Re: [PATCH 1/2] arm64: dts: qcom: hamoa-iot-evk: Add eDP display overlay

On Mon, Aug 10, 2026 at 08:11:12AM -0500, Bjorn Andersson wrote:
> On Mon, Aug 10, 2026 at 01:52:42PM +0530, Anand Tiwari wrote:
> > On Fri, Aug 07, 2026 at 04:01:05PM -0500, Bjorn Andersson wrote:
> > > On Fri, Aug 07, 2026 at 03:30:13PM +0530, Anand Tiwari wrote:
> > > > On Thu, Aug 06, 2026 at 07:05:37PM -0500, Bjorn Andersson wrote:
> > > > > On Thu, Aug 06, 2026 at 10:01:20PM +0530, Anand Tiwari wrote:
> > > > > > Move the eDP panel configuration and related power, backlight, and pinctrl
> > > > > > nodes into a separate overlay. Keep the base DTB suitable for headless
> > > > > > variants and provide a composite DTB for headed variants.
> > > > > 
> > > > > To quote:
> > > > > https://docs.kernel.org/process/submitting-patches.html#describe-your-changes
> > > > > 
> > > > > """
> > > > > Describe your problem. Whether your patch is a one-line bug fix or 5000
> > > > > lines of a new feature, there must be an underlying problem that
> > > > > motivated you to do this work. Convince the reviewer that there is a
> > > > > problem worth fixing and that it makes sense for them to read past the
> > > > > first paragraph.
> > > > > """
> > > > > 
> > > > > In fact, I'm not even able to guess what the problem you're fixing here.
> > > > > 
> > > > > Regards,
> > > > > Bjorn
> > > > >
> > > > 
> > > > The problem is that the base device trees currently describe the eDP
> > > > display hardware unconditionally. However, Hamoa and Purwa IoT EVKs
> > > > are also available in headless variants without an eDP panel.
> > > > 
> > > > With current base DTBs, headless variants are not stable where eDP panel is not
> > > > physically connected with board and userspace is trying to use it.
> 
> If this is the configuration that is shipping, then it better be
> stable...
>

Agreed.
 
> > > > 
> > > > This series fixes the inaccurate hardware description by keeping the
> > > > base DTBs limited to hardware common to both variants and moving the
> > > > eDP-specific nodes into separate overlays. The base DTB can therefore
> > > > be used for headless variants, while the overlay is selected for
> > > > headed variants.
> > > >  
> > > 
> > > I see, then it seems your change does make sense. I don't understand why
> > > you didn't explained this in the commit message though.
> > >
> > 
> > I will update the commit msg with details, currently details added only in 
> > series cover letter. 
> > 
> > > 
> > > I further do not understand how it can be that nobody noticed that the
> > > display was missing on their Hamoa EVK during the 11 months since I
> > > merged the offending patch - this is completely unacceptable.
> > > 
> > > Regards,
> > > Bjorn
> > >
> > 
> > Internally, we have always required the eDP panel to be connected during 
> > testing. In theory, this platform is expected to have a display attached 
> > by default.
> > 1. The only difference is that the eDP connector on this board is externally
> > accessible and can be disconnected quite easily.
> > 2. Also, only the boards shipped to customers are delivered without an eDP 
> > panel attached by default, which is why this requirement exists. 
> 
> In particular since you're shipping it without display, that is the
> configuration that should be the dts (and that you should test...).
> 
> Common additional configurations can then be added as overlays - while
> the developer can add their own overlays as needed.
> 
> Regards,
> Bjorn

Agreed. The headless configuration is now the base DTS, with the eDP
nodes moved to an optional overlay. Testing is complete for both
headless and headed configurations.