On Wed, 30 Jan 2019 13:01:44 +0000 Schrempf Frieder <[email protected]> wrote:
> From: Frieder Schrempf <[email protected]> > > The information about where the manufacturer puts the bad block > markers inside the bad block and in the OOB data is stored in > different places. Let's move this information to nand_chip.options > and nand_chip.badblockpos. > > As this chip-specific information is not directly related to the > bad block table (BBT), we also rename the flags to NAND_BBM_*. > > Signed-off-by: Frieder Schrempf <[email protected]> > --- > drivers/mtd/nand/onenand/onenand_base.c | 2 +- > drivers/mtd/nand/onenand/onenand_bbt.c | 2 +- > drivers/mtd/nand/raw/nand_amd.c | 2 +- > drivers/mtd/nand/raw/nand_base.c | 12 ++++++------ > drivers/mtd/nand/raw/nand_bbt.c | 4 ++-- > drivers/mtd/nand/raw/nand_esmt.c | 2 +- > drivers/mtd/nand/raw/nand_hynix.c | 4 ++-- > drivers/mtd/nand/raw/nand_macronix.c | 2 +- > drivers/mtd/nand/raw/nand_micron.c | 2 +- > drivers/mtd/nand/raw/nand_samsung.c | 4 ++-- > drivers/mtd/nand/raw/nand_toshiba.c | 2 +- > drivers/mtd/nand/raw/sh_flctl.c | 4 ++-- > include/linux/mtd/bbm.h | 14 +------------- > include/linux/mtd/rawnand.h | 15 +++++++++++++++ Might be better to split that in 2 patches: - update onenand logic - update rawnand logic > 14 files changed, 37 insertions(+), 34 deletions(-) > > diff --git a/drivers/mtd/nand/onenand/onenand_base.c > b/drivers/mtd/nand/onenand/onenand_base.c > index 4ca4b194e7d7..d6701b8f031f 100644 > --- a/drivers/mtd/nand/onenand/onenand_base.c > +++ b/drivers/mtd/nand/onenand/onenand_base.c > @@ -2458,7 +2458,7 @@ static int onenand_default_block_markbad(struct > mtd_info *mtd, loff_t ofs) > bbm->bbt[block >> 2] |= 0x01 << ((block & 0x03) << 1); > > /* We write two bytes, so we don't have to mess with 16-bit access */ > - ofs += mtd->oobsize + (bbm->badblockpos & ~0x01); > + ofs += mtd->oobsize + (this->badblockpos & ~0x01); Looks like the ->badblockpos field does not exist in struct onenand_chip, which means you didn't compile test this part ;-). > /* FIXME : What to do when marking SLC block in partition > * with MLC erasesize? For now, it is not advisable to > * create partitions containing both SLC and MLC regions. > diff --git a/drivers/mtd/nand/onenand/onenand_bbt.c > b/drivers/mtd/nand/onenand/onenand_bbt.c > index dde20487937d..880b0abd36c8 100644 > --- a/drivers/mtd/nand/onenand/onenand_bbt.c > +++ b/drivers/mtd/nand/onenand/onenand_bbt.c > @@ -191,7 +191,7 @@ static int onenand_scan_bbt(struct mtd_info *mtd, struct > nand_bbt_descr *bd) > return -ENOMEM; > > /* Set the bad block position */ > - bbm->badblockpos = ONENAND_BADBLOCK_POS; > + this->badblockpos = NAND_BBM_POS_ONENAND; This should be done in onenand_base.c not onenand_bbt.c. > > /* Set erase shift */ > bbm->bbt_erase_shift = this->erase_shift; ... > diff --git a/drivers/mtd/nand/raw/nand_bbt.c b/drivers/mtd/nand/raw/nand_bbt.c > index 1b722fe9213c..862eaa3a0552 100644 > --- a/drivers/mtd/nand/raw/nand_bbt.c > +++ b/drivers/mtd/nand/raw/nand_bbt.c > @@ -466,7 +466,7 @@ static int create_bbt(struct nand_chip *this, uint8_t > *buf, > > pr_info("Scanning device for bad blocks\n"); > > - if (bd->options & NAND_BBT_SCAN2NDPAGE) > + if (this->options & NAND_BBM_SECONDPAGE) > numpages = 2; > else > numpages = 1; > @@ -487,7 +487,7 @@ static int create_bbt(struct nand_chip *this, uint8_t > *buf, > from = (loff_t)startblock << this->bbt_erase_shift; > } > > - if (this->bbt_options & NAND_BBT_SCANLASTPAGE) > + if (this->bbt_options & NAND_BBM_LASTPAGE) You're not checking the right field here: if (this->options & NAND_BBM_LASTPAGE) > from += mtd->erasesize - (mtd->writesize * numpages); > > for (i = startblock; i < numblocks; i++) {
