External interrupt limitation due to the NVIC #4624
Replies: 1 comment
-
Posted at 2018-03-14 by PaddeK My understanding was PA.1 and PA.2 is fine but not PA.1 and PB.1 . Thats how i understand the 2. Note in the setWatch reference. Posted at 2018-03-14 by @yerpj whoops sorry, you're right. Corrected :-) Posted at 2018-03-14 by @yerpj Well thanks for pointing the note in the Posted at 2018-03-14 by PaddeK This would be my guess too, since its hardware related. Maybe you could add some layer on top of it, like you mentioned with polling, but i imagine this would come at a hefty price speedwise. Posted at 2018-03-16 by @gfwilliams Yes, I'm not sure if it can be worked around :( It's not specific to the Cortex M because the Nordic chips don't have the limitation (you can attach an interrupt to any pin - but then you have less interrupts in total). While STM32F1/F4 have the issue I wonder whether maybe some other ST devices don't. |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
Posted at 2018-03-14 by @yerpj
Hi guys,
I was discussing with my colleagues the problem of the NVIC multiplexer within the STm32 (and probably every CortexM) that limits the number of external interrupts.
For example,
EXTI1_IRQHandler
can be triggered by any pinPx.1
(mutually exclusive) among the available ports, a limitation is therefore that you cannot enable an interrupt simultaneously onPA.1
andPB.1
.Then came Espruino into the discussion, and we wondered how is it possible to do a
setWatch
onPA.1
in addition to anothersetWatch
on pinPB.1
?If possible, what is the trick? Does Espruino poll the inputs and generate virtual interrupts?
Beta Was this translation helpful? Give feedback.
All reactions