[SOLVED] can linux be set to ignore unused USB 1.0 hardware

[midigark]

I'm back...

I have an old motherboard - MSI 760G-P23 Rev 3. The southbridge contains USB 1.1 and USB 2.0 hardware, but only the USB 2.0 is attached to physical ports or connectors.

Is it possible to force the system to "ignore" the USB 1.1 hardware and for that matter, any other unimplemented devices?
I'm thinking sort of a similar method like Windows has in the device manager - "Do not use this device".

Thanks,
Mark

LM21.3 Virginia Cinnamon
AMD FX8320E @ 3.6GHz CoolerMaster Hyper 212 EVO
Team Group DDR3 1600 16Gb
EVGA GTX660 OC Nvidia 470.223.02
ASMedia PCIe x1 SATA 3 interface

[SMG]

I have an old motherboard - MSI 760G-P23 Rev 3. The southbridge contains USB 1.1 and USB 2.0 hardware, but only the USB 2.0 is attached to physical ports or connectors.

Is it possible to force the system to "ignore" the USB 1.1 hardware and for that matter, any other unimplemented devices?

If it is not attached to anything, why would it show up? Does it show up when you run ls usb?

[midigark]

Thank you for the reply.

Yes, Bus 6 and Bus 7, but as it turns out are both part of already initialized "pairs". (1 on each of the 2 motherboard USB headers)

But, things just got a little stranger...
The motherboard 760GM-P23(FX) (MS-7641) MSI comes with 6 USB 2.0 connections on the back and 2 USB 2.0 headers (4 USB 2.0 connections) on the motherboard, for a total of 10 USB 2.0 ports.

Output from lsusb which is confirmed by Hard Info - System Information: (sorted)

Code: Select all

Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 001 Device 007: ID 1b3f:2247 Generalplus Technology Inc. GENERAL WEBCAM [USB 2.0]

Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 002 Device 002: ID 2109:0715 VIA Labs, Inc. VL817 SATA Adaptor [USB 2.1 - Physically USB 3]
Bus 002 Device 003: ID 1058:1021 Western Digital Technologies, Inc. Elements Desktop (WDBAAU) [USB2.0]

Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 003 Device 002: ID 0b05:190e ASUSTek Computer, Inc. ASUS USB-BT500 [USB 2.0] ???
Bus 003 Device 004: ID 046d:c223 Logitech, Inc. G11/G15 Keyboard / USB Hub [USB 1.1]
Bus 003 Device 005: ID 046d:c226 Logitech, Inc. G15 Refresh Keyboard
Bus 003 Device 006: ID 046d:c227 Logitech, Inc. G15 Refresh Keyboard

Bus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 004 Device 002: ID 1b1c:1b2e Corsair Corsair Corsair Gaming M65 Pro RGB Mouse [USB 2.0] ???
Bus 004 Device 003: ID 046d:0a5c Logitech, Inc. Logitech G633 Gaming Headset [USB 2.0] ???

Bus 005 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 005 Device 002: ID 046d:c215 Logitech, Inc. Extreme 3D Pro [USB 1.1]

Bus 006 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub

Bus 007 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub

The motherboard southbridge (AMD SB710) has the following USB specs
USB Controllers
5 OHCI and 2 EHCI Host controllers to support(s) 12 USB 2.0 ports and 2 dedicated USB 1.1 ports
> ACPI S1 ~ S5 supported
> Legacy Keyboard/Mouse support
> USB debug port
> Port disable supported with individual control

The motherboard manual states that all 10 of the available USB ports are USB 2.0 capable.

The question now becomes:
Why is the OS assigning the root hubs as USB 1.1 when the attached devices are USB 2.0?
Bus 003 Device 002 ASUS USB-BT500 Bluetooth adapter [USB 2.0]
(this would explain the painfully slow picture transfer to/from my phone and painfully glitchy audio support)
Bus 004 Device 002 Corsair M65 Pro RGB Mouse [USB 2.0]
Bus 004 Device 003 Logitech G633 Gaming Headset [USB 2.0]

Seems that every time I dig for answers, I find new questions. lol
Hopefully, this motherboard will be going away in a week or so. Just looking for answers to why its been behaving the way it has.

Thanks, as always!
Mark

[SMG]

But, things just got a little stranger...
The motherboard 760GM-P23(FX) (MS-7641) MSI comes with 6 USB 2.0 connections on the back and 2 USB 2.0 headers (4 USB 2.0 connections) on the motherboard, for a total of 10 USB 2.0 ports.

You have the H.H BIOS version installed on it? (That is what I found listed as the newest one.)

I did a search on USB 2.0 ports showing as 1.1 and there might be a setting in the BIOS to set them as a higher speed (usb enhanced controller?). I've not run into this type of situation so I don't know for sure.

I did find this which indicates:

On Linux

USB 2.0 will use the ehci_hcd module
USB 1.x will use either ohci_hcd or uhci_hcd modules.

You can find out which module is being used by lspci -v or lshw.

Perhaps a better method (if you're interested in particular ports/devices) is to use "lsusb -t"

[midigark]

Thanks for the timely response.

Ghaaaa...CURSE MSI and their silly BIOS naming conventions...

I have what is known as
American Megatrends Inc. (AMI)
BIOS Version V17.17
With a date of 04/22/2015

This corresponds with MSI BIOS release
AMI BIOS Version H.H
With a release date of 04/23/2015

There doesn't seem to be an "Enhanced USB Controller" setting.
But, the setting might be "Enable Legacy USB support".
Might give that a try and see what happens. Worst case is I will lose the USB keyboard before POST completes.
I still have an old serial one in arms reach.

Viewing https://docs.kernel.org/usb/index.html
Neither the OHCI or the EHCI "chapters" really explain how or why a USB 2.0 device connected to a USB 2.0 controller would be assigned a USB 1.1 driver.
I'm going to give the BIOS setting a once over as the manual really does not go into detail.
MSI as always been weaker here than other brands like ASUS to name just one. (sighs)

Wish me luck, it's still St. Patrick's day...maybe I need to rub a shamrock.
Mark

[midigark]

Back again with some corrections that do not help things.

I am in "USB Legacy Support Mode" [DISABLED]
No change as far as I can detect.

Corrections to the USB data:

Code: Select all

Bus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
Bus 003 Device 002: ID 0b05:190e ASUSTek Computer, Inc. ASUS USB-BT500 [USB 2.0] ???
Bus 003 Device 004: ID 046d:c223 Logitech, Inc. G11/G15 Keyboard / USB Hub [USB 1.1]
(the above is the internal Logiech G15 HUB to 2 Low Power USB 1.1 connectors on the back of the keyboard)
Bus 003 Device 005: ID 046d:c226 Logitech, Inc. G15 Refresh Keyboard [USB 2.0] (probably the main keyboard)
Bus 003 Device 006: ID 046d:c227 Logitech, Inc. G15 Refresh Keyboard [USB 2.0] (probably the display and additional Macro Keys)

Next step is to swap some of the connections around to see if what is connected, affects if the root hubs are assigned as USB 1.1 or USB 2.0.
Or if the HUB configuration remains static. I can do without the webcam and we know that hub is already assigned as USB 2.0. I'll start with the ASUS BT-500 dongle and go from there.

Mark

[midigark]

back again with more info.

Apparently, the problem is the Linux kernel and how it detects the hubs.
viewtopic.php?t=83548
This goes on and on, finally ending up with nothing answered, but another post (lost the link) describes how on a dual boot system (Win 8.1) the ports behave differently between Linux and Microsoft. These posts are from early 2000's and the people describing the problem are not exactly using correct terminology when stating speeds, which only makes the reading more painful.

I've been mapping out the ports, being suspicious that the port is assigned the USB version with the lowest common denominator. If this were true, it would be down right stupid, as a USB 2.0 HUB is quite capable of supporting USB 1.1 devices.


From a users point of view, what I don't understand and quite frankly, sounds of either laziness or utterly craptastic programming, is if a motherboard chipset USB controller supports USB 1.1 and/or USB 2.0, (aside from critical power savings) why would anyone in their right mind even consider setting it to be a USB 1.1 HUB?

As it stands now, under Linux Mint 21.3 Virginia, my computer has 4 x USB 2.0 EHCI ports and the other 6 are USB 1.1 OHCI.
When the manual clearly states that all 10 of the motherboard USB ports are USB 2.0, there must be something rotten under the hood.
Either MSI or Linux is pulling my leg.

Across the I/O shield:
+---------+---------+---------+
| USB 1.1 | USB 2.0 | USB 2.0 |
+---------+---------+---------+
| USB 1.1 | USB 2.0 | USB 2.0 |
+---------+---------+---------+
And, both the motherboard headers show up as 2 USB 1.1 ports each

I'll admit that some of the specifications of the SB710 southbridge are more than a bit confusing, but I don't believe that there could be a USB 2.0 HUB operating as USB 2.0 OHCI. A USB 2.0 HUB should always be able to manage a USB 1.1 OHCI device -> as a USB 1.1 OHCI device.
If I'm mistaken, please correct me.

A USB 2.0 port is a USB 2.0 is a USB 2.0 port and will always host a USB 1.1 device properly.
It's not just me saying that, the industry says it quite clearly.
That's why it's called "downwards compatibility".

So, why Linux, why?

[SMG]

...but another post (lost the link) describes how on a dual boot system (Win 8.1) the ports behave differently between Linux and Microsoft.

I also came across information indicating that has happened.

Most of the information I found was quite old and was mainly from the time when usb 2.0 and 3.0 were coming out.

I don't have any other ideas. Maybe someone else will.

[midigark]

Thank you again.

As I said, this motherboard will be going away, hopefully in a week. It's just been a "bug" that's been pestering me since I got the USB 5.0 adapter. And, I guess I was correct to be "bugged" about it.

Using a program named "hw-probe" it's clear that, according to Linux, there are not 10 USB 2.0 ports.


$ sudo -E hw-probe -all -upload

Provides a secure link to view your machines data.
Just a clip of it.

As well as links to just about every known listing/probing tool available in linux.

Thank you again for your time and efforts.
Mark