On Tue, Jul 08, 2025 at 08:43:37PM +0530, Nautiyal, Ankit K wrote:
On 7/8/2025 7:44 PM, Ville Syrjälä wrote:
On Sun, Jul 06, 2025 at 11:01:49AM +0530, Ankit Nautiyal wrote:
Refine the logic introduced in commit 584cf613c24a ("drm/i915/dp: Reject HBR3 when sink doesn't support TPS4") to allow HBR3 on eDP panels that report DPCD revision 1.4, even if TPS4 is not supported. This aligns with the DisplayPort specification, which does not mandate TPS4 support for eDP with DPCD rev 1.4.
This change avoids regressions on panels that require HBR3 to operate at their native resolution but do not advertise TPS4 support.
Additionally, some ICL/TGL platforms with combo PHY ports suffer from signal integrity issues at HBR3. While certain systems include a Parade PS8461 mux to mitigate this, its presence cannot be reliably detected. Furthermore, broken or missing VBT entries make it unsafe to rely on VBT for enforcing link rate limits.
To address the HBR3-related issues on such platforms and eDP panels, introduce a device specific quirk to cap the eDP link rate to HBR2 (540000 kHz). This will override any higher advertised rates from the sink or DPCD for specific devices.
Currently, the quirk is added for Dell XPS 13 7390 2-in-1 which is reported in gitlab issue #5969 [1].
[1] https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/5969 [2] https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/14517
Fixes: 584cf613c24a ("drm/i915/dp: Reject HBR3 when sink doesn't support TPS4") Cc: Jani Nikula jani.nikula@linux.intel.com Cc: Ville Syrjälä ville.syrjala@linux.intel.com Cc: stable@vger.kernel.org # v6.15+ Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/5969 Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/14517 Signed-off-by: Ankit Nautiyal ankit.k.nautiyal@intel.com
drivers/gpu/drm/i915/display/intel_dp.c | 31 +++++++++++++++++++-- drivers/gpu/drm/i915/display/intel_quirks.c | 9 ++++++ drivers/gpu/drm/i915/display/intel_quirks.h | 1 + 3 files changed, 39 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/i915/display/intel_dp.c b/drivers/gpu/drm/i915/display/intel_dp.c index f48912f308df..362e376fca27 100644 --- a/drivers/gpu/drm/i915/display/intel_dp.c +++ b/drivers/gpu/drm/i915/display/intel_dp.c @@ -171,6 +171,15 @@ int intel_dp_link_symbol_clock(int rate) return DIV_ROUND_CLOSEST(rate * 10, intel_dp_link_symbol_size(rate)); } +static bool intel_dp_reject_hbr3_due_to_tps4(struct intel_dp *intel_dp) +{
- /* TPS4 is not mandatory for eDP with DPCD rev 1.4 */
- if (intel_dp_is_edp(intel_dp) && intel_dp->dpcd[DP_DPCD_REV] == 0x14)
return false;- return !drm_dp_tps4_supported(intel_dp->dpcd);
+}
This feels like it's getty too messy for comfort. I think we should just revert the whole thing and quirk that one icl machine.
Alright sure.
Can this be done is same patch, I mean no need for a revert and a separate quirk patch, right?
Separate revert seems cleaner to me. And then one can actually backport the quirk without having to also backport the already reverted commit. Though I haven't checked how far the bad commit already got backported.