Nico Huber has posted comments on this change. ( https://review.coreboot.org/c/coreboot/+/31498 )
Change subject: sb/intel/bd82x6x: Fix default IRQ mapping ......................................................................
Patch Set 3:
(1 comment)
https://review.coreboot.org/#/c/31498/1//COMMIT_MSG Commit Message:
https://review.coreboot.org/#/c/31498/1//COMMIT_MSG@11 PS1, Line 11: some register bits).
Can we mention the functions that additionally have their interrupt pin routed after this change?
Done.
Is there any risk that we could trigger interrupt on hardware that will not have their OS driver loaded? I don't remember what would happen then, that GSI/PIRQ line may end up being disabled completely.
For these functions, it is rather unlikely (PCIe root ports and disabled/hidden SATA 2). But generally you are right, Linux would disable the interrupt vector. Also, I have never encountered Intel-chipset-integrated devices that would trigger interrupts without being asked to. Buggy payload could forget to disable them again, though.