Thank you for your donation!


Cloudsmith graciously provides open-source package management and distribution for our project.


Problem: RPi B4 slowdown with using USB 3.0
#11
(07-12-2023, 08:55 AM)DRONE7 Wrote:
(07-12-2023, 08:02 AM)STUDI Wrote:
(07-12-2023, 06:47 AM)DRONE7 Wrote:
(07-12-2023, 06:20 AM)STUDI Wrote:
(07-11-2023, 10:24 PM)DRONE7 Wrote: This is a power supply problem well canvassed regarding Pi4.
https://forums.raspberrypi.com/viewtopic.php?t=257220
https://www.reddit.com/r/raspberry_pi/co...e_causing/
https://forums.raspberrypi.com/viewtopic.php?t=248572


"The USB ports on the Raspberry Pi can only supply up to 1.2A from all 4 ports combined (with no per-port limits). That's not enough for two USB powered hard drives. You'll need to power at least one of the drives externally."

Bad suggestion. All you need is the mouse dongle itself. Inserted into USB 2.0 ports, no slowdown, but inserted into a USB 3.0 port, we immediately have a slowdown, and in the case of a mouse, it makes the GUI practically unusable. As a pendrive (32GB, Sandisk, standard an nano model) as mass storage, identical behavior. Of course, it's the same only for the USB SSD itself (no udervoltage warnings). There are no network connections for this case (Local GUI).

Two devices, pendrive and DAC Topping D10s, as long as they are connected to USB 2.0 ports, we have normal correct operation. But it is enough that only one of them (no matter which one) is plugged into the USB 3.0 port and the work of the Raspberry Pi is slowed down. There are no network connections for this case (Local GUI).

The pendrive (and mouse dongle) does not draw more than 1.2A of current from the USB port (just to clarify: no other peripherals connected to the USB ports).



I only see 2 possibilities:

1) A construction error in this initial 4B Raspberry Pi series that interferes with the software (kernel, drivers) which manifests itself in slowdowns, many lags.

2) Damage to the Raspberry Pi I have that causes the software to behave this way.

I don't have a second Raspberry Pi, it's practically impossible to buy (out of stock) and even if they were available, they would be in a different hardware version.

I thought it was a known problem (hence the question here on the forum), testing with buying another (already more expensive) Raspberry Pi is not an optimal solution.

In some time I can do some tests with generating logs. But first, job gains priority.
Bad reply ;-)
So have you ever connected a powered hub between any of these devices and the Pi ?
Have you examined the USB cables for current carrying capacity ?
Have you posted your power supply details for your Pi ?

I am aware of the poor quality of USB cables (significant voltage drops resulting from small cross-sections of wires, even in very expensive and supposedly high-quality USB cables). I used the HAT DAC extension with a 5.5/2.1 power socket plus a homemade power cable, e.g. from a power bank (1.5mm2 wire cross-section). There were no significant voltage drops that would have triggered an undervoltage warning (4.85 - 4.9V on power socket in Raspberry Pi system).

I am an electronics engineer by passion, so the problem of the quality of the USB cable is too well known to me (once I needed to power the PLC with HMI from powerbank - to support the creation of the program, they will be somewhere outside the house, so I reworked this topic and the conclusion was to make USB cables as power cables myself).
Whoa..wind that back a bit ..."HAT DAC extension "  ??  seems there are several peripheral interactions here with unreported hardware.
Suggest you post a full summary of your Pi and all its attendant hardware...otherwise we just keep trying to troubleshoot on the basic system.


But what interactions? It doesn't even mention playback. Just any device not related to mass storage or USB DAC causes a problem. Even if it's just a bare Rasbperry Pi board. Anything used and plugged into a USB 3.0 port. Even an ordinary dongle from a wireless mouse or a small keyboard. General problem, not specific with USB mass storage, USB DAC, etc. In addition, the other combinations described preclude the simple diagnosis that it is only the power supply.

This means that the problem is not "external", i.e. dependent on the connected devices, but "internal", i.e. inherent in the Raspberry Pi itself, perhaps only in the one I have. The second is 2B version 1.1. If these were problems with these extensions, the specific model of the connected device would also appear on the Raspberry Pi. Once on this kicking Rabperry Pi 4B I even checked the 32-bit version of Mooode. The issue in question occurred on this version of Moodeaudio. In the Internet there are mentions that the first Raspberry PI 4B had problems with USB support with the suggestion of a construction error.
Reply
#12
I plugged in a wireless kbd/trackpad into Pi-4B USB3 port and am not able to repro you issue. The Pi has a USB DAC connected to USB2 and Local Display on.

Code:
pi@sig:~ $ moodeutl --revision
0xb03111        4B      1.1     2GB     Sony UK BCM2711

pi@sig:~ $ lsusb
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub
Bus 001 Device 004: ID 20b1:0008 XMOS Ltd Revolution
Bus 001 Device 003: ID 046d:c52b Logitech, Inc. Unifying Receiver
Bus 001 Device 002: ID 2109:3431 VIA Labs, Inc. Hub
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
pi@sig:~ $

Are there other steps to try and reproduce the issue?
Enjoy the Music!
moodeaudio.org | Mastodon Feed | GitHub
Reply
#13
(07-12-2023, 09:20 AM)STUDI Wrote:
(07-12-2023, 08:55 AM)DRONE7 Wrote:
(07-12-2023, 08:02 AM)STUDI Wrote:
(07-12-2023, 06:47 AM)DRONE7 Wrote:
(07-12-2023, 06:20 AM)STUDI Wrote: Bad suggestion. All you need is the mouse dongle itself. Inserted into USB 2.0 ports, no slowdown, but inserted into a USB 3.0 port, we immediately have a slowdown, and in the case of a mouse, it makes the GUI practically unusable. As a pendrive (32GB, Sandisk, standard an nano model) as mass storage, identical behavior. Of course, it's the same only for the USB SSD itself (no udervoltage warnings). There are no network connections for this case (Local GUI).

Two devices, pendrive and DAC Topping D10s, as long as they are connected to USB 2.0 ports, we have normal correct operation. But it is enough that only one of them (no matter which one) is plugged into the USB 3.0 port and the work of the Raspberry Pi is slowed down. There are no network connections for this case (Local GUI).

The pendrive (and mouse dongle) does not draw more than 1.2A of current from the USB port (just to clarify: no other peripherals connected to the USB ports).



I only see 2 possibilities:

1) A construction error in this initial 4B Raspberry Pi series that interferes with the software (kernel, drivers) which manifests itself in slowdowns, many lags.

2) Damage to the Raspberry Pi I have that causes the software to behave this way.

I don't have a second Raspberry Pi, it's practically impossible to buy (out of stock) and even if they were available, they would be in a different hardware version.

I thought it was a known problem (hence the question here on the forum), testing with buying another (already more expensive) Raspberry Pi is not an optimal solution.

In some time I can do some tests with generating logs. But first, job gains priority.
Bad reply ;-)
So have you ever connected a powered hub between any of these devices and the Pi ?
Have you examined the USB cables for current carrying capacity ?
Have you posted your power supply details for your Pi ?

I am aware of the poor quality of USB cables (significant voltage drops resulting from small cross-sections of wires, even in very expensive and supposedly high-quality USB cables). I used the HAT DAC extension with a 5.5/2.1 power socket plus a homemade power cable, e.g. from a power bank (1.5mm2 wire cross-section). There were no significant voltage drops that would have triggered an undervoltage warning (4.85 - 4.9V on power socket in Raspberry Pi system).

I am an electronics engineer by passion, so the problem of the quality of the USB cable is too well known to me (once I needed to power the PLC with HMI from powerbank - to support the creation of the program, they will be somewhere outside the house, so I reworked this topic and the conclusion was to make USB cables as power cables myself).
Whoa..wind that back a bit ..."HAT DAC extension "  ??  seems there are several peripheral interactions here with unreported hardware.
Suggest you post a full summary of your Pi and all its attendant hardware...otherwise we just keep trying to troubleshoot on the basic system.


But what interactions? It doesn't even mention playback. Just any device not related to mass storage or USB DAC causes a problem. Even if it's just a bare Rasbperry Pi board. Anything used and plugged into a USB 3.0 port. Even an ordinary dongle from a wireless mouse or a small keyboard. General problem, not specific with USB mass storage, USB DAC, etc. In addition, the other combinations described preclude the simple diagnosis that it is only the power supply.

This means that the problem is not "external", i.e. dependent on the connected devices, but "internal", i.e. inherent in the Raspberry Pi itself, perhaps only in the one I have. The second is 2B version 1.1. If these were problems with these extensions, the specific model of the connected device would also appear on the Raspberry Pi. Once on this kicking Rabperry Pi 4B I even checked the 32-bit version of Mooode. The issue in question occurred on this version of Moodeaudio. In the Internet there are mentions that the first Raspberry PI 4B had problems with USB support with the suggestion of a construction error.
I have the same early version of the Pi4, updated the firmware soon after it came out and AFAIK MoOde does this too, and I have not had the problem you report.
Perhaps, as you have mentioned, you have a defective unit.
----------
bob
Reply
#14
I can't spend time fiddling with the Raspberry Pi at the moment. Professional work has its priorities. But I'll take a closer look when I have more time. I think that dmesg can help because at this stage I assume a problem at the hardware level and the operating system at the kernel level.
Reply


Forum Jump: