Sorry, I forgot to reply. My problem was probably due to concurrent access to lwIP, since my low_level_output function might block if the ETH Tx buffer is full, waiting for a Tx trigger. I've created an access queue based on messages to simulate a lock in the high-level driver.
Vitor Em sex, 26 de out de 2018 às 18:29, Jan Menzel <men...@peperoni-light.de> escreveu: > On 26.10.2018 15:57, vr roriz wrote: > >> Vitor, make sure your processes are never cycled by the scheduler > > otherwise you'll violate the threading requirements. I'd suggest to just > > have a single thread for lwip and use a mailbox or similar to trigger > > certain actions. The OS will take care that you can write into the > > mailbox from any interrupt/thread/timer. > > > > Hi Jan, I do have a "mailbox", I am using a message-passing OS. All my > > interrupts just send messages to processes that execute lwIP code, all > > these processes have the same priority. What do you exactly mean by " > > never cycle"? The scheduling algo is priority level based, without > > time-slices. > > Just make sure your scheduler never interrupts one process in favour for > an other with same priority. I would consider this non standard. > Alternatively you could have one lock shared between all your processes > to force the scheduler to never interrupt any lwip function to execute > any other. > > Jan > > _______________________________________________ > lwip-users mailing list > lwip-users@nongnu.org > https://lists.nongnu.org/mailman/listinfo/lwip-users >
_______________________________________________ lwip-users mailing list lwip-users@nongnu.org https://lists.nongnu.org/mailman/listinfo/lwip-users
