Re: Commit 9a11899 (USB: OHCI: add missing PCI PM callbacks to ohci-pci.c)breaks several builds

[Guenter Roeck]

On 08/25/2013 09:26 PM, Greg Kroah-Hartman wrote:
> On Sun, Aug 25, 2013 at 09:02:15PM -0700, Guenter Roeck wrote:
> > On 08/25/2013 08:30 PM, Guenter Roeck wrote:
> > > Broken builds:
> > >       mips:ath79_defconfig
> > >       parisc:defconfig
> > >       sparc32:defconfig
> > >       sparc64:defconfig
> > >       tile:defconfig
> > Add:
> > 	powerpc:ppc64e_defconfig
> > 	powerpc:cell_defconfig
> > 	powerps:maple_defconfig
> >
> > That makes it 8 out of 82 builds, or roughly 10% of all builds.
> >
> > My qemu test build for powerpc (which has its own config file)  fails as well :(.
>
> Ugh, I got no reports of this from linux-next or the 0-day build system,
> odd.

Maybe the build system isn't as comprehensive as mine, or it just takes a bit longer.
My system caught it because you updated linux-stable to 3.11-rc7, which triggers all builds.

The functions are only declared if CONFIG_PM is defined, yet are called unconditionally.
This means the patch breaks in all configurations where CONFIG_PM is undefined and
CONFIG_USB_OHCI_HCD_PCI is defined.

Add this to the agenda at the kernel summit: We need more automated test coverage.
It is a bit scary that my little pc-based server farm catches problems like this
faster than everything else out there. And it isn't even the kind of problem
I am looking for, but rather a side effect of the builds I am running.

Guenter

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/