Re: git: ba1904937d9a - main - acpica: Extract _OSC parsing to a common file

From: Kristof Provost <kp_at_FreeBSD.org>
Date: Fri, 27 Dec 2024 17:26:18 UTC
On 26 Dec 2024, at 21:44, Renato Botelho wrote:
> On 12/12/24 15:19, Andrew Turner wrote:
>> The branch main has been updated by andrew:
>>
>> URL: 
>> https://cgit.FreeBSD.org/src/commit/?id=ba1904937d9ae0539e39001467a1519b17177118
>>
>> commit ba1904937d9ae0539e39001467a1519b17177118
>> Author:     Andrew Turner <andrew@FreeBSD.org>
>> AuthorDate: 2024-12-12 16:29:45 +0000
>> Commit:     Andrew Turner <andrew@FreeBSD.org>
>> CommitDate: 2024-12-12 18:00:22 +0000
>>
>>      acpica: Extract _OSC parsing to a common file
>>          This will be used by pci_host_generic_acpi.c so needs to be 
>> in a
>>      common location.
>>          Reviewed by:    imp, jhb
>>      Sponsored by:   Arm Ltd
>>      Differential Revision:  https://reviews.freebsd.org/D48044
>
> My dev system was running CURRENT from Oct 31 and broke after I 
> upgrade to most recent version on Dec 20.  After bisecting I found 
> this commit as the one that introduced the problem.
>
> It starts booting but can't found any disk.  Pressing ? at mount 
> prompt shows nothing.  This system is a VM running on KVM and I've 
> attached dmesg from successfully boot
>
> FreeBSD idaho.arrakis.com.br 15.0-CURRENT FreeBSD 15.0-CURRENT #22 
> n274182-938e4b131c4: Thu Dec 26 18:10:12 -03 2024 
> root@idaho.arrakis.com.br:/usr/obj/usr/src/amd64.amd64/sys/GENERIC-NODEBUG 
> amd64
>

I’m seeing similar boot issues on VMWare aarch64 (i.e. freebsd in 
VMware on a Mac). My bisect pointed to:

	commit 1f5c50a8617355758510675cb9412f56fed12efc
	Author: Andrew Turner <andrew@FreeBSD.org>
	Date:   Thu Dec 12 16:30:39 2024 +0000

	    pci_host_generic:Add pcib_request_feature on ACPI

	    In the ACPI attachment add support for the pcib_request_feature 
method.
	    This uses the common _OSC handling.

	    Reviewed by:    imp, jhb
	    Sponsored by:   Arm Ltd
	    Differential Revision:  https://reviews.freebsd.org/D48048

I’ve not yet been able to capture a dmesg, but it looks like the nvme 
storage device just isn’t found, leading to the zpool failing to 
mount.

I’m currently working around that with this patch:

	diff --git a/sys/dev/pci/pci_host_generic_acpi.c 
b/sys/dev/pci/pci_host_generic_acpi.c
	index 1ff3a6bda707..c990edf26a3c 100644
	--- a/sys/dev/pci/pci_host_generic_acpi.c
	+++ b/sys/dev/pci/pci_host_generic_acpi.c
	@@ -543,7 +543,7 @@ static device_method_t generic_pcie_acpi_methods[] 
= {
	        DEVMETHOD(pcib_release_msix,    
generic_pcie_acpi_release_msix),
	        DEVMETHOD(pcib_map_msi,         generic_pcie_acpi_map_msi),
	        DEVMETHOD(pcib_get_id,          generic_pcie_acpi_get_id),
	-       DEVMETHOD(pcib_request_feature, 
generic_pcie_acpi_request_feature),
	+       //DEVMETHOD(pcib_request_feature,       
generic_pcie_acpi_request_feature),

	        DEVMETHOD_END
	 };

Best regards,
Kristof