Re: [PATCH RFC v2 0/4] phy: qcom-mipi-csi2: C-PHY for X1E80100

[Bryan O'Donoghue]

On 16/10/2026 12:32, George Neacsu wrote:
> > > The kernel was 7.1.5, with linux-next's CAMSS and CSI2 PHY built as
> > > modules; the test build needed shims only for API differences.
> > What's this ?
> >
> > "It builds" I can verify myself, "it works" is the thing. Just slow down
> > and get something you can test on hardware.
> Fair point. The frames in the cover letter were captured on the
> hardware, but the kernel underneath was the 7.1.5 one this machine runs
> day to day, with linux-next's CAMSS and CSI2 PHY sources built as modules
> for it. That is not the same as running linux-next. Before v3 I will
> boot an actual linux-next kernel on the Surface Pro 11, retest both
> cameras there, and say exactly what was run.
>
> For 4/4 I will also move the fsleep() change into its own patch with the
> reasoning, and put the lanes_enable() decomposition in a precursor
> patch, as you asked. Thanks for the reviews of 2/4 and 3/4.
>
> Thanks,
> George

The bit I'm a bit alarmed about is "stuff was stubbed" which makes me 
wonder if what lands in -next can then said to be actually verified 
meaningfully.

Could you git rebase linux-stable/master for your 7.1.5 tree maybe ?

---
bod