Gerwin Klein

> Potentially some other code could exists in code base which appears only with some optimization options. That would not be a problem. Switching off optimisation is not supported and...

Ok, so for the discussion wether this is dead code or not: I can confirm that this is not dead code and it is compiled and tested regularly. It is...

It's definitely there. `objdump -d kernel.elf` shows me (this is with the default `-O2`): ```asm ffffffff808071a0 : ffffffff808071a0: b8 20 00 00 00 movl $32, %eax ffffffff808071a5: 48 83 e8...

Picking up this very old thread.. From a higher-level view the question here seem to be: should `VMCheckBoundNotification` properly receive the message on the notification (resetting the notification state as...

It does look like it's a left-over, yes. @michaelmcinerney and @mktnk3 do you remember why `mcsPreemptionPoint` has an `irq` parameter? It seems to be unused inside.

That'd be very cool. I can help if you want.

> @lsf37 : What is the trick to no longer run the sel4test-sim with `SPIKE64_debug_gcc_64` but use my new RISCVVIRT64 then? I can have a look at that. Does it...

> @lsf37 : What is the trick to no longer run the sel4test-sim with `SPIKE64_debug_gcc_64` but use my new RISCVVIRT64 then? So to get sel4test-sim to run on GitHub with...

> > Does it make sense to test both platforms (spike and qemu-riscv-virt) or do we want to fully remove spike? > > I think the virt platform is enough,...

> > Ok, I'm happy with that. Are you fine with me adding all this to CI or did you want to first continue the experiment on your branch? >...