| Message ID | 1335380771-27805-7-git-send-email-mikedunn@newsguy.com |
|---|---|
| State | Accepted |
| Commit | e2788c98b98269a3131bffd2b57599280d7abd73 |
| Headers | show |
On Wed, Apr 25, 2012 at 12:06 PM, Mike Dunn <mikedunn@newsguy.com> wrote: > This patch adds sanity checks that ensure that drivers for controllers with > hardware ECC set the 'strength' element in struct nand_ecc_ctrl. Also stylistic > changes to the line that calculates strength for software ECC. > > This v2 simplifies the check. Thanks Brian! [1] Thank you for going through the review process! This series provides a good general fix for something that's been kind of masked in hardware like mine (which has its own threshold for suppressing bitflip messages). > [1] http://lists.infradead.org/pipermail/linux-mtd/2012-April/040890.html > > Signed-off-by: Mike Dunn <mikedunn@newsguy.com> If we're looking for Ack's to rebase with: Acked-by: Brian Norris <computersforpeace@gmail.com>
diff --git a/drivers/mtd/nand/nand_base.c b/drivers/mtd/nand/nand_base.c index 3137abd..dac0afa 100644 --- a/drivers/mtd/nand/nand_base.c +++ b/drivers/mtd/nand/nand_base.c @@ -3345,8 +3345,13 @@ int nand_scan_tail(struct mtd_info *mtd) if (!chip->ecc.write_oob) chip->ecc.write_oob = nand_write_oob_syndrome; - if (mtd->writesize >= chip->ecc.size) + if (mtd->writesize >= chip->ecc.size) { + if (!chip->ecc.strength) { + pr_warn("Driver must set ecc.strength when using hardware ECC\n"); + BUG(); + } break; + } pr_warn("%d byte HW ECC not possible on " "%d byte page size, fallback to SW ECC\n", chip->ecc.size, mtd->writesize); @@ -3401,7 +3406,7 @@ int nand_scan_tail(struct mtd_info *mtd) BUG(); } chip->ecc.strength = - chip->ecc.bytes*8 / fls(8*chip->ecc.size); + chip->ecc.bytes * 8 / fls(8 * chip->ecc.size); break; case NAND_ECC_NONE:
This patch adds sanity checks that ensure that drivers for controllers with hardware ECC set the 'strength' element in struct nand_ecc_ctrl. Also stylistic changes to the line that calculates strength for software ECC. This v2 simplifies the check. Thanks Brian! [1] [1] http://lists.infradead.org/pipermail/linux-mtd/2012-April/040890.html Signed-off-by: Mike Dunn <mikedunn@newsguy.com> --- drivers/mtd/nand/nand_base.c | 7 ++++++- 1 files changed, 6 insertions(+), 1 deletions(-)