Re: CMA related memory questions

From: Joonsoo Kim <js1304@gmail.com>
To: Heesub Shin <heesub.shin@samsung.com>
Cc: Jun Nie <jun.nie@linaro.org>, Arnd Bergmann <arnd@arndb.de>,
	Marek Szyprowski <m.szyprowski@samsung.com>,
	Shawn Guo <shawn.guo@linaro.org>,
	"mark.brown@linaro.org, wan.zhijun" <wan.zhijun@zte.com.cn>,
	linux-arm-kernel@lists.infradead.org,
	Linux Memory Management List <linux-mm@kvack.org>,
	sunae.seo@samsung.com, cmlaika.kim@samsung.com
Subject: Re: CMA related memory questions
Date: Fri, 30 Jan 2015 21:58:48 +0900	[thread overview]
Message-ID: <CAAmzW4OfrKt0XK5bDO0TNR5Ofv=rLDkpS7HAOCk0NRV1WdOhNg@mail.gmail.com> (raw)
In-Reply-To: <54CAF9A4.1040606@samsung.com>

2015-01-30 12:25 GMT+09:00 Heesub Shin <heesub.shin@samsung.com>:
>
>
> On 01/30/2015 11:57 AM, Jun Nie wrote:
>>
>> On 2015年01月30日 10:36, Jun Nie wrote:
>>>
>>> Hi Marek & Arnd,
>>>
>>> Did you ever know issue that free CMA memory is high, but system is
>>> hungry for memory and page cache is very low? I am enabling CMA in
>>> Android on my board with 512MB memory and see FreeMem in /proc/meminfo
>>> increase a lot with CMA comparing the reservation solution on boot. But
>>> I find system is not borrowing memory from CMA pool when running 3dmark
>>> (high webkit workload at start). Because the FreeMem size is high, but
>>> cache size decreasing significantly to several MB during benchmark run,
>>> I suppose system is trying to reclaim memory from pagecache for new
>>> allocation. My question is that what API that page cache and webkit
>>> related functionality are using to allocate memory. Maybe page cache
>>> require memory that is not movable/reclaimable memory, where we may have
>>> optimization to go thru dma_alloc_xxx to borrow CMA memory? I suppose
>>> app level memory allocation shall be movable/reclaimable memory and can
>>> borrow from CMA pool, but not sure whether the flag match the
>>> movable/reclaimable memory and go thru the right path.

Hello,

Maybe, you experienced the problem what I tried to solve.
CMA freepage allocation logic in mainline doesn't work well now since
they are only allocated in fallback case. See below link for detailed
explanation.

https://lkml.org/lkml/2014/5/28/64

This problem can be solved by aggressive allocation approach in that link, but,
there are too many issues left in CMA. So, I'm trying to implement ZONE_CMA
now. Prototyping is nearly finished so I will send it soon.

Thanks.

--
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>