Re: [PATCH v1] net: liquidio: resolve VF pci_dev on demand for FLR requests

[Simon Horman <horms@kernel.org>]

On Sun, Apr 19, 2026 at 10:33:04PM -0400, Yuho Choi wrote:
> The PF SR-IOV enable path caches VF pci_dev pointers in
> dpiring_to_vfpcidev_lut[] by iterating with pci_get_device(). Those
> entries do not own a reference, because the iterator drops the previous
> device reference on each step. The cached pointer is then dereferenced
> later when handling OCTEON_VF_FLR_REQUEST.
> 
> This can leave stale VF pci_dev pointers in the lookup table and makes
> the FLR path rely on a PCI device object whose lifetime is not pinned.
> 
> Drop the long-lived lookup table and resolve the VF pci_dev only when an
> FLR request arrives. Use the PF's SR-IOV metadata to derive the VF's
> bus/devfn, get a referenced pci_dev for immediate use, issue the FLR,
> and then drop the reference.
> 
> Fixes: ca6139ffc67ee ("liquidio CN23XX: sysfs VF config support")
> Fixes: 8c978d059224 ("liquidio CN23XX: Mailbox support")
> Co-developed-by: Myeonghun Pak <mhun512@gmail.com>
> Signed-off-by: Myeonghun Pak <mhun512@gmail.com>
> Co-developed-by: Ijae Kim <ae878000@gmail.com>
> Signed-off-by: Ijae Kim <ae878000@gmail.com>
> Co-developed-by: Taegyu Kim <tmk5904@psu.edu>
> Signed-off-by: Taegyu Kim <tmk5904@psu.edu>
> Signed-off-by: Yuho Choi <dbgh9129@gmail.com>

As this fixes code present in the net tree, it should be targeted
at that tree, like this:

Subject: [PATCH net] ...

In this case the CI defaulted to the net-next tree.
Which might be harmless. But please keep this in mind for next time.

...

> diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.c b/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.c
> index ad685f5d0a136..b967c7928b4a7 100644
> --- a/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.c
> +++ b/drivers/net/ethernet/cavium/liquidio/octeon_mailbox.c
> @@ -26,6 +26,29 @@
>  #include "octeon_mailbox.h"
>  #include "cn23xx_pf_device.h"
>  
> +static struct pci_dev *lio_vf_pci_dev_by_qno(struct octeon_device *oct, u32 q_no)
> +{
> +	int vfidx, bus, devfn;
> +
> +	if (!oct->sriov_info.rings_per_vf)
> +		return NULL;
> +
> +	if (q_no % oct->sriov_info.rings_per_vf)
> +		return NULL;
> +
> +	vfidx = q_no / oct->sriov_info.rings_per_vf;
> +	if (vfidx >= oct->sriov_info.num_vfs_alloced)
> +		return NULL;
> +
> +	bus = pci_iov_virtfn_bus(oct->pci_dev, vfidx);

When applied against net-next this causes a linker error with x86_64
allmodconfig (at least) because pci_iov_virtfn_bus is not defined.

> +	devfn = pci_iov_virtfn_devfn(oct->pci_dev, vfidx);
> +	if (bus < 0 || devfn < 0)
> +		return NULL;
> +
> +	return pci_get_domain_bus_and_slot(pci_domain_nr(oct->pci_dev->bus),
> +					   bus, devfn);
> +}
> +
>  /**
>   * octeon_mbox_read:
>   * @mbox: Pointer mailbox

-- 
pw-bot: changes-requested