Richi, On Mon, Jun 8, 2020 at 8:52 AM Sebastian Huber <sebastian.hu...@embedded-brains.de> wrote: > > On 08/06/2020 16:33, Richi Dubey wrote: > > > Thanks for telling me about the uniprocessor rule for having a single > > scheduler for all the tasks. It makes sense. However I still cannot > > find the code which actually links our processor to that scheduler! > > Talking on a high level, I cannot find a code anywhere that tells > > kernel/processor to use Simple scheduler for all the tasks when we > > write #define CONFIGURE_SCHEDULER_SIMPLE inside a testcase(I could not > > find anything related in the scheduler.h configuration file too). I > > hope my doubt makes sense. > > Depending on the configuration options, the header file > <rtems/confdefs/scheduler.h> defines a couple of data structures. One is > the _Scheduler_Table: > > https://git.rtems.org/rtems/tree/cpukit/include/rtems/confdefs/scheduler.h#n239 > > This is how the scheduler is defined for an application. You can look a > the preprocessed header file to see all the data structures. >
You will need to learn some techniques for navigating a large code base. I don't know if there is a good tutorial out there, but check out information about: * ctags * cscope They can integrate with editors on *nix and provide you with identifier/name/text search in an indexed database built from a filesystem directory tree. Or you can use them from the command line also. This is the best way to find code definitions and to navigate code quickly. You can push/pop searches on a stack also, so you can easily follow a function's control flow down, and then backtrace up. UNIX utilities: * grep * ack (an updated variant of grep) * find Gedare _______________________________________________ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel