linux-kernel - Re: [PATCH v9 3/5] x86/KASLR: Randomize virtual address separately

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <20160617082026.GA4791@gmail.com>
Date:	Fri, 17 Jun 2016 10:20:27 +0200
From:	Ingo Molnar <mingo@...nel.org>
To:	Kees Cook <keescook@...omium.org>
Cc:	Baoquan He <bhe@...hat.com>, Borislav Petkov <bp@...e.de>,
	Yinghai Lu <yinghai@...nel.org>,
	"H. Peter Anvin" <hpa@...or.com>,
	Thomas Gleixner <tglx@...utronix.de>,
	Ingo Molnar <mingo@...hat.com>,
	"x86@...nel.org" <x86@...nel.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Josh Poimboeuf <jpoimboe@...hat.com>,
	Andrey Ryabinin <aryabinin@...tuozzo.com>,
	"H.J. Lu" <hjl.tools@...il.com>,
	Dmitry Vyukov <dvyukov@...gle.com>,
	LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v9 3/5] x86/KASLR: Randomize virtual address separately


* Kees Cook <keescook@...omium.org> wrote:

> -unsigned char *choose_random_location(unsigned long input,
> -				      unsigned long input_size,
> -				      unsigned long output,
> -				      unsigned long output_size)
> +void choose_random_location(unsigned long input,
> +			    unsigned long input_size,
> +			    unsigned long *output,
> +			    unsigned long output_size,
> +			    unsigned long *virt_addr)
>  {
> -	unsigned long choice = output;
>  	unsigned long random_addr;
>  
> +	/* By default, keep output position unchanged. */
> +	*virt_addr = *output;

So I applied this, after fixing a conflict with a recent hibernation related 
change, but it would be nice to further clean up the types in this file, in 
particular could we please propagate 'const' for all input-only pointers?

For example in the above function it would be obvious at a glance if it said 
something like:

 void choose_random_location(unsigned long input,
			    unsigned long input_size,
			    const unsigned long *output,
			    unsigned long output_size,
			    unsigned long *virt_addr)

when reading such a function prototype I can immediately tell: 'yeah, while it's 
named "output", it's in fact a read-only input parameter - the _real_ output of 
the function is 'virt_addr'.)

In addition to that it would also be useful to eliminate the 'virt_addr' parameter 
altogether, and use an 'unsigned long' return value to set virt_addr in misc.c.

Ok?

Thanks,

	Ingo