search

dangowrt / owrt-ubi-installer

OpenWrt firmware installer for the Linksys E8450 aka. Belkin RT3200
GNU General Public License v2.0
395 stars 52 forks source link

no lights when powering on,am i bricked #166

Open ned42 opened 11 months ago

ned42 commented 11 months ago

got openwrt worked on my e8450 with ubi installer sucessfully for like a month just cant boot on now when i powering it on and i cant see any light turning on either front or back tried press reset button with hold and turn on for like 3-4 times but still got nothing tired connect with a pc on 192.168.1.x to confirm whether the system is working got nothing wanna know how to solve this kind of situation thanks

ned42 commented 10 months ago

so i checked ttl output,it says "BL2: Failed to load image id 3" just like this:https://forum.openwrt.org/t/need-help-with-a-bricked-linksys-e8450/144836 it sounds like i need to plug 20 pins into jtag to do the recover, but in openocd-scripts here says:

If the bootloader is broken, make sure BootROM has output "System halt!" before moving to step 2

so...any chance i can get rid of 20 pins jtag? @dangowrt

usrlocalben commented 9 months ago

There is an outline for recovering the device via JTAG here: https://forum.openwrt.org/t/belkin-rt3200-linksys-e8450-wifi-ax-discussion/94302/4010

My impression is that @dangowrt is working on some new boot tech. (Also described in that thread) If progress is made there then I'll recover mine and put it in service--otherwise it goes in the trash since it's not reliable.

ned42 commented 9 months ago

@usrlocalben thanks for the tips! i've sent it to a local workshop and got it repaired by flashing some copy files(sort of programmer reflashing idk) from a new machine. and yes, i still dont even know how it just went out during normal working, just hope this time i wont need to send it back that soon ：）

schuettecarsten commented 8 months ago

I have the same problem - no lights turn on, but the USB port has power. For me, turning the device off, wait some seconds or minutes and then turn it on sometimes helps. If not, try again and again. So, the device is not really bricked, but the bootloader seems to be buggy.

@dangowrt Is that bad cell count output normal?

Fri Mar  8 13:22:07 2024 kern.info kernel: [    0.367531] jffs2: version 2.2 (NAND) (SUMMARY) (LZMA) (RTIME) (CMODE_PRIORITY) (c) 2001-2006 Red Hat, Inc.
Fri Mar  8 13:22:07 2024 kern.err kernel: [    0.411148] mt7622-pinctrl 10211000.pinctrl: invalid group "pwm_ch7_2" for function "pwm"
Fri Mar  8 13:22:07 2024 kern.err kernel: [    0.424004] mt-pmic-pwrap 10001000.pwrap: unexpected interrupt int=0x1
Fri Mar  8 13:22:07 2024 kern.info kernel: [    0.439782] Serial: 8250/16550 driver, 3 ports, IRQ sharing disabled
Fri Mar  8 13:22:07 2024 kern.info kernel: [    0.447150] printk: console [ttyS0] disabled
Fri Mar  8 13:22:07 2024 kern.info kernel: [    0.471627] 11002000.serial: ttyS0 at MMIO 0x11002000 (irq = 118, base_baud = 1562500) is a ST16650V2
Fri Mar  8 13:22:07 2024 kern.info kernel: [    0.480954] printk: console [ttyS0] enabled
Fri Mar  8 13:22:07 2024 kern.info kernel: [    0.489321] printk: bootconsole [uart8250] disabled
Fri Mar  8 13:22:07 2024 kern.info kernel: [    0.519982] 11004000.serial: ttyS1 at MMIO 0x11004000 (irq = 119, base_baud = 1562500) is a ST16650V2
Fri Mar  8 13:22:07 2024 kern.info kernel: [    0.530617] mtk_rng 1020f000.rng: registered RNG driver
Fri Mar  8 13:22:07 2024 kern.notice kernel: [    0.536335] random: crng init done
Fri Mar  8 13:22:07 2024 kern.info kernel: [    0.543740] loop: module loaded
Fri Mar  8 13:22:07 2024 kern.info kernel: [    0.547458] mtk-ecc 1100e000.ecc: probed
Fri Mar  8 13:22:07 2024 kern.info kernel: [    0.555261] spi-nand spi2.0: Fidelix SPI NAND was found.
Fri Mar  8 13:22:07 2024 kern.info kernel: [    0.560613] spi-nand spi2.0: 128 MiB, block size: 128 KiB, page size: 2048, OOB size: 64
Fri Mar  8 13:22:07 2024 kern.info kernel: [    0.568745] mtk-snand 1100d000.spi: ECC strength: 4 bits per 512 bytes
Fri Mar  8 13:22:07 2024 kern.notice kernel: [    0.575632] 2 fixed-partitions partitions found on MTD device spi2.0
Fri Mar  8 13:22:07 2024 kern.err kernel: [    0.582038] OF: Bad cell count for /spi@1100d000/flash@0/partitions
Fri Mar  8 13:22:07 2024 kern.err kernel: [    0.588311] OF: Bad cell count for /spi@1100d000/flash@0/partitions
Fri Mar  8 13:22:07 2024 kern.notice kernel: [    0.594824] Creating 2 MTD partitions on "spi2.0":
Fri Mar  8 13:22:07 2024 kern.notice kernel: [    0.599645] 0x000000000000-0x000000080000 : "bl2"
Fri Mar  8 13:22:07 2024 kern.notice kernel: [    0.605428] 0x000000080000-0x000008000000 : "ubi"
Fri Mar  8 13:22:07 2024 kern.notice kernel: [    0.743171] ubi0: default fastmap pool size: 50
Fri Mar  8 13:22:07 2024 kern.notice kernel: [    0.747713] ubi0: default fastmap WL pool size: 25
Fri Mar  8 13:22:07 2024 kern.notice kernel: [    0.752525] ubi0: attaching mtd1
Fri Mar  8 13:22:07 2024 kern.notice kernel: [    1.034551] ubi0: scanning is finished
Fri Mar  8 13:22:07 2024 kern.notice kernel: [    1.043104] ubi0: attached mtd1 (name "ubi", size 127 MiB)
Fri Mar  8 13:22:07 2024 kern.notice kernel: [    1.048603] ubi0: PEB size: 131072 bytes (128 KiB), LEB size: 126976 bytes
Fri Mar  8 13:22:07 2024 kern.notice kernel: [    1.055489] ubi0: min./max. I/O unit sizes: 2048/2048, sub-page size 2048
Fri Mar  8 13:22:07 2024 kern.notice kernel: [    1.062274] ubi0: VID header offset: 2048 (aligned 2048), data offset: 4096
Fri Mar  8 13:22:07 2024 kern.notice kernel: [    1.069227] ubi0: good PEBs: 1020, bad PEBs: 0, corrupted PEBs: 0
Fri Mar  8 13:22:07 2024 kern.notice kernel: [    1.075314] ubi0: user volume: 8, internal volumes: 1, max. volumes count: 128
Fri Mar  8 13:22:07 2024 kern.notice kernel: [    1.082531] ubi0: max/mean erase counter: 87/58, WL threshold: 4096, image sequence number: 268174460
Fri Mar  8 13:22:07 2024 kern.notice kernel: [    1.091749] ubi0: available PEBs: 0, total reserved PEBs: 1020, PEBs reserved for bad PEB handling: 20
Fri Mar  8 13:22:07 2024 kern.notice kernel: [    1.101060] ubi0: background thread "ubi_bgt0d" started, PID 206
Fri Mar  8 13:22:07 2024 kern.err kernel: [    1.101289] OF: Bad cell count for /spi@1100d000/flash@0/partitions
Fri Mar  8 13:22:07 2024 kern.err kernel: [    1.113515] OF: Bad cell count for /spi@1100d000/flash@0/partitions
Fri Mar  8 13:22:07 2024 kern.info kernel: [    1.120459] block ubiblock0_5: created from ubi0:5(fit)
dangowrt commented 8 months ago

Is that bad cell count output normal?

Yes, that's not related, just a formality related to DTS, no functional impact what-so-ever.

Sounds more like you are also affected by https://github.com/mtk-openwrt/arm-trusted-firmware/pull/7

dalutou commented 8 months ago

I have also encountered this problem because of a sudden power outage, and the E8450 has never bootup again. After recovery with jtag, I still use the same openwrt version (23.05.0). When executing the reboot command, there is a high probability that it will not be able to bootup.

But after adding the following command to /etc/rc.local

echo 812500 > /sys/devices/system/cpu/cpufreq/policy0/scaling_min_freq

The problem never appeared again.

Remove scaling_min_freq setting, bootup failure will occur again.

@dangowrt

schuettecarsten commented 8 months ago

But after adding the following command to /etc/rc.local

echo 812500 > /sys/devices/system/cpu/cpufreq/policy0/scaling_min_freq

Do you also set a different governor? Unfortunately, this did not help for me.

I also found this: https://forum.openwrt.org/t/belkin-rt3200-linksys-e8450-wifi-ax-discussion/94302/4062

schuettecarsten commented 8 months ago

My device is in a very strange state now. I can set whatever I want, change CPU frequencies or governors, nothing seems to help yet. On every reboot, regardless if it is a soft or a hard reboot, the device seems to be dead. Solution is to unplug it competely, wait 5 to 10 minutes, reconnect and it will boot. On the next reboot, it will look dead again, so unplug, wait, and it boots again....

dalutou commented 8 months ago

My device is in a very strange state now. I can set whatever I want, change CPU frequencies or governors, nothing seems to help yet. On every reboot, regardless if it is a soft or a hard reboot, the device seems to be dead. Solution is to unplug it competely, wait 5 to 10 minutes, reconnect and it will boot. On the next reboot, it will look dead again, so unplug, wait, and it boots again....

schuettecarsten commented 8 months ago

Is it possible to update the bootloader from a running OpenWRT system? I do not want to run into reboot issues while upgrading using the installer and my device is not yet "opened".

dangowrt commented 8 months ago

Is it possible to update the bootloader from a running OpenWRT system?

Yes, by force-flashing the v1.1.0-or-later installer.

schuettecarsten commented 8 months ago

Is it possible to update the bootloader from a running OpenWRT system?

Yes, by force-flashing the v1.1.0-or-later installer.

But then I have to reboot - i hoped it is possible to flash a possibly fixed bootloader directly without needing to boot from a broken one.

dangowrt commented 8 months ago

I get your point, but it's not possible because what the installer does is much more than just writing a new bootloader. The size of the UBI MTD partition also changed, and factory and fip need to be written into static UBI volumes instead of MTD partitions...

schuettecarsten commented 8 months ago

Ok, and if I already ran the 1.1.0 installer some weeks ago, as I am using custom snapshot builds? Is manual update to the 1.1.1 bootloader possible?

dangowrt commented 8 months ago

Is manual update to the 1.1.1 bootloader possible?

Yes, that's easy, but not needed (as in: it won't improve anything). You can do it in the bootloader menu while providing the files via TFTP.

schuettecarsten commented 8 months ago

Hmm, ok. So I hope for a good idea and a fix. 🤷