From mboxrd@z Thu Jan  1 00:00:00 1970
From: Ingo Molnar <mingo@kernel.org>
Subject: Re: [PATCH] x86: Use entire page for the per-cpu GDT only if
 paravirt-enabled
Date: Mon, 28 Sep 2015 09:58:51 +0200
Message-ID: <20150928075851.GA23998@gmail.com>
References: <1443290440-14930-1-git-send-email-dvlasenk@redhat.com>
 <706E9982-24E3-442B-808A-172909449DD4@zytor.com>
 <56070241.2030407@redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Cc: "H. Peter Anvin" <hpa@zytor.com>,
	Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
	Boris Ostrovsky <boris.ostrovsky@oracle.com>,
	David Vrabel <david.vrabel@citrix.com>,
	Joerg Roedel <joro@8bytes.org>, Gleb Natapov <gleb@kernel.org>,
	Paolo Bonzini <pbonzini@redhat.com>, kvm@vger.kernel.org,
	x86@kernel.org, linux-kernel@vger.kernel.org,
	Andy Lutomirski <luto@kernel.org>,
	Brian Gerst <brgerst@gmail.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Borislav Petkov <bp@alien8.de>,
	Peter Zijlstra <a.p.zijlstra@chello.nl>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Kees Cook <keescook@chromium.org>
To: Denys Vlasenko <dvlasenk@redhat.com>
Return-path: <linux-kernel-owner@vger.kernel.org>
Content-Disposition: inline
In-Reply-To: <56070241.2030407@redhat.com>
Sender: linux-kernel-owner@vger.kernel.org
List-Id: kvm.vger.kernel.org


* Denys Vlasenko <dvlasenk@redhat.com> wrote:

> On 09/26/2015 09:50 PM, H. Peter Anvin wrote:
> > NAK.  We really should map the GDT read-only on all 64 bit systems,
> > since we can't hide the address from SLDT.  Same with the IDT.
> 
> Sorry, I don't understand your point.

So the problem is that right now the SGDT instruction (which is unprivileged) 
leaks the real address of the kernel image:

 fomalhaut:~> ./sgdt 
 SGDT: ffff88303fd89000 / 007f

that 'ffff88303fd89000' is a kernel address.

 fomalhaut:~> cat sgdt.c 
 #include <stdio.h>
 #include <stdlib.h>

 int main(void)
 {
         struct gdt_desc {
                 unsigned short  limit;
                 unsigned long   addr;
         } __attribute__((packed)) gdt_desc = { -1, -1 };

         asm volatile("sgdt %0": "=m" (gdt_desc));

         printf("SGDT: %016lx / %04x\n", gdt_desc.addr, gdt_desc.limit);

         return 0;
 }

Your observation in the changelog and your patch:

> >> It is page-sized because of paravirt. [...]

... conflicts with the intention to mark (remap) the primary GDT address read-only 
on native kernels as well.

So what we should do instead is to use the page alignment properly and remap the 
GDT to a read-only location, and load that one.

This would have a couple of advantages:

 - This would give kernel address randomization more teeth on x86.

 - An additional advantage would be that rootkits overwriting the GDT would have 
   a bit more work to do.

 - A third advantage would be that for NUMA systems we could 'mirror' the GDT into
   node-local memory and load those. This makes GDT load cache-misses a bit less
   expensive.

The IDT is already remapped:

 fomalhaut:~> ./sidt 
 Sidt: ffffffffff57b000 / 0fff
 fomalhaut:~> cat sidt.c
 #include <stdio.h>
 #include <stdlib.h>

 int main(void)
 {
         struct idt_desc {
                 unsigned short  limit;
                 unsigned long   addr;
         } __attribute__((packed)) idt_desc = { -1, -1 };

         asm volatile("sidt %0": "=m" (idt_desc));

         printf("Sidt: %016lx / %04x\n", idt_desc.addr, idt_desc.limit);

         return 0;
 }

Thanks,

	Ingo