On 20.08.20 13:12, Pali Rohár wrote:
On Thursday 20 August 2020 10:51:28 Pali Rohár wrote:
On Thursday 20 August 2020 10:17:55 Stefan Roese wrote:
On 20.08.20 09:40, Pali Rohár wrote:
On Thursday 20 August 2020 07:02:18 Stefan Roese wrote:
Can't you just use "mtdparts=" kernel cmdline parameter instead to pass
the MTD layout to the kernel?
Maybe it is possible too, I have not tried it.
I thought that more common is to update DTS file by uboot when loading
kernel as it is already done e.g. for ethernet MAC address on
Espressobin.
Updating the MAC address is very common, yes. But passing the
MTD partition layout via the mtdparts= cmdline is also very
common and used very frequently. From my experience, its the
defacto recommended method to pass this info and also easier than
generating these DTS lines via some code.
Please take a look at the mtdparts U-Boot command and its usage
in other targets.
Also I see that uboot has function fdt_fixup_mtdparts() via
CONFIG_FDT_FIXUP_PARTITIONS option which do this, but uses uboot MTD
code which IIRC cannot initialize SPI NOR.
Why not?
I had an impression that U-Boot does not support.
SPI NOR is also integrated into the U-Boot MTD world as
well. Please see this GARDENA MT7688 target, which uses SPI NOR &
SPI NAND:
Ok, I'm going to look at it.
With following config
CONFIG_MTD=y
CONFIG_CMD_MTD=y
CONFIG_CMD_MTDPARTS=y
CONFIG_SPI_FLASH_MTD=y
CONFIG_MTDIDS_DEFAULT="nor0=nor0"
I get following result:
=> mtd list
List of MTD devices:
No MTD device found
=> mtdparts
Device nor0 not found!
=> sf probe
SF: Detected w25q32dw with page size 256 Bytes, erase size 4 KiB, total 4 MiB
=> mtdparts
=> mtd list
List of MTD devices:
* nor0
- type: NOR flash
- block size: 0x1000 bytes
- min I/O: 0x1 bytes
- 0x000000000000-0x000000400000 : "nor0"
So main problem is that MTD does not work until I call 'sf probe' in
uboot command line.
Then a dependency seems to be missing here. On my MT7688 board I get
this:
U-Boot 2020.07-rc5-00075-g0b7d95531c-dirty (Jun 30 2020 - 09:51:39 +0200)
CPU: MediaTek MT7688A ver:1 eco:2
Boot: DDR2, SPI-NOR 3-Byte Addr, CPU clock from XTAL
Clock: CPU: 580MHz, Bus: 193MHz, XTAL: 40MHz
Model: GARDENA smart Gateway (MT7688)
DRAM: 128 MiB
WDT: Started with servicing (60s timeout)
Loading Environment from SPI Flash... SF: Detected XM25QH64A with page
size 256 Bytes, erase size 4 KiB, total 8 MiB
OK
F-Data:factory-data version 1 detected
Net: eth0: eth@10110000
Hit any key to stop autoboot: 0
=> dm tree
Class Index Probed Driver Name
-----------------------------------------------------------
root 0 [ + ] root_driver root_driver
...
spi 0 [ + ] mt7621_spi | |-- spi@b00
spi_flash 0 [ + ] spi_flash_std | | |-- spi-flash@0
mtd 0 [ ] spi_nand | | `-- spi-nand@1
...
=> mtd list
List of MTD devices:
* nor0
- type: NOR flash
- block size: 0x1000 bytes
- min I/O: 0x1 bytes
- 0x000000000000-0x000000800000 : "nor0"
- 0x000000000000-0x0000000a0000 : "uboot"
- 0x0000000a0000-0x0000000b0000 : "uboot_env0"
- 0x0000000b0000-0x0000000c0000 : "uboot_env1"
- 0x0000000c0000-0x0000000d0000 : "factory"
- 0x0000000d0000-0x000000800000 : "unused"
* spi-nand0
- device: spi-nand@1
- parent: spi@b00
- driver: spi_nand
- type: NAND flash
- block size: 0x20000 bytes
- min I/O: 0x800 bytes
- OOB size: 128 bytes
- OOB available: 63 bytes
- 0x000000000000-0x000008000000 : "spi-nand0"
- 0x000000000000-0x000008000000 : "nand"
=> dm tree
Class Index Probed Driver Name
-----------------------------------------------------------
root 0 [ + ] root_driver root_driver
...
spi 0 [ + ] mt7621_spi | |-- spi@b00
spi_flash 0 [ + ] spi_flash_std | | |-- spi-flash@0
mtd 0 [ + ] spi_nand | | `-- spi-nand@1
...
As you see, spi_nand is automatically probed by "mtd list". SPI NOR
is most likely already probed since its used for ENV storage.
It would be great, if you could check, if and where such a dependency
is missing in your case.
Thanks,
Stefan
