[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH] i386/acpi: restore device paths for pre-5.1 vms
From: |
Michael S. Tsirkin |
Subject: |
Re: [PATCH] i386/acpi: restore device paths for pre-5.1 vms |
Date: |
Tue, 23 Mar 2021 10:48:43 -0400 |
The issue is with people who installed a VM using 5.1 qemu,
migrated to 5.2, booted there and set a config on a device
e.g. IP on a NIC.
They now have a 5.1 machine type but changing uid back
like we do will break these VMs.
Unlikley to be common but let's at least create a way for these people
to used these VMs.
On Mon, Mar 22, 2021 at 06:49:09PM +0300, Vitaly Cheptsov wrote:
> Hi Michael,
>
> That makes little sense in my opinion, these people can simply upgrade the VM
> version, which apparently looks the right way to do it in my opinion.
>
> Best regards,
> Vitaly
>
> > 22 марта 2021 г., в 18:43, Michael S. Tsirkin <mst@redhat.com> написал(а):
> >
> > On Mon, Mar 01, 2021 at 10:59:18PM +0300, Vitaly Cheptsov wrote:
> >> After fixing the _UID value for the primary PCI root bridge in
> >> af1b80ae it was discovered that this change updates Windows
> >> configuration in an incompatible way causing network configuration
> >> failure unless DHCP is used. More details provided on the list:
> >>
> >> https://lists.gnu.org/archive/html/qemu-devel/2021-02/msg08484.html
> >>
> >> This change reverts the _UID update from 1 to 0 for q35 and i440fx
> >> VMs before version 5.2 to maintain the original behaviour when
> >> upgrading.
> >>
> >> Cc: qemu-stable@nongnu.org
> >> Cc: qemu-devel@nongnu.org
> >> Reported-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
> >> Suggested-by: Michael S. Tsirkin <mst@redhat.com>
> >> Signed-off-by: Vitaly Cheptsov <cheptsov@ispras.ru>
> >> ---
> >> hw/i386/acpi-build.c | 4 ++--
> >> hw/i386/pc_piix.c | 2 ++
> >> hw/i386/pc_q35.c | 2 ++
> >> include/hw/i386/pc.h | 1 +
> >> 4 files changed, 7 insertions(+), 2 deletions(-)
> >>
> >> diff --git a/hw/i386/acpi-build.c b/hw/i386/acpi-build.c
> >> index 31a5f6f4a5..442b4629a9 100644
> >> --- a/hw/i386/acpi-build.c
> >> +++ b/hw/i386/acpi-build.c
> >> @@ -1277,7 +1277,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
> >> 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(0)));
> >> + aml_append(dev, aml_name_decl("_UID",
> >> aml_int(pcmc->pci_root_uid)));
> >> aml_append(sb_scope, dev);
> >> aml_append(dsdt, sb_scope);
> >>
> >> @@ -1296,7 +1296,7 @@ build_dsdt(GArray *table_data, BIOSLinker *linker,
> >> 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(0)));
> >> + aml_append(dev, aml_name_decl("_UID",
> >> aml_int(pcmc->pci_root_uid)));
> >> aml_append(dev, build_q35_osc_method());
> >> aml_append(sb_scope, dev);
> >> if (mcfg_valid) {
> >> diff --git a/hw/i386/pc_piix.c b/hw/i386/pc_piix.c
> >> index 2904b40163..46cc951073 100644
> >> --- a/hw/i386/pc_piix.c
> >> +++ b/hw/i386/pc_piix.c
> >> @@ -405,6 +405,7 @@ static void pc_i440fx_machine_options(MachineClass *m)
> >> {
> >> PCMachineClass *pcmc = PC_MACHINE_CLASS(m);
> >> pcmc->default_nic_model = "e1000";
> >> + pcmc->pci_root_uid = 0;
> >>
> >> m->family = "pc_piix";
> >> m->desc = "Standard PC (i440FX + PIIX, 1996)";
> >> @@ -448,6 +449,7 @@ static void pc_i440fx_5_1_machine_options(MachineClass
> >> *m)
> >> compat_props_add(m->compat_props, hw_compat_5_1, hw_compat_5_1_len);
> >> compat_props_add(m->compat_props, pc_compat_5_1, pc_compat_5_1_len);
> >> pcmc->kvmclock_create_always = false;
> >> + pcmc->pci_root_uid = 1;
> >> }
> >>
> >> DEFINE_I440FX_MACHINE(v5_1, "pc-i440fx-5.1", NULL,
> >> diff --git a/hw/i386/pc_q35.c b/hw/i386/pc_q35.c
> >> index 0a212443aa..53450190f5 100644
> >> --- a/hw/i386/pc_q35.c
> >> +++ b/hw/i386/pc_q35.c
> >> @@ -329,6 +329,7 @@ static void pc_q35_machine_options(MachineClass *m)
> >> {
> >> PCMachineClass *pcmc = PC_MACHINE_CLASS(m);
> >> pcmc->default_nic_model = "e1000e";
> >> + pcmc->pci_root_uid = 0;
> >>
> >> m->family = "pc_q35";
> >> m->desc = "Standard PC (Q35 + ICH9, 2009)";
> >> @@ -375,6 +376,7 @@ static void pc_q35_5_1_machine_options(MachineClass *m)
> >> compat_props_add(m->compat_props, hw_compat_5_1, hw_compat_5_1_len);
> >> compat_props_add(m->compat_props, pc_compat_5_1, pc_compat_5_1_len);
> >> pcmc->kvmclock_create_always = false;
> >> + pcmc->pci_root_uid = 1;
> >> }
> >>
> >> DEFINE_Q35_MACHINE(v5_1, "pc-q35-5.1", NULL,
> >> diff --git a/include/hw/i386/pc.h b/include/hw/i386/pc.h
> >> index c9d194a5e7..d4c3d73c11 100644
> >> --- a/include/hw/i386/pc.h
> >> +++ b/include/hw/i386/pc.h
> >> @@ -105,6 +105,7 @@ struct PCMachineClass {
> >> int legacy_acpi_table_size;
> >> unsigned acpi_data_size;
> >> bool do_not_add_smb_acpi;
> >> + int pci_root_uid;
> >>
> >> /* SMBIOS compat: */
> >> bool smbios_defaults;
> >
> > So this upstream, but I think we should also have a property
> > that allows people to have VMs installed with the old
> > machine type booted with a new machine type.
> >
> > E.g. "compat-pci-root-uid".
> >
> > Vitaly could you take a look?
> >
> >
> >> --
> >> 2.24.3 (Apple Git-128)
> >
>