Merge branch 'develop' of https://github.com/quickwritereader/OpenBLAS into develop

Author: quickwritereader

.travis.yml

@@ -149,7 +149,7 @@ matrix:
 
     - &test-macos
       os: osx
-      osx_image: xcode8.3
+      osx_image: xcode10.1
       before_script:
         - COMMON_FLAGS="DYNAMIC_ARCH=1 TARGET=NEHALEM NUM_THREADS=32"
         - brew update
@@ -160,6 +160,7 @@ matrix:
         - BTYPE="BINARY=64 INTERFACE64=1"
 
     - <<: *test-macos
+      osx_image: xcode8.3
       env:
         - BTYPE="BINARY=32"
 

CMakeLists.txt

@@ -42,6 +42,19 @@ endif()
 
 #######
 
+if(MSVC AND MSVC_STATIC_CRT)
+    set(CompilerFlags
+            CMAKE_CXX_FLAGS
+            CMAKE_CXX_FLAGS_DEBUG
+            CMAKE_CXX_FLAGS_RELEASE
+            CMAKE_C_FLAGS
+            CMAKE_C_FLAGS_DEBUG
+            CMAKE_C_FLAGS_RELEASE
+            )
+    foreach(CompilerFlag ${CompilerFlags})
+      string(REPLACE "/MD" "/MT" ${CompilerFlag} "${${CompilerFlag}}")
+    endforeach()
+endif()
 
 message(WARNING "CMake support is experimental. It does not yet support all build options and may not produce the same Makefiles that OpenBLAS ships with.")
 
@@ -62,10 +75,10 @@ endif ()
 
 set(SUBDIRS	${BLASDIRS})
 if (NOT NO_LAPACK)
-  list(APPEND SUBDIRS lapack)
   if(BUILD_RELAPACK)
     list(APPEND SUBDIRS relapack/src)
   endif()
+  list(APPEND SUBDIRS lapack)
 endif ()
 
 # set which float types we want to build for
@@ -134,7 +147,7 @@ endif ()
 
 # Only generate .def for dll on MSVC and always produce pdb files for debug and release
 if(MSVC)
-  if (${CMAKE_MAJOR_VERSION}.${CMAKE_MINOR_VERSION} LESS 3.4)
+  if (${CMAKE_MAJOR_VERSION}.${CMAKE_MINOR_VERSION} VERSION_LESS 3.4)
     set(OpenBLAS_DEF_FILE "${PROJECT_BINARY_DIR}/openblas.def")
   endif()
   set(CMAKE_C_FLAGS_RELEASE "${CMAKE_C_FLAGS_RELEASE} /Zi")
@@ -149,15 +162,9 @@ if (${DYNAMIC_ARCH})
   endforeach()
 endif ()
 
-# Only build shared libs for MSVC
-if (MSVC)
-  set(BUILD_SHARED_LIBS ON)
-endif()
-
-
 # add objects to the openblas lib
 add_library(${OpenBLAS_LIBNAME} ${LA_SOURCES} ${LAPACKE_SOURCES} ${RELA_SOURCES} ${TARGET_OBJS} ${OpenBLAS_DEF_FILE})
-target_include_directories(${OpenBLAS_LIBNAME} INTERFACE $<INSTALL_INTERFACE:include>)
+target_include_directories(${OpenBLAS_LIBNAME} INTERFACE $<INSTALL_INTERFACE:include/openblas${SUFFIX64}>)
 
 # Android needs to explicitly link against libm
 if(ANDROID)
@@ -166,7 +173,7 @@ endif()
 
 # Handle MSVC exports
 if(MSVC AND BUILD_SHARED_LIBS)
-  if (${CMAKE_MAJOR_VERSION}.${CMAKE_MINOR_VERSION} LESS 3.4)
+  if (${CMAKE_MAJOR_VERSION}.${CMAKE_MINOR_VERSION} VERSION_LESS 3.4)
     include("${PROJECT_SOURCE_DIR}/cmake/export.cmake")
   else()
     # Creates verbose .def file (51KB vs 18KB)
@@ -217,6 +224,14 @@ set_target_properties(${OpenBLAS_LIBNAME} PROPERTIES
   SOVERSION ${OpenBLAS_MAJOR_VERSION}
 )
 
+if (BUILD_SHARED_LIBS AND BUILD_RELAPACK)
+  if (NOT MSVC)
+    target_link_libraries(${OpenBLAS_LIBNAME} "-Wl,-allow-multiple-definition")
+  else()
+    target_link_libraries(${OpenBLAS_LIBNAME} "/FORCE:MULTIPLE")
+  endif()
+endif()
+
 if (BUILD_SHARED_LIBS AND NOT ${SYMBOLPREFIX}${SYMBOLSUFIX} STREQUAL "")
 if (NOT DEFINED ARCH)
   set(ARCH_IN "x86_64")
@@ -314,7 +329,7 @@ install (FILES ${OPENBLAS_CONFIG_H} DESTINATION ${CMAKE_INSTALL_INCLUDEDIR})
 if(NOT NOFORTRAN)
   message(STATUS "Generating f77blas.h in ${CMAKE_INSTALL_INCLUDEDIR}")
 
-  set(F77BLAS_H ${CMAKE_BINARY_DIR}/f77blas.h)
+  set(F77BLAS_H ${CMAKE_BINARY_DIR}/generated/f77blas.h)
   file(WRITE  ${F77BLAS_H} "#ifndef OPENBLAS_F77BLAS_H\n")
   file(APPEND ${F77BLAS_H} "#define OPENBLAS_F77BLAS_H\n")
   file(APPEND ${F77BLAS_H} "#include \"openblas_config.h\"\n")
@@ -327,10 +342,11 @@ endif()
 if(NOT NO_CBLAS)
 	message (STATUS "Generating cblas.h in ${CMAKE_INSTALL_INCLUDEDIR}")
 
+	set(CBLAS_H ${CMAKE_BINARY_DIR}/generated/cblas.h)
 	file(READ ${CMAKE_CURRENT_SOURCE_DIR}/cblas.h CBLAS_H_CONTENTS)
 	string(REPLACE "common" "openblas_config" CBLAS_H_CONTENTS_NEW "${CBLAS_H_CONTENTS}")
-	file(WRITE ${CMAKE_BINARY_DIR}/cblas.tmp "${CBLAS_H_CONTENTS_NEW}")
-	install (FILES ${CMAKE_BINARY_DIR}/cblas.tmp DESTINATION ${CMAKE_INSTALL_INCLUDEDIR} RENAME cblas.h)
+	file(WRITE ${CBLAS_H} "${CBLAS_H_CONTENTS_NEW}")
+	install (FILES ${CBLAS_H} DESTINATION ${CMAKE_INSTALL_INCLUDEDIR})
 endif()
 
 if(NOT NO_LAPACKE)

Makefile

@@ -96,7 +96,7 @@ endif
 	@echo
 
 shared :
-ifndef NO_SHARED
+ifneq ($(NO_SHARED), 1)
 ifeq ($(OSNAME), $(filter $(OSNAME),Linux SunOS Android Haiku))
 	@$(MAKE) -C exports so
 	@ln -fs $(LIBSONAME) $(LIBPREFIX).so

Makefile.arm64

@@ -38,3 +38,8 @@ ifeq ($(CORE), THUNDERX2T99)
 CCOMMON_OPT += -march=armv8.1-a -mtune=thunderx2t99
 FCOMMON_OPT += -march=armv8.1-a -mtune=thunderx2t99
 endif
+
+ifeq ($(CORE), TSV110)
+CCOMMON_OPT += -march=armv8.2-a -mtune=tsv110
+FCOMMON_OPT += -march=armv8.2-a -mtune=tsv110
+endif

Makefile.install

@@ -58,14 +58,14 @@ ifndef NO_LAPACKE
 endif
 
 #for install static library
-ifndef NO_STATIC
+ifneq ($(NO_STATIC),1)
 	@echo Copying the static library to $(DESTDIR)$(OPENBLAS_LIBRARY_DIR)
 	@install -pm644 $(LIBNAME) "$(DESTDIR)$(OPENBLAS_LIBRARY_DIR)"
 	@cd "$(DESTDIR)$(OPENBLAS_LIBRARY_DIR)" ; \
 	ln -fs $(LIBNAME) $(LIBPREFIX).$(LIBSUFFIX)
 endif
 #for install shared library
-ifndef NO_SHARED
+ifneq ($(NO_SHARED),1)
 	@echo Copying the shared library to $(DESTDIR)$(OPENBLAS_LIBRARY_DIR)
 ifeq ($(OSNAME), $(filter $(OSNAME),Linux SunOS Android Haiku))
 	@install -pm755 $(LIBSONAME) "$(DESTDIR)$(OPENBLAS_LIBRARY_DIR)"
@@ -106,14 +106,14 @@ ifndef NO_LAPACKE
 endif
 
 #for install static library
-ifndef NO_STATIC
+ifneq ($(NO_STATIC),1)
 	@echo Copying the static library to $(DESTDIR)$(OPENBLAS_LIBRARY_DIR)
 	@installbsd -c -m 644 $(LIBNAME) "$(DESTDIR)$(OPENBLAS_LIBRARY_DIR)"
 	@cd "$(DESTDIR)$(OPENBLAS_LIBRARY_DIR)" ; \
 	ln -fs $(LIBNAME) $(LIBPREFIX).$(LIBSUFFIX)
 endif
 #for install shared library
-ifndef NO_SHARED
+ifneq ($(NO_SHARED),1)
 	@echo Copying the shared library to $(DESTDIR)$(OPENBLAS_LIBRARY_DIR)
 	@installbsd -c -m 755 $(LIBSONAME) "$(DESTDIR)$(OPENBLAS_LIBRARY_DIR)"
 	@cd "$(DESTDIR)$(OPENBLAS_LIBRARY_DIR)" ; \
@@ -138,7 +138,7 @@ endif
 	@echo "SET(OpenBLAS_VERSION \"${VERSION}\")" > "$(DESTDIR)$(OPENBLAS_CMAKE_DIR)/$(OPENBLAS_CMAKE_CONFIG)"
 	@echo "SET(OpenBLAS_INCLUDE_DIRS ${OPENBLAS_INCLUDE_DIR})" >> "$(DESTDIR)$(OPENBLAS_CMAKE_DIR)/$(OPENBLAS_CMAKE_CONFIG)"
 
-ifndef NO_SHARED
+ifneq ($(NO_SHARED),1)
 #ifeq logical or
 ifeq ($(OSNAME), $(filter $(OSNAME),Linux FreeBSD NetBSD OpenBSD DragonFly))
 	@echo "SET(OpenBLAS_LIBRARIES ${OPENBLAS_LIBRARY_DIR}/$(LIBPREFIX).so)" >> "$(DESTDIR)$(OPENBLAS_CMAKE_DIR)/$(OPENBLAS_CMAKE_CONFIG)"

Makefile.rule

@@ -48,6 +48,8 @@ VERSION = 0.3.6.dev
 # HOSTCC = gcc
 
 # If you need 32bit binary, define BINARY=32, otherwise define BINARY=64
+# Please note that AVX is not available on 32-bit.
+# Setting BINARY=32 disables AVX/AVX2/AVX-512.
 # BINARY=64
 
 # About threaded BLAS. It will be automatically detected if you don't
@@ -57,7 +59,7 @@ VERSION = 0.3.6.dev
 # USE_THREAD = 0
 
 # If you're going to use this library with OpenMP, please comment it in.
-# This flag is always set for POWER8. Don't modify the flag 
+# This flag is always set for POWER8. Don't set USE_OPENMP = 0 if you're targeting POWER8.
 # USE_OPENMP = 1
 
 # The OpenMP scheduler to use - by default this is "static" and you
@@ -68,36 +70,45 @@ VERSION = 0.3.6.dev
 # allow you to select the scheduler from the environment variable OMP_SCHEDULE
 # CCOMMON_OPT += -DOMP_SCHED=dynamic
 
-# You can define maximum number of threads. Basically it should be
-# less than actual number of cores. If you don't specify one, it's
-# automatically detected by the the script.
+# You can define the maximum number of threads. Basically it should be less
+# than or equal to the number of CPU threads. If you don't specify one, it's
+# automatically detected by the build system.
+# If SMT (aka. HT) is enabled on the system, it may or may not be beneficial to 
+# restrict NUM_THREADS to the number of physical cores. By default, the automatic 
+# detection includes logical CPUs, thus allowing the use of SMT.
+# Users may opt at runtime to use less than NUM_THREADS threads.
+#
+# Note for package maintainers: you can build OpenBLAS with a large NUM_THREADS
+# value (eg. 32-256) if you expect your users to use that many threads. Due to the way
+# some internal structures are allocated, using a large NUM_THREADS value has a RAM
+# footprint penalty, even if users reduce the actual number of threads at runtime.
 # NUM_THREADS = 24
 
 # If you have enabled USE_OPENMP and your application would call
-# OpenBLAS's calculation API from multi threads, please comment it in.
-# This flag defines how many instances of OpenBLAS's calculation API can 
-# actually run in parallel. If more threads call OpenBLAS's calculation API,
+# OpenBLAS's calculation API from multiple threads, please comment this in.
+# This flag defines how many instances of OpenBLAS's calculation API can actually
+# run in parallel. If more than NUM_PARALLEL threads call OpenBLAS's calculation API,
 # they need to wait for the preceding API calls to finish or risk data corruption.
 # NUM_PARALLEL = 2
 
-# if you don't need to install the static library, please comment it in.
+# If you don't need to install the static library, please comment this in.
 # NO_STATIC = 1
 
-# if you don't need generate the shared library, please comment it in.
+# If you don't need to generate the shared library, please comment this in.
 # NO_SHARED = 1
 
-# If you don't need CBLAS interface, please comment it in.
+# If you don't need the CBLAS interface, please comment this in.
 # NO_CBLAS = 1
 
-# If you only want CBLAS interface without installing Fortran compiler,
-# please comment it in.
+# If you only want the CBLAS interface without installing a Fortran compiler,
+# please comment this in.
 # ONLY_CBLAS = 1
 
-# If you don't need LAPACK, please comment it in.
-# If you set NO_LAPACK=1, the library automatically sets NO_LAPACKE=1.
+# If you don't need LAPACK, please comment this in.
+# If you set NO_LAPACK=1, the build system automatically sets NO_LAPACKE=1.
 # NO_LAPACK = 1
 
-# If you don't need LAPACKE (C Interface to LAPACK), please comment it in.
+# If you don't need LAPACKE (C Interface to LAPACK), please comment this in.
 # NO_LAPACKE = 1
 
 # Build LAPACK Deprecated functions since LAPACK 3.6.0
@@ -106,7 +117,7 @@ BUILD_LAPACK_DEPRECATED = 1
 # Build RecursiveLAPACK on top of LAPACK
 # BUILD_RELAPACK = 1
 
-# If you want to use legacy threaded Level 3 implementation.
+# If you want to use the legacy threaded Level 3 implementation.
 # USE_SIMPLE_THREADED_LEVEL3 = 1
 
 # If you want to use the new, still somewhat experimental code that uses
@@ -116,19 +127,27 @@ BUILD_LAPACK_DEPRECATED = 1
 # USE_TLS = 1
 
 # If you want to drive whole 64bit region by BLAS. Not all Fortran
-# compiler supports this. It's safe to keep comment it out if you
-# are not sure(equivalent to "-i8" option).
+# compilers support this. It's safe to keep this commented out if you
+# are not sure. (This is equivalent to the "-i8" ifort option).
 # INTERFACE64 = 1
 
 # Unfortunately most of kernel won't give us high quality buffer.
 # BLAS tries to find the best region before entering main function,
 # but it will consume time. If you don't like it, you can disable one.
 NO_WARMUP = 1
 
-# If you want to disable CPU/Memory affinity on Linux.
+# Comment this in if you want to disable OpenBLAS's CPU/Memory affinity handling.
+# This feature is only implemented on Linux, and is always disabled on other platforms.
+# Enabling affinity handling may improve performance, especially on NUMA systems, but 
+# it may conflict with certain applications that also try to manage affinity.
+# This conflict can result in threads of the application calling OpenBLAS ending up locked
+# to the same core(s) as OpenBLAS, possibly binding all threads to a single core.
+# For this reason, affinity handling is disabled by default. Can be safely enabled if nothing
+# else modifies affinity settings.
+# Note: enabling affinity has been known to cause problems with NumPy and R
 NO_AFFINITY = 1
 
-# if you are compiling for Linux and you have more than 16 numa nodes or more than 256 cpus
+# If you are compiling for Linux and you have more than 16 numa nodes or more than 256 cpus
 # BIGNUMA = 1
 
 # Don't use AVX kernel on Sandy Bridge. It is compatible with old compilers
@@ -180,7 +199,7 @@ NO_AFFINITY = 1
 # been reported to be optimal for certain workloads (50 is the recommended value for Julia).
 # GEMM_MULTITHREAD_THRESHOLD = 4
 
-# If you need santy check by comparing reference BLAS. It'll be very
+# If you need sanity check by comparing results to reference BLAS. It'll be very
 # slow (Not implemented yet).
 # SANITY_CHECK = 1
 

Makefile.system

@@ -95,6 +95,9 @@ endif
 ifeq ($(TARGET), ZEN)
 GETARCH_FLAGS := -DFORCE_BARCELONA
 endif
+ifeq ($(TARGET), ARMV8)
+GETARCH_FLAGS := -DFORCE_ARMV7
+endif
 endif
 
 
@@ -152,7 +155,8 @@ GETARCH_FLAGS	+= -DNO_AVX
 endif
 
 ifeq ($(BINARY), 32)
-GETARCH_FLAGS	+= -DNO_AVX
+GETARCH_FLAGS	+= -DNO_AVX -DNO_AVX2 -DNO_AVX512
+NO_AVX512 = 1
 endif
 
 ifeq ($(NO_AVX2), 1)

Makefile.zarch

@@ -4,3 +4,7 @@ CCOMMON_OPT += -march=z13 -mzvector
 FCOMMON_OPT += -march=z13 -mzvector
 endif
 
+ifeq ($(CORE), Z14)
+CCOMMON_OPT += -march=z14 -mzvector
+FCOMMON_OPT += -march=z14 -mzvector
+endif

TargetList.txt

@@ -91,7 +91,9 @@ CORTEXA73
 FALKOR
 THUNDERX
 THUNDERX2T99
+TSV110
 
 9.System Z:
 ZARCH_GENERIC
 Z13
+Z14

appveyor.yml

@@ -53,9 +53,9 @@ before_build:
   - ps: if (-Not (Test-Path .\build)) { mkdir build }
   - cd build
   - if [%COMPILER%]==[cl] cmake -G "Visual Studio 15 2017 Win64" ..
-  - if [%WITH_FORTRAN%]==[no] cmake -G "Ninja" -DCMAKE_CXX_COMPILER=clang-cl -DCMAKE_C_COMPILER=clang-cl ..
+  - if [%WITH_FORTRAN%]==[no] cmake -G "Ninja" -DCMAKE_CXX_COMPILER=clang-cl -DCMAKE_C_COMPILER=clang-cl -DMSVC_STATIC_CRT=ON ..
   - if [%WITH_FORTRAN%]==[yes] cmake -G "Ninja" -DCMAKE_CXX_COMPILER=clang-cl -DCMAKE_C_COMPILER=clang-cl -DCMAKE_Fortran_COMPILER=flang -DBUILD_WITHOUT_LAPACK=no -DNOFORTRAN=0 ..
-  - if [%DYNAMIC_ARCH%]==[ON] cmake -DDYNAMIC_ARCH=ON ..
+  - if [%DYNAMIC_ARCH%]==[ON] cmake -DDYNAMIC_ARCH=ON -DDYNAMIC_LIST='CORE2;NEHALEM;SANDYBRIDGE;BULLDOZER;HASWELL' ..
 
 build_script:
   - cmake --build .