On 5/10/21 1:45 PM, Paul Menzel wrote:
Dear Piotr,
Am 10.05.21 um 10:51 schrieb Piotr Król:
On 5/8/21 9:24 AM, Paul Menzel wrote:
Am 28.04.21 um 00:38 schrieb Piotr Król:
(...)
This is one part of the problem, other is specifications compatibility where ACPI is one that breaks things often. coreboot moves with ACPI compiler faster then most of BSD systems, what lead to problems with BSD-based firewalls.
Please excuse my ignorance. I am still not fully understanding the problem, so it’d be great if more concrete examples could be given. For example, what ACPI change caused an OS problem?
- ACPI CPU
https://review.coreboot.org/c/coreboot/+/36258 here we tried to reduce problem https://review.coreboot.org/c/coreboot/+/36258
That is the same URL. I wasn’t able to find the ID or the commit hash in any commit messages in the master branch, and there is no comment in CB:36258 referencing the fixup.
Sorry, I meant this: https://review.coreboot.org/c/coreboot/+/39698
Above patch introduce compliance with most recent version of ACPI spec breaking in the same way BSD (FreeBSD < 12.2 and < 13, most of downstream firewall distros which are based on older FreeBSD).
- ACPI hostbridge
https://review.coreboot.org/c/coreboot/+/36318 further reverted https://review.coreboot.org/c/coreboot/+/37710 we also tried to fix problem according to spec https://review.coreboot.org/c/coreboot/+/37835 but since PCS This caused PCI detection issues.
Those are examples of problems we faced with ACPI.
I would have thought, that the “payload interface” and coreboot tables are the main problems.
Agree.
Kind regards,
Paul