Ouss4 commented on pull request #5336:
URL: https://github.com/apache/incubator-nuttx/pull/5336#issuecomment-1042711844


   > I have modify up_enable_irq with this patch
   
   That's the part I was referring to as going to be a bit messy.  :)  It's 
just because we are going to treat the software interrupt as a special case 
compared to the rest of interrupts.  However, I think it would be a bit less 
messier if we enabled the CPU interrupt without using `up_enable_irq` and 
directly calling `xtensa_enable_cpuint` for both CPUs.
   
   > So my question is, we must find an elegant way,
   to solve g_irqmap must be g_irqmap[CONFIG_SMP_NCPUS] for internal interrupts.
   
   I agree that it needs some work.  The original idea of having the same map 
was to be able to enable/disable internal CPU interrupts from a different CPU.  
When one CPU enabled an interrupt the same CPU needed to disable that 
interrupt.  This was done using an inter CPU interrupt as is done for the CXD56 
chip.  However, that part was removed and peripheral interrupts are only 
controlled by the interrupt matrix.
   So I don't think right now we need to map the IRQ to a CPU+CPU interrupt 
anymore when it comes to internal interrupts.  I think what you are suggesting 
could work and make things a bit simpler.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


Reply via email to