Re: [PATCH v2 10/11] vsock_test: skip read() in test_stream*close tests on a VMCI host

From: Stefano Garzarella
Date: Thu Aug 22 2019 - 09:45:23 EST

On Tue, Aug 20, 2019 at 09:32:03AM +0100, Stefan Hajnoczi wrote:
> On Thu, Aug 01, 2019 at 05:25:40PM +0200, Stefano Garzarella wrote:
> > When VMCI transport is used, if the guest closes a connection,
> > all data is gone and EOF is returned, so we should skip the read
> > of data written by the peer before closing the connection.
> All transports should aim for identical semantics. I think virtio-vsock
> should behave the same as VMCI since userspace applications should be
> transport-independent.

Yes, it is a good point!

> Let's view this as a vsock bug. Is it feasible to change the VMCI
> behavior so it's more like TCP sockets? If not, let's change the
> virtio-vsock behavior to be compatible with VMCI.

I'm not sure it is feasible to change the VMCI behavior. IIUC reading the
Jorgen's answer [1], this was a decision made during the implementation.

@Jorgen: please, can you confirm? or not :-)

If it is the case, I'll change virtio-vsock to the same behavior.