Technically all tasks in SCons run in parallel, thanks to a GIL. What run in parallel are the subprocess calls to a command line ( as these turn into waits from the python point of view). The easy way to do what you suggest its to add locks around this shared data to force the exeution of one task at a time.
Jason -----Original Message----- From: scons-dev-boun...@scons.org [mailto:scons-dev-boun...@scons.org] On Behalf Of alex.bur...@csiro.au Sent: Wednesday, November 20, 2013 5:03 PM To: scons-dev@scons.org Subject: [Scons-dev] SCons with shared resource Hi, I am successfully using scons for batch processing of simulation work. Unfortunately some of the tasks share resources that prevent them being executed in parallel. Other tasks are can be run in parallel. This means that I need to run scons in single processor mode, meaning some tasks are unnecessarily serialised. I am considering modifying the scons code to allow specification of a shared resource/ mutex for certain tasks. Does anyone have any advise as to how this is best implemented ? Thanks, Alex Alex Burton Research Engineer - Solar Thermal - Energy Technology - CSIRO +61 2 4960 6110 +61 416 594 752 _______________________________________________ Scons-dev mailing list Scons-dev@scons.org http://two.pairlist.net/mailman/listinfo/scons-dev _______________________________________________ Scons-dev mailing list Scons-dev@scons.org http://two.pairlist.net/mailman/listinfo/scons-dev
