Re: [RFC PATCH 06/10] x86/entry: add CFI hint undwarf annotations

[Andy Lutomirski <luto@kernel.org>]

On Wed, May 31, 2017 at 10:44 PM, Josh Poimboeuf <jpoimboe@redhat.com> wrote:
> Add CFI hint undwarf annotations to entry_64.S.  This will enable the
> undwarf unwinder to unwind through any location in the entry code
> including syscalls, interrupts, and exceptions.
>
> Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
> ---
>  arch/x86/entry/Makefile   |  1 -
>  arch/x86/entry/calling.h  |  5 +++++
>  arch/x86/entry/entry_64.S | 56 ++++++++++++++++++++++++++++++++++++++++++-----
>  3 files changed, 55 insertions(+), 7 deletions(-)
>
> diff --git a/arch/x86/entry/Makefile b/arch/x86/entry/Makefile
> index 9976fce..af28a8a 100644
> --- a/arch/x86/entry/Makefile
> +++ b/arch/x86/entry/Makefile
> @@ -2,7 +2,6 @@
>  # Makefile for the x86 low level entry code
>  #
>
> -OBJECT_FILES_NON_STANDARD_entry_$(BITS).o   := y
>  OBJECT_FILES_NON_STANDARD_entry_64_compat.o := y
>
>  CFLAGS_syscall_64.o            += $(call cc-option,-Wno-override-init,)
> diff --git a/arch/x86/entry/calling.h b/arch/x86/entry/calling.h
> index 05ed3d3..bbec02e 100644
> --- a/arch/x86/entry/calling.h
> +++ b/arch/x86/entry/calling.h
> @@ -1,4 +1,6 @@
>  #include <linux/jump_label.h>
> +#include <asm/undwarf.h>
> +
>
>  /*
>

Just to make sure I understand this, if we unwind from...

> @@ -112,6 +114,7 @@ For 32-bit we have the following conventions - kernel is built with
>         movq %rdx, 12*8+\offset(%rsp)
>         movq %rsi, 13*8+\offset(%rsp)

...here..., will objtool think that rdx and rsi (etc) still live in
their respective regs, or will it find them in the on-stack data given
by CFI_REGS?  If the former, how does undwarf deal with the
corresponding pops?

>         movq %rdi, 14*8+\offset(%rsp)
> +       CFI_REGS offset=\offset extra=0

> @@ -414,6 +424,7 @@ ENTRY(ret_from_fork)
>  2:
>         movq    %rsp, %rdi
>         call    syscall_return_slowpath /* returns with IRQs disabled */
> +       CFI_REGS

I'm confused.  syscall_return_slowpath didn't change anything relevant
to unwinding, right?  What's CFI_REGS here for?