Attention is currently required from: Nico Huber, Wonkyu Kim, Tim Wawrzynczak. Kyösti Mälkki has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/60213 )
Change subject: [WIP] cpu/x86/lapic: Fix choice X2APIC_ONLY ......................................................................
Patch Set 2:
(2 comments)
Commit Message:
https://review.coreboot.org/c/coreboot/+/60213/comment/c8b7888c_cfe0fbef PS2, Line 11: avoid entering a printk() and acquiring console_lock and : dead-locking.
just to be clear, have you observed this happening before?
We have similar comments in lapic_cpu_stop.c:stop_this_cpu(). The comments and current implementation is mine, but it originates way back, CB:776.
Maybe there is some more proper way, like some fence or other barrier?
File src/cpu/x86/mp_init.c:
https://review.coreboot.org/c/coreboot/+/60213/comment/da03288e_84a12c43 PS2, Line 652: lapic_send_ipi(LAPIC_INT_ASSERT | LAPIC_DM_SMI, lapicid());
it looks like this is the only place we send IPI to self. […]
We can do that. There is also stop_this_cpu(), like mentioned before.
I will try if the delay is still necessary if we use the "Self IPI" destination shorthand that does not require passing the target APIC ID.