* [Qemu-devel] [RfC PATCH] hid: handle full ptr queues in post_load
@ 2014-11-27 9:04 Gerd Hoffmann
2014-11-27 9:25 ` Gonglei
0 siblings, 1 reply; 7+ messages in thread
From: Gerd Hoffmann @ 2014-11-27 9:04 UTC (permalink / raw)
To: qemu-devel; +Cc: Gonglei, Gerd Hoffmann, Dr. David Alan Gilbert
Cc: Gonglei <arei.gonglei@huawei.com>
Cc: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
---
hw/input/hid.c | 21 +++++++++++++++++++++
1 file changed, 21 insertions(+)
diff --git a/hw/input/hid.c b/hw/input/hid.c
index 8f6fbb3..6fb963f 100644
--- a/hw/input/hid.c
+++ b/hw/input/hid.c
@@ -519,6 +519,27 @@ static int hid_post_load(void *opaque, int version_id)
HIDState *s = opaque;
hid_set_next_idle(s);
+
+ if (s->n == QUEUE_LENGTH && (s->kind == HID_TABLET ||
+ s->kind == HID_MOUSE)) {
+ /*
+ * Handle ptr device migration from old qemu with full queue.
+ *
+ * Throw away everything but the last event, so we propagate
+ * at least the current button state to the guest. Also keep
+ * current position for the tablet, signal "no motion" for the
+ * mouse.
+ */
+ HIDPointerEvent evt;
+ evt = s->ptr.queue[(s->head+s->n) & QUEUE_MASK];
+ if (s->kind == HID_MOUSE) {
+ evt.xdx = 0;
+ evt.ydy = 0;
+ }
+ s->ptr.queue[0] = evt;
+ s->head = 0;
+ s->n = 1;
+ }
return 0;
}
--
1.8.3.1
^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [Qemu-devel] [RfC PATCH] hid: handle full ptr queues in post_load
2014-11-27 9:04 [Qemu-devel] [RfC PATCH] hid: handle full ptr queues in post_load Gerd Hoffmann
@ 2014-11-27 9:25 ` Gonglei
2014-11-27 11:04 ` Gerd Hoffmann
0 siblings, 1 reply; 7+ messages in thread
From: Gonglei @ 2014-11-27 9:25 UTC (permalink / raw)
To: Gerd Hoffmann; +Cc: qemu-devel, Dr. David Alan Gilbert
On 2014/11/27 17:04, Gerd Hoffmann wrote:
> Cc: Gonglei <arei.gonglei@huawei.com>
> Cc: Dr. David Alan Gilbert <dgilbert@redhat.com>
> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
> ---
> hw/input/hid.c | 21 +++++++++++++++++++++
> 1 file changed, 21 insertions(+)
>
> diff --git a/hw/input/hid.c b/hw/input/hid.c
> index 8f6fbb3..6fb963f 100644
> --- a/hw/input/hid.c
> +++ b/hw/input/hid.c
> @@ -519,6 +519,27 @@ static int hid_post_load(void *opaque, int version_id)
> HIDState *s = opaque;
>
> hid_set_next_idle(s);
> +
> + if (s->n == QUEUE_LENGTH && (s->kind == HID_TABLET ||
> + s->kind == HID_MOUSE)) {
> + /*
> + * Handle ptr device migration from old qemu with full queue.
> + *
> + * Throw away everything but the last event, so we propagate
> + * at least the current button state to the guest. Also keep
> + * current position for the tablet, signal "no motion" for the
> + * mouse.
> + */
> + HIDPointerEvent evt;
> + evt = s->ptr.queue[(s->head+s->n) & QUEUE_MASK];
s->n is QUEUE_LENGTH, can we directly delete it?
evt = s->ptr.queue[s->head & QUEUE_MASK]
Best regards,
-Gonglei
> + if (s->kind == HID_MOUSE) {
> + evt.xdx = 0;
> + evt.ydy = 0;
> + }
> + s->ptr.queue[0] = evt;
> + s->head = 0;
> + s->n = 1;
> + }
> return 0;
> }
>
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Qemu-devel] [RfC PATCH] hid: handle full ptr queues in post_load
2014-11-27 9:25 ` Gonglei
@ 2014-11-27 11:04 ` Gerd Hoffmann
2014-11-27 11:16 ` Gonglei
0 siblings, 1 reply; 7+ messages in thread
From: Gerd Hoffmann @ 2014-11-27 11:04 UTC (permalink / raw)
To: Gonglei; +Cc: qemu-devel, Dr. David Alan Gilbert
Hi,
> > + evt = s->ptr.queue[(s->head+s->n) & QUEUE_MASK];
>
> s->n is QUEUE_LENGTH, can we directly delete it?
> evt = s->ptr.queue[s->head & QUEUE_MASK]
I prefer to make clear in the code that we want the last ring element
not the first and leave in the "+n", even if we could take it out.
cheers,
Gerd
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Qemu-devel] [RfC PATCH] hid: handle full ptr queues in post_load
2014-11-27 11:04 ` Gerd Hoffmann
@ 2014-11-27 11:16 ` Gonglei
2014-11-27 11:19 ` Gerd Hoffmann
0 siblings, 1 reply; 7+ messages in thread
From: Gonglei @ 2014-11-27 11:16 UTC (permalink / raw)
To: Gerd Hoffmann; +Cc: qemu-devel, Dr. David Alan Gilbert
On 2014/11/27 19:04, Gerd Hoffmann wrote:
> Hi,
>
>>> + evt = s->ptr.queue[(s->head+s->n) & QUEUE_MASK];
>>
>> s->n is QUEUE_LENGTH, can we directly delete it?
>> evt = s->ptr.queue[s->head & QUEUE_MASK]
>
> I prefer to make clear in the code that we want the last ring element
> not the first and leave in the "+n", even if we could take it out.
>
OK.
Another question, whether or not we handle this scenario
before calling hid_set_next_idle(s) ? Maybe it is safer, because
hid_idle_timer will call hid_pointer_event(), callback function
of hs->event(hs).
Thanks,
-Gonglei
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Qemu-devel] [RfC PATCH] hid: handle full ptr queues in post_load
2014-11-27 11:16 ` Gonglei
@ 2014-11-27 11:19 ` Gerd Hoffmann
2014-11-27 11:22 ` Gonglei
0 siblings, 1 reply; 7+ messages in thread
From: Gerd Hoffmann @ 2014-11-27 11:19 UTC (permalink / raw)
To: Gonglei; +Cc: qemu-devel, Dr. David Alan Gilbert
On Do, 2014-11-27 at 19:16 +0800, Gonglei wrote:
> On 2014/11/27 19:04, Gerd Hoffmann wrote:
>
> > Hi,
> >
> >>> + evt = s->ptr.queue[(s->head+s->n) & QUEUE_MASK];
> >>
> >> s->n is QUEUE_LENGTH, can we directly delete it?
> >> evt = s->ptr.queue[s->head & QUEUE_MASK]
> >
> > I prefer to make clear in the code that we want the last ring element
> > not the first and leave in the "+n", even if we could take it out.
> >
> OK.
>
> Another question, whether or not we handle this scenario
> before calling hid_set_next_idle(s) ? Maybe it is safer, because
> hid_idle_timer will call hid_pointer_event(), callback function
> of hs->event(hs).
Not needed, the timer will not called before vmload is completely
finished.
cheers,
Gerd
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Qemu-devel] [RfC PATCH] hid: handle full ptr queues in post_load
2014-11-27 11:19 ` Gerd Hoffmann
@ 2014-11-27 11:22 ` Gonglei
2014-11-27 12:28 ` Gonglei
0 siblings, 1 reply; 7+ messages in thread
From: Gonglei @ 2014-11-27 11:22 UTC (permalink / raw)
To: Gerd Hoffmann; +Cc: qemu-devel, Dr. David Alan Gilbert
On 2014/11/27 19:19, Gerd Hoffmann wrote:
> On Do, 2014-11-27 at 19:16 +0800, Gonglei wrote:
>> On 2014/11/27 19:04, Gerd Hoffmann wrote:
>>
>>> Hi,
>>>
>>>>> + evt = s->ptr.queue[(s->head+s->n) & QUEUE_MASK];
>>>>
>>>> s->n is QUEUE_LENGTH, can we directly delete it?
>>>> evt = s->ptr.queue[s->head & QUEUE_MASK]
>>>
>>> I prefer to make clear in the code that we want the last ring element
>>> not the first and leave in the "+n", even if we could take it out.
>>>
>> OK.
>>
>> Another question, whether or not we handle this scenario
>> before calling hid_set_next_idle(s) ? Maybe it is safer, because
>> hid_idle_timer will call hid_pointer_event(), callback function
>> of hs->event(hs).
>
> Not needed, the timer will not called before vmload is completely
> finished.
>
Yep, as the Qemu big lock.
I will test this patch later, thanks for your work. :)
Regards,
-Gonglei
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [Qemu-devel] [RfC PATCH] hid: handle full ptr queues in post_load
2014-11-27 11:22 ` Gonglei
@ 2014-11-27 12:28 ` Gonglei
0 siblings, 0 replies; 7+ messages in thread
From: Gonglei @ 2014-11-27 12:28 UTC (permalink / raw)
To: Gerd Hoffmann; +Cc: qemu-devel, Dr. David Alan Gilbert
On 2014/11/27 19:22, Gonglei wrote:
> On 2014/11/27 19:19, Gerd Hoffmann wrote:
>
>> On Do, 2014-11-27 at 19:16 +0800, Gonglei wrote:
>>> On 2014/11/27 19:04, Gerd Hoffmann wrote:
>>>
>>>> Hi,
>>>>
>>>>>> + evt = s->ptr.queue[(s->head+s->n) & QUEUE_MASK];
>>>>>
>>>>> s->n is QUEUE_LENGTH, can we directly delete it?
>>>>> evt = s->ptr.queue[s->head & QUEUE_MASK]
>>>>
>>>> I prefer to make clear in the code that we want the last ring element
>>>> not the first and leave in the "+n", even if we could take it out.
>>>>
>>> OK.
>>>
>>> Another question, whether or not we handle this scenario
>>> before calling hid_set_next_idle(s) ? Maybe it is safer, because
>>> hid_idle_timer will call hid_pointer_event(), callback function
>>> of hs->event(hs).
>>
>> Not needed, the timer will not called before vmload is completely
>> finished.
>>
>
> Yep, as the Qemu big lock.
> I will test this patch later, thanks for your work. :)
>
Patch works for me. Thanks.
Tested-by: Gonglei <arei.gonglei@huawei.com>
Reviewed-by: Gonglei <arei.gonglei@huawei.com>
Regards,
-Gonglei
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2014-11-27 12:29 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2014-11-27 9:04 [Qemu-devel] [RfC PATCH] hid: handle full ptr queues in post_load Gerd Hoffmann
2014-11-27 9:25 ` Gonglei
2014-11-27 11:04 ` Gerd Hoffmann
2014-11-27 11:16 ` Gonglei
2014-11-27 11:19 ` Gerd Hoffmann
2014-11-27 11:22 ` Gonglei
2014-11-27 12:28 ` Gonglei
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.