[Intel-gfx] [PATCH 2/3] drm/i915/bios: parse dsi devices in parse_ddi_ports()

[Ville Syrjälä]

On Thu, Mar 07, 2019 at 12:18:15PM +0200, Jani Nikula wrote:
> On Wed, 06 Mar 2019, Ville Syrjälä <ville.syrjala at linux.intel.com> wrote:
> > On Wed, Mar 06, 2019 at 05:34:15PM +0200, Jani Nikula wrote:
> >> For the time being this is only for completeness and better debug
> >> logging of DSI ports.
> >> 
> >> Signed-off-by: Jani Nikula <jani.nikula at intel.com>
> >> ---
> >>  drivers/gpu/drm/i915/i915_drv.h   |  1 +
> >>  drivers/gpu/drm/i915/intel_bios.c | 14 ++++++++------
> >>  2 files changed, 9 insertions(+), 6 deletions(-)
> >> 
> >> diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
> >> index eeeb0d9cfdcd..e14563418aad 100644
> >> --- a/drivers/gpu/drm/i915/i915_drv.h
> >> +++ b/drivers/gpu/drm/i915/i915_drv.h
> >> @@ -951,6 +951,7 @@ struct ddi_vbt_port_info {
> >>  	u8 supports_hdmi:1;
> >>  	u8 supports_dp:1;
> >>  	u8 supports_edp:1;
> >> +	u8 supports_dsi:1;
> >>  	u8 supports_typec_usb:1;
> >>  	u8 supports_tbt:1;
> >>  
> >> diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c
> >> index fc27ffe81c14..9beff569b010 100644
> >> --- a/drivers/gpu/drm/i915/intel_bios.c
> >> +++ b/drivers/gpu/drm/i915/intel_bios.c
> >> @@ -1333,9 +1333,9 @@ static enum port dvo_port_to_port(u8 dvo_port)
> >>  	 * so look for all the possible values for each port.
> >>  	 */
> >>  	static const int dvo_ports[][3] = {
> >> -		[PORT_A] = { DVO_PORT_HDMIA, DVO_PORT_DPA, -1},
> >> -		[PORT_B] = { DVO_PORT_HDMIB, DVO_PORT_DPB, -1},
> >> -		[PORT_C] = { DVO_PORT_HDMIC, DVO_PORT_DPC, -1},
> >> +		[PORT_A] = { DVO_PORT_HDMIA, DVO_PORT_DPA, DVO_PORT_MIPIA },
> >> +		[PORT_B] = { DVO_PORT_HDMIB, DVO_PORT_DPB, DVO_PORT_MIPIB },
> >> +		[PORT_C] = { DVO_PORT_HDMIC, DVO_PORT_DPC, DVO_PORT_MIPIC },
> >
> > I wonder if this is going to end up badly on CHV...
> 
> Yeah. We also have a bunch of these mapping tables, with various
> DVO_PORT_* combos. There's a certain appeal to consolidating them.
> 
> So I can't but wonder why we have duplicate information with child
> device type and child device specific port numbers. Is there a case
> where the distinction helps?

The one case I remeber is 7a17995a3dc8 ("drm/i915: Assume non-DP++
port if dvo_port is HDMI and there's no AUX ch specified in the VBT")

-- 
Ville Syrjälä
Intel