Nouveau - Jul 2014 - [PATCH 09/17] drm/radeon: use common fence implementation for fences

> Regardless of the fence implementation, why would it be a good idea to do a
full lockup recovery when some other driver is
> calling your wait function? That doesn't seem to be a nice thing to do,
so I think a timeout is the best error you could return here,
> other drivers have to deal with that anyway.
The problem is that we need to guarantee that the lockup will be 
resolved eventually.

Just imagine an application using prime is locking up Radeon and because 
of that gets killed by the user. Nothing else in the system would use 
the Radeon hardware any more and so radeon gets only called by another 
driver waiting patiently for radeon to finish rendering which never 
happens because the whole thing is locked up and we don't get a chance 
to recover.

Christian.

Am 23.07.2014 09:51, schrieb Maarten Lankhorst:
> op 23-07-14 09:37, Christian K?nig schreef:
>> Am 23.07.2014 09:31, schrieb Daniel Vetter:
>>> On Wed, Jul 23, 2014 at 9:26 AM, Christian K?nig
>>> <deathsimple at vodafone.de> wrote:
>>>> It's not a locking problem I'm talking about here.
Radeons lockup handling
>>>> kicks in when anything calls into the driver from the outside,
if you have a
>>>> fence wait function that's called from the outside but
doesn't handle
>>>> lockups you essentially rely on somebody else calling another
radeon
>>>> function for the lockup to be resolved.
>>> So you don't have a timer in radeon that periodically checks
whether
>>> progress is still being made? That's the approach we're
using in i915,
>>> together with some tricks to kick any stuck waiters so that we can
>>> reliably step in and grab locks for the reset.
>> We tried this approach, but it didn't worked at all.
>>
>> I already considered trying it again because of the upcoming fence
implementation, but reconsidering that when a driver is forced to change
it's handling because of the fence implementation that's just another
hint that there is something wrong here.
> As far as I can tell it wouldn't need to be reworked for the fence
implementation currently, only the moment you want to allow callers outside of
radeon. :-)
> Doing a GPU lockup recovery in the wait function would be messy even right
now, you would hit a deadlock in ttm_bo_delayed_delete ->
ttm_bo_cleanup_refs_and_unlock.
>
> Regardless of the fence implementation, why would it be a good idea to do a
full lockup recovery when some other driver is
> calling your wait function? That doesn't seem to be a nice thing to do,
so I think a timeout is the best error you could return here,
> other drivers have to deal with that anyway.
>
> ~Maarten
>

On Wed, Jul 23, 2014 at 9:58 AM, Christian K?nig
<deathsimple at vodafone.de> wrote:
> Just imagine an application using prime is locking up Radeon and because of
> that gets killed by the user. Nothing else in the system would use the
> Radeon hardware any more and so radeon gets only called by another driver
> waiting patiently for radeon to finish rendering which never happens
because
> the whole thing is locked up and we don't get a chance to recover.
But isn't that possible already without fences? X hangs radeon, user
crashes X for unrelated reasons before radeon will notice the hang.
Then no one uses radeon any longer and the hang stays undetected.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch

Am 23.07.2014 10:07, schrieb Daniel Vetter:
> On Wed, Jul 23, 2014 at 9:58 AM, Christian K?nig
> <deathsimple at vodafone.de> wrote:
>> Just imagine an application using prime is locking up Radeon and
because of
>> that gets killed by the user. Nothing else in the system would use the
>> Radeon hardware any more and so radeon gets only called by another
driver
>> waiting patiently for radeon to finish rendering which never happens
because
>> the whole thing is locked up and we don't get a chance to recover.
> But isn't that possible already without fences? X hangs radeon, user
> crashes X for unrelated reasons before radeon will notice the hang.
> Then no one uses radeon any longer and the hang stays undetected.
Yeah, especially with multimedia application. But I don't really care 
about this problem because the next time an application tries to use the 
block in question we actually do the reset and everything is fine.

In your example we would do the reset when the next X server starts, 
before that point nobody would care because nobody uses the hardware.

An additional problem here is that resets are something perfect normal 
for radeon. For example UVD can "crash" when you feed it with invalid 
bitstream data, (ok actually it send an interrupt and stops any 
processing for the driver to investigate). To continue processing you 
need to go through a rather complicated reset procedure.

Christian.
> -Daniel