[PATCH] drm/i915: skip eDP hw probe when connector forced off via cmdline

[PATCH] drm/i915: skip eDP hw probe when connector forced off via cmdline

[LIOU, Mei Fan]

When eDP connector is forced off (video=eDP-1:d),
intel_edp_init_connector() ignores the force flag and still probes
the AUX channel for DPCD and EDID.

video=eDP-1:d sets connector->force = DRM_FORCE_OFF via
drm_connector_get_cmdline_mode() inside drm_connector_init_with_ddc(),
which logs 'forcing eDP-1 connector off'. However,
intel_edp_init_connector() is called immediately after and ignores
the force flag, so AUX/PPS probing still occurs.

Fix this by checking connector->base.force == DRM_FORCE_OFF in
intel_edp_init_connector() after the intel_dp_is_edp() check, before
any AUX/DPCD probing is attempted. When the connector is forced off,
log an informational message and goto out_vdd_off to properly clean
up PPS state and skip all remaining hardware probing.

Suggested-by: Arun R Murthy <arun.r.murthy@intel.com>
Signed-off-by: LIOU, Mei Fan <mei.fan.liou@intel.com>
---
 drivers/gpu/drm/i915/display/intel_dp.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 3569e61e7fee..fa61b299f83e 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -7057,6 +7057,17 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
 
 	intel_alpm_init(intel_dp);
 
+	/*
+	 * If the connector has been forced off via the kernel cmdline
+	 * (e.g. video=eDP-1:d), skip DPCD/AUX probing.
+	 */
+	if (connector->base.force == DRM_FORCE_OFF) {
+		drm_info(display->drm,
+			 "[ENCODER:%d:%s] eDP disabled via cmdline, skipping eDP init\n",
+			 encoder->base.base.id, encoder->base.name);
+		goto out_vdd_off;
+	}
+
 	/* Cache DPCD and EDID for edp. */
 	has_dpcd = intel_edp_init_dpcd(intel_dp, connector);
 
-- 
2.52.0

Re: [PATCH] drm/i915: skip eDP hw probe when connector forced off via cmdline

[Ville Syrjälä]

On Tue, Jul 07, 2026 at 04:57:04PM +0800, LIOU, Mei Fan wrote:
> When eDP connector is forced off (video=eDP-1:d),
> intel_edp_init_connector() ignores the force flag and still probes
> the AUX channel for DPCD and EDID.
> 
> video=eDP-1:d sets connector->force = DRM_FORCE_OFF via
> drm_connector_get_cmdline_mode() inside drm_connector_init_with_ddc(),
> which logs 'forcing eDP-1 connector off'. However,
> intel_edp_init_connector() is called immediately after and ignores
> the force flag, so AUX/PPS probing still occurs.
> 
> Fix this by checking connector->base.force == DRM_FORCE_OFF in
> intel_edp_init_connector() after the intel_dp_is_edp() check, before
> any AUX/DPCD probing is attempted. When the connector is forced off,
> log an informational message and goto out_vdd_off to properly clean
> up PPS state and skip all remaining hardware probing.
> 
> Suggested-by: Arun R Murthy <arun.r.murthy@intel.com>
> Signed-off-by: LIOU, Mei Fan <mei.fan.liou@intel.com>
> ---
>  drivers/gpu/drm/i915/display/intel_dp.c | 11 +++++++++++
>  1 file changed, 11 insertions(+)
> 
> diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
> index 3569e61e7fee..fa61b299f83e 100644
> --- a/drivers/gpu/drm/i915/display/intel_dp.c
> +++ b/drivers/gpu/drm/i915/display/intel_dp.c
> @@ -7057,6 +7057,17 @@ static bool intel_edp_init_connector(struct intel_dp *intel_dp,
>  
>  	intel_alpm_init(intel_dp);
>  
> +	/*
> +	 * If the connector has been forced off via the kernel cmdline
> +	 * (e.g. video=eDP-1:d), skip DPCD/AUX probing.
> +	 */
> +	if (connector->base.force == DRM_FORCE_OFF) {
> +		drm_info(display->drm,
> +			 "[ENCODER:%d:%s] eDP disabled via cmdline, skipping eDP init\n",
> +			 encoder->base.base.id, encoder->base.name);
> +		goto out_vdd_off;
> +	}

This doesn't looks sane if the eDP panel is actually there. We need to
turn it off gracefully.

> +
>  	/* Cache DPCD and EDID for edp. */
>  	has_dpcd = intel_edp_init_dpcd(intel_dp, connector);
>  
> -- 
> 2.52.0

-- 
Ville Syrjälä
Intel

RE: [PATCH] drm/i915: skip eDP hw probe when connector forced off via cmdline

[Murthy, Arun R]

> On Tue, Jul 07, 2026 at 04:57:04PM +0800, LIOU, Mei Fan wrote:
> > When eDP connector is forced off (video=eDP-1:d),
> > intel_edp_init_connector() ignores the force flag and still probes the
> > AUX channel for DPCD and EDID.
> >
> > video=eDP-1:d sets connector->force = DRM_FORCE_OFF via
> > drm_connector_get_cmdline_mode() inside drm_connector_init_with_ddc(),
> > which logs 'forcing eDP-1 connector off'. However,
> > intel_edp_init_connector() is called immediately after and ignores the
> > force flag, so AUX/PPS probing still occurs.
> >
> > Fix this by checking connector->base.force == DRM_FORCE_OFF in
> > intel_edp_init_connector() after the intel_dp_is_edp() check, before
> > any AUX/DPCD probing is attempted. When the connector is forced off,
> > log an informational message and goto out_vdd_off to properly clean up
> > PPS state and skip all remaining hardware probing.
> >
> > Suggested-by: Arun R Murthy <arun.r.murthy@intel.com>
> > Signed-off-by: LIOU, Mei Fan <mei.fan.liou@intel.com>
> > ---
> >  drivers/gpu/drm/i915/display/intel_dp.c | 11 +++++++++++
> >  1 file changed, 11 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/i915/display/intel_dp.c
> > b/drivers/gpu/drm/i915/display/intel_dp.c
> > index 3569e61e7fee..fa61b299f83e 100644
> > --- a/drivers/gpu/drm/i915/display/intel_dp.c
> > +++ b/drivers/gpu/drm/i915/display/intel_dp.c
> > @@ -7057,6 +7057,17 @@ static bool intel_edp_init_connector(struct
> > intel_dp *intel_dp,
> >
> >  	intel_alpm_init(intel_dp);
> >
> > +	/*
> > +	 * If the connector has been forced off via the kernel cmdline
> > +	 * (e.g. video=eDP-1:d), skip DPCD/AUX probing.
> > +	 */
> > +	if (connector->base.force == DRM_FORCE_OFF) {
> > +		drm_info(display->drm,
> > +			 "[ENCODER:%d:%s] eDP disabled via cmdline, skipping
> eDP init\n",
> > +			 encoder->base.base.id, encoder->base.name);
> > +		goto out_vdd_off;
> > +	}
> 
> This doesn't looks sane if the eDP panel is actually there. We need to turn it off
> gracefully.
> 
eDP panel is not connected.
Hence adding the parameter video=edp-1:d should bypass the edp probe. But even with adding this parameter edp is being probed and failure/error messages are being cascaded in dmesg.
Eventually trying to fix the parameter video=edp-1:d !

Thanks and Regards,
Arun R Murthy
-------------------
> > +
> >  	/* Cache DPCD and EDID for edp. */
> >  	has_dpcd = intel_edp_init_dpcd(intel_dp, connector);
> >
> > --
> > 2.52.0
> 
> --
> Ville Syrjälä
> Intel