Re: [PATCH 14/29] ibmvfc: add NVMe/FC Process Login support

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] Self-deadlock on `host_lock` if a management CRQ event fails to send.
--

commit ee31c1669f64561bd4a4e52947529fea4cf6545e
Author: Tyrel Datwyler <tyreld@linux.ibm.com>

ibmvfc: add NVMe/FC Process Login support

This commit extends the PRLI handling code to support NVMe/FC targets.
It adds NVMe-specific service parameters and sets the remote-port roles
upon successful process login. The existing SCSI PRLI flow is preserved.

> diff --git a/drivers/scsi/ibmvscsi/ibmvfc-core.c b/drivers/scsi/ibmvscsi/ibmvfc-core.c
> index 2c54d0b9add49..b45cd0183fb51 100644
> --- a/drivers/scsi/ibmvscsi/ibmvfc-core.c
> +++ b/drivers/scsi/ibmvscsi/ibmvfc-core.c
[ ... ]
> @@ -4172,25 +4190,30 @@ static void ibmvfc_tgt_send_prli(struct ibmvfc_target *tgt)
[ ... ]
>  	ibmvfc_set_tgt_action(tgt, IBMVFC_TGT_ACTION_INIT_WAIT);
>  	if (ibmvfc_send_event(evt, vhost, default_timeout)) {

[Severity: High]
This is a pre-existing issue, but can this call path trigger a deadlock?

When ibmvfc_tgt_send_prli() is called by the target's job step from
ibmvfc_do_work(), the host_lock is already held.

If ibmvfc_send_event() fails (e.g. due to a hypercall failure), it
synchronously invokes the completion callback evt->done(evt).

Because the event was initialized with IBMVFC_MAD_FORMAT via
ibmvfc_init_event(), evt->done is set to ibmvfc_locked_done(), which
attempts to acquire the host_lock again:

    spin_lock_irqsave(evt->vhost->host->host_lock, flags);

Would this cause a hard deadlock on the CPU if the underlying hardware
encounters an error and fails to send the process login?

>  		vhost->discovery_threads--;
>  		ibmvfc_set_tgt_action(tgt, IBMVFC_TGT_ACTION_NONE);
>  		kref_put(&tgt->kref, ibmvfc_release_tgt);
>  	} else
> -		tgt_dbg(tgt, "Sent process login\n");
> +		tgt_dbg(tgt, "%s Sent process login\n", proto_type[tgt->protocol]);
>  }

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260623013035.3436640-1-tyreld@linux.ibm.com?part=14