Jeff Layton <jlayton@xxxxxxxxxx> wrote: > > + rreq->work.func = netfs_rreq_work; > > + > > ^^^ > This seems like it should be an INIT_WORK call. I assume you're moving > this here this because you intend to use netfs_alloc_request for writes > too? Interesting question. INIT_WORK() was called in netfs_alloc_request(), so the lockdep state has already been initialised and may even have been used already (say, for instance, we do an RMW cycle buffering in the same request struct). David -- Linux-cachefs mailing list Linux-cachefs@xxxxxxxxxx https://listman.redhat.com/mailman/listinfo/linux-cachefs
