On Tue, Apr 10, 2012 at 4:00 PM, NightStrike <nightstr...@gmail.com> wrote: > On Tue, Apr 10, 2012 at 9:56 AM, Richard Guenther > <richard.guent...@gmail.com> wrote: >> On Tue, Apr 10, 2012 at 3:25 PM, NightStrike <nightstr...@gmail.com> wrote: >>> On Tue, Apr 10, 2012 at 8:38 AM, Jonathan Wakely <jwakely....@gmail.com> >>> wrote: >>>> On 10 April 2012 13:11, NightStrike wrote: >>>>> Generally speaking, I've tried to help people get us a clean build of >>>>> gcc warning-wise for the windows targets. This has historically been >>>>> challenging mainly due to printf. Kai added a lot of support for >>>>> handling whacky windows printfs, and we were doing very well for a >>>>> long time. Currently, however, there are two instances where giant >>>>> massive piles of warnings spit out due to enums not being handled in a >>>>> switch. Is this a warning that we could just disable by default for >>>>> building gcc? My builds are filled with this: >>>>> >>>>> >>>>> ../../../../build/gcc/src/gcc/ipa-pure-const.c:458: warning: >>>>> enumeration value `BUILT_IN_NONE' not handled in switch >>>> >>>> Did you need to paste hundreds of lines of the warning to ask the question? >>>> >>>> Did you try adding -Wno-switch-enum to the build flags? >>> >>> I was asking if that could be disabled by default, so that it wouldn't >>> have to be manually added to the build flags each time, since the >>> massive warnings are intentional. >> >> I don't see them. >> >> Richard. > > Hmm.. Looking closer, it seems that these occur on our cygwin hosted > buildbots running gcc 3.4.4. Was the enum switch warning moved out of > -Wall in later versions of gcc? This happens during make all-gcc, > which should still be using the host compiler in a cross compiler > configuration.
I'm not sure - the ipa-pure-const.c use looks like it could need fixing. Maybe the warning code changed? Richard.
