On (Wed) Apr 28 2010 [08:25:59], Anthony Liguori wrote:
> On 04/28/2010 02:29 AM, Amit Shah wrote:
>> On (Tue) Apr 27 2010 [12:41:27], Anthony Liguori wrote:
>>
>>> On 04/27/2010 07:34 AM, Amit Shah wrote:
>>>
>>>> From: Marcelo Tosatti<mtosa...@redhat.com>
>>>>
>>>> Wake up iothread when buffers are consumed.
>>>>
>>>> Signed-off-by: Marcelo Tosatti<mtosa...@redhat.com>
>>>> Signed-off-by: Amit Shah<amit.s...@redhat.com>
>>>>
>>>>
>>> What's the race here? This looks very odd to me.
>>>
>> When the guest indicates it has added buffers to the vq, the iothread
>> can then start consuming them. Without this notification, the iothread
>> only polls for free buffers when it times out or gets woken up
>> otherwise.
>>
>
> When you say, polls for free buffers, what do you mean by that?
>
> You mean, there's a can_read() somewhere that checks for free buffers?
Not just can_read(), in-qemu applications too can query when writes to
guests will go through.
> I think switching to qemu_set_fd_handler() would be a better solution.
Hm, there's no fd here. How to signal to apps (ports) when a guest
becomes writable?
Amit
