[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20230814154730.GA3072@dev-arch.thelio-3990X>
Date: Mon, 14 Aug 2023 08:47:30 -0700
From: Nathan Chancellor <nathan@...nel.org>
To: Peter Xu <peterx@...hat.com>, lkp@...el.com
Cc: llvm@...ts.linux.dev, oe-kbuild-all@...ts.linux.dev,
linux-kernel@...r.kernel.org,
Andrew Morton <akpm@...ux-foundation.org>,
Linux Memory Management List <linux-mm@...ck.org>,
"Huang, Ying" <ying.huang@...el.com>
Subject: Re: Cannot find symbol for section 69: .text.arch_max_swapfile_size.
On Mon, Aug 14, 2023 at 11:38:18AM -0400, Peter Xu wrote:
> On Sat, Aug 12, 2023 at 09:11:33AM +0800, kernel test robot wrote:
> > tree: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
> > head: 190bf7b14b0cf3df19c059061be032bd8994a597
> > commit: be45a4902c7caa717fee6b2f671e59b396ed395c mm/swap: cache maximum swapfile size when init swap
> > date: 11 months ago
> > config: mips-randconfig-r014-20230811 (https://download.01.org/0day-ci/archive/20230812/202308120906.Efohswpa-lkp@intel.com/config)
> > compiler: clang version 17.0.0 (https://github.com/llvm/llvm-project.git 4a5ac14ee968ff0ad5d2cc1ffa0299048db4c88a)
> > reproduce: (https://download.01.org/0day-ci/archive/20230812/202308120906.Efohswpa-lkp@intel.com/reproduce)
> >
> > If you fix the issue in a separate patch/commit (i.e. not just a new version of
> > the same patch/commit), kindly add following tags
> > | Reported-by: kernel test robot <lkp@...el.com>
> > | Closes: https://lore.kernel.org/oe-kbuild-all/202308120906.Efohswpa-lkp@intel.com/
> >
> > All errors (new ones prefixed by >>):
> >
> > >> Cannot find symbol for section 69: .text.arch_max_swapfile_size.
> > mm/swapfile.o: failed
>
> Hmm.. I don't really know what's the issue here, neither can I reproduce
> this locally - the cross build seems to all work with the reproducer and I
> can see mm/swapfile.o correctly generated.
This is https://github.com/ClangBuiltLinux/linux/issues/1830 and
https://github.com/ClangBuiltLinux/linux/issues/981, which is a bug in
recordmcount with regards to section symbols that occurs when building
with clang's integrated assembler. It happens when the there are only
weak symbols in a .text section, which occurs more often with MIPS due
to LD_DEAD_CODE_DATA_ELIMINATION implying -ffunction-sections (i.e., any
__weak symbol will trigger this) and I can see this randconfig setting
it.
Unfortunately, the robot seems to have tripped over this a lot over the
weekend :/ Intel folks, can this type of warning please be added to your
filter for at least MIPS + LD_DEAD_CODE_DATA_ELIMINATION so that people
don't get bothered?
https://lore.kernel.org/202308121810.APAJIBxn-lkp@intel.com/
https://lore.kernel.org/202308121535.otQuZept-lkp@intel.com/
https://lore.kernel.org/202308121810.APAJIBxn-lkp@intel.com/
https://lore.kernel.org/202308121246.0TYkHC1K-lkp@intel.com/
https://lore.kernel.org/202308120553.IRzzfaXL-lkp@intel.com/
https://lore.kernel.org/202308111956.QRc8XJVr-lkp@intel.com/
Cheers,
Nathan
Powered by blists - more mailing lists