diff mbox

drm/i915: Fall back to zero vswing/preemph if the sink doesn't like the last good values

Message ID 1446223656-27792-1-git-send-email-ville.syrjala@linux.intel.com (mailing list archive)
State New, archived
Headers show

Commit Message

Ville Syrjälä Oct. 30, 2015, 4:47 p.m. UTC 
From: Ville Syrjälä <ville.syrjala@linux.intel.com>

My Lenovo STM STDP3100 miniDP->VGA dongle doesn't seem to like it when
we try to start link training with non-zero vswing/preemphasis. So when
the initial link training DPCD write fails, retry it with zero values.

Fixes a bunch of errors like so:
[drm:intel_dp_start_link_train [i915]] *ERROR* failed to enable link training

Cc: Mika Kahola <mika.kahola@intel.com>
Cc: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com>
Cc: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
Fixes: 5fa836a9d859 ("drm/i915: DP link training optimization")
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
---
 drivers/gpu/drm/i915/intel_dp.c | 7 +++++++
 1 file changed, 7 insertions(+)

Comments

Jani Nikula Nov. 2, 2015, 8:15 a.m. UTC | #1 
On Fri, 30 Oct 2015, ville.syrjala@linux.intel.com wrote:
> From: Ville Syrjälä <ville.syrjala@linux.intel.com>
>
> My Lenovo STM STDP3100 miniDP->VGA dongle doesn't seem to like it when
> we try to start link training with non-zero vswing/preemphasis. So when
> the initial link training DPCD write fails, retry it with zero values.
>
> Fixes a bunch of errors like so:
> [drm:intel_dp_start_link_train [i915]] *ERROR* failed to enable link training
>
> Cc: Mika Kahola <mika.kahola@intel.com>
> Cc: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com>
> Cc: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
> Fixes: 5fa836a9d859 ("drm/i915: DP link training optimization")
> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>

Possibly

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=91393


> ---
>  drivers/gpu/drm/i915/intel_dp.c | 7 +++++++
>  1 file changed, 7 insertions(+)
>
> diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
> index ba4cbf5..9529a6e 100644
> --- a/drivers/gpu/drm/i915/intel_dp.c
> +++ b/drivers/gpu/drm/i915/intel_dp.c
> @@ -3750,10 +3750,17 @@ intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp)
>  
>  	DP |= DP_PORT_EN;
>  
> +again:
>  	/* clock recovery */
>  	if (!intel_dp_reset_link_train(intel_dp, &DP,
>  				       DP_TRAINING_PATTERN_1 |
>  				       DP_LINK_SCRAMBLING_DISABLE)) {
> +		if (intel_dp->train_set_valid) {
> +			DRM_DEBUG_KMS("Sink rejected link training request, trying again with zero values\n");
> +			intel_dp->train_set_valid = false;
> +			goto again;
> +		}
> +
>  		DRM_ERROR("failed to enable link training\n");
>  		return;
>  	}
Ander Conselvan de Oliveira Nov. 2, 2015, 1:21 p.m. UTC | #2 
On Fri, 2015-10-30 at 18:47 +0200, ville.syrjala@linux.intel.com wrote:
> From: Ville Syrjälä <ville.syrjala@linux.intel.com>

> 

> My Lenovo STM STDP3100 miniDP->VGA dongle doesn't seem to like it when

> we try to start link training with non-zero vswing/preemphasis. So when

> the initial link training DPCD write fails, retry it with zero values.


Does the device NACKs the request?

Ander

> 

> Fixes a bunch of errors like so:

> [drm:intel_dp_start_link_train [i915]] *ERROR* failed to enable link training

> 

> Cc: Mika Kahola <mika.kahola@intel.com>

> Cc: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com>

> Cc: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>

> Fixes: 5fa836a9d859 ("drm/i915: DP link training optimization")

> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>

> ---

>  drivers/gpu/drm/i915/intel_dp.c | 7 +++++++

>  1 file changed, 7 insertions(+)

> 

> diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c

> index ba4cbf5..9529a6e 100644

> --- a/drivers/gpu/drm/i915/intel_dp.c

> +++ b/drivers/gpu/drm/i915/intel_dp.c

> @@ -3750,10 +3750,17 @@ intel_dp_link_training_clock_recovery(struct intel_dp

> *intel_dp)

>  

>  	DP |= DP_PORT_EN;

>  

> +again:

>  	/* clock recovery */

>  	if (!intel_dp_reset_link_train(intel_dp, &DP,

>  				       DP_TRAINING_PATTERN_1 |

>  				       DP_LINK_SCRAMBLING_DISABLE)) {

> +		if (intel_dp->train_set_valid) {

> +			DRM_DEBUG_KMS("Sink rejected link training request,

> trying again with zero values\n");

> +			intel_dp->train_set_valid = false;

> +			goto again;

> +		}

> +

>  		DRM_ERROR("failed to enable link training\n");

>  		return;

>  	}
---------------------------------------------------------------------
Intel Finland Oy
Registered Address: PL 281, 00181 Helsinki 
Business Identity Code: 0357606 - 4 
Domiciled in Helsinki 

This e-mail and any attachments may contain confidential material for
the sole use of the intended recipient(s). Any review or distribution
by others is strictly prohibited. If you are not the intended
recipient, please contact the sender and delete all copies.
Ville Syrjälä Nov. 2, 2015, 2:55 p.m. UTC | #3 
On Mon, Nov 02, 2015 at 01:21:23PM +0000, Conselvan De Oliveira, Ander wrote:
> On Fri, 2015-10-30 at 18:47 +0200, ville.syrjala@linux.intel.com wrote:
> > From: Ville Syrjälä <ville.syrjala@linux.intel.com>
> > 
> > My Lenovo STM STDP3100 miniDP->VGA dongle doesn't seem to like it when
> > we try to start link training with non-zero vswing/preemphasis. So when
> > the initial link training DPCD write fails, retry it with zero values.
> 
> Does the device NACKs the request?

That was my assumption, but actually it just defers it to death. I had to
bump the retry count to ~48 to make it work without this patch. That was
with drm.debug=0, with drm.debug=0xe it seems to work even with the
current retry count of 32. Which also makes me think we should run the
bat stuff both with debugs on and off to catch more problems.

> 
> Ander
> 
> > 
> > Fixes a bunch of errors like so:
> > [drm:intel_dp_start_link_train [i915]] *ERROR* failed to enable link training
> > 
> > Cc: Mika Kahola <mika.kahola@intel.com>
> > Cc: Sivakumar Thulasimani <sivakumar.thulasimani@intel.com>
> > Cc: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
> > Fixes: 5fa836a9d859 ("drm/i915: DP link training optimization")
> > Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
> > ---
> >  drivers/gpu/drm/i915/intel_dp.c | 7 +++++++
> >  1 file changed, 7 insertions(+)
> > 
> > diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
> > index ba4cbf5..9529a6e 100644
> > --- a/drivers/gpu/drm/i915/intel_dp.c
> > +++ b/drivers/gpu/drm/i915/intel_dp.c
> > @@ -3750,10 +3750,17 @@ intel_dp_link_training_clock_recovery(struct intel_dp
> > *intel_dp)
> >  
> >  	DP |= DP_PORT_EN;
> >  
> > +again:
> >  	/* clock recovery */
> >  	if (!intel_dp_reset_link_train(intel_dp, &DP,
> >  				       DP_TRAINING_PATTERN_1 |
> >  				       DP_LINK_SCRAMBLING_DISABLE)) {
> > +		if (intel_dp->train_set_valid) {
> > +			DRM_DEBUG_KMS("Sink rejected link training request,
> > trying again with zero values\n");
> > +			intel_dp->train_set_valid = false;
> > +			goto again;
> > +		}
> > +
> >  		DRM_ERROR("failed to enable link training\n");
> >  		return;
> >  	}
diff mbox

Patch

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index ba4cbf5..9529a6e 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -3750,10 +3750,17 @@  intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp)
 
 	DP |= DP_PORT_EN;
 
+again:
 	/* clock recovery */
 	if (!intel_dp_reset_link_train(intel_dp, &DP,
 				       DP_TRAINING_PATTERN_1 |
 				       DP_LINK_SCRAMBLING_DISABLE)) {
+		if (intel_dp->train_set_valid) {
+			DRM_DEBUG_KMS("Sink rejected link training request, trying again with zero values\n");
+			intel_dp->train_set_valid = false;
+			goto again;
+		}
+
 		DRM_ERROR("failed to enable link training\n");
 		return;
 	}