Re: [PATCH v6] PCI: hv: Make sure the bus domain is really unique
From: Bjorn Helgaas
Date: Wed Apr 11 2018 - 16:16:45 EST
I would make the subject:
Revert "PCI: hv: Use device serial number as PCI domain"
so it matches the original commit. Otherwise you really need quite a
lot of driver internal knowledge to connect them.
On Wed, Apr 11, 2018 at 12:36:25AM +0000, Sridhar Pitchai wrote:
> When Linux runs as a guest VM in Hyper-V and Hyper-V adds the virtual
> PCI bus to the guest, Hyper-V always provides unique PCI domain.
>
> commit 4a9b0933bdfc ("PCI: hv: Use device serial number as PCI domain")
> overrode unique domain with the serial number of the first device added
> to the virtual PCI bus. The reason for that patch is to have a consistent
> and short name for the device. But commit 4a9b0933bdfc ("PCI: hv: Use
> device serial number as PCI domain") will not guarantee unique domain id.
Possible alternate text:
The reason for that patch was to have a consistent and short name
for the device, but Hyper-V doesn't provide unique serial numbers
(in spite of the code comment claiming that it does). Using
non-unique serial numbers as domain IDs leads to duplicate device
addresses, which causes PCI bus registration to fail.
> For example, if the serial number of the device is 0 and there exists a
> PCI bus with domain 0 already, this will cause the PCI bus registration
> with kernel fails.
>
> commit 4a9b0933bdfc ("PCI: hv: Use device serial number as PCI
> domain") need to be reverted.
I think the above sentence is redundant and could be removed.
> Revert commit 4a9b0933bdfc ("PCI: hv: Use device serial number as PCI
> domain") so we can reliably support multiple devices being assigned to
> a guest.
>
> Fixes: 4a9b0933bdfc ("PCI: hv: Use device serial number as PCI domain")
> Signed-off-by: Sridhar Pitchai <sridhar.pitchai@xxxxxxxxxxxxx>
> Cc: stable@xxxxxxxxxxxxxxx
Regardless of what you do with the feedback above,
Reviewed-by: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>
Thanks a lot for persisting with this!
> ---
> Changes in v6:
> * fix the commit comment. [Lorenzo Pieralisi, Bjorn Helgaas]
> ---
> drivers/pci/host/pci-hyperv.c | 11 -----------
> 1 file changed, 11 deletions(-)
>
> diff --git a/drivers/pci/host/pci-hyperv.c b/drivers/pci/host/pci-hyperv.c
> index 2faf38eab785..ac67e56e451a 100644
> --- a/drivers/pci/host/pci-hyperv.c
> +++ b/drivers/pci/host/pci-hyperv.c
> @@ -1518,17 +1518,6 @@ static struct hv_pci_dev *new_pcichild_device(struct hv_pcibus_device *hbus,
> get_pcichild(hpdev, hv_pcidev_ref_childlist);
> spin_lock_irqsave(&hbus->device_list_lock, flags);
>
> - /*
> - * When a device is being added to the bus, we set the PCI domain
> - * number to be the device serial number, which is non-zero and
> - * unique on the same VM. The serial numbers start with 1, and
> - * increase by 1 for each device. So device names including this
> - * can have shorter names than based on the bus instance UUID.
> - * Only the first device serial number is used for domain, so the
> - * domain number will not change after the first device is added.
> - */
> - if (list_empty(&hbus->children))
> - hbus->sysdata.domain = desc->ser;
> list_add_tail(&hpdev->list_entry, &hbus->children);
> spin_unlock_irqrestore(&hbus->device_list_lock, flags);
> return hpdev;
> --
> 2.14.1
>