Re: [PATCH] arm: Use kernel mm when updating section permissions

[Laura Abbott <labbott@redhat.com>]

On 11/05/2015 05:15 PM, Kees Cook wrote:
> On Thu, Nov 5, 2015 at 5:05 PM, Laura Abbott <labbott@redhat.com> wrote:
>> On 11/05/2015 08:27 AM, Russell King - ARM Linux wrote:
>>>
>>> On Thu, Nov 05, 2015 at 08:20:42AM -0800, Laura Abbott wrote:
>>>>
>>>> On 11/05/2015 01:46 AM, Russell King - ARM Linux wrote:
>>>>>
>>>>> On Wed, Nov 04, 2015 at 05:00:39PM -0800, Laura Abbott wrote:
>>>>>>
>>>>>> Currently, read only permissions are not being applied even
>>>>>> when CONFIG_DEBUG_RODATA is set. This is because section_update
>>>>>> uses current->mm for adjusting the page tables. current->mm
>>>>>> need not be equivalent to the kernel version. Use pgd_offset_k
>>>>>> to get the proper page directory for updating.
>>>>>
>>>>>
>>>>> What are you trying to achieve here?  You can't use these functions
>>>>> at run time (after the first thread has been spawned) to change
>>>>> permissions, because there will be multiple copies of the kernel
>>>>> section mappings, and those copies will not get updated.
>>>>>
>>>>> In any case, this change will probably break kexec and ftrace, as
>>>>> the running thread will no longer see the updated page tables.
>>>>>
>>>>
>>>> I think I was hitting that exact problem with multiple copies
>>>> not getting updated. The section_update code was being called
>>>> and I was seeing the tables get updated but nothing was being
>>>> applied when I tried to write to text or check the debugfs
>>>> page table. The current flow is:
>>>>
>>>> rest_init -> kernel_thread(kernel_init) and from that thread
>>>> mark_rodata_ro. So mark_rodata_ro is always going to happen
>>>> in a thread.
>>>>
>>>> Do we need to update for both init_mm and the first running
>>>> thread?
>>>
>>>
>>> The "first running thread" is merely coincidental for things like kexec.
>>>
>>> Hmm.  Actually, I think the existing code _should_ be fine.  At the
>>> point where mark_rodata_ro() is, we should still be using init_mm, so
>>> updating the current threads page tables should actually be updating
>>> the swapper_pg_dir.
>>
>>
>> That doesn't seem to hold true. Based on what I'm seeing, we lose
>> the the guarantee of init_mm after the first exec. If usermodehelper
>> gets called to load a module, that triggers an exec and the kernel
>> thread is no longer using init_mm after that. I'm testing with the
>> multi-v7 defconfig which uses the smsc911x driver which loads a
>> module during initcall. That gets called before mark_rodata_ro so
>> the init_mm is never updated. I verified that disabling smsc911x
>> makes things work as expected. I suspect the testing was never done
>> with a driver that tried to call usermodehelper during init time.
>
> Ooooh. Nice catch. Yeah, my testing didn't include that case.
>
>> I got as far as narrowing it down that it happens after the usermodehelper
>> but I wasn't able to pinpoint where exactly the switch happened. It seems
>> like we need to have the page tables set up before any initcalls
>> happen otherwise we risk having an exec create stray processes which we
>> can't update.
>
> Can we just make mark_rodata_ro() a no-op and do the RO setting
> earlier when we do the NX setting?
>

Unfortunately no. The time we are doing the nx setting is before we've finished
with the initmem so we need the initmem to be finished and freed before we can
mark anything RO.

More importantly, the NX settings are also not getting set. Compare before:

---[ Kernel Mapping ]---
0xc0000000-0xc0300000           3M     RW NX
0xc0300000-0xc1300000          16M     RW x
0xc1300000-0xcc000000         173M     RW NX
0xcc000000-0xcc040000         256K     RW NX     MEM/BUFFERABLE/WC
0xcc040000-0xcc100000         768K     RW NX     MEM/CACHED/WBRA
0xcc100000-0xcc280000        1536K     RW NX     MEM/BUFFERABLE/WC
0xcc280000-0xd0000000       62976K     RW NX     MEM/CACHED/WBRA
0xd0000000-0xd0200000           2M     RW NX

and after

---[ Kernel Mapping ]---
0xc0000000-0xc0300000           3M     RW NX
0xc0300000-0xc0c00000           9M     ro x
0xc0c00000-0xc1100000           5M     ro NX
0xc1100000-0xcc000000         175M     RW NX
0xcc000000-0xcc040000         256K     RW NX     MEM/BUFFERABLE/WC
0xcc040000-0xcc100000         768K     RW NX     MEM/CACHED/WBRA
0xcc100000-0xcc280000        1536K     RW NX     MEM/BUFFERABLE/WC
0xcc280000-0xd0000000       62976K     RW NX     MEM/CACHED/WBRA
0xd0000000-0xd0200000           2M     RW NX


with my test patch. I think setting both current->active_mm and &init_mm
is sufficient. Maybe explicitly setting swapper_pg_dir would be cleaner?

Is there a test that should be running in a CI somewhere to catch cases like
this where the permissions are not working as expected

My test patch that seems to be working:

----8<-----

diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index 8a63b4c..6276b234 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -627,12 +627,10 @@ static struct section_perm ro_perms[] = {
   * safe to be called with preemption disabled, as under stop_machine().
   */
  static inline void section_update(unsigned long addr, pmdval_t mask,
-				  pmdval_t prot)
+				  pmdval_t prot, struct mm_struct *mm)
  {
-	struct mm_struct *mm;
  	pmd_t *pmd;
  
-	mm = current->active_mm;
  	pmd = pmd_offset(pud_offset(pgd_offset(mm, addr), addr), addr);
  
  #ifdef CONFIG_ARM_LPAE
@@ -656,7 +654,7 @@ static inline bool arch_has_strict_perms(void)
  	return !!(get_cr() & CR_XP);
  }
  
-#define set_section_perms(perms, field)	{				\
+#define set_section_perms(perms, field, all)	{			\
  	size_t i;							\
  	unsigned long addr;						\
  									\
@@ -674,31 +672,35 @@ static inline bool arch_has_strict_perms(void)
  									\
  		for (addr = perms[i].start;				\
  		     addr < perms[i].end;				\
-		     addr += SECTION_SIZE)				\
+		     addr += SECTION_SIZE) {				\
  			section_update(addr, perms[i].mask,		\
-				       perms[i].field);			\
+				       perms[i].field, current->active_mm); \
+			if (all)					\
+				section_update(addr, perms[i].mask,	\
+					perms[i].field, &init_mm);	\
+		}							\
  	}								\
  }
  
-static inline void fix_kernmem_perms(void)
+void fix_kernmem_perms(void)
  {
-	set_section_perms(nx_perms, prot);
+	set_section_perms(nx_perms, prot, true);
  }
  
  #ifdef CONFIG_DEBUG_RODATA
  void mark_rodata_ro(void)
  {
-	set_section_perms(ro_perms, prot);
+	set_section_perms(ro_perms, prot, true);
  }
  
  void set_kernel_text_rw(void)
  {
-	set_section_perms(ro_perms, clear);
+	set_section_perms(ro_perms, clear, false);
  }
  
  void set_kernel_text_ro(void)
  {
-	set_section_perms(ro_perms, prot);
+	set_section_perms(ro_perms, prot, false);
  }
  #endif /* CONFIG_DEBUG_RODATA */