| Message ID | 20180709210937.30150-1-boris.brezillon@bootlin.com |
|---|---|
| Headers | show |
| Series | mtd: rawnand: micron: Fix on-die ECC | expand |
On Mon, 9 Jul 2018 23:09:35 +0200 Boris Brezillon <boris.brezillon@bootlin.com> wrote: > Chris, Bean, > > Here are 2 patches for you to review/test. The first one is fixing > the layout definition, and unless I missed something it should be > correct. > > The second one is just my understanding of how byte 5 of READ_ID works > based on our experience with the 4bit/512 on-die ECC chip we have > worked on and the other datasheet I had a look at. > I'm not 100% sure this will work for all chips, but might work for the > 2 chips we support right now. I tested it on a MT29F2G08ABAEAH4 (4bit/512bytes on-die ECC) and it works as expected. When I do the SET_FEATURES(ECC_EN) the ECC enabled bit in READID byte 5 is set, and when I do SET_FEATURES(ECC_DIS), the bit is cleared. Now we need to make sure it works correctly on MT29F1G08ABAFAWP. > So please test and/or review it and let me know if this approach works. > > Regards, > > Boris > > Boris Brezillon (2): > mtd: rawnand: micron: Define the proper layout for 8bit/512bytes > on-die ECC > mtd: rawnand: micron: Fix on-die ECC detection logic > > drivers/mtd/nand/raw/nand_micron.c | 110 +++++++++++++++++++++++++------------ > 1 file changed, 75 insertions(+), 35 deletions(-) >
Hi Boris, On 10/07/18 20:11, Boris Brezillon wrote: > On Mon, 9 Jul 2018 23:09:35 +0200 > Boris Brezillon <boris.brezillon@bootlin.com> wrote: > >> Chris, Bean, >> >> Here are 2 patches for you to review/test. The first one is fixing >> the layout definition, and unless I missed something it should be >> correct. >> >> The second one is just my understanding of how byte 5 of READ_ID works >> based on our experience with the 4bit/512 on-die ECC chip we have >> worked on and the other datasheet I had a look at. >> I'm not 100% sure this will work for all chips, but might work for the >> 2 chips we support right now. > > I tested it on a MT29F2G08ABAEAH4 (4bit/512bytes on-die ECC) and it > works as expected. When I do the SET_FEATURES(ECC_EN) the ECC enabled > bit in READID byte 5 is set, and when I do SET_FEATURES(ECC_DIS), the > bit is cleared. > > Now we need to make sure it works correctly on MT29F1G08ABAFAWP. I'll test that as soon as I can get access to the hardware (probably Monday next week). > >> So please test and/or review it and let me know if this approach works. >> >> Regards, >> >> Boris >> >> Boris Brezillon (2): >> mtd: rawnand: micron: Define the proper layout for 8bit/512bytes >> on-die ECC >> mtd: rawnand: micron: Fix on-die ECC detection logic >> >> drivers/mtd/nand/raw/nand_micron.c | 110 +++++++++++++++++++++++++------------ >> 1 file changed, 75 insertions(+), 35 deletions(-) >> > >
On 10/07/18 09:09, Boris Brezillon wrote: > Chris, Bean, > > Here are 2 patches for you to review/test. The first one is fixing > the layout definition, and unless I missed something it should be > correct. > > The second one is just my understanding of how byte 5 of READ_ID works > based on our experience with the 4bit/512 on-die ECC chip we have > worked on and the other datasheet I had a look at. > I'm not 100% sure this will work for all chips, but might work for the > 2 chips we support right now. > So please test and/or review it and let me know if this approach works. > Hi Boris, I went to test this today and found that I could no-longer mount the Micron MT29F1G08ABAFAWP-ITE chip. However it doesn't appear to these changes but d73e5d29 ("mtd: rawnand: micron: Disable ECC earlier in the read path"). [root@linuxbox ~]# ubiformat -q -y /dev/mtd0 [root@linuxbox ~]# ubiattach -p /dev/mtd0 ubi0: attaching mtd0 ubi0: scanning is finished ubi0 error: ubi_read_volume_table: the layout volume was not found ubi0 error: ubi_attach_mtd_dev: failed to attach mtd0, error -22 ubiattach: error!: cannot attach "/dev/mtd0" error 22 (Invalid argument) [root@linuxbox ~]# ubimkvol /dev/ubi0 -N user -m libubi: error!: cannot get information about "/dev/ubi0" error 2 (No such file or directory) ubimkvol: error!: error while probing "/dev/ubi0" error 2 (No such file or directory) I'll try juggling some of the nand/next patches around to see if I can test these two without d73e5d29.
Chris, Bean, Here are 2 patches for you to review/test. The first one is fixing the layout definition, and unless I missed something it should be correct. The second one is just my understanding of how byte 5 of READ_ID works based on our experience with the 4bit/512 on-die ECC chip we have worked on and the other datasheet I had a look at. I'm not 100% sure this will work for all chips, but might work for the 2 chips we support right now. So please test and/or review it and let me know if this approach works. Regards, Boris Boris Brezillon (2): mtd: rawnand: micron: Define the proper layout for 8bit/512bytes on-die ECC mtd: rawnand: micron: Fix on-die ECC detection logic drivers/mtd/nand/raw/nand_micron.c | 110 +++++++++++++++++++++++++------------ 1 file changed, 75 insertions(+), 35 deletions(-)