Re: ECC nand 74 code error in barebox 2018.04.10

[Mihaita Ivascu <mihaita_ivascu@trimble.com>]

Hello,

    Yes both linux and barebox drivers use bbt. In fact the message
above related to ECC error in BBT I see them as well in dmesg when
loading kernel:

>nand_bbt: ECC error in BBT at 0x00001ffe0005
>nand_bbt: ECC error in BBT at 0x00001ffc0005


  I have disabled hardware ECC and enabled Software BCH ECC in barebox
menuconfig with the intention to have the same ECC as in kernel where
I have only software ECC option and now I have the following situation
in barebox:

Board: Phytec phyCORE-i.MX6 Ultra Lite SOM
detected i.MX6 UltraLite revision 1.2
mdio_bus: miibus0: probed
eth0: got preset MAC address: 50:2d:f4:14:96:09
nand: ONFI flash detected
nand: NAND device: Manufacturer ID: 0x2c, Chip ID: 0xdc (Micron
MT29F4G08ABADAH4), 512MiB, page size: 2048, OOB size: 64
nand: Invalid NAND_ECC_MODE 2
BUG: failure at
~/Projects/barebox/2018.04.0-phy1-r7.0/git/drivers/mtd/nand/nand_base.c:3699/nand_scan_tail()!
BUG!
[<9fd71451>] (unwind_backtrace+0x1/0x58) from [<9fd00d9d>] (panic+0x1d/0x34)
[<9fd00d9d>] (panic+0x1d/0x34) from [<9fd1272d>] (nand_scan_tail+0x99/0x3dc)
[<9fd1272d>] (nand_scan_tail+0x99/0x3dc) from [<9fd15cf9>]
(mxs_nand_probe+0x40d/0x43c)
[<9fd15cf9>] (mxs_nand_probe+0x40d/0x43c) from [<9fd0c595>]
(device_probe+0x21/0x68)
[<9fd0c595>] (device_probe+0x21/0x68) from [<9fd0c601>] (match.part.0+0x25/0x2c)
[<9fd0c601>] (match.part.0+0x25/0x2c) from [<9fd0c7df>]
(register_driver+0x5b/0x78)
[<9fd0c7df>] (register_driver+0x5b/0x78) from [<9fd00acb>]
(start_barebox+0x1b/0x98)
[<9fd00acb>] (start_barebox+0x1b/0x98) from [<9fd6faa1>]
(barebox_non_pbl_start+0xd5/0xf8)
[<9fd6faa1>] (barebox_non_pbl_start+0xd5/0xf8) from [<9fd00005>]
(__bare_init_start+0x1/0xc)
### ERROR ### Please RESET the board ###

   I cannot recover it now.

Thanks,
      Mihaita


On Thu, Apr 4, 2019 at 9:51 PM Sascha Hauer <s.hauer@pengutronix.de> wrote:
>
> On Wed, Apr 03, 2019 at 05:31:09PM +0200, Mihaita Ivascu wrote:
> > Hello,
> >
> >    I have the following situation:
> >
> >    I use imx6ul platforms.
> >    kernel 4.9.11 from NXP
> >    barebox 2018.04.10
> >
> >    I create and update an FIT image(*.itb) using
> > ubiformat/ubiattach/ubimkvol/ubiupdatevol commands
> >
> >    I also do this operation from Linux once the kernel was booted.
> >    Everything worked for a while but at some point the FITIMAGE size
> > increased over 50 MB and I adapted the nand table in device tree
> > sources.
> >
> >   Now the image update from barebox is still working, I update, the
> > kernel loads and then at the next reboot the target boots correctly.
> >   But if do the the FIt image update from linux, the update is
> > finished without warnings but at the next reboot, the barebox will
> > complained as listed in the attachment and the target won't boot from
> > nand anymore. Then i have to do the update again but only from
> > barebox.
> >
> >   nand partition table is the same in kernel and barebox dts.
> >   But for images larger than 50 MB somehow if I update from linux, the
> > barebox will complain about ECC code 74.
> >   I have enabled the other ECC options(software with/without BCH,
> > hardware) both in barebox and linux but no change.
> >   I don't know if/how should I play with the PEB settings.
> >
> >  Does anybody know how could I get around this issue  or how to
> > continue with my investigation?
> >
> > Thanks in advance,
> >           Mihaita Ivascu
>
> > barebox 2018.04.0 #2 Wed Apr 3 07:55:40 PDT 2019
> >
> >
> > Board: Phytec phyCORE-i.MX6 Ultra Lite SOM
> > detected i.MX6 UltraLite revision 1.2
> > mdio_bus: miibus0: probed
> > eth0: got preset MAC address: 50:2d:f4:14:96:09
> > nand: ONFI flash detected
> > nand: NAND device: Manufacturer ID: 0x2c, Chip ID: 0xdc (Micron MT29F4G08ABADAH4), 512MiB, page size: 2048, OOB size: 64
> > Bad block table found at page 262080, version 0x01
> > Bad block table found at page 262016, version 0x01
> > nand_bbt: ECC error in BBT at 0x00001ffe0005
> > nand_bbt: ECC error in BBT at 0x00001ffc0005
>
> Things go wrong here already. Could it be that you use a flash based bbt
> in barebox but not in Linux?
>
> Generally have you made sure that you can write a page in barebox and
> read it in Linux, then afterwards the other way round?
>
> I have no idea what the Freescale Kernel does. It might have some
> options to use other ECC modes or something like that. Looking at the
> device tree nodes might give a clue.
>
> Sascha
>
> --
> Pengutronix e.K.                           |                             |
> Industrial Linux Solutions                 | http://www.pengutronix.de/  |
> Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
> Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox