Comment # 2 on bug 1224402 from Miroslav Beneš

Strange and I would need to take a look at sources because

000000000000abb0 <nv_register_procfs>:
[...]
    adbc:       e8 00 00 00 00          call   adc1 <nv_register_procfs+0x211>
                        adbd: R_X86_64_PC32     __stack_chk_fail-0x4
    adc1:       4c 8b 05 00 00 00 00    mov    0x0(%rip),%r8        # adc8
<nv_register_procfs+0x218>
                        adc4: R_X86_64_PC32     .bss+0x1284
    adc8:       48 c7 c1 00 00 00 00    mov    $0x0,%rcx
                        adcb: R_X86_64_32S      .rodata+0x420
    adcf:       48 89 c2                mov    %rax,%rdx
    add2:       be 24 81 00 00          mov    $0x8124,%esi
    add7:       e8 00 00 00 00          call   addc <nv_register_procfs+0x22c>
                        add8: R_X86_64_PC32     proc_create_data-0x4
    addc:       0f 1f 40 00             nopl   0x0(%rax)

It really ends with a call to proc_create_data() which, obviously, is not a
noreturn function. So it falls through to the next one. If it ended with the
call to __stack_chk_fail() a couple of lines above, like most of the other
functions, everything would be fine, because __stack_chk_fail() is marked as
noreturn and objtool knows about it.

The code between the call to __stack_chk_fail() and the call to
proc_create_data() fills arguments. adc1 address is a target of intra-function
jump so looks like a valid code.

000000000000ade0 <nv_unregister_procfs>:
    ade0:       e8 00 00 00 00          call   ade5 <nv_unregister_procfs+0x5>
                        ade1: R_X86_64_PC32     __fentry__-0x4
    ade5:       48 8b 3d 00 00 00 00    mov    0x0(%rip),%rdi        # adec
<nv_unregister_procfs+0xc>
                        ade8: R_X86_64_PC32     .bss+0x1274
    adec:       e9 00 00 00 00          jmp    adf1 <nv_unregister_procfs+0x11>
                        aded: R_X86_64_PC32     proc_remove-0x4
    adf1:       66 66 2e 0f 1f 84 00    data16 cs nopw 0x0(%rax,%rax,1)
    adf8:       00 00 00 00 
    adfc:       0f 1f 40 00             nopl   0x0(%rax)

I left it here as an example. There is jmp at the end which is correct. It does
not return. Sibling jump. All good.

I am only guessing without sources though. Could you also appends logs from the
compilation, please?