[Qemu-devel] Re: [PATCH] pci: allow hotplug removal of cold-plugged devices

[Cam Macdonell <cam@cs.ualberta.ca>]

On Sun, Nov 14, 2010 at 7:18 AM, Michael S. Tsirkin <mst@redhat.com> wrote:
> This patch fixes 5beb8ad503c88a76f2b8106c3b74b4ce485a60e1
> which broke hotplug removal of cold plugged devices:
>
> - pass addition/removal state to hotplug callbacks
> - use that in piix and pcie
>
> This also fixes an assert on hotplug removal of coldplugged
> express devices.
>
> Reported-by: by Cam Macdonell <cam@cs.ualberta.ca>.
> Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

Acked-by: Cam Macdonell <cam@cs.ualberta.ca>

> ---
>
> So I think the below would be the cleanest way
> to fix the bug as we keep the hot/cold plug logic
> in a central palce. Untested. Comments? Cam?

Yes, it seems to fix the problem.

>
>
> diff --git a/hw/acpi_piix4.c b/hw/acpi_piix4.c
> index 66c7885..f549089 100644
> --- a/hw/acpi_piix4.c
> +++ b/hw/acpi_piix4.c
> @@ -585,7 +585,8 @@ static void pciej_write(void *opaque, uint32_t addr, uint32_t val)
>     PIIX4_DPRINTF("pciej write %x <== %d\n", addr, val);
>  }
>
> -static int piix4_device_hotplug(DeviceState *qdev, PCIDevice *dev, int state);
> +static int piix4_device_hotplug(DeviceState *qdev, PCIDevice *dev,
> +                                PCIHotplugState state);
>
>  static void piix4_acpi_system_hot_add_init(PCIBus *bus, PIIX4PMState *s)
>  {
> @@ -615,18 +616,23 @@ static void disable_device(PIIX4PMState *s, int slot)
>     s->pci0_status.down |= (1 << slot);
>  }
>
> -static int piix4_device_hotplug(DeviceState *qdev, PCIDevice *dev, int state)
> +static int piix4_device_hotplug(DeviceState *qdev, PCIDevice *dev,
> +                               PCIHotplugState state)
>  {
>     int slot = PCI_SLOT(dev->devfn);
>     PIIX4PMState *s = DO_UPCAST(PIIX4PMState, dev,
>                                 DO_UPCAST(PCIDevice, qdev, qdev));
>
> -    if (!dev->qdev.hotplugged)
> +    /* Don't send event when device is enabled during qemu machine creation:
> +     * it is present on boot, no hotplug event is necessary. We do send an
> +     * event when the device is disabled later. */
> +    if (state == PCI_COLDPLUG_ENABLED) {
>         return 0;
> +    }
>
>     s->pci0_status.up = 0;
>     s->pci0_status.down = 0;
> -    if (state) {
> +    if (state == PCI_HOTPLUG_ENABLED) {
>         enable_device(s, slot);
>     } else {
>         disable_device(s, slot);
> diff --git a/hw/pci.c b/hw/pci.c
> index 30e1603..316b24f 100644
> --- a/hw/pci.c
> +++ b/hw/pci.c
> @@ -1566,8 +1566,11 @@ static int pci_qdev_init(DeviceState *qdev, DeviceInfo *base)
>     pci_add_option_rom(pci_dev);
>
>     if (bus->hotplug) {
> -        /* lower layer must check qdev->hotplugged */
> -        rc = bus->hotplug(bus->hotplug_qdev, pci_dev, 1);
> +        /* Let buses differentiate between hotplug and when device is
> +         * enabled during qemu machine creation. */
> +        rc = bus->hotplug(bus->hotplug_qdev, pci_dev,
> +                          qdev->hotplugged ? PCI_HOTPLUG_ENABLED:
> +                          PCI_COLDPLUG_ENABLED);
>         if (rc != 0) {
>             int r = pci_unregister_device(&pci_dev->qdev);
>             assert(!r);
> @@ -1581,7 +1584,8 @@ static int pci_unplug_device(DeviceState *qdev)
>  {
>     PCIDevice *dev = DO_UPCAST(PCIDevice, qdev, qdev);
>
> -    return dev->bus->hotplug(dev->bus->hotplug_qdev, dev, 0);
> +    return dev->bus->hotplug(dev->bus->hotplug_qdev, dev,
> +                             PCI_HOTPLUG_DISABLED);
>  }
>
>  void pci_qdev_register(PCIDeviceInfo *info)
> diff --git a/hw/pci.h b/hw/pci.h
> index 7100804..09b3e4c 100644
> --- a/hw/pci.h
> +++ b/hw/pci.h
> @@ -214,7 +214,15 @@ int pci_device_load(PCIDevice *s, QEMUFile *f);
>
>  typedef void (*pci_set_irq_fn)(void *opaque, int irq_num, int level);
>  typedef int (*pci_map_irq_fn)(PCIDevice *pci_dev, int irq_num);
> -typedef int (*pci_hotplug_fn)(DeviceState *qdev, PCIDevice *pci_dev, int state);
> +
> +typedef enum {
> +    PCI_HOTPLUG_DISABLED,
> +    PCI_HOTPLUG_ENABLED,
> +    PCI_COLDPLUG_ENABLED,
> +} PCIHotplugState;
> +
> +typedef int (*pci_hotplug_fn)(DeviceState *qdev, PCIDevice *pci_dev,
> +                              PCIHotplugState state);
>  void pci_bus_new_inplace(PCIBus *bus, DeviceState *parent,
>                          const char *name, int devfn_min);
>  PCIBus *pci_bus_new(DeviceState *parent, const char *name, int devfn_min);
> diff --git a/hw/pcie.c b/hw/pcie.c
> index 35918f7..4df48b8 100644
> --- a/hw/pcie.c
> +++ b/hw/pcie.c
> @@ -192,14 +192,16 @@ static void pcie_cap_slot_event(PCIDevice *dev, PCIExpressHotPlugEvent event)
>  }
>
>  static int pcie_cap_slot_hotplug(DeviceState *qdev,
> -                                 PCIDevice *pci_dev, int state)
> +                                 PCIDevice *pci_dev, PCIHotplugState state)
>  {
>     PCIDevice *d = DO_UPCAST(PCIDevice, qdev, qdev);
>     uint8_t *exp_cap = d->config + d->exp.exp_cap;
>     uint16_t sltsta = pci_get_word(exp_cap + PCI_EXP_SLTSTA);
>
> -    if (!pci_dev->qdev.hotplugged) {
> -        assert(state); /* this case only happens at machine creation. */
> +    /* Don't send event when device is enabled during qemu machine creation:
> +     * it is present on boot, no hotplug event is necessary. We do send an
> +     * event when the device is disabled later. */
> +    if (state == PCI_COLDPLUG_ENABLED) {
>         pci_word_test_and_set_mask(exp_cap + PCI_EXP_SLTSTA,
>                                    PCI_EXP_SLTSTA_PDS);
>         return 0;
>
>