[Intel-gfx] [PATCH] drm/i915: bdw expands ACTHD to 64bit

[Daniel Vetter]

On Thu, Mar 20, 2014 at 5:28 PM, Chris Wilson <chris at chris-wilson.co.uk> wrote:
>> > I wasn't sure either, but I thought since we didn't do anything special
>> > for BBADDR, to leave ACTHD alone.
>> >
>> > I wonder if it would help splitting it up, having to count 8 extra
>> > leading zeros is going to be a nightmare (especially as the decoder
>> > doesn't pad it right either...)
>> >
>> > For reading, I think I would prefer
>> > err_printf(m, "  ACTHD: 0x%016llx [0x%08x %08x]\n",
>> >            ring->acthd, (u32)(ring->acthd>>32), (u32)(ring->acthd));
>> >
>> > or maybe just
>> > err_printf(m, "  ACTHD: 0x%08x %08x\n",
>> >            (u32)(ring->acthd>>32), (u32)(ring->acthd));
>>
>> I expect this to lead to a day-long wtf session once we have 64b ppgtt
>> enabled and a leaking X hung ;-) I use g* a lot in vim when reading error
>> states, so dropping information is dangerous.
>
> I'm not following, where is the information loss? If you haven't already
> guessed, we are seeing ACTHD above 4GiB already.

Utter failure to read your diff. I agree that some form of split would
look nice.

/me hides in shame
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch