Re: [PATCH v4] iio: chemical: sps30: Replace manual locking with RAII locking

[Jonathan Cameron]

On Wed, 20 May 2026 11:49:33 -0500
Maxwell Doose <m32285159@xxxxxxxxx> wrote:

> On Wed, May 20, 2026 at 7:22 AM Stepan Ionichev <sozdayvek@xxxxxxxxx> wrote:
> >
> > The hunk in sps30_read_raw() adds an empty line before "default:" in
> > the switch, which is unrelated to the RAII conversion and probably
> > worth dropping.
> >  
> 
> First one is a stray change, my bad.
> 
> 
> > Also, moving guard(mutex) into sps30_do_meas() changes its contract:
> > previously callers had to hold state->lock around the call, now the
> > function takes it itself. A line in the commit log would help future
> > readers avoid the obvious "I'll just wrap it in mutex_lock()" pitfall.
> >
> > Stepan  
> 
> See:
> >> Add guard(mutex)() into sps30_do_meas() as every caller locks it's call.  
> 
> Basic but still denotes that sps30_do_meas() holds the lock.

I'd tweak to something a tiny bit more detailed like

Move mutex locking (and switch to guard()) into sps30_do_meas() as every
caller takes the lock just for this call.

> 
> best regards,
> max
>