On Tue, 31 Dec 2024, Imre Deak <imre.deak@xxxxxxxxx> wrote:
> On Thu, Dec 19, 2024 at 11:34:00PM +0200, Jani Nikula wrote:
>> Write the DP2 specific VFREQ registers.
>>
>> This is preparation for enabling 128b/132b SST. This path is not
>> reachable yet.
>>
>> Signed-off-by: Jani Nikula <jani.nikula@xxxxxxxxx>
>> ---
>> drivers/gpu/drm/i915/display/intel_ddi.c | 12 ++++++++++++
>> 1 file changed, 12 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/i915/display/intel_ddi.c b/drivers/gpu/drm/i915/display/intel_ddi.c
>> index 64528ff8856e..91e6cd91e91f 100644
>> --- a/drivers/gpu/drm/i915/display/intel_ddi.c
>> +++ b/drivers/gpu/drm/i915/display/intel_ddi.c
>> @@ -3467,8 +3467,20 @@ static void intel_ddi_enable(struct intel_atomic_state *state,
>> {
>> struct intel_display *display = to_intel_display(encoder);
>> struct intel_crtc *pipe_crtc;
>> + enum transcoder cpu_transcoder = crtc_state->cpu_transcoder;
>> int i;
>>
>> + /* 128b/132b SST */
>> + if (intel_dp_is_uhbr(crtc_state)) {
>
> Hm, not sure how this would work on HDMI (FRL), but to be sure shouldn't
> this check that crtc_state is for a DP (SST) mode?
DP MST does not call intel_ddi_enable().
Yes, it's all very subtle, and I hate the way all of the
DDI/SST/MST/etc. is intertwined, but I feel like it's overall a bigger
refactoring to clean up.
BR,
Jani.
>
>> + const struct drm_display_mode *adjusted_mode = &crtc_state->hw.adjusted_mode;
>> + u64 crtc_clock_hz = KHz(adjusted_mode->crtc_clock);
>> +
>> + intel_de_write(display, TRANS_DP2_VFREQHIGH(cpu_transcoder),
>> + TRANS_DP2_VFREQ_PIXEL_CLOCK(crtc_clock_hz >> 24));
>> + intel_de_write(display, TRANS_DP2_VFREQLOW(cpu_transcoder),
>> + TRANS_DP2_VFREQ_PIXEL_CLOCK(crtc_clock_hz & 0xffffff));
>> + }
>> +
>> intel_ddi_enable_transcoder_func(encoder, crtc_state);
>>
>> /* Enable/Disable DP2.0 SDP split config before transcoder */
>> --
>> 2.39.5
>>
--
Jani Nikula, Intel
