linux-kernel - Re: [PATCH] fortify: Use WARN instead of BUG for now

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: <CAGXu5jJrAWKAsDEchyV+wFX_530YpbeBtnrV-kdKO6OP3S=_Hg@mail.gmail.com>
Date:   Wed, 26 Jul 2017 10:17:23 -0700
From:   Kees Cook <keescook@...omium.org>
To:     Linus Torvalds <torvalds@...ux-foundation.org>
Cc:     Andrew Morton <akpm@...ux-foundation.org>,
        Daniel Micay <danielmicay@...il.com>,
        Dan Williams <dan.j.williams@...el.com>,
        Mika Westerberg <mika.westerberg@...ux.intel.com>,
        Al Viro <viro@...iv.linux.org.uk>,
        David Howells <dhowells@...hat.com>,
        Heikki Krogerus <heikki.krogerus@...ux.intel.com>,
        Bjorn Helgaas <bhelgaas@...gle.com>,
        Arnd Bergmann <arnd@...db.de>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        Mauro Carvalho Chehab <mchehab@...nel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] fortify: Use WARN instead of BUG for now

On Wed, Jul 26, 2017 at 10:10 AM, Linus Torvalds
<torvalds@...ux-foundation.org> wrote:
> On Tue, Jul 25, 2017 at 8:50 PM, Kees Cook <keescook@...omium.org> wrote:
>> +
>> +void fortify_read_overflow(const char *func)
>>  {
>> -       pr_emerg("detected buffer overflow in %s\n", name);
>> -       BUG();
>> +       WARN(1, "detected read beyond size of object passed as 1st parameter in %s\n", func);
>>  }
>
> Side note: have you actually checked the code generation of this all?
>
> In particular, do you have any reason to use the out-of-line
> functions? Our WARN() code isn't horrible, and isn't likely to be
> noticeably worse than your own explicit out-of-lining. And you'd get
> the "unlikely()" for free, so you'll possibly get smaller code that
> runs better too.
>
> And it would even *look* better. This:
>
>         if (p_size < size)
>                fortify_read_overflow(__func__);
>
> would become
>
>        WARN(p_size < size, "kmemdup size overflow");
>
> or something.

I did, yeah. It's actually slightly smaller code size to out-of-line these:

$ size vmlinux.fortify*
   text    data     bss     dec     hex filename
10903767        5605009 13930496        30439272        1d07768
vmlinux.fortify-off
10944795        5617801 13930496        30493092        1d149a4 vmlinux.fortify
10950117        5626725 13930496        30507338        1d1814a
vmlinux.fortify-inline

If the readability improvement is preferred over the growth in size, I
can certainly respin it.

-Kees

-- 
Kees Cook
Pixel Security