RE: [Intel-wired-lan] [PATCH net-next v2 08/13] bnxt: use snapshot in bnxt_cfg_rx_mode

[Loktionov, Aleksandr]

> -----Original Message-----
> From: Intel-wired-lan <intel-wired-lan-bounces@xxxxxxxxxx> On Behalf
> Of Stanislav Fomichev
> Sent: Wednesday, March 18, 2026 4:03 PM
> To: netdev@xxxxxxxxxxxxxxx
> Cc: davem@xxxxxxxxxxxxx; edumazet@xxxxxxxxxx; kuba@xxxxxxxxxx;
> pabeni@xxxxxxxxxx; horms@xxxxxxxxxx; corbet@xxxxxxx;
> skhan@xxxxxxxxxxxxxxxxxxx; andrew+netdev@xxxxxxx;
> michael.chan@xxxxxxxxxxxx; pavan.chebbi@xxxxxxxxxxxx; Nguyen, Anthony
> L <anthony.l.nguyen@xxxxxxxxx>; Kitszel, Przemyslaw
> <przemyslaw.kitszel@xxxxxxxxx>; saeedm@xxxxxxxxxx; tariqt@xxxxxxxxxx;
> mbloch@xxxxxxxxxx; alexanderduyck@xxxxxx; kernel-team@xxxxxxxx;
> johannes@xxxxxxxxxxxxxxxx; sd@xxxxxxxxxxxxxxx; jianbol@xxxxxxxxxx;
> dtatulea@xxxxxxxxxx; sdf@xxxxxxxxxxx; mohsin.bashr@xxxxxxxxx; Keller,
> Jacob E <jacob.e.keller@xxxxxxxxx>; willemb@xxxxxxxxxx;
> skhawaja@xxxxxxxxxx; bestswngs@xxxxxxxxx; linux-doc@xxxxxxxxxxxxxxx;
> linux-kernel@xxxxxxxxxxxxxxx; intel-wired-lan@xxxxxxxxxxxxxxxx; linux-
> rdma@xxxxxxxxxxxxxxx; linux-wireless@xxxxxxxxxxxxxxx; linux-
> kselftest@xxxxxxxxxxxxxxx; leon@xxxxxxxxxx
> Subject: [Intel-wired-lan] [PATCH net-next v2 08/13] bnxt: use
> snapshot in bnxt_cfg_rx_mode
> 
> With the introduction of ndo_set_rx_mode_async (as discussed in [0])
> we can call bnxt_cfg_rx_mode directly. Convert bnxt_cfg_rx_mode to use
> uc/mc snapshots and move its call in bnxt_sp_task to the section that
> resets BNXT_STATE_IN_SP_TASK. Switch to direct call in
> bnxt_set_rx_mode.
> 
> 0:
> https://lore.kernel.org/netdev/CACKFLi=5vj8hPqEUKDd8RTw3au5G+zRgQEqjF+
> 6NZnyoNm90KA@xxxxxxxxxxxxxx/
> 
> Cc: Michael Chan <michael.chan@xxxxxxxxxxxx>
> Cc: Pavan Chebbi <pavan.chebbi@xxxxxxxxxxxx>
> Signed-off-by: Stanislav Fomichev <sdf@xxxxxxxxxxx>
> ---
>  drivers/net/ethernet/broadcom/bnxt/bnxt.c | 24 ++++++++++++++--------
> -
>  1 file changed, 15 insertions(+), 9 deletions(-)

...

> -static int bnxt_cfg_rx_mode(struct bnxt *bp)
> +static int bnxt_cfg_rx_mode(struct bnxt *bp, struct
> netdev_hw_addr_list *uc,
> +			    struct netdev_hw_addr_list *mc)
>  {
>  	struct net_device *dev = bp->dev;
>  	struct bnxt_vnic_info *vnic = &bp-
> >vnic_info[BNXT_VNIC_DEFAULT];
> @@ -13623,7 +13625,7 @@ static int bnxt_cfg_rx_mode(struct bnxt *bp)
>  	bool uc_update;
> 
>  	netif_addr_lock_bh(dev);
> -	uc_update = bnxt_uc_list_updated(bp, &dev->uc);
> +	uc_update = bnxt_uc_list_updated(bp, uc);
>  	netif_addr_unlock_bh(dev);
> 
>  	if (!uc_update)
> @@ -13642,7 +13644,7 @@ static int bnxt_cfg_rx_mode(struct bnxt *bp)
>  	if (netdev_uc_count(dev) > (BNXT_MAX_UC_ADDRS - 1)) {
>  		vnic->rx_mask |=
> CFA_L2_SET_RX_MASK_REQ_MASK_PROMISCUOUS;
This limit check uses the live device list, dev->uc.
In the new async model, the live list can differ from the snapshot.

>  	} else {
> -		netdev_for_each_uc_addr(ha, dev) {
> +		netdev_hw_addr_list_for_each(ha, uc) {
This loop iterates the snapshot list, uc. */
So, the guard above and the loop below are checking different data.

>  			memcpy(vnic->uc_list + off, ha->addr, ETH_ALEN);
>  			off += ETH_ALEN;
>  			vnic->uc_filter_count++;
> @@ -14600,6 +14602,7 @@ static void bnxt_ulp_restart(struct bnxt *bp)
> static void bnxt_sp_task(struct work_struct *work)  {
>  	struct bnxt *bp = container_of(work, struct bnxt, sp_task);
> +	struct net_device *dev = bp->dev;
> 
>  	set_bit(BNXT_STATE_IN_SP_TASK, &bp->state);
>  	smp_mb__after_atomic();
> @@ -14613,9 +14616,6 @@ static void bnxt_sp_task(struct work_struct
> *work)
>  		bnxt_reenable_sriov(bp);
>  	}

...

> 2.53.0