On Thu, Aug 1, 2019 at 10:33 AM Ricardo Gomes (1161078) <1161...@isep.ipp.pt> wrote: > > Helo, > > First of all, thank you for your response. > You are correct, it freezes at the point it jumps to RTEMS entry point > address. > I cannot understand why it works with the non-smp samples and fails with smp > samples, should I change the address? > You might use objdump to verify the expected instructions at the start address. I don't see that the starting point should differ between SMP and UP, but there are some SMP-specific instructions in the start sequence (bsps/arm/shared/start/start.S), are they being reached?
I don't have experience with the RPi2 or access to the board, so I can't offer much beyond that. > > ________________________________ > De: Gedare Bloom <ged...@rtems.org> > Enviado: 1 de agosto de 2019 04:17 > Para: Ricardo Gomes (1161078) <1161...@isep.ipp.pt> > Cc: devel@rtems.org <devel@rtems.org> > Assunto: Re: RTEMS SMP on Raspberry Pi 2 > > On Thu, Aug 1, 2019 at 4:49 AM Ricardo Gomes (1161078) > <1161...@isep.ipp.pt> wrote: > > > > Hello everyone, > > > > I'm trying to collect some results of some experiments that I'm doing > > related with synchronization mechanisms for SMPs. > > So far, I have the results for QEMU but I'm trying to use a Raspberry Pi 2 > > to obtain results in real hardware. > > > > I saw the new patches that were recently added for the raspberry pi bsp and > > as mentioned in the emails related to the new patches I changed the start > > address to 0x200000. > > > > However, I have a problem running my samples in the board. The kernel > > compiles without any problem. I'm using the following commands to compile > > it: > > > > arm-rtems5-objcopy -Obinary > > > > > > > > Now the issue occurs when I try to execute my tests for SMP, as the kernel > > freezes during the boot loader first stage. > > > > As a side note, when I compile and execute something para single processor > > na RPi2 I don't have any issue. The issue only occurs when dealing with SMP. > > > > Can someone help me in figuring out what I'm doing wrong? > > > Can you clarify what is the "boot loader first stage"? Usually, that > is pre-kernel code, since it will load the binary image of the kernel, > so it is strange to see different behavior based on different compiled > kernels. Unless it is freezing at the point it jumps to RTEMS entry > start address. > > > Thanks in advance. > > Regards, > > _______________________________________________ > > 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
