Hi All,
I've been playing around with the new intel (v12.0) compilers and have
come across a couple of issues with compilation that appear to be
solved through the use of some additional flags. Specifically:
1) The '-shared-intel' flag is needed during fortran (maybe also C,but
definitely fortran) link steps for dynamic linking of libraries
(
http://archive.ambermd.org/201012/0085.html).
2) The '-std=c++0x' flag is needed during compilation of c++ files
that include <iomanip> when the header files belong to gnu > 4.4
(
http://software.intel.com/en-us/forums/showthread.php?t=78677&p=1&wapkw=%28Multi+Threading+.NET*+Environment%29).
I've modified the configure script to add these flags to 'intel' if
the compiler version is >= 11. It seems to work OK (other than some
problems with mtkpp compilation; I've already sent Martin email
regarding those issues).
This brought up another minor issue with flags in general. There are a
lot of flags now in config.h, and it appears that not all of the
Makefiles have been updated to make use of these flags properly. For
example, in 'reduce' there are a bunch of rules that look like:
$(CXX) -c $(CFLAGS) $(AMBERCXXFLAGS) -o $. $<
where CFLAGS is mixed with AMBERCXXFLAGS for c++ compilation (which
seems bad if not outright wrong). Maybe there should be a discussion
at the amber meeting about making the overall build more uniform in
terms of what flags are used in what cases?
-Dan
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Wed Feb 02 2011 - 16:00:04 PST
