mbox series

[v3,0/2] mtd: rawnand: vf610_nfc: fix reads on chips with more than 64 bytes of OOB

Message ID 20260901073907.2443698-1-mehmet.fide@gmail.com
Headers show
Series mtd: rawnand: vf610_nfc: fix reads on chips with more than 64 bytes of OOB | expand

Message

Mehmet Fide Sept. 1, 2026, 7:39 a.m. UTC 
From: Mehmet Fide <mehmet.fide@screeningeagle.com>

The driver only implements the 64-byte OOB format the controller
transfers, so chips with a larger OOB (the Colibri VF61's MX30LF4G28AC
has 112 bytes) stopped working when nanddev_init() began restoring
mtd->oobsize after ->attach_chip(): the parity moved and every
ECC-protected read failed, including the BBT and everything UBI needs.

The detected OOB size stays, the driver gets its own mtd_ooblayout_ops
computed on the first 64 OOB bytes, and the data paths keep
transferring exactly those 64 spare bytes, so the on-flash format stays
identical to U-Boot and to the kernels that clamped.

v3 addresses Miquel's review and the confirmed finding from the Sashiko
report: ECC page reads now fill the tail of oob_poi with 0xff, since
mtd->oobsize bytes of it may reach userspace while only 64 are
transferred. Retested on both boards. Colibri VF61 (112 bytes of OOB):
BBT found and read clean, UBI attaches and the UBIFS root mounts, 8 MiB
write/read-back intact, and OOB reads return 0xff in the 48 bytes past
the transferred area on every page sampled (6000+ pages, written and
erased). Colibri VF50 (64-byte chip, tail fill is a no-op): layout
unchanged, corrected counter stays zero across a full dump.

v2: https://lore.kernel.org/linux-mtd/20260828085337.3916199-1-mehmet.fide@gmail.com/
v1: https://lore.kernel.org/linux-mtd/20260818114208.2780311-1-mehmet.fide@gmail.com/


Mehmet Fide (2):
  mtd: rawnand: vf610_nfc: fix reads on chips with more than 64 bytes of
    OOB
  mtd: rawnand: vf610_nfc: fix false bitflips on reads of erased pages

 drivers/mtd/nand/raw/vf610_nfc.c | 85 ++++++++++++++++++++++++++------
 1 file changed, 70 insertions(+), 15 deletions(-)

Comments

Miquel Raynal Sept. 4, 2026, 5:38 p.m. UTC | #1 
On Tue, 01 Sep 2026 09:39:05 +0200, Mehmet Fide wrote:
> From: Mehmet Fide <mehmet.fide@screeningeagle.com>
> 
> The driver only implements the 64-byte OOB format the controller
> transfers, so chips with a larger OOB (the Colibri VF61's MX30LF4G28AC
> has 112 bytes) stopped working when nanddev_init() began restoring
> mtd->oobsize after ->attach_chip(): the parity moved and every
> ECC-protected read failed, including the BBT and everything UBI needs.
> 
> [...]

Applied to mtd/fixes, thanks!

[1/2] mtd: rawnand: vf610_nfc: fix reads on chips with more than 64 bytes of OOB
      commit: 1400e88023f4551032df6a3397a376dcdf67f3d4
[2/2] mtd: rawnand: vf610_nfc: fix false bitflips on reads of erased pages
      commit: 8d40b1261e2c967173b88c2f5d1c97cfeed1a2e5

Patche(s) should be available on mtd/linux.git and will be
part of the next PR (provided that no robot complains by then).

Kind regards,
Miquèl