On Thu, Feb 13, 2020 at 09:54:00AM +0100, Rasmus Villemoes wrote: > After this was made buildable for something other than PPC32, kbuild > starts warning > > drivers/usb/host/fhci-hcd.c:398:8: warning: this statement may fall > through [-Wimplicit-fallthrough=] > > I don't know this code, but from the construction (initializing size > with 0 and explicitly using "size +=" in the PIPE_BULK case) I assume > that fallthrough is indeed intended. > > Reported-by: kbuild test robot <lkp@xxxxxxxxx> > Fixes: 5a35435ef4e6 (soc: fsl: qe: remove PPC32 dependency from CONFIG_QUICC_ENGINE) > Fixes: a035d552a93b (Makefile: Globally enable fall-through warning) > Signed-off-by: Rasmus Villemoes <linux@xxxxxxxxxxxxxxxxxx> > --- > > Two different Fixes: Obviously my 5a35435ef4e6 is the one that started > making kbuild complain, but that's just because apparently kbuild > doesn't cover a PPC32+USB_FHCI_HCD .config. Note for -stable folks, > just in case 5.3.y is still maintained somewhere: a035d552a93b > appeared in 5.3, but the #define fallthrough that I'm using here > wasn't introduced until 5.4 (294f69e662d15). So either ignore this, > make it /* fallthrough */, or backport 294f69e662d15 to 5.3.y as well. > > drivers/usb/host/fhci-hcd.c | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/drivers/usb/host/fhci-hcd.c b/drivers/usb/host/fhci-hcd.c > index 04733876c9c6..a8e1048278d0 100644 > --- a/drivers/usb/host/fhci-hcd.c > +++ b/drivers/usb/host/fhci-hcd.c > @@ -396,6 +396,7 @@ static int fhci_urb_enqueue(struct usb_hcd *hcd, struct urb *urb, > case PIPE_CONTROL: > /* 1 td fro setup,1 for ack */ > size = 2; > + fallthrough; We have an attribute for that? Shouldn't this be /* fall through */ instead? Gustavo, what's the best practice here, I count only a few "fallthrough;" instances in the kernel, although one is in our coding style document, and thousands of the /* */ version. thanks, greg k-h
