linux-kernel - Re: [PATCH] fs, elf: Make sure to page align bss in load_elf

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:   Fri, 6 Jul 2018 12:23:38 -0700
From:   Kees Cook <keescook@...omium.org>
To:     Oscar Salvador <osalvador@...hadventures.net>
Cc:     "linux-fsdevel@...r.kernel.org" <linux-fsdevel@...r.kernel.org>,
        Linux-MM <linux-mm@...ck.org>,
        LKML <linux-kernel@...r.kernel.org>,
        Michal Hocko <mhocko@...e.com>,
        Tetsuo Handa <penguin-kernel@...ove.sakura.ne.jp>,
        Nicolas Pitre <nicolas.pitre@...aro.org>,
        Oscar Salvador <osalvador@...e.de>,
        Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [PATCH] fs, elf: Make sure to page align bss in load_elf_library

On Fri, Jul 6, 2018 at 12:13 AM, Oscar Salvador
<osalvador@...hadventures.net> wrote:
> On Thu, Jul 05, 2018 at 08:44:18AM -0700, Kees Cook wrote:
>> On Thu, Jul 5, 2018 at 7:55 AM,  <osalvador@...hadventures.net> wrote:
>> > From: Oscar Salvador <osalvador@...e.de>
>> >
>> > The current code does not make sure to page align bss before calling
>> > vm_brk(), and this can lead to a VM_BUG_ON() in __mm_populate()
>> > due to the requested lenght not being correctly aligned.
>> >
>> > Let us make sure to align it properly.
>> >
>> > Signed-off-by: Oscar Salvador <osalvador@...e.de>
>> > Tested-by: Tetsuo Handa <penguin-kernel@...ove.sakura.ne.jp>
>> > Reported-by: syzbot+5dcb560fe12aa5091c06@...kaller.appspotmail.com
>>
>> Wow. CONFIG_USELIB? I'm surprised distros are still using this. 32-bit
>> only, and libc5 and earlier only.
>>
>> Regardless, this appears to match the current bss alignment logic in
>> the main elf loader, so:
>>
>> Acked-by: Kees Cook <keescook@...omium.org>
>>
>> -Kees
>>
>> > ---
>> >  fs/binfmt_elf.c | 5 ++---
>> >  1 file changed, 2 insertions(+), 3 deletions(-)
>> >
>> > diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
>> > index 0ac456b52bdd..816cc921cf36 100644
>> > --- a/fs/binfmt_elf.c
>> > +++ b/fs/binfmt_elf.c
>> > @@ -1259,9 +1259,8 @@ static int load_elf_library(struct file *file)
>> >                 goto out_free_ph;
>> >         }
>> >
>> > -       len = ELF_PAGESTART(eppnt->p_filesz + eppnt->p_vaddr +
>> > -                           ELF_MIN_ALIGN - 1);
>> > -       bss = eppnt->p_memsz + eppnt->p_vaddr;
>> > +       len = ELF_PAGEALIGN(eppnt->p_filesz + eppnt->p_vaddr);
>> > +       bss = ELF_PAGEALIGN(eppnt->p_memsz + eppnt->p_vaddr);
>> >         if (bss > len) {
>> >                 error = vm_brk(len, bss - len);
>> >                 if (error)
>> > --
>> > 2.13.6
>> >
> CC Andrew
>
> Hi Andrew,
>
> in case this patch gets accepted, does it have to go through your tree?
> Or is it for someone else to take it?

(FWIW, binfmt_elf changes have traditionally gone through -mm, yes.)

-Kees

-- 
Kees Cook
Pixel Security