Re: [linus:master] BUILD REGRESSION 68abbe729567cef128b2c2141f2ed2567f3b8372

From: Philip Li
Date: Mon Jun 11 2018 - 03:16:56 EST


On Sat, Jun 09, 2018 at 08:29:17AM +0800, Philip Li wrote:
> On Fri, Jun 08, 2018 at 08:48:29AM -0700, Linus Torvalds wrote:
> > On Fri, Jun 8, 2018 at 8:12 AM kbuild test robot <lkp@xxxxxxxxx> wrote:
> > >
> > > arch/sh/include/asm/io.h:246:32: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]
> > > arch/x86/net/bpf_jit_comp32.c:2449 do_jit() warn: potential off by one 'addrs[]' limit 'bpf_prog->len'
> > > drivers/acpi/battery.c:720:3-9: preceding lock on line 701
> > > drivers/gpu/drm/amd/amdkfd/Kconfig:8: unknown option "imply"
> > > drivers/gpu/drm/amd/amdkfd/Kconfig:9: syntax error
> > > .. lots of these ..
> >
> > Hmm? Is the kbuild robot not feeling well? This looks odd.
> Thanks Linus, we will follow up to check what possibly went wrong here,
> this does look unreasonable.
Hi Linus, this is caused by some cherry picked old commits are compiled on one
randconfig, and triggers bisect. The error is discussed before as you may know,
at https://patchwork.kernel.org/patch/9459363/ or
http://lkml.iu.edu/hypermail/linux/kernel/1611.2/01337.html.

The bisect considers the errors belong to mainline tree, meanwhile, there's a
corner case, the bot sends out this mail externally which is supposed to send
only internally for double check.

Kindly ignore this. We will continue the improvement of the bot.

>
> >
> > Linus
> >