#2 installtion error on ubuntu-8.10
Dear NanoHive experts,
Hi!
When I install NanoHive-1.2 on ubuntu-8.10, the configure successfully runs. I run the command:
./configure --enable-mpqc-sclib --enable-netcdf-dataset --disable-console-command F77=gfortran
However, make command gives the error:
many of outputs ...
In file included from ./../include/NanoHiveInterface/PhysicalInteractionPlugin.h:26,
from PIP_Thread.h:34,
from PIP_ThreadPool.h:34,
from PIP_ThreadPool.cpp:20:
./../include/NanoHiveUtil/CommandResult.h:40:17: warning: "/*" within comment
PIP_ThreadPool.cpp: In member function ‘nanohive::PIP_Thread* nanohive::PIP_ThreadPool::getThread(int, int)’:
PIP_ThreadPool.cpp:104: error: cast from ‘nanohive::PIP_Thread*’ to ‘int’ loses precision
PIP_ThreadPool.cpp: In member function ‘bool nanohive::PIP_ThreadPool::hasRunningThreads()’:
PIP_ThreadPool.cpp:120: warning: comparison between signed and unsigned integer expressions
./../include/NanoHiveUtil/Utility.h: At global scope:
./../include/NanoHiveUtil/Utility.h:117: warning: ‘nanohive::NSPRcodeToMeaning’ defined but not used
make[2]: *** [NanoHive-PIP_ThreadPool.o] Error 1
make[2]: Leaving directory `/opt/NanoHive/src'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/opt/NanoHive/src'
make: *** [all] Error 2
Please help me!
You have to pass the -fpermissive flag to g++ to solve the loss of precision error (do this in the Makefile), however this is generally not a good idea, although it worked for me on Debian Wheezy (stable).