On 12/30/24 00:26, Alex Williamson wrote:
> On Sun, 29 Dec 2024 23:51:40 +0800
> Tomita Moeko <tomitamoeko@xxxxxxxxx> wrote:
>
>> igd device can either expose as a VGA controller or display controller
>> depending on whether it is configured as the primary display device in
>> BIOS. In both cases, the OpRegion may be present. Also checks if the
>> device is at bdf 00:02.0 to avoid setting up igd-specific regions on
>> Intel discrete GPUs.
>>
>> Signed-off-by: Tomita Moeko <tomitamoeko@xxxxxxxxx>
>> ---
>> drivers/vfio/pci/vfio_pci.c | 6 ++++--
>> 1 file changed, 4 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/vfio/pci/vfio_pci.c b/drivers/vfio/pci/vfio_pci.c
>> index e727941f589d..051ef4ad3f43 100644
>> --- a/drivers/vfio/pci/vfio_pci.c
>> +++ b/drivers/vfio/pci/vfio_pci.c
>> @@ -111,9 +111,11 @@ static int vfio_pci_open_device(struct vfio_device *core_vdev)
>> if (ret)
>> return ret;
>>
>> - if (vfio_pci_is_vga(pdev) &&
>> + if (IS_ENABLED(CONFIG_VFIO_PCI_IGD) &&
>> pdev->vendor == PCI_VENDOR_ID_INTEL &&
>> - IS_ENABLED(CONFIG_VFIO_PCI_IGD)) {
>> + ((pdev->class >> 8) == PCI_CLASS_DISPLAY_VGA ||
>
> The above is vfio_pci_is_vga(pdev), maybe below should have a similar
> helper.
There isn't, shall I create a new helper function? or just keep it as
the match only happens here.
>> + (pdev->class >> 8) == PCI_CLASS_DISPLAY_OTHER) &&
>> + pdev == pci_get_domain_bus_and_slot(0, 0, PCI_DEVFN(2, 0))) {
>
> This increments the reference count on the device:
>
> * Given a PCI domain, bus, and slot/function number, the desired PCI
> * device is located in the list of PCI devices. If the device is
> * found, its reference count is increased and this function returns a
> * pointer to its data structure. The caller must decrement the
> * reference count by calling pci_dev_put().
Sorry I missed that, will fix in v2.
>> ret = vfio_pci_igd_init(vdev);
>> if (ret && ret != -ENODEV) {
>> pci_warn(pdev, "Failed to setup Intel IGD regions\n");
>
