Re: [PATCH v7 3/4] x86/dumpstack: Introduce die_addr() for die() with #GP fault address

[Borislav Petkov]

On Thu, Dec 19, 2019 at 12:11:49AM +0100, Jann Horn wrote:
> diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
> index c8b4ae6aed5b..4c691bb9e0d9 100644
> --- a/arch/x86/kernel/traps.c
> +++ b/arch/x86/kernel/traps.c
> @@ -621,7 +621,10 @@ do_general_protection(struct pt_regs *regs, long error_code)
>  				 "maybe for address",
>  				 gp_addr);

 
> -		die(desc, regs, error_code);

I've added here:

                /*
                 * KASAN is interested only in the non-canonical case, clear it
                 * otherwise.
                 */

> +		if (hint != GP_NON_CANONICAL)
> +			gp_addr = 0;


otherwise you have:

	if (hint != GP_NO_HINT)
		...

	if (hint != GP_NON_CANONICAL)
		...

which is kinda confusing at a first glance and one has to follow the
code into die_addr() to figure out the usage of the address argument.

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette