Re: [PATCH 3/3] objtool: Generate ORC data for __pfx code

[Peter Zijlstra <peterz@infradead.org>]

On Wed, Apr 12, 2023 at 01:26:15PM -0700, Josh Poimboeuf wrote:
> Allow unwinding from prefix code by copying the CFI from the starting
> instruction of the corresponding function.  Even when the NOPs are
> replaced, they're still stack-invariant instructions so the same ORC
> entry can be reused everywhere.
> 
> Signed-off-by: Josh Poimboeuf <jpoimboe@kernel.org>
> ---
>  tools/objtool/check.c | 16 +++++++++++++++-
>  1 file changed, 15 insertions(+), 1 deletion(-)
> 
> diff --git a/tools/objtool/check.c b/tools/objtool/check.c
> index 2f3136145b2e..3f27a0278bf8 100644
> --- a/tools/objtool/check.c
> +++ b/tools/objtool/check.c
> @@ -4123,6 +4123,7 @@ static bool ignore_unreachable_insn(struct objtool_file *file, struct instructio
>  static int add_prefix_symbol(struct objtool_file *file, struct symbol *func)
>  {
>  	struct instruction *insn, *prev;
> +	struct cfi_state *cfi;
>  
>  	insn = find_insn(file, func->sec, func->offset);
>  	if (!insn)
> @@ -4151,6 +4152,19 @@ static int add_prefix_symbol(struct objtool_file *file, struct symbol *func)
>  	if (!prev)
>  		return -1;
>  
> +	if (!insn->cfi) {
> +		/*
> +		 * This can happen if stack validation isn't enabled or the
> +		 * function is annotated with STACK_FRAME_NON_STANDARD.
> +		 */
> +		return 0;
> +	}
> +
> +	/* Propagate insn->cfi to the prefix code */
> +	cfi = cfi_hash_find_or_add(insn->cfi);
> +	for (; prev != insn; prev = next_insn_same_sec(file, prev))
> +		prev->cfi = cfi;
> +
>  	return 0;
>  }

FWIW, this makes the whole thing hard rely on the prefix being single
byte NOPs -- which they are, but perhaps we should assert this?