[PATCH v3 1/2] phy: qcom: edp: Add eDP/DP mode switch support

[Dmitry Baryshkov]

On Mon, Apr 20, 2026 at 08:47:09PM +0800, Yongxing Mou wrote:
> 
> 
> On 3/20/2026 2:36 PM, Dmitry Baryshkov wrote:
> > On Mon, Mar 02, 2026 at 04:28:29PM +0800, Yongxing Mou wrote:
> > > The eDP PHY supports both eDP&DP modes, each requires a different table.
> > > The current driver doesn't fully support every combo PHY mode and use
> > > either the eDP or DP table when enable the platform. In addition, some
> > > platforms mismatch between the mode and the table where DP mode uses
> > > the eDP table or eDP mode use the DP table.
> > > 
> > > Clean up and correct the tables for currently supported platforms based on
> > > the HPG specification.
> > > 
> > > Here lists the tables can be reused across current platforms.
> > > DP mode：
> > > 	-sa8775p/sc7280/sc8280xp/x1e80100
> > > 	-glymur
> > > eDP mode(low vdiff):
> > 
> > Separate question: should we extend phy_configure_dp_opts with the
> > low/high vdiff? Is there a point in providing the ability to toggle
> > between low vdiff and high vdiff?
> > 
> Emm ,i haven't found any platform using high vdiff so far, and I'm not clear
> in which cases switching between low and high vdiff would be needed.

>From my (maybe incorrect) understanding of eDP B.3, the high vs low
vdiff selection should be based on the cable length. 

> 
> > > 	-glymur/sa8775p/sc8280xp/x1e80100
> > > 	-sc7280
> > 
> > I understand your wish to perform all the changes in a single patch, but
> > there is one problem with that. Consider this patch regresses one of the
> > platforms (I'm looking at Kodiak and SC8180X as they get the biggest set
> > of changes). It would be almost impossible to separate, which particular
> > change caused the regression. I'd suggest splitting this patch into a
> > set of more atomic changes. E.g. the AUX_CFG8 is definitely a separate
> > change. Writing swing / pre_emph tables on Kodiak and SC8180X is a
> > separate change (or two). Switching each of the platforms to the
> > corrected set of tables ideally also should come as a separate change,
> > so that in case of a regression the issue would be easier to identify.
> > 
> Thank for point this, will separate the change.
> I mostly overlooked SC8180X here, since I assumed it shares the same PHY as
> SC7280. However, they are using different PHY sub‑versions. Will add proper
> support for it in the next version.

Thanks!

> > > 
> > > Cc: stable at vger.kernel.org
> > > Fixes: f199223cb490 ("phy: qcom: Introduce new eDP PHY driver")
> > > Signed-off-by: Yongxing Mou <yongxing.mou at oss.qualcomm.com>
> > > ---
> > >   drivers/phy/qualcomm/phy-qcom-edp.c | 90 ++++++++++++++++++++++---------------
> > >   1 file changed, 53 insertions(+), 37 deletions(-)
> > > 

-- 
With best wishes
Dmitry