Re: OVMF and PCI0 UID

["Michael S. Tsirkin" <mst@redhat.com>]

Thanks, I sent the patches on the list now!

On Wed, Jul 22, 2020 at 01:53:23PM +0000, vit9696 via wrote:
> Hello,
> 
> We can confirm that the suggested change to zero UIDs resolves the problem. It
> will be great if you could handle the rest as you see fit. Thank you!
> 
> Best regards,
> Vitaly
> 
> 
> В вт, июля 21, 2020 в 12:24, vit9696 <vit9696@protonmail.com> пишет:
> 
>     Thank you, we will provide an update whether this solves the problem.
> 
>     Besides, this is not the only case where UIDs are wrong for the PCI bus.
>     In hw/arm/virt-acpi-build.c there is the following code:
> 
>         Aml *dev = aml_device("%s", "PCI0");
>         aml_append(dev, aml_name_decl("_HID", aml_string("PNP0A08")));
>         aml_append(dev, aml_name_decl("_CID", aml_string("PNP0A03")));
>         aml_append(dev, aml_name_decl("_SEG", aml_int(0)));
>         aml_append(dev, aml_name_decl("_BBN", aml_int(0)));
>         aml_append(dev, aml_name_decl("_ADR", aml_int(0)));
>         aml_append(dev, aml_name_decl("_UID", aml_string("PCI0")));
>         aml_append(dev, aml_name_decl("_STR", aml_unicode("PCIe 0 Device")));
>         aml_append(dev, aml_name_decl("_CCA", aml_int(1)));
> 
>     https://github.com/qemu/qemu/blob/2c1fb4d/hw/arm/virt-acpi-build.c#
>     L168-L175
> 
>     It makes UID on ARM builds a string, which is certainly not expected. We do
>     not have ARM test setups, but I hope this can be useful too.
> 
>     Best wishes,
>     Vitaly
> 
> 
>         21 июля 2020 г., в 09:58, Michael S. Tsirkin <mst@redhat.com> написал
>         (а):
> 
> 
>         On Mon, Jul 20, 2020 at 11:25:58PM +0200, Laszlo Ersek wrote:
> 
>             Hi Vitaly,
> 
>             adding Igor, Michael, Marcel, and qemu-devel.
> 
>             On 07/20/20 11:06, vit9696 wrote:
> 
>                 Hello,
> 
>                 I discovered an issue with inconsistent QEMU/OVMF device paths,
>                 and
>                 while I am unsure whether directing this e-mail is appropriate
>                 to you,
>                 I believe that you likely have the contacts you could forward
>                 this
>                 e-mail to.
> 
>                 macOS uses ACPI UIDs to build the DevicePath for NVRAM boot
>                 options,
>                 while OVMF firmware gets them via an internal channel through
>                 QEMU.
>                 Due to a bug in QEMU (or OVMF) currently UEFI firmware and ACPI
>                 have
>                 different values, and this makes the underlying operating
>                 system
>                 unable to report its boot option.
> 
>                 The particular node in question is the primary PciRoot (PCI0 in
>                 ACPI),
>                 which for some reason gets assigned 1 in ACPI UID and 0 in the
>                 DevicePath. To me this looks like a bug here:
>                 https://github.com/qemu/qemu/blob/8f06f22/hw/i386/acpi-build.c#
>                 L1511-L1515
>                 Which does not correspond to the primary PCI identifier here:
>                 https://github.com/qemu/qemu/blob/5a79d10/hw/pci/pci.c#
>                 L160-L162
> 
>                 Reference with the device paths, OVMF startup logs, and ACPI
>                 table
>                 dumps (SysReport):
>                 https://github.com/acidanthera/bugtracker/issues/1050
> 
>                 Would you be able to forward this to the right people or
>                 perhaps keep
>                 an eye on the issue itself?
> 
> 
>             I think you are right.
> 
>             In UEFI v2.8, section "10.4.2 Rules with ACPI _HID and _UID" ends
>             with
>             the paragraph,
> 
>                Root PCI bridges will use the plug and play ID of PNP0A03, This
>             will
>                be stored in the ACPI Device Path _HID field, or in the Expanded
>                ACPI Device Path _CID field to match the ACPI name space. The
>             _UID
>                in the ACPI Device Path structure must match the _UID in the
>             ACPI
>                name space.
> 
>             (See especially the last sentence.)
> 
>             Considering *extra* root bridges / root buses (with bus number >
>             0),
>             QEMU's ACPI generator actually does the right thing; since QEMU
>             commit
>             c96d9286a6d7 ("i386/acpi-build: more traditional _UID and _HID for
>             PXB
>             root buses", 2015-06-11).
> 
>             However, the _UID values for root bridge zero (on both i440fx and
>             q35)
>             have always been "wrong" (from UEFI perspective), going back in
>             QEMU to
>             commit 74523b850189 ("i386: add ACPI table files from seabios",
>             2013-10-14).
> 
>             Even in SeaBIOS, these _UID values have always been 1; see commit
>             a4d357638c57 ("Port rombios32 code from bochs-bios.", 2008-03-08)
>             for
>             i440fx, and commit ecbe3fd61511 ("seabios: q35: add dsdt",
>             2012-12-01)
>             for q35.
> 
>             Does the following patch work for you? (I can see you proposed the
>             same
>             in
>             <https://github.com/acidanthera/bugtracker/issues/1050
>             issuecomment-660734139>)
> 
> 
>                 diff --git a/hw/i386/acpi-build.c b/hw/i386/acpi-build.c
>                 index b7bcbbbb2a35..7a5a8b3521b0 100644
>                 --- a/hw/i386/acpi-build.c
>                 +++ b/hw/i386/acpi-build.c
>                 @@ -1496,9 +1496,9 @@ build_dsdt(GArray *table_data, BIOSLinker
>                 *linker,
>                         sb_scope = aml_scope("_SB");
>                         dev = aml_device("PCI0");
>                         aml_append(dev, aml_name_decl("_HID", aml_eisaid
>                 ("PNP0A03")));
>                         aml_append(dev, aml_name_decl("_ADR", aml_int(0)));
>                 -        aml_append(dev, aml_name_decl("_UID", aml_int(1)));
>                 +        aml_append(dev, aml_name_decl("_UID", aml_int(0)));
>                         aml_append(sb_scope, dev);
>                         aml_append(dsdt, sb_scope);
> 
>                         build_hpet_aml(dsdt);
>                 @@ -1511,9 +1511,9 @@ build_dsdt(GArray *table_data, BIOSLinker
>                 *linker,
>                         dev = aml_device("PCI0");
>                         aml_append(dev, aml_name_decl("_HID", aml_eisaid
>                 ("PNP0A08")));
>                         aml_append(dev, aml_name_decl("_CID", aml_eisaid
>                 ("PNP0A03")));
>                         aml_append(dev, aml_name_decl("_ADR", aml_int(0)));
>                 -        aml_append(dev, aml_name_decl("_UID", aml_int(1)));
>                 +        aml_append(dev, aml_name_decl("_UID", aml_int(0)));
>                         aml_append(dev, build_q35_osc_method());
>                         aml_append(sb_scope, dev);
>                         aml_append(dsdt, sb_scope);
> 
> 
>             If it does, I suggest submitting the above patch to qemu-devel, and
>             /or
>             filing a bug for upstream QEMU at <https://bugs.launchpad.net/qemu/
>             >.
> 
> 
>         Or even just reporting whether the above helps you, we can
>         take it from there.
> 
> 
>             (Note: I didn't even compile the above change.)
> 
>             Thanks
>             Laszlo
> 
> 
> 
> 
>