Re: {standard input}:924: Error: unrecognized opcode: `eieio'

[Geert Uytterhoeven]

Hi Pali,

On Mon, Apr 24, 2023 at 8:49 PM Pali Rohár <pali@xxxxxxxxxx> wrote:
> On Monday 24 April 2023 09:49:07 Geert Uytterhoeven wrote:
> > On Sat, Apr 22, 2023 at 4:03 PM Pali Rohár <pali@xxxxxxxxxx> wrote:
> > > On Saturday 22 April 2023 21:53:23 kernel test robot wrote:
> > > > tree:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
> > > > head:   8e41e0a575664d26bb87e012c39435c4c3914ed9
> > > > commit: ff7c76f66d8bad4e694c264c789249e1d3a8205d powerpc/boot: Don't always pass -mcpu=powerpc when building 32-bit uImage
> > > > date:   3 months ago
> > > > config: powerpc-buildonly-randconfig-r006-20230422 (https://download.01.org/0day-ci/archive/20230422/202304222120.WcBMaHXo-lkp@xxxxxxxxx/config)
> > > > compiler: powerpc-linux-gcc (GCC) 12.1.0
> > > > reproduce (this is a W=1 build):
> > > >         wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
> > > >         chmod +x ~/bin/make.cross
> > > >         # https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=ff7c76f66d8bad4e694c264c789249e1d3a8205d
> > > >         git remote add linus https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
> > > >         git fetch --no-tags linus master
> > > >         git checkout ff7c76f66d8bad4e694c264c789249e1d3a8205d
> > > >         # save the config file
> > > >         mkdir build_dir && cp config build_dir/.config
> > > >         COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-12.1.0 make.cross W=1 O=build_dir ARCH=powerpc olddefconfig
> > > >         COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-12.1.0 make.cross W=1 O=build_dir ARCH=powerpc SHELL=/bin/bash
> > > >
> > > > If you fix the issue, kindly add following tag where applicable
> > > > | Reported-by: kernel test robot <lkp@xxxxxxxxx>
> > > > | Link: https://lore.kernel.org/oe-kbuild-all/202304222120.WcBMaHXo-lkp@xxxxxxxxx/
> > > >
> > > > All errors (new ones prefixed by >>):
> > > >
> > > >    {standard input}: Assembler messages:
> > > > >> {standard input}:924: Error: unrecognized opcode: `eieio'
> > > >    {standard input}:1094: Error: unrecognized opcode: `eieio'
> > > >
> > > > --
> > > > 0-DAY CI Kernel Test Service
> > > > https://github.com/intel/lkp-tests
> > >
> > > Hello! What is that? In this error report there is no compile log, nor
> > > any file / line. And "Link:" line from above error report does not work.
> > >
> > > Intel, can you look at your robot? This looks some bug in error reporting.
> >
> > Looks fine to me: this is not a compile problem, but an assembler problem.
> > Presumably this is when building a kernel for a PPC platform that does
> > not support the eieio instruction?
>
> What is fine on this report? There is completely missing command which
> generated this error; whole error message; input files which caused them

The file name is indeed missing.

> and also change or commit.

That is mentioned, and looks like a valid possible introducer.

> And what presumably happened? I'm really not going to spend more time on
> guessing what reporter could probably might wanted to write into the
> report and start looking into crystal ball what could be an issue.
>
> I can say that this issue presumably happened when email operator of
> this robot pushed incorrect button and instead of the real report, it
> sent test report with content of some random error line from the random
> log file.
>
> Are we on the technical discussion about exactness? Or are going to
> discuss what who think that what could happened?

It's randconfig, and the config is provided, incl. steps to reproduce.

> > The Link: works fine for me, sometimes you just have to be a little bit
> > patient for the archiver to catch up...
>
> Sorry, I do not have time to wait undefined amount of time to access
> additional resources which are based on reported details required for
> debugging.

The Link is just a reference to the email you received in your mailbox,
and to all follow-ups.

> Intel, could you please react on this issue? If not, I'm going to add
> your robot to my black list. It is not first time when I received such
> useless report from you and nobody reacted on it.

Well, you are the one who authored the offending commit, and at
first glance, it looks like a reasonable culprit that could introduce the
failure mentioned ;-)

Presumably you forgot to cater for a very specific combination of
config option values?

Gr{oetje,eeting}s,

                        Geert

-- 
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds