[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-devel] [PULL v3 40/44] spapr_pci: perform unplug via the hotplug h
From: |
Michael S. Tsirkin |
Subject: |
[Qemu-devel] [PULL v3 40/44] spapr_pci: perform unplug via the hotplug handler |
Date: |
Thu, 20 Dec 2018 13:40:10 -0500 |
From: David Hildenbrand <address@hidden>
Introduce and use the "unplug" callback.
This is a preparation for multi-stage hotplug handlers, whereby the bus
hotplug handler is overwritten by the machine hotplug handler. This handler
will then pass control to the bus hotplug handler. So to get this running
cleanly, we also have to make sure to go via the hotplug handler chain when
actually unplugging a device after an unplug request. Lookup the hotplug
handler and call "unplug".
Reviewed-by: Greg Kurz <address@hidden>
Reviewed-by: Igor Mammedov <address@hidden>
Acked-by: David Gibson <address@hidden>
Signed-off-by: David Hildenbrand <address@hidden>
Reviewed-by: Michael S. Tsirkin <address@hidden>
Signed-off-by: Michael S. Tsirkin <address@hidden>
---
hw/ppc/spapr_pci.c | 33 +++++++++++++++++++++------------
1 file changed, 21 insertions(+), 12 deletions(-)
diff --git a/hw/ppc/spapr_pci.c b/hw/ppc/spapr_pci.c
index 2374d55fc1..bfb02ee96b 100644
--- a/hw/ppc/spapr_pci.c
+++ b/hw/ppc/spapr_pci.c
@@ -1370,18 +1370,9 @@ static int spapr_create_pci_child_dt(sPAPRPHBState *phb,
PCIDevice *dev,
/* Callback to be called during DRC release. */
void spapr_phb_remove_pci_device_cb(DeviceState *dev)
{
- /* some version guests do not wait for completion of a device
- * cleanup (generally done asynchronously by the kernel) before
- * signaling to QEMU that the device is safe, but instead sleep
- * for some 'safe' period of time. unfortunately on a busy host
- * this sleep isn't guaranteed to be long enough, resulting in
- * bad things like IRQ lines being left asserted during final
- * device removal. to deal with this we call reset just prior
- * to finalizing the device, which will put the device back into
- * an 'idle' state, as the device cleanup code expects.
- */
- pci_device_reset(PCI_DEVICE(dev));
- object_unparent(OBJECT(dev));
+ HotplugHandler *hotplug_ctrl = qdev_get_hotplug_handler(dev);
+
+ hotplug_handler_unplug(hotplug_ctrl, dev, &error_abort);
}
static sPAPRDRConnector *spapr_phb_get_pci_func_drc(sPAPRPHBState *phb,
@@ -1490,6 +1481,23 @@ out:
}
}
+static void spapr_pci_unplug(HotplugHandler *plug_handler,
+ DeviceState *plugged_dev, Error **errp)
+{
+ /* some version guests do not wait for completion of a device
+ * cleanup (generally done asynchronously by the kernel) before
+ * signaling to QEMU that the device is safe, but instead sleep
+ * for some 'safe' period of time. unfortunately on a busy host
+ * this sleep isn't guaranteed to be long enough, resulting in
+ * bad things like IRQ lines being left asserted during final
+ * device removal. to deal with this we call reset just prior
+ * to finalizing the device, which will put the device back into
+ * an 'idle' state, as the device cleanup code expects.
+ */
+ pci_device_reset(PCI_DEVICE(plugged_dev));
+ object_unparent(OBJECT(plugged_dev));
+}
+
static void spapr_pci_unplug_request(HotplugHandler *plug_handler,
DeviceState *plugged_dev, Error **errp)
{
@@ -1965,6 +1973,7 @@ static void spapr_phb_class_init(ObjectClass *klass, void
*data)
dc->user_creatable = true;
set_bit(DEVICE_CATEGORY_BRIDGE, dc->categories);
hp->plug = spapr_pci_plug;
+ hp->unplug = spapr_pci_unplug;
hp->unplug_request = spapr_pci_unplug_request;
}
--
MST
- [Qemu-devel] [PULL v3 29/44] hw: acpi: Remove AcpiRsdpDescriptor and fix tests, (continued)
- [Qemu-devel] [PULL v3 29/44] hw: acpi: Remove AcpiRsdpDescriptor and fix tests, Michael S. Tsirkin, 2018/12/20
- [Qemu-devel] [PULL v3 25/44] hw: arm: Carry RSDP specific data through AcpiRsdpData, Michael S. Tsirkin, 2018/12/20
- [Qemu-devel] [PULL v3 33/44] s390x/pci: rename hotplug handler callbacks, Michael S. Tsirkin, 2018/12/20
- [Qemu-devel] [PULL v3 26/44] hw: arm: Convert the RSDP build to the buid_append_foo() API, Michael S. Tsirkin, 2018/12/20
- [Qemu-devel] [PULL v3 37/44] pci/pcie: perform unplug via the hotplug handler, Michael S. Tsirkin, 2018/12/20
- [Qemu-devel] [PULL v3 36/44] pci/pcihp: perform unplug via the hotplug handler, Michael S. Tsirkin, 2018/12/20
- [Qemu-devel] [PULL v3 30/44] hw/i386: Remove deprecated machines pc-0.10 and pc-0.11, Michael S. Tsirkin, 2018/12/20
- [Qemu-devel] [PULL v3 35/44] pci/pcihp: overwrite hotplug handler recursively from the start, Michael S. Tsirkin, 2018/12/20
- [Qemu-devel] [PULL v3 34/44] pci/pcihp: perform check for bus capability in pre_plug handler, Michael S. Tsirkin, 2018/12/20
- [Qemu-devel] [PULL v3 39/44] pci/shpc: perform unplug via the hotplug handler, Michael S. Tsirkin, 2018/12/20
- [Qemu-devel] [PULL v3 40/44] spapr_pci: perform unplug via the hotplug handler,
Michael S. Tsirkin <=
- [Qemu-devel] [PULL v3 38/44] pci: Reuse pci-bridge hotplug handler handlers for pcie-pci-bridge, Michael S. Tsirkin, 2018/12/20
- [Qemu-devel] [PULL v3 41/44] pci: Adjust PCI config limit based on bus topology, Michael S. Tsirkin, 2018/12/20
- [Qemu-devel] [PULL v3 44/44] x86-iommu: turn on IR by default if proper, Michael S. Tsirkin, 2018/12/20
- [Qemu-devel] [PULL v3 43/44] x86-iommu: switch intr_supported to OnOffAuto type, Michael S. Tsirkin, 2018/12/20
- [Qemu-devel] [PULL v3 42/44] q35: set split kernel irqchip as default, Michael S. Tsirkin, 2018/12/20
- Re: [Qemu-devel] [PULL v3 00/44] pci, pc, virtio: fixes, features, Peter Maydell, 2018/12/21
- Re: [Qemu-devel] [PULL v3 00/44] pci, pc, virtio: fixes, features, no-reply, 2018/12/26