All of lore.kernel.org
 help / color / mirror / Atom feed
From: Boris Brezillon <boris.brezillon@free-electrons.com>
To: Kyle Roeschley <kyle.roeschley@ni.com>
Cc: richard@nod.at, dwmw2@infradead.org, computersforpeace@gmail.com,
	beanhuo@micron.com, linux-mtd@lists.infradead.org,
	linux-kernel@vger.kernel.org, nathan.sullivan@ni.com,
	xander.huff@ni.com, peterpansjtu@gmail.com
Subject: Re: [PATCH v8 1/2] mtd: nand_bbt: Move BBT block selection logic out of write_bbt()
Date: Wed, 17 Aug 2016 21:59:19 +0200	[thread overview]
Message-ID: <20160817215919.45b8db96@bbrezillon> (raw)
In-Reply-To: <1471303322-2479-1-git-send-email-kyle.roeschley@ni.com>

On Mon, 15 Aug 2016 18:22:01 -0500
Kyle Roeschley <kyle.roeschley@ni.com> wrote:

> From: Boris Brezillon <boris.brezillon@free-electrons.com>
> 
> This clarifies the write_bbt() by removing the write label and
> clarifying the error/exit path.

Applied both.

Thanks,

Boris

> 
> Signed-off-by: Boris Brezillon <boris.brezillon@free-electrons.com>
> Tested-by: Kyle Roeschley <kyle.roeschley@ni.com>
> ---
> v8: Move the chip indexing change back into patch 2
> 
> v7: Move all code cleanup into first patch
>     Correct documentation of mark_bbt_block_bad
>     Make pr_warn messages consistent
>     Add Tested-bys
> 
> v6: Split functionality of write_bbt out into new functions
> 
> v5: De-duplicate bad block handling
> 
> v4: Don't ignore write protection while marking bad BBT blocks
>     Correctly call block_markbad
>     Minor cleanups
> 
> v3: Don't overload mtd->priv
>     Keep nand_erase_nand from erroring on protected BBT blocks
> 
> v2: Mark OOB area in each block as well as BBT
>     Avoid marking read-only, bad address, or known bad blocks as bad
> 
>  drivers/mtd/nand/nand_bbt.c | 110 +++++++++++++++++++++++++++++---------------
>  1 file changed, 74 insertions(+), 36 deletions(-)
> 
> diff --git a/drivers/mtd/nand/nand_bbt.c b/drivers/mtd/nand/nand_bbt.c
> index 2fbb523..9fbd664 100644
> --- a/drivers/mtd/nand/nand_bbt.c
> +++ b/drivers/mtd/nand/nand_bbt.c
> @@ -605,6 +605,69 @@ static void search_read_bbts(struct mtd_info *mtd, uint8_t *buf,
>  }
>  
>  /**
> + * get_bbt_block - Get the first valid eraseblock suitable to store a BBT
> + * @this: the NAND device
> + * @td: the BBT description
> + * @md: the mirror BBT descriptor
> + * @chip: the CHIP selector
> + *
> + * This functions returns a positive block number pointing a valid eraseblock
> + * suitable to store a BBT (i.e. in the range reserved for BBT), or -ENOSPC if
> + * all blocks are already used of marked bad. If td->pages[chip] was already
> + * pointing to a valid block we re-use it, otherwise we search for the next
> + * valid one.
> + */
> +static int get_bbt_block(struct nand_chip *this, struct nand_bbt_descr *td,
> +			 struct nand_bbt_descr *md, int chip)
> +{
> +	int startblock, dir, page, numblocks, i;
> +
> +	/*
> +	 * There was already a version of the table, reuse the page. This
> +	 * applies for absolute placement too, as we have the page number in
> +	 * td->pages.
> +	 */
> +	if (td->pages[chip] != -1)
> +		return td->pages[chip] >>
> +				(this->bbt_erase_shift - this->page_shift);
> +
> +	numblocks = (int)(this->chipsize >> this->bbt_erase_shift);
> +	if (!(td->options & NAND_BBT_PERCHIP))
> +		numblocks *= this->numchips;
> +
> +	/*
> +	 * Automatic placement of the bad block table. Search direction
> +	 * top -> down?
> +	 */
> +	if (td->options & NAND_BBT_LASTBLOCK) {
> +		startblock = numblocks * (chip + 1) - 1;
> +		dir = -1;
> +	} else {
> +		startblock = chip * numblocks;
> +		dir = 1;
> +	}
> +
> +	for (i = 0; i < td->maxblocks; i++) {
> +		int block = startblock + dir * i;
> +
> +		/* Check, if the block is bad */
> +		switch (bbt_get_entry(this, block)) {
> +		case BBT_BLOCK_WORN:
> +		case BBT_BLOCK_FACTORY_BAD:
> +			continue;
> +		}
> +
> +		page = block << (this->bbt_erase_shift - this->page_shift);
> +
> +		/* Check, if the block is used by the mirror table */
> +		if (!md || md->pages[chip] != page)
> +			return block;
> +	}
> +
> +	return -ENOSPC;
> +}
> +
> +/**
>   * write_bbt - [GENERIC] (Re)write the bad block table
>   * @mtd: MTD device structure
>   * @buf: temporary buffer
> @@ -621,7 +684,7 @@ static int write_bbt(struct mtd_info *mtd, uint8_t *buf,
>  	struct nand_chip *this = mtd_to_nand(mtd);
>  	struct erase_info einfo;
>  	int i, res, chip = 0;
> -	int bits, startblock, dir, page, offs, numblocks, sft, sftmsk;
> +	int bits, page, offs, numblocks, sft, sftmsk;
>  	int nrchips, pageoffs, ooboffs;
>  	uint8_t msk[4];
>  	uint8_t rcode = td->reserved_block_code;
> @@ -653,45 +716,20 @@ static int write_bbt(struct mtd_info *mtd, uint8_t *buf,
>  
>  	/* Loop through the chips */
>  	for (; chip < nrchips; chip++) {
> -		/*
> -		 * There was already a version of the table, reuse the page
> -		 * This applies for absolute placement too, as we have the
> -		 * page nr. in td->pages.
> -		 */
> -		if (td->pages[chip] != -1) {
> -			page = td->pages[chip];
> -			goto write;
> +		int block;
> +
> +		block = get_bbt_block(this, td, md, chip);
> +		if (block < 0) {
> +			pr_err("No space left to write bad block table\n");
> +			res = block;
> +			goto outerr;
>  		}
>  
>  		/*
> -		 * Automatic placement of the bad block table. Search direction
> -		 * top -> down?
> +		 * get_bbt_block() returns a block number, shift the value to
> +		 * get a page number.
>  		 */
> -		if (td->options & NAND_BBT_LASTBLOCK) {
> -			startblock = numblocks * (chip + 1) - 1;
> -			dir = -1;
> -		} else {
> -			startblock = chip * numblocks;
> -			dir = 1;
> -		}
> -
> -		for (i = 0; i < td->maxblocks; i++) {
> -			int block = startblock + dir * i;
> -			/* Check, if the block is bad */
> -			switch (bbt_get_entry(this, block)) {
> -			case BBT_BLOCK_WORN:
> -			case BBT_BLOCK_FACTORY_BAD:
> -				continue;
> -			}
> -			page = block <<
> -				(this->bbt_erase_shift - this->page_shift);
> -			/* Check, if the block is used by the mirror table */
> -			if (!md || md->pages[chip] != page)
> -				goto write;
> -		}
> -		pr_err("No space left to write bad block table\n");
> -		return -ENOSPC;
> -	write:
> +		page = block << (this->bbt_erase_shift - this->page_shift);
>  
>  		/* Set up shift count and masks for the flash table */
>  		bits = td->options & NAND_BBT_NRBITS_MSK;

      parent reply	other threads:[~2016-08-17 19:59 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-08-15 23:22 [PATCH v8 1/2] mtd: nand_bbt: Move BBT block selection logic out of write_bbt() Kyle Roeschley
2016-08-15 23:22 ` [PATCH v8 2/2] mtd: nand_bbt: scan for next free bbt block if writing bbt fails Kyle Roeschley
2016-08-17 19:59 ` Boris Brezillon [this message]

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=20160817215919.45b8db96@bbrezillon \
    --to=boris.brezillon@free-electrons.com \
    --cc=beanhuo@micron.com \
    --cc=computersforpeace@gmail.com \
    --cc=dwmw2@infradead.org \
    --cc=kyle.roeschley@ni.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mtd@lists.infradead.org \
    --cc=nathan.sullivan@ni.com \
    --cc=peterpansjtu@gmail.com \
    --cc=richard@nod.at \
    --cc=xander.huff@ni.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.