Regression in marvell-nand

[Daniel Mack <daniel@xxxxxxxxxx>]

Hi Boris and all,

I'm seeing a severe regression with the NAND driver on a PXA3xx platform
caused by b25251414f6e00 ("mtd: rawnand: marvell: Stop implementing
->select_chip()"). Since that commit, I'm seeing lots of messages like
this during boot:

[    5.480730] Bad eraseblock 702 at 0x0000057c0000
[    5.485936] Bad eraseblock 703 at 0x0000057e0000
[    5.491354] Bad eraseblock 704 at 0x000005800000
[    5.496544] Bad eraseblock 705 at 0x000005820000

Eventually followed by the UBI layer acting up:

[    9.334403] ubi0 error: ubi_io_read: error -74 (ECC error) while
reading 64 bytes from PEB 5:0, read 64 bytes
[    9.344361] CPU: 0 PID: 1 Comm: swapper Not tainted 5.0.2+ #551
[    9.350285] Hardware name: Marvell PXA3xx (Device Tree Support)
[    9.356206] [<c0107c14>] (unwind_backtrace) from [<c0105a5c>]
(show_stack+0x10/0x14)
[    9.363981] [<c0105a5c>] (show_stack) from [<c04322c8>]
(ubi_io_read+0x1dc/0x2c8)
[    9.371512] [<c04322c8>] (ubi_io_read) from [<c04325e8>]
(ubi_io_read_ec_hdr+0x6c/0x220)
[    9.379633] [<c04325e8>] (ubi_io_read_ec_hdr) from [<c0437568>]
(ubi_attach+0x18c/0x137c)
[    9.387839] [<c0437568>] (ubi_attach) from [<c042c71c>]
(ubi_attach_mtd_dev+0x654/0xad0)
[    9.395891] [<c042c71c>] (ubi_attach_mtd_dev) from [<c09153b8>]
(ubi_init+0x148/0x200)
[    9.403834] [<c09153b8>] (ubi_init) from [<c0102508>]
(do_one_initcall+0x74/0x178)
[    9.411456] [<c0102508>] (do_one_initcall) from [<c0900d8c>]
(kernel_init_freeable+0x108/0x1d4)
[    9.420177] [<c0900d8c>] (kernel_init_freeable) from [<c0665da0>]
(kernel_init+0x8/0x110)
[    9.428385] [<c0665da0>] (kernel_init) from [<c01010e0>]
(ret_from_fork+0x14/0x34)
[    9.435900] Exception stack(0xc603dfb0 to 0xc603dff8)


The commit before is working just fine. I haven't dug any further than
bisecting yet however.

Any idea?


Thanks,
Daniel

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/