From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <5397CBF9.800@atmel.com> Date: Wed, 11 Jun 2014 11:24:41 +0800 From: Bo Shen MIME-Version: 1.0 To: Josh Wu Subject: Re: [PATCH] mtd: atmel_nand: make ecc parameters same as definition References: <1402041728-12056-1-git-send-email-voice.shen@atmel.com> <53958CCE.5080307@atmel.com> In-Reply-To: <53958CCE.5080307@atmel.com> Content-Type: text/plain; charset="ISO-8859-1"; format=flowed Content-Transfer-Encoding: 7bit Cc: linux-mtd@lists.infradead.org, computersforpeace@gmail.com, nicolas.ferre@atmel.com, linux-arm-kernel@lists.infradead.org List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Hi Josh, On 06/09/2014 06:30 PM, Josh Wu wrote: > Hi, Bo > > Thanks for the patch. > > On 6/6/2014 4:02 PM, Bo Shen wrote: >> If the ecc parameter is not the same as definition, when the >> mtd core check these parameters, it will give the unexpected >> result. > Could you be more specific to tell what kind of result will happened? For example: When try to calculate how many bits can correct in one page according to the definition, it will give the wrong result. one page correct bits = (mtd->writesize * ecc->strength) / ecc->size; take the following use case as an example: mtd->writesize = 2048 bytes, ecc->strength = 4 bytes, (for 512 bytes) before this patch, the ecc->size = 2048, so the result is 4 bytes. after this patch, the ecc->size = 512, so the result is 16 bytes. The maximum correct bytes is 16 bytes while not 4 bytes. So, if not align Best Regards, Bo Shen From mboxrd@z Thu Jan 1 00:00:00 1970 From: voice.shen@atmel.com (Bo Shen) Date: Wed, 11 Jun 2014 11:24:41 +0800 Subject: [PATCH] mtd: atmel_nand: make ecc parameters same as definition In-Reply-To: <53958CCE.5080307@atmel.com> References: <1402041728-12056-1-git-send-email-voice.shen@atmel.com> <53958CCE.5080307@atmel.com> Message-ID: <5397CBF9.800@atmel.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Hi Josh, On 06/09/2014 06:30 PM, Josh Wu wrote: > Hi, Bo > > Thanks for the patch. > > On 6/6/2014 4:02 PM, Bo Shen wrote: >> If the ecc parameter is not the same as definition, when the >> mtd core check these parameters, it will give the unexpected >> result. > Could you be more specific to tell what kind of result will happened? For example: When try to calculate how many bits can correct in one page according to the definition, it will give the wrong result. one page correct bits = (mtd->writesize * ecc->strength) / ecc->size; take the following use case as an example: mtd->writesize = 2048 bytes, ecc->strength = 4 bytes, (for 512 bytes) before this patch, the ecc->size = 2048, so the result is 4 bytes. after this patch, the ecc->size = 512, so the result is 16 bytes. The maximum correct bytes is 16 bytes while not 4 bytes. So, if not align Best Regards, Bo Shen