Right, the picking up the system copies is probably because some
additional environment variables need to be set, it seems to be different
on different computers but some notes I'm adding to the installation
instructions (not yet included in the gzipped tar file) are:
"To get compilation of Python to work with the distributed versions
of Tcl/Tk you might need to set the following environment variables
first:
setenv LD_LIBRARY_PATH
${CCPNMR_TOP_DIR}/tcl8.3/lib:${CCPNMR_TOP_DIR}/tk8.3/lib
where CCPNMR_TOP_DIR is is the ccpnmr directory created by the
unpacking of the tar file."
and in order to get it to run:
setenv LD_LIBRARY_PATH
${CCPNMR_TOP_DIR}/tcl8.3/lib:${CCPNMR_TOP_DIR}/tk8.3/lib:${CCPNMR_TOP_DIR}/mesa4.0/lib
setenv TCL_LIBRARY ${CCPNMR_TOP_DIR}/tcl8.3/lib/tcl8.3
setenv TK_LIBRARY ${CCPNMR_TOP_DIR}/tk8.3/lib/tk8.3
(each setenv should be on one line, email will probably mangle it).
It's possible the code compiled against your system tcl/tk/gl. This might
not necessarily be a bad thing! (Given that Mesa seems to be having
problems on some people's computers.)
I have seen that BadMatch error in the circumstances you mention, I forget
what this is about now because I can't reproduce it now, it's possibly a
timing issue.
On Fri, 26 Sep 2003, Brian Smith wrote:
> Wayne,
>
> Tried the modification you suggested, but no dice (or contours).
>
> I will investigate the relationship with various X/OpenGL
> libraries & drivers. I am using an NVidia card and their OpenGL libraries
> in order to get hardware graphics acceleration (which makes a huge
> difference to Ansig4OpenGL, Molmol etc.). I notice that although I've now
> compiled with all the source bundles from your downloads, I'm still
> picking up the system copies of libGLU/GL/glut/tcl/tk (according to lsof).
> The tcl & tk bits should be OK since I'm also on a RH8 system with tcl/tk
> 8.3.something. I need to play with linker flags etc. by the looks of
> things.
>
> Incidentally - I noticed that if I create a horizontal or vertical
> separator in my blank window and then delete it analysis crashes out with
> the following message:
>
> >>> X Error of failed request: BadMatch (invalid parameter attributes)
> Major opcode of failed request: 145 (GLX)
> Minor opcode of failed request: 5 (X_GLXMakeCurrent)
> Serial number of failed request: 8149
> Current serial number in output stream: 8149
>
> Brian
>
> --
> Dr. Brian O. Smith ------------------------ [log in to unmask]
> Division of Biochemistry & Molecular Biology,
> Institute Biomolecular & Life Sciences,
> University of Glasgow, Glasgow G12 8QQ, UK.
> Tel: 0141 330 5167 Fax: 0131 330 8640
>
|