Re: [Gems-users] Using Multiple CC Protocols

[Greg Byrd <gbyrd@xxxxxxxx>]

For some reason, the ruby Makefile does not do the move properly for Simics 3.x, but you can easily change that.  Just copy and adjust the steps used for 2.x to copy files into the DESTINATION directory.  

I believe that the SIMICS_EXTRA_LIB step works the same for 3.x as it does for 2.x.  I switched to 3.x for my class this spring, and my students didn't report any problems with that approach.

...gb


Edward Lee wrote:
> Hi Dan,
>
> Thanks for your reply.
>
> However, DESTINATION=blah does not work for Simics 3.0.x specifically 
> for the "movemodule" part. That's why I tried to that manually. And if I 
> don't, when I compile the new protocol it replaces the old ruby.so in 
> "$HOST_TYPE/lib/" . And I always run protocol P1 from its own directory 
> as well as P2 from its own directory. So, is there any problem wih what 
> I am doing here?
> Regards,
>
> Ed
>
>
> On Sat, May 30, 2009 at 1:13 AM, Dan Gibson <degibson@xxxxxxxx 
> <mailto:degibson@xxxxxxxx>> wrote:
>     This is exactly the purpose of the DESTINATION=blah portion of 'make
>     PROTOCOL=prot DESTINATION=blah', shown in the examples on the wiki.
>     We strongly recommend you keep to this practice, and run protocol P1
>     from $GEMS/simics/home/P1 and protocol P2 from $GEMS/simics/home/P2.
>     You should have no need to copy .so files around.
>
>     Regards,
>     Dan
>
>     On Fri, May 29, 2009 at 11:43 PM, Edward Lee <edwl202@xxxxxxxxx
>     <mailto:edwl202@xxxxxxxxx>> wrote:
>
>         Hi,
>
>         I have a problem running multiple protocols from GEMS workspace.
>         I know this had been discussed before and I am pretty much
>         following the suggestions but no success so far.
>
>         So this is what I am doing:
>
>         Although the new compiled *.so files do not go to target
>         protocol's folder, I copy them to the specific protocol's
>         "modules" directory. And SIMICS_EXTRA_LIB is directing the
>         ./modules so as far as I understand RUBY should be loaded from
>         there which has the corresponding dynamic libraries. But anytime
>         I run a test, I get results from the last compiled CC protocol.
>         And if I delete the ruby.so from workspace/$(HOST_TYPE)/lib/
>         ruby has failed to initialize, so looks like it is not really
>         looking for "./modules" at all. Any idea what I am missing here?
>
>         Regards,
>
>         Ed
>
>         _______________________________________________
>         Gems-users mailing list
>         Gems-users@xxxxxxxxxxx <mailto:Gems-users@xxxxxxxxxxx>
>         https://lists.cs.wisc.edu/mailman/listinfo/gems-users
>         Use Google to search the GEMS Users mailing list by adding
>         "site:https://lists.cs.wisc.edu/archive/gems-users/"; to your search.
>
>
>
>
>
>     -- 
>     http://www.cs.wisc.edu/~gibson <http://www.cs.wisc.edu/%7Egibson>
>     [esc]:wq!
>
>     _______________________________________________
>     Gems-users mailing list
>     Gems-users@xxxxxxxxxxx <mailto:Gems-users@xxxxxxxxxxx>
>     https://lists.cs.wisc.edu/mailman/listinfo/gems-users
>     Use Google to search the GEMS Users mailing list by adding
>     "site:https://lists.cs.wisc.edu/archive/gems-users/"; to your search.
>
>
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> Gems-users mailing list
> Gems-users@xxxxxxxxxxx
> https://lists.cs.wisc.edu/mailman/listinfo/gems-users
> Use Google to search the GEMS Users mailing list by adding "site:https://lists.cs.wisc.edu/archive/gems-users/"; to your search.