linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andy Lutomirski <luto@amacapital.net>
To: Dave Hansen <dave.hansen@intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>,
	Dave Hansen <dave.hansen@linux.intel.com>,
	Al Viro <viro@zeniv.linux.org.uk>, X86 ML <x86@kernel.org>,
	Hugh Dickins <hughd@google.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Linux API <linux-api@vger.kernel.org>,
	Ingo Molnar <mingo@kernel.org>,
	Mel Gorman <mgorman@techsingularity.net>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	linux-arch <linux-arch@vger.kernel.org>,
	"linux-mm@kvack.org" <linux-mm@kvack.org>,
	Arnd Bergmann <arnd@arndb.de>,
	Peter Zijlstra <a.p.zijlstra@chello.nl>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"H. Peter Anvin" <hpa@zytor.com>
Subject: Re: [PATCH 6/9] x86, pkeys: add pkey set/get syscalls
Date: Tue, 12 Jul 2016 15:55:45 -0700	[thread overview]
Message-ID: <CALCETrVfmYm5jzM=JWCS0NjBA4VFouren2X22w7M+gLBQF-W4w@mail.gmail.com> (raw)
In-Reply-To: <578524E0.6080401@intel.com>

On Tue, Jul 12, 2016 at 10:12 AM, Dave Hansen <dave.hansen@intel.com> wrote:
> On 07/12/2016 09:32 AM, Andy Lutomirski wrote:
>> I think it's more or less impossible to get sensible behavior passing
>> pkey != 0 data to legacy functions.  If you call:
>>
>> void frob(struct foo *p);
>>
>> If frob in turn passes p to a thread, what PKRU is it supposed to use?
>
> The thread inheritance of PKRU can be nice.  It actually gives things a
> good chance of working if you can control PKRU before clone().  I'd
> describe the semantics like this:
>
>         PKRU values are inherited at the time of a clone() system
>         call.  Threads unaware of protection keys may work on
>         protection-key-protected data as long as PKRU is set up in
>         advance of the clone() and never needs to be changed inside the
>         thread.
>
>         If a thread is created before PKRU is set appropriately, the
>         thread may not be able to act on protection-key-protected data.

Given the apparent need for seccomp's TSYNC, I'm a bit nervous that
this will be restrictive to a problematic degree.

>
> Otherwise, the semantics are simpler, but they basically give threads no
> chance of ever working:
>
>         Threads unaware of protection keys and which can not manage
>         PKRU may not operate on data where a non-zero key has been
>         passed to pkey_mprotect().
>
> It isn't clear to me that one of these is substantially better than the
> other.  It's fairly easy in either case for an app that cares to get the
> behavior of the other.
>
> But, one is clearly easier to implement in the kernel. :)
>
>>>> So how is user code supposed lock down all of its threads?
>>>>
>>>> seccomp has TSYNC for this, but I don't think that PKRU allows
>>>> something like that.
>>>
>>> I'm not sure this is possible for PKRU.  Think of a simple PKRU
>>> manipulation in userspace:
>>>
>>>         pkru = rdpkru();
>>>         pkru |= PKEY_DENY_ACCESS<<key*2;
>>>         wrpkru(pkru);
>>>
>>> If we push a PKRU value into a thread between the rdpkru() and wrpkru(),
>>> we'll lose the content of that "push".  I'm not sure there's any way to
>>> guarantee this with a user-controlled register.
>>
>> We could try to insist that user code uses some vsyscall helper that
>> tracks which bits are as-yet-unassigned.  That's quite messy, though.
>
> Yeah, doable, but not without some new data going out to userspace, plus
> the vsyscall code itself.
>
>> We could also arbitrarily partition the key space into
>> initially-wide-open, initially-read-only, and initially-no-access and
>> let pkey_alloc say which kind it wants.
>
> The point is still that wrpkru destroyed the 'push' operation.  You
> always end up with a PKRU that (at least temporarily) ignored the 'push'.
>

Not with my partitioning proposal.  We'd never asynchronously modify
another thread's state -- we'd start start with a mask that gives us a
good chance of having the initial state always be useful.  To be
completely precise, the initial state would be something like:

0 = all access, 1 (PROT_EXEC) = deny read and write, 2-11: deny read
and write, 12-21: deny write, 22-31: all access

Then pkru_alloc would take a parameter giving the requested initial
state, and it would only work if a key with that initial state is
available.

If we went with the vdso approach, the API could look like:

pkru_state_t prev = pkru_push(mask, value);

...

pkru_pop(prev); // or pkru_pop(mask, prev)?

This doesn't fundamentally require the vdso, except that implementing
bitwise operations on PKRU can't be done atomically with RDPKRU /
WRPKRU.  Grr.  This also falls apart pretty badly when sigreturn
happens, so I don't think I like this approach.

--Andy

  reply	other threads:[~2016-07-12 22:56 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-07-07 12:47 [PATCH 0/9] [REVIEW-REQUEST] [v4] System Calls for Memory Protection Keys Dave Hansen
2016-07-07 12:47 ` [PATCH 1/9] x86, pkeys: add fault handling for PF_PK page fault bit Dave Hansen
2016-07-07 14:40   ` Mel Gorman
2016-07-07 15:42     ` Dave Hansen
2016-07-07 12:47 ` [PATCH 2/9] mm: implement new pkey_mprotect() system call Dave Hansen
2016-07-07 14:40   ` Mel Gorman
2016-07-07 16:51     ` Dave Hansen
2016-07-08 10:15       ` Mel Gorman
2016-07-07 12:47 ` [PATCH 3/9] x86, pkeys: make mprotect_key() mask off additional vm_flags Dave Hansen
2016-07-07 12:47 ` [PATCH 4/9] x86: wire up mprotect_key() system call Dave Hansen
2016-07-07 12:47 ` [PATCH 5/9] x86, pkeys: allocation/free syscalls Dave Hansen
2016-07-07 14:40   ` Mel Gorman
2016-07-07 15:38     ` Dave Hansen
2016-07-07 12:47 ` [PATCH 6/9] x86, pkeys: add pkey set/get syscalls Dave Hansen
2016-07-07 14:45   ` Mel Gorman
2016-07-07 17:33     ` Dave Hansen
2016-07-08  7:18       ` Ingo Molnar
2016-07-08 16:32         ` Dave Hansen
2016-07-09  8:37           ` Ingo Molnar
2016-07-11  4:25             ` Andy Lutomirski
2016-07-11  7:35               ` Ingo Molnar
2016-07-11 14:28                 ` Dave Hansen
2016-07-12  7:13                   ` Ingo Molnar
2016-07-12 15:39                     ` Dave Hansen
2016-07-11 14:50                 ` Andy Lutomirski
2016-07-11 14:34               ` Dave Hansen
2016-07-11 14:45                 ` Andy Lutomirski
2016-07-11 15:48                   ` Dave Hansen
2016-07-12 16:32                     ` Andy Lutomirski
2016-07-12 17:12                       ` Dave Hansen
2016-07-12 22:55                         ` Andy Lutomirski [this message]
2016-07-13  7:56                       ` Ingo Molnar
2016-07-13 18:43                         ` Andy Lutomirski
2016-07-14  8:07                           ` Ingo Molnar
2016-07-18  4:43                             ` Andy Lutomirski
2016-07-18  9:56                               ` Ingo Molnar
2016-07-18 18:02             ` Dave Hansen
2016-07-18 20:12             ` Dave Hansen
2016-07-08 19:26         ` Dave Hansen
2016-07-08 10:22       ` Mel Gorman
2016-07-07 12:47 ` [PATCH 7/9] generic syscalls: wire up memory protection keys syscalls Dave Hansen
2016-07-07 12:47 ` [PATCH 8/9] pkeys: add details of system call use to Documentation/ Dave Hansen
2016-07-07 12:47 ` [PATCH 9/9] x86, pkeys: add self-tests Dave Hansen
2016-07-07 14:47 ` [PATCH 0/9] [REVIEW-REQUEST] [v4] System Calls for Memory Protection Keys Mel Gorman
2016-07-08 18:38 ` Hugh Dickins
  -- strict thread matches above, loose matches on Subject: below --
2016-06-09  0:01 [PATCH 0/9] [v3] " Dave Hansen
2016-06-09  0:01 ` [PATCH 6/9] x86, pkeys: add pkey set/get syscalls Dave Hansen
2016-06-07 20:47 [PATCH 0/9] [v2] System Calls for Memory Protection Keys Dave Hansen
2016-06-07 20:47 ` [PATCH 6/9] x86, pkeys: add pkey set/get syscalls Dave Hansen

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='CALCETrVfmYm5jzM=JWCS0NjBA4VFouren2X22w7M+gLBQF-W4w@mail.gmail.com' \
    --to=luto@amacapital.net \
    --cc=a.p.zijlstra@chello.nl \
    --cc=akpm@linux-foundation.org \
    --cc=arnd@arndb.de \
    --cc=dave.hansen@intel.com \
    --cc=dave.hansen@linux.intel.com \
    --cc=hpa@zytor.com \
    --cc=hughd@google.com \
    --cc=linux-api@vger.kernel.org \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@techsingularity.net \
    --cc=mingo@kernel.org \
    --cc=tglx@linutronix.de \
    --cc=torvalds@linux-foundation.org \
    --cc=viro@zeniv.linux.org.uk \
    --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).