Follow dmesg Raspbian from distro.libre.computer
Your bootloader is from Armbian: BIOS 2024.01-armbian 01/01/2024
Armbian is of the first pastbin. Now I downloaded the image of Raspbian from distro.libre.computer to the sdcard. I do the upgrade before dmesg. I have Armbian on emmc but i do the boot from sdcard. It wrong because of this?
You are using the bootloader (Armbian) from eMMC.
See now please. Raspbian (from distro.libre.computer) from emmc.
[ 34.707591] usb 1-2.4: USB disconnect, device number 4
And there’s the on-board hub hard reset bug in all its glory, followed by reconnects, disconnects, … until the power is killed. I never tried using a keyboard or mouse (always used the console serial port), but this is interesting - it’s not just storage devices that the on-board hub hates, it’s any USB device.
For a workaround, how about trying an external hub on the “OTG” port? Bet that will work for you, assuming adequate power for the devices.
I am using an external usb hub (Orico, without external power) on the OTG port. The intention is to recover the other 3 ports, if can.
Thanks for the tip. This problem is interesting, I’ll check the log more closely and try to look for a solution as well. I have unparalleled support from the Libre Computer team.
USB disconnect issue is usually power related.
USB disconnect issue is usually power related.
That may be true, but there are plenty of us running into disconnects that aren’t power related. From what he’s said, I’m pretty sure Edilson is just fighting the disconnects that Libre has known about for months, and which have nothing to do with USB power.
April 17th: La Frite does not have a USB hub so it’s not affected by the timeout issues experienced by the hub. We are in the middle of fixing those issues for Sweet Potato and other boards. And if La Frite meets your needs, that proves to be the case. Sadly, I thought I’d hedge my bet on how much hardware I needed and got a Sweet Potato first. 
May 2nd: We plan to fix this by the time we roll out Linux 6.6 soon. To paraphrase the late Bob Newhart, “What is soon, Mr. librecomputer?”
Anyway, I pulled the AML boards off the shelf again, and verified that no, none of the handful of USB boot devices I’ve been working with have any trouble booting with Sweet Potato (using the OTG port) or La Frite. But even the little 8GB, USB2 thumb drive fails to boot if it’s on one of the hub ports. The console log goes something like this:
[ 33.582312] usbcore: registered new device driver onboard-usb-hub
[ 33.584906] meson-ir c8100580.ir: receiver initialized
[ 33.588738] usb 1-1.1: USB disconnect, device number 3
I don’t have a console log for that same thumb drive in the OTG, because it’s unnecessary: it just boots.
I bought my third power supply now from the Libre Computer store on Aliexpress to test, I should receive it in about twenty days (Brazil).
I have no hope that it is the source because I am using a very stable 5.1V/3A, as in the photo, it has 4.8V on the USB output. I’ve already used an external USB hub (Orico) with external power but no result.
The booting from USB issues you point out are unrelated to Edilson’s issue. Edilson’s issue is that none of his hub USB ports work. You are conflating issues and causing confusion. Stop polluting the thread with unrelated issues.
Stop the boot process in u-boot. Does u-boot detect your keyboard and storage devices on the USB hub ports? If not, the USB hub’s IO pins might be fried due to high voltages. The datasheet for the hub is here: http://www.sunnyqi.com/upLoad/product/month_1308/GL852G.pdf
VIN Input Voltage for digital I/O(Ovcur1-4,Pself,Reset) pins -0.5 +5.25 V
Sorry, I was actually responding more to your demonstrably incorrect claim that USB disconnects are usually power related. I got a little carried away summarizing the history of USB problems as I knew them, and of course those were about boot devices.
So by way of apology, I plugged a keyboard and mouse into a couple of the hub ports, and somewhat to my surprise they worked. Did NOT work as a keyboard once Linux came up, but they were known to hidraw. More to the point, this exercise lets me answer Edilson’s question from a while ago:
I don’t know how to see if keyboard and mouse detect on u-boot. Can you guide me to check?
You have to connect to the console port (header 2J1) with your trusty serial adapter in order to see the pre-EFI boot. If you do, you might see something like this if you have a keyboard and mouse plugged into the hub, and all the parts are working:
U-Boot 2023.07+ (Oct 13 2023 - 16:21:04 -0400) Libre Computer AML-S905X-CC-V2
Model: Libre Computer AML-S905X-CC V2
SoC: Amlogic Meson GXL (S905X) Revision 21:d (84:2)
DRAM: 2 GiB
Core: 181 devices, 32 uclasses, devicetree: separate
WDT: Not starting watchdog@98d0
MMC: mmc@72000: 1, mmc@74000: 0
Loading Environment from FAT... Card did not respond to voltage select! : -110
** Bad device specification mmc 0 **
Card did not respond to voltage select! : -110
** Bad device specification mmc 0 **
Couldn't find partition mmc 0
Card did not respond to voltage select! : -110
** Bad device specification mmc 0 **
Couldn't find partition mmc 0
Error: could not access storage.
[BL31]: tee size: 0
[BL31]: tee size: 0
starting USB...
Bus usb@c9000000: dwc3_meson_gxl_get_phys: usb2 ports: 2
Register 2000140 NbrPorts 2
Starting the controller
USB XHCI 1.00
scanning bus usb@c9000000 for devices... 4 USB Device(s) found
scanning usb for storage devices... 0 Storage Device(s) found
Hit any key to stop autoboot: 0
=>
NB: this was preceded by the boot loader’s stuff. And I interrupted uboot from the USB keyboard, so I know it actually accepted the USB keyboard.
This has been covered repeatedly on other threads and it’s not relevant to this thread. USB disconnects except for USB root devices is associated with power issues. The context which we made that statement is relevant to that thread and the context we made this statement is relevant to this thread. You are taking something we posted in another thread about another issue and repeating it here. This is confusing to the people we are trying to help, which is not helping. That is why we are asking you to not post about that issue here.
OpenEcoLabLasAlpuj
what does my dmesg log say about my situation?
is not enough?
Enough, in combination with my more recent experience, to hazard a guess. Is that screenshot from a terminal emulator talking to the Potato’s console? If so, the USB keyboard is recognized, but Linux isn’t using it for the console input, so it won’t do anything visible in that context. See also my previous post about finding and using the USB keyboard during uboot.