From: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx>
We need to get rid of the vbt.ports[] array. The main
reason being the bogus VBTs found on many ADL laptops
that declare both eDP+HDMI child devices for the same
port. The goal is to probe each of those in order and
stick to the first one that works. But the vbt.ports[]
array gets populated before we do any output probing
and, being indexed with the port, can't handle any
aliasing child devices.
Here's a bit of prep work to reduce our reliance on
vbt.ports[], mainly by expanding the encoder->devdata
(a direct pointer to the correct vbt child device from
the encoder) usage.
Ville Syrjälä (10):
drm/i915: Pass the whole encoder to hotplug_enables()
drm/i915: Move variables to loop context
drm/i915: Replace intel_bios_is_lspcon_present() with
intel_bios_encoder_is_lspcon()
drm/i915: Replace intel_bios_is_lane_reversal_needed() with
intel_bios_encoder_lane_reversal()
drm/i915: Replace intel_bios_is_port_hpd_inverted() with
intel_bios_encoder_hpd_invert()
drm/i915: Consult the registested encoders for the ICL combo PHY w/a
drm/i915: Populate encoder->devdata for g4x+ DP/HDMI ports
drm/i915: Pass devdata to intel_bios_port_aux_ch()
drm/i915: Iterate all child devs in intel_bios_is_port_present()
drm/i915: Use encoder->devdata in eDP init
drivers/gpu/drm/i915/display/g4x_dp.c | 12 +-
drivers/gpu/drm/i915/display/g4x_hdmi.c | 12 +-
drivers/gpu/drm/i915/display/intel_bios.c | 128 ++++++------------
drivers/gpu/drm/i915/display/intel_bios.h | 14 +-
drivers/gpu/drm/i915/display/intel_ddi.c | 6 +-
.../i915/display/intel_display_power_well.c | 15 +-
drivers/gpu/drm/i915/display/intel_dp.c | 30 ++--
drivers/gpu/drm/i915/display/intel_hdmi.c | 2 +-
drivers/gpu/drm/i915/display/intel_lspcon.c | 2 +-
drivers/gpu/drm/i915/i915_irq.c | 59 ++++----
10 files changed, 134 insertions(+), 146 deletions(-)
--
2.39.1
