When a VF (Virtual Function) driver is removed, the PF (Physical Function)
driver continues to retain stale VF-specific information. This can lead to
inconsistencies or unexpected behavior when the VF is re-initialized or
reassigned.

This patch ensures that the PF driver clears the corresponding VF info
when the VF driver is removed, maintaining a clean state and preventing
potential issues.

Fixes: cde29af9e68e ("octeon_ep: add PF-VF mailbox communication")
Signed-off-by: Sathesh B Edara <sedara@marvell.com>
---
Changes:
V2:
  - Commit header format corrected.

 drivers/net/ethernet/marvell/octeon_ep/octep_pfvf_mbox.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_pfvf_mbox.c b/drivers/net/ethernet/marvell/octeon_ep/octep_pfvf_mbox.c
index ebecdd29f3bd..f2759d2073d1 100644
--- a/drivers/net/ethernet/marvell/octeon_ep/octep_pfvf_mbox.c
+++ b/drivers/net/ethernet/marvell/octeon_ep/octep_pfvf_mbox.c
@@ -205,6 +205,8 @@ static void octep_pfvf_dev_remove(struct octep_device *oct,  u32 vf_id,
 {
 	int err;
 
+	/* Reset VF-specific information maintained by the PF */
+	memset(&oct->vf_info[vf_id], 0, sizeof(struct octep_pfvf_info));
 	err = octep_ctrl_net_dev_remove(oct, vf_id);
 	if (err) {
 		rsp->s.type = OCTEP_PFVF_MBOX_TYPE_RSP_NACK;
-- 
2.36.0

On Tue, Sep 16, 2025 at 06:12:25AM -0700, Sathesh B Edara wrote:
> When a VF (Virtual Function) driver is removed, the PF (Physical Function)
> driver continues to retain stale VF-specific information. This can lead to
> inconsistencies or unexpected behavior when the VF is re-initialized or
> reassigned.
> 
> This patch ensures that the PF driver clears the corresponding VF info
> when the VF driver is removed, maintaining a clean state and preventing
> potential issues.
> 
> Fixes: cde29af9e68e ("octeon_ep: add PF-VF mailbox communication")
> Signed-off-by: Sathesh B Edara <sedara@marvell.com>
> ---
> Changes:
> V2:
>   - Commit header format corrected.

Hi,

I feel that I must be missing something terribly obvious.
But this patch seems to be a subset of the one at the link below.

* [net PATCH v2] octeon_ep: fix VF MAC address lifecycle handling
  https://lore.kernel.org/netdev/20250916133207.21737-1-sedara@marvell.com/

> 
>  drivers/net/ethernet/marvell/octeon_ep/octep_pfvf_mbox.c | 2 ++
>  1 file changed, 2 insertions(+)
> 
> diff --git a/drivers/net/ethernet/marvell/octeon_ep/octep_pfvf_mbox.c b/drivers/net/ethernet/marvell/octeon_ep/octep_pfvf_mbox.c
> index ebecdd29f3bd..f2759d2073d1 100644
> --- a/drivers/net/ethernet/marvell/octeon_ep/octep_pfvf_mbox.c
> +++ b/drivers/net/ethernet/marvell/octeon_ep/octep_pfvf_mbox.c
> @@ -205,6 +205,8 @@ static void octep_pfvf_dev_remove(struct octep_device *oct,  u32 vf_id,
>  {
>  	int err;
>  
> +	/* Reset VF-specific information maintained by the PF */
> +	memset(&oct->vf_info[vf_id], 0, sizeof(struct octep_pfvf_info));
>  	err = octep_ctrl_net_dev_remove(oct, vf_id);
>  	if (err) {
>  		rsp->s.type = OCTEP_PFVF_MBOX_TYPE_RSP_NACK;
> -- 
> 2.36.0
> 
>