[gmx-users] Testing (was: Tyrosine problems again)

dichan at ucalgary.ca dichan at ucalgary.ca
Sat Aug 2 01:13:43 CEST 2008


Hello once again

As per the suggestions, I went ahead and installed an older version of the
gcc compiler: version 3.4.

I then re-installed fftw and gromacs3.3.3 using the following commands:

./configure --enable-float --enable-threads --enable-sse
--prefix=/home/david/fftw-gcc3 CC=/usr/lib64/ccache/gcc34
CXX=/usr/lib64/ccache/g++34

./configure LDFLAGS=-L/home/david/fftw-gcc3/lib
CPPFLAGS=-I/home/david/fftw-gcc3/include
--prefix=/usr/local/gromacs-single-3.3.3-gcc3 CC=/usr/lib64/ccache/gcc34
CXX=/usr/lib64/ccache/g++34

The compilation seems to finish, but I do still find a couple of error
instances:

conftest.c: In function `main':
conftest.c:42: error: `not' undeclared (first use in this function)
conftest.c:42: error: (Each undeclared identifier is reported only once
conftest.c:42: error: for each function it appears in.)
conftest.c:42: error: syntax error before "big"

conftest.c: In function `main':
conftest.c:51: error: void value not ignored as it ought to be

Wno-unused -funroll-all-loops -I/home/david/fftw-gcc3/include conftest.c >&5
conftest.c:74: error: syntax error before "ac__type_new_"
conftest.c:74: warning: type defaults to `int' in declaration of
`ac__type_new_'
conftest.c:74: warning: data definition has no type or storage class
conftest.c: In function `main':
conftest.c:86: error: syntax error before ')' token

Do these indicate my configure run failed? Sorry, I just don't know what
to make of these.

Gromacs runs after, but I am still getting odd tyrosine behaviour. The
dihedral once again resembles a chair at the beginning and then settles
down during extended equilibration. Based on previous experience I am not
used to this sort of behaviour - can anyone confirm this is something to
worry about? (www.ucalgary.ca/~dichan/gcc3_h2o_eq.jpg)

Cheers,

David




More information about the gromacs.org_gmx-users mailing list