Re: mm/gup.o: warning: objtool: faultin_page_range+0x100: unreachable instruction

[Philip Li <philip.li@intel.com>]

On Tue, Jul 23, 2024 at 10:28:15AM +0200, David Hildenbrand wrote:
> On 23.07.24 00:56, kernel test robot wrote:
> > tree:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
> > head:   933069701c1b507825b514317d4edd5d3fd9d417
> > commit: 631426ba1d45a8672b177ee85ad4cabe760dd131 mm/madvise: make MADV_POPULATE_(READ|WRITE) handle VM_FAULT_RETRY properly
> > date:   3 months ago
> > config: loongarch-randconfig-001-20240722 (https://download.01.org/0day-ci/archive/20240723/202407230630.3vGhnlFB-lkp@intel.com/config)
> > compiler: loongarch64-linux-gcc (GCC) 14.1.0
> > reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20240723/202407230630.3vGhnlFB-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@intel.com>
> > | Closes: https://lore.kernel.org/oe-kbuild-all/202407230630.3vGhnlFB-lkp@intel.com/
> > 
> > All warnings (new ones prefixed by >>):
> > 
> >     mm/gup.o: warning: objtool: get_user_pages_remote+0x160: unreachable instruction
> >     mm/gup.o: warning: objtool: get_user_pages+0x12c: unreachable instruction
> >     mm/gup.o: warning: objtool: get_user_pages_unlocked+0x138: unreachable instruction
> >     mm/gup.o: warning: objtool: __gup_longterm_locked+0x3d8: unreachable instruction
> > > > mm/gup.o: warning: objtool: faultin_page_range+0x100: unreachable instruction
> > 
> > 
> > objdump-func vmlinux.o faultin_page_range:
> > 
> 
> I'm afraid I don't know how serious I should take this report :/

Sorry, kindly ignore this report which could be related to tool chain upgrade to
loongarch64-linux-gcc (GCC) 14.1.0.

We will further check whether the issue can be reproduced on gcc-13.

> 
> The commit is 3 month old and was found on some toolchain with some
> randconfig on some architecture, in code that is not particularly
> architecture specific.
> 
> I wish I could even understand from the report what is happening here
> ("unreachable instruction" -- isn't that the compilers fault if it generates
> that such that objtool would complain? Why is that a C code issue?)
> 
> Please, someone shout if I should take a closer look at this.
> 
> -- 
> Cheers,
> 
> David / dhildenb
> 
>