[PATCH 2/6] drm/radeon: remove HDMI interrupts on Evergreen

[Rafał Miłecki]

2013/4/21 Alex Deucher <alexdeucher at gmail.com>:
> On Sun, Apr 21, 2013 at 3:15 PM, Rafał Miłecki <zajec5 at gmail.com> wrote:
>> 2013/4/21 Rafał Miłecki <zajec5 at gmail.com>:
>>> 2013/4/14 Alex Deucher <alexdeucher at gmail.com>:
>>>> On Sat, Apr 13, 2013 at 7:26 PM, Rafał Miłecki <zajec5 at gmail.com> wrote:
>>>>> We need interrupts on format change for R6xx only, where hardware seems
>>>>> to be somehow bugged and requires setting audio info manually.
>>>>
>>>> Can you confirm that this is actually needed on older chips?  AFAIK,
>>>> it shouldn't be required for any chips.  It's mainly for debugging.
>>>
>>> I've finally managed to install, run and debug fglrx on RV620. That
>>> was really painful and ignoring Xorg 1.13 release in fglrx legacy
>>> didn't help :|
>>>
>>> Please take a look at attached log from fglrx. It seems that fglrx on
>>> every audio format change does:
>>> 1) Read audio status (R600_AUDIO_STATUS_BITS, R600_AUDIO_RATE_BPS_CHANNEL)
>>> 2) Writes sth to HDMI0_STATUS and HDMI0_INFOFRAME_CONTROL0
>>> 3) Updates audio info frame
>>> 4) Recalculates checksum of audio info frame
>>> 5) Triggers update with HDMI0_AUDIO_INFO_UPDATE bit
>>
>> Btw. could we get a meaning of bit 0x100 in HDMI0_STATUS?
>
> There is no bit 8 defined in that register.  Also FWIW, HDMI0_STATUS
> is a read only register.

Ahh, sorry. 0x7400 is HDMI0_CONTROL not HDMI0_STATUS! My silly
mistake. It makes sense of course:
#       define HDMI0_ERROR_ACK       (1 << 8)

Still I can't see how we could drop interrupts and keep updating audio
infoframe.

-- 
Rafał