arm64: Ensure that thread0's PCB flags are initialized

[markj]

Summary

On arm64, the PCB is stored at the top of the thread stack. For thread0
this comes from the static "initstack" region, which is placed in the
.init_pagetable section, which is not part of the BSS and thus doesn't
get zeroed by locore. (See the comment in ldscript.arm64.) It is thus
possible for the pcb_flags field to be uninitialized, which can result
in PCB_SINGLE_STEP being set.

Fix this by simply initializing the field. Alternately we could move
initstack to the BSS. OTOH, kernel stacks for all other threads also
should not be assumed to be zero-filled, so it's probably better to be
consistent.

[andrew]

I think it's an accident it's in the .init_pagetable section. I don't see a reason to not move it to .bss if it fixes issues where the pcb has uninitialised data.

[markj]

In D39343#895990, @andrew wrote:

I think it's an accident it's in the .init_pagetable section. I don't see a reason to not move it to .bss if it fixes issues where the pcb has uninitialised data.

I suspect we should simply do both. Explicit initialization here is better, but there's no reason for initstack to be in init_pagetable.

[markj]

In D39343#896044, @markj wrote:

In D39343#895990, @andrew wrote:

I think it's an accident it's in the .init_pagetable section. I don't see a reason to not move it to .bss if it fixes issues where the pcb has uninitialised data.

I suspect we should simply do both. Explicit initialization here is better, but there's no reason for initstack to be in init_pagetable.

https://reviews.freebsd.org/D39367