linux-kernel - Re: [PATCHv4 1/5] x86/mm: split arch_mmap

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]

Date:   Sat, 11 Feb 2017 00:28:54 +0300
From:   Dmitry Safonov <0x7f454c46@...il.com>
To:     Thomas Gleixner <tglx@...utronix.de>
Cc:     Borislav Petkov <bp@...en8.de>,
        Dmitry Safonov <dsafonov@...tuozzo.com>,
        open list <linux-kernel@...r.kernel.org>,
        Ingo Molnar <mingo@...hat.com>,
        "H. Peter Anvin" <hpa@...or.com>,
        Andy Lutomirski <luto@...nel.org>,
        Borislav Petkov <bp@...e.de>, X86 ML <x86@...nel.org>,
        linux-mm@...ck.org
Subject: Re: [PATCHv4 1/5] x86/mm: split arch_mmap_rnd() on compat/native versions

2017-02-10 23:10 GMT+03:00 Thomas Gleixner <tglx@...utronix.de>:
> On Thu, 9 Feb 2017, Borislav Petkov wrote:
>> I can't say that I'm thrilled about the ifdeffery this is adding.
>>
>> But I can't think of a cleaner approach at a quick glance, though -
>> that's generic and arch-specific code intertwined muck. Sad face.
>
> It's trivial enough to do ....
>
> Thanks,
>
>         tglx
>
> ---
>  arch/x86/mm/mmap.c |   22 ++++++++++------------
>  1 file changed, 10 insertions(+), 12 deletions(-)
>
> --- a/arch/x86/mm/mmap.c
> +++ b/arch/x86/mm/mmap.c
> @@ -55,6 +55,10 @@ static unsigned long stack_maxrandom_siz
>  #define MIN_GAP (128*1024*1024UL + stack_maxrandom_size())
>  #define MAX_GAP (TASK_SIZE/6*5)
>
> +#ifndef CONFIG_COMPAT
> +# define mmap_rnd_compat_bits  mmap_rnd_bits
> +#endif
> +

>From my POV, I can't say that it's clearer to shadow mmap_compat_bits
like that then to have two functions with native/compat names.
But if you insist, I'll resend patches set with your version.

>  static int mmap_is_legacy(void)
>  {
>         if (current->personality & ADDR_COMPAT_LAYOUT)
> @@ -66,20 +70,14 @@ static int mmap_is_legacy(void)
>         return sysctl_legacy_va_layout;
>  }
>
> -unsigned long arch_mmap_rnd(void)
> +static unsigned long arch_rnd(unsigned int rndbits)
>  {
> -       unsigned long rnd;
> -
> -       if (mmap_is_ia32())
> -#ifdef CONFIG_COMPAT
> -               rnd = get_random_long() & ((1UL << mmap_rnd_compat_bits) - 1);
> -#else
> -               rnd = get_random_long() & ((1UL << mmap_rnd_bits) - 1);
> -#endif
> -       else
> -               rnd = get_random_long() & ((1UL << mmap_rnd_bits) - 1);
> +       return (get_random_long() & ((1UL << rndbits) - 1)) << PAGE_SHIFT;
> +}
>
> -       return rnd << PAGE_SHIFT;
> +unsigned long arch_mmap_rnd(void)
> +{
> +       return arch_rnd(mmap_is_ia32() ? mmap_rnd_compat_bits : mmap_rnd_bits);
>  }
>
>  static unsigned long mmap_base(unsigned long rnd)

-- 
             Dmitry