[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250520164339.GC365796@horms.kernel.org>
Date: Tue, 20 May 2025 17:43:39 +0100
From: Simon Horman <horms@...nel.org>
To: Hariprasad Kelam <hkelam@...vell.com>
Cc: netdev@...r.kernel.org, linux-kernel@...r.kernel.org,
Sunil Goutham <sgoutham@...vell.com>,
Geetha sowjanya <gakula@...vell.com>,
Subbaraya Sundeep <sbhatta@...vell.com>,
Bharat Bhushan <bbhushan2@...vell.com>,
Andrew Lunn <andrew+netdev@...n.ch>,
"David S. Miller" <davem@...emloft.net>,
Eric Dumazet <edumazet@...gle.com>,
Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>,
Naveen Mamindlapalli <naveenm@...vell.com>
Subject: Re: [net] octeontx2-pf: QOS: Fix HTB queue deletion on reboot
On Tue, May 20, 2025 at 01:05:23PM +0530, Hariprasad Kelam wrote:
> During a system reboot, the interface receives TC_HTB_LEAF_DEL
> and TC_HTB_LEAF_DEL_LAST callbacks to delete its HTB queues.
> In the case of TC_HTB_LEAF_DEL_LAST, although the same send queue
> is reassigned to the parent, the current logic still attempts to update
> the real number of queues, leadning to below warnings
>
> New queues can't be registered after device unregistration.
> WARNING: CPU: 0 PID: 6475 at net/core/net-sysfs.c:1714
> netdev_queue_update_kobjects+0x1e4/0x200
>
> Fixes: 5e6808b4c68d ("octeontx2-pf: Add support for HTB offload")
> Signed-off-by: Hariprasad Kelam <hkelam@...vell.com>
> ---
> drivers/net/ethernet/marvell/octeontx2/nic/qos.c | 4 +---
> 1 file changed, 1 insertion(+), 3 deletions(-)
>
> diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/qos.c b/drivers/net/ethernet/marvell/octeontx2/nic/qos.c
> index 35acc07bd964..5765bac119f0 100644
> --- a/drivers/net/ethernet/marvell/octeontx2/nic/qos.c
> +++ b/drivers/net/ethernet/marvell/octeontx2/nic/qos.c
> @@ -1638,6 +1638,7 @@ static int otx2_qos_leaf_del_last(struct otx2_nic *pfvf, u16 classid, bool force
> if (!node->is_static)
> dwrr_del_node = true;
>
> + WRITE_ONCE(node->qid, OTX2_QOS_QID_INNER);
Hi Hariprasad,
Perhaps a comment is warranted regarding the line above.
It would probably be more valuable than the one on the line below.
> /* destroy the leaf node */
> otx2_qos_disable_sq(pfvf, qid);
> otx2_qos_destroy_node(pfvf, node);
> @@ -1682,9 +1683,6 @@ static int otx2_qos_leaf_del_last(struct otx2_nic *pfvf, u16 classid, bool force
> }
> kfree(new_cfg);
>
> - /* update tx_real_queues */
> - otx2_qos_update_tx_netdev_queues(pfvf);
> -
> return 0;
> }
>
> --
> 2.34.1
>
>
Powered by blists - more mailing lists