On 19/01/2016 17:52, Daniel P. Berrange wrote: > I'll have a look at this and see what I can do. It shouldn't block > merging of this patch I assume, since the current call to > qemu_coroutine_yield() should be fine - it just my comment suggestion > that would not work.
Yes, that is tied to negotiations being always in the main loop. I'm not entirely sure that is the case (e.g. a call bdrv_reopen could perhaps be from somewhere else) but it's anyway no worse than the current state. Paolo
