Re: [AMBER] problems compiling ambertool

From: Per Jr. Greisen <pgreisen.gmail.com>
Date: Tue, 17 Mar 2009 08:10:20 +0000

Hi,

Thanks - I added the -lm in the Makefile in the antechamber/ directory as
suggested

cc -lm ncsu-penalties.o bondtype.o -o bondtype

but then the last directory to compile comes with some new error messages

installing /reduce_src...
CC -c -O3 -DAROMATICS_ACCEPT_HBONDS -DCHARFUNCMACROS -DBOOLPREDEFINED
-DBRACKETOPERPARMS -DLEFT_JUSTIFY_NUC_RES_OK -DLISTFRIENDFIX
-DINCTEMPLATEDEFNS
-DHET_DICTIONARY='"/xbar/nas1/gappl/gappl/ggamess/amber10/amber10/dat/reduce_wwPDB_het_dict.txt"'
-I../libpdb -I../toolclasses -o GraphToHoldScores.o GraphToHoldScores.cpp
"ResBlk.h", line 40: Error: Could not find a match for
std::multimap<std::string, PDBrec*, std::less<std::string>,
std::allocator<std::pair<const std::string,
PDBrec*>>>::insert(std::pair<std::string, PDBrec*>) needed in
ResBlk::addPrevRec(PDBrec*).
"ResBlk.h", line 43: Error: Could not find a match for
std::multimap<std::string, PDBrec*, std::less<std::string>,
std::allocator<std::pair<const std::string,
PDBrec*>>>::insert(std::pair<std::string, PDBrec*>) needed in
ResBlk::addNextRec(PDBrec*).
"AtomPositions.h", line 67: Error: Could not find a match for
std::multimap<LocBlk, PDBrec*, std::less<LocBlk>,
std::allocator<std::pair<const LocBlk, PDBrec*>>>::insert(std::pair<LocBlk,
PDBrec*>) needed in AtomPositions::put(PDBrec*).
3 Error(s) detected.
*** Error code 3
make: Fatal error: Command failed for target `GraphToHoldScores.o'
Current working directory
/xbar/nas1/gappl/gappl/ggamess/amber10/amber10/src/reduce/reduce_src
*** Error code 1
The following command caused the error:
case '' in *[ik]*) set +e;; esac; \
for i in toolclasses libpdb reduce_src ;\
do \
        (cd $i ; echo "installing" "/$i..."; \
                make 'CXXDEBUGFLAGS=' install); \
done
make: Fatal error: Command failed for target `install'
Current working directory
/xbar/nas1/gappl/gappl/ggamess/amber10/amber10/src/reduce
*** Error code 1
make: Fatal error: Command failed for target `install'






On Mon, Mar 16, 2009 at 11:50 PM, Bill Ross <ross.cgl.ucsf.edu> wrote:

> > ...[lots of warnings that can be ignored]...
> > cc ncsu-penalties.o bondtype.o -o bondtype [no -lm flag]
> > Undefined first referenced
> > symbol in file
> > fabs bondtype.o
> > ld: fatal: Symbol referencing errors. No output written to bondtype
> > *** Error code 1
>
> Ironic that a missing math library routine would be what stops the
> compile. Note that other compiles use the math lib (-lm):
>
> > cc atomtype.o -lm -o atomtype
>
> But this flag is missing from the bondtype compile above.
>
> Bill
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Mar 18 2009 - 01:14:46 PDT
Custom Search