I have struggled with the PRU scratchpad, persisting because it seems that 
it could be a very powerful feature. In particular, the XCHG instruction 
(pseudo-op) has caused me hours of frustration to identify and diagnose a 
problem. It seems that XCHG (swapping current PRU register values for those 
in a specified scratchpad bank) does not work - it is effectively an XIN 
op. I researched and found only one bit of content about this, here 
<https://e2e.ti.com/support/arm/sitara_arm/f/791/p/340589/1189854>: The 
accepted answer was provided by a (supposed) TI employee. I find it hard to 
believe...

Is it possible that he PASM compiler accepts the XCHG instruction, and all 
the PRU documentation from TI documents this instruction, but it is not 
even implemented? I understand the "PRU is not officially supported" claim.

-- 
For more options, visit http://beagleboard.org/discuss
--- 
You received this message because you are subscribed to the Google Groups 
"BeagleBoard" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/beagleboard/42441994-13c9-4309-80f3-16b61c2f6e77%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to