Re: [RFC 1/2] staging: ion: shrink page-pool by page unit

From: Devendra Naga
Date: Thu Oct 23 2014 - 19:50:04 EST


On Fri, Oct 24, 2014 at 5:01 AM, Gioh Kim <gioh.kim@xxxxxxx> wrote:
>
>
> 2014-10-24 ìì 8:16, Gioh Kim ì ê:
>
>>
>>
>> 2014-10-24 ìì 5:36, Devendra Naga ì ê:
>>>
>>> Hello,
>>>
>>> On Thu, Oct 23, 2014 at 6:22 PM, Gioh Kim <gioh.kim@xxxxxxx> wrote:
>>>> sys_heap = container_of(heap, struct ion_system_heap, heap);
>>>> @@ -219,7 +219,10 @@ static int ion_system_heap_shrink(struct ion_heap
>>>> *heap, gfp_t gfp_mask,
>>>> for (i = 0; i < num_orders; i++) {
>>>> struct ion_page_pool *pool = sys_heap->pools[i];
>>>>
>>>> - nr_total += ion_page_pool_shrink(pool, gfp_mask,
>>>> nr_to_scan);
>>>> + nr_freed += ion_page_pool_shrink(pool, gfp_mask,
>>>> nr_to_scan);
>>>
>>>
>>> nr_freed is not initialised to zero. This might result in updating of
>>> a garbage value stored initially after nr_freed is created.
>>
>>
>> Yes, your're right. It's my mistake.
>> I testes this on my platform with custom kernel and copied codes.
>> I missed a line for it.
>> I'm sorry.
>
>
> Following is the correct code:
> nr_freed = ion_page_pool_shrink(pool, gfp_mask, nr_to_scan);
>

Thanks that is correct. You can send another patch with this fix (it
can be patchset v2 or rfc v2).

>
>>
>>>
>>>> + nr_total += nr_freed;
>>>> + /* nr_to_scan can be negative */
>>>> + nr_to_scan -= nr_freed;
>>>> }
>>>>
>>>> return nr_total;
>>>> --
>>>> 1.7.9.5
>>>>
>>>> --
>>>> To unsubscribe from this list: send the line "unsubscribe linux-kernel"
>>>> in
>>>> the body of a message to majordomo@xxxxxxxxxxxxxxx
>>>> More majordomo info at http://vger.kernel.org/majordomo-info.html
>>>> Please read the FAQ at http://www.tux.org/lkml/
>>>
>>>
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/