 
   CXXFLAGS += -g
   LDFLAGS += -rdynamic
-  CPPFLAGS += -D_DEBUG
+  COMMON_FLAGS += -D_DEBUG
   CUFLAGS += -O0 -g -use_fast_math -lineinfo  $(GENCODE_FLAGS)
 endif
 
@@ -174,7 +176,7 @@ ifeq ("$(BUILDTYPE)","release")
 
   CXXFLAGS += -g -O4
   LDFLAGS += -rdynamic
-  CPPFLAGS += -DNDEBUG
+  COMMON_FLAGS += -DNDEBUG
   CUFLAGS += -O3 -g -use_fast_math -lineinfo $(GENCODE_FLAGS)
 endif
 
@@ -245,7 +247,7 @@ MATH_SRC +=\
 	$(SOURCEDIR)/Math/GPUSparseMatrix.cu \
 	$(SOURCEDIR)/Math/GPUWatcher.cu \
 	$(SOURCEDIR)/Math/MatrixQuantizerGPU.cu \
-	$(SOURCEDIR)/Math/CuDnnConvolutionEngine.cpp \
+	$(SOURCEDIR)/Math/CuDnnConvolutionEngine.cu \
 	$(SOURCEDIR)/Math/GPUDataTransferer.cpp \
 
 else
@@ -469,7 +471,7 @@ endif
 
   INCLUDEPATH += $(SOURCEDIR)/1BitSGD
 
-  CPPFLAGS += -DQUANTIZED_GRADIENT_AGGREGATION
+  COMMON_FLAGS += -DQUANTIZED_GRADIENT_AGGREGATION
 endif
 
 ########################################
@@ -549,13 +551,13 @@ $(OBJDIR)/%.o : %.cu Makefile
 	@echo $(SEPARATOR)
 	@echo creating $@ for $(ARCH) with build type $(BUILDTYPE)
 	@mkdir -p $(dir $@)
-	$(NVCC) -c $< -o $@  $(CUFLAGS) $(INCLUDEPATH:%=-I%) -Xcompiler "-fPIC -Werror"
+	$(NVCC) -c $< -o $@ $(COMMON_FLAGS) $(CUFLAGS) $(INCLUDEPATH:%=-I%) -Xcompiler "-fPIC -Werror"
 
 $(OBJDIR)/%.o : %.cpp Makefile
 	@echo $(SEPARATOR)
 	@echo creating $@ for $(ARCH) with build type $(BUILDTYPE)
 	@mkdir -p $(dir $@)
-	$(CXX) -c $< -o $@ $(CPPFLAGS) $(CXXFLAGS) $(INCLUDEPATH:%=-I%) -MD -MP -MF ${@:.o=.d}
+	$(CXX) -c $< -o $@ $(COMMON_FLAGS) $(CPPFLAGS) $(CXXFLAGS) $(INCLUDEPATH:%=-I%) -MD -MP -MF ${@:.o=.d}
 
 .PHONY: force clean buildall all
 

diff --git a/Source/Math/CuDnnConvolutionEngine.cuh b/Source/Math/CuDnnConvolutionEngine.cuh
@@ -119,7 +119,7 @@ namespace Microsoft { namespace MSR { namespace CNTK {
 #endif
     }
 
-    struct Operations
+    namespace Operations
     {
         template <typename T>
         struct RSqrt
@@ -147,7 +147,7 @@ namespace Microsoft { namespace MSR { namespace CNTK {
                 return rsqrt(a);
             }
         };
-    };
+    }
 
     // This function is used to select correct unroll factor.
     // REVIEW alexeyk: ask our C++ gurus (Marko/Amit) if there is better way.

diff --git a/Source/Math/GPUMatrix.cu b/Source/Math/GPUMatrix.cu
@@ -4300,7 +4300,7 @@ void GPUMatrix<ElemType>::CreateCurandObject(unsigned long seed, const char* cal
     if (s_curandGenerator == NULL)
     {
         unsigned long long cudaSeed = (seed == USE_TIME_BASED_SEED) ? time(NULL) : seed;
-        fprintf(stderr, "%s (GPU): creating curand object with seed %llu, sizeof(ElemType)==%u\n",
+        fprintf(stderr, "%s (GPU): creating curand object with seed %llu, sizeof(ElemType)==%lu\n",
                 caller, cudaSeed, sizeof(ElemType));
         s_curandGenerator = new curandGenerator_t;
         // Create pseudo-random number generator