[PATCH v10 1/4] media: dt-bindings: update clocks for sc7280-camss

[Dmitry Baryshkov]

On Tue, Dec 17, 2024 at 04:30:37PM +0000, Bryan O'Donoghue wrote:
> On 17/12/2024 16:23, Krzysztof Kozlowski wrote:
> > On 17/12/2024 17:12, Bryan O'Donoghue wrote:
> > > On 17/12/2024 14:10, Krzysztof Kozlowski wrote:
> > > > On 17/12/2024 15:06, Vikram Sharma wrote:
> > > > > This patch change clock names to make it consistent with
> > > > > existing platforms as gcc_cam_hf_axi -> gcc_axi_hf.
> > > > > This also adds gcc_axi_sf and remove gcc_camera_ahb.
> > > > 
> > > > Don't combine ABI changes with some less important things.
> > > > 
> > > > You miss here explanation why doing the ABI change in the first place.
> > > > Without that explanation I find it rather churn. But anyway, reason for
> > > > ABI break and impact should be documented in commit msg.
> > > > 
> > > > > 
> > > > > Signed-off-by: Vikram Sharma <quic_vikramsa at quicinc.com>
> > > > > ---
> > > > 
> > > > Best regards,
> > > > Krzysztof
> > > 
> > > This change should be fine since we haven't committed and upstream dtsi,
> > > so there's no ABI to break yet.
> > > 
> > > Agree the cover letter should have been explicit in explaining.
> > 
> > So these are recently added bindings (sc7280 is not particularly new)?
> > If so mention in the commit msg that no users are affected because of this.
> > 
> > Best regards,
> > Krzysztof
> 
> Agreed.
> 
> The commit log should make clear that the ABI hasn't been cemented yet.
> 
> 20241217140656.965235-4-quic_vikramsa at quicinc.com <- is still pending

If it hasn't been comitted yet, isn't it better to post a fixed version
rather than committing a version which has known issues?

> 
> ---
> bod

-- 
With best wishes
Dmitry