[Intel-gfx] [PATCH] drm/i915/dp: Read link status more times when EQ not done

[Jani Nikula]

On Thu, 02 Feb 2017, "Lee, Shawn C" <shawn.c.lee at intel.com> wrote:
> From: "Lee, Shawn C" <shawn.c.lee at intel.com>
>
> When user space link status, display driver read DPCD register
> 0x202, 0x203 and 0x204 to identify sink status. When PSR exit
> is ongoing before EQ done. Panel will report EQ & symbol lock
> not done. Both of them are under progressing at the same time
> to cause this issue.
>
> This WA try to read link status more times if EQ not done.
> Panel spec request at least 1000us for fast link train when
> PSR exit. So driver will wait 1000us~1500us then retrieve
> sink link status again.

I'm not sure about the patch, but no matter what it needs to be rebased
on top of current drm-tip to even be considered.

If you're working around a specific eDP PSR issue, why do this on *all*
DP?

BR,
Jani.


>
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=99639
> TEST=Reboot DUT and no flicking on local display at login screen
>
> Cc: Cooper Chiou <cooper.chiou at intel.com>
> Cc: Wei Shun Chen <wei.shun.chang at intel.com>
> Cc: Gary C Wang <gary.c.wang at intel.com>
> Cc: Jani Nikula <jani.nikula at intel.com>
>
> Signed-off-by: Lee, Shawn C <shawn.c.lee at intel.com>
> ---
>  drivers/gpu/drm/i915/intel_dp.c |   33 ++++++++++++++++++++++++++++-----
>  1 file changed, 28 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
> index e80d620846c8..a9a6ce476438 100644
> --- a/drivers/gpu/drm/i915/intel_dp.c
> +++ b/drivers/gpu/drm/i915/intel_dp.c
> @@ -4136,13 +4136,36 @@ static void intel_dp_handle_test_request(struct intel_dp *intel_dp)
>  		return;
>  
>  	/* if link training is requested we should perform it always */
> -	if ((intel_dp->compliance.test_type == DP_TEST_LINK_TRAINING) ||
> -	    (!drm_dp_channel_eq_ok(link_status, intel_dp->lane_count))) {
> -		DRM_DEBUG_KMS("%s: channel EQ not ok, retraining\n",
> -			      intel_encoder->base.name);
> +	if (intel_dp->compliance.test_type == DP_TEST_LINK_TRAINING) {
> +		DRM_DEBUG_KMS("%s: compliance test mode, retraining\n",
> +			intel_encoder->base.name);
> +	} else {
> +		u8 retry;
> +
> +		for (retry = 0; retry < 3; retry++) {
> +			if (!drm_dp_channel_eq_ok(link_status, intel_dp->lane_count)) {
> +				/*
> +				 * EQ not ok may caused by fast link train while exit PSR active,
> +				 * wait at least 1000 us then read it again.
> +				 */
> +				DRM_DEBUG_KMS("%s: channel EQ not ok, retry = %d, DPCD 0x202 = 0x%x, 0x203 = 0x%x, 0x204 = 0x%x\n",
> +					intel_encoder->base.name, retry, link_status[0], link_status[1], link_status[2]);
> +				usleep_range(1000, 1500);
> +				if (!intel_dp_get_link_status(intel_dp, link_status)) {
> +					DRM_ERROR("Failed to get link status\n");
> +					return;
> +				}
> +			} else {
> +				/* channel EQ is fine */
> +				return;
> +			}
> +		}
>  
> -		intel_dp_retrain_link(intel_dp);
> +		DRM_DEBUG_KMS("%s: channel EQ not ok, retraining\n",
> +			intel_encoder->base.name);
>  	}
> +
> +	intel_dp_retrain_link(intel_dp);
>  }
>  
>  /*

-- 
Jani Nikula, Intel Open Source Technology Center