On Fri, 2017-01-13 at 14:08 -0500, Chuck Lever wrote:
> I was thinking of wiring something into call_timeout, but xprt_timer
> looks like it would perform the same job, and there is already a
> per-xprt hook. I'll have a look.
>
> Is it safe to call rpc_run_task while transport_lock is held? If not
> I can simply schedule a generic worker thread to construct and send
> the NULL.
You shouldn't call rpc_run_task() from a spinlock context, but let's
consider moving that spinlock into the xprt->ops->timer() callback. I
don't really see how it is needed for the cases where you don't have a
callback.
--
Trond Myklebust
Linux NFS client maintainer, PrimaryData
trond.myklebust@xxxxxxxxxxxxxxx
��.n��������+%������w��{.n�����{��w��� jg��� �����ݢj����G���� ���j:+v���w�m������w�������h�����٥
