linux-kernel - Re: [RFC PATCH 2/2] x86/kprobes: Prohibit probing on compiler generated CFI checking code

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 for Android: free password hash cracker in your pocket

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

Date:   Mon, 10 Jul 2023 18:21:06 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     "Masami Hiramatsu (Google)" <mhiramat@...nel.org>
Cc:     Petr Pavlu <petr.pavlu@...e.com>, tglx@...utronix.de,
        mingo@...hat.com, bp@...en8.de, dave.hansen@...ux.intel.com,
        hpa@...or.com, samitolvanen@...gle.com, x86@...nel.org,
        linux-trace-kernel@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH 2/2] x86/kprobes: Prohibit probing on compiler
 generated CFI checking code

On Mon, Jul 10, 2023 at 06:16:43PM +0200, Peter Zijlstra wrote:

> diff --git a/kernel/cfi.c b/kernel/cfi.c
> index 08caad776717..2656e6ffa013 100644
> --- a/kernel/cfi.c
> +++ b/kernel/cfi.c
> @@ -31,16 +31,22 @@ static inline unsigned long trap_address(s32 *p)
>  	return (unsigned long)((long)p + (long)*p);
>  }
>  
> -static bool is_trap(unsigned long addr, s32 *start, s32 *end)
> +static long cfi_trap_distance(unsigned long addr, s32 *start, s32 *end)
>  {
> +	long dist = LONG_MAX;
>  	s32 *p;
>  
>  	for (p = start; p < end; ++p) {
> -		if (trap_address(p) == addr)
> -			return true;
> +		long d = trap_address(p) - addr;
> +
> +		if (abs(dist) < abs(d)) {

Not that I expect anybody will care, but that should obviously be:

		abs(d) < abs(dist)

> +			dist = d;
> +			if (dist == 0)
> +				return 0;
> +		}
>  	}
>  
> -	return false;
> +	return dist;
>  }