linux-kernel - Re: objtool warnings in prerelease clang-9

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed, 10 Jul 2019 16:42:43 -0700
From:   Nick Desaulniers <ndesaulniers@...gle.com>
To:     Josh Poimboeuf <jpoimboe@...hat.com>
Cc:     Thomas Gleixner <tglx@...utronix.de>,
        Peter Zijlstra <peterz@...radead.org>,
        clang-built-linux <clang-built-linux@...glegroups.com>,
        Kees Cook <keescook@...gle.com>,
        LKML <linux-kernel@...r.kernel.org>,
        Craig Topper <craig.topper@...el.com>,
        Alexander Potapenko <glider@...gle.com>,
        Bill Wendling <morbo@...gle.com>,
        Stephen Hines <srhines@...gle.com>
Subject: Re: objtool warnings in prerelease clang-9

On Wed, Jul 10, 2019 at 4:22 PM Josh Poimboeuf <jpoimboe@...hat.com> wrote:
>
> On Sat, Jul 06, 2019 at 10:50:01AM -0500, Josh Poimboeuf wrote:
> > On Tue, Jul 02, 2019 at 11:58:27PM +0200, Thomas Gleixner wrote:
> > > platform-quirks.o:
> > >
> > >         if (x86_platform.set_legacy_features)
> > >   74:   4c 8b 1d 00 00 00 00    mov    0x0(%rip),%r11        # 7b <x86_early_init_platform_quirks+0x7b>
> > >   7b:   4d 85 db                test   %r11,%r11
> > >   7e:   0f 85 00 00 00 00       jne    84 <x86_early_init_platform_quirks+0x84>
> > >                 x86_platform.set_legacy_features();
> > > }
> > >   84:   c3                      retq
> > >
> > > That jne jumps to __x86_indirect_thunk_r11, aka. ratpoutine.
> > >
> > > No idea why objtool thinks that the instruction at 0x84 is not
> > > reachable. Josh?
> >
> > That's a conditional tail call, which is something GCC never does.
> > Objtool doesn't understand that, so we'll need to fix it.
>
> Can somebody test this patch to see if it fixes the platform-quirks.o
> warning?

$ make CC=clang -j71 2>&1 | grep platform-quirks
  CC      arch/x86/kernel/platform-quirks.o
arch/x86/kernel/platform-quirks.o: warning: objtool:
x86_early_init_platform_quirks()+0x84: unreachable instruction
$ git am /tmp/objtool.patch
$ make CC=clang -j71 clean
$ make CC=clang -j71 2>&1 | grep platform-quirks
  CC      arch/x86/kernel/platform-quirks.o
arch/x86/kernel/platform-quirks.o: warning: objtool:
x86_early_init_platform_quirks()+0x84: unreachable instruction

:(

$ llvm-objdump -dr arch/x86/kernel/platform-quirks.o

arch/x86/kernel/platform-quirks.o: file format ELF64-x86-64


Disassembly of section .init.text:

0000000000000000 x86_early_init_platform_quirks:
       0: 48 b8 02 00 00 00 01 00 00 00 movabsq $4294967298, %rax
       a: 48 89 05 00 00 00 00          movq %rax, (%rip)
000000000000000d:  R_X86_64_PC32 x86_platform+84
      11: c7 05 00 00 00 00 01 00 00 00 movl $1, (%rip)
0000000000000013:  R_X86_64_PC32 x86_platform+88
      1b: 48 b8 00 00 00 00 01 00 00 00 movabsq $4294967296, %rax
      25: 48 89 05 00 00 00 00          movq %rax, (%rip)
0000000000000028:  R_X86_64_PC32 x86_platform+100
      2c: 8b 05 00 00 00 00            movl (%rip), %eax
000000000000002e:  R_X86_64_PC32 boot_params+568
      32: 8d 48 fd                      leal -3(%rax), %ecx
      35: 83 f9 02                      cmpl $2, %ecx
      38: 72 15                        jb 21
<x86_early_init_platform_quirks+0x4f>
      3a: 83 f8 02                      cmpl $2, %eax
      3d: 74 27                        je 39
<x86_early_init_platform_quirks+0x66>
      3f: 85 c0                        testl %eax, %eax
      41: 75 31                        jne 49
<x86_early_init_platform_quirks+0x74>
      43: c7 05 00 00 00 00 01 00 00 00 movl $1, (%rip)
0000000000000045:  R_X86_64_PC32 x86_platform+96
      4d: eb 25                        jmp 37
<x86_early_init_platform_quirks+0x74>
      4f: c7 05 00 00 00 00 00 00 00 00 movl $0, (%rip)
0000000000000051:  R_X86_64_PC32 x86_platform+100
      59: 48 c7 05 00 00 00 00 00 00 00 00      movq $0, (%rip)
000000000000005c:  R_X86_64_PC32 x86_platform+80
      64: eb 0e                        jmp 14
<x86_early_init_platform_quirks+0x74>
      66: 31 c0                        xorl %eax, %eax
      68: 89 05 00 00 00 00            movl %eax, (%rip)
000000000000006a:  R_X86_64_PC32 x86_platform+104
      6e: 89 05 00 00 00 00            movl %eax, (%rip)
0000000000000070:  R_X86_64_PC32 x86_platform+88
      74: 4c 8b 1d 00 00 00 00          movq (%rip), %r11
0000000000000077:  R_X86_64_PC32 x86_platform+108
      7b: 4d 85 db                      testq %r11, %r11
      7e: 0f 85 00 00 00 00            jne 0
<x86_early_init_platform_quirks+0x84>
0000000000000080:  R_X86_64_PC32 __x86_indirect_thunk_r11-4
      84: c3                            retq

I've sent you the .o file off thread as well.  Thanks for taking a
look into this. :D
-- 
Thanks,
~Nick Desaulniers