[PATCH v5 07/14] nvmet-fcloop: access fcpreq only when holding reqlock

[Daniel Wagner]

On Thu, Apr 24, 2025 at 12:15:20PM +0200, Hannes Reinecke wrote:
> > @@ -663,9 +667,10 @@ fcloop_fcp_abort_recv_work(struct work_struct *work)
> >   	unsigned long flags;
> >   	spin_lock_irqsave(&tfcp_req->reqlock, flags);
> > -	fcpreq = tfcp_req->fcpreq;
> >   	switch (tfcp_req->inistate) {
> >   	case INI_IO_ABORTED:
> > +		fcpreq = tfcp_req->fcpreq;
> > +		tfcp_req->fcpreq = NULL;
> >   		break;
> >   	case INI_IO_COMPLETED:
> >   		completed = true;
> > @@ -688,10 +693,6 @@ fcloop_fcp_abort_recv_work(struct work_struct *work)
> >   		nvmet_fc_rcv_fcp_abort(tfcp_req->tport->targetport,
> >   					&tfcp_req->tgt_fcp_req);
> > -	spin_lock_irqsave(&tfcp_req->reqlock, flags);
> > -	tfcp_req->fcpreq = NULL;
> > -	spin_unlock_irqrestore(&tfcp_req->reqlock, flags);
> > -
> What happens for INI_IO_COMPLETED?

The request was completed before the abort handler was running. Thus
nothing to do here.

> Don't we need to clear the 'fcpreq' pointer in that case, too?

The normal code path has already taken care of this request and the only
thing left to do is to give back the refcount on the tfcp_req to free
the memory.