Fixed #374.

Merge branch 'TimothyGu-develop' into develop

diff --git a/Changelog.txt b/Changelog.txt
index 198c2d8..f53cadc 100644 (file)
--- a/Changelog.txt
+++ b/Changelog.txt
@@ -51,7 +51,7 @@ common:
           parallelization model is used by OpenBLAS. (Thank grisuthedragon)
        * Detect LLVM/Clang compiler. The default compiler is Clang on Mac OS X.
         * Change LIBSUFFIX from .lib to .a on windows.
-       * A walk round for dtrti_U single thread bug. Replace it with LAPACK codes. (#191)
+       * A work-around for dtrti_U single thread bug. Replace it with LAPACK codes. (#191)
 
 x86/x86-64:
        * Optimize c/zgemm, trsm, dgemv_n, ddot, daxpy, dcopy on 
@@ -284,7 +284,7 @@ x86/x86_64:
        * Fixed #28 a wrong result of dsdot on x86_64.
        * Fixed #32 a SEGFAULT bug of zdotc with gcc-4.6.
        * Fixed #33 ztrmm bug on Nehalem.
-       * Walk round #27 the low performance axpy issue with small imput size & multithreads.
+       * Work-around #27 the low performance axpy issue with small imput size & multithreads.
 
 MIPS64:
        * Fixed #28 a wrong result of dsdot on Loongson3A/MIPS64. 
@@ -308,7 +308,7 @@ common:
 
 x86/x86_64:
        * On x86 32bits, fixed a bug in zdot_sse2.S line 191. This would casue 
-         zdotu & zdotc failures.Instead,Walk around it. (Refs issue #8 #9 on github)
+         zdotu & zdotc failures. Instead, work-around it. (Refs issue #8 #9 on github)
        * Modified ?axpy functions to return same netlib BLAS results 
          when incx==0 or incy==0 (Refs issue #7 on github)
        * Modified ?swap functions to return same netlib BLAS results 

diff --git a/Makefile b/Makefile
index 01bedaf..2f5d032 100644 (file)
--- a/Makefile
+++ b/Makefile
@@ -129,6 +129,11 @@ endif
 ifeq ($(NOFORTRAN), 1)
        $(error OpenBLAS: Detecting fortran compiler failed. Please install fortran compiler, e.g. gfortran, ifort, openf90.)
 endif
+ifeq ($(NO_STATIC), 1)
+ifeq ($(NO_SHARED), 1)
+       $(error OpenBLAS: neither static nor shared are enabled.)
+endif
+endif
        @-ln -fs $(LIBNAME) $(LIBPREFIX).$(LIBSUFFIX)
        @for d in $(SUBDIRS) ; \
        do if test -d $$d; then \

diff --git a/Makefile.install b/Makefile.install
index 6fcbcc3..e041120 100644 (file)
--- a/Makefile.install
+++ b/Makefile.install
@@ -50,10 +50,12 @@ ifndef NO_LAPACKE
 endif
 
 #for install static library 
+ifndef NO_STATIC
        @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
        @echo Copying the shared library to $(DESTDIR)$(OPENBLAS_LIBRARY_DIR)
@@ -80,6 +82,7 @@ ifeq ($(OSNAME), Darwin)
 endif
 ifeq ($(OSNAME), WINNT)
        @-cp $(LIBDLLNAME) $(OPENBLAS_BINARY_DIR)
+       @-cp $(LIBDLLNAME).a $(OPENBLAS_LIBRARY_DIR)
 endif
 ifeq ($(OSNAME), CYGWIN_NT)
        @-cp $(LIBDLLNAME) $(OPENBLAS_BINARY_DIR)

diff --git a/Makefile.rule b/Makefile.rule
index 6b604b7..cf631d8 100644 (file)
--- a/Makefile.rule
+++ b/Makefile.rule
@@ -48,6 +48,9 @@ VERSION = 0.2.9.rc2
 # automatically detected by the the script.
 # NUM_THREADS = 24
 
+# if you don't need to install the static library, please comment it in.
+# NO_STATIC = 1
+
 # if you don't need generate the shared library, please comment it in.
 # NO_SHARED = 1
 

diff --git a/exports/Makefile b/exports/Makefile
index e09b3c3..ce28ab5 100644 (file)
--- a/exports/Makefile
+++ b/exports/Makefile
@@ -83,15 +83,9 @@ dll  : ../$(LIBDLLNAME)
 # For more details see: https://github.com/xianyi/OpenBLAS/issues/127.
 ../$(LIBDLLNAME) : ../$(LIBNAME) libopenblas.def dllinit.$(SUFFIX)
        $(RANLIB) ../$(LIBNAME)
-ifeq ($(BINARY32), 1)
-       $(DLLWRAP) -o ../$(LIBDLLNAME) --def libopenblas.def \
-       --entry _dllinit@12 -s dllinit.$(SUFFIX) --dllname $(@F) ../$(LIBNAME) $(EXTRALIB)
-       -lib /machine:i386 /def:libopenblas.def
-else
-       $(DLLWRAP) -o ../$(LIBDLLNAME) --def libopenblas.def \
-       --entry $(FU)dllinit    -s dllinit.$(SUFFIX) --dllname $(@F) ../$(LIBNAME) $(EXTRALIB)
-       -lib /machine:X64 /def:libopenblas.def 
-endif
+       $(CC) $(CFLAGS) $(LDFLAGS) libopenblas.def dllinit.$(SUFFIX) \
+       -shared -o ../$(LIBDLLNAME) -Wl,--out-implib,../$(LIBDLLNAME).a \
+       -Wl,--whole-archive ../$(LIBNAME) -Wl,--no-whole-archive $(FEXTRALIB)
 
 libopenblas.def : gensymbol
        perl ./gensymbol win2k    $(ARCH) dummy $(EXPRECISION) $(NO_CBLAS) $(NO_LAPACK) $(NO_LAPACKE) $(NEED2UNDERSCORES) $(ONLY_CBLAS) > $(@F)

diff --git a/exports/dllinit.c b/exports/dllinit.c
index 54ec1c3..0f25824 100644 (file)
--- a/exports/dllinit.c
+++ b/exports/dllinit.c
@@ -41,7 +41,7 @@
 void gotoblas_init(void);
 void gotoblas_quit(void);
 
-BOOL APIENTRY dllinit(HINSTANCE hInst, DWORD reason, LPVOID reserved) {
+BOOL APIENTRY DllMain(HINSTANCE hInst, DWORD reason, LPVOID reserved) {
 
   if (reason == DLL_PROCESS_ATTACH) {
     gotoblas_init();

diff --git a/interface/axpy.c b/interface/axpy.c
index 82b0ee2..5e288e3 100644 (file)
--- a/interface/axpy.c
+++ b/interface/axpy.c
@@ -86,7 +86,8 @@ void CNAME(blasint n, FLOAT alpha, FLOAT *x, blasint incx, FLOAT *y, blasint inc
   if (incx == 0 || incy == 0)
          nthreads = 1;
  
-  //Temporarily walk around the low performance issue with small imput size & multithreads.
+  //Temporarily work-around the low performance issue with small imput size &
+  //multithreads.
   if (n <= 10000)
          nthreads = 1;