[BPI-R3] Identify / Verify NOR, NAND, eMMC & SD Devices

mikew · February 25, 2023, 3:01pm

Hi,

In order to restore the installation in NOR I messed up when trying to update OpenWRT, I want to verify which the path to each device. I want to mount NOR when I booted from NAND and attempt to restore NOR by copying over the “backup” installation. So far I found in the wiki but could not find more details:

NOR: ???
NAND: /dev/mtdblock0
eMMC: /dev/mmcblk0boot0
eMMC Boot Partition: /dev/mmcblk0

https://wiki.banana-pi.org/Getting_Started_with_BPI-R3

Thanks Mike

frank-w · February 25, 2023, 3:06pm

Nand/nor are connected via same spi-bus like sd/emmc are accessable via 1 mmc-controller. I have a uboot-way to determine which device is connected each and load the right ftd overlay if this helps.

github.com

frank-w/u-boot/blob/2023-01-bpi/uEnv_r3.txt#L36


#console=${console}


useusb=if usb info; then usb reset;else usb start;fi; setenv device usb;setenv partition 0:1;ls ${device} ${partition}


useinitrd=setenv root "/dev/ram0 rw";setenv bootopts "${bootopts} initrd=${rdaddr},20M";fatload ${device} ${partition} ${rdaddr} ${initrd};
checkrd=if printenv initrd;then run useinitrd;fi


newboot=run checkrd; run setbootconf; run buildargs;if printenv fit;then setenv kernel ${fit};if run loadkernel; then bootm ${kaddr}${bootconf}; fi;fi;


checkmmc=if mmc partconf 0; then echo "emmc available";mmcdev="emmc";else echo "sd available";mmcdev="sd";fi
checkspi=if sf probe; then echo "NOR available";spidev="nor";else echo "NAND available";spidev="nand";fi


setbootconf=run checkmmc checkspi;echo $mmcdev $spidev;setenv bootconf "#conf-$mmcdev-$spidev"


wrspimnand=if printenv bl2file;then if fatload $device $partition $loadaddr $bl2file;then mtd erase spi-nand0 0x0 0x100000;mtd write spi-nand0 $loadaddr 0x0 0x100000;else echo "loading bl2 $bl2file failed";fi;else echo "bl2file not defined";fi;if printenv fipfile;then if fatload $device $partition $loadaddr $fipfile;then mtd erase spi-nand0 0x380000 0x200000;mtd write spi-nand0 $loadaddr 0x380000 0x200000;else echo "loading fip $fipfile failed";fi;else echo "fipfile not defined";fi
wremmc=if printenv bl2file;then mmc partconf 0 1 1 1;if fatload $device $partition $loadaddr $bl2file;then mmc erase 0x0 0x400;mmc write ${loadaddr} 0x0 0x400;else echo "loading bl2 $bl2file failed";fi;else echo "bl2file not defined";fi;mmc partconf 0 1 1 0;if printenv fipfile;then if fatload $device $partition $loadaddr $fipfile;then mmc erase 0x3400 0x1000;mmc write ${loadaddr} 0x3400 0x1000;else echo "loading fip $fipfile failed";fi;else echo "fipfile not defined";fi


bootdelay=0
bootcmd=setenv bootdelay 3; run loadenv;bootmenu;


##untested

you cannot mount nor when booted from nand…either nand or nor is accessable on spi (same for sd/emmc) and they will get same device-number if the fdt is loaded correctly

ericwoud · February 25, 2023, 3:18pm

Use the spi-tools package? Did you check dmesg ?

lsblk and check the size? Nand and nor are different size I believe.

mikew · February 25, 2023, 7:32pm

lsblk isn’t available on the vanilla OpenWRT Snapshoot for BPI-R3 it seems. I have seen the spi-tools but couldn’t find anything specific detailing out it capabilities. Just have seen the spi-tools being used once in the plethora of posts, wikis and tutorials out there. Going to check the spi-tools out if I fail to utilize @frank-w tremendous supporting comments.

dangowrt · February 26, 2023, 2:44am

Also OpenWrt uses probing within U-Boot to communicate the position of switch C and D to OpenWrt/Linux (by selecting the appropriate device tree overlay).

While in theory it would be possible to have both, NAND and NOR connected in a usable way at the same time on the same bus (using SPI chip-select lines), this is not how it works on the BPi-R3.

If switch C is in position 1 for NAND, you won’t be able to access NOR at all.

If switch C is in position 0 for NOR, you won’t be able to access NAND at all.

This is because both devices use the same chip-select line, and sadly instead of having two chip-select lines (CS0 and CS1) and using the dip switch C to swap CS0 and CS1, the dip switch directly selects which of the two is connected to CS0 (and there is no CS1 at all). So this was a decision of the board designers, supposedly to not waste an additional SoC pin for CS1 and in order not to need another multiplexer IC for swapping CS0 and CS1.

A similar situation applies for eMMC vs. microSD, also here you can only use one of them at the time, this time a multiplexer IC is used, but it switches the first 4 data lines to be connected either to microSD or to the eMMC.

To also answer your question, applying the restrictions explained above, the Linux device names of each storage chip are:

eMMC /dev/mmcblk0, /dev/mmcblk0p*, /dev/mmcblk0boot0, /dev/mmcblk0boo1 and /dev/mmcblk0rpmb

microSD /dev/mmcblk0, /dev/mmcblk0p*

NOR /dev/mtd*, /dev/mtdblock*

NAND /dev/mtd*, /dev/mtdblock*, /dev/ubi?_?, /dev/ubiblock?_?