include/linux/device.h:925:12: error: 'dev_vprintk_emit' defined

Hi Joe,

FYI, kernel build failed on

tree:   git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core.git driver-core-next
head:   e0f21e6d52cc245e7d4f7e02ca4b7b6571660ec2
commit: 05e4e5b87aab74f994acf9dd4bed4a8f1367cd09 [4/11] dev: Add dev_vprintk_emit and dev_printk_emit
config: sh-allnoconfig

All error/warnings:

include/linux/device.h:925:12: error: 'dev_vprintk_emit' defined but not used [-Werror=unused-function]
cc1: all warnings being treated as errors

vim +925 include/linux/device.h

b9075fa9 (Joe Perches 2011-10-31  919) int dev_notice(const struct device *dev, const char *fmt, ...);
b9075fa9 (Joe Perches 2011-10-31  920) extern __printf(2, 3)
b9075fa9 (Joe Perches 2011-10-31  921) int _dev_info(const struct device *dev, const char *fmt, ...);
99bcf217 (Joe Perches 2010-06-27  922) 
99bcf217 (Joe Perches 2010-06-27  923) #else
99bcf217 (Joe Perches 2010-06-27  924) 
05e4e5b8 (Joe Perches 2012-09-12 @925) static int dev_vprintk_emit(int level, const struct device *dev,
05e4e5b8 (Joe Perches 2012-09-12  926) 			    const char *fmt, va_list args)
05e4e5b8 (Joe Perches 2012-09-12  927) { return 0; }
05e4e5b8 (Joe Perches 2012-09-12  928) static inline __printf(3, 4)

---
0-DAY kernel build testing backend Open Source Technology Centre
Fengguang Wu, Yuanhan Liu Intel Corporation

I can't duplicate this here, what is your .config?

And treating warnings as errors? Good luck :)

greg k-h

Hi Fengguang.

My fault, sorry. It was fixed in a later commit.

commit 0a18b05043acc01d1d6a4bac459e62c79628881c

Author: Joe Perches <[email protected]>
Date:   Tue Sep 25 18:19:57 2012 -0700

device.h: Add missing inline to #ifndef CONFIG_PRINTK dev_vprintk_emit

make ARCH=sh allnoconfig

Sorry, but it's a bit complicated: driver-core-next HEAD is actually fine. The commit 05e4e5b87aab74f994acf9dd4bed4a8f1367cd09 is broken.

Since I'm not doing bisectibility tests on driver-core, I checked the log and find that the error is actually first found when testing the HEAD commit of

        git://git.linaro.org/people/shawnguo/linux-2.6.git for-3.8

It seems to be based on linux-next, which includes an older version of your tree. The build system bisected down the error to the first bad commit 05e4e5b87aab74f994acf9dd4bed4a8f1367cd09, and then find that it's not really in shawnguo's tree, but part of your tree. So it ends up reporting against driver-core rather than shawnguo's tree.

Everything is working fine above, except that your tree HEAD has advanced to a new point that the error already get fixed.

I guess some dir in ARCH sh does that -Werror thing.

Thanks,
Fengguang