Re: [RFC] hv_netvsc: safer orderly shutdown

2017-11-13 Thread Stephen Hemminger
On Mon, 13 Nov 2017 11:57:47 +0100 Vitaly Kuznetsov wrote: > Stephen Hemminger writes: > > > > > The NAPI disable is already handled by rndis close. > > Sorry, but I'm probably missing something: I can only see > netif_napi_del() call in netvsc_device_remove() but this happens much > later.

Re: [RFC] hv_netvsc: safer orderly shutdown

2017-11-13 Thread Vitaly Kuznetsov
Stephen Hemminger writes: > > The NAPI disable is already handled by rndis close. Sorry, but I'm probably missing something: I can only see netif_napi_del() call in netvsc_device_remove() but this happens much later. And I don see us doing napi_disable() anywhere on the path. But I'm probably mi

Re: [RFC] hv_netvsc: safer orderly shutdown

2017-11-10 Thread Vitaly Kuznetsov
Stephen Hemminger writes: > Several types of control operations require that the underlying RNDIS > infrastructure be restarted. This patch changes the ordering of the > shutdown to avoid race conditions. > Stop all transmits before doing RNDIS halt. This involves stopping the > network device tr

[RFC] hv_netvsc: safer orderly shutdown

2017-11-09 Thread Stephen Hemminger
Several types of control operations require that the underlying RNDIS infrastructure be restarted. This patch changes the ordering of the shutdown to avoid race conditions. Stop all transmits before doing RNDIS halt. This involves stopping the network device transmit queues, then waiting for all o