On Wed, Jul 18, 2018 at 1:01 PM Linus Torvalds
<torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:
>
> On Wed, Jul 18, 2018 at 9:52 AM Dmitry Torokhov
> <dmitry.torokhov@xxxxxxxxx> wrote:
> >
> > OK, I see it, it is busted conversion to struct_size(). The macro
> > expects that the array is an array of pointers, whereas in input leds
> > helper we have an array of real structs, and struct_size() calculates
> > too small size for them.
>
> Which one exactly?
Hmm, might be I am full of crap. I'll blame it on the fact that I'm
flying trying to get home since Monday Europe time.
I still suspect that conversion though as it is the only thing that
changed in input-leds.c in the last few releases.
>
> > I'd recommend reverting commit acafe7e30216 ("treewide: Use
> > struct_size() for kmalloc()-family") as used Coccinelle scripts seem
> > to be too greedy.
>
> Ugh. Then we'll just end up re-doing it later.
>
> I think it might be better to try to find the coccinelle bug, and then
> in a private tree do a "revert, reapply using fixed coccielle script,
> then generate inter-diff".
>
> That way we'd have less humongous big patches (just one, rather than
> three: orig, revert, new) in history that people actually see, and it
> will be easier to see what the problem was anyway.
>
> But yes, we want to find the script bug first, because fixing just one
> place and leaving other potential wrongly auto-converted places is not
> wonderful, and then the revert would be preferred.
>
> Kees?
>
> Linus
--
Dmitry
--
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
