[Linaro-mm-sig] [PATCHv16 0/9] Contiguous Memory Allocator

[Maxime Coquelin]

On 10/11/2011 12:50 PM, Marek Szyprowski wrote:
> Hello,
>
> On Tuesday, October 11, 2011 9:30 AM Maxime Coquelin wrote:
>
>> On 10/11/2011 09:17 AM, Marek Szyprowski wrote:
>>> On Monday, October 10, 2011 2:08 PM Maxime Coquelin wrote:
>>>
>>>        During our stress tests, we encountered some problems :
>>>
>>>        1) Contiguous allocation lockup:
>>>            When system RAM is full of Anon pages, if we try to allocate a
>>> contiguous buffer greater than the min_free value, we face a
>>> dma_alloc_from_contiguous lockup.
>>>            The expected result would be dma_alloc_from_contiguous() to fail.
>>>            The problem is reproduced systematically on our side.
>>> Thanks for the report. Do you use Android's lowmemorykiller? I haven't
>>> tested CMA on Android kernel yet. I have no idea how it will interfere
>>> with Android patches.
>>>
>> The software used for this test (v16) is a generic 3.0 Kernel and a
>> minimal filesystem using Busybox.
> I'm really surprised. Could you elaborate a bit how to trigger this issue?

At system startup, I drop caches (sync && echo 3 > 
/proc/sys/vm/drop_caches) and check how much memory is free.
For example, in my case, only 15MB is used on the 270MB available on the 
system, so I got 255MB of free memory. Note that the min_free is 4MB in 
my case.
In userspace, I allocate 230MB using malloc(), the free memory is now 25MB.
Finaly, I ask for a contiguous allocation of 64MB using CMA, the result 
is a lockup in dma_alloc_from_contiguous().

> I've did several tests and I never get a lockup. Allocation failed from time
> to time though.
When it succeed, what is the behaviour on your side? Is the OOM triggered?

Regards,
Maxime