Re: [PATCH V2 net] net: mana: Fix error handling in mana_create_txq/rxq's NAPI cleanup

[Shradha Gupta]

On Fri, Aug 23, 2024 at 02:44:29AM -0700, Souradeep Chakrabarti wrote:
> Currently napi_disable() gets called during rxq and txq cleanup,
> even before napi is enabled and hrtimer is initialized. It causes
> kernel panic.
> 
> ? page_fault_oops+0x136/0x2b0
>   ? page_counter_cancel+0x2e/0x80
>   ? do_user_addr_fault+0x2f2/0x640
>   ? refill_obj_stock+0xc4/0x110
>   ? exc_page_fault+0x71/0x160
>   ? asm_exc_page_fault+0x27/0x30
>   ? __mmdrop+0x10/0x180
>   ? __mmdrop+0xec/0x180
>   ? hrtimer_active+0xd/0x50
>   hrtimer_try_to_cancel+0x2c/0xf0
>   hrtimer_cancel+0x15/0x30
>   napi_disable+0x65/0x90
>   mana_destroy_rxq+0x4c/0x2f0
>   mana_create_rxq.isra.0+0x56c/0x6d0
>   ? mana_uncfg_vport+0x50/0x50
>   mana_alloc_queues+0x21b/0x320
>   ? skb_dequeue+0x5f/0x80
> 
> Fixes: e1b5683ff62e ("net: mana: Move NAPI from EQ to CQ")
> Signed-off-by: Souradeep Chakrabarti <schakrabarti@xxxxxxxxxxxxxxxxxxx>
> ---
> V2 -> V1:
> Addressed the comment on cleaning up napi for the queues,
> where queue creation was successful.
> ---
>  drivers/net/ethernet/microsoft/mana/mana_en.c | 22 +++++++++++--------
>  1 file changed, 13 insertions(+), 9 deletions(-)
> 
> diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c
> index 39f56973746d..7448085fd49e 100644
> --- a/drivers/net/ethernet/microsoft/mana/mana_en.c
> +++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
> @@ -1872,10 +1872,11 @@ static void mana_destroy_txq(struct mana_port_context *apc)
>  
>  	for (i = 0; i < apc->num_queues; i++) {
>  		napi = &apc->tx_qp[i].tx_cq.napi;
> -		napi_synchronize(napi);
> -		napi_disable(napi);
> -		netif_napi_del(napi);
> -
> +		if (napi->dev == apc->ndev) {
> +			napi_synchronize(napi);
> +			napi_disable(napi);
> +			netif_napi_del(napi);
> +		}
>  		mana_destroy_wq_obj(apc, GDMA_SQ, apc->tx_qp[i].tx_object);
>  
>  		mana_deinit_cq(apc, &apc->tx_qp[i].tx_cq);
> @@ -2023,14 +2024,17 @@ static void mana_destroy_rxq(struct mana_port_context *apc,
>  
>  	napi = &rxq->rx_cq.napi;
>  
> -	if (validate_state)
> -		napi_synchronize(napi);
> +	if (napi->dev == apc->ndev) {
>  
> -	napi_disable(napi);
> +		if (validate_state)
> +			napi_synchronize(napi);
>  
> -	xdp_rxq_info_unreg(&rxq->xdp_rxq);
> +		napi_disable(napi);
>  
> -	netif_napi_del(napi);
> +		netif_napi_del(napi);
> +	}
> +
> +	xdp_rxq_info_unreg(&rxq->xdp_rxq);
>  
>  	mana_destroy_wq_obj(apc, GDMA_RQ, rxq->rxobj);

Reviewed-by: Shradha Gupta <shradhagupta@xxxxxxxxxxxxxxxxxxx>
>  
> -- 
> 2.34.1
> 
>