Re: [OMPI devel] Aggregate MCA Parameter Sets
I'm coming back on this feature. Whatever it does, it doesn't do it right. If we support a new feature, then at least it should be supported by all launchers (HINT it should go somewhere in the base). As I read the PLS code, only the RSH PLS can have this feature. That's called discrimination !!! Either we support it everywhere, or we completely remove it. george. On Mar 2, 2007, at 3:58 PM, Jeff Squyres wrote: On Mar 2, 2007, at 3:37 PM, Josh Hursey wrote: Certainly the end user could specify another file to use in addition to the default MCA parameter files (mca_param_files), but then they must also specify the other locations that already exist in that path (e.g., $HOME/.openmpi/mca-params.conf:$SYSCONFDIR/openmpi-mca- params.conf). This is a short cut in a sense, so the end user doesn't have to know all of this uglyness every time they want to run a benchmark, or ... This is the main point of AMCA's: user convenience. Yes, you could supply 20 MCA params on the command line, or set them in the environment, or add them to your $HOME/.openmpi .conf file, or add an individual file to your MCA param file list, or ... The --amca param is effectively syntactic sugar for adding a .conf file to the MCA param file list. It seemed a whole lot easier to do: mpirun --amca my_benchmark_settings.conf ... Than foo=`ompi_info --param mca base --parsable | grep mca_param_files:value | cut -d: -f7-` mpirun --mca mca_base_path my_benchmark_settings.conf:$foo ... The intent is that various AMCA files can be introduced and installed around the code base (I have one for the openib BTL for benchmarking, for example -- settings that you wouldn't use in production, but you would use with netpipe). -- Jeff Squyres Server Virtualization Business Unit Cisco Systems ___ devel mailing list de...@open-mpi.org http://www.open-mpi.org/mailman/listinfo.cgi/devel smime.p7s Description: S/MIME cryptographic signature
Re: [OMPI devel] Aggregate MCA Parameter Sets
Discriminate we did. Not intentionally, but you are correct that it is not supported by all the launchers yet. We are not removing it, but I don't have the cycles at the moment to add the support to all the different launchers. It is about 10 lines that need to be added to the launchers, and shouldn't take much time. Are you volunteering? -- Josh On Apr 5, 2007, at 1:10 PM, George Bosilca wrote: I'm coming back on this feature. Whatever it does, it doesn't do it right. If we support a new feature, then at least it should be supported by all launchers (HINT it should go somewhere in the base). As I read the PLS code, only the RSH PLS can have this feature. That's called discrimination !!! Either we support it everywhere, or we completely remove it. george. On Mar 2, 2007, at 3:58 PM, Jeff Squyres wrote: On Mar 2, 2007, at 3:37 PM, Josh Hursey wrote: Certainly the end user could specify another file to use in addition to the default MCA parameter files (mca_param_files), but then they must also specify the other locations that already exist in that path (e.g., $HOME/.openmpi/mca-params.conf:$SYSCONFDIR/openmpi-mca- params.conf). This is a short cut in a sense, so the end user doesn't have to know all of this uglyness every time they want to run a benchmark, or ... This is the main point of AMCA's: user convenience. Yes, you could supply 20 MCA params on the command line, or set them in the environment, or add them to your $HOME/.openmpi .conf file, or add an individual file to your MCA param file list, or ... The --amca param is effectively syntactic sugar for adding a .conf file to the MCA param file list. It seemed a whole lot easier to do: mpirun --amca my_benchmark_settings.conf ... Than foo=`ompi_info --param mca base --parsable | grep mca_param_files:value | cut -d: -f7-` mpirun --mca mca_base_path my_benchmark_settings.conf:$foo ... The intent is that various AMCA files can be introduced and installed around the code base (I have one for the openib BTL for benchmarking, for example -- settings that you wouldn't use in production, but you would use with netpipe). -- Jeff Squyres Server Virtualization Business Unit Cisco Systems ___ devel mailing list de...@open-mpi.org http://www.open-mpi.org/mailman/listinfo.cgi/devel ___ devel mailing list de...@open-mpi.org http://www.open-mpi.org/mailman/listinfo.cgi/devel
Re: [OMPI devel] Aggregate MCA Parameter Sets
I think you didn't get the idea. I'm absolutely not volunteering. I'm just stating that it get into the trunk way to early in a stage where it shouldn't. And that's starting to become a very common problem in Open MPI nowadays. Moreover, as it is supposed to be a generic aggregate file why should we copy&paste (even 10 lines of code)? It shouldn't be in any of the components, if it's general then it should go in the base. george. On Apr 5, 2007, at 1:25 PM, Josh Hursey wrote: Discriminate we did. Not intentionally, but you are correct that it is not supported by all the launchers yet. We are not removing it, but I don't have the cycles at the moment to add the support to all the different launchers. It is about 10 lines that need to be added to the launchers, and shouldn't take much time. Are you volunteering? -- Josh On Apr 5, 2007, at 1:10 PM, George Bosilca wrote: I'm coming back on this feature. Whatever it does, it doesn't do it right. If we support a new feature, then at least it should be supported by all launchers (HINT it should go somewhere in the base). As I read the PLS code, only the RSH PLS can have this feature. That's called discrimination !!! Either we support it everywhere, or we completely remove it. george. On Mar 2, 2007, at 3:58 PM, Jeff Squyres wrote: On Mar 2, 2007, at 3:37 PM, Josh Hursey wrote: Certainly the end user could specify another file to use in addition to the default MCA parameter files (mca_param_files), but then they must also specify the other locations that already exist in that path (e.g., $HOME/.openmpi/mca-params.conf:$SYSCONFDIR/openmpi-mca- params.conf). This is a short cut in a sense, so the end user doesn't have to know all of this uglyness every time they want to run a benchmark, or ... This is the main point of AMCA's: user convenience. Yes, you could supply 20 MCA params on the command line, or set them in the environment, or add them to your $HOME/.openmpi .conf file, or add an individual file to your MCA param file list, or ... The --amca param is effectively syntactic sugar for adding a .conf file to the MCA param file list. It seemed a whole lot easier to do: mpirun --amca my_benchmark_settings.conf ... Than foo=`ompi_info --param mca base --parsable | grep mca_param_files:value | cut -d: -f7-` mpirun --mca mca_base_path my_benchmark_settings.conf:$foo ... The intent is that various AMCA files can be introduced and installed around the code base (I have one for the openib BTL for benchmarking, for example -- settings that you wouldn't use in production, but you would use with netpipe). -- Jeff Squyres Server Virtualization Business Unit Cisco Systems ___ devel mailing list de...@open-mpi.org http://www.open-mpi.org/mailman/listinfo.cgi/devel ___ devel mailing list de...@open-mpi.org http://www.open-mpi.org/mailman/listinfo.cgi/devel ___ devel mailing list de...@open-mpi.org http://www.open-mpi.org/mailman/listinfo.cgi/devel smime.p7s Description: S/MIME cryptographic signature
