diff options
author | Sascha Hauer <s.hauer@pengutronix.de> | 2018-01-17 12:32:09 +0100 |
---|---|---|
committer | Boris Brezillon <boris.brezillon@bootlin.com> | 2018-02-12 23:35:34 +0100 |
commit | 67b87f66cae0bca0dd6602739383381c10b0c9c4 (patch) | |
tree | 652d1ddfa1d00303b8a4bbf03899ed7e2249e564 /drivers/hid/hid-tmff.c | |
parent | 1af0b3147448954c50e555496b80d94a0800a5e4 (diff) | |
download | kernel_replicant_linux-67b87f66cae0bca0dd6602739383381c10b0c9c4.tar.gz kernel_replicant_linux-67b87f66cae0bca0dd6602739383381c10b0c9c4.tar.bz2 kernel_replicant_linux-67b87f66cae0bca0dd6602739383381c10b0c9c4.zip |
mtd: nand: mxc: Fix failed/corrected values for v2/v3 controllers
Currently nand_read_page_hwecc() from nand_base calls
mxc_nand_correct_data_v2_v3() for each subpage, but in this function we
return the corrected/failed results for the whole page instead
of a single subpage. On a 2k page size Nand this leads to results which
are 4 times too high.
The whole ecc.calculate/ecc.correct mechanism used by
nand_read_page_hwecc() is not suitable for devices which correct the
data in hardware, so fix this by using a driver specific read_page
function which does the right thing. Also add read_page_raw and read_oob
For proper raw and oob read support.
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Boris Brezillon <boris.brezillon@bootlin.com>
Diffstat (limited to 'drivers/hid/hid-tmff.c')
0 files changed, 0 insertions, 0 deletions