[PATCH] drm/i915/dp: remove redundant link state check

[PATCH] drm/i915/dp: remove redundant link state check

[tarund18303]

From: tarun-tarun143 <tarund18303@gmail.com>

The DP link state is now checked and recovered after link training,
including after modesets. The same recovery path is also used from
the hotplug and HPD IRQ paths.

The link state check from intel_dp_detect() is therefore redundant.
Remove it and rely on the existing link recovery paths.

Signed-off-by: tarun-tarun143 <tarund18303@gmail.com>
---
 drivers/gpu/drm/i915/display/intel_dp.c | 12 ------------
 1 file changed, 12 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
index 113d767e62e9..ffd39206b958 100644
--- a/drivers/gpu/drm/i915/display/intel_dp.c
+++ b/drivers/gpu/drm/i915/display/intel_dp.c
@@ -6379,18 +6379,6 @@ intel_dp_detect(struct drm_connector *_connector,
 		status = connector_status_disconnected;
 		goto out_unset_edid;
 	}
-
-	/*
-	 * Some external monitors do not signal loss of link synchronization
-	 * with an IRQ_HPD, so force a link status check.
-	 *
-	 * TODO: this probably became redundant, so remove it: the link state
-	 * is rechecked/recovered now after modesets, where the loss of
-	 * synchronization tends to occur.
-	 */
-	if (!intel_dp_is_edp(intel_dp))
-		intel_dp_check_link_state(intel_dp);
-
 	/*
 	 * Clearing NACK and defer counts to get their exact values
 	 * while reading EDID which are required by Compliance tests
-- 
2.43.0

Re: [PATCH] drm/i915/dp: remove redundant link state check

[sashiko-bot]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Removing the DP link state check from `intel_dp_detect()` breaks recovery for monitors that drop link synchronization silently (without sending an HPD IRQ).
--

commit 26b9e2acc7df07dcc2e2026421a5c54d7a72b52e
Author: tarun-tarun143 <tarund18303@gmail.com>

drm/i915/dp: remove redundant link state check

This commit removes the DP link state check from intel_dp_detect(), relying
instead on the link recovery paths after modesets and hotplug events.

> diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c
> index 113d767e62e9c..ffd39206b958e 100644
> --- a/drivers/gpu/drm/i915/display/intel_dp.c
> +++ b/drivers/gpu/drm/i915/display/intel_dp.c
> @@ -6379,18 +6379,6 @@ intel_dp_detect(struct drm_connector *_connector,
>  		status = connector_status_disconnected;
>  		goto out_unset_edid;
>  	}
> -
> -	/*
> -	 * Some external monitors do not signal loss of link synchronization
> -	 * with an IRQ_HPD, so force a link status check.
> -	 *
> -	 * TODO: this probably became redundant, so remove it: the link state
> -	 * is rechecked/recovered now after modesets, where the loss of
> -	 * synchronization tends to occur.
> -	 */
> -	if (!intel_dp_is_edp(intel_dp))
> -		intel_dp_check_link_state(intel_dp);
> -

[Severity: Medium]
Does removing this check break recovery for monitors that drop link
synchronization silently?

While the link state is recovered after modesets and hotplug events, some
non-compliant monitors fail to assert an HPD IRQ when the main link drops
due to issues like EMI. When this hardware failure happens, a user or
desktop environment running a probe calls intel_dp_detect().

Before this patch, intel_dp_detect() would read the DPCD link status and
queue a link retrain, acting as a manual fallback. By removing this check,
intel_dp_detect() succeeds via the independent AUX channel but ignores the
broken main link. Will this leave users with a permanent blank screen until
a full modeset or physical replug is performed?

>  	/*
>  	 * Clearing NACK and defer counts to get their exact values
>  	 * while reading EDID which are required by Compliance tests

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260826044533.9292-1-tarund18303@gmail.com?part=1

✗ LGCI.VerificationFailed: failure for drm/i915/dp: remove redundant link state check

[Patchwork]

== Series Details ==

Series: drm/i915/dp: remove redundant link state check
URL   : https://patchwork.freedesktop.org/series/172818/
State : failure

== Summary ==

Series author address 'tarund18303@gmail.com' is not on the allowlist, which prevents CI from being automatically triggered.
If you want CI to run for this series, ask Patchwork project owners to click 'retest' on the series in Patchwork.
Exception occurred during validation, bailing out!
Build URL: http://intel-gfx-ci-public.igk.intel.com:8080/job/xe_pw_trigger/1262186/ (on master)