From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from smtp.nokia.com ([131.228.20.173] helo=mgw-ext14.nokia.com)
	by canuck.infradead.org with esmtps (Exim 4.63 #1 (Red Hat Linux))
	id 1IG9ys-0001Pq-AV
	for linux-mtd@lists.infradead.org; Wed, 01 Aug 2007 04:54:59 -0400
Message-ID: <46B049B1.3000608@nokia.com>
Date: Wed, 01 Aug 2007 11:52:01 +0300
From: Adrian Hunter <ext-adrian.hunter@nokia.com>
MIME-Version: 1.0
To: kmpark@infradead.org
Subject: Re: [RFC PATCH] [MTD] [OneNAND] Cache Read support
References: <001501c7d3d1$bcf66b40$e1ac580a@swcenter.sec.samsung.co.kr>
	<46B0272C.6040504@nokia.com>
	<001c01c7d40a$e57aa4d0$e1ac580a@swcenter.sec.samsung.co.kr>
In-Reply-To: <001c01c7d40a$e57aa4d0$e1ac580a@swcenter.sec.samsung.co.kr>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Cc: linux-mtd@lists.infradead.org
List-Id: Linux MTD discussion mailing list <linux-mtd.lists.infradead.org>
List-Unsubscribe: <http://lists.infradead.org/mailman/listinfo/linux-mtd>,
	<mailto:linux-mtd-request@lists.infradead.org?subject=unsubscribe>
List-Archive: <http://lists.infradead.org/pipermail/linux-mtd>
List-Post: <mailto:linux-mtd@lists.infradead.org>
List-Help: <mailto:linux-mtd-request@lists.infradead.org?subject=help>
List-Subscribe: <http://lists.infradead.org/mailman/listinfo/linux-mtd>,
	<mailto:linux-mtd-request@lists.infradead.org?subject=subscribe>

ext Kyungmin Park wrote:
>  
>> ext Kyungmin Park wrote:
>>> This patch supports the Cache Read feature in OneNAND.
>>> It's similar with read-while-loading except while read it does sensing the page buffer in NAND
>> core.
>>> So it's called Transfer-While-Sensing. you can find it in OneNAND Spec. in detail.
>>>
>>> Now there's no big performance gain in our test board, Apollon(OMAP2). But others are maybe
>> different.
>>> Any comments are welcome.
>> Cool!
>>
>> Presume you have run the NAND tests
> 
> Sure, it passed NAND tests.
> 
> Could you review the code?

I will try to find some time to look at it maybe this week.

> I wonder why there's no performance gain. It's faster than read-while-loading in the Spec.
> Software overhead? Or there's not many 2 or more pages read?

How are you measuring performance?  I would suggest either dd (if you don't have bad blocks)
or write your own program.

As I see it, unless the read from dataRAM is faster than the sensing/transfer, there won't be
a performance improvement.  So the faster the OneNAND frequency the better.  Ditto bus and
CPU frequencies probably (for memcpy).