Merge branch 'develop' into betterPowerGEMVTail

Author: Chip Kerchner

CMakeLists.txt

@@ -8,7 +8,7 @@ project(OpenBLAS C ASM)
 
 set(OpenBLAS_MAJOR_VERSION 0)
 set(OpenBLAS_MINOR_VERSION 3)
-set(OpenBLAS_PATCH_VERSION 27.dev)
+set(OpenBLAS_PATCH_VERSION 28.dev)
 
 set(OpenBLAS_VERSION "${OpenBLAS_MAJOR_VERSION}.${OpenBLAS_MINOR_VERSION}.${OpenBLAS_PATCH_VERSION}")
 
@@ -22,6 +22,8 @@ option(BUILD_WITHOUT_LAPACK "Do not build LAPACK and LAPACKE (Only BLAS or CBLAS
 
 option(BUILD_LAPACK_DEPRECATED "When building LAPACK, include also some older, deprecated routines" ON)
 
+set(LAPACK_STRLEN "" CACHE STRING "When building LAPACK, use this type (e.g. \"int\") for character lengths (defaults to size_t)")
+
 option(BUILD_TESTING "Build LAPACK testsuite when building LAPACK" ON)
 
 option(BUILD_BENCHMARKS "Build the collection of BLAS/LAPACK benchmarks" OFF)
@@ -30,7 +32,7 @@ option(C_LAPACK "Build LAPACK from C sources instead of the original Fortran" OF
 
 option(BUILD_WITHOUT_CBLAS "Do not build the C interface (CBLAS) to the BLAS functions" OFF)
 
-option(DYNAMIC_ARCH "Include support for multiple CPU targets, with automatic selection at runtime (x86/x86_64, aarch64 or ppc only)" OFF)
+option(DYNAMIC_ARCH "Include support for multiple CPU targets, with automatic selection at runtime (x86/x86_64, aarch64, ppc or RISCV64-RVV1.0 only)" OFF)
 
 option(DYNAMIC_OLDER "Include specific support for older x86 cpu models (Penryn,Dunnington,Atom,Nano,Opteron) with DYNAMIC_ARCH" OFF)
 
@@ -256,6 +258,10 @@ if (${CMAKE_SYSTEM_NAME} MATCHES "AIX|Android|Linux|FreeBSD|OpenBSD|NetBSD|Drago
   endif()
 endif()
 
+if (APPLE AND BUILD_SHARED_LIBS)
+set(CMAKE_MACOSX_RPATH ON)
+endif()
+
 # Seems that this hack doesn't required since macOS 11 Big Sur
 if (APPLE AND BUILD_SHARED_LIBS AND CMAKE_HOST_SYSTEM_VERSION VERSION_LESS 20)
   set (CMAKE_C_USE_RESPONSE_FILE_FOR_OBJECTS 1)

Changelog.txt

@@ -1,4 +1,127 @@
 OpenBLAS ChangeLog
+====================================================================
+Version 0.3.28
+ 8-Aug-2024
+
+general:
+- Reworked the unfinished implementation of HUGETLB from GotoBLAS
+  for allocating huge memory pages as buffers on suitable systems
+- Changed the unfinished implementation of GEMM3M for the generic
+  target on all architectures to at least forward to regular GEMM
+- Improved multithreaded GEMM performance for large non-skinny matrices
+- Improved BLAS3 performance on larger multicore systems through improved
+  parallelism
+- Improved performance of the initial memory allocation by reducing
+  locking overhead
+- Improved performance of GBMV at small problem sizes by introducing
+  a size barrier for the switch to multithreading
+- Added an implementation of the CBLAS_GEMM_BATCH extension
+- Fixed miscompilation of CAXPYC and ZAXPYC on all architectures in 
+  CMAKE builds (error introduced in 0.3.27)
+- Fixed corner cases involving the handling of NAN and INFINITY
+  arguments in ?SCAL on all architectures
+- Added support for cross-compiling to WEBM with CMAKE (in addition
+  to the already present makefile support)
+- Fixed NAN handling and potential accuracy issues in compilations with
+  Intel ICX by supplying a suitable fp-model option by default
+- The contents of the github project wiki have been converted into
+  a new set of documentation included with the source code.
+- It is now possible to register a callback function that replaces
+  the built-in support for multithreading with an external backend
+  like TBB (openblas_set_threads_callback_function)
+- Fixed potential duplication of suffixes in shared library naming
+- Improved C compiler detection by the build system to tolerate more
+  naming variants for gcc builds
+- Fixed an unnecessary dependency of the utest on CBLAS
+- Fixed spurious error reports from the BLAS extensions utest
+- Fixed unwanted invocation of the GEMM3M tests in cross-compilation
+- Fixed a flaw in the makefile build that could lead to the pkgconfig
+  file containing an entry of UNKNOWN for the target cpu after installing
+- Integrated fixes from the Reference-LAPACK project:
+  - Fixed uninitialized variables in the LAPACK tests for ?QP3RK (PR 961)
+  - Fixed potential bounds error in ?UNHR_COL/?ORHR_COL (PR 1018)
+  - Fixed potential infinite loop in the LAPACK testsuite (PR 1024)
+  - Make the variable type used for hidden length arguments configurable (PR 1025)  
+  - Fixed SYTRD workspace computation and various typos (PR 1030)
+  - Prevent compiler use of FMA that could increase numerical error in ?GEEVX (PR 1033)
+
+x86-64:
+- reverted thread management under Windows to its state before 0.3.26
+  due to signs of race conditions in some circumstances now under study
+- fixed accidental selection of the unoptimized generic SBGEMM kernel
+  in CMAKE builds for CooperLake and SapphireRapids targets
+- fixed a potential thread buffer overrun in SBSTOBF16 on small systems
+- fixed an accuracy issue in ZSCAL introduced in 0.3.26
+- fixed compilation with CMAKE and recent releases of LLVM
+- added support for Intel Emerald Rapids and Meteor Lake cpus
+- added autodetection support for the Zhaoxin KX-7000 cpu
+- fixed autodetection of Intel Prescott (probably broken since 0.3.19)
+- fixed compilation for older targets with the Yocto SDK
+- fixed compilation of the converter-generated C versions
+  of the LAPACK sources with gcc-14
+- improved compiler options when building with CMAKE and LLVM for
+  AVX512-capable targets
+- added support for supplying the L2 cache size via an environment
+  variable (OPENBLAS_L2_SIZE) in case it is not correctly reported
+  (as in some VM configurations)
+- improved the error message shown when thread creation fails on startup
+- fixed setting the rpath entry of the dylib in CMAKE builds on MacOS
+
+arm:
+- fixed building for baremetal targets with make
+
+arm64:
+- Added a fast path forwarding SGEMM and DGEMM calls with a 1xN or Mx1
+  matrix to the corresponding GEMV kernel 
+- added optimized SGEMV and DGEMV kernels for A64FX
+- added optimized SVE kernels for small-matrix GEMM
+- added A64FX to the cpu list for DYNAMIC_ARCH
+- fixed building with support for cpu affinity
+- worked around accuracy problems with C/ZNRM2 on NeoverseN1 and
+  Apple M targets
+- improved GEMM performance on Neoverse V1
+- fixed compilation for NEOVERSEN2 with older compilers
+- fixed potential miscompilation of the SVE SDOT and DDOT kernels
+- fixed potential miscompilation of the non-SVE CDOT and ZDOT kernels
+- fixed a potential overflow when using very large user-defined BUFFERSIZE
+- fixed setting the rpath entry of the dylib in CMAKE builds on MacOS
+
+power:
+- Added a fast path forwarding SGEMM and DGEMM calls with a 1xN or Mx1
+  matrix to the corresponding GEMV kernel 
+- significantly improved performance of SBGEMM on POWER10
+- fixed compilation with OpenMP and the XLF compiler
+- fixed building of the BLAS extension utests under AIX
+- fixed building of parts of the LAPACK testsuite with XLF
+- fixed CSWAP/ZSWAP on big-endian POWER10 targets
+- fixed a performance regression in SAXPY on POWER10 with OpenXL
+- fixed accuracy issues in CSCAL/ZSCAL when compiled with LLVM
+- fixed building for POWER9 under FreeBSD
+- fixed a potential overflow when using very large user-defined BUFFERSIZE
+- fixed an accuracy issue in the POWER6 kernels for GEMM and GEMV
+
+riscv64:
+- Added a fast path forwarding SGEMM and DGEMM calls with a 1xN or Mx1
+  matrix to the corresponding GEMV kernel 
+- fixed building for RISCV64_GENERIC with OpenMP enabled
+- added DYNAMIC_ARCH support (comprising GENERIC_RISCV64 and the two
+  RVV 1.0 targets with vector length of 128 and 256)
+- worked around the ZVL128B kernels for AXPBY mishandling the special
+  case of zero Y increment
+
+loongarch64:
+- improved GEMM performance on servers of the 3C5000 generation
+- improved performance and stability of DGEMM
+- improved GEMV and TRSM kernels for LSX and LASX vector ABIs
+- fixed CMAKE compilation with the INTERFACE64 option set
+- fixed compilation with CMAKE
+- worked around spurious errors flagged by the BLAS3 tests
+- worked around a miscompilation of the POTRS utest by gcc 14.1
+
+mips64:
+- fixed ASUM and SUM kernels to accept negative step sizes in X
+- fixed complex GEMV kernels for MSA
+
 ====================================================================
 Version 0.3.27
  4-Apr-2024

Makefile

@@ -45,6 +45,10 @@ else
 LAPACK_NOOPT := $(filter-out -O0 -O1 -O2 -O3 -Ofast -O -Og -Os,$(LAPACK_FFLAGS))
 endif
 
+ifdef LAPACK_STRLEN
+LAPACK_FFLAGS += -DLAPACK_STRLEN=$(LAPACK_STRLEN)
+endif
+
 SUBDIRS_ALL = $(SUBDIRS) test ctest utest exports benchmark ../laswp ../bench cpp_thread_test
 
 .PHONY : all libs netlib $(RELA) test ctest shared install

Makefile.install

@@ -178,7 +178,7 @@ endif
 	@echo 'libnamesuffix='$(LIBNAMESUFFIX) >> "$(PKGFILE)"
 	@echo 'libsuffix='$(SYMBOLSUFFIX) >> "$(PKGFILE)"
 	@echo 'includedir='$(OPENBLAS_INCLUDE_DIR) >> "$(PKGFILE)"
-	@echo 'openblas_config= USE_64BITINT='$(INTERFACE64) 'DYNAMIC_ARCH='$(DYNAMIC_ARCH) 'DYNAMIC_OLDER='$(DYNAMIC_OLDER) 'NO_CBLAS='$(NO_CBLAS) 'NO_LAPACK='$(NO_LAPACK) 'NO_LAPACKE='$(NO_LAPACKE) 'NO_AFFINITY='$(NO_AFFINITY) 'USE_OPENMP='$(USE_OPENMP) $(CORE) 'MAX_THREADS='$(NUM_THREADS)>> "$(PKGFILE)"
+	@echo 'openblas_config= USE_64BITINT='$(INTERFACE64) 'DYNAMIC_ARCH='$(DYNAMIC_ARCH) 'DYNAMIC_OLDER='$(DYNAMIC_OLDER) 'NO_CBLAS='$(NO_CBLAS) 'NO_LAPACK='$(NO_LAPACK) 'NO_LAPACKE='$(NO_LAPACKE) 'NO_AFFINITY='$(NO_AFFINITY) 'USE_OPENMP='$(USE_OPENMP) $(TARGET) 'MAX_THREADS='$(NUM_THREADS)>> "$(PKGFILE)"
 	@echo 'version='$(VERSION) >> "$(PKGFILE)"
 	@echo 'extralib='$(PKG_EXTRALIB) >> "$(PKGFILE)"
 	@cat openblas.pc.in >> "$(PKGFILE)"

Makefile.rule

@@ -3,7 +3,7 @@
 #
 
 # This library's version
-VERSION = 0.3.27.dev
+VERSION = 0.3.28.dev
 
 # If you set this prefix, the library name will be lib$(LIBNAMESUFFIX)openblas.a
 # and lib$(LIBNAMESUFFIX)openblas.so, with a matching soname in the shared library
@@ -134,6 +134,12 @@ VERSION = 0.3.27.dev
 # Build LAPACK Deprecated functions since LAPACK 3.6.0
 BUILD_LAPACK_DEPRECATED = 1
 
+# The variable type assumed for the length of character arguments when passing
+# data between Fortran LAPACK and C BLAS (defaults to "size_t", but older GCC
+# versions used "int"). Mismatches will not cause runtime failures but may result
+# in build warnings or errors when building with link-time optimization (LTO)
+# LAPACK_STRLEN=int
+
 # Build RecursiveLAPACK on top of LAPACK
 # BUILD_RELAPACK = 1
 # Have RecursiveLAPACK actually replace standard LAPACK routines instead of 

Makefile.system

@@ -277,6 +277,12 @@ endif
 ifeq ($(ARCH), arm64)
 GEMM_GEMV_FORWARD = 1
 endif
+ifeq ($(ARCH), riscv)
+GEMM_GEMV_FORWARD = 1
+endif
+ifeq ($(ARCH), power)
+GEMM_GEMV_FORWARD = 1
+endif
 
 ifeq ($(SMALL_MATRIX_OPT), 1)
 CCOMMON_OPT += -DSMALL_MATRIX_OPT

cmake/arch.cmake

@@ -57,7 +57,11 @@ if (DYNAMIC_ARCH)
 	  set(DYNAMIC_CORE POWER6 POWER8 POWER9 POWER10)
 	  set(CCOMMON_OPT "${CCOMMON_OPT} -DHAVE_P10_SUPPORT")
   endif ()
-  
+ 
+  if (RISCV64)
+	  set(DYNAMIC_CORE RISCV64_GENERIC RISCV64_ZVL128B RISCV64_ZVL256B) 
+  endif ()
+
   if (X86)
     set(DYNAMIC_CORE KATMAI COPPERMINE NORTHWOOD PRESCOTT BANIAS CORE2 PENRYN DUNNINGTON NEHALEM ATHLON OPTERON OPTERON_SSE3 BARCELONA BOBCAT ATOM NANO)
   endif ()

cmake/system.cmake

@@ -403,7 +403,7 @@ if (SMALL_MATRIX_OPT)
 endif ()
 
 if (DYNAMIC_ARCH)
-  if (X86 OR X86_64 OR ARM64 OR POWER)
+  if (X86 OR X86_64 OR ARM64 OR POWER OR RISCV64)
     set(CCOMMON_OPT "${CCOMMON_OPT} -DDYNAMIC_ARCH")
     if (DYNAMIC_OLDER)
       set(CCOMMON_OPT "${CCOMMON_OPT} -DDYNAMIC_OLDER")
@@ -621,7 +621,10 @@ set(CMAKE_Fortran_FLAGS "${CMAKE_Fortran_FLAGS} ${FCOMMON_OPT}")
 set(FPFLAGS "${FPFLAGS} ${FCOMMON_OPT} ${COMMON_PROF}")
 
 #For LAPACK Fortran codes.
-set(LAPACK_FFLAGS "${LAPACK_FFLAGS} ${CMAKE_Fortran_FLAGS}")
+set(LAPACK_FFLAGS "${LAPACK_FFLAGS} ${CMAKE_Fortran_FLAGS}" )
+if (LAPACK_STRLEN)
+	set (LAPACK_FFLAGS "${LAPACK_FFLAGS} -DLAPACK_STRLEN=${LAPACK_STRLEN}")
+endif()
 set(LAPACK_FPFLAGS "${LAPACK_FPFLAGS} ${FPFLAGS}")
 
 #Disable -fopenmp for LAPACK Fortran codes on Windows.

common_thread.h

@@ -111,8 +111,8 @@ typedef struct blas_queue {
   struct blas_queue *next;
 
 #if defined( __WIN32__) || defined(__CYGWIN32__) || defined(_WIN32) || defined(__CYGWIN__)
-  // CRITICAL_SECTION lock;
-  // HANDLE finish;
+   CRITICAL_SECTION lock;
+   HANDLE finish;
   volatile int finished;
 #else
   pthread_mutex_t	 lock;

driver/others/CMakeLists.txt

@@ -52,6 +52,8 @@ if (DYNAMIC_ARCH)
     list(APPEND COMMON_SOURCES dynamic_arm64.c)
   elseif (POWER)
     list(APPEND COMMON_SOURCES dynamic_power.c)
+  elseif (RISCV64)
+    list(APPEND COMMON_SOURCES dynamic_riscv64.c detect_riscv64.c)
   else ()  
     list(APPEND COMMON_SOURCES dynamic.c)
   endif ()