Dear Dr. Bloom, I have added my name on the GSoC Tracking list and have shared a draft proposal with you. Please suggest changes in my proposal: https://docs.google.com/document/d/1bth9sh3qH2vMTks1yiGpZVskUcD-g5jED4a3zJx1Klc/edit?usp=sharing
Also, currently the extension ideas I have in my mind are following: 1) Add support for multiple criticalilty level for tasks, and work on another scheduler which would takes into account criticality level and the multicore system. For example: EDF-VD by Dr. Baruah as it is 100% schedulable. 2) Add priority ordering for asynchronous task according to Audsley's optimal priority ordering (https://www.sciencedirect.com/science/article/pii/S0020019000001654) Please let me know your thoughts. Thanks, Richi. On Fri, Mar 13, 2020 at 12:53 AM Gedare Bloom <ged...@rtems.org> wrote: > > Hi Richi, > > Welcome. The "Improve the SMP Scheduler" will require strong C > programming skills before the summer. You will be expected to > demonstrate that during the proposal period. I'm the only likely > mentor for that project this year, and I haven't decided if I will > mentor any projects, or which project I might mentor if I do. So it > can be a bit risky to pursue that project, but if you are passionate > about it and have the required capabilities, then you should start to > prepare a proposal and convince me that I should mentor you :). > > The project itself is a straightforward implementation from the > current skeleton of the scheduler. So you will want to also include > some "extension" activities in case you complete the scheduler > quickly. > > You should add yourself to https://devel.rtems.org/wiki/GSoC/2020 > > Gedare > > On Thu, Mar 12, 2020 at 1:04 PM Richi Dubey <richidu...@gmail.com> wrote: > > > > Hey everyone, > > > > Could someone please tell me if this project > > (https://devel.rtems.org/ticket/2510) titled > > "Improve the SMP scheduler with arbitrary processor affinity support" > > still open? I went through the mpi-sws paper and I found the concept > > of implementing > > task's affinity to a processor to increase its efficiency really > > interesting. Can someone please help me with this? > > > > Thanks, > > Richi. > > _______________________________________________ > > devel mailing list > > devel@rtems.org > > http://lists.rtems.org/mailman/listinfo/devel _______________________________________________ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel
