linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Andy Lutomirski <luto@kernel.org>
To: Daniel Axtens <dja@axtens.net>
Cc: Dmitry Vyukov <dvyukov@google.com>,
	kasan-dev <kasan-dev@googlegroups.com>,
	 Linux-MM <linux-mm@kvack.org>,
	"the arch/x86 maintainers" <x86@kernel.org>,
	 Andrey Ryabinin <aryabinin@virtuozzo.com>,
	Alexander Potapenko <glider@google.com>,
	 Andy Lutomirski <luto@kernel.org>
Subject: Re: [PATCH 3/3] x86/kasan: support KASAN_VMALLOC
Date: Thu, 25 Jul 2019 09:32:32 -0700	[thread overview]
Message-ID: <CALCETrXW_=6sPd8gcdkZtYAmCTYhoOYMYhp6_yVd-8Wd5zYsrA@mail.gmail.com> (raw)
In-Reply-To: <87lfwmgm2v.fsf@dja-thinkpad.axtens.net>

On Thu, Jul 25, 2019 at 8:39 AM Daniel Axtens <dja@axtens.net> wrote:
>
>
> >> Would it make things simpler if we pre-populate the top level page
> >> tables for the whole vmalloc region? That would be
> >> (16<<40)/4096/512/512*8 = 131072 bytes?
> >> The check in vmalloc_fault in not really a big burden, so I am not
> >> sure. Just brining as an option.
> >
> > I prefer pre-populating them. In particular, I have already spent far too much time debugging the awful explosions when the stack doesn’t have KASAN backing, and the vmap stack code is very careful to pre-populate the stack pgds — vmalloc_fault fundamentally can’t recover when the stack itself isn’t mapped.
> >
> > So the vmalloc_fault code, if it stays, needs some careful analysis to make sure it will actually survive all the various context switch cases.  Or you can pre-populate it.
> >
>
> No worries - I'll have another crack at prepopulating them for v2.
>
> I tried prepopulating them at first, but because I'm really a powerpc
> developer rather than an x86 developer (and because I find mm code
> confusing at the best of times) I didn't have a lot of luck. I think on
> reflection I stuffed up the pgd/p4d stuff and I think I know how to fix
> it. So I'll give it another go and ask for help here if I get stuck :)
>

I looked at this a bit more, and I think the vmalloc_fault approach is
fine with one tweak.  In prepare_switch_to(), you'll want to add
something like:

kasan_probe_shadow(next->thread.sp);

where kasan_probe_shadow() is a new function that, depending on kernel
config, either does nothing or reads the shadow associated with the
passed-in address.  Also, if you take this approach, I think you
should refactor vmalloc_fault() to push the address check to a new
helper:

static bool is_vmalloc_fault_addr(unsigned long addr)
{
  if (addr >= VMALLOC_START && addr < VMALLOC_END)
    return true;

#ifdef CONFIG_WHATEVER
  if (addr >= whatever && etc)
    return true;
#endif

 return false;
}

and call that from vmalloc_fault() rather than duplicating the logic.

Also, thanks for doing this series!


      reply	other threads:[~2019-07-25 16:32 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-07-25  5:55 [PATCH 0/3] kasan: support backing vmalloc space with real shadow memory Daniel Axtens
2019-07-25  5:55 ` [PATCH 1/3] " Daniel Axtens
2019-07-25  7:35   ` Dmitry Vyukov
2019-07-25  7:51     ` Dmitry Vyukov
2019-07-25 10:06       ` Marco Elver
2019-07-25 10:11         ` Mark Rutland
2019-07-25 11:38           ` Marco Elver
2019-07-25 15:25         ` Daniel Axtens
2019-07-26  5:11           ` Daniel Axtens
2019-07-26  9:55             ` Marco Elver
2019-07-26 10:32       ` Marco Elver
2019-07-29 10:15     ` Daniel Axtens
2019-07-29 10:28       ` Dmitry Vyukov
2019-07-25  5:55 ` [PATCH 2/3] fork: support VMAP_STACK with KASAN_VMALLOC Daniel Axtens
2019-07-25  7:37   ` Dmitry Vyukov
2019-07-25  5:55 ` [PATCH 3/3] x86/kasan: support KASAN_VMALLOC Daniel Axtens
2019-07-25  7:49   ` Dmitry Vyukov
2019-07-25 15:08     ` Andy Lutomirski
2019-07-25 15:39       ` Daniel Axtens
2019-07-25 16:32         ` Andy Lutomirski [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CALCETrXW_=6sPd8gcdkZtYAmCTYhoOYMYhp6_yVd-8Wd5zYsrA@mail.gmail.com' \
    --to=luto@kernel.org \
    --cc=aryabinin@virtuozzo.com \
    --cc=dja@axtens.net \
    --cc=dvyukov@google.com \
    --cc=glider@google.com \
    --cc=kasan-dev@googlegroups.com \
    --cc=linux-mm@kvack.org \
    --cc=x86@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).