Is it possible to build and run sel4-test against the "experimental"
kernel tree? if so, whats the process?
On RefOS for my port, I'm seeing a weird problem with the timer code.
It is mapping in the right phys addr for my timer device, and I can
read the right ID register out, so it appears to be mapping the right
memory, but the code to do a timer reset, which sets a reset bit
then polls for the bit to return to zero, is getting stuck in an infinite
loop. I verified that the loop isn't being altered by the compiler, and I
verified that the same code runs fine in sel4 test. At this point I
suspect one of two things
1) refos is allowing the mapping to happen with caching on (even
though I asked it not to) or
2) something is wrong elsewhere with the "experimental" kernel
(sel4-test is running on the "master" tree, and refos is using
the kernel from the "experimental" tree).
is there anything else I could be missing? any thoughts?
--
Tim Newsham | www.thenewsh.com/~newsham | @newshtwit | thenewsh.blogspot.com
_______________________________________________
Devel mailing list
[email protected]
https://sel4.systems/lists/listinfo/devel
