Hi, FYI, now building the GPU accelerated version in a non-clean build tree (used to build the CPU version) should work as well as the other way around.
_However_, be warned that in the latter case the CPU build-related parameters do _not_ get reset to their default values (e.g. GMX_ACCELERATION != auto, GMX_THREADS != on)! The safest way (and what I strongly recommend) is to have different build trees for different builds or, as Rossen suggested, just remove the cache before building something different. Regards, -- Szilárd On Tue, Aug 24, 2010 at 2:39 PM, Rossen Apostolov <rossen.aposto...@cbr.su.se> wrote: > Hi Alan, > > With OpenMM you don't need SSE kernels. If you look at CMakeCache.txt you'll > see that > > >> Hi there, >> >> I've tried before and the error still basically the same (line error in >> 'nb_kernel400_x86_64_sse.c' was 630, now's 629) . >> >> I am using gmx 4.5 release-4-5-patches branch. >> >> git pull >> rm -fr build >> mkdir -p build >> cd build >> cmake -D BUILD_SHARED_LIBS=ON -DGMX_OPENMM=OFF .. > > Since GMX_OPENMM=OFF, GMX_ACCELERATION is automatically set to SSE. >> >> make clean > > This command doesn't remove CMaceCache.txt! >> >> make -j 2 >> sudo make install # all fine till here >> make clean >> export OPENMM_ROOT_DIR=/usr/local/openmm >> cmake -D BUILD_SHARED_LIBS=OFF -DGMX_OPENMM=ON .. > > Now GMX_OPENMM is ON, *but* GMX_ACCELERATION is still SSE (the value is read > from the cache). This is an issue with the cmake system. You should either > remove CMakeCache.txt or add an explicit -DGMX_ACCELERATION=OFF to the cmake > command. > > This is a workaround at the moment but the issue should be fixed. > > Rossen >> >> make mdrun >> >> [snip] >> [ 56%] Building C object src/gmxlib/CMakeFiles/gmx.dir/version.c.o >> [ 56%] Building C object >> src/gmxlib/CMakeFiles/gmx.dir/nonbonded/nb_kernel_x86_64_sse/nb_kernel400_x86_64_sse.c.o >> >> /Users/alan/Programmes/gromacs/src/gmxlib/nonbonded/nb_kernel_x86_64_sse/nb_kernel400_x86_64_sse.c: >> In function ‘nb_kernel400nf_x86_64_sse’: >> >> /Users/alan/Programmes/gromacs/src/gmxlib/nonbonded/nb_kernel_x86_64_sse/nb_kernel400_x86_64_sse.c:629: >> error: ‘gmx_invsqrt_exptab’ undeclared (first use in this function) >> >> /Users/alan/Programmes/gromacs/src/gmxlib/nonbonded/nb_kernel_x86_64_sse/nb_kernel400_x86_64_sse.c:629: >> error: (Each undeclared identifier is reported only once >> >> /Users/alan/Programmes/gromacs/src/gmxlib/nonbonded/nb_kernel_x86_64_sse/nb_kernel400_x86_64_sse.c:629: >> error: for each function it appears in.) >> >> /Users/alan/Programmes/gromacs/src/gmxlib/nonbonded/nb_kernel_x86_64_sse/nb_kernel400_x86_64_sse.c:629: >> error: ‘gmx_invsqrt_fracttab’ undeclared (first use in this function) >> make[3]: *** >> [src/gmxlib/CMakeFiles/gmx.dir/nonbonded/nb_kernel_x86_64_sse/nb_kernel400_x86_64_sse.c.o] >> Error 1 >> make[2]: *** [src/gmxlib/CMakeFiles/gmx.dir/all] Error 2 >> make[1]: *** [src/kernel/CMakeFiles/mdrun.dir/rule] Error 2 >> make: *** [mdrun] Error 2 >> >> Any help would be appreciated. >> >> Thanks, >> >> Alan >> -- >> Alan Wilter S. da Silva, D.Sc. - CCPN Research Associate >> Department of Biochemistry, University of Cambridge. >> 80 Tennis Court Road, Cambridge CB2 1GA, UK. >> >>http://www.bio.cam.ac.uk/~awd28 <http://www.bio.cam.ac.uk/%7Eawd28><< > > -- > gmx-users mailing list gmx-us...@gromacs.org > http://lists.gromacs.org/mailman/listinfo/gmx-users > Please search the archive at http://www.gromacs.org/search before posting! > Please don't post (un)subscribe requests to the list. Use the www interface > or send it to gmx-users-requ...@gromacs.org. > Can't post? Read http://www.gromacs.org/mailing_lists/users.php > -- gmx-users mailing list gmx-users@gromacs.org http://lists.gromacs.org/mailman/listinfo/gmx-users Please search the archive at http://www.gromacs.org/search before posting! Please don't post (un)subscribe requests to the list. Use the www interface or send it to gmx-users-requ...@gromacs.org. Can't post? Read http://www.gromacs.org/mailing_lists/users.php