[PATCH v2] acpi: fix acpi_index migration

[PATCH v2] acpi: fix acpi_index migration

[Igor Mammedov]

From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>

vmstate_acpi_pcihp_use_acpi_index() was expecting AcpiPciHpState
as state but it actually received PIIX4PMState, because
VMSTATE_PCI_HOTPLUG is a macro and not another struct.
So it ended up accessing random pointer, which resulted
in 'false' return value and acpi_index field wasn't ever
sent.

However in 7.0 that pointer de-references to value > 0, and
destination QEMU starts to expect the field which isn't
sent in migratioon stream from older QEMU (6.2 and older).
As result migration fails with:
  qemu-system-x86_64: Missing section footer for 0000:00:01.3/piix4_pm
  qemu-system-x86_64: load of migration failed: Invalid argument

In addition with QEMU-6.2, destination due to not expected
state, also never expects the acpi_index field in migration
stream.

Q35 is not affected as it always sends/expects the field as
long as acpi based PCI hotplug is enabled.

Fix issue by introducing compat knob to never send/expect
acpi_index in migration stream for 6.2 and older PC machine
types and always send it for 7.0 and newer PC machine types.

Diagnosed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Fixes: b32bd76 ("pci: introduce acpi-index property for PCI device")
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/932
Signed-off-by: Igor Mammedov <imammedo@redhat.com>
---
 include/hw/acpi/pcihp.h         |  2 --
 hw/acpi/acpi-pci-hotplug-stub.c |  4 ----
 hw/acpi/pcihp.c                 |  6 ------
 hw/acpi/piix4.c                 | 15 ++++++++++++++-
 hw/core/machine.c               |  4 +++-
 5 files changed, 17 insertions(+), 14 deletions(-)

diff --git a/include/hw/acpi/pcihp.h b/include/hw/acpi/pcihp.h
index af1a169fc3..7e268c2c9c 100644
--- a/include/hw/acpi/pcihp.h
+++ b/include/hw/acpi/pcihp.h
@@ -73,8 +73,6 @@ void acpi_pcihp_reset(AcpiPciHpState *s, bool acpihp_root_off);
 
 extern const VMStateDescription vmstate_acpi_pcihp_pci_status;
 
-bool vmstate_acpi_pcihp_use_acpi_index(void *opaque, int version_id);
-
 #define VMSTATE_PCI_HOTPLUG(pcihp, state, test_pcihp, test_acpi_index) \
         VMSTATE_UINT32_TEST(pcihp.hotplug_select, state, \
                             test_pcihp), \
diff --git a/hw/acpi/acpi-pci-hotplug-stub.c b/hw/acpi/acpi-pci-hotplug-stub.c
index 734e4c5986..a43f6dafc9 100644
--- a/hw/acpi/acpi-pci-hotplug-stub.c
+++ b/hw/acpi/acpi-pci-hotplug-stub.c
@@ -41,7 +41,3 @@ void acpi_pcihp_reset(AcpiPciHpState *s, bool acpihp_root_off)
     return;
 }
 
-bool vmstate_acpi_pcihp_use_acpi_index(void *opaque, int version_id)
-{
-    return false;
-}
diff --git a/hw/acpi/pcihp.c b/hw/acpi/pcihp.c
index 6351bd3424..bf65bbea49 100644
--- a/hw/acpi/pcihp.c
+++ b/hw/acpi/pcihp.c
@@ -554,12 +554,6 @@ void acpi_pcihp_init(Object *owner, AcpiPciHpState *s, PCIBus *root_bus,
                                    OBJ_PROP_FLAG_READ);
 }
 
-bool vmstate_acpi_pcihp_use_acpi_index(void *opaque, int version_id)
-{
-     AcpiPciHpState *s = opaque;
-     return s->acpi_index;
-}
-
 const VMStateDescription vmstate_acpi_pcihp_pci_status = {
     .name = "acpi_pcihp_pci_status",
     .version_id = 1,
diff --git a/hw/acpi/piix4.c b/hw/acpi/piix4.c
index cc37fa3416..fe5625d07a 100644
--- a/hw/acpi/piix4.c
+++ b/hw/acpi/piix4.c
@@ -82,6 +82,7 @@ struct PIIX4PMState {
     AcpiPciHpState acpi_pci_hotplug;
     bool use_acpi_hotplug_bridge;
     bool use_acpi_root_pci_hotplug;
+    bool not_migrate_acpi_index;
 
     uint8_t disable_s3;
     uint8_t disable_s4;
@@ -267,6 +268,16 @@ static bool piix4_vmstate_need_smbus(void *opaque, int version_id)
     return pm_smbus_vmstate_needed();
 }
 
+/*
+ * This is a fudge to turn off the acpi_index field,
+ * whose test was always broken on piix4 with 6.2 and older machine types.
+ */
+static bool vmstate_test_migrate_acpi_index(void *opaque, int version_id)
+{
+    PIIX4PMState *s = PIIX4_PM(opaque);
+    return s->use_acpi_hotplug_bridge && !s->not_migrate_acpi_index;
+}
+
 /* qemu-kvm 1.2 uses version 3 but advertised as 2
  * To support incoming qemu-kvm 1.2 migration, change version_id
  * and minimum_version_id to 2 below (which breaks migration from
@@ -297,7 +308,7 @@ static const VMStateDescription vmstate_acpi = {
             struct AcpiPciHpPciStatus),
         VMSTATE_PCI_HOTPLUG(acpi_pci_hotplug, PIIX4PMState,
                             vmstate_test_use_acpi_hotplug_bridge,
-                            vmstate_acpi_pcihp_use_acpi_index),
+                            vmstate_test_migrate_acpi_index),
         VMSTATE_END_OF_LIST()
     },
     .subsections = (const VMStateDescription*[]) {
@@ -652,6 +663,8 @@ static Property piix4_pm_properties[] = {
     DEFINE_PROP_BOOL("memory-hotplug-support", PIIX4PMState,
                      acpi_memory_hotplug.is_enabled, true),
     DEFINE_PROP_BOOL("smm-compat", PIIX4PMState, smm_compat, false),
+    DEFINE_PROP_BOOL("x-not-migrate-acpi-index", PIIX4PMState,
+                      not_migrate_acpi_index, false),
     DEFINE_PROP_END_OF_LIST(),
 };
 
diff --git a/hw/core/machine.c b/hw/core/machine.c
index d856485cb4..1e23fdc14b 100644
--- a/hw/core/machine.c
+++ b/hw/core/machine.c
@@ -37,7 +37,9 @@
 #include "hw/virtio/virtio.h"
 #include "hw/virtio/virtio-pci.h"
 
-GlobalProperty hw_compat_6_2[] = {};
+GlobalProperty hw_compat_6_2[] = {
+    { "PIIX4_PM", "x-not-migrate-acpi-index", "on"},
+};
 const size_t hw_compat_6_2_len = G_N_ELEMENTS(hw_compat_6_2);
 
 GlobalProperty hw_compat_6_1[] = {
-- 
2.31.1

Re: [PATCH v2] acpi: fix acpi_index migration

[Dr. David Alan Gilbert]

* Igor Mammedov (imammedo@redhat.com) wrote:
> From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
> 
> vmstate_acpi_pcihp_use_acpi_index() was expecting AcpiPciHpState
> as state but it actually received PIIX4PMState, because
> VMSTATE_PCI_HOTPLUG is a macro and not another struct.
> So it ended up accessing random pointer, which resulted
> in 'false' return value and acpi_index field wasn't ever
> sent.
> 
> However in 7.0 that pointer de-references to value > 0, and
> destination QEMU starts to expect the field which isn't
> sent in migratioon stream from older QEMU (6.2 and older).
> As result migration fails with:
>   qemu-system-x86_64: Missing section footer for 0000:00:01.3/piix4_pm
>   qemu-system-x86_64: load of migration failed: Invalid argument
> 
> In addition with QEMU-6.2, destination due to not expected
> state, also never expects the acpi_index field in migration
> stream.
> 
> Q35 is not affected as it always sends/expects the field as
> long as acpi based PCI hotplug is enabled.
> 
> Fix issue by introducing compat knob to never send/expect
> acpi_index in migration stream for 6.2 and older PC machine
> types and always send it for 7.0 and newer PC machine types.
> 
> Diagnosed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
> Fixes: b32bd76 ("pci: introduce acpi-index property for PCI device")
> Resolves: https://gitlab.com/qemu-project/qemu/-/issues/932
> Signed-off-by: Igor Mammedov <imammedo@redhat.com>

Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

> ---
>  include/hw/acpi/pcihp.h         |  2 --
>  hw/acpi/acpi-pci-hotplug-stub.c |  4 ----
>  hw/acpi/pcihp.c                 |  6 ------
>  hw/acpi/piix4.c                 | 15 ++++++++++++++-
>  hw/core/machine.c               |  4 +++-
>  5 files changed, 17 insertions(+), 14 deletions(-)
> 
> diff --git a/include/hw/acpi/pcihp.h b/include/hw/acpi/pcihp.h
> index af1a169fc3..7e268c2c9c 100644
> --- a/include/hw/acpi/pcihp.h
> +++ b/include/hw/acpi/pcihp.h
> @@ -73,8 +73,6 @@ void acpi_pcihp_reset(AcpiPciHpState *s, bool acpihp_root_off);
>  
>  extern const VMStateDescription vmstate_acpi_pcihp_pci_status;
>  
> -bool vmstate_acpi_pcihp_use_acpi_index(void *opaque, int version_id);
> -
>  #define VMSTATE_PCI_HOTPLUG(pcihp, state, test_pcihp, test_acpi_index) \
>          VMSTATE_UINT32_TEST(pcihp.hotplug_select, state, \
>                              test_pcihp), \
> diff --git a/hw/acpi/acpi-pci-hotplug-stub.c b/hw/acpi/acpi-pci-hotplug-stub.c
> index 734e4c5986..a43f6dafc9 100644
> --- a/hw/acpi/acpi-pci-hotplug-stub.c
> +++ b/hw/acpi/acpi-pci-hotplug-stub.c
> @@ -41,7 +41,3 @@ void acpi_pcihp_reset(AcpiPciHpState *s, bool acpihp_root_off)
>      return;
>  }
>  
> -bool vmstate_acpi_pcihp_use_acpi_index(void *opaque, int version_id)
> -{
> -    return false;
> -}
> diff --git a/hw/acpi/pcihp.c b/hw/acpi/pcihp.c
> index 6351bd3424..bf65bbea49 100644
> --- a/hw/acpi/pcihp.c
> +++ b/hw/acpi/pcihp.c
> @@ -554,12 +554,6 @@ void acpi_pcihp_init(Object *owner, AcpiPciHpState *s, PCIBus *root_bus,
>                                     OBJ_PROP_FLAG_READ);
>  }
>  
> -bool vmstate_acpi_pcihp_use_acpi_index(void *opaque, int version_id)
> -{
> -     AcpiPciHpState *s = opaque;
> -     return s->acpi_index;
> -}
> -
>  const VMStateDescription vmstate_acpi_pcihp_pci_status = {
>      .name = "acpi_pcihp_pci_status",
>      .version_id = 1,
> diff --git a/hw/acpi/piix4.c b/hw/acpi/piix4.c
> index cc37fa3416..fe5625d07a 100644
> --- a/hw/acpi/piix4.c
> +++ b/hw/acpi/piix4.c
> @@ -82,6 +82,7 @@ struct PIIX4PMState {
>      AcpiPciHpState acpi_pci_hotplug;
>      bool use_acpi_hotplug_bridge;
>      bool use_acpi_root_pci_hotplug;
> +    bool not_migrate_acpi_index;
>  
>      uint8_t disable_s3;
>      uint8_t disable_s4;
> @@ -267,6 +268,16 @@ static bool piix4_vmstate_need_smbus(void *opaque, int version_id)
>      return pm_smbus_vmstate_needed();
>  }
>  
> +/*
> + * This is a fudge to turn off the acpi_index field,
> + * whose test was always broken on piix4 with 6.2 and older machine types.
> + */
> +static bool vmstate_test_migrate_acpi_index(void *opaque, int version_id)
> +{
> +    PIIX4PMState *s = PIIX4_PM(opaque);
> +    return s->use_acpi_hotplug_bridge && !s->not_migrate_acpi_index;
> +}
> +
>  /* qemu-kvm 1.2 uses version 3 but advertised as 2
>   * To support incoming qemu-kvm 1.2 migration, change version_id
>   * and minimum_version_id to 2 below (which breaks migration from
> @@ -297,7 +308,7 @@ static const VMStateDescription vmstate_acpi = {
>              struct AcpiPciHpPciStatus),
>          VMSTATE_PCI_HOTPLUG(acpi_pci_hotplug, PIIX4PMState,
>                              vmstate_test_use_acpi_hotplug_bridge,
> -                            vmstate_acpi_pcihp_use_acpi_index),
> +                            vmstate_test_migrate_acpi_index),
>          VMSTATE_END_OF_LIST()
>      },
>      .subsections = (const VMStateDescription*[]) {
> @@ -652,6 +663,8 @@ static Property piix4_pm_properties[] = {
>      DEFINE_PROP_BOOL("memory-hotplug-support", PIIX4PMState,
>                       acpi_memory_hotplug.is_enabled, true),
>      DEFINE_PROP_BOOL("smm-compat", PIIX4PMState, smm_compat, false),
> +    DEFINE_PROP_BOOL("x-not-migrate-acpi-index", PIIX4PMState,
> +                      not_migrate_acpi_index, false),
>      DEFINE_PROP_END_OF_LIST(),
>  };
>  
> diff --git a/hw/core/machine.c b/hw/core/machine.c
> index d856485cb4..1e23fdc14b 100644
> --- a/hw/core/machine.c
> +++ b/hw/core/machine.c
> @@ -37,7 +37,9 @@
>  #include "hw/virtio/virtio.h"
>  #include "hw/virtio/virtio-pci.h"
>  
> -GlobalProperty hw_compat_6_2[] = {};
> +GlobalProperty hw_compat_6_2[] = {
> +    { "PIIX4_PM", "x-not-migrate-acpi-index", "on"},
> +};
>  const size_t hw_compat_6_2_len = G_N_ELEMENTS(hw_compat_6_2);
>  
>  GlobalProperty hw_compat_6_1[] = {
> -- 
> 2.31.1
> 
-- 
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK

Re: [PATCH v2] acpi: fix acpi_index migration

[Peter Maydell]

On Wed, 6 Apr 2022 at 19:59, Dr. David Alan Gilbert <dgilbert@redhat.com> wrote:
>
> * Igor Mammedov (imammedo@redhat.com) wrote:
> > From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
> >
> > vmstate_acpi_pcihp_use_acpi_index() was expecting AcpiPciHpState
> > as state but it actually received PIIX4PMState, because
> > VMSTATE_PCI_HOTPLUG is a macro and not another struct.
> > So it ended up accessing random pointer, which resulted
> > in 'false' return value and acpi_index field wasn't ever
> > sent.
> >
> > However in 7.0 that pointer de-references to value > 0, and
> > destination QEMU starts to expect the field which isn't
> > sent in migratioon stream from older QEMU (6.2 and older).
> > As result migration fails with:
> >   qemu-system-x86_64: Missing section footer for 0000:00:01.3/piix4_pm
> >   qemu-system-x86_64: load of migration failed: Invalid argument
> >
> > In addition with QEMU-6.2, destination due to not expected
> > state, also never expects the acpi_index field in migration
> > stream.
> >
> > Q35 is not affected as it always sends/expects the field as
> > long as acpi based PCI hotplug is enabled.
> >
> > Fix issue by introducing compat knob to never send/expect
> > acpi_index in migration stream for 6.2 and older PC machine
> > types and always send it for 7.0 and newer PC machine types.
> >
> > Diagnosed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
> > Fixes: b32bd76 ("pci: introduce acpi-index property for PCI device")
> > Resolves: https://gitlab.com/qemu-project/qemu/-/issues/932
> > Signed-off-by: Igor Mammedov <imammedo@redhat.com>
>
> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

Applied to master for rc3, thanks.

-- PMM

Re: [PATCH v2] acpi: fix acpi_index migration

[Michael S. Tsirkin]

On Wed, Apr 06, 2022 at 09:29:02PM +0100, Peter Maydell wrote:
> On Wed, 6 Apr 2022 at 19:59, Dr. David Alan Gilbert <dgilbert@redhat.com> wrote:
> >
> > * Igor Mammedov (imammedo@redhat.com) wrote:
> > > From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
> > >
> > > vmstate_acpi_pcihp_use_acpi_index() was expecting AcpiPciHpState
> > > as state but it actually received PIIX4PMState, because
> > > VMSTATE_PCI_HOTPLUG is a macro and not another struct.
> > > So it ended up accessing random pointer, which resulted
> > > in 'false' return value and acpi_index field wasn't ever
> > > sent.
> > >
> > > However in 7.0 that pointer de-references to value > 0, and
> > > destination QEMU starts to expect the field which isn't
> > > sent in migratioon stream from older QEMU (6.2 and older).
> > > As result migration fails with:
> > >   qemu-system-x86_64: Missing section footer for 0000:00:01.3/piix4_pm
> > >   qemu-system-x86_64: load of migration failed: Invalid argument
> > >
> > > In addition with QEMU-6.2, destination due to not expected
> > > state, also never expects the acpi_index field in migration
> > > stream.
> > >
> > > Q35 is not affected as it always sends/expects the field as
> > > long as acpi based PCI hotplug is enabled.
> > >
> > > Fix issue by introducing compat knob to never send/expect
> > > acpi_index in migration stream for 6.2 and older PC machine
> > > types and always send it for 7.0 and newer PC machine types.
> > >
> > > Diagnosed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
> > > Fixes: b32bd76 ("pci: introduce acpi-index property for PCI device")
> > > Resolves: https://gitlab.com/qemu-project/qemu/-/issues/932
> > > Signed-off-by: Igor Mammedov <imammedo@redhat.com>
> >
> > Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
> 
> Applied to master for rc3, thanks.
> 
> -- PMM

Oh. I had this in my queue with a different commit log.
But sure. Feel free to add:
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>

-- 
MST