From: Yu-cheng Yu <yu-cheng.yu@intel.com>
To: Dave Hansen <dave.hansen@linux.intel.com>,
x86@kernel.org, "H. Peter Anvin" <hpa@zytor.com>,
Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@redhat.com>,
linux-kernel@vger.kernel.org, linux-doc@vger.kernel.org,
linux-mm@kvack.org, linux-arch@vger.kernel.org,
linux-api@vger.kernel.org, Arnd Bergmann <arnd@arndb.de>,
Andy Lutomirski <luto@amacapital.net>,
Balbir Singh <bsingharora@gmail.com>,
Cyrill Gorcunov <gorcunov@gmail.com>,
Florian Weimer <fweimer@redhat.com>,
"H.J. Lu" <hjl.tools@gmail.com>, Jann Horn <jannh@google.com>,
Jonathan Corbet <corbet@lwn.net>,
Kees Cook <keescook@chromiun.org>,
Mike Kravetz <mike.kravetz@oracle.com>,
Nadav Amit <nadav.amit@gmail.com>,
Oleg Nesterov <oleg@redhat.com>, Pavel Machek <pavel@ucw.cz>,
Peter Zijlstra <peterz@infradead.org>,
"Ravi V. Shankar" <ravi.v.shankar@intel.com>,
Vedvyas Shanbhogue <vedvyas.shanbhogue@intel.com>
Subject: Re: [RFC PATCH v2 16/27] mm: Modify can_follow_write_pte/pmd for shadow stack
Date: Thu, 19 Jul 2018 10:06:03 -0700 [thread overview]
Message-ID: <1532019963.16711.61.camel@intel.com> (raw)
In-Reply-To: <f4c90626-51d8-5551-5b77-baaff81f16bb@linux.intel.com>
On Wed, 2018-07-18 at 17:06 -0700, Dave Hansen wrote:
> >
> > >
> > > >
> > > > -static inline bool can_follow_write_pte(pte_t pte, unsigned
> > > > int flags)
> > > > +static inline bool can_follow_write(pte_t pte, unsigned int
> > > > flags,
> > > > + struct vm_area_struct
> > > > *vma)
> > > > {
> > > > - return pte_write(pte) ||
> > > > - ((flags & FOLL_FORCE) && (flags & FOLL_COW)
> > > > && pte_dirty(pte));
> > > > + if (!is_shstk_mapping(vma->vm_flags)) {
> > > > + if (pte_write(pte))
> > > > + return true;
> > > Let me see if I can say this another way.
> > >
> > > The bigger issue is that these patches change the semantics of
> > > pte_write(). Before these patches, it meant that you *MUST*
> > > have this
> > > bit set to write to the page controlled by the PTE. Now, it
> > > means: you
> > > can write if this bit is set *OR* the shadowstack bit
> > > combination is set.
> > Here, we only figure out (1) if the page is pointed by a writable
> > PTE; or
> > (2) if the page is pointed by a RO PTE (data or SHSTK) and it has
> > been
> > copied and it still exists. We are not trying to
> > determine if the
> > SHSTK PTE is writable (we know it is not).
> Please think about the big picture. I'm not just talking about this
> patch, but about every use of pte_write() in the kernel.
>
> >
> > >
> > > That's the fundamental problem. We need some code in the kernel
> > > that
> > > logically represents the concept of "is this PTE a shadowstack
> > > PTE or a
> > > PTE with the write bit set", and we will call that pte_write(),
> > > or maybe
> > > pte_writable().
> > >
> > > You *have* to somehow rectify this situation. We can absolutely
> > > no
> > > leave pte_write() in its current, ambiguous state where it has
> > > no real
> > > meaning or where it is used to mean _both_ things depending on
> > > context.
> > True, the processor can always write to a page through a shadow
> > stack
> > PTE, but it must do that with a CALL instruction. Can we define
> > a
> > write operation as: MOV r1, *(r2). Then we don't have any doubt
> > on
> > pte_write() any more.
> No, we can't just move the target. :)
>
> You can define it this way, but then you also need to go to every
> spot
> in the kernel that calls pte_write() (and _PAGE_RW in fact) and
> audit it
> to ensure it means "mov ..." and not push.
Which pte_write() do you think is right?
bool is_shstk_pte(pte) {
return (_PAGE_RW not set) &&
(_PAGE_DIRTY_HW set);
}
int pte_write_1(pte) {
return (_PAGE_RW set) && !is_shstk_pte(pte);
}
int pte_write_2(pte) {
return (_PAGE_RW set) || is_shstk_pte(pte);
}
Yu-cheng
next prev parent reply other threads:[~2018-07-19 17:09 UTC|newest]
Thread overview: 123+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-07-10 22:26 [RFC PATCH v2 00/27] Control Flow Enforcement (CET) Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 01/27] x86/cpufeatures: Add CPUIDs for Control-flow Enforcement Technology (CET) Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 02/27] x86/fpu/xstate: Change some names to separate XSAVES system and user states Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 03/27] x86/fpu/xstate: Enable XSAVES system states Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 04/27] x86/fpu/xstate: Add XSAVES system states for shadow stack Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 05/27] Documentation/x86: Add CET description Yu-cheng Yu
2018-07-11 8:27 ` Pavel Machek
2018-07-11 15:25 ` Yu-cheng Yu
2018-07-11 9:57 ` Florian Weimer
2018-07-11 13:47 ` H.J. Lu
2018-07-11 14:53 ` Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 06/27] x86/cet: Control protection exception handler Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 07/27] x86/cet/shstk: Add Kconfig option for user-mode shadow stack Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 08/27] mm: Introduce VM_SHSTK for shadow stack memory Yu-cheng Yu
2018-07-11 8:34 ` Peter Zijlstra
2018-07-11 16:15 ` Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 09/27] x86/mm: Change _PAGE_DIRTY to _PAGE_DIRTY_HW Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 10/27] x86/mm: Introduce _PAGE_DIRTY_SW Yu-cheng Yu
2018-07-11 8:45 ` Peter Zijlstra
2018-07-11 9:21 ` Peter Zijlstra
2018-07-10 22:26 ` [RFC PATCH v2 11/27] x86/mm: Modify ptep_set_wrprotect and pmdp_set_wrprotect for _PAGE_DIRTY_SW Yu-cheng Yu
2018-07-10 22:44 ` Dave Hansen
2018-07-10 23:23 ` Nadav Amit
2018-07-10 23:52 ` Dave Hansen
2018-07-11 8:48 ` Peter Zijlstra
2018-07-10 22:26 ` [RFC PATCH v2 12/27] x86/mm: Shadow stack page fault error checking Yu-cheng Yu
2018-07-10 22:52 ` Dave Hansen
2018-07-11 17:28 ` Yu-cheng Yu
2018-07-10 23:24 ` Dave Hansen
2018-07-10 22:26 ` [RFC PATCH v2 13/27] mm: Handle shadow stack page fault Yu-cheng Yu
2018-07-10 23:06 ` Dave Hansen
2018-07-11 9:06 ` Peter Zijlstra
2018-08-14 21:28 ` Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 14/27] mm: Handle THP/HugeTLB " Yu-cheng Yu
2018-07-10 23:08 ` Dave Hansen
2018-07-11 9:10 ` Peter Zijlstra
2018-07-11 16:11 ` Yu-cheng Yu
2018-07-20 14:20 ` Dave Hansen
2018-07-20 14:58 ` Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 15/27] mm/mprotect: Prevent mprotect from changing shadow stack Yu-cheng Yu
2018-07-10 23:10 ` Dave Hansen
2018-07-11 9:12 ` Peter Zijlstra
2018-07-11 16:07 ` Yu-cheng Yu
2018-07-11 16:22 ` Dave Hansen
2018-07-10 22:26 ` [RFC PATCH v2 16/27] mm: Modify can_follow_write_pte/pmd for " Yu-cheng Yu
2018-07-10 23:37 ` Dave Hansen
2018-07-11 17:05 ` Yu-cheng Yu
2018-07-13 18:26 ` Dave Hansen
2018-07-17 23:03 ` Yu-cheng Yu
2018-07-17 23:11 ` Dave Hansen
2018-07-17 23:15 ` Dave Hansen
2018-07-18 20:14 ` Yu-cheng Yu
2018-07-18 21:45 ` Dave Hansen
2018-07-18 23:10 ` Yu-cheng Yu
2018-07-19 0:06 ` Dave Hansen
2018-07-19 17:06 ` Yu-cheng Yu [this message]
2018-07-19 19:31 ` Dave Hansen
2018-07-11 9:29 ` Peter Zijlstra
2018-07-17 23:00 ` Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 17/27] x86/cet/shstk: User-mode shadow stack support Yu-cheng Yu
2018-07-10 23:40 ` Dave Hansen
2018-07-11 9:34 ` Peter Zijlstra
2018-07-11 15:45 ` Yu-cheng Yu
2018-07-11 9:36 ` Peter Zijlstra
2018-07-11 21:10 ` Jann Horn
2018-07-11 21:34 ` Andy Lutomirski
2018-07-11 21:51 ` Jann Horn
2018-07-11 22:21 ` Andy Lutomirski
2018-07-13 18:03 ` Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 18/27] x86/cet/shstk: Introduce WRUSS instruction Yu-cheng Yu
2018-07-10 23:48 ` Dave Hansen
2018-07-12 22:59 ` Yu-cheng Yu
2018-07-12 23:49 ` Dave Hansen
2018-07-13 1:50 ` Dave Hansen
2018-07-13 2:21 ` Andy Lutomirski
2018-07-13 4:16 ` Dave Hansen
2018-07-13 4:18 ` Dave Hansen
2018-07-13 17:39 ` Yu-cheng Yu
2018-07-13 5:55 ` Andy Lutomirski
2018-07-11 9:44 ` Peter Zijlstra
2018-07-11 15:06 ` Yu-cheng Yu
2018-07-11 15:30 ` Peter Zijlstra
2018-07-11 9:45 ` Peter Zijlstra
2018-07-11 14:58 ` Yu-cheng Yu
2018-07-11 15:27 ` Peter Zijlstra
2018-07-11 15:41 ` Yu-cheng Yu
2018-07-13 12:12 ` Dave Hansen
2018-07-13 17:37 ` Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 19/27] x86/cet/shstk: Signal handling for shadow stack Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 20/27] x86/cet/shstk: ELF header parsing of CET Yu-cheng Yu
2018-07-11 11:12 ` Florian Weimer
2018-07-11 19:37 ` Jann Horn
2018-07-11 20:53 ` Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 21/27] x86/cet/ibt: Add Kconfig option for user-mode Indirect Branch Tracking Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 22/27] x86/cet/ibt: User-mode indirect branch tracking support Yu-cheng Yu
2018-07-11 0:11 ` Dave Hansen
2018-07-11 22:10 ` Yu-cheng Yu
2018-07-11 22:40 ` Dave Hansen
2018-07-11 23:00 ` Yu-cheng Yu
2018-07-11 23:16 ` Dave Hansen
2018-07-13 17:56 ` Yu-cheng Yu
2018-07-13 18:05 ` Dave Hansen
2018-07-11 21:07 ` Jann Horn
2018-07-10 22:26 ` [RFC PATCH v2 23/27] mm/mmap: Add IBT bitmap size to address space limit check Yu-cheng Yu
2018-07-10 23:57 ` Dave Hansen
2018-07-11 16:56 ` Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 24/27] x86: Insert endbr32/endbr64 to vDSO Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 25/27] x86/cet: Add PTRACE interface for CET Yu-cheng Yu
2018-07-11 10:20 ` Ingo Molnar
2018-07-11 15:40 ` Yu-cheng Yu
2018-07-12 14:03 ` Ingo Molnar
2018-07-12 22:37 ` Yu-cheng Yu
2018-07-12 23:08 ` Thomas Gleixner
2018-07-13 16:07 ` Yu-cheng Yu
2018-07-13 6:28 ` Pavel Machek
2018-07-13 13:33 ` Ingo Molnar
2018-07-14 6:27 ` Pavel Machek
2018-07-10 22:26 ` [RFC PATCH v2 26/27] x86/cet/shstk: Handle thread shadow stack Yu-cheng Yu
2018-07-10 22:26 ` [RFC PATCH v2 27/27] x86/cet: Add arch_prctl functions for CET Yu-cheng Yu
2018-07-11 12:19 ` Florian Weimer
2018-07-11 21:02 ` Yu-cheng Yu
2018-07-11 19:45 ` Jann Horn
2018-07-11 20:55 ` Yu-cheng Yu
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=1532019963.16711.61.camel@intel.com \
--to=yu-cheng.yu@intel.com \
--cc=arnd@arndb.de \
--cc=bsingharora@gmail.com \
--cc=corbet@lwn.net \
--cc=dave.hansen@linux.intel.com \
--cc=fweimer@redhat.com \
--cc=gorcunov@gmail.com \
--cc=hjl.tools@gmail.com \
--cc=hpa@zytor.com \
--cc=jannh@google.com \
--cc=keescook@chromiun.org \
--cc=linux-api@vger.kernel.org \
--cc=linux-arch@vger.kernel.org \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=luto@amacapital.net \
--cc=mike.kravetz@oracle.com \
--cc=mingo@redhat.com \
--cc=nadav.amit@gmail.com \
--cc=oleg@redhat.com \
--cc=pavel@ucw.cz \
--cc=peterz@infradead.org \
--cc=ravi.v.shankar@intel.com \
--cc=tglx@linutronix.de \
--cc=vedvyas.shanbhogue@intel.com \
--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).