Re: [kbuild-all] [PATCH v3 1/3] mfd: intel_soc_pmic: Fix a mess with compilation units

[Fengguang Wu]

On Mon, Apr 03, 2017 at 12:26:54PM +0300, Andy Shevchenko wrote:
> On Mon, Apr 3, 2017 at 12:20 PM, Lee Jones <lee.jones@xxxxxxxxxx> wrote:
> > On Sun, 02 Apr 2017, Andy Shevchenko wrote:
> >
> > > On Sun, Apr 2, 2017 at 11:03 PM, kbuild test robot <lkp@xxxxxxxxx> wrote:
> > > > Hi Andy,
> > > >
> > > > [auto build test ERROR on ljones-mfd/for-mfd-next]
> > > > [also build test ERROR on v4.11-rc4 next-20170331]
> > > > [if your patch is applied to the wrong git tree, please drop us a note to help improve the system]
> > >
> > > Thank you for report, though...
> > >
> > > >
> > > > url:    https://github.com/0day-ci/linux/commits/Andy-Shevchenko/mfd-intel_soc_pmic-Fix-a-mess-with-compilation-units/20170320-153539
> > > > base:   https://git.kernel.org/pub/scm/linux/kernel/git/lee/mfd.git for-mfd-next
> > > > config: tile-allyesconfig (attached as .config)
> > > > compiler: tilegx-linux-gcc (GCC) 4.6.2
> > > > reproduce:
> > > >         wget https://raw.githubusercontent.com/01org/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
> > > >         chmod +x ~/bin/make.cross
> > > >         # save the attached .config to linux build tree
> > >
> > > >         make.cross ARCH=tile
> > >
> > > ...I doubt how below is related to my patch. This is obvious x86 code
> > > which was initially submitted without
> > > depend on X86
> > > line in corresponding Kconfig.
> >
> > I think you should probably specify the architecture in the Kconfig
> > entry, or else some `randconfig`s will fail.
>
> Yes, I understand how to fix this, though it's not a problem of this
> patch per se.
>
> Fengguang, is there any possible way to determine the relation between
> an error and a root cause change?

Andy, our 0day build error reports are all bisected ones. It means the
error happens since the reported patch/commit, which is either
responsible for the root cause, or somehow triggers (or changes the
exact form of) an old bug.

I just manual confirmed the bisect is correct:

=============== commit 162c61d1a ===============
/home/wfg/linux
HEAD is now at 162c61d... mfd: intel_soc_pmic: Fix a mess with compilation units
/home/wfg/linux/obj-compiletest

make ARCH=tile allyesconfig
make ARCH=tile drivers/mfd/

!!! BUILD ERROR !!!
make ARCH=tile M=drivers/mfd/

!!! BUILD ERROR !!!
grep -a -F drivers/mfd/ /tmp/build-err-162c61d1a76aa5ccfb398225b3f6f2e896fb58b0-wfg --color
../drivers/mfd/intel_soc_pmic_bxtwc.c:24:31: fatal error: asm/intel_pmc_ipc.h: No such file or directory
compilation terminated.
make[2]: *** [drivers/mfd/intel_soc_pmic_bxtwc.o] Error 1
make[2]: Target '__build' not remade because of errors.
make[1]: *** [drivers/mfd/] Error 2
make: *** [sub-make] Error 2
make[2]: *** No rule to make target 'drivers/mfd//intel_soc_pmic_bxtwc.o', needed by 'drivers/mfd//built-in.o'.
make[2]: Target '__build' not remade because of errors.
make[1]: *** [_module_drivers/mfd/] Error 2
make[1]: Target '_all' not remade because of errors.
make: *** [sub-make] Error 2

=============== PREV commit e93c10211d03c35271896b03a40d3eca4a674770 ===============
/home/wfg/linux
Previous HEAD position was 162c61d... mfd: intel_soc_pmic: Fix a mess with compilation units
HEAD is now at e93c102... mfd: lpc_ich: Enable watchdog on Intel Apollo Lake PCH
/home/wfg/linux/obj-compiletest

make ARCH=tile allyesconfig
make ARCH=tile drivers/mfd/

make ARCH=tile M=drivers/mfd/

grep -a -F drivers/mfd/ /tmp/build-err-e93c10211d03c35271896b03a40d3eca4a674770-wfg --color

Thanks,
Fengguang