devicetree.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Ivan Djelic <ivan.djelic@parrot.com>
To: "Philip, Avinash" <avinashphilip@ti.com>
Cc: "dwmw2@infradead.org" <dwmw2@infradead.org>,
	"artem.bityutskiy@linux.intel.com"
	<artem.bityutskiy@linux.intel.com>,
	"tony@atomide.com" <tony@atomide.com>,
	"afzal@ti.com" <afzal@ti.com>,
	"linux-mtd@lists.infradead.org" <linux-mtd@lists.infradead.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-omap@vger.kernel.org" <linux-omap@vger.kernel.org>,
	"linux-arm-kernel@lists.infradead.org"
	<linux-arm-kernel@lists.infradead.org>,
	"linux-doc@vger.kernel.org" <linux-doc@vger.kernel.org>,
	"devicetree-discuss@lists.ozlabs.org"
	<devicetree-discuss@lists.ozlabs.org>
Subject: Re: [PATCH 4/4] mtd: nand: omap2: Add data correction support
Date: Wed, 3 Oct 2012 21:20:45 +0200	[thread overview]
Message-ID: <20121003192044.GB27502@parrot.com> (raw)
In-Reply-To: <1349274589-11389-5-git-send-email-avinashphilip@ti.com>

On Wed, Oct 03, 2012 at 03:29:49PM +0100, Philip, Avinash wrote:
> ELM module can be used for error correction of BCH 4 & 8 bit. Also
> support read & write page in one shot by adding custom read_page &
> write_page methods. This helps in optimizing code.
> 
> New structure member "is_elm_used" is added to know the status of
> whether the ELM module is used for error correction or not.
> 
> Note:
> ECC layout of BCH8 uses 14 bytes for 512 byte of data to make compatible
> with RBL ECC layout, even though the requirement was only 13 byte. This
> results a common ecc layout across RBL, U-boot & Linux.
> 

See a few comments below,

(...)
> +static int omap_elm_correct_data(struct mtd_info *mtd, u_char *dat,
> +                               u_char *read_ecc, u_char *calc_ecc)
> +{
> +       struct omap_nand_info *info = container_of(mtd, struct omap_nand_info,
> +                       mtd);
> +       int eccsteps = info->nand.ecc.steps;
> +       int i , j, stat = 0;
> +       int eccsize, eccflag, size;
> +       struct elm_errorvec err_vec[ERROR_VECTOR_MAX];
> +       u_char *ecc_vec = calc_ecc;
> +       enum bch_ecc type;
> +       bool is_error_reported = false;
> +
> +       /* initialize elm error vector to zero */
> +       memset(err_vec, 0, sizeof(err_vec));
> +       if (info->nand.ecc.strength == BCH8_MAX_ERROR) {
> +               size = BCH8_SIZE;
> +               eccsize = BCH8_ECC_OOB_BYTES;
> +               type = BCH8_ECC;
> +       } else {
> +               size = BCH4_SIZE;
> +               eccsize = BCH4_SIZE;
> +               type = BCH4_ECC;
> +       }
> +
> +       for (i = 0; i < eccsteps ; i++) {
> +               eccflag = 0;    /* initialize eccflag */
> +
> +               for (j = 0; (j < eccsize); j++) {
> +                       if (read_ecc[j] != 0xFF) {
> +                               eccflag = 1;    /* data area is flashed */

Just a reminder: this way of checking if a page has been programmed is not robust to bitflips,
so you may get into trouble with UBIFS on a fairly recent device.

(...)
> @@ -1039,14 +1180,45 @@ static void omap3_enable_hwecc_bch(struct mtd_info *mtd, int mode)
> 
>         nerrors = info->nand.ecc.strength;
>         dev_width = (chip->options & NAND_BUSWIDTH_16) ? 1 : 0;
> +#ifdef CONFIG_MTD_NAND_OMAP_BCH
> +       if (info->is_elm_used) {
> +               /*
> +                * Program GPMC to perform correction on (steps * 512) byte
> +                * sector at a time.
> +                */
> +               gpmc_enable_hwecc_bch(info->gpmc_cs, mode, dev_width,
> +                               info->nand.ecc.steps, nerrors);
> +               return;
> +       }
> +#endif
>         /*
> -        * Program GPMC to perform correction on one 512-byte sector at a time.
> -        * Using 4 sectors at a time (i.e. ecc.size = 2048) is also possible and
> -        * gives a slight (5%) performance gain (but requires additional code).
> +        * Program GPMC to perform correction on one 512-byte sector at
> +        * a time.

Why removing the comment about 4-sector perf gain ? :-)

(...)
> @@ -1146,35 +1402,62 @@ static int omap3_init_bch(struct mtd_info *mtd, int ecc_opt)
>                 goto fail;
>         }
> 
> -       /* initialize GPMC BCH engine */
> -       ret = gpmc_init_hwecc_bch(info->gpmc_cs, 1, max_errors);
> -       if (ret)
> -               goto fail;
> -
> -       /* software bch library is only used to detect and locate errors */
> -       info->bch = init_bch(13, max_errors, 0x201b /* hw polynomial */);
> -       if (!info->bch)
> -               goto fail;
> +       info->nand.ecc.size = 512;
> +       info->nand.ecc.hwctl = omap3_enable_hwecc_bch;
> +       info->nand.ecc.mode = NAND_ECC_HW;
> +       info->nand.ecc.strength = hw_errors;
> 
> -       info->nand.ecc.size    = 512;
> -       info->nand.ecc.hwctl   = omap3_enable_hwecc_bch;
> -       info->nand.ecc.correct = omap3_correct_data_bch;
> -       info->nand.ecc.mode    = NAND_ECC_HW;
> +       if (info->is_elm_used && (mtd->writesize <= 4096)) {
> +               enum bch_ecc bch_type;
> 
> -       /*
> -        * The number of corrected errors in an ecc block that will trigger
> -        * block scrubbing defaults to the ecc strength (4 or 8).
> -        * Set mtd->bitflip_threshold here to define a custom threshold.
> -        */
> +               if (hw_errors == BCH8_MAX_ERROR) {
> +                       bch_type = BCH8_ECC;
> +                       info->nand.ecc.bytes = BCH8_SIZE;
> +               } else {
> +                       bch_type = BCH4_ECC;
> +                       info->nand.ecc.bytes = BCH4_SIZE;
> +               }
> 
> -       if (max_errors == 8) {
> -               info->nand.ecc.strength  = 8;
> -               info->nand.ecc.bytes     = 13;
> -               info->nand.ecc.calculate = omap3_calculate_ecc_bch8;
> +               info->nand.ecc.correct = omap_elm_correct_data;
> +               info->nand.ecc.calculate = omap3_calculate_ecc_bch;
> +               info->nand.ecc.read_page = omap_read_page_bch;
> +               info->nand.ecc.write_page = omap_write_page_bch;
> +               info->elm_dev = elm_request(bch_type);
> +               if (!info->elm_dev) {
> +                       pr_err("Request to elm module failed\n");
> +                       goto fail;
> +               }
>         } else {
> -               info->nand.ecc.strength  = 4;
> -               info->nand.ecc.bytes     = 7;
> -               info->nand.ecc.calculate = omap3_calculate_ecc_bch4;
> +
> +               /* initialize GPMC BCH engine */
> +               ret = gpmc_init_hwecc_bch(info->gpmc_cs, 1, max_errors);
> +               if (ret)
> +                       goto fail;
> +
> +               /*
> +                * software bch library is only used to detect and
> +                * locateerrors

s/locateerrors/locate errors/

BTW, did you check that your patch does not break the software BCH case (i.e. no ELM) ?

BR,
--
Ivan

  reply	other threads:[~2012-10-03 19:20 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-10-03 14:29 [PATCH 0/4] mtd: nand: OMAP: Add support to use ELM as error correction module Philip, Avinash
2012-10-03 14:29 ` [PATCH 1/4] mtd: nand: omap2: Update nerrors using ecc.strength Philip, Avinash
2012-10-15 18:56   ` Peter Korsgaard
2012-10-23 10:17     ` Philip, Avinash
2012-10-03 14:29 ` [PATCH 2/4] mtd: devices: elm: Add support for ELM error correction Philip, Avinash
2012-10-03 15:10   ` Peter Meerwald
2012-10-04  7:49     ` Philip, Avinash
2012-10-15 19:40   ` Peter Korsgaard
2012-10-23 10:17     ` Philip, Avinash
2012-10-03 14:29 ` [PATCH 3/4] ARM: OMAP2: gpmc: Add support for BCH ECC scheme Philip, Avinash
2012-10-03 18:54   ` Ivan Djelic
2012-10-04  8:03     ` Philip, Avinash
2012-10-04 12:04       ` Ivan Djelic
2012-10-15 18:48   ` Peter Korsgaard
2012-10-23 10:18     ` Philip, Avinash
2012-10-03 14:29 ` [PATCH 4/4] mtd: nand: omap2: Add data correction support Philip, Avinash
2012-10-03 19:20   ` Ivan Djelic [this message]
2012-10-04 10:22     ` Philip, Avinash
2012-10-05  8:51     ` Philip, Avinash
2012-10-05 14:23       ` Ivan Djelic
2012-10-09 12:36         ` Philip, Avinash
2012-10-10 17:08           ` Ivan Djelic
2012-10-11  5:27             ` Philip, Avinash

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20121003192044.GB27502@parrot.com \
    --to=ivan.djelic@parrot.com \
    --cc=afzal@ti.com \
    --cc=artem.bityutskiy@linux.intel.com \
    --cc=avinashphilip@ti.com \
    --cc=devicetree-discuss@lists.ozlabs.org \
    --cc=dwmw2@infradead.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mtd@lists.infradead.org \
    --cc=linux-omap@vger.kernel.org \
    --cc=tony@atomide.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).