[PATCH] drm/i915/cnl: DDIA Lane capability bit not set in clone mode

[PATCH] drm/i915/cnl: DDIA Lane capability bit not set in clone mode

[Rodrigo Vivi]

From: Clint Taylor <clinton.a.taylor@intel.com>

DDIA Lane capability control 4 lane bit is not being set by firmware during
clone mode boot. This occurs when multiple monitors are connected during
boot. The driver will configure the port for 2 lane maximum width if this
bit is not set.

Cc: Mika Kahola <mika.kahola@intel.com>
Signed-off-by: Clint Taylor <clinton.a.taylor@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
---
 drivers/gpu/drm/i915/intel_ddi.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index a9c0c16e3838..0ad915d71132 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -2791,9 +2791,10 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
 	 * configuration so that we use the proper lane count for our
 	 * calculations.
 	 */
-	if (IS_GEN9_LP(dev_priv) && port == PORT_A) {
+	if ((IS_GEN9_LP(dev_priv) || IS_CANNONLAKE(dev_priv)) &&
+	    port == PORT_A) {
 		if (!(intel_dig_port->saved_port_bits & DDI_A_4_LANES)) {
-			DRM_DEBUG_KMS("BXT BIOS forgot to set DDI_A_4_LANES for port A; fixing\n");
+			DRM_DEBUG_KMS("BIOS forgot to set DDI_A_4_LANES for port A\n");
 			intel_dig_port->saved_port_bits |= DDI_A_4_LANES;
 			max_lanes = 4;
 		}
-- 
2.13.5

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

✓ Fi.CI.BAT: success for drm/i915/cnl: DDIA Lane capability bit not set in clone mode (rev2)

[Patchwork]

== Series Details ==

Series: drm/i915/cnl: DDIA Lane capability bit not set in clone mode (rev2)
URL   : https://patchwork.freedesktop.org/series/28204/
State : success

== Summary ==

Series 28204v2 drm/i915/cnl: DDIA Lane capability bit not set in clone mode
https://patchwork.freedesktop.org/api/1.0/series/28204/revisions/2/mbox/

Test gem_sync:
        Subgroup basic-store-all:
                pass       -> FAIL       (fi-ivb-3520m) fdo#100007

fdo#100007 https://bugs.freedesktop.org/show_bug.cgi?id=100007

fi-bdw-5557u     total:289  pass:268  dwarn:0   dfail:0   fail:0   skip:21  time:438s
fi-bdw-gvtdvm    total:289  pass:265  dwarn:0   dfail:0   fail:0   skip:24  time:454s
fi-blb-e6850     total:289  pass:223  dwarn:1   dfail:0   fail:0   skip:65  time:369s
fi-bsw-n3050     total:289  pass:243  dwarn:0   dfail:0   fail:0   skip:46  time:527s
fi-bwr-2160      total:289  pass:183  dwarn:0   dfail:0   fail:0   skip:106 time:264s
fi-bxt-dsi       total:289  pass:259  dwarn:0   dfail:0   fail:0   skip:30  time:497s
fi-bxt-j4205     total:289  pass:260  dwarn:0   dfail:0   fail:0   skip:29  time:491s
fi-byt-j1900     total:289  pass:253  dwarn:1   dfail:0   fail:0   skip:35  time:496s
fi-byt-n2820     total:289  pass:249  dwarn:1   dfail:0   fail:0   skip:39  time:476s
fi-elk-e7500     total:289  pass:229  dwarn:0   dfail:0   fail:0   skip:60  time:419s
fi-gdg-551       total:289  pass:178  dwarn:1   dfail:0   fail:1   skip:109 time:249s
fi-glk-1         total:289  pass:261  dwarn:0   dfail:0   fail:0   skip:28  time:574s
fi-hsw-4770r     total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:422s
fi-ilk-650       total:289  pass:228  dwarn:0   dfail:0   fail:0   skip:61  time:428s
fi-ivb-3520m     total:289  pass:259  dwarn:0   dfail:0   fail:1   skip:29  time:489s
fi-ivb-3770      total:289  pass:260  dwarn:0   dfail:0   fail:0   skip:29  time:459s
fi-kbl-7500u     total:289  pass:264  dwarn:1   dfail:0   fail:0   skip:24  time:492s
fi-kbl-7560u     total:289  pass:270  dwarn:0   dfail:0   fail:0   skip:19  time:576s
fi-kbl-7567u     total:289  pass:269  dwarn:0   dfail:0   fail:0   skip:20  time:475s
fi-kbl-r         total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:582s
fi-pnv-d510      total:289  pass:222  dwarn:1   dfail:0   fail:0   skip:66  time:545s
fi-skl-6260u     total:289  pass:269  dwarn:0   dfail:0   fail:0   skip:20  time:449s
fi-skl-6700hq    total:289  pass:263  dwarn:0   dfail:0   fail:0   skip:26  time:646s
fi-skl-6700k     total:289  pass:265  dwarn:0   dfail:0   fail:0   skip:24  time:526s
fi-skl-6770hq    total:289  pass:269  dwarn:0   dfail:0   fail:0   skip:20  time:492s
fi-skl-gvtdvm    total:289  pass:266  dwarn:0   dfail:0   fail:0   skip:23  time:455s
fi-snb-2520m     total:289  pass:250  dwarn:0   dfail:0   fail:0   skip:39  time:577s
fi-snb-2600      total:289  pass:249  dwarn:0   dfail:0   fail:0   skip:40  time:421s

0673a9da7359e1575ac4581254110a0b4bc06c96 drm-tip: 2017y-10m-16d-20h-13m-40s UTC integration manifest
21b047663119 drm/i915/cnl: DDIA Lane capability bit not set in clone mode

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_6060/
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH] drm/i915/cnl: DDIA Lane capability bit not set in clone mode

[Jani Nikula]

On Mon, 16 Oct 2017, Rodrigo Vivi <rodrigo.vivi@intel.com> wrote:
> From: Clint Taylor <clinton.a.taylor@intel.com>
>
> DDIA Lane capability control 4 lane bit is not being set by firmware during
> clone mode boot. This occurs when multiple monitors are connected during
> boot. The driver will configure the port for 2 lane maximum width if this
> bit is not set.

Please be more specific about what you mean with clone mode.

>
> Cc: Mika Kahola <mika.kahola@intel.com>
> Signed-off-by: Clint Taylor <clinton.a.taylor@intel.com>
> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
> ---
>  drivers/gpu/drm/i915/intel_ddi.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
> index a9c0c16e3838..0ad915d71132 100644
> --- a/drivers/gpu/drm/i915/intel_ddi.c
> +++ b/drivers/gpu/drm/i915/intel_ddi.c
> @@ -2791,9 +2791,10 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
>  	 * configuration so that we use the proper lane count for our
>  	 * calculations.
>  	 */

The comment above needs updating too. Also, it doesn't say anything
about cloning.

BR,
Jani.

> -	if (IS_GEN9_LP(dev_priv) && port == PORT_A) {
> +	if ((IS_GEN9_LP(dev_priv) || IS_CANNONLAKE(dev_priv)) &&
> +	    port == PORT_A) {
>  		if (!(intel_dig_port->saved_port_bits & DDI_A_4_LANES)) {
> -			DRM_DEBUG_KMS("BXT BIOS forgot to set DDI_A_4_LANES for port A; fixing\n");
> +			DRM_DEBUG_KMS("BIOS forgot to set DDI_A_4_LANES for port A\n");
>  			intel_dig_port->saved_port_bits |= DDI_A_4_LANES;
>  			max_lanes = 4;
>  		}

-- 
Jani Nikula, Intel Open Source Technology Center
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH] drm/i915/cnl: DDIA Lane capability bit not set in clone mode

[Ville Syrjälä]

On Mon, Oct 16, 2017 at 04:44:49PM -0700, Rodrigo Vivi wrote:
> From: Clint Taylor <clinton.a.taylor@intel.com>
> 
> DDIA Lane capability control 4 lane bit is not being set by firmware during
> clone mode boot. This occurs when multiple monitors are connected during
> boot. The driver will configure the port for 2 lane maximum width if this
> bit is not set.
> 
> Cc: Mika Kahola <mika.kahola@intel.com>
> Signed-off-by: Clint Taylor <clinton.a.taylor@intel.com>
> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
> ---
>  drivers/gpu/drm/i915/intel_ddi.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
> index a9c0c16e3838..0ad915d71132 100644
> --- a/drivers/gpu/drm/i915/intel_ddi.c
> +++ b/drivers/gpu/drm/i915/intel_ddi.c
> @@ -2791,9 +2791,10 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
>  	 * configuration so that we use the proper lane count for our
>  	 * calculations.
>  	 */
> -	if (IS_GEN9_LP(dev_priv) && port == PORT_A) {
> +	if ((IS_GEN9_LP(dev_priv) || IS_CANNONLAKE(dev_priv)) &&
> +	    port == PORT_A) {
>  		if (!(intel_dig_port->saved_port_bits & DDI_A_4_LANES)) {
> -			DRM_DEBUG_KMS("BXT BIOS forgot to set DDI_A_4_LANES for port A; fixing\n");
> +			DRM_DEBUG_KMS("BIOS forgot to set DDI_A_4_LANES for port A\n");
>  			intel_dig_port->saved_port_bits |= DDI_A_4_LANES;
>  			max_lanes = 4;
>  		}

CNL has DDI E so this doesn't make sense. If there are CNLs out there
that don't actually use DDI E but forget to configure the bifurcation
coreectly, then we'll need a fancier way to detect that. Ie. we need to
be sure DDI E isn't going to be used before we can force DDI A to use
4 lanes.

-- 
Ville Syrjälä
Intel OTC
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

✗ Fi.CI.IGT: failure for drm/i915/cnl: DDIA Lane capability bit not set in clone mode (rev2)

[Patchwork]

== Series Details ==

Series: drm/i915/cnl: DDIA Lane capability bit not set in clone mode (rev2)
URL   : https://patchwork.freedesktop.org/series/28204/
State : failure

== Summary ==

Test kms_frontbuffer_tracking:
        Subgroup fbcpsr-2p-scndscrn-pri-indfb-draw-mmap-cpu:
                skip       -> INCOMPLETE (shard-hsw)
Test kms_flip:
        Subgroup flip-vs-modeset-interruptible:
                pass       -> DMESG-WARN (shard-hsw) fdo#102557
Test gem_exec_params:
        Subgroup invalid-bsd2-flag-on-render:
                pass       -> INCOMPLETE (shard-hsw)
Test perf:
        Subgroup oa-exponents:
                fail       -> PASS       (shard-hsw) fdo#102254

fdo#102557 https://bugs.freedesktop.org/show_bug.cgi?id=102557
fdo#102254 https://bugs.freedesktop.org/show_bug.cgi?id=102254

shard-hsw        total:2349 pass:1262 dwarn:1   dfail:0   fail:9   skip:978 time:8125s

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_6060/shards.html
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH] drm/i915/cnl: DDIA Lane capability bit not set in clone mode

[Ville Syrjälä]

On Tue, Oct 17, 2017 at 03:10:16PM +0300, Ville Syrjälä wrote:
> On Mon, Oct 16, 2017 at 04:44:49PM -0700, Rodrigo Vivi wrote:
> > From: Clint Taylor <clinton.a.taylor@intel.com>
> > 
> > DDIA Lane capability control 4 lane bit is not being set by firmware during
> > clone mode boot. This occurs when multiple monitors are connected during
> > boot. The driver will configure the port for 2 lane maximum width if this
> > bit is not set.
> > 
> > Cc: Mika Kahola <mika.kahola@intel.com>
> > Signed-off-by: Clint Taylor <clinton.a.taylor@intel.com>
> > Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
> > ---
> >  drivers/gpu/drm/i915/intel_ddi.c | 5 +++--
> >  1 file changed, 3 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
> > index a9c0c16e3838..0ad915d71132 100644
> > --- a/drivers/gpu/drm/i915/intel_ddi.c
> > +++ b/drivers/gpu/drm/i915/intel_ddi.c
> > @@ -2791,9 +2791,10 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
> >  	 * configuration so that we use the proper lane count for our
> >  	 * calculations.
> >  	 */
> > -	if (IS_GEN9_LP(dev_priv) && port == PORT_A) {
> > +	if ((IS_GEN9_LP(dev_priv) || IS_CANNONLAKE(dev_priv)) &&
> > +	    port == PORT_A) {
> >  		if (!(intel_dig_port->saved_port_bits & DDI_A_4_LANES)) {
> > -			DRM_DEBUG_KMS("BXT BIOS forgot to set DDI_A_4_LANES for port A; fixing\n");
> > +			DRM_DEBUG_KMS("BIOS forgot to set DDI_A_4_LANES for port A\n");
> >  			intel_dig_port->saved_port_bits |= DDI_A_4_LANES;
> >  			max_lanes = 4;
> >  		}
> 
> CNL has DDI E so this doesn't make sense. If there are CNLs out there
> that don't actually use DDI E but forget to configure the bifurcation
> coreectly, then we'll need a fancier way to detect that. Ie. we need to
> be sure DDI E isn't going to be used before we can force DDI A to use
> 4 lanes.

Oh, we'd really need some way to make sure all four lanes from DDI A are
really connected to the panel. Otherwise link training would surely
fail if we try to use four lanes.

-- 
Ville Syrjälä
Intel OTC
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH] drm/i915/cnl: DDIA Lane capability bit not set in clone mode

[Rodrigo Vivi]

On Tue, Oct 17, 2017 at 12:10:16PM +0000, Ville Syrjälä wrote:
> On Mon, Oct 16, 2017 at 04:44:49PM -0700, Rodrigo Vivi wrote:
> > From: Clint Taylor <clinton.a.taylor@intel.com>
> > 
> > DDIA Lane capability control 4 lane bit is not being set by firmware during
> > clone mode boot. This occurs when multiple monitors are connected during
> > boot. The driver will configure the port for 2 lane maximum width if this
> > bit is not set.
> > 
> > Cc: Mika Kahola <mika.kahola@intel.com>
> > Signed-off-by: Clint Taylor <clinton.a.taylor@intel.com>
> > Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
> > ---
> >  drivers/gpu/drm/i915/intel_ddi.c | 5 +++--
> >  1 file changed, 3 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
> > index a9c0c16e3838..0ad915d71132 100644
> > --- a/drivers/gpu/drm/i915/intel_ddi.c
> > +++ b/drivers/gpu/drm/i915/intel_ddi.c
> > @@ -2791,9 +2791,10 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
> >  	 * configuration so that we use the proper lane count for our
> >  	 * calculations.
> >  	 */
> > -	if (IS_GEN9_LP(dev_priv) && port == PORT_A) {
> > +	if ((IS_GEN9_LP(dev_priv) || IS_CANNONLAKE(dev_priv)) &&
> > +	    port == PORT_A) {
> >  		if (!(intel_dig_port->saved_port_bits & DDI_A_4_LANES)) {
> > -			DRM_DEBUG_KMS("BXT BIOS forgot to set DDI_A_4_LANES for port A; fixing\n");
> > +			DRM_DEBUG_KMS("BIOS forgot to set DDI_A_4_LANES for port A\n");
> >  			intel_dig_port->saved_port_bits |= DDI_A_4_LANES;
> >  			max_lanes = 4;
> >  		}
> 
> CNL has DDI E so this doesn't make sense. If there are CNLs out there
> that don't actually use DDI E but forget to configure the bifurcation
> coreectly, then we'll need a fancier way to detect that. Ie. we need to
> be sure DDI E isn't going to be used before we can force DDI A to use
> 4 lanes.

We got the confirmation that DDI E is not there for any of the current SKUs
what support is currently merged.

DDI E is only available on the SKUs that posted yesterday... the one that
supports DDI F that is the proper split. Also when DDI F is in use DDI E
cannot be used because the interrupts handling.

So apparently this is not going to be used at all. But I agree that it
would be good if we could have a smarter way to detect it... Any idea?

Thanks,
Rodrigo.

> 
> -- 
> Ville Syrjälä
> Intel OTC
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH] drm/i915/cnl: DDIA Lane capability bit not set in clone mode

[Rodrigo Vivi]

On Tue, Oct 17, 2017 at 08:04:04AM +0000, Jani Nikula wrote:
> On Mon, 16 Oct 2017, Rodrigo Vivi <rodrigo.vivi@intel.com> wrote:
> > From: Clint Taylor <clinton.a.taylor@intel.com>
> >
> > DDIA Lane capability control 4 lane bit is not being set by firmware during
> > clone mode boot. This occurs when multiple monitors are connected during
> > boot. The driver will configure the port for 2 lane maximum width if this
> > bit is not set.
> 
> Please be more specific about what you mean with clone mode.

Oh... yes, we need to change the comment.

I didn't see any relation here to the clone mode.
If eDP comes back on boot and HDMI comes later everything worked.

If HDMI is also present during boot at some point during initialization
we have probably a long pulse on edp that hits our check of max_level
and change that to 2, because this bit is unset as we were expecting.

I will grab more information here on the flow and proper update the commit message.

> 
> >
> > Cc: Mika Kahola <mika.kahola@intel.com>
> > Signed-off-by: Clint Taylor <clinton.a.taylor@intel.com>
> > Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
> > ---
> >  drivers/gpu/drm/i915/intel_ddi.c | 5 +++--
> >  1 file changed, 3 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
> > index a9c0c16e3838..0ad915d71132 100644
> > --- a/drivers/gpu/drm/i915/intel_ddi.c
> > +++ b/drivers/gpu/drm/i915/intel_ddi.c
> > @@ -2791,9 +2791,10 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
> >  	 * configuration so that we use the proper lane count for our
> >  	 * calculations.
> >  	 */
> 
> The comment above needs updating too. Also, it doesn't say anything
> about cloning.

I will update this comment also. But also not related to clone mode I believe.

> 
> BR,
> Jani.
> 
> > -	if (IS_GEN9_LP(dev_priv) && port == PORT_A) {
> > +	if ((IS_GEN9_LP(dev_priv) || IS_CANNONLAKE(dev_priv)) &&
> > +	    port == PORT_A) {
> >  		if (!(intel_dig_port->saved_port_bits & DDI_A_4_LANES)) {
> > -			DRM_DEBUG_KMS("BXT BIOS forgot to set DDI_A_4_LANES for port A; fixing\n");
> > +			DRM_DEBUG_KMS("BIOS forgot to set DDI_A_4_LANES for port A\n");
> >  			intel_dig_port->saved_port_bits |= DDI_A_4_LANES;
> >  			max_lanes = 4;
> >  		}
> 
> -- 
> Jani Nikula, Intel Open Source Technology Center
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH] drm/i915/cnl: DDIA Lane capability bit not set in clone mode

[Ville Syrjälä]

On Tue, Oct 17, 2017 at 09:27:39AM -0700, Rodrigo Vivi wrote:
> On Tue, Oct 17, 2017 at 12:10:16PM +0000, Ville Syrjälä wrote:
> > On Mon, Oct 16, 2017 at 04:44:49PM -0700, Rodrigo Vivi wrote:
> > > From: Clint Taylor <clinton.a.taylor@intel.com>
> > > 
> > > DDIA Lane capability control 4 lane bit is not being set by firmware during
> > > clone mode boot. This occurs when multiple monitors are connected during
> > > boot. The driver will configure the port for 2 lane maximum width if this
> > > bit is not set.
> > > 
> > > Cc: Mika Kahola <mika.kahola@intel.com>
> > > Signed-off-by: Clint Taylor <clinton.a.taylor@intel.com>
> > > Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
> > > ---
> > >  drivers/gpu/drm/i915/intel_ddi.c | 5 +++--
> > >  1 file changed, 3 insertions(+), 2 deletions(-)
> > > 
> > > diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
> > > index a9c0c16e3838..0ad915d71132 100644
> > > --- a/drivers/gpu/drm/i915/intel_ddi.c
> > > +++ b/drivers/gpu/drm/i915/intel_ddi.c
> > > @@ -2791,9 +2791,10 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
> > >  	 * configuration so that we use the proper lane count for our
> > >  	 * calculations.
> > >  	 */
> > > -	if (IS_GEN9_LP(dev_priv) && port == PORT_A) {
> > > +	if ((IS_GEN9_LP(dev_priv) || IS_CANNONLAKE(dev_priv)) &&
> > > +	    port == PORT_A) {
> > >  		if (!(intel_dig_port->saved_port_bits & DDI_A_4_LANES)) {
> > > -			DRM_DEBUG_KMS("BXT BIOS forgot to set DDI_A_4_LANES for port A; fixing\n");
> > > +			DRM_DEBUG_KMS("BIOS forgot to set DDI_A_4_LANES for port A\n");
> > >  			intel_dig_port->saved_port_bits |= DDI_A_4_LANES;
> > >  			max_lanes = 4;
> > >  		}
> > 
> > CNL has DDI E so this doesn't make sense. If there are CNLs out there
> > that don't actually use DDI E but forget to configure the bifurcation
> > coreectly, then we'll need a fancier way to detect that. Ie. we need to
> > be sure DDI E isn't going to be used before we can force DDI A to use
> > 4 lanes.
> 
> We got the confirmation that DDI E is not there for any of the current SKUs
> what support is currently merged.
> 
> DDI E is only available on the SKUs that posted yesterday... the one that
> supports DDI F that is the proper split. Also when DDI F is in use DDI E
> cannot be used because the interrupts handling.
> 
> So apparently this is not going to be used at all. But I agree that it
> would be good if we could have a smarter way to detect it... Any idea?

Looks like DDI E doesn't have a hardware strap, so I suppose you could
just check vbt.ddi_port_info. That still doesn't protect us against
accidentally using 4 lanes when we should have used 2. Not sure there's
anything in VBT that could help us there.

-- 
Ville Syrjälä
Intel OTC
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH] drm/i915/cnl: DDIA Lane capability bit not set in clone mode

[Rodrigo Vivi]

we confirmed that with current supported CNL skus using IS_CANNONLAKE
is enogh for now.

So feel free to change that if and use

Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>

on the v2....


On Tue, Aug 1, 2017 at 1:23 PM, Vivi, Rodrigo <rodrigo.vivi@intel.com> wrote:
> + Art, couple questions below
>
> On Tue, 2017-08-01 at 09:56 -0700, clinton.a.taylor@intel.com wrote:
>> From: Clint Taylor <clinton.a.taylor@intel.com>
>>
>> DDIA Lane capability control 4 lane bit is not being set by firmware during
>> clone mode boot. This occurs when multiple monitors are connected during
>> boot. The driver will configure the port for 2 lane maximum width if this
>> bit is not set.
>>
>> Once DDIA/E lane split is supported in vbt and the i915 driver we will need
>> to revisit this code.
>>
>> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
>> Signed-off-by: Clint Taylor <clinton.a.taylor@intel.com>
>> ---
>>  drivers/gpu/drm/i915/intel_ddi.c | 5 +++--
>>  1 file changed, 3 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
>> index 494fbe0..e7644b4 100644
>> --- a/drivers/gpu/drm/i915/intel_ddi.c
>> +++ b/drivers/gpu/drm/i915/intel_ddi.c
>> @@ -2713,9 +2713,10 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
>
> we would need to fix more comment lines:
>
> /*
> * Bspec says that DDI_A_4_LANES is the only supported
> configuration
>          * for Broxton.  Yet some BIOS fail to set this bit on port A if
> eDP
>          * wasn't lit up at boot.  Force this bit on in our internal
>>        * configuration so that we use the proper lane count for our
>>        * calculations.
>>        */
>
> But I believe the approach we currently have with this might not be
> optimal. If BIOS is not bringing the port A up why should we expect it
> to set this bit ever? We probably need to be able to do it by ourselves,
> without expecting others to do...
>
> However, I've never seen a production BIOS that never brings port A when
> it is present. And it is also true that spec tells: "This field must be
> programmed at system boot based on board configuration and may not be
> changed afterwards."
>
> So I have kind of mixed feelings here on this bit.
>
> For BXT it is so clear: "Not supported on Broxton."
>
> Also few other future cases the 0 won't be valid, but not entirely sure
> if this is always true. Art?
>
> I just noticed on spec that DDI E for CNL-U and CNL-Y are marked as "not
> supported", what means that for that SKU we should be able to add this
> workaround here, but at same time means that we need to remove the power
> wells support for DDI-E for these SKUs probably.
>
> Art, could/should we set this bit blindly for all CNL?
>
> and if yes:
>
> Art, will it always be a decision based by SKUs now on? Or should we
> really rely on BIOS?
>
> Is there another way to detect the port E presence? I don't believe VBT
> has that info in a reliable way, does it? Otherwise we would be using it
> already for the missed straps...
>
>
> Thanks,
> Rodrigo.
>
>> -     if (IS_GEN9_LP(dev_priv) && port == PORT_A) {
>> +     if ((IS_GEN9_LP(dev_priv) || INTEL_GEN(dev_priv) >= 10) &&
>> +         port == PORT_A) {
>>               if (!(intel_dig_port->saved_port_bits & DDI_A_4_LANES)) {
>> -                     DRM_DEBUG_KMS("BXT BIOS forgot to set DDI_A_4_LANES for port A; fixing\n");
>> +                     DRM_DEBUG_KMS("BIOS forgot to set DDI_A_4_LANES for port A\n");
>>                       intel_dig_port->saved_port_bits |= DDI_A_4_LANES;
>>                       max_lanes = 4;
>>               }
>
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx



-- 
Rodrigo Vivi
Blog: http://blog.vivi.eng.br
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH] drm/i915/cnl: DDIA Lane capability bit not set in clone mode

[Vivi, Rodrigo]

+ Art, couple questions below

On Tue, 2017-08-01 at 09:56 -0700, clinton.a.taylor@intel.com wrote:
> From: Clint Taylor <clinton.a.taylor@intel.com>
> 
> DDIA Lane capability control 4 lane bit is not being set by firmware during
> clone mode boot. This occurs when multiple monitors are connected during
> boot. The driver will configure the port for 2 lane maximum width if this
> bit is not set.
> 
> Once DDIA/E lane split is supported in vbt and the i915 driver we will need
> to revisit this code.
> 
> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
> Signed-off-by: Clint Taylor <clinton.a.taylor@intel.com>
> ---
>  drivers/gpu/drm/i915/intel_ddi.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
> index 494fbe0..e7644b4 100644
> --- a/drivers/gpu/drm/i915/intel_ddi.c
> +++ b/drivers/gpu/drm/i915/intel_ddi.c
> @@ -2713,9 +2713,10 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)

we would need to fix more comment lines:

/*
* Bspec says that DDI_A_4_LANES is the only supported
configuration                                                                       
         * for Broxton.  Yet some BIOS fail to set this bit on port A if
eDP                                                                       
         * wasn't lit up at boot.  Force this bit on in our internal 
>  	 * configuration so that we use the proper lane count for our
>  	 * calculations.
>  	 */

But I believe the approach we currently have with this might not be
optimal. If BIOS is not bringing the port A up why should we expect it
to set this bit ever? We probably need to be able to do it by ourselves,
without expecting others to do...

However, I've never seen a production BIOS that never brings port A when
it is present. And it is also true that spec tells: "This field must be
programmed at system boot based on board configuration and may not be
changed afterwards."

So I have kind of mixed feelings here on this bit.

For BXT it is so clear: "Not supported on Broxton."

Also few other future cases the 0 won't be valid, but not entirely sure
if this is always true. Art?

I just noticed on spec that DDI E for CNL-U and CNL-Y are marked as "not
supported", what means that for that SKU we should be able to add this
workaround here, but at same time means that we need to remove the power
wells support for DDI-E for these SKUs probably.

Art, could/should we set this bit blindly for all CNL? 

and if yes:

Art, will it always be a decision based by SKUs now on? Or should we
really rely on BIOS?

Is there another way to detect the port E presence? I don't believe VBT
has that info in a reliable way, does it? Otherwise we would be using it
already for the missed straps...


Thanks,
Rodrigo.

> -	if (IS_GEN9_LP(dev_priv) && port == PORT_A) {
> +	if ((IS_GEN9_LP(dev_priv) || INTEL_GEN(dev_priv) >= 10) &&
> +	    port == PORT_A) {
>  		if (!(intel_dig_port->saved_port_bits & DDI_A_4_LANES)) {
> -			DRM_DEBUG_KMS("BXT BIOS forgot to set DDI_A_4_LANES for port A; fixing\n");
> +			DRM_DEBUG_KMS("BIOS forgot to set DDI_A_4_LANES for port A\n");
>  			intel_dig_port->saved_port_bits |= DDI_A_4_LANES;
>  			max_lanes = 4;
>  		}

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

[PATCH] drm/i915/cnl: DDIA Lane capability bit not set in clone mode

[clinton.a.taylor]

From: Clint Taylor <clinton.a.taylor@intel.com>

DDIA Lane capability control 4 lane bit is not being set by firmware during
clone mode boot. This occurs when multiple monitors are connected during
boot. The driver will configure the port for 2 lane maximum width if this
bit is not set.

Once DDIA/E lane split is supported in vbt and the i915 driver we will need
to revisit this code.

Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Clint Taylor <clinton.a.taylor@intel.com>
---
 drivers/gpu/drm/i915/intel_ddi.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 494fbe0..e7644b4 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -2713,9 +2713,10 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port)
 	 * configuration so that we use the proper lane count for our
 	 * calculations.
 	 */
-	if (IS_GEN9_LP(dev_priv) && port == PORT_A) {
+	if ((IS_GEN9_LP(dev_priv) || INTEL_GEN(dev_priv) >= 10) &&
+	    port == PORT_A) {
 		if (!(intel_dig_port->saved_port_bits & DDI_A_4_LANES)) {
-			DRM_DEBUG_KMS("BXT BIOS forgot to set DDI_A_4_LANES for port A; fixing\n");
+			DRM_DEBUG_KMS("BIOS forgot to set DDI_A_4_LANES for port A\n");
 			intel_dig_port->saved_port_bits |= DDI_A_4_LANES;
 			max_lanes = 4;
 		}
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx