Re: [PATCH v7 0/9] More Rust bindings for device property reads

[Rob Herring <robh@kernel.org>]

On Fri, May 30, 2025 at 11:50:39PM +0200, Danilo Krummrich wrote:
> On Fri, May 30, 2025 at 11:45:38PM +0200, Remo Senekowitsch wrote:
> > On Fri May 30, 2025 at 9:56 PM CEST, Danilo Krummrich wrote:
> > > On Fri, May 30, 2025 at 09:28:47PM +0200, Remo Senekowitsch wrote:
> > >> changes in v7:
> > >> * Fix a typo in a commit message.
> > >> * Fix bug in `FwNode::display_path`. I took a slightly different
> > >>   approach than the one suggested, using `Either` to handle the
> > >>   owned and borrowed case. That also removes the conditional
> > >>   `fwnode_handle_put` at the end.
> > >
> > > That's a good idea, but also a bit unfortunate; there are efforts to remove
> > > Either [1] in favor of using - more descriptive - custom enum types.
> > >
> > > Can you please replace this with e.g. an enum Node with a Borrowed and Owned
> > > variant?
> > >
> > > [1] https://lore.kernel.org/lkml/20250519124304.79237-1-lossin@kernel.org/
> > 
> > Sure, that seems reasonable.
> > 
> > Btw. what's the normal waiting time before posting a new version of a
> > patch series? The requested changes have been getting fewer and I could
> > crank these out much faster, but my gut feeling tells me not to spam the
> > list too much. Or is that wrong and people can deal with quick updates
> > just fine?
> 
> I think the pace was appropriate. For the current state, I don't expect much
> more feedback, so it'd be fine to send an update for the enum change rather
> quicky.

Yes. General rules are no more frequent than 24 hours, but generally 1-2 
weeks. It also is a function of amount of review. No review, wait. If 
there's enough review that the current version isn't going to get more 
review, then go ahead and send another version. When things are close to 
merging and the changes are small, you can pick up the pace a little 
(outside of the merge window). 

> 
> However, we're anyways in the merge window currently, so I'd recomment to leave
> the patch series as is and send a v8 once the merge window closes -- I'll pick
> it up then unless there's some further feedback.

You'll need to fixup the user drivers/cpufreq/rcpufreq_dt.rs that 
landed in the merge window.

Overall, this all looks great to me. Thanks for continuing to push this 
forward.

Rob