Re: [tip:x86/core 1/1] vmlinux.o: warning: objtool: rcar_pcie_probe+0x13e: no-cfi indirect call!

[Peter Zijlstra <peterz@infradead.org>]

On Thu, Oct 09, 2025 at 08:20:01PM -0700, Nathan Chancellor wrote:
> Hi Peter,
> 
> On Thu, Oct 09, 2025 at 09:07:02PM +0800, kernel test robot wrote:
> > tree:   https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git x86/core
> > head:   6c6e6a5416471498d8aafc050110bec9467e4da7
> > commit: 6c6e6a5416471498d8aafc050110bec9467e4da7 [1/1] Merge branch 'linus' into x86/core, to resolve conflicts
> 
> It appears that this got bisected to the merge because the configuration
> has CONFIG_CFI=y, which needs the rename that is in Linus's tree. This
> is reproducible with just commit 894af4a1cde6 ("objtool: Validate kCFI
> calls") cherry-picked onto 6.17 and CONFIG_CFI_CLANG=y in the config
> provided at the link below.
> 
> > config: x86_64-buildonly-randconfig-001-20251009 (https://download.01.org/0day-ci/archive/20251009/202510092124.O2IX0Jek-lkp@intel.com/config)
> > compiler: clang version 20.1.8 (https://github.com/llvm/llvm-project 87f0227cb60147a26a1eeb4fb06e3b505e9c7261)
> > reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20251009/202510092124.O2IX0Jek-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/202510092124.O2IX0Jek-lkp@intel.com/
> > 
> > All warnings (new ones prefixed by >>):
> > 
> > >> vmlinux.o: warning: objtool: rcar_pcie_probe+0x13e: no-cfi indirect call!
> 
> I do see an indirect call in rcar_pcie_probe() with ->host_init_fn() but
> rcar_pcie_probe() is not __nocfi... +0x13e is 0x8d9d7e in this build.
> 
>     8d9d6f: 31 c0                         xorl    %eax, %eax
>     8d9d71: 49 89 86 58 06 00 00          movq    %rax, 0x658(%r14)
>     8d9d78: 4c 89 ff                      movq    %r15, %rdi
>     8d9d7b: 45 31 db                      xorl    %r11d, %r11d
>     8d9d7e: 2e e8 00 00 00 00             callq   0x8d9d84 <rcar_pcie_probe+0x144>
>                   00000000008d9d80:  R_X86_64_PLT32       __x86_indirect_thunk_r11-0x4
> 
> Is this an issue with the objtool check or is clang not generating the
> right code?

So going by the asm above, objtool is right, this is not a CFI adorned
indirect call.

Let me go build this thing.