[Intel-gfx] Deadlock in intel_user_framebuffer_destroy()

[Jani Nikula]

On Mon, 29 Jun 2015, Lukas Wunner <lukas at wunner.de> wrote:
> Hi Jani,
>
> On Mon, Jun 29, 2015 at 02:24:19PM +0300, Jani Nikula wrote:
>> On Mon, 15 Jun 2015, Chris Wilson <chris at chris-wilson.co.uk> wrote:
>> > There wasn't, I just rewrote it incorrectly. There's also a
>> > drm_framebuffer_remove() called by intelfb_alloc which needs to be moved
>> > out of the mutex. A much larger disentangling of the functions involved
>> > here is required. Tvrstko volunteered himself. Brave, very brave :)
>> -ETIMEDOUT
>> IIUC we still need the fix, but there's no update from Lucas addressing
>> review. Is that right?
>
> Sorry, I was not copied on Tvrtko's e-mail of June 15 ("drm/i915:
> Fix failure paths around initial fbdev allocation") and though
> I'm subscribed to intel-gfx I didn't realize that particular
> e-mail required a reaction from me.

No worries, I didn't realize these two threads were connected!

BR,
Jani.

>
> I've just sent out my review and I'll also test the patch and
> report back.
>
> Thanks,
>
> Lukas

-- 
Jani Nikula, Intel Open Source Technology Center