Re: [PATCH rdma-next v4 08/10] RDMA/mana_ib: Flush and notify CQs when kernel QPs enter ERR

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Low] rdma/mana_ib: dead code - qp->state is write-only and never read
--

commit 77c71a02dc85cc117e4f58d2a8f85635639fc4ac
Author: Konstantin Taranov <kotaranov@microsoft.com>

    RDMA/mana_ib: Flush and notify CQs when kernel QPs enter ERR
    
    Complete the RC software flush path by adding a kernel QP to the CQ
    error lists when it transitions to the ERR state.
    
    Invoke the CQ handlers after the transition to flush pending work
    requests.

> diff --git a/drivers/infiniband/hw/mana/mana_ib.h b/drivers/infiniband/hw/mana/mana_ib.h
> index 55899b1e2f868..a72bed94c08d7 100644
> --- a/drivers/infiniband/hw/mana/mana_ib.h
> +++ b/drivers/infiniband/hw/mana/mana_ib.h
> @@ -271,6 +271,7 @@ struct mana_ib_qp {
>  	bool pending_mmq_fence;
>  	bool sq_sig_all;
>  	enum ib_mtu mtu;
> +	enum ib_qp_state state;
>  
>  	/* Serializes receive WR posting. */
>  	spinlock_t rq_lock;

[ ... ]

> diff --git a/drivers/infiniband/hw/mana/qp.c b/drivers/infiniband/hw/mana/qp.c
> index 74cf80074e54f..31e91cc657ac7 100644
> --- a/drivers/infiniband/hw/mana/qp.c
> +++ b/drivers/infiniband/hw/mana/qp.c

[ ... ]

> @@ -1049,6 +1071,11 @@ static void mana_ib_modify_qp_state(struct ib_qp *ibqp, struct ib_qp_attr *attr,
>  		default:
>  			break;
>  		}
> +		qp->state = attr->qp_state;

[Severity: Low]
Is this state field ever read? 

It appears to be write-only in the driver, including in subsequent patches in
the series. Could this field be removed to save memory and avoid unnecessary
instructions?

> +		if (attr->qp_state == IB_QPS_ERR) {
> +			mana_add_qp_to_error_cqs(qp);
> +			notify = true;
> +		}
>  	}
>  
>  	if (attr_mask & IB_QP_SQ_PSN) {

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261010065443.3554193-1-kotaranov@linux.microsoft.com?part=8