Skip to contentSkip to conversation

enable pnv-phb user created devices

PATCH In review Daniel Henrique Barboza 3 participants 28 messages Aug 11, 2022 – Aug 16, 2022

lore
Share & tools
lore
#0 lore raw

v4 cover enable pnv-phb user created devices

Hi,

This version contains changes based on Cedric's v3 feedback. The biggest change was made in patch 4, where a new helper was added to handle the logic where a PHB is added to a chip.

Changes from v3:
- patch 4:
- added Error **errp parameter to pnv_parent_bus_fixup() and pnv_phb_user_device_init()
- added pnv_chip_add_phb() helper
- patch 5:
- changed pnv_chip_power8_instance_init() to use an 'Object *phb' pointer
- patch 6:
- added the default PHBs under an "if (defaults_enabled())" case
- patch 7:
- pnv_phb4_get_pec() was moved to hw/ppc/pnv.c
- patch 9:
- use g_assert_not_reached() in pnv_phb4_get_pec()
- v3 link: https://lists.gnu.org/archive/html/qemu-devel/2022-08/msg01548.html

Daniel Henrique Barboza (11):
ppc/pnv: add phb-id/chip-id PnvPHB3RootBus properties ppc/pnv: add phb-id/chip-id PnvPHB4RootBus properties ppc/pnv: set root port chassis and slot using Bus properties ppc/pnv: add helpers for pnv-phb user devices ppc/pnv: turn chip8->phbs[] into a PnvPHB* array ppc/pnv: enable user created pnv-phb for powernv8 ppc/pnv: add PHB4 helpers for user created pnv-phb ppc/pnv: enable user created pnv-phb for powernv9 ppc/pnv: change pnv_phb4_get_pec() to also retrieve chip10->pecs ppc/pnv: user creatable pnv-phb for powernv10 ppc/pnv: fix QOM parenting of user creatable root ports

 hw/pci-host/pnv_phb.c          | 120 +++++++++++++++++++++++++++------
 hw/pci-host/pnv_phb3.c         |  50 ++++++++++++++
 hw/pci-host/pnv_phb4.c         |  51 ++++++++++++++
 hw/pci-host/pnv_phb4_pec.c     |   6 +-
 hw/ppc/pnv.c                   | 103 +++++++++++++++++++++++++---
 include/hw/pci-host/pnv_phb3.h |   9 ++-
 include/hw/pci-host/pnv_phb4.h |  10 +++
 include/hw/ppc/pnv.h           |   7 +-
 8 files changed, 325 insertions(+), 31 deletions(-)
Signature · 1 line
2.36.1
#1 lore raw

PATCH v4 1/11 ppc/pnv: add phb-id/chip-id PnvPHB3RootBus properties

We rely on the phb-id and chip-id, which are PHB properties, to assign chassis and slot to the root port. For default devices this is no big
deal: the root port is being created under pnv_phb_realize() and the values are being passed on via the 'index' and 'chip-id' of the pnv_phb_attach_root_port() helper.

If we want to implement user created root ports we have a problem. The user created root port will not be aware of which PHB it belongs to, unless we're willing to violate QOM best practices and access the PHB via dev->parent_bus->parent. What we can do is to access the root bus parent bus.

Since we're already assigning the root port as QOM child of the bus, and the bus is initiated using PHB properties, let's add phb-id and chip-id as properties of the bus. This will allow us trivial access to them, for both user-created and default root ports, without doing anything too shady with QOM.

---
 hw/pci-host/pnv_phb3.c         | 50 ++++++++++++++++++++++++++++++++++
 include/hw/pci-host/pnv_phb3.h |  9 +++++-
 2 files changed, 58 insertions(+), 1 deletion(-)
hw/pci-host/pnv_phb3.c +50 -0
index d4c04a281a..af8575c007 100644
@@ -1006,6 +1006,11 @@ void pnv_phb3_bus_init(DeviceState *dev, PnvPHB3 *phb)
                                      &phb->pci_mmio, &phb->pci_io,
                                      0, 4, TYPE_PNV_PHB3_ROOT_BUS);
 
+    object_property_set_int(OBJECT(pci->bus), "phb-id", phb->phb_id,
+                            &error_abort);
+    object_property_set_int(OBJECT(pci->bus), "chip-id", phb->chip_id,
+                            &error_abort);
+
     pci_setup_iommu(pci->bus, pnv_phb3_dma_iommu, phb);
 }
 
@@ -1105,10 +1110,55 @@ static const TypeInfo pnv_phb3_type_info = {
     .instance_init = pnv_phb3_instance_init,
 };
 
+static void pnv_phb3_root_bus_get_prop(Object *obj, Visitor *v,
+                                       const char *name,
+                                       void *opaque, Error **errp)
+{
+    PnvPHB3RootBus *bus = PNV_PHB3_ROOT_BUS(obj);
+    uint64_t value = 0;
+
+    if (strcmp(name, "phb-id") == 0) {
+        value = bus->phb_id;
+    } else {
+        value = bus->chip_id;
+    }
+
+    visit_type_size(v, name, &value, errp);
+}
+
+static void pnv_phb3_root_bus_set_prop(Object *obj, Visitor *v,
+                                       const char *name,
+                                       void *opaque, Error **errp)
+
+{
+    PnvPHB3RootBus *bus = PNV_PHB3_ROOT_BUS(obj);
+    uint64_t value;
+
+    if (!visit_type_size(v, name, &value, errp)) {
+        return;
+    }
+
+    if (strcmp(name, "phb-id") == 0) {
+        bus->phb_id = value;
+    } else {
+        bus->chip_id = value;
+    }
+}
+
 static void pnv_phb3_root_bus_class_init(ObjectClass *klass, void *data)
 {
     BusClass *k = BUS_CLASS(klass);
 
+    object_class_property_add(klass, "phb-id", "int",
+                              pnv_phb3_root_bus_get_prop,
+                              pnv_phb3_root_bus_set_prop,
+                              NULL, NULL);
+
+    object_class_property_add(klass, "chip-id", "int",
+                              pnv_phb3_root_bus_get_prop,
+                              pnv_phb3_root_bus_set_prop,
+                              NULL, NULL);
+
     /*
      * PHB3 has only a single root complex. Enforce the limit on the
      * parent bus
include/hw/pci-host/pnv_phb3.h +8 -1
index bff69201d9..4854f6d2f6 100644
@@ -104,9 +104,16 @@ struct PnvPBCQState {
 };
 
 /*
- * PHB3 PCIe Root port
+ * PHB3 PCIe Root Bus
  */
 #define TYPE_PNV_PHB3_ROOT_BUS "pnv-phb3-root"
+struct PnvPHB3RootBus {
+    PCIBus parent;
+
+    uint32_t chip_id;
+    uint32_t phb_id;
+};
+OBJECT_DECLARE_SIMPLE_TYPE(PnvPHB3RootBus, PNV_PHB3_ROOT_BUS)
 
 /*
  * PHB3 PCIe Host Bridge for PowerNV machines (POWER8)
Signature · 1 line
2.36.1
Reviewed-by on v4 1/11
#2 lore raw

Re: [PATCH for-7.2 v4 01/11] ppc/pnv: add phb-id/chip-id PnvPHB3RootBus properties

Quoted 18 lines · Daniel Henrique Barboza

We rely on the phb-id and chip-id, which are PHB properties, to assign chassis and slot to the root port. For default devices this is no big
deal: the root port is being created under pnv_phb_realize() and the values are being passed on via the 'index' and 'chip-id' of the pnv_phb_attach_root_port() helper.

If we want to implement user created root ports we have a problem. The user created root port will not be aware of which PHB it belongs to, unless we're willing to violate QOM best practices and access the PHB via dev->parent_bus->parent. What we can do is to access the root bus parent bus.

Since we're already assigning the root port as QOM child of the bus, and the bus is initiated using PHB properties, let's add phb-id and chip-id as properties of the bus. This will allow us trivial access to them, for both user-created and default root ports, without doing anything too shady with QOM.

Quoted 3 lines

---

Fred

Quoted 98 lines
  hw/pci-host/pnv_phb3.c         | 50 ++++++++++++++++++++++++++++++++++
  include/hw/pci-host/pnv_phb3.h |  9 +++++-
  2 files changed, 58 insertions(+), 1 deletion(-)
hw/pci-host/pnv_phb3.c +50 -0
index d4c04a281a..af8575c007 100644
@@ -1006,6 +1006,11 @@ void pnv_phb3_bus_init(DeviceState *dev, PnvPHB3 *phb)
                                       &phb->pci_mmio, &phb->pci_io,
                                       0, 4, TYPE_PNV_PHB3_ROOT_BUS);
 
+    object_property_set_int(OBJECT(pci->bus), "phb-id", phb->phb_id,
+                            &error_abort);
+    object_property_set_int(OBJECT(pci->bus), "chip-id", phb->chip_id,
+                            &error_abort);
+
      pci_setup_iommu(pci->bus, pnv_phb3_dma_iommu, phb);
  }
 
@@ -1105,10 +1110,55 @@ static const TypeInfo pnv_phb3_type_info = {
      .instance_init = pnv_phb3_instance_init,
  };
 
+static void pnv_phb3_root_bus_get_prop(Object *obj, Visitor *v,
+                                       const char *name,
+                                       void *opaque, Error **errp)
+{
+    PnvPHB3RootBus *bus = PNV_PHB3_ROOT_BUS(obj);
+    uint64_t value = 0;
+
+    if (strcmp(name, "phb-id") == 0) {
+        value = bus->phb_id;
+    } else {
+        value = bus->chip_id;
+    }
+
+    visit_type_size(v, name, &value, errp);
+}
+
+static void pnv_phb3_root_bus_set_prop(Object *obj, Visitor *v,
+                                       const char *name,
+                                       void *opaque, Error **errp)
+
+{
+    PnvPHB3RootBus *bus = PNV_PHB3_ROOT_BUS(obj);
+    uint64_t value;
+
+    if (!visit_type_size(v, name, &value, errp)) {
+        return;
+    }
+
+    if (strcmp(name, "phb-id") == 0) {
+        bus->phb_id = value;
+    } else {
+        bus->chip_id = value;
+    }
+}
+
  static void pnv_phb3_root_bus_class_init(ObjectClass *klass, void *data)
  {
      BusClass *k = BUS_CLASS(klass);
 
+    object_class_property_add(klass, "phb-id", "int",
+                              pnv_phb3_root_bus_get_prop,
+                              pnv_phb3_root_bus_set_prop,
+                              NULL, NULL);
+
+    object_class_property_add(klass, "chip-id", "int",
+                              pnv_phb3_root_bus_get_prop,
+                              pnv_phb3_root_bus_set_prop,
+                              NULL, NULL);
+
      /*
       * PHB3 has only a single root complex. Enforce the limit on the
       * parent bus
include/hw/pci-host/pnv_phb3.h +8 -1
index bff69201d9..4854f6d2f6 100644
@@ -104,9 +104,16 @@ struct PnvPBCQState {
  };
 
  /*
- * PHB3 PCIe Root port
+ * PHB3 PCIe Root Bus
   */
  #define TYPE_PNV_PHB3_ROOT_BUS "pnv-phb3-root"
+struct PnvPHB3RootBus {
+    PCIBus parent;
+
+    uint32_t chip_id;
+    uint32_t phb_id;
+};
+OBJECT_DECLARE_SIMPLE_TYPE(PnvPHB3RootBus, PNV_PHB3_ROOT_BUS)
 
  /*
   * PHB3 PCIe Host Bridge for PowerNV machines (POWER8)
#3 lore raw

PATCH v4 2/11 ppc/pnv: add phb-id/chip-id PnvPHB4RootBus properties

In reply to Daniel Henrique Barboza

The same rationale provided in the PHB3 bus case applies here.

Note: we could have merged both buses in a single object, like we did with the root ports, and spare some boilerplate. The reason we opted to preserve both buses objects is twofold:

- there's not user side advantage in doing so. Unifying the root ports presents a clear user QOL change when we enable user created devices back. The buses objects, aside from having a different QOM name, is transparent to the user;

- we leave a door opened in case we want to increase the root port limit for phb4/5 later on without having to deal with phb3 code.

---
 hw/pci-host/pnv_phb4.c         | 51 ++++++++++++++++++++++++++++++++++
 include/hw/pci-host/pnv_phb4.h | 10 +++++++
 2 files changed, 61 insertions(+)
hw/pci-host/pnv_phb4.c +51 -0
index b98c394713..824e1a73fb 100644
@@ -1551,6 +1551,12 @@ void pnv_phb4_bus_init(DeviceState *dev, PnvPHB4 *phb)
                                      pnv_phb4_set_irq, pnv_phb4_map_irq, phb,
                                      &phb->pci_mmio, &phb->pci_io,
                                      0, 4, TYPE_PNV_PHB4_ROOT_BUS);
+
+    object_property_set_int(OBJECT(pci->bus), "phb-id", phb->phb_id,
+                            &error_abort);
+    object_property_set_int(OBJECT(pci->bus), "chip-id", phb->chip_id,
+                            &error_abort);
+
     pci_setup_iommu(pci->bus, pnv_phb4_dma_iommu, phb);
     pci->bus->flags |= PCI_BUS_EXTENDED_CONFIG_SPACE;
 }
@@ -1708,10 +1714,55 @@ static const TypeInfo pnv_phb5_type_info = {
     .instance_size = sizeof(PnvPHB4),
 };
 
+static void pnv_phb4_root_bus_get_prop(Object *obj, Visitor *v,
+                                       const char *name,
+                                       void *opaque, Error **errp)
+{
+    PnvPHB4RootBus *bus = PNV_PHB4_ROOT_BUS(obj);
+    uint64_t value = 0;
+
+    if (strcmp(name, "phb-id") == 0) {
+        value = bus->phb_id;
+    } else {
+        value = bus->chip_id;
+    }
+
+    visit_type_size(v, name, &value, errp);
+}
+
+static void pnv_phb4_root_bus_set_prop(Object *obj, Visitor *v,
+                                       const char *name,
+                                       void *opaque, Error **errp)
+
+{
+    PnvPHB4RootBus *bus = PNV_PHB4_ROOT_BUS(obj);
+    uint64_t value;
+
+    if (!visit_type_size(v, name, &value, errp)) {
+        return;
+    }
+
+    if (strcmp(name, "phb-id") == 0) {
+        bus->phb_id = value;
+    } else {
+        bus->chip_id = value;
+    }
+}
+
 static void pnv_phb4_root_bus_class_init(ObjectClass *klass, void *data)
 {
     BusClass *k = BUS_CLASS(klass);
 
+    object_class_property_add(klass, "phb-id", "int",
+                              pnv_phb4_root_bus_get_prop,
+                              pnv_phb4_root_bus_set_prop,
+                              NULL, NULL);
+
+    object_class_property_add(klass, "chip-id", "int",
+                              pnv_phb4_root_bus_get_prop,
+                              pnv_phb4_root_bus_set_prop,
+                              NULL, NULL);
+
     /*
      * PHB4 has only a single root complex. Enforce the limit on the
      * parent bus
include/hw/pci-host/pnv_phb4.h +10 -0
index 20aa4819d3..50d4faa001 100644
@@ -45,7 +45,17 @@ typedef struct PnvPhb4DMASpace {
     QLIST_ENTRY(PnvPhb4DMASpace) list;
 } PnvPhb4DMASpace;
 
+/*
+ * PHB4 PCIe Root Bus
+ */
 #define TYPE_PNV_PHB4_ROOT_BUS "pnv-phb4-root"
+struct PnvPHB4RootBus {
+    PCIBus parent;
+
+    uint32_t chip_id;
+    uint32_t phb_id;
+};
+OBJECT_DECLARE_SIMPLE_TYPE(PnvPHB4RootBus, PNV_PHB4_ROOT_BUS)
 
 /*
  * PHB4 PCIe Host Bridge for PowerNV machines (POWER9)
Signature · 1 line
2.36.1
Reviewed-by on v4 2/11
#4 lore raw

Re: [PATCH for-7.2 v4 02/11] ppc/pnv: add phb-id/chip-id PnvPHB4RootBus properties

Quoted 11 lines · Daniel Henrique Barboza

The same rationale provided in the PHB3 bus case applies here.

Note: we could have merged both buses in a single object, like we did with the root ports, and spare some boilerplate. The reason we opted to preserve both buses objects is twofold:

- there's not user side advantage in doing so. Unifying the root ports presents a clear user QOL change when we enable user created devices back. The buses objects, aside from having a different QOM name, is transparent to the user;

Quoted 6 lines

- we leave a door opened in case we want to increase the root port limit for phb4/5 later on without having to deal with phb3 code.

---

Fred

Quoted 99 lines
  hw/pci-host/pnv_phb4.c         | 51 ++++++++++++++++++++++++++++++++++
  include/hw/pci-host/pnv_phb4.h | 10 +++++++
  2 files changed, 61 insertions(+)
hw/pci-host/pnv_phb4.c +51 -0
index b98c394713..824e1a73fb 100644
@@ -1551,6 +1551,12 @@ void pnv_phb4_bus_init(DeviceState *dev, PnvPHB4 *phb)
                                       pnv_phb4_set_irq, pnv_phb4_map_irq, phb,
                                       &phb->pci_mmio, &phb->pci_io,
                                       0, 4, TYPE_PNV_PHB4_ROOT_BUS);
+
+    object_property_set_int(OBJECT(pci->bus), "phb-id", phb->phb_id,
+                            &error_abort);
+    object_property_set_int(OBJECT(pci->bus), "chip-id", phb->chip_id,
+                            &error_abort);
+
      pci_setup_iommu(pci->bus, pnv_phb4_dma_iommu, phb);
      pci->bus->flags |= PCI_BUS_EXTENDED_CONFIG_SPACE;
  }
@@ -1708,10 +1714,55 @@ static const TypeInfo pnv_phb5_type_info = {
      .instance_size = sizeof(PnvPHB4),
  };
 
+static void pnv_phb4_root_bus_get_prop(Object *obj, Visitor *v,
+                                       const char *name,
+                                       void *opaque, Error **errp)
+{
+    PnvPHB4RootBus *bus = PNV_PHB4_ROOT_BUS(obj);
+    uint64_t value = 0;
+
+    if (strcmp(name, "phb-id") == 0) {
+        value = bus->phb_id;
+    } else {
+        value = bus->chip_id;
+    }
+
+    visit_type_size(v, name, &value, errp);
+}
+
+static void pnv_phb4_root_bus_set_prop(Object *obj, Visitor *v,
+                                       const char *name,
+                                       void *opaque, Error **errp)
+
+{
+    PnvPHB4RootBus *bus = PNV_PHB4_ROOT_BUS(obj);
+    uint64_t value;
+
+    if (!visit_type_size(v, name, &value, errp)) {
+        return;
+    }
+
+    if (strcmp(name, "phb-id") == 0) {
+        bus->phb_id = value;
+    } else {
+        bus->chip_id = value;
+    }
+}
+
  static void pnv_phb4_root_bus_class_init(ObjectClass *klass, void *data)
  {
      BusClass *k = BUS_CLASS(klass);
 
+    object_class_property_add(klass, "phb-id", "int",
+                              pnv_phb4_root_bus_get_prop,
+                              pnv_phb4_root_bus_set_prop,
+                              NULL, NULL);
+
+    object_class_property_add(klass, "chip-id", "int",
+                              pnv_phb4_root_bus_get_prop,
+                              pnv_phb4_root_bus_set_prop,
+                              NULL, NULL);
+
      /*
       * PHB4 has only a single root complex. Enforce the limit on the
       * parent bus
include/hw/pci-host/pnv_phb4.h +10 -0
index 20aa4819d3..50d4faa001 100644
@@ -45,7 +45,17 @@ typedef struct PnvPhb4DMASpace {
      QLIST_ENTRY(PnvPhb4DMASpace) list;
  } PnvPhb4DMASpace;
 
+/*
+ * PHB4 PCIe Root Bus
+ */
  #define TYPE_PNV_PHB4_ROOT_BUS "pnv-phb4-root"
+struct PnvPHB4RootBus {
+    PCIBus parent;
+
+    uint32_t chip_id;
+    uint32_t phb_id;
+};
+OBJECT_DECLARE_SIMPLE_TYPE(PnvPHB4RootBus, PNV_PHB4_ROOT_BUS)
 
  /*
   * PHB4 PCIe Host Bridge for PowerNV machines (POWER9)
#5 lore raw

PATCH v4 3/11 ppc/pnv: set root port chassis and slot using Bus properties

In reply to Daniel Henrique Barboza

For default root ports we have a way of accessing chassis and slot, before root_port_realize(), via pnv_phb_attach_root_port(). For the future user created root ports this won't be the case: we can't use this helper because we don't have access to the PHB phb-id/chip-id values.

In earlier patches we've added phb-id and chip-id to pnv-phb-root-bus objects. We're now able to use the bus to retrieve them. The bus is reachable for both user created and default devices, so we're changing all the code paths. This also allow us to validate these changes with the existing default devices.

---
 hw/pci-host/pnv_phb.c | 25 ++++++++++++++++---------
 1 file changed, 16 insertions(+), 9 deletions(-)
hw/pci-host/pnv_phb.c +16 -9
index c47ed92462..826c0c144e 100644
@@ -25,21 +25,19 @@
  * QOM id. 'chip_id' is going to be used as PCIE chassis for the
  * root port.
  */
-static void pnv_phb_attach_root_port(PCIHostState *pci, int index, int chip_id)
+static void pnv_phb_attach_root_port(PCIHostState *pci)
 {
     PCIDevice *root = pci_new(PCI_DEVFN(0, 0), TYPE_PNV_PHB_ROOT_PORT);
-    g_autofree char *default_id = g_strdup_printf("%s[%d]",
-                                                  TYPE_PNV_PHB_ROOT_PORT,
-                                                  index);
     const char *dev_id = DEVICE(root)->id;
+    g_autofree char *default_id = NULL;
+    int index;
+
+    index = object_property_get_int(OBJECT(pci->bus), "phb-id", &error_fatal);
+    default_id = g_strdup_printf("%s[%d]", TYPE_PNV_PHB_ROOT_PORT, index);
 
     object_property_add_child(OBJECT(pci->bus), dev_id ? dev_id : default_id,
                               OBJECT(root));
 
-    /* Set unique chassis/slot values for the root port */
-    qdev_prop_set_uint8(DEVICE(root), "chassis", chip_id);
-    qdev_prop_set_uint16(DEVICE(root), "slot", index);
-
     pci_realize_and_unref(root, pci->bus, &error_fatal);
 }
 
@@ -93,7 +91,7 @@ static void pnv_phb_realize(DeviceState *dev, Error **errp)
         pnv_phb4_bus_init(dev, PNV_PHB4(phb->backend));
     }
 
-    pnv_phb_attach_root_port(pci, phb->phb_id, phb->chip_id);
+    pnv_phb_attach_root_port(pci);
 }
 
 static const char *pnv_phb_root_bus_path(PCIHostState *host_bridge,
@@ -162,9 +160,18 @@ static void pnv_phb_root_port_realize(DeviceState *dev, Error **errp)
 {
     PCIERootPortClass *rpc = PCIE_ROOT_PORT_GET_CLASS(dev);
     PnvPHBRootPort *phb_rp = PNV_PHB_ROOT_PORT(dev);
+    PCIBus *bus = PCI_BUS(qdev_get_parent_bus(dev));
     PCIDevice *pci = PCI_DEVICE(dev);
     uint16_t device_id = 0;
     Error *local_err = NULL;
+    int chip_id, index;
+
+    chip_id = object_property_get_int(OBJECT(bus), "chip-id", &error_fatal);
+    index = object_property_get_int(OBJECT(bus), "phb-id", &error_fatal);
+
+    /* Set unique chassis/slot values for the root port */
+    qdev_prop_set_uint8(dev, "chassis", chip_id);
+    qdev_prop_set_uint16(dev, "slot", index);
 
     rpc->parent_realize(dev, &local_err);
     if (local_err) {
Signature · 1 line
2.36.1
Reviewed-by on v4 3/11
#6 lore raw

Re: [PATCH for-7.2 v4 03/11] ppc/pnv: set root port chassis and slot using Bus properties

Quoted 12 lines · Daniel Henrique Barboza

For default root ports we have a way of accessing chassis and slot, before root_port_realize(), via pnv_phb_attach_root_port(). For the future user created root ports this won't be the case: we can't use this helper because we don't have access to the PHB phb-id/chip-id values.

In earlier patches we've added phb-id and chip-id to pnv-phb-root-bus objects. We're now able to use the bus to retrieve them. The bus is reachable for both user created and default devices, so we're changing all the code paths. This also allow us to validate these changes with the existing default devices.

Quoted 3 lines

---

Fred

Quoted 63 lines
  hw/pci-host/pnv_phb.c | 25 ++++++++++++++++---------
  1 file changed, 16 insertions(+), 9 deletions(-)
hw/pci-host/pnv_phb.c +16 -9
index c47ed92462..826c0c144e 100644
@@ -25,21 +25,19 @@
   * QOM id. 'chip_id' is going to be used as PCIE chassis for the
   * root port.
   */
-static void pnv_phb_attach_root_port(PCIHostState *pci, int index, int chip_id)
+static void pnv_phb_attach_root_port(PCIHostState *pci)
  {
      PCIDevice *root = pci_new(PCI_DEVFN(0, 0), TYPE_PNV_PHB_ROOT_PORT);
-    g_autofree char *default_id = g_strdup_printf("%s[%d]",
-                                                  TYPE_PNV_PHB_ROOT_PORT,
-                                                  index);
      const char *dev_id = DEVICE(root)->id;
+    g_autofree char *default_id = NULL;
+    int index;
+
+    index = object_property_get_int(OBJECT(pci->bus), "phb-id", &error_fatal);
+    default_id = g_strdup_printf("%s[%d]", TYPE_PNV_PHB_ROOT_PORT, index);
 
      object_property_add_child(OBJECT(pci->bus), dev_id ? dev_id : default_id,
                                OBJECT(root));
 
-    /* Set unique chassis/slot values for the root port */
-    qdev_prop_set_uint8(DEVICE(root), "chassis", chip_id);
-    qdev_prop_set_uint16(DEVICE(root), "slot", index);
-
      pci_realize_and_unref(root, pci->bus, &error_fatal);
  }
 
@@ -93,7 +91,7 @@ static void pnv_phb_realize(DeviceState *dev, Error **errp)
          pnv_phb4_bus_init(dev, PNV_PHB4(phb->backend));
      }
 
-    pnv_phb_attach_root_port(pci, phb->phb_id, phb->chip_id);
+    pnv_phb_attach_root_port(pci);
  }
 
  static const char *pnv_phb_root_bus_path(PCIHostState *host_bridge,
@@ -162,9 +160,18 @@ static void pnv_phb_root_port_realize(DeviceState *dev, Error **errp)
  {
      PCIERootPortClass *rpc = PCIE_ROOT_PORT_GET_CLASS(dev);
      PnvPHBRootPort *phb_rp = PNV_PHB_ROOT_PORT(dev);
+    PCIBus *bus = PCI_BUS(qdev_get_parent_bus(dev));
      PCIDevice *pci = PCI_DEVICE(dev);
      uint16_t device_id = 0;
      Error *local_err = NULL;
+    int chip_id, index;
+
+    chip_id = object_property_get_int(OBJECT(bus), "chip-id", &error_fatal);
+    index = object_property_get_int(OBJECT(bus), "phb-id", &error_fatal);
+
+    /* Set unique chassis/slot values for the root port */
+    qdev_prop_set_uint8(dev, "chassis", chip_id);
+    qdev_prop_set_uint16(dev, "slot", index);
 
      rpc->parent_realize(dev, &local_err);
      if (local_err) {
#7 lore raw

PATCH v4 4/11 ppc/pnv: add helpers for pnv-phb user devices

In reply to Daniel Henrique Barboza

pnv_parent_qom_fixup() and pnv_parent_bus_fixup() are versions of the helpers that were reverted by commit 9c10d86fee "ppc/pnv: Remove user-created PHB{3,4,5} devices". They are needed to amend the QOM and bus hierarchies of user created pnv-phbs, matching them with default pnv-phbs.

A new helper pnv_phb_user_device_init() is created to handle user-created devices setup. We're going to call it inside pnv_phb_realize() in case we're realizing an user created device. This will centralize all user device realated in a single spot, leaving the realize functions of the phb3/phb4 backends untouched.

Another helper called pnv_chip_add_phb() was added to handle the particularities of each chip version when adding a new PHB.

---
 hw/pci-host/pnv_phb.c | 75 +++++++++++++++++++++++++++++++++++++++++++
 hw/ppc/pnv.c          | 20 ++++++++++++
 include/hw/ppc/pnv.h  |  1 +
 3 files changed, 96 insertions(+)
hw/pci-host/pnv_phb.c +75 -0
index 826c0c144e..5dc44f45d1 100644
@@ -18,6 +18,38 @@
 #include "hw/qdev-properties.h"
 #include "qom/object.h"
 
+
+/*
+ * Set the QOM parent of an object child. If the device state
+ * associated with the child has an id, use it as QOM id. Otherwise
+ * use object_typename[index] as QOM id.
+ */
+static void pnv_parent_qom_fixup(Object *parent, Object *child, int index)
+{
+    g_autofree char *default_id =
+        g_strdup_printf("%s[%d]", object_get_typename(child), index);
+    const char *dev_id = DEVICE(child)->id;
+
+    if (child->parent == parent) {
+        return;
+    }
+
+    object_ref(child);
+    object_unparent(child);
+    object_property_add_child(parent, dev_id ? dev_id : default_id, child);
+    object_unref(child);
+}
+
+static void pnv_parent_bus_fixup(DeviceState *parent, DeviceState *child,
+                                 Error **errp)
+{
+    BusState *parent_bus = qdev_get_parent_bus(parent);
+
+    if (!qdev_set_parent_bus(child, parent_bus, errp)) {
+        return;
+    }
+}
+
 /*
  * Attach a root port device.
  *
@@ -41,6 +73,39 @@ static void pnv_phb_attach_root_port(PCIHostState *pci)
     pci_realize_and_unref(root, pci->bus, &error_fatal);
 }
 
+/*
+ * User created devices won't have the initial setup that default
+ * devices have. This setup consists of assigning a parent device
+ * (chip for PHB3, PEC for PHB4/5) that will be the QOM/bus parent
+ * of the PHB.
+ */
+static bool pnv_phb_user_device_init(PnvPHB *phb, Error **errp)
+{
+    PnvMachineState *pnv = PNV_MACHINE(qdev_get_machine());
+    PnvChip *chip = pnv_get_chip(pnv, phb->chip_id);
+    Object *parent = NULL;
+
+    if (!chip) {
+        error_setg(errp, "invalid chip id: %d", phb->chip_id);
+        return false;
+    }
+
+    parent = pnv_chip_add_phb(chip, phb, errp);
+    if (!parent) {
+        return false;
+    }
+
+    /*
+     * Reparent user created devices to the chip to build
+     * correctly the device tree. pnv_xscom_dt() needs every
+     * PHB to be a child of the chip to build the DT correctly.
+     */
+    pnv_parent_qom_fixup(parent, OBJECT(phb), phb->phb_id);
+    pnv_parent_bus_fixup(DEVICE(chip), DEVICE(phb), errp);
+
+    return true;
+}
+
 static void pnv_phb_realize(DeviceState *dev, Error **errp)
 {
     PnvPHB *phb = PNV_PHB(dev);
@@ -74,6 +139,16 @@ static void pnv_phb_realize(DeviceState *dev, Error **errp)
     object_property_set_uint(phb->backend, "chip-id", phb->chip_id, errp);
     object_property_set_link(phb->backend, "phb-base", OBJECT(phb), errp);
 
+    /*
+     * Handle user created devices. User devices will not have a
+     * pointer to a chip (PHB3) and a PEC (PHB4/5).
+     */
+    if (!phb->chip && !phb->pec) {
+        if (!pnv_phb_user_device_init(phb, errp)) {
+            return;
+        }
+    }
+
     if (phb->version == 3) {
         object_property_set_link(phb->backend, "chip",
                                  OBJECT(phb->chip), errp);
hw/ppc/pnv.c +20 -0
index f9e5a3d248..2deaac17f7 100644
@@ -281,6 +281,26 @@ static void pnv_dt_icp(PnvChip *chip, void *fdt, uint32_t pir,
     g_free(reg);
 }
 
+/*
+ * Adds a PnvPHB to the chip. Returns the parent obj of the
+ * PHB which varies with each version (phb version 3 is parented
+ * by the chip, version 4 and 4 are parented by the PEC
+ * device).
+ *
+ * TODO: for version 3 we're still parenting the PHB with the
+ * chip. We should parent with a (so far not implemented)
+ * PHB3 PEC device.
+ */
+Object *pnv_chip_add_phb(PnvChip *chip, PnvPHB *phb, Error **errp)
+{
+    if (phb->version == 3) {
+        return OBJECT(chip);
+    } else {
+        /* phb4 support will be added later */
+        return NULL;
+    }
+}
+
 static void pnv_chip_power8_dt_populate(PnvChip *chip, void *fdt)
 {
     static const char compat[] = "ibm,power8-xscom\0ibm,xscom";
include/hw/ppc/pnv.h +1 -0
index 033d907287..781d0acffa 100644
@@ -231,6 +231,7 @@ struct PnvMachineState {
 };
 
 PnvChip *pnv_get_chip(PnvMachineState *pnv, uint32_t chip_id);
+Object *pnv_chip_add_phb(PnvChip *chip, PnvPHB *phb, Error **errp);
 
 #define PNV_FDT_ADDR          0x01000000
 #define PNV_TIMEBASE_FREQ     512000000ULL
Signature · 1 line
2.36.1
Reviewed-by on v4 4/11
#8 lore raw

Re: [PATCH for-7.2 v4 04/11] ppc/pnv: add helpers for pnv-phb user devices

Quoted 12 lines · Daniel Henrique Barboza

pnv_parent_qom_fixup() and pnv_parent_bus_fixup() are versions of the helpers that were reverted by commit 9c10d86fee "ppc/pnv: Remove user-created PHB{3,4,5} devices". They are needed to amend the QOM and bus hierarchies of user created pnv-phbs, matching them with default pnv-phbs.

A new helper pnv_phb_user_device_init() is created to handle user-created devices setup. We're going to call it inside pnv_phb_realize() in case we're realizing an user created device. This will centralize all user device realated in a single spot, leaving the realize functions of the phb3/phb4 backends untouched.

Quoted 4 lines

Another helper called pnv_chip_add_phb() was added to handle the particularities of each chip version when adding a new PHB.

Thanks,

C.

Quoted 149 lines
---
  hw/pci-host/pnv_phb.c | 75 +++++++++++++++++++++++++++++++++++++++++++
  hw/ppc/pnv.c          | 20 ++++++++++++
  include/hw/ppc/pnv.h  |  1 +
  3 files changed, 96 insertions(+)
hw/pci-host/pnv_phb.c +75 -0
index 826c0c144e..5dc44f45d1 100644
@@ -18,6 +18,38 @@
  #include "hw/qdev-properties.h"
  #include "qom/object.h"
 
+
+/*
+ * Set the QOM parent of an object child. If the device state
+ * associated with the child has an id, use it as QOM id. Otherwise
+ * use object_typename[index] as QOM id.
+ */
+static void pnv_parent_qom_fixup(Object *parent, Object *child, int index)
+{
+    g_autofree char *default_id =
+        g_strdup_printf("%s[%d]", object_get_typename(child), index);
+    const char *dev_id = DEVICE(child)->id;
+
+    if (child->parent == parent) {
+        return;
+    }
+
+    object_ref(child);
+    object_unparent(child);
+    object_property_add_child(parent, dev_id ? dev_id : default_id, child);
+    object_unref(child);
+}
+
+static void pnv_parent_bus_fixup(DeviceState *parent, DeviceState *child,
+                                 Error **errp)
+{
+    BusState *parent_bus = qdev_get_parent_bus(parent);
+
+    if (!qdev_set_parent_bus(child, parent_bus, errp)) {
+        return;
+    }
+}
+
  /*
   * Attach a root port device.
   *
@@ -41,6 +73,39 @@ static void pnv_phb_attach_root_port(PCIHostState *pci)
      pci_realize_and_unref(root, pci->bus, &error_fatal);
  }
 
+/*
+ * User created devices won't have the initial setup that default
+ * devices have. This setup consists of assigning a parent device
+ * (chip for PHB3, PEC for PHB4/5) that will be the QOM/bus parent
+ * of the PHB.
+ */
+static bool pnv_phb_user_device_init(PnvPHB *phb, Error **errp)
+{
+    PnvMachineState *pnv = PNV_MACHINE(qdev_get_machine());
+    PnvChip *chip = pnv_get_chip(pnv, phb->chip_id);
+    Object *parent = NULL;
+
+    if (!chip) {
+        error_setg(errp, "invalid chip id: %d", phb->chip_id);
+        return false;
+    }
+
+    parent = pnv_chip_add_phb(chip, phb, errp);
+    if (!parent) {
+        return false;
+    }
+
+    /*
+     * Reparent user created devices to the chip to build
+     * correctly the device tree. pnv_xscom_dt() needs every
+     * PHB to be a child of the chip to build the DT correctly.
+     */
+    pnv_parent_qom_fixup(parent, OBJECT(phb), phb->phb_id);
+    pnv_parent_bus_fixup(DEVICE(chip), DEVICE(phb), errp);
+
+    return true;
+}
+
  static void pnv_phb_realize(DeviceState *dev, Error **errp)
  {
      PnvPHB *phb = PNV_PHB(dev);
@@ -74,6 +139,16 @@ static void pnv_phb_realize(DeviceState *dev, Error **errp)
      object_property_set_uint(phb->backend, "chip-id", phb->chip_id, errp);
      object_property_set_link(phb->backend, "phb-base", OBJECT(phb), errp);
 
+    /*
+     * Handle user created devices. User devices will not have a
+     * pointer to a chip (PHB3) and a PEC (PHB4/5).
+     */
+    if (!phb->chip && !phb->pec) {
+        if (!pnv_phb_user_device_init(phb, errp)) {
+            return;
+        }
+    }
+
      if (phb->version == 3) {
          object_property_set_link(phb->backend, "chip",
                                   OBJECT(phb->chip), errp);
hw/ppc/pnv.c +20 -0
index f9e5a3d248..2deaac17f7 100644
@@ -281,6 +281,26 @@ static void pnv_dt_icp(PnvChip *chip, void *fdt, uint32_t pir,
      g_free(reg);
  }
 
+/*
+ * Adds a PnvPHB to the chip. Returns the parent obj of the
+ * PHB which varies with each version (phb version 3 is parented
+ * by the chip, version 4 and 4 are parented by the PEC
+ * device).
+ *
+ * TODO: for version 3 we're still parenting the PHB with the
+ * chip. We should parent with a (so far not implemented)
+ * PHB3 PEC device.
+ */
+Object *pnv_chip_add_phb(PnvChip *chip, PnvPHB *phb, Error **errp)
+{
+    if (phb->version == 3) {
+        return OBJECT(chip);
+    } else {
+        /* phb4 support will be added later */
+        return NULL;
+    }
+}
+
  static void pnv_chip_power8_dt_populate(PnvChip *chip, void *fdt)
  {
      static const char compat[] = "ibm,power8-xscom\0ibm,xscom";
include/hw/ppc/pnv.h +1 -0
index 033d907287..781d0acffa 100644
@@ -231,6 +231,7 @@ struct PnvMachineState {
  };
 
  PnvChip *pnv_get_chip(PnvMachineState *pnv, uint32_t chip_id);
+Object *pnv_chip_add_phb(PnvChip *chip, PnvPHB *phb, Error **errp);
 
  #define PNV_FDT_ADDR          0x01000000
  #define PNV_TIMEBASE_FREQ     512000000ULL
Reviewed-by on v4 4/11
#9 lore raw

Re: [PATCH for-7.2 v4 04/11] ppc/pnv: add helpers for pnv-phb user devices

In reply to Daniel Henrique Barboza
Quoted 12 lines · Daniel Henrique Barboza

pnv_parent_qom_fixup() and pnv_parent_bus_fixup() are versions of the helpers that were reverted by commit 9c10d86fee "ppc/pnv: Remove user-created PHB{3,4,5} devices". They are needed to amend the QOM and bus hierarchies of user created pnv-phbs, matching them with default pnv-phbs.

A new helper pnv_phb_user_device_init() is created to handle user-created devices setup. We're going to call it inside pnv_phb_realize() in case we're realizing an user created device. This will centralize all user device realated in a single spot, leaving the realize functions of the phb3/phb4 backends untouched.

Quoted 5 lines

Another helper called pnv_chip_add_phb() was added to handle the particularities of each chip version when adding a new PHB.

---

Just a minor typo in a comment below.

hw/ppc/pnv.c static void pnv_dt_icp(PnvChip *chip, void *fdt, uint32_t pir, · line 287

Quoted 12 lines
hw/ppc/pnv.c +4 -0
index f9e5a3d248..2deaac17f7 100644
@@ -281,6 +281,26 @@ static void pnv_dt_icp(PnvChip *chip, void *fdt, uint32_t pir,
      g_free(reg);
  }
 
+/*
+ * Adds a PnvPHB to the chip. Returns the parent obj of the
+ * PHB which varies with each version (phb version 3 is parented
+ * by the chip, version 4 and 4 are parented by the PEC

typo-----------------------------^

Fred

Quoted 31 lines
hw/ppc/pnv.c +16 -0
+ * device).
+ *
+ * TODO: for version 3 we're still parenting the PHB with the
+ * chip. We should parent with a (so far not implemented)
+ * PHB3 PEC device.
+ */
+Object *pnv_chip_add_phb(PnvChip *chip, PnvPHB *phb, Error **errp)
+{
+    if (phb->version == 3) {
+        return OBJECT(chip);
+    } else {
+        /* phb4 support will be added later */
+        return NULL;
+    }
+}
+
  static void pnv_chip_power8_dt_populate(PnvChip *chip, void *fdt)
  {
      static const char compat[] = "ibm,power8-xscom\0ibm,xscom";
include/hw/ppc/pnv.h +1 -0
index 033d907287..781d0acffa 100644
@@ -231,6 +231,7 @@ struct PnvMachineState {
  };
 
  PnvChip *pnv_get_chip(PnvMachineState *pnv, uint32_t chip_id);
+Object *pnv_chip_add_phb(PnvChip *chip, PnvPHB *phb, Error **errp);
 
  #define PNV_FDT_ADDR          0x01000000
  #define PNV_TIMEBASE_FREQ     512000000ULL
#10 lore raw

PATCH v4 5/11 ppc/pnv: turn chip8->phbs[] into a PnvPHB* array

In reply to Daniel Henrique Barboza

When enabling user created PHBs (a change reverted by commit 9c10d86fee) we were handling PHBs created by default versus by the user in different manners. The only difference between these PHBs is that one will have a valid phb3->chip that is assigned during pnv_chip_power8_realize(), while the user created needs to search which chip it belongs to.

Aside from that there shouldn't be any difference. Making the default PHBs behave in line with the user created ones will make it easier to re-introduce them later on. It will also make the code easier to follow since we are dealing with them in equal manner.

The first step is to turn chip8->phbs[] into a PnvPHB3 pointer array. This will allow us to assign user created PHBs into it later on. The way we initilize the default case is now more in line with that would happen with the user created case: the object is created, parented by the chip because pnv_xscom_dt() relies on it, and then assigned to the array.

---
 hw/ppc/pnv.c         | 27 ++++++++++++++++++++++-----
 include/hw/ppc/pnv.h |  6 +++++-
 2 files changed, 27 insertions(+), 6 deletions(-)
hw/ppc/pnv.c +22 -5
index 2deaac17f7..e53e9e297d 100644
@@ -294,6 +294,13 @@ static void pnv_dt_icp(PnvChip *chip, void *fdt, uint32_t pir,
 Object *pnv_chip_add_phb(PnvChip *chip, PnvPHB *phb, Error **errp)
 {
     if (phb->version == 3) {
+        Pnv8Chip *chip8 = PNV8_CHIP(chip);
+
+        phb->chip = chip;
+
+        chip8->phbs[chip8->num_phbs] = phb;
+        chip8->num_phbs++;
+
         return OBJECT(chip);
     } else {
         /* phb4 support will be added later */
@@ -681,7 +688,7 @@ static void pnv_chip_power8_pic_print_info(PnvChip *chip, Monitor *mon)
     ics_pic_print_info(&chip8->psi.ics, mon);
 
     for (i = 0; i < chip8->num_phbs; i++) {
-        PnvPHB *phb = &chip8->phbs[i];
+        PnvPHB *phb = chip8->phbs[i];
         PnvPHB3 *phb3 = PNV_PHB3(phb->backend);
 
         pnv_phb3_msi_pic_print_info(&phb3->msis, mon);
@@ -1174,7 +1181,17 @@ static void pnv_chip_power8_instance_init(Object *obj)
     chip8->num_phbs = pcc->num_phbs;
 
     for (i = 0; i < chip8->num_phbs; i++) {
-        object_initialize_child(obj, "phb[*]", &chip8->phbs[i], TYPE_PNV_PHB);
+        Object *phb = object_new(TYPE_PNV_PHB);
+
+        /*
+         * We need the chip to parent the PHB to allow the DT
+         * to build correctly (via pnv_xscom_dt()).
+         *
+         * TODO: the PHB should be parented by a PEC device that, at
+         * this moment, is not modelled powernv8/phb3.
+         */
+        object_property_add_child(obj, "phb[*]", phb);
+        chip8->phbs[i] = PNV_PHB(phb);
     }
 
 }
@@ -1290,7 +1307,7 @@ static void pnv_chip_power8_realize(DeviceState *dev, Error **errp)
 
     /* PHB controllers */
     for (i = 0; i < chip8->num_phbs; i++) {
-        PnvPHB *phb = &chip8->phbs[i];
+        PnvPHB *phb = chip8->phbs[i];
 
         object_property_set_int(OBJECT(phb), "index", i, &error_fatal);
         object_property_set_int(OBJECT(phb), "chip-id", chip->chip_id,
@@ -1958,7 +1975,7 @@ static ICSState *pnv_ics_get(XICSFabric *xi, int irq)
         }
 
         for (j = 0; j < chip8->num_phbs; j++) {
-            PnvPHB *phb = &chip8->phbs[j];
+            PnvPHB *phb = chip8->phbs[j];
             PnvPHB3 *phb3 = PNV_PHB3(phb->backend);
 
             if (ics_valid_irq(&phb3->lsis, irq)) {
@@ -1997,7 +2014,7 @@ static void pnv_ics_resend(XICSFabric *xi)
         ics_resend(&chip8->psi.ics);
 
         for (j = 0; j < chip8->num_phbs; j++) {
-            PnvPHB *phb = &chip8->phbs[j];
+            PnvPHB *phb = chip8->phbs[j];
             PnvPHB3 *phb3 = PNV_PHB3(phb->backend);
 
             ics_resend(&phb3->lsis);
include/hw/ppc/pnv.h +5 -1
index 781d0acffa..49433281d7 100644
@@ -81,7 +81,11 @@ struct Pnv8Chip {
     PnvHomer     homer;
 
 #define PNV8_CHIP_PHB3_MAX 4
-    PnvPHB       phbs[PNV8_CHIP_PHB3_MAX];
+    /*
+     * The array is used to allow quick access to the phbs by
+     * pnv_ics_get_child() and pnv_ics_resend_child().
+     */
+    PnvPHB       *phbs[PNV8_CHIP_PHB3_MAX];
     uint32_t     num_phbs;
 
     XICSFabric    *xics;
Signature · 1 line
2.36.1
Reviewed-by on v4 5/11
#11 lore raw

Re: [PATCH for-7.2 v4 05/11] ppc/pnv: turn chip8->phbs[] into a PnvPHB* array

Quoted 11 lines · Daniel Henrique Barboza

When enabling user created PHBs (a change reverted by commit 9c10d86fee) we were handling PHBs created by default versus by the user in different manners. The only difference between these PHBs is that one will have a valid phb3->chip that is assigned during pnv_chip_power8_realize(), while the user created needs to search which chip it belongs to.

Aside from that there shouldn't be any difference. Making the default PHBs behave in line with the user created ones will make it easier to re-introduce them later on. It will also make the code easier to follow since we are dealing with them in equal manner.

Quoted 7 lines

The first step is to turn chip8->phbs[] into a PnvPHB3 pointer array. This will allow us to assign user created PHBs into it later on. The way we initilize the default case is now more in line with that would happen with the user created case: the object is created, parented by the chip because pnv_xscom_dt() relies on it, and then assigned to the array.

Thanks,

C.

Quoted 95 lines
---
  hw/ppc/pnv.c         | 27 ++++++++++++++++++++++-----
  include/hw/ppc/pnv.h |  6 +++++-
  2 files changed, 27 insertions(+), 6 deletions(-)
hw/ppc/pnv.c +22 -5
index 2deaac17f7..e53e9e297d 100644
@@ -294,6 +294,13 @@ static void pnv_dt_icp(PnvChip *chip, void *fdt, uint32_t pir,
  Object *pnv_chip_add_phb(PnvChip *chip, PnvPHB *phb, Error **errp)
  {
      if (phb->version == 3) {
+        Pnv8Chip *chip8 = PNV8_CHIP(chip);
+
+        phb->chip = chip;
+
+        chip8->phbs[chip8->num_phbs] = phb;
+        chip8->num_phbs++;
+
          return OBJECT(chip);
      } else {
          /* phb4 support will be added later */
@@ -681,7 +688,7 @@ static void pnv_chip_power8_pic_print_info(PnvChip *chip, Monitor *mon)
      ics_pic_print_info(&chip8->psi.ics, mon);
 
      for (i = 0; i < chip8->num_phbs; i++) {
-        PnvPHB *phb = &chip8->phbs[i];
+        PnvPHB *phb = chip8->phbs[i];
          PnvPHB3 *phb3 = PNV_PHB3(phb->backend);
 
          pnv_phb3_msi_pic_print_info(&phb3->msis, mon);
@@ -1174,7 +1181,17 @@ static void pnv_chip_power8_instance_init(Object *obj)
      chip8->num_phbs = pcc->num_phbs;
 
      for (i = 0; i < chip8->num_phbs; i++) {
-        object_initialize_child(obj, "phb[*]", &chip8->phbs[i], TYPE_PNV_PHB);
+        Object *phb = object_new(TYPE_PNV_PHB);
+
+        /*
+         * We need the chip to parent the PHB to allow the DT
+         * to build correctly (via pnv_xscom_dt()).
+         *
+         * TODO: the PHB should be parented by a PEC device that, at
+         * this moment, is not modelled powernv8/phb3.
+         */
+        object_property_add_child(obj, "phb[*]", phb);
+        chip8->phbs[i] = PNV_PHB(phb);
      }
 
  }
@@ -1290,7 +1307,7 @@ static void pnv_chip_power8_realize(DeviceState *dev, Error **errp)
 
      /* PHB controllers */
      for (i = 0; i < chip8->num_phbs; i++) {
-        PnvPHB *phb = &chip8->phbs[i];
+        PnvPHB *phb = chip8->phbs[i];
 
          object_property_set_int(OBJECT(phb), "index", i, &error_fatal);
          object_property_set_int(OBJECT(phb), "chip-id", chip->chip_id,
@@ -1958,7 +1975,7 @@ static ICSState *pnv_ics_get(XICSFabric *xi, int irq)
          }
 
          for (j = 0; j < chip8->num_phbs; j++) {
-            PnvPHB *phb = &chip8->phbs[j];
+            PnvPHB *phb = chip8->phbs[j];
              PnvPHB3 *phb3 = PNV_PHB3(phb->backend);
 
              if (ics_valid_irq(&phb3->lsis, irq)) {
@@ -1997,7 +2014,7 @@ static void pnv_ics_resend(XICSFabric *xi)
          ics_resend(&chip8->psi.ics);
 
          for (j = 0; j < chip8->num_phbs; j++) {
-            PnvPHB *phb = &chip8->phbs[j];
+            PnvPHB *phb = chip8->phbs[j];
              PnvPHB3 *phb3 = PNV_PHB3(phb->backend);
 
              ics_resend(&phb3->lsis);
include/hw/ppc/pnv.h +5 -1
index 781d0acffa..49433281d7 100644
@@ -81,7 +81,11 @@ struct Pnv8Chip {
      PnvHomer     homer;
 
  #define PNV8_CHIP_PHB3_MAX 4
-    PnvPHB       phbs[PNV8_CHIP_PHB3_MAX];
+    /*
+     * The array is used to allow quick access to the phbs by
+     * pnv_ics_get_child() and pnv_ics_resend_child().
+     */
+    PnvPHB       *phbs[PNV8_CHIP_PHB3_MAX];
      uint32_t     num_phbs;
 
      XICSFabric    *xics;
Reviewed-by on v4 5/11
#12 lore raw

Re: [PATCH for-7.2 v4 05/11] ppc/pnv: turn chip8->phbs[] into a PnvPHB* array

In reply to Daniel Henrique Barboza
Quoted 11 lines · Daniel Henrique Barboza

When enabling user created PHBs (a change reverted by commit 9c10d86fee) we were handling PHBs created by default versus by the user in different manners. The only difference between these PHBs is that one will have a valid phb3->chip that is assigned during pnv_chip_power8_realize(), while the user created needs to search which chip it belongs to.

Aside from that there shouldn't be any difference. Making the default PHBs behave in line with the user created ones will make it easier to re-introduce them later on. It will also make the code easier to follow since we are dealing with them in equal manner.

Quoted 8 lines

The first step is to turn chip8->phbs[] into a PnvPHB3 pointer array. This will allow us to assign user created PHBs into it later on. The way we initilize the default case is now more in line with that would happen with the user created case: the object is created, parented by the chip because pnv_xscom_dt() relies on it, and then assigned to the array.

---

Fred

Quoted 94 lines
  hw/ppc/pnv.c         | 27 ++++++++++++++++++++++-----
  include/hw/ppc/pnv.h |  6 +++++-
  2 files changed, 27 insertions(+), 6 deletions(-)
hw/ppc/pnv.c +22 -5
index 2deaac17f7..e53e9e297d 100644
@@ -294,6 +294,13 @@ static void pnv_dt_icp(PnvChip *chip, void *fdt, uint32_t pir,
  Object *pnv_chip_add_phb(PnvChip *chip, PnvPHB *phb, Error **errp)
  {
      if (phb->version == 3) {
+        Pnv8Chip *chip8 = PNV8_CHIP(chip);
+
+        phb->chip = chip;
+
+        chip8->phbs[chip8->num_phbs] = phb;
+        chip8->num_phbs++;
+
          return OBJECT(chip);
      } else {
          /* phb4 support will be added later */
@@ -681,7 +688,7 @@ static void pnv_chip_power8_pic_print_info(PnvChip *chip, Monitor *mon)
      ics_pic_print_info(&chip8->psi.ics, mon);
 
      for (i = 0; i < chip8->num_phbs; i++) {
-        PnvPHB *phb = &chip8->phbs[i];
+        PnvPHB *phb = chip8->phbs[i];
          PnvPHB3 *phb3 = PNV_PHB3(phb->backend);
 
          pnv_phb3_msi_pic_print_info(&phb3->msis, mon);
@@ -1174,7 +1181,17 @@ static void pnv_chip_power8_instance_init(Object *obj)
      chip8->num_phbs = pcc->num_phbs;
 
      for (i = 0; i < chip8->num_phbs; i++) {
-        object_initialize_child(obj, "phb[*]", &chip8->phbs[i], TYPE_PNV_PHB);
+        Object *phb = object_new(TYPE_PNV_PHB);
+
+        /*
+         * We need the chip to parent the PHB to allow the DT
+         * to build correctly (via pnv_xscom_dt()).
+         *
+         * TODO: the PHB should be parented by a PEC device that, at
+         * this moment, is not modelled powernv8/phb3.
+         */
+        object_property_add_child(obj, "phb[*]", phb);
+        chip8->phbs[i] = PNV_PHB(phb);
      }
 
  }
@@ -1290,7 +1307,7 @@ static void pnv_chip_power8_realize(DeviceState *dev, Error **errp)
 
      /* PHB controllers */
      for (i = 0; i < chip8->num_phbs; i++) {
-        PnvPHB *phb = &chip8->phbs[i];
+        PnvPHB *phb = chip8->phbs[i];
 
          object_property_set_int(OBJECT(phb), "index", i, &error_fatal);
          object_property_set_int(OBJECT(phb), "chip-id", chip->chip_id,
@@ -1958,7 +1975,7 @@ static ICSState *pnv_ics_get(XICSFabric *xi, int irq)
          }
 
          for (j = 0; j < chip8->num_phbs; j++) {
-            PnvPHB *phb = &chip8->phbs[j];
+            PnvPHB *phb = chip8->phbs[j];
              PnvPHB3 *phb3 = PNV_PHB3(phb->backend);
 
              if (ics_valid_irq(&phb3->lsis, irq)) {
@@ -1997,7 +2014,7 @@ static void pnv_ics_resend(XICSFabric *xi)
          ics_resend(&chip8->psi.ics);
 
          for (j = 0; j < chip8->num_phbs; j++) {
-            PnvPHB *phb = &chip8->phbs[j];
+            PnvPHB *phb = chip8->phbs[j];
              PnvPHB3 *phb3 = PNV_PHB3(phb->backend);
 
              ics_resend(&phb3->lsis);
include/hw/ppc/pnv.h +5 -1
index 781d0acffa..49433281d7 100644
@@ -81,7 +81,11 @@ struct Pnv8Chip {
      PnvHomer     homer;
 
  #define PNV8_CHIP_PHB3_MAX 4
-    PnvPHB       phbs[PNV8_CHIP_PHB3_MAX];
+    /*
+     * The array is used to allow quick access to the phbs by
+     * pnv_ics_get_child() and pnv_ics_resend_child().
+     */
+    PnvPHB       *phbs[PNV8_CHIP_PHB3_MAX];
      uint32_t     num_phbs;
 
      XICSFabric    *xics;
#13 lore raw

PATCH v4 6/11 ppc/pnv: enable user created pnv-phb for powernv8

In reply to Daniel Henrique Barboza

The bulk of the work was already done by previous patches.

Use defaults_enabled() to determine whether we need to create the default devices or not.

---
 hw/pci-host/pnv_phb.c |  9 +++++++--
 hw/ppc/pnv.c          | 32 ++++++++++++++++++--------------
 2 files changed, 25 insertions(+), 16 deletions(-)
hw/pci-host/pnv_phb.c +7 -2
index 5dc44f45d1..1f53ff77c5 100644
@@ -17,6 +17,7 @@
 #include "hw/ppc/pnv.h"
 #include "hw/qdev-properties.h"
 #include "qom/object.h"
+#include "sysemu/sysemu.h"
 
 
 /*
@@ -166,6 +167,10 @@ static void pnv_phb_realize(DeviceState *dev, Error **errp)
         pnv_phb4_bus_init(dev, PNV_PHB4(phb->backend));
     }
 
+    if (phb->version == 3 && !defaults_enabled()) {
+        return;
+    }
+
     pnv_phb_attach_root_port(pci);
 }
 
@@ -201,7 +206,7 @@ static void pnv_phb_class_init(ObjectClass *klass, void *data)
     dc->realize = pnv_phb_realize;
     device_class_set_props(dc, pnv_phb_properties);
     set_bit(DEVICE_CATEGORY_BRIDGE, dc->categories);
-    dc->user_creatable = false;
+    dc->user_creatable = true;
 }
 
 static void pnv_phb_root_port_reset(DeviceState *dev)
@@ -292,7 +297,7 @@ static void pnv_phb_root_port_class_init(ObjectClass *klass, void *data)
     device_class_set_parent_reset(dc, pnv_phb_root_port_reset,
                                   &rpc->parent_reset);
     dc->reset = &pnv_phb_root_port_reset;
-    dc->user_creatable = false;
+    dc->user_creatable = true;
 
     k->vendor_id = PCI_VENDOR_ID_IBM;
     /* device_id will be written during realize() */
hw/ppc/pnv.c +18 -14
index e53e9e297d..e82d6522f0 100644
@@ -1178,20 +1178,22 @@ static void pnv_chip_power8_instance_init(Object *obj)
 
     object_initialize_child(obj, "homer", &chip8->homer, TYPE_PNV8_HOMER);
 
-    chip8->num_phbs = pcc->num_phbs;
-
-    for (i = 0; i < chip8->num_phbs; i++) {
-        Object *phb = object_new(TYPE_PNV_PHB);
-
-        /*
-         * We need the chip to parent the PHB to allow the DT
-         * to build correctly (via pnv_xscom_dt()).
-         *
-         * TODO: the PHB should be parented by a PEC device that, at
-         * this moment, is not modelled powernv8/phb3.
-         */
-        object_property_add_child(obj, "phb[*]", phb);
-        chip8->phbs[i] = PNV_PHB(phb);
+    if (defaults_enabled()) {
+        chip8->num_phbs = pcc->num_phbs;
+
+        for (i = 0; i < chip8->num_phbs; i++) {
+            Object *phb = object_new(TYPE_PNV_PHB);
+
+            /*
+             * We need the chip to parent the PHB to allow the DT
+             * to build correctly (via pnv_xscom_dt()).
+             *
+             * TODO: the PHB should be parented by a PEC device that, at
+             * this moment, is not modelled powernv8/phb3.
+             */
+            object_property_add_child(obj, "phb[*]", phb);
+            chip8->phbs[i] = PNV_PHB(phb);
+        }
     }
 
 }
@@ -2130,6 +2132,8 @@ static void pnv_machine_power8_class_init(ObjectClass *oc, void *data)
 
     pmc->compat = compat;
     pmc->compat_size = sizeof(compat);
+
+    machine_class_allow_dynamic_sysbus_dev(mc, TYPE_PNV_PHB);
 }
 
 static void pnv_machine_power9_class_init(ObjectClass *oc, void *data)
Signature · 1 line
2.36.1
Reviewed-by on v4 6/11
#14 lore raw

Re: [PATCH for-7.2 v4 06/11] ppc/pnv: enable user created pnv-phb for powernv8

Quoted 8 lines · Daniel Henrique Barboza

The bulk of the work was already done by previous patches.

Use defaults_enabled() to determine whether we need to create the default devices or not.

---

The QOM relationship for user-created root port is not ideal, but it's addressed in the last patch of the series.

Fred

Quoted 95 lines
  hw/pci-host/pnv_phb.c |  9 +++++++--
  hw/ppc/pnv.c          | 32 ++++++++++++++++++--------------
  2 files changed, 25 insertions(+), 16 deletions(-)
hw/pci-host/pnv_phb.c +7 -2
index 5dc44f45d1..1f53ff77c5 100644
@@ -17,6 +17,7 @@
  #include "hw/ppc/pnv.h"
  #include "hw/qdev-properties.h"
  #include "qom/object.h"
+#include "sysemu/sysemu.h"
 
 
  /*
@@ -166,6 +167,10 @@ static void pnv_phb_realize(DeviceState *dev, Error **errp)
          pnv_phb4_bus_init(dev, PNV_PHB4(phb->backend));
      }
 
+    if (phb->version == 3 && !defaults_enabled()) {
+        return;
+    }
+
      pnv_phb_attach_root_port(pci);
  }
 
@@ -201,7 +206,7 @@ static void pnv_phb_class_init(ObjectClass *klass, void *data)
      dc->realize = pnv_phb_realize;
      device_class_set_props(dc, pnv_phb_properties);
      set_bit(DEVICE_CATEGORY_BRIDGE, dc->categories);
-    dc->user_creatable = false;
+    dc->user_creatable = true;
  }
 
  static void pnv_phb_root_port_reset(DeviceState *dev)
@@ -292,7 +297,7 @@ static void pnv_phb_root_port_class_init(ObjectClass *klass, void *data)
      device_class_set_parent_reset(dc, pnv_phb_root_port_reset,
                                    &rpc->parent_reset);
      dc->reset = &pnv_phb_root_port_reset;
-    dc->user_creatable = false;
+    dc->user_creatable = true;
 
      k->vendor_id = PCI_VENDOR_ID_IBM;
      /* device_id will be written during realize() */
hw/ppc/pnv.c +18 -14
index e53e9e297d..e82d6522f0 100644
@@ -1178,20 +1178,22 @@ static void pnv_chip_power8_instance_init(Object *obj)
 
      object_initialize_child(obj, "homer", &chip8->homer, TYPE_PNV8_HOMER);
 
-    chip8->num_phbs = pcc->num_phbs;
-
-    for (i = 0; i < chip8->num_phbs; i++) {
-        Object *phb = object_new(TYPE_PNV_PHB);
-
-        /*
-         * We need the chip to parent the PHB to allow the DT
-         * to build correctly (via pnv_xscom_dt()).
-         *
-         * TODO: the PHB should be parented by a PEC device that, at
-         * this moment, is not modelled powernv8/phb3.
-         */
-        object_property_add_child(obj, "phb[*]", phb);
-        chip8->phbs[i] = PNV_PHB(phb);
+    if (defaults_enabled()) {
+        chip8->num_phbs = pcc->num_phbs;
+
+        for (i = 0; i < chip8->num_phbs; i++) {
+            Object *phb = object_new(TYPE_PNV_PHB);
+
+            /*
+             * We need the chip to parent the PHB to allow the DT
+             * to build correctly (via pnv_xscom_dt()).
+             *
+             * TODO: the PHB should be parented by a PEC device that, at
+             * this moment, is not modelled powernv8/phb3.
+             */
+            object_property_add_child(obj, "phb[*]", phb);
+            chip8->phbs[i] = PNV_PHB(phb);
+        }
      }
 
  }
@@ -2130,6 +2132,8 @@ static void pnv_machine_power8_class_init(ObjectClass *oc, void *data)
 
      pmc->compat = compat;
      pmc->compat_size = sizeof(compat);
+
+    machine_class_allow_dynamic_sysbus_dev(mc, TYPE_PNV_PHB);
  }
 
  static void pnv_machine_power9_class_init(ObjectClass *oc, void *data)
#15 lore raw

PATCH v4 7/11 ppc/pnv: add PHB4 helpers for user created pnv-phb

In reply to Daniel Henrique Barboza

The PHB4 backend relies on a link with the corresponding PEC element. This is trivial to do during machine_init() time for default devices, but not so much for user created ones.

pnv_phb4_get_pec() is a small variation of the function that was reverted by commit 9c10d86fee "ppc/pnv: Remove user-created PHB{3,4,5} devices". We'll use it to determine the appropriate PEC for a given user created pnv-phb that uses a PHB4 backend.

This is done during realize() time, in pnv_phb_user_device_init().

---
 hw/ppc/pnv.c | 35 ++++++++++++++++++++++++++++++++---
 1 file changed, 32 insertions(+), 3 deletions(-)
hw/ppc/pnv.c +32 -3
index e82d6522f0..0644f45a1d 100644
@@ -281,6 +281,34 @@ static void pnv_dt_icp(PnvChip *chip, void *fdt, uint32_t pir,
     g_free(reg);
 }
 
+static PnvPhb4PecState *pnv_phb4_get_pec(PnvChip *chip, PnvPHB4 *phb,
+                                         Error **errp)
+{
+    Pnv9Chip *chip9 = PNV9_CHIP(chip);
+    int chip_id = phb->chip_id;
+    int index = phb->phb_id;
+    int i, j;
+
+    for (i = 0; i < chip->num_pecs; i++) {
+        /*
+         * For each PEC, check the amount of phbs it supports
+         * and see if the given phb4 index matches an index.
+         */
+        PnvPhb4PecState *pec = &chip9->pecs[i];
+
+        for (j = 0; j < pec->num_phbs; j++) {
+            if (index == pnv_phb4_pec_get_phb_id(pec, j)) {
+                return pec;
+            }
+        }
+    }
+    error_setg(errp,
+               "pnv-phb4 chip-id %d index %d didn't match any existing PEC",
+               chip_id, index);
+
+    return NULL;
+}
+
 /*
  * Adds a PnvPHB to the chip. Returns the parent obj of the
  * PHB which varies with each version (phb version 3 is parented
@@ -302,10 +330,11 @@ Object *pnv_chip_add_phb(PnvChip *chip, PnvPHB *phb, Error **errp)
         chip8->num_phbs++;
 
         return OBJECT(chip);
-    } else {
-        /* phb4 support will be added later */
-        return NULL;
     }
+
+    phb->pec = pnv_phb4_get_pec(chip, PNV_PHB4(phb->backend), errp);
+
+    return OBJECT(phb->pec);
 }
 
 static void pnv_chip_power8_dt_populate(PnvChip *chip, void *fdt)
Signature · 1 line
2.36.1
Reviewed-by on v4 7/11
#16 lore raw

Re: [PATCH for-7.2 v4 07/11] ppc/pnv: add PHB4 helpers for user created pnv-phb

Quoted 4 lines · Daniel Henrique Barboza

The PHB4 backend relies on a link with the corresponding PEC element. This is trivial to do during machine_init() time for default devices, but not so much for user created ones.

Quoted 8 lines

pnv_phb4_get_pec() is a small variation of the function that was reverted by commit 9c10d86fee "ppc/pnv: Remove user-created PHB{3,4,5} devices". We'll use it to determine the appropriate PEC for a given user created pnv-phb that uses a PHB4 backend.

This is done during realize() time, in pnv_phb_user_device_init().

Thanks,

C.

Quoted 58 lines
---
  hw/ppc/pnv.c | 35 ++++++++++++++++++++++++++++++++---
  1 file changed, 32 insertions(+), 3 deletions(-)
hw/ppc/pnv.c +32 -3
index e82d6522f0..0644f45a1d 100644
@@ -281,6 +281,34 @@ static void pnv_dt_icp(PnvChip *chip, void *fdt, uint32_t pir,
      g_free(reg);
  }
 
+static PnvPhb4PecState *pnv_phb4_get_pec(PnvChip *chip, PnvPHB4 *phb,
+                                         Error **errp)
+{
+    Pnv9Chip *chip9 = PNV9_CHIP(chip);
+    int chip_id = phb->chip_id;
+    int index = phb->phb_id;
+    int i, j;
+
+    for (i = 0; i < chip->num_pecs; i++) {
+        /*
+         * For each PEC, check the amount of phbs it supports
+         * and see if the given phb4 index matches an index.
+         */
+        PnvPhb4PecState *pec = &chip9->pecs[i];
+
+        for (j = 0; j < pec->num_phbs; j++) {
+            if (index == pnv_phb4_pec_get_phb_id(pec, j)) {
+                return pec;
+            }
+        }
+    }
+    error_setg(errp,
+               "pnv-phb4 chip-id %d index %d didn't match any existing PEC",
+               chip_id, index);
+
+    return NULL;
+}
+
  /*
   * Adds a PnvPHB to the chip. Returns the parent obj of the
   * PHB which varies with each version (phb version 3 is parented
@@ -302,10 +330,11 @@ Object *pnv_chip_add_phb(PnvChip *chip, PnvPHB *phb, Error **errp)
          chip8->num_phbs++;
 
          return OBJECT(chip);
-    } else {
-        /* phb4 support will be added later */
-        return NULL;
      }
+
+    phb->pec = pnv_phb4_get_pec(chip, PNV_PHB4(phb->backend), errp);
+
+    return OBJECT(phb->pec);
  }
 
  static void pnv_chip_power8_dt_populate(PnvChip *chip, void *fdt)
Reviewed-by on v4 7/11
#17 lore raw

Re: [PATCH for-7.2 v4 07/11] ppc/pnv: add PHB4 helpers for user created pnv-phb

In reply to Daniel Henrique Barboza
Quoted 9 lines · Daniel Henrique Barboza

The PHB4 backend relies on a link with the corresponding PEC element. This is trivial to do during machine_init() time for default devices, but not so much for user created ones.

pnv_phb4_get_pec() is a small variation of the function that was reverted by commit 9c10d86fee "ppc/pnv: Remove user-created PHB{3,4,5} devices". We'll use it to determine the appropriate PEC for a given user created pnv-phb that uses a PHB4 backend.

Quoted 4 lines

This is done during realize() time, in pnv_phb_user_device_init().

---

Fred

Quoted 57 lines
  hw/ppc/pnv.c | 35 ++++++++++++++++++++++++++++++++---
  1 file changed, 32 insertions(+), 3 deletions(-)
hw/ppc/pnv.c +32 -3
index e82d6522f0..0644f45a1d 100644
@@ -281,6 +281,34 @@ static void pnv_dt_icp(PnvChip *chip, void *fdt, uint32_t pir,
      g_free(reg);
  }
 
+static PnvPhb4PecState *pnv_phb4_get_pec(PnvChip *chip, PnvPHB4 *phb,
+                                         Error **errp)
+{
+    Pnv9Chip *chip9 = PNV9_CHIP(chip);
+    int chip_id = phb->chip_id;
+    int index = phb->phb_id;
+    int i, j;
+
+    for (i = 0; i < chip->num_pecs; i++) {
+        /*
+         * For each PEC, check the amount of phbs it supports
+         * and see if the given phb4 index matches an index.
+         */
+        PnvPhb4PecState *pec = &chip9->pecs[i];
+
+        for (j = 0; j < pec->num_phbs; j++) {
+            if (index == pnv_phb4_pec_get_phb_id(pec, j)) {
+                return pec;
+            }
+        }
+    }
+    error_setg(errp,
+               "pnv-phb4 chip-id %d index %d didn't match any existing PEC",
+               chip_id, index);
+
+    return NULL;
+}
+
  /*
   * Adds a PnvPHB to the chip. Returns the parent obj of the
   * PHB which varies with each version (phb version 3 is parented
@@ -302,10 +330,11 @@ Object *pnv_chip_add_phb(PnvChip *chip, PnvPHB *phb, Error **errp)
          chip8->num_phbs++;
 
          return OBJECT(chip);
-    } else {
-        /* phb4 support will be added later */
-        return NULL;
      }
+
+    phb->pec = pnv_phb4_get_pec(chip, PNV_PHB4(phb->backend), errp);
+
+    return OBJECT(phb->pec);
  }
 
  static void pnv_chip_power8_dt_populate(PnvChip *chip, void *fdt)
#18 lore raw

PATCH v4 8/11 ppc/pnv: enable user created pnv-phb for powernv9

In reply to Daniel Henrique Barboza

Enable pnv-phb user created devices for powernv9 now that we have everything in place.

---
 hw/pci-host/pnv_phb.c      | 2 +-
 hw/pci-host/pnv_phb4_pec.c | 6 ++++--
 hw/ppc/pnv.c               | 2 ++
 3 files changed, 7 insertions(+), 3 deletions(-)
hw/pci-host/pnv_phb.c +1 -1
index 1f53ff77c5..17d9960aa1 100644
@@ -167,7 +167,7 @@ static void pnv_phb_realize(DeviceState *dev, Error **errp)
         pnv_phb4_bus_init(dev, PNV_PHB4(phb->backend));
     }
 
-    if (phb->version == 3 && !defaults_enabled()) {
+    if (!defaults_enabled()) {
         return;
     }
 
hw/pci-host/pnv_phb4_pec.c +4 -2
index 8dc363d69c..9871f462cd 100644
@@ -146,8 +146,10 @@ static void pnv_pec_realize(DeviceState *dev, Error **errp)
     pec->num_phbs = pecc->num_phbs[pec->index];
 
     /* Create PHBs if running with defaults */
-    for (i = 0; i < pec->num_phbs; i++) {
-        pnv_pec_default_phb_realize(pec, i, errp);
+    if (defaults_enabled()) {
+        for (i = 0; i < pec->num_phbs; i++) {
+            pnv_pec_default_phb_realize(pec, i, errp);
+        }
     }
 
     /* Initialize the XSCOM regions for the PEC registers */
hw/ppc/pnv.c +2 -0
index 0644f45a1d..ec0558ed1c 100644
@@ -2188,6 +2188,8 @@ static void pnv_machine_power9_class_init(ObjectClass *oc, void *data)
     pmc->compat = compat;
     pmc->compat_size = sizeof(compat);
     pmc->dt_power_mgt = pnv_dt_power_mgt;
+
+    machine_class_allow_dynamic_sysbus_dev(mc, TYPE_PNV_PHB);
 }
 
 static void pnv_machine_power10_class_init(ObjectClass *oc, void *data)
Signature · 1 line
2.36.1
Reviewed-by on v4 8/11
#19 lore raw

Re: [PATCH for-7.2 v4 08/11] ppc/pnv: enable user created pnv-phb for powernv9

Quoted 6 lines · Daniel Henrique Barboza

Enable pnv-phb user created devices for powernv9 now that we have everything in place.

---

Same comment as in patch 6 regarding the QOM relationship of the user-created root port.

Fred

Quoted 48 lines
  hw/pci-host/pnv_phb.c      | 2 +-
  hw/pci-host/pnv_phb4_pec.c | 6 ++++--
  hw/ppc/pnv.c               | 2 ++
  3 files changed, 7 insertions(+), 3 deletions(-)
hw/pci-host/pnv_phb.c +1 -1
index 1f53ff77c5..17d9960aa1 100644
@@ -167,7 +167,7 @@ static void pnv_phb_realize(DeviceState *dev, Error **errp)
          pnv_phb4_bus_init(dev, PNV_PHB4(phb->backend));
      }
 
-    if (phb->version == 3 && !defaults_enabled()) {
+    if (!defaults_enabled()) {
          return;
      }
 
hw/pci-host/pnv_phb4_pec.c +4 -2
index 8dc363d69c..9871f462cd 100644
@@ -146,8 +146,10 @@ static void pnv_pec_realize(DeviceState *dev, Error **errp)
      pec->num_phbs = pecc->num_phbs[pec->index];
 
      /* Create PHBs if running with defaults */
-    for (i = 0; i < pec->num_phbs; i++) {
-        pnv_pec_default_phb_realize(pec, i, errp);
+    if (defaults_enabled()) {
+        for (i = 0; i < pec->num_phbs; i++) {
+            pnv_pec_default_phb_realize(pec, i, errp);
+        }
      }
 
      /* Initialize the XSCOM regions for the PEC registers */
hw/ppc/pnv.c +2 -0
index 0644f45a1d..ec0558ed1c 100644
@@ -2188,6 +2188,8 @@ static void pnv_machine_power9_class_init(ObjectClass *oc, void *data)
      pmc->compat = compat;
      pmc->compat_size = sizeof(compat);
      pmc->dt_power_mgt = pnv_dt_power_mgt;
+
+    machine_class_allow_dynamic_sysbus_dev(mc, TYPE_PNV_PHB);
  }
 
  static void pnv_machine_power10_class_init(ObjectClass *oc, void *data)
#20 lore raw

PATCH v4 9/11 ppc/pnv: change pnv_phb4_get_pec() to also retrieve chip10->pecs

In reply to Daniel Henrique Barboza

The function assumes that we're always dealing with a PNV9_CHIP() object. This is not the case when the pnv-phb device belongs to a powernv10 machine.

Change pnv_phb4_get_pec() to be able to work with PNV10_CHIP() if necessary.

---
 hw/ppc/pnv.c | 17 +++++++++++++++--
 1 file changed, 15 insertions(+), 2 deletions(-)
hw/ppc/pnv.c +15 -2
index ec0558ed1c..e6c14fe231 100644
@@ -284,17 +284,30 @@ static void pnv_dt_icp(PnvChip *chip, void *fdt, uint32_t pir,
 static PnvPhb4PecState *pnv_phb4_get_pec(PnvChip *chip, PnvPHB4 *phb,
                                          Error **errp)
 {
-    Pnv9Chip *chip9 = PNV9_CHIP(chip);
+    PnvPHB *phb_base = phb->phb_base;
+    PnvPhb4PecState *pecs = NULL;
     int chip_id = phb->chip_id;
     int index = phb->phb_id;
     int i, j;
 
+    if (phb_base->version == 4) {
+        Pnv9Chip *chip9 = PNV9_CHIP(chip);
+
+        pecs = chip9->pecs;
+    } else if (phb_base->version == 5) {
+        Pnv10Chip *chip10 = PNV10_CHIP(chip);
+
+        pecs = chip10->pecs;
+    } else {
+        g_assert_not_reached();
+    }
+
     for (i = 0; i < chip->num_pecs; i++) {
         /*
          * For each PEC, check the amount of phbs it supports
          * and see if the given phb4 index matches an index.
          */
-        PnvPhb4PecState *pec = &chip9->pecs[i];
+        PnvPhb4PecState *pec = &pecs[i];
 
         for (j = 0; j < pec->num_phbs; j++) {
             if (index == pnv_phb4_pec_get_phb_id(pec, j)) {
Signature · 1 line
2.36.1
Reviewed-by on v4 9/11
#21 lore raw

Re: [PATCH for-7.2 v4 09/11] ppc/pnv: change pnv_phb4_get_pec() to also retrieve chip10->pecs

Quoted 8 lines · Daniel Henrique Barboza

The function assumes that we're always dealing with a PNV9_CHIP() object. This is not the case when the pnv-phb device belongs to a powernv10 machine.

Change pnv_phb4_get_pec() to be able to work with PNV10_CHIP() if necessary.

Thanks,

C.

Quoted 41 lines
---
  hw/ppc/pnv.c | 17 +++++++++++++++--
  1 file changed, 15 insertions(+), 2 deletions(-)
hw/ppc/pnv.c +15 -2
index ec0558ed1c..e6c14fe231 100644
@@ -284,17 +284,30 @@ static void pnv_dt_icp(PnvChip *chip, void *fdt, uint32_t pir,
  static PnvPhb4PecState *pnv_phb4_get_pec(PnvChip *chip, PnvPHB4 *phb,
                                           Error **errp)
  {
-    Pnv9Chip *chip9 = PNV9_CHIP(chip);
+    PnvPHB *phb_base = phb->phb_base;
+    PnvPhb4PecState *pecs = NULL;
      int chip_id = phb->chip_id;
      int index = phb->phb_id;
      int i, j;
 
+    if (phb_base->version == 4) {
+        Pnv9Chip *chip9 = PNV9_CHIP(chip);
+
+        pecs = chip9->pecs;
+    } else if (phb_base->version == 5) {
+        Pnv10Chip *chip10 = PNV10_CHIP(chip);
+
+        pecs = chip10->pecs;
+    } else {
+        g_assert_not_reached();
+    }
+
      for (i = 0; i < chip->num_pecs; i++) {
          /*
           * For each PEC, check the amount of phbs it supports
           * and see if the given phb4 index matches an index.
           */
-        PnvPhb4PecState *pec = &chip9->pecs[i];
+        PnvPhb4PecState *pec = &pecs[i];
 
          for (j = 0; j < pec->num_phbs; j++) {
              if (index == pnv_phb4_pec_get_phb_id(pec, j)) {
Reviewed-by on v4 9/11
#22 lore raw

Re: [PATCH for-7.2 v4 09/11] ppc/pnv: change pnv_phb4_get_pec() to also retrieve chip10->pecs

In reply to Daniel Henrique Barboza
Quoted 9 lines · Daniel Henrique Barboza

The function assumes that we're always dealing with a PNV9_CHIP() object. This is not the case when the pnv-phb device belongs to a powernv10 machine.

Change pnv_phb4_get_pec() to be able to work with PNV10_CHIP() if necessary.

---

Fred

Quoted 40 lines
  hw/ppc/pnv.c | 17 +++++++++++++++--
  1 file changed, 15 insertions(+), 2 deletions(-)
hw/ppc/pnv.c +15 -2
index ec0558ed1c..e6c14fe231 100644
@@ -284,17 +284,30 @@ static void pnv_dt_icp(PnvChip *chip, void *fdt, uint32_t pir,
  static PnvPhb4PecState *pnv_phb4_get_pec(PnvChip *chip, PnvPHB4 *phb,
                                           Error **errp)
  {
-    Pnv9Chip *chip9 = PNV9_CHIP(chip);
+    PnvPHB *phb_base = phb->phb_base;
+    PnvPhb4PecState *pecs = NULL;
      int chip_id = phb->chip_id;
      int index = phb->phb_id;
      int i, j;
 
+    if (phb_base->version == 4) {
+        Pnv9Chip *chip9 = PNV9_CHIP(chip);
+
+        pecs = chip9->pecs;
+    } else if (phb_base->version == 5) {
+        Pnv10Chip *chip10 = PNV10_CHIP(chip);
+
+        pecs = chip10->pecs;
+    } else {
+        g_assert_not_reached();
+    }
+
      for (i = 0; i < chip->num_pecs; i++) {
          /*
           * For each PEC, check the amount of phbs it supports
           * and see if the given phb4 index matches an index.
           */
-        PnvPhb4PecState *pec = &chip9->pecs[i];
+        PnvPhb4PecState *pec = &pecs[i];
 
          for (j = 0; j < pec->num_phbs; j++) {
              if (index == pnv_phb4_pec_get_phb_id(pec, j)) {
#23 lore raw

PATCH v4 10/11 ppc/pnv: user creatable pnv-phb for powernv10

In reply to Daniel Henrique Barboza

Given that powernv9 and powernv10 uses the same pnv-phb backend, the logic to allow user created pnv-phbs for powernv10 is already in place. Let's flip the switch.

---
 hw/ppc/pnv.c | 2 ++
 1 file changed, 2 insertions(+)
hw/ppc/pnv.c +2 -0
index e6c14fe231..9bf35ca9d6 100644
@@ -2226,6 +2226,8 @@ static void pnv_machine_power10_class_init(ObjectClass *oc, void *data)
     pmc->dt_power_mgt = pnv_dt_power_mgt;
 
     xfc->match_nvt = pnv10_xive_match_nvt;
+
+    machine_class_allow_dynamic_sysbus_dev(mc, TYPE_PNV_PHB);
 }
 
 static bool pnv_machine_get_hb(Object *obj, Error **errp)
Signature · 1 line
2.36.1
Reviewed-by on v4 10/11
#24 lore raw

Re: [PATCH for-7.2 v4 10/11] ppc/pnv: user creatable pnv-phb for powernv10

Quoted 7 lines · Daniel Henrique Barboza

Given that powernv9 and powernv10 uses the same pnv-phb backend, the logic to allow user created pnv-phbs for powernv10 is already in place. Let's flip the switch.

---

Fred

Quoted 16 lines
  hw/ppc/pnv.c | 2 ++
  1 file changed, 2 insertions(+)
hw/ppc/pnv.c +2 -0
index e6c14fe231..9bf35ca9d6 100644
@@ -2226,6 +2226,8 @@ static void pnv_machine_power10_class_init(ObjectClass *oc, void *data)
      pmc->dt_power_mgt = pnv_dt_power_mgt;
 
      xfc->match_nvt = pnv10_xive_match_nvt;
+
+    machine_class_allow_dynamic_sysbus_dev(mc, TYPE_PNV_PHB);
  }
 
  static bool pnv_machine_get_hb(Object *obj, Error **errp)
#25 lore raw

PATCH v4 11/11 ppc/pnv: fix QOM parenting of user creatable root ports

In reply to Daniel Henrique Barboza

User creatable root ports are being parented by the 'peripheral' or the 'peripheral-anon' container. This happens because this is the regular QOM schema for sysbus devices that are added via the command line.

Let's make this QOM hierarchy similar to what we have with default root ports, i.e. the root port must be parented by the pnv-root-bus. To do that we change the qom and bus parent of the root port during root_port_realize(). The realize() is shared by the default root port code path, so we can remove the code inside pnv_phb_attach_root_port() that was adding the root port as a child of the bus as well.

While we're at it, change pnv_phb_attach_root_port() to receive a PCIBus instead of a PCIHostState to make it clear that the function does not make use of the PHB.

---
 hw/pci-host/pnv_phb.c | 35 +++++++++++++++--------------------
 1 file changed, 15 insertions(+), 20 deletions(-)
hw/pci-host/pnv_phb.c +15 -20
index 17d9960aa1..65ed1f9eb4 100644
@@ -51,27 +51,11 @@ static void pnv_parent_bus_fixup(DeviceState *parent, DeviceState *child,
     }
 }
 
-/*
- * Attach a root port device.
- *
- * 'index' will be used both as a PCIE slot value and to calculate
- * QOM id. 'chip_id' is going to be used as PCIE chassis for the
- * root port.
- */
-static void pnv_phb_attach_root_port(PCIHostState *pci)
+static void pnv_phb_attach_root_port(PCIBus *bus)
 {
     PCIDevice *root = pci_new(PCI_DEVFN(0, 0), TYPE_PNV_PHB_ROOT_PORT);
-    const char *dev_id = DEVICE(root)->id;
-    g_autofree char *default_id = NULL;
-    int index;
-
-    index = object_property_get_int(OBJECT(pci->bus), "phb-id", &error_fatal);
-    default_id = g_strdup_printf("%s[%d]", TYPE_PNV_PHB_ROOT_PORT, index);
-
-    object_property_add_child(OBJECT(pci->bus), dev_id ? dev_id : default_id,
-                              OBJECT(root));
 
-    pci_realize_and_unref(root, pci->bus, &error_fatal);
+    pci_realize_and_unref(root, bus, &error_fatal);
 }
 
 /*
@@ -171,7 +155,7 @@ static void pnv_phb_realize(DeviceState *dev, Error **errp)
         return;
     }
 
-    pnv_phb_attach_root_port(pci);
+    pnv_phb_attach_root_port(pci->bus);
 }
 
 static const char *pnv_phb_root_bus_path(PCIHostState *host_bridge,
@@ -240,12 +224,18 @@ static void pnv_phb_root_port_realize(DeviceState *dev, Error **errp)
 {
     PCIERootPortClass *rpc = PCIE_ROOT_PORT_GET_CLASS(dev);
     PnvPHBRootPort *phb_rp = PNV_PHB_ROOT_PORT(dev);
-    PCIBus *bus = PCI_BUS(qdev_get_parent_bus(dev));
+    BusState *qbus = qdev_get_parent_bus(dev);
+    PCIBus *bus = PCI_BUS(qbus);
     PCIDevice *pci = PCI_DEVICE(dev);
     uint16_t device_id = 0;
     Error *local_err = NULL;
     int chip_id, index;
 
+    /*
+     * 'index' will be used both as a PCIE slot value and to calculate
+     * QOM id. 'chip_id' is going to be used as PCIE chassis for the
+     * root port.
+     */
     chip_id = object_property_get_int(OBJECT(bus), "chip-id", &error_fatal);
     index = object_property_get_int(OBJECT(bus), "phb-id", &error_fatal);
 
@@ -253,6 +243,11 @@ static void pnv_phb_root_port_realize(DeviceState *dev, Error **errp)
     qdev_prop_set_uint8(dev, "chassis", chip_id);
     qdev_prop_set_uint16(dev, "slot", index);
 
+    pnv_parent_qom_fixup(OBJECT(bus), OBJECT(dev), index);
+    if (!qdev_set_parent_bus(dev, qbus, &error_fatal)) {
+        return;
+    }
+
     rpc->parent_realize(dev, &local_err);
     if (local_err) {
         error_propagate(errp, local_err);
Signature · 1 line
2.36.1
#26 lore raw

Re: [PATCH for-7.2 v4 11/11] ppc/pnv: fix QOM parenting of user creatable root ports

Quoted 84 lines · Daniel Henrique Barboza

User creatable root ports are being parented by the 'peripheral' or the 'peripheral-anon' container. This happens because this is the regular QOM schema for sysbus devices that are added via the command line.

Let's make this QOM hierarchy similar to what we have with default root ports, i.e. the root port must be parented by the pnv-root-bus. To do that we change the qom and bus parent of the root port during root_port_realize(). The realize() is shared by the default root port code path, so we can remove the code inside pnv_phb_attach_root_port() that was adding the root port as a child of the bus as well.

While we're at it, change pnv_phb_attach_root_port() to receive a PCIBus instead of a PCIHostState to make it clear that the function does not make use of the PHB.

---
  hw/pci-host/pnv_phb.c | 35 +++++++++++++++--------------------
  1 file changed, 15 insertions(+), 20 deletions(-)
hw/pci-host/pnv_phb.c +10 -20
index 17d9960aa1..65ed1f9eb4 100644
@@ -51,27 +51,11 @@ static void pnv_parent_bus_fixup(DeviceState *parent, DeviceState *child,
      }
  }
 
-/*
- * Attach a root port device.
- *
- * 'index' will be used both as a PCIE slot value and to calculate
- * QOM id. 'chip_id' is going to be used as PCIE chassis for the
- * root port.
- */
-static void pnv_phb_attach_root_port(PCIHostState *pci)
+static void pnv_phb_attach_root_port(PCIBus *bus)
  {
      PCIDevice *root = pci_new(PCI_DEVFN(0, 0), TYPE_PNV_PHB_ROOT_PORT);
-    const char *dev_id = DEVICE(root)->id;
-    g_autofree char *default_id = NULL;
-    int index;
-
-    index = object_property_get_int(OBJECT(pci->bus), "phb-id", &error_fatal);
-    default_id = g_strdup_printf("%s[%d]", TYPE_PNV_PHB_ROOT_PORT, index);
-
-    object_property_add_child(OBJECT(pci->bus), dev_id ? dev_id : default_id,
-                              OBJECT(root));
 
-    pci_realize_and_unref(root, pci->bus, &error_fatal);
+    pci_realize_and_unref(root, bus, &error_fatal);
  }
 
  /*
@@ -171,7 +155,7 @@ static void pnv_phb_realize(DeviceState *dev, Error **errp)
          return;
      }
 
-    pnv_phb_attach_root_port(pci);
+    pnv_phb_attach_root_port(pci->bus);
  }
 
  static const char *pnv_phb_root_bus_path(PCIHostState *host_bridge,
@@ -240,12 +224,18 @@ static void pnv_phb_root_port_realize(DeviceState *dev, Error **errp)
  {
      PCIERootPortClass *rpc = PCIE_ROOT_PORT_GET_CLASS(dev);
      PnvPHBRootPort *phb_rp = PNV_PHB_ROOT_PORT(dev);
-    PCIBus *bus = PCI_BUS(qdev_get_parent_bus(dev));
+    BusState *qbus = qdev_get_parent_bus(dev);
+    PCIBus *bus = PCI_BUS(qbus);
      PCIDevice *pci = PCI_DEVICE(dev);
      uint16_t device_id = 0;
      Error *local_err = NULL;
      int chip_id, index;
 
+    /*
+     * 'index' will be used both as a PCIE slot value and to calculate
+     * QOM id. 'chip_id' is going to be used as PCIE chassis for the
+     * root port.
+     */
      chip_id = object_property_get_int(OBJECT(bus), "chip-id", &error_fatal);
      index = object_property_get_int(OBJECT(bus), "phb-id", &error_fatal);
 

hw/pci-host/pnv_phb.c static void pnv_phb_root_port_realize(DeviceState *dev, Error **errp) · line 247

Quoted 5 lines
hw/pci-host/pnv_phb.c +2 -0
@@ -253,6 +243,11 @@ static void pnv_phb_root_port_realize(DeviceState *dev, Error **errp)
      qdev_prop_set_uint8(dev, "chassis", chip_id);
      qdev_prop_set_uint16(dev, "slot", index);
 
+    pnv_parent_qom_fixup(OBJECT(bus), OBJECT(dev), index);
+    if (!qdev_set_parent_bus(dev, qbus, &error_fatal)) {

That line looks surprising at first, because we got qbus from qdev_get_parent_bus() just above, so it looks like a noop. I talked to Daniel about it: the device<->bus relationship is correct when entering the function but the call to pnv_parent_qom_fixup() interferes with the bus relationship, so it needs to be re-established. Short of a better suggestion, it probably need a comment.

Fred

Quoted 6 lines
hw/pci-host/pnv_phb.c +3 -0
+        return;
+    }
+
      rpc->parent_realize(dev, &local_err);
      if (local_err) {
          error_propagate(errp, local_err);
#27 lore raw

Re: [PATCH for-7.2 v4 11/11] ppc/pnv: fix QOM parenting of user creatable root ports

Quoted 79 lines · Frederic Barrat
Quoted 76 lines · Daniel Henrique Barboza

User creatable root ports are being parented by the 'peripheral' or the 'peripheral-anon' container. This happens because this is the regular QOM schema for sysbus devices that are added via the command line.

Let's make this QOM hierarchy similar to what we have with default root ports, i.e. the root port must be parented by the pnv-root-bus. To do that we change the qom and bus parent of the root port during root_port_realize(). The realize() is shared by the default root port code path, so we can remove the code inside pnv_phb_attach_root_port() that was adding the root port as a child of the bus as well.

While we're at it, change pnv_phb_attach_root_port() to receive a PCIBus instead of a PCIHostState to make it clear that the function does not make use of the PHB.

---
  hw/pci-host/pnv_phb.c | 35 +++++++++++++++--------------------

  1 file changed, 15 insertions(+), 20 deletions(-)
hw/pci-host/pnv_phb.c +10 -20
index 17d9960aa1..65ed1f9eb4 100644
@@ -51,27 +51,11 @@ static void pnv_parent_bus_fixup(DeviceState *parent, DeviceState *child,
       }
   }
-/*
- * Attach a root port device.
- *
- * 'index' will be used both as a PCIE slot value and to calculate
- * QOM id. 'chip_id' is going to be used as PCIE chassis for the
- * root port.
- */
-static void pnv_phb_attach_root_port(PCIHostState *pci)
+static void pnv_phb_attach_root_port(PCIBus *bus)
   {
       PCIDevice *root = pci_new(PCI_DEVFN(0, 0), TYPE_PNV_PHB_ROOT_PORT);
-    const char *dev_id = DEVICE(root)->id;
-    g_autofree char *default_id = NULL;
-    int index;
-
-    index = object_property_get_int(OBJECT(pci->bus), "phb-id", &error_fatal);
-    default_id = g_strdup_printf("%s[%d]", TYPE_PNV_PHB_ROOT_PORT, index);
-
-    object_property_add_child(OBJECT(pci->bus), dev_id ? dev_id : default_id,
-                              OBJECT(root));
-    pci_realize_and_unref(root, pci->bus, &error_fatal);
+    pci_realize_and_unref(root, bus, &error_fatal);
   }
   /*
@@ -171,7 +155,7 @@ static void pnv_phb_realize(DeviceState *dev, Error **errp)
           return;
       }
-    pnv_phb_attach_root_port(pci);
+    pnv_phb_attach_root_port(pci->bus);
   }
   static const char *pnv_phb_root_bus_path(PCIHostState *host_bridge,
@@ -240,12 +224,18 @@ static void pnv_phb_root_port_realize(DeviceState *dev, Error **errp)
   {
       PCIERootPortClass *rpc = PCIE_ROOT_PORT_GET_CLASS(dev);
       PnvPHBRootPort *phb_rp = PNV_PHB_ROOT_PORT(dev);
-    PCIBus *bus = PCI_BUS(qdev_get_parent_bus(dev));
+    BusState *qbus = qdev_get_parent_bus(dev);
+    PCIBus *bus = PCI_BUS(qbus);
       PCIDevice *pci = PCI_DEVICE(dev);
       uint16_t device_id = 0;
       Error *local_err = NULL;
       int chip_id, index;
+    /*
+     * 'index' will be used both as a PCIE slot value and to calculate
+     * QOM id. 'chip_id' is going to be used as PCIE chassis for the
+     * root port.
+     */
       chip_id = object_property_get_int(OBJECT(bus), "chip-id", &error_fatal);
       index = object_property_get_int(OBJECT(bus), "phb-id", &error_fatal);
Quoted 8 lines

hw/pci-host/pnv_phb.c static void pnv_phb_root_port_realize(DeviceState *dev, Error **errp) · line 246

Quoted 5 lines
hw/pci-host/pnv_phb.c +2 -0
@@ -253,6 +243,11 @@ static void pnv_phb_root_port_realize(DeviceState *dev, Error **errp)
       qdev_prop_set_uint8(dev, "chassis", chip_id);
       qdev_prop_set_uint16(dev, "slot", index);
+    pnv_parent_qom_fixup(OBJECT(bus), OBJECT(dev), index);
+    if (!qdev_set_parent_bus(dev, qbus, &error_fatal)) {

That line looks surprising at first, because we got qbus from qdev_get_parent_bus() just above, so it looks like a noop. I talked to Daniel about it: the device<->bus relationship is correct when entering the function but the call to pnv_parent_qom_fixup() interferes with the bus relationship, so it needs to be re-established.

The more detailed version: when adding a child property
using object_property_add_child(), object_property_try_add_child() is called. This function calls internally a object_property_try_add() function with lots of parameters, including a release callback named 'object_finalize_child_property'.

This release callback is implemented like this:

static void object_finalize_child_property(Object *obj, const char *name,
void *opaque)

{
    Object *child = opaque;

    if (child->class->unparent) {
        (child->class->unparent)(child);
    }
    child->parent = NULL;
    object_unref(child);
}

If you're adding a device as a child, which is our case here, child->class->unparent is device_unparent(). Note that device_unparent() removes the device from its parent_bus:

static void device_unparent(Object *obj)

{
    DeviceState *dev = DEVICE(obj);
    BusState *bus;

    if (dev->realized) {
        qdev_unrealize(dev);
    }
    while (dev->num_child_bus) {
        bus = QLIST_FIRST(&dev->child_bus);
        object_unparent(OBJECT(bus));
    }
    if (dev->parent_bus) {
        bus_remove_child(dev->parent_bus, dev);
        object_unref(OBJECT(dev->parent_bus));
        dev->parent_bus = NULL;
    }
}

So, back in our qom_fixup() helper, we're doing an object_unparent(). This function calls object_property_del_child(), which in turn calls the property release callback prop->release if it's defined. It is defined, and in our case it's the object_finalize_child_property() from above that will end up calling device_unparent(), which will remove the device from the bus.

This is why I'm having to do a qdev_set_parent_bus() to assign the root port to the same parent bus it had before. We want to fixup it's QOM parent without changing its parent bus.

Also, note that we're doing the same thing for the user created pnv-phbs in pnv_phb_user_device_init():

    pnv_parent_qom_fixup(parent, OBJECT(phb), phb->phb_id);
    pnv_parent_bus_fixup(DEVICE(chip), DEVICE(phb));

The difference here is that the QOM parent is not the same as the parent bus, so it's less apparent what we're doing here.

All that said, since we're always calling these 2 fixup helpers together, I think it's a good idea to create a single helper that handles both the QOM parenting and the parent bus. I can also add some comments about what I've said here to explain why we need to set the parent bus after doing an object_unparent().

Quoted 1 line

Short of a better suggestion, it probably need a comment.

I don't think we have wiggle room since we're dealing with deep QOM and qdev internals in this case. I'm all ears for suggestions though.

Thanks,

Daniel

Quoted 11 lines

  Fred

Quoted 6 lines
hw/pci-host/pnv_phb.c +3 -0
+        return;
+    }
+
       rpc->parent_realize(dev, &local_err);
       if (local_err) {
           error_propagate(errp, local_err);