On Mon, Aug 5, 2013 at 7:01 PM, Joe Perches <joe@xxxxxxxxxxx> wrote: > On Mon, 2013-08-05 at 18:50 -0700, Bryan Wu wrote: >> On Mon, Aug 5, 2013 at 6:37 PM, Jingoo Han <jg1.han@xxxxxxxxxxx> wrote: >> > On Friday, August 02, 2013 3:53 PM, Joe Perches wrote: >> >> On Fri, 2013-08-02 at 08:54 +0900, Jingoo Han wrote: >> >> > On Friday, August 02, 2013 7:24 AM, Bryan Wu wrote: >> >> > > On Wed, Jul 31, 2013 at 11:03 PM, Jingoo Han <jg1.han@xxxxxxxxxxx> wrote: >> >> > > > Don't mix different enum types to fix the following sparse warnings. >> >> > > > >> >> > > > drivers/leds/leds-lm355x.c:180:49: warning: mixing different enum types >> >> > > > drivers/leds/leds-lm355x.c:180:49: int enum lm355x_tx2 versus >> >> > > > drivers/leds/leds-lm355x.c:180:49: int enum lm355x_ntc >> >> > > > drivers/leds/leds-lm355x.c:191:49: warning: mixing different enum types >> >> > > > drivers/leds/leds-lm355x.c:191:49: int enum lm355x_tx2 versus >> >> > > > drivers/leds/leds-lm355x.c:191:49: int enum lm355x_ntc >> >> >> >> > > Although there are some warning from sparse, changing a beautiful one >> >> > > line code into multiple lines code looks not good to me. Is there any >> >> > > better way to solve this issue. >> >> > >> >> > +CC Joe Perches, >> >> > >> >> > Hi Joe Perches, >> >> >> >> Hello. >> >> >> >> > Do you know any better way to fix sparse warnings, without changing a one >> >> > line into multiple lines? >> >> >> >> sparse warnings are ignorable. >> >> Ignore warnings when you know better than the tool. >> >> >> >> > Maybe if 'enum' is changed to '#define', it will not make the sparse >> >> > warnings. But, it requires many code changes. >> >> >> >> That works but loses some type checking. >> >> >> >> Maybe you could add something like this >> >> to the Makefile to make sparse not report >> >> enum mismatches. >> > >> > Hi Joe Perches, >> > >> > I really appreciate your answer. >> > It is very helpful. :-) >> > >> > I think that it is better to ignore these warnings >> > without any code changes. >> > >> > Bryan, how about you? >> > >> > Best regards, >> > Jingoo Han >> > >> >> >> >> --- >> >> >> >> diff --git a/drivers/leds/Makefile b/drivers/leds/Makefile >> >> index ac28977..d15716c 100644 >> >> --- a/drivers/leds/Makefile >> >> +++ b/drivers/leds/Makefile >> >> @@ -52,6 +52,9 @@ obj-$(CONFIG_LEDS_ASIC3) += leds-asic3.o >> >> obj-$(CONFIG_LEDS_RENESAS_TPU) += leds-renesas-tpu.o >> >> obj-$(CONFIG_LEDS_MAX8997) += leds-max8997.o >> >> obj-$(CONFIG_LEDS_LM355x) += leds-lm355x.o >> >> +ifdef CONFIG_LEDS_LM355x >> >> +CHECKFLAGS += -Wno-enum-mismatch >> >> I’m not sure about this. So if we turn on CONFIG_LEDS_LM355x, we will >> ignore enum mismatch warning with sparse for other source code as well >> as this leds-lm355x.c, right? Then we will miss to check other files's >> enum mismatching? > > Yup. > > I didn't say it was perfect, I said "something like". ;) > Yeah, that's why I'm not sure abut this. -:)) > Right now there's no generic Makefile ability to have > per-file CHECKFLAGS. > > Maybe you could add one if it appeals to you. > > Otherwise, it's a trivial sparse warning, ignore it. > I think we can probably ignore it now. Anyway, thanks for the CHECKFLAGS trick. -Bryan -- To unsubscribe from this list: send the line "unsubscribe linux-leds" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
