Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII;
 format=flowed
Content-Transfer-Encoding: 7bit
Date:   Tue, 08 May 2018 12:56:13 +0530
From:   Abhishek Sahu <absahu@codeaurora.org>
To:     Boris Brezillon <boris.brezillon@bootlin.com>
Cc:     Archit Taneja <architt@codeaurora.org>,
        Richard Weinberger <richard@nod.at>,
        linux-arm-msm@vger.kernel.org, linux-kernel@vger.kernel.org,
        Marek Vasut <marek.vasut@gmail.com>,
        linux-mtd@lists.infradead.org,
        Miquel Raynal <miquel.raynal@bootlin.com>,
        Andy Gross <andy.gross@linaro.org>,
        Brian Norris <computersforpeace@gmail.com>,
        David Woodhouse <dwmw2@infradead.org>
Subject: Re: [PATCH v2 04/14] mtd: rawnand: qcom: use the ecc strength from
 device parameter
In-Reply-To: <20180507102840.3624fe01@bbrezillon>
References: <1525350041-22995-1-git-send-email-absahu@codeaurora.org>
 <1525350041-22995-5-git-send-email-absahu@codeaurora.org>
 <20180507102840.3624fe01@bbrezillon>
Message-ID: <3d9cc66c8344850c435d761c0c5b92d1@codeaurora.org>
User-Agent: Roundcube Webmail/1.2.5
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 2018-05-07 13:58, Boris Brezillon wrote:
> On Thu,  3 May 2018 17:50:31 +0530
> Abhishek Sahu <absahu@codeaurora.org> wrote:
> 
>> Currently the driver uses the ECC strength specified in DT.
>> The QPIC/EBI2 NAND supports 4 or 8-bit ECC correction. The same
>> kind of board can have different NAND parts so use the ECC
>> strength from device parameters if it is not specified in DT.
>> 
>> Signed-off-by: Abhishek Sahu <absahu@codeaurora.org>
>> ---
>> * Changes from v1:
>> 
>>   1. Removed the custom logic and used the helper fuction.
>> 
>>  drivers/mtd/nand/raw/qcom_nandc.c | 31 
>> ++++++++++++++++++++++---------
>>  1 file changed, 22 insertions(+), 9 deletions(-)
>> 
>> diff --git a/drivers/mtd/nand/raw/qcom_nandc.c 
>> b/drivers/mtd/nand/raw/qcom_nandc.c
>> index b554fb6..a8d71ce 100644
>> --- a/drivers/mtd/nand/raw/qcom_nandc.c
>> +++ b/drivers/mtd/nand/raw/qcom_nandc.c
>> @@ -2315,13 +2315,21 @@ static int qcom_nand_ooblayout_free(struct 
>> mtd_info *mtd, int section,
>>  	.free = qcom_nand_ooblayout_free,
>>  };
>> 
>> +static int
>> +qcom_nandc_calc_ecc_bytes(int step_size, int strength)
>> +{
>> +	return strength == 4 ? 12 : 16;
>> +}
>> +NAND_ECC_CAPS_SINGLE(qcom_nandc_ecc_caps, qcom_nandc_calc_ecc_bytes,
>> +		     NANDC_STEP_SIZE, 4, 8);
>> +
>>  static int qcom_nand_host_setup(struct qcom_nand_host *host)
>>  {
>>  	struct nand_chip *chip = &host->chip;
>>  	struct mtd_info *mtd = nand_to_mtd(chip);
>>  	struct nand_ecc_ctrl *ecc = &chip->ecc;
>>  	struct qcom_nand_controller *nandc = get_qcom_nand_controller(chip);
>> -	int cwperpage, bad_block_byte;
>> +	int cwperpage, bad_block_byte, ret;
>>  	bool wide_bus;
>>  	int ecc_mode = 1;
>> 
>> @@ -2334,8 +2342,20 @@ static int qcom_nand_host_setup(struct 
>> qcom_nand_host *host)
>>  		return -EINVAL;
>>  	}
>> 
>> +	cwperpage = mtd->writesize / ecc->size;
> 
> Looks like you're still expecting nand-ecc-step-size to be defined in
> the DT, which does not really make sense since you only support one
> size: NANDC_STEP_SIZE.
> 
> You should remove the
> 
> 	if (ecc->size != NANDC_STEP_SIZE) {
> 		dev_err(nandc->dev, "invalid ecc size\n");
> 		return -EINVAL;
> 	}
> 
> block, then do:
> 
> 	cwperpage = mtd->writesize / NANDC_STEP_SIZE;
> 
> and finally let the nand_ecc_param_setup() function choose the best
> config for you.
> 

  Correct Boris.
  It only supports one step size so we can remove this DT
  property. I will make the changes.

>> +
>> +	/*
>> +	 * Each CW has 4 available OOB bytes which will be protected with 
>> ECC
>> +	 * so remaining bytes can be used for ECC.
>> +	 */
>> +	ret = nand_ecc_param_setup(chip, &qcom_nandc_ecc_caps,
>> +				   mtd->oobsize - (cwperpage << 2));
> 
> Please stop doing useless optimizations like. That brings nothing and
> obfuscates the code a bit more. You say in the comment that each
> codeword has 4 protected OOB bytes that can be used by the upper layer,
> so just do (cwperpage * 4) and let gcc optimize that for you.
> 

  Sure. I will change it.

  Thanks,
  Abhishek

>> +	if (ret) {
>> +		dev_err(nandc->dev, "No valid ecc settings possible\n");
>> +		return ret;
>> +	}
>> +
>>  	wide_bus = chip->options & NAND_BUSWIDTH_16 ? true : false;
>> -
>>  	if (ecc->strength >= 8) {
>>  		/* 8 bit ECC defaults to BCH ECC on all platforms */
>>  		host->bch_enabled = true;
>> @@ -2403,7 +2423,6 @@ static int qcom_nand_host_setup(struct 
>> qcom_nand_host *host)
>> 
>>  	mtd_set_ooblayout(mtd, &qcom_nand_ooblayout_ops);
>> 
>> -	cwperpage = mtd->writesize / ecc->size;
>>  	nandc->max_cwperpage = max_t(unsigned int, nandc->max_cwperpage,
>>  				     cwperpage);
>> 
>> @@ -2419,12 +2438,6 @@ static int qcom_nand_host_setup(struct 
>> qcom_nand_host *host)
>>  	 * for 8 bit ECC
>>  	 */
>>  	host->cw_size = host->cw_data + ecc->bytes;
>> -
>> -	if (ecc->bytes * (mtd->writesize / ecc->size) > mtd->oobsize) {
>> -		dev_err(nandc->dev, "ecc data doesn't fit in OOB area\n");
>> -		return -EINVAL;
>> -	}
>> -
>>  	bad_block_byte = mtd->writesize - host->cw_size * (cwperpage - 1) + 
>> 1;
>> 
>>  	host->cfg0 = (cwperpage - 1) << CW_PER_PAGE