On Tue, 2018-07-10 at 12:38 -0600, Jason Gunthorpe wrote: > On Tue, Jul 10, 2018 at 10:32:00AM -0700, Bart Van Assche wrote: > > diff --git a/drivers/infiniband/ulp/srpt/ib_srpt.c b/drivers/infiniband/ulp/srpt/ib_srpt.c > > index 325bae29e90d..705f6a992d82 100644 > > +++ b/drivers/infiniband/ulp/srpt/ib_srpt.c > > @@ -2152,6 +2152,7 @@ static int srpt_cm_req_recv(struct srpt_device *const sdev, > > } > > > > kref_init(&ch->kref); > > + kref_get(&ch->kref); > > Oh that is ugly, can you put the 'get' closer to the the place that > stores the ch pointer this kref is protecting? Perhaps it is one of > the list_add's in this function? > > Guessing the kref created kref_init should probably belong to target_core's > 'se_sess->fabric_sess_ptr' .. Hello Jason, Can you clarify your second paragraph? The ib_srpt driver has been implemented such that sess->fabric_sess_ptr == ch as long as a session exists. An ib_srpt RDMA channel object (ch in the above code) must stay around as long as the associated target core session (se_sess) exists and also as long as the target core has not yet called srpt_close_session(). Hence the initialization of ch->kref to 2 just before an RDMA channel is registered with the target core. Hence also the kref_put() calls in srpt_close_session() and srpt_release_channel_work(). Bart.
