Re: [Qemu-devel] [PATCH] tap: avoid deadlocking rx

2014-03-11 Thread Stefan Hajnoczi
On Sat, Mar 08, 2014 at 04:00:43PM +0100, Stefan Hajnoczi wrote: > The net subsystem has a control flow mechanism so peer NetClientStates > can tell each other to stop sending packets. This is used to stop > monitoring the tap file descriptor for incoming packets if the guest rx > ring has no spar

Re: [Qemu-devel] [PATCH] tap: avoid deadlocking rx

2014-03-10 Thread Neil Skrypuch
On Saturday 08 March 2014 16:00:43 Stefan Hajnoczi wrote: > The net subsystem has a control flow mechanism so peer NetClientStates > can tell each other to stop sending packets. This is used to stop > monitoring the tap file descriptor for incoming packets if the guest rx > ring has no spare buffe

[Qemu-devel] [PATCH] tap: avoid deadlocking rx

2014-03-08 Thread Stefan Hajnoczi
The net subsystem has a control flow mechanism so peer NetClientStates can tell each other to stop sending packets. This is used to stop monitoring the tap file descriptor for incoming packets if the guest rx ring has no spare buffers. There is a corner case when tap_can_send() is true at the beg