==================================================================================
HoW to build Intel Decimal Floating Point Library under Linux (archive - libbid.a)
==================================================================================

1) Extract [Intel Decimal Floating Point Library] (https://www.intel.com/content/www/us/en/developer/articles/tool/intel-decimal-floating-point-math-library.html)
to some directory

2) In IntelRDFPMathLib20U2/LIBRARY execute the following commands:
make CC=gcc CALL_BY_REF=0 GLOBAL_RND=0 GLOBAL_FLAGS=0 UNCHANGED_BINARY_FLAGS=0
rm *.o

Note: in the sample above all build flags are set to 0. See IntelRDFPMathLib20U2/README for details.

3) As result there is libbid.a library generated. You can use it with TestCppClientStatic.


=========================================================================================
HoW to build Intel Decimal Floating Point Library under Linux (shared object - libbid.so)
=========================================================================================

1) Extract [Intel Decimal Floating Point Library] (https://www.intel.com/content/www/us/en/developer/articles/tool/intel-decimal-floating-point-math-library.html)
to some directory

2) In IntelRDFPMathLib20U2/LIBRARY modify "makefile":
a)
Line 370:
change
BID_LIB = $(LIB_DIR)/libbid.$A
to
BID_LIB = $(LIB_DIR)/libbid.so

b)
Line 377:
change
	$(AR_CMD) $(AR_OUT)$@ $^
to
	gcc -o $@ $^ -shared

c)
Line 112:
change
_CFLAGS_OPT :=
to
_CFLAGS_OPT := -fPIC


2) In IntelRDFPMathLib20U2/LIBRARY execute the following commands:
make CC=gcc CALL_BY_REF=0 GLOBAL_RND=0 GLOBAL_FLAGS=0 UNCHANGED_BINARY_FLAGS=0
rm *.o

Note: in the sample above all build flags are set to 0. See IntelRDFPMathLib20U2/README for details.

3) As result there is libbid.so library generated. You can use it with TestCppClient.
