[PATCH v2] kasan: avoid overflowing quarantine size on low memory systems

[PATCH v2] kasan: avoid overflowing quarantine size on low memory systems

[Alexander Potapenko]

If the total amount of memory assigned to quarantine is less than the
amount of memory assigned to per-cpu quarantines, |new_quarantine_size|
may overflow. Instead, set it to zero.

Reported-by: Dmitry Vyukov <dvyukov@google.com>
Fixes: 55834c59098d ("mm: kasan: initial memory quarantine
implementation")
Signed-off-by: Alexander Potapenko <glider@google.com>
---
v2: Don't print the warning if we don't have enough memory for quarantine.
---
 mm/kasan/quarantine.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/mm/kasan/quarantine.c b/mm/kasan/quarantine.c
index 65793f1..8c29938 100644
--- a/mm/kasan/quarantine.c
+++ b/mm/kasan/quarantine.c
@@ -196,7 +196,7 @@ void quarantine_put(struct kasan_free_meta *info, struct kmem_cache *cache)
 
 void quarantine_reduce(void)
 {
-	size_t new_quarantine_size;
+	size_t new_quarantine_size, percpu_quarantines;
 	unsigned long flags;
 	struct qlist_head to_free = QLIST_INIT;
 	size_t size_to_free = 0;
@@ -214,7 +214,9 @@ void quarantine_reduce(void)
 	 */
 	new_quarantine_size = (READ_ONCE(totalram_pages) << PAGE_SHIFT) /
 		QUARANTINE_FRACTION;
-	new_quarantine_size -= QUARANTINE_PERCPU_SIZE * num_online_cpus();
+	percpu_quarantines = QUARANTINE_PERCPU_SIZE * num_online_cpus();
+	new_quarantine_size = (new_quarantine_size < percpu_quarantines) ?
+		0 : new_quarantine_size - percpu_quarantines;
 	WRITE_ONCE(quarantine_size, new_quarantine_size);
 
 	last = global_quarantine.head;
-- 
2.8.0.rc3.226.g39d4020

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH v2] kasan: avoid overflowing quarantine size on low memory systems

[Andrey Ryabinin]

On 08/02/2016 01:27 PM, Alexander Potapenko wrote:
> If the total amount of memory assigned to quarantine is less than the
> amount of memory assigned to per-cpu quarantines, |new_quarantine_size|
> may overflow. Instead, set it to zero.
> 
> Reported-by: Dmitry Vyukov <dvyukov@google.com>
> Fixes: 55834c59098d ("mm: kasan: initial memory quarantine
> implementation")
> Signed-off-by: Alexander Potapenko <glider@google.com>

Acked-by: Andrey Ryabinin <aryabinin@virtuozzo.com>

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH v2] kasan: avoid overflowing quarantine size on low memory systems

[Andrew Morton]

On Tue,  2 Aug 2016 12:27:00 +0200 Alexander Potapenko <glider@google.com> wrote:

> If the total amount of memory assigned to quarantine is less than the
> amount of memory assigned to per-cpu quarantines, |new_quarantine_size|
> may overflow. Instead, set it to zero.
> 
> ...
>
> --- a/mm/kasan/quarantine.c
> +++ b/mm/kasan/quarantine.c
> @@ -196,7 +196,7 @@ void quarantine_put(struct kasan_free_meta *info, struct kmem_cache *cache)
>  
>  void quarantine_reduce(void)
>  {
> -	size_t new_quarantine_size;
> +	size_t new_quarantine_size, percpu_quarantines;
>  	unsigned long flags;
>  	struct qlist_head to_free = QLIST_INIT;
>  	size_t size_to_free = 0;
> @@ -214,7 +214,9 @@ void quarantine_reduce(void)
>  	 */
>  	new_quarantine_size = (READ_ONCE(totalram_pages) << PAGE_SHIFT) /
>  		QUARANTINE_FRACTION;
> -	new_quarantine_size -= QUARANTINE_PERCPU_SIZE * num_online_cpus();
> +	percpu_quarantines = QUARANTINE_PERCPU_SIZE * num_online_cpus();
> +	new_quarantine_size = (new_quarantine_size < percpu_quarantines) ?
> +		0 : new_quarantine_size - percpu_quarantines;
>  	WRITE_ONCE(quarantine_size, new_quarantine_size);
>  
>  	last = global_quarantine.head;

Confused.  Which kernel version is this supposed to apply to?

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH v2] kasan: avoid overflowing quarantine size on low memory systems

[Alexander Potapenko]

On Thu, Aug 11, 2016 at 12:50 AM, Andrew Morton
<akpm@linux-foundation.org> wrote:
> On Tue,  2 Aug 2016 12:27:00 +0200 Alexander Potapenko <glider@google.com> wrote:
>
>> If the total amount of memory assigned to quarantine is less than the
>> amount of memory assigned to per-cpu quarantines, |new_quarantine_size|
>> may overflow. Instead, set it to zero.
>>
>> ...
>>
>> --- a/mm/kasan/quarantine.c
>> +++ b/mm/kasan/quarantine.c
>> @@ -196,7 +196,7 @@ void quarantine_put(struct kasan_free_meta *info, struct kmem_cache *cache)
>>
>>  void quarantine_reduce(void)
>>  {
>> -     size_t new_quarantine_size;
>> +     size_t new_quarantine_size, percpu_quarantines;
>>       unsigned long flags;
>>       struct qlist_head to_free = QLIST_INIT;
>>       size_t size_to_free = 0;
>> @@ -214,7 +214,9 @@ void quarantine_reduce(void)
>>        */
>>       new_quarantine_size = (READ_ONCE(totalram_pages) << PAGE_SHIFT) /
>>               QUARANTINE_FRACTION;
>> -     new_quarantine_size -= QUARANTINE_PERCPU_SIZE * num_online_cpus();
>> +     percpu_quarantines = QUARANTINE_PERCPU_SIZE * num_online_cpus();
>> +     new_quarantine_size = (new_quarantine_size < percpu_quarantines) ?
>> +             0 : new_quarantine_size - percpu_quarantines;
>>       WRITE_ONCE(quarantine_size, new_quarantine_size);
>>
>>       last = global_quarantine.head;
>
> Confused.  Which kernel version is this supposed to apply to?
This is the second version of the patch which should've been applied
to the mainline instead of v1.
But since v1 has already hit upstream, this patch makes sense no more.
If WARN_ONCE (which is currently present in this code) is a big deal,
I can send a new patch that removes it.


-- 
Alexander Potapenko
Software Engineer

Google Germany GmbH
Erika-Mann-Straße, 33
80636 München

Geschäftsführer: Matthew Scott Sucherman, Paul Terence Manicle
Registergericht und -nummer: Hamburg, HRB 86891
Sitz der Gesellschaft: Hamburg

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH v2] kasan: avoid overflowing quarantine size on low memory systems

[Andrey Ryabinin]

On 08/11/2016 04:42 PM, Alexander Potapenko wrote:
> On Thu, Aug 11, 2016 at 12:50 AM, Andrew Morton
> <akpm@linux-foundation.org> wrote:
>> On Tue,  2 Aug 2016 12:27:00 +0200 Alexander Potapenko <glider@google.com> wrote:
>>
>>> If the total amount of memory assigned to quarantine is less than the
>>> amount of memory assigned to per-cpu quarantines, |new_quarantine_size|
>>> may overflow. Instead, set it to zero.
>>>
>>> ...
>>>
>>> --- a/mm/kasan/quarantine.c
>>> +++ b/mm/kasan/quarantine.c
>>> @@ -196,7 +196,7 @@ void quarantine_put(struct kasan_free_meta *info, struct kmem_cache *cache)
>>>
>>>  void quarantine_reduce(void)
>>>  {
>>> -     size_t new_quarantine_size;
>>> +     size_t new_quarantine_size, percpu_quarantines;
>>>       unsigned long flags;
>>>       struct qlist_head to_free = QLIST_INIT;
>>>       size_t size_to_free = 0;
>>> @@ -214,7 +214,9 @@ void quarantine_reduce(void)
>>>        */
>>>       new_quarantine_size = (READ_ONCE(totalram_pages) << PAGE_SHIFT) /
>>>               QUARANTINE_FRACTION;
>>> -     new_quarantine_size -= QUARANTINE_PERCPU_SIZE * num_online_cpus();
>>> +     percpu_quarantines = QUARANTINE_PERCPU_SIZE * num_online_cpus();
>>> +     new_quarantine_size = (new_quarantine_size < percpu_quarantines) ?
>>> +             0 : new_quarantine_size - percpu_quarantines;
>>>       WRITE_ONCE(quarantine_size, new_quarantine_size);
>>>
>>>       last = global_quarantine.head;
>>
>> Confused.  Which kernel version is this supposed to apply to?
> This is the second version of the patch which should've been applied
> to the mainline instead of v1.
> But since v1 has already hit upstream, this patch makes sense no more.
> If WARN_ONCE (which is currently present in this code) is a big deal,
> I can send a new patch that removes it.
> 

Probably not a big deal, but please, send the patch anyway.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>