cmp18, Here is the DTS in pretty-print standard format for Linux. You could compile this and compare with your DTB to verify it's acurate.by bodhi - Rescue System
cergei, > any questions, can I > ask them in this thread? Sure. Any time you have question, post in this thread.by bodhi - Debian
> For a combined boot, I want the system > to boot from the hdd if there is a rootfs on the > hard drive, and if there is no rootfs on the hdd, > it should boot from the flash drive. Is this > possible? > > > fw_setenv bootdev ide > fw_setenv devices 'ide usb' > > > Will this command be correct? Yes, it is correct. However, if the Hby bodhi - Debian
BTW, this version is the latest for the Stora: U-Boot 2017.07-tld-1 (Sep 05 2017 - 00:38:05 -0700) Netgear Stora MS2110 I have not rebased this u-boot to 2025.10.by bodhi - Debian
cergei, > I plan to have the rootfs on a usb flash drive, > and the hdds will work as raid 0 drives to > increase the size of the hard drives without > mirroring. You have already booted this configuration above. Power up, in serial console, interrupt u-boot countdown and setenv devices 'usb ide' setenv bootdev usb boot The setenvs are transient. Afterby bodhi - Debian
> I removed two HDDs and booted from a rootfs flash > drive. Cool! That's pretty good. ===== The problem with this previous error EHCI timed out on TD - token=0xb4008d80 is that old u-boot code is not very robust dealing with USB. When you have large HDDs in the slots, a lot of power is used initializing them. Especially with this Stora box, it is problematic. Doby bodhi - Debian
cergei, Power up, in serial console, interrupt u-boot countdown and setenv devices 'usb ide' setenv bootdev usb and then boot Now it should boot into the USB rootfs.by bodhi - Debian
cergei, Looks like your envs are all messed up. Power up, in serial console, interrupt u-boot countdown at Hit any key to stop autoboot: 0 And then, printenvby bodhi - Debian
All, I'm rolling a new u-boot for this box. Hopefully it will remove the shackle of not being able to save u-boot envs, and therefore have to self-impose a kernel size limitation of 5MB on SPI flash. Please get these info. Boot with serial console, interrupt u-boot countdown and, printenv help bdinfo md.l f1018000 8 md.l f1018100 1 md.l f1018140 1by bodhi - Debian
cergei, Go ahead and connect serial console. Power up and let it boot automatically. And then post the entire serial console log here.by bodhi - Debian
> Yes, the download stops here. Because I previously > damaged fw_env. If I open the terminal with the Rx > pin disabled, the information will be displayed > incorrectly on the screen, which, in my opinion, > indicates that the connection is normal. Not sure I understand the above. Did you mean disconnecting wire to the RX pin on the board serial header? It seems you haveby bodhi - uBoot
Kenneth, Llooks good! > I didn't remove usb from devices since everything > seems OK. If it ain't broken... No fear. Once u-boot was flashed, it's done. Once in a while you should shutdown and backup the rootfs. QuoteWiki thread Backup and Cloning rootfs Back up a live rootfs using rsync CrashPlan 4.3.0 Backup/Restore rootfs using tar commandby bodhi - Debian
I'm afraid that for this box, like other 32-bit processors with single CPU, the performance might get worse. We've been running this kernel as non-SMP to gain a bit of performance. But looks like everything will be SMP in kernel 6.17. IOW, the kernel might have to be configured as SMP (running as numCPU =1), to avoid potential new bugs. https://lwn.net/ml/all/20250528080924.227by bodhi - Debian
tiaderosa, Congrats! > Personally I'd like to: > 1) Recover the stock firmware > 2) make a backup of the mtd partitions so I can > always go back to stock if I want to I would skip this. Later, you can hopefully install OpenWRT. And use it as rescue system. It's more productive if you can work on booting Debian on USB rootfs. That should be the end goal.by bodhi - Rescue System
irod, > Reading data from 0x3ff800 -- 100% complete. > 3145728 bytes read: OK > ## Booting image at 00800000 ... > Image Name: Linux-2.6.31.8 > Created: 2019-06-17 12:13:03 UTC > Image Type: ARM Linux Kernel Image > (uncompressed) > Data Size: 2502832 Bytes = 2.4 MB > Load Address: 00008000 > Entry Point: 00008000 >by bodhi - uBoot
Kenneth, Going back to the installation draft. https://forum.doozan.com/read.php?2,140220,140326#msg-140326 Quote3. Adjust the Linux env config file Edit /etc/fw_env.config to have the following content: cat /etc/fw_env.config # MTD device name Device offset Env. size Flash sector size Number of sectors # Lacie Netspace V2 /dev/mtd1 0x00000 0x1000 0x1000 After you'veby bodhi - Debian
tiaderosa, > I'm sorry to be so long but since if something > goes wrong the only way I have to resume my stora > is jtag an as far as I understand it s complex and > time consuming process, before going haed with the > nand erase command I'd like to share what I have > found up to now and some doubts I still have. > The doubts are not related to the procby bodhi - Rescue System
tiaderosa, My bad. I thought I've answered all your questions. Disclaimer: As I mentioned, I don't have this NAS and the unbricking sessions were done by others. So I'll do the best I can to read those posts and understand them. But there is no guarantee in my advice here. I could be mistaken and give you a wrong advice (or you could have a typo), that could hard brick thisby bodhi - Rescue System
> > Unfortunately, it's not in mainline u-boot yet. > > You mean this u-boot, or any rk3399 u-boot? This u-boot specifically. There are quite a few other rk3399 u-boots in mainline, but I don't see the rk3399-kobol-helios64.dts being used anywhere. > But somehow the host and otg mode can be combined. > When I boot to linux, I can login over serial, and >by bodhi - uBoot
Kenneth, Thanks for a great job testing. I'm declaring victory! To complete checking the flashing, while in Debian, make sure the env location is correct. Note: the "Flash sector size" should be 0x1000. cat /etc/fw_env.config # MTD device name Device offset Env. size Flash sector size Number of sectors # Lacie Netspace V2 /dev/mtd1 0x00000 0x1000 0x1000 And thenby bodhi - Debian
Mijzelf, > I'm trying to load files over tftp to my Kobol > Helios64. This box has 2 ethernet ports, one > 2.5Gbit, and one 1Gbit. When I give any network > related command, it gives a repeated > 'dw_dm_mdio_init: mdio node is missing, > registering legacy mdio busdw_dm_mdio_init'. Do mii info mdio list > In Linux that MAC address is the addreby bodhi - uBoot
Kenneth, > That was it! I can now reboot as expected. > Shut down, cut power and power up again also boots > the box. > > The output was: > > root@debian:~# cat /proc/mtd > dev: size erasesize name > mtd0: 00070000 00001000 "u-boot" > mtd1: 00001000 00001000 "u-boot-env" > root@debian:~# dd if=/dev/zero ibs=1 count=458752by bodhi - Debian
adrien, > My SSD doesn't have cache, it is a Patriot P120 > which has no DRAM cache. So it's good for rootfs (where mostly random read/writes occur), but not for network share (where most sequential read/writes occur). > I already tried disabling flow control, it doesn't > change, I have lot of buffer overrun errors with > flow control enabled or disabledby bodhi - Debian
Kenneth, Please try again this version. I've found that this flash chip mx25l4005a was not considered when u-boot unprotects all SPI flash upon starting. Just that it was old and lacking maintenance. So I added it to the list. Let's see if it will help. ========= uboot.2025.07-rc5-tld-1.netspace_mini_v2.bodhi.251020.tar sha256: a9dfb1cfb77dfc5465715a12a4202929a3a8dda6by bodhi - Debian
adrien, > root@readynas-104:~# dmesg | grep -i ata Everything looks normal. Does your SDD have cache, how big? > root@readynas-104:~# ethtool eth1 > Advertised pause frame use: Symmetric > Link partner advertised pause frame use: Symmetric Looks like flow control is enabled, that will slow down the xfer. dmesg | grep flow Unless you have a lot of netby bodhi - Debian
And also test both raw speed read for both disks hdparm -Tt /dev/sda hdparm -Tt /dev/sdbby bodhi - Debian
adrien, > root@readynas-104:~# dd if=/dev/zero > of=/Disque_1/test1.img bs=10M count=100 > 100+0 records in > 100+0 records out > 1048576000 bytes (1.0 GB, 1000 MiB) copied, > 14.4151 s, 72.7 MB/s That's more like it. In the ball park. Please post the output of these commands: dmesg | grep -i ata ethtool eth0 mount sysctl -a | grep -iE 'free|swapby bodhi - Debian
ZooMoo, > I looked at the Synology kernel source for how it > handles the kernel parameters, but I didn't get > much insight as it's nested through a series of > functions. I couldn't understand how it worked vs > what a stock kernel does with DTS. Stock kernel does not use DTS (it was too old). Usually you will find a xxx_setup.c in the old mainline kerby bodhi - Debian
adrien, root@readynas-104:~# dd if=/dev/zero of=/Disque_1/test1.img bs=10M count=100 oflag=dsync Should be dd if=/dev/zero of=/Disque_1/test1.img bs=10M count=100 With that dsync flag, you'd cut the HDD/SSD speed in half. sync option is not realistic. Also, to maximize speed, the shares should be mounted async,noatime. > Network speed seem to be good. In certain setup youby bodhi - Debian
adrien, > I see a lot of "buffer overrun" erros when I write > to the NAS, could this limit the transfer speed ? Definitely. Buffer overrun errors could occur if the flow control are not the same in the client and server. There are 3 parts in tuning NFS or SMB. 1. Run hdparm and dd tests to measure the hard disk speed. This is where you adjust the vm.min_free_kbytes.by bodhi - Debian