Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location

From	Thomas Garnier <thgarnie@google.com>
Newsgroups	linux.kernel
Subject	Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location
Date	2017-01-05 23:00 +0100
Message-ID	<sWoeK-ic-15@gated-at.bofh.it> (permalink)
References	(3 earlier) <sWlAe-791-39@gated-at.bofh.it> <sWlJT-7ej-21@gated-at.bofh.it> <sWmFY-7Qg-23@gated-at.bofh.it> <sWnsm-8rY-39@gated-at.bofh.it> <sWnLH-83-15@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Thu, Jan 5, 2017 at 1:19 PM, Andy Lutomirski <luto@amacapital.net> wrote:
> On Thu, Jan 5, 2017 at 1:08 PM, Thomas Garnier <thgarnie@google.com> wrote:
>> On Thu, Jan 5, 2017 at 12:18 PM, Andy Lutomirski <luto@kernel.org> wrote:
>>> On Thu, Jan 5, 2017 at 11:03 AM, Thomas Garnier <thgarnie@google.com> wrote:
>>>> On Thu, Jan 5, 2017 at 10:58 AM, Arjan van de Ven <arjan@linux.intel.com> wrote:
>>>>> On 1/5/2017 9:54 AM, Thomas Garnier wrote:
>>>>>
>>>>>>
>>>>>> That's my goal too. I started by doing a RO remap and got couple
>>>>>> problems with hibernation. I can try again for the next iteration or
>>>>>> delay it for another patch. I also need to look at KVM GDT usage, I am
>>>>>> not familiar with it yet.
>>>>>
>>>>>
>>>>> don't we write to the GDT as part of the TLS segment stuff for glibc ?
>>>>>
>>>>
>>>> Not sure which glibc feature it is.
>>>>
>>>> In this design, you can write to the GDT per-cpu variable that will
>>>> remain read-write. You just need to make the remapping writeable when
>>>> we load task registers (ltr) then the processor use the current GDT
>>>> address. At least that the case I know, I might find more through
>>>> testing.
>>>
>>> Hmm.  I bet that if we preset the accessed bits in all the segments
>>> then we don't need it to be writable in general.  But your point about
>>> set_thread_area (TLS) is well taken.  However, I strongly suspect that
>>> we could make set_thread_area unconditionally set the accessed bit and
>>> no one would ever notice.
>>
>> Not sure I fully understood and I don't want to miss an important
>> point. Do you mean making GDT (remapping and per-cpu) read-only and
>> switch the writeable flag only when we write to the per-cpu entry?
>>
>
> What I mean is: write to the GDT through normal percpu access (or
> whatever the normal mapping is) but load a read-only alias into the
> GDT register.  As long as nothing ever tries to write through the GDTR
> alias, no page faults will be generated.  So we just need to make sure
> that nothing ever writes to it through GDTR.  AFAIK the only reason
> the CPU ever writes to the address in GDTR is to set an accessed bit.
>

A write is made when we use load_TR_desc (ltr). I didn't see any other yet.

> --Andy



-- 
Thomas

Thread

[RFC] x86/mm/KASLR: Remap GDTs at fixed location Thomas Garnier <thgarnie@google.com> - 2017-01-04 23:20 +0100
  Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Ingo Molnar <mingo@kernel.org> - 2017-01-05 09:30 +0100
    Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Arjan van de Ven <arjan@linux.intel.com> - 2017-01-05 16:10 +0100
      Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Thomas Garnier <thgarnie@google.com> - 2017-01-05 17:50 +0100
        Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Arjan van de Ven <arjan@linux.intel.com> - 2017-01-05 20:10 +0100
          Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Thomas Garnier <thgarnie@google.com> - 2017-01-05 20:10 +0100
        Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Borislav Petkov <bp@alien8.de> - 2017-01-06 18:50 +0100
          Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Thomas Garnier <thgarnie@google.com> - 2017-01-06 19:10 +0100
    Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Thomas Garnier <thgarnie@google.com> - 2017-01-05 17:40 +0100
      Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Ingo Molnar <mingo@kernel.org> - 2017-01-06 07:40 +0100
  Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Thomas Garnier <thgarnie@google.com> - 2017-01-05 19:20 +0100
    Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Andy Lutomirski <luto@amacapital.net> - 2017-01-05 19:40 +0100
      Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Thomas Garnier <thgarnie@google.com> - 2017-01-05 19:40 +0100
    Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Arjan van de Ven <arjan@linux.intel.com> - 2017-01-05 20:10 +0100
      Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Thomas Garnier <thgarnie@google.com> - 2017-01-05 20:20 +0100
        Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Andy Lutomirski <luto@kernel.org> - 2017-01-05 21:20 +0100
          Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Thomas Garnier <thgarnie@google.com> - 2017-01-05 22:10 +0100
            Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Andy Lutomirski <luto@amacapital.net> - 2017-01-05 22:30 +0100
              Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Thomas Garnier <thgarnie@google.com> - 2017-01-05 23:00 +0100
                Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Ingo Molnar <mingo@kernel.org> - 2017-01-06 08:00 +0100
                Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Thomas Garnier <thgarnie@google.com> - 2017-01-06 19:10 +0100
                Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Andy Lutomirski <luto@kernel.org> - 2017-01-06 23:30 +0100
                Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Thomas Garnier <thgarnie@google.com> - 2017-01-07 00:00 +0100
                Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Andy Lutomirski <luto@kernel.org> - 2017-01-07 00:40 +0100
                Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Ingo Molnar <mingo@kernel.org> - 2017-01-07 08:50 +0100
                Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Andy Lutomirski <luto@amacapital.net> - 2017-01-07 17:00 +0100
                Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Ingo Molnar <mingo@kernel.org> - 2017-01-07 08:40 +0100
                Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Andy Lutomirski <luto@amacapital.net> - 2017-01-07 17:10 +0100
                Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Thomas Garnier <thgarnie@google.com> - 2017-01-09 23:40 +0100
                Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Ingo Molnar <mingo@kernel.org> - 2017-01-10 11:30 +0100
                Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Thomas Garnier <thgarnie@google.com> - 2017-01-10 18:20 +0100
          Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Linus Torvalds <torvalds@linux-foundation.org> - 2017-01-06 00:10 +0100
            Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Thomas Garnier <thgarnie@google.com> - 2017-01-06 00:30 +0100
            Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Andy Lutomirski <luto@amacapital.net> - 2017-01-06 03:40 +0100
              Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Thomas Garnier <thgarnie@google.com> - 2017-01-06 19:10 +0100
                Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Andy Lutomirski <luto@kernel.org> - 2017-01-06 23:00 +0100
                Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Ingo Molnar <mingo@kernel.org> - 2017-01-07 08:50 +0100
      Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Ingo Molnar <mingo@kernel.org> - 2017-01-06 07:50 +0100
  Re: [RFC] x86/mm/KASLR: Remap GDTs at fixed location Andy Lutomirski <luto@amacapital.net> - 2017-01-05 19:30 +0100

csiph-web