Message ID | 20170714092540.1217397-1-arnd@arndb.de |
---|---|
Headers | show |
Series | gcc-7 warnings | expand |
On Fri, 2017-07-14 at 11:31 +0200, Arnd Bergmann wrote: > When we pass the result of a multiplication as the timeout, we > can get a warning: > > drivers/mmc/host/bcm2835.c:596:149: error: '*' in boolean context, suggest '&&' instead [-Werror=int-in-bool-context] > drivers/mfd/arizona-core.c:247:195: error: '*' in boolean context, suggest '&&' instead [-Werror=int-in-bool-context] > > This is easy to avoid by comparing the timeout to zero instead, > making it a boolean expression. Perhaps this is better as != 0 if the multiply is signed. > diff --git a/include/linux/iopoll.h b/include/linux/iopoll.h [] > @@ -48,7 +48,8 @@ > (val) = op(addr); \ > if (cond) \ > break; \ > - if (timeout_us && ktime_compare(ktime_get(), timeout) > 0) { \ > + if ((timeout_us) > 0 && \ > + ktime_compare(ktime_get(), timeout) > 0) { \ > (val) = op(addr); \ > break; \ > } \ etc...
On Fri, Jul 14, 2017 at 11:25:12AM +0200, Arnd Bergmann wrote: > This series should shut up all warnings introduced by gcc-6 or gcc-7 on > today's linux-next, as observed in "allmodconfig" builds on x86, > arm and arm64. > > I have sent some of these before, but some others are new, as I had > at some point disabled the -Wint-in-bool-context warning in my > randconfig testing and did not notice the other warnings. > > I have another series to address all -Wformat-overflow warnings, > and one more patch to turn off the -Wformat-truncation warnings > unless we build with "make W=1". I'll send that separately. > > Most of these are consist of trivial refactoring of the code to > shut up false-positive warnings, the one exception being > "staging:iio:resolver:ad2s1210 fix negative IIO_ANGL_VEL read", > which fixes a regression against linux-3.1 that has gone > unnoticed since then. Still, review from subsystem maintainers > would be appreciated. > > I would suggest that Andrew Morton can pick these up into linux-mm > so we can make sure they all make it into the release. Alternatively > Linus might feel like picking them all up himself. > > While I did not mark the harmless ones for stable backports, > Greg may also want to pick them up once they go upstream, to > help build-test the stable kernels with gcc-7. Thanks for these, I'll keep an eye out for them to get into the stable trees, so I can eventually update my test-build box to gcc-7. thanks, greg k-h