Re: [PATCH 20/39] hw/ppc: replace assert(false) with g_assert_not_reached()

[Pierrick Bouvier]

On 9/11/24 07:10, BALATON Zoltan wrote:
> On Tue, 10 Sep 2024, Pierrick Bouvier wrote:
>
> > Signed-off-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
> > ---
> > hw/ppc/spapr_events.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/hw/ppc/spapr_events.c b/hw/ppc/spapr_events.c
> > index cb0eeee5874..38ac1cb7866 100644
> > --- a/hw/ppc/spapr_events.c
> > +++ b/hw/ppc/spapr_events.c
> > @@ -645,7 +645,7 @@ static void spapr_hotplug_req_event(uint8_t hp_id, uint8_t 
> > hp_action,
> >          /* we shouldn't be signaling hotplug events for resources
> >           * that don't support them
> >           */
> > -        g_assert(false);
> > +        g_assert_not_reached();
> >          return;
> >      }
>
> If break does not make sense after g_assert_not_reached() and removed then
> return is the same here.
>
> It may make the series shorter and easier to check that none of these are
> missed if this is done in the same patch where the assert is changed
> instead of separate patches. It's unlikely that the assert change and
> removal of the following break or return would need to be reverted
> separately so it's a simple enough change to put in one patch in my
> opinion but I don't mink if it's kept separate either.
>
> Regards,
> BALATON Zoltan

Mostly done this way because it's easy for creating many commits.