From s.kr713@yahoo.com Tue May 29 06:26:55 2012 Received: from nm22.bullet.mail.ne1.yahoo.com ([98.138.90.85]) by stoneboat.aleph1.co.uk with smtp (Exim 4.72) (envelope-from ) id 1SZExI-0006fG-HW for yaffs@lists.aleph1.co.uk; Tue, 29 May 2012 06:26:55 +0100 Received: from [98.138.90.52] by nm22.bullet.mail.ne1.yahoo.com with NNFMP; 29 May 2012 05:26:46 -0000 Received: from [98.138.89.160] by tm5.bullet.mail.ne1.yahoo.com with NNFMP; 29 May 2012 05:26:46 -0000 Received: from [127.0.0.1] by omp1016.mail.ne1.yahoo.com with NNFMP; 29 May 2012 05:26:46 -0000 X-Yahoo-Newman-Property: ymail-3 X-Yahoo-Newman-Id: 219050.47152.bm@omp1016.mail.ne1.yahoo.com Received: (qmail 92123 invoked by uid 60001); 29 May 2012 05:26:46 -0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=yahoo.com; s=s1024; t=1338269205; bh=o9Q/J8E5Y4QDl1//HEmyMWmd6LEncQHhWOIlykNsRXs=; h=X-YMail-OSG:Received:X-Mailer:References:Message-ID:Date:From:Reply-To:Subject:To:In-Reply-To:MIME-Version:Content-Type; b=sDs1h5hfCBuefvq/plmPysz77KGDTFndLzvDjwjrdPVIifF9sp6KEt6cDQo3Q2BP590Kf+CArq7ovTctSGN/K0Cpl/4JeC7Ds13MiYymNWh8l/Gbb7/2DBnvnoYjVkIAvHTjJNMsIbFLxPX3xeUsbgwL9tBQfjq/7L94RfE8tyY= DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=X-YMail-OSG:Received:X-Mailer:References:Message-ID:Date:From:Reply-To:Subject:To:In-Reply-To:MIME-Version:Content-Type; b=AsaxNIUlyCP4sHCXfbaeg23VEygEIZZ4yBb0mIoCIqYc4+l66J3RHVaVrsr7UyjAtij6Qeev6khbWJGrwKNyx4RkONwoUN561NKiQMlZY7Af/zuXDO2wnpbuRuWs6Uqei0CuHjcBcQVqZ5MghW5NKVebgtJqa8p8vn59ZEA5O40=; X-YMail-OSG: SBaeGn0VM1lkXI9dVXg93O2cNj8zlkCVCn5gePh.JZRq.l0 dFjekXlSl2fYPAjzn33xhZq12NyjG5Ln4gx3rau_87bD9tknz44x.ngPt16_ l8nFy7tuS009K33fUqAT5rDkzmZkwodxrvDO.R2yOoUMwawD87l7VUE596q0 PwJnBVarM3TrjvroHfuMEq_zZu4sXejKgV0NRosXTHQVdQwVCrVOM9v5IjUA EG5aOTSyQaiVO1vcg29Tj_5YRt_I1QU6vsOX87b.0HJkIM9izDl7yJKe6sVg ucmiIL5.NO1phRu1dLprjTrbS9yojpgODshsC6G.2eHYGtxQWInZuOtceDWf 7AcX6PAfVAR0wS7zxTrhVrWwPonKlmzrXvcyXlnylrGCTgmVXB7svT8M5I8. AH52efFDxckRB5h0- Received: from [217.218.56.12] by web121202.mail.ne1.yahoo.com via HTTP; Mon, 28 May 2012 22:26:45 PDT X-Mailer: YahooMailWebService/0.8.118.349524 References: <1338014291.23453.YahooMailNeo@web121201.mail.ne1.yahoo.com> <201205280916.30589.manningc2@actrix.gen.nz> Message-ID: <1338269205.83558.YahooMailNeo@web121202.mail.ne1.yahoo.com> Date: Mon, 28 May 2012 22:26:45 -0700 (PDT) From: "S. K." To: Charles Manning , "yaffs@lists.aleph1.co.uk" In-Reply-To: <201205280916.30589.manningc2@actrix.gen.nz> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="88058567-1600776949-1338269205=:83558" X-SA-Exim-Connect-IP: 98.138.90.85 X-SA-Exim-Mail-From: s.kr713@yahoo.com X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on stoneboat.aleph1.co.uk X-Spam-Level: * X-Spam-Status: No, score=1.7 required=4.5 tests=AWL,BAYES_00, FREEMAIL_ENVFROM_END_DIGIT,FREEMAIL_FROM,FREEMAIL_REPLYTO_END_DIGIT, HTML_MESSAGE, RCVD_IN_DNSWL_NONE, REPTO_QUOTE_YAHOO autolearn=no version=3.3.1 X-SA-Exim-Version: 4.2.1 (built Mon, 22 Mar 2010 06:51:10 +0000) X-SA-Exim-Scanned: Yes (on stoneboat.aleph1.co.uk) Subject: Re: [Yaffs] yaffs2 ECC algorithm X-BeenThere: yaffs@lists.aleph1.co.uk X-Mailman-Version: 2.1.13 Precedence: list Reply-To: "S. K." List-Id: Discussion of YAFFS NAND flash filesystem List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 29 May 2012 05:26:55 -0000 --88058567-1600776949-1338269205=:83558 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable =0A=0AThanks for reply!=0A=0A"use_nand_ecc" is set to 1 and "no_tags_ecc" i= s set to 0 in /proc/yaffs. Does this path stores configuration setting?=0AI= try to set "use_nand_ecc" to 0 but I can not. How can I change it?=0A=0A= =0Aafter setting the flages, How can I be sure that ECC works? How can I se= e what it is doing with data?=0A=0A=0AAnd is there any where that I select = the mode?(I mounted yaffs2, so I thought that I can just use it!)=0A=0A=0A= =0A=0A=0A=0A________________________________=0A From: Charles Manning =0ATo: yaffs@lists.aleph1.co.uk; S. K. = =0ASent: Monday, May 28, 2012 1:46 AM=0ASubject: Re: [Yaffs] yaffs2 ECC al= gorithm=0A =0AOn Saturday 26 May 2012 18:38:11 S. K. wrote:=0A> Hello every= one!=0A>=0A> I have already mounted yaffs2 on my ubuntu12.04 with nand simu= lator .There=0A> is something not clear to me. Yaffs2 has ECC algorithm but= , Where it will=0A> apply?=0A=0AThe ECC algorithm is used in different ways= depending on configuration =0Asettings.=0A=0AIn yaffs1 mode, the ECC is us= ed if use_nand_ecc is NOT selected.=0A=0AIn yaffs2 mode, the ecc is used to= protect the tags if no_tags_ecc is NOT set.=0A=0A=0A>=0A> I want to check = performance of its ECC algorithm on nand simulator and on=0A> flash device.= can you help me exactly what I have to do?=0A=0AWithout more information i= t is difficult to advise you.=0A=0A>=0A> and another question: I understood= that flash memories can be SLC or MLC.=0A> SLC uses binary algorithms and= =A0 MLC uses nonbinary one. Is it correct?=0A=0AWell all flash cells are re= ally analogue devices - not digitial.=0A=0ASLC devices store only one bit p= er cell so they only have one threshold. MLC =0Adevices store multiple bits= per cell so have multiple thresholds.=0A=0A=0A=0A> Is =0A> it influence th= e ECC algorithm? If yes what is the yaffs ECC=0A> algorithm?(Hamming or ...= ) what kind of flashs does yaffs=0A> support?(simulator and device)=0A=0ASi= nce MLC has less margin for error, MLC typically requires more robust ECC.= =0A=0ASLC devices will typically use 1-bit ECC using hamming codes (thats w= hat =0Ayaffs_ecc.c does). MLC will typically use BCH codes.=0A=0ABCH codes = are computationally expensive and will often use hardware in the =0ANAND co= ntroller.=0A=0A-- Charles --88058567-1600776949-1338269205=:83558 Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable

Thanks for r= eply!

"use_nand_ecc" is set to 1 and "no_tags_ecc"= is set to 0 in /proc/yaffs. Does this path stores configuration setting?
I try to set "use_nand_ecc" to 0 but I can not. How can I change i= t?


after setting the flages, How can I be sure= that ECC works? How can I see what it is doing with data?


And is there any where that I select the mode?(I mo= unted yaffs2, so I thought that I can just use it!)
=0A

=0A=0A



From= : Charles Manning <cdhmanning@gmail.com>
To: yaffs@lists.aleph1.co.uk; S. K. <= ;s.kr713@yahoo.com>
Sent: Monday, May 28, 2012 1:46 AM
Subject: Re: [Yaffs] yaffs2 ECC algorithm
=
=0AOn Saturday 26 May 2012 18:38:11 S. K. wrote:
> Hello everyon= e!
>
> I have already mounted yaffs2 on my ubuntu12.04 with nan= d simulator .There
> is something not clear to me. Yaffs2 has ECC alg= orithm but, Where it will
> apply?

The ECC algorithm is used i= n different ways depending on configuration
settings.

In yaffs1 = mode, the ECC is used if use_nand_ecc is NOT selected.

In yaffs2 mod= e, the ecc is used to protect the tags if no_tags_ecc is NOT set.

>
> I want to check performance of its ECC algorithm on nand sim= ulator and on
> flash device. can you help me exactly what I have to = do?

Without more information it is difficult to advise you.

&= gt;
> and another question: I understood that flash memories can be S= LC or MLC.
> SLC uses binary algorithms and  MLC uses nonbinary = one. Is it correct?

Well all flash cells are really analogue devices - not digitial.

SLC devices store only one bit per cell so they onl= y have one threshold. MLC
devices store multiple bits per cell so have = multiple thresholds.



> Is
> it influence the ECC a= lgorithm? If yes what is the yaffs ECC
> algorithm?(Hamming or ...) w= hat kind of flashs does yaffs
> support?(simulator and device)
Since MLC has less margin for error, MLC typically requires more robust EC= C.

SLC devices will typically use 1-bit ECC using hamming codes (tha= ts what
yaffs_ecc.c does). MLC will typically use BCH codes.

BCH= codes are computationally expensive and will often use hardware in the NAND controller.

-- Charles



--88058567-1600776949-1338269205=:83558--