> -----Original Message----- > From: Doug Ledford [mailto:dledford@xxxxxxxxxx] > Sent: Monday, February 04, 2019 1:20 PM > To: Dalessandro, Dennis <dennis.dalessandro@xxxxxxxxx>; jgg@xxxxxxxx > Cc: Dixit, Ashutosh <ashutosh.dixit@xxxxxxxxx>; linux-rdma@xxxxxxxxxxxxxxx; > Mitko Haralanov <mitko.haralanov@xxxxxxxxx>; Marciniszyn, Mike > <mike.marciniszyn@xxxxxxxxx>; Wan, Kaike <kaike.wan@xxxxxxxxx> > Subject: Re: [PATCH for-next 02/17] IB/hfi1: TID RDMA flow allocation > > On Wed, 2019-01-23 at 19:29 -0800, Dennis Dalessandro wrote: > > + * Return: > > + * false if all the conditions below are statisfied: > > + * 1. The list is empty or > > + * 2. The indicated qp is at the head of the list and the > > + * HFI1_S_WAIT_TID_SPACE bit is set in qp->s_flags. > > + * true is returned otherwise. > > + */ > > +static bool kernel_tid_waiters(struct hfi1_ctxtdata *rcd, > > + struct tid_queue *queue, struct rvt_qp *qp) > > + __must_hold(&rcd->exp_lock) __must_hold(&qp->s_lock) { > > + struct rvt_qp *fqp; > > + bool ret = true; > > + > > + lockdep_assert_held(&qp->s_lock); > > + lockdep_assert_held(&rcd->exp_lock); > > + fqp = first_qp(rcd, queue); > > + if (!fqp || (fqp == qp && (qp->s_flags & HFI1_S_WAIT_TID_SPACE))) > > + ret = false; > > + rvt_put_qp(fqp); > > + return ret; > > +} > > + > > The comment and the code don't match. I can fix it up when applying, or if > this series needs a respin you can fix it then. We will fix it in next spin. Thank you, Kaike > > -- > Doug Ledford <dledford@xxxxxxxxxx> > GPG KeyID: B826A3330E572FDD > Key fingerprint = AE6B 1BDA 122B 23B4 265B 1274 B826 A333 0E57 2FDD
