http://bretjohnson.us/forum

by **Dinosaur** » Thu Jul 30, 2009 12:39 pm

Hi Bret

Just to refresh, I am using an Industrial cpu board with on board usb controller,which I have tested with a mouse.

I have two issues.
1.Joystick not recognised
=====================
My Joystick is connected to a usb gamecontroller that is configurable for buttons and Axis.Here is the link.
http://www.u-hid.com/index.php?option=c ... &Itemid=19
Autoexec.bat
usbuhci
usbprint
usbjstik (have also tried usbmouse here)
C:\Dos32\ctmouse

UsbDevice output:

Code: Select all: USBDEVIC 0.05, (C) 2008, Bret E. Johnson. Program to display information about Devices attached to the USB Host(s). DEVICE ADDRESSES ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Host Index: 0 Host Type: UHCI Bus Type: PCI IRQ#: 10 Root Hub Ports: 2 Vendor: 1106h = VIA Technologies Inc Product: 3038h ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ DEVICES INTERFACES ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ L C I A O ADRS o P o n l w ÍÍÍÍ (hex) S o BUS n t t n Test VEND PROD Sub Pro p USB HUB r POWR f f I e Sub Pro RWak ID ID Cls Cls col d VER ADR t (mA) g c n DESCRIPTION d Cls Cls col ÍÍÍÍ ÍÍÍÍ ÍÍÍÍ ÍÍÍ ÍÍÍ ÍÍÍ Í ÍÍÍ ÍÍÍ Í ÍÍÍÍ Í Í Í ÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Í ÍÍÍ ÍÍÍ ÍÍÍ 1 1106 3038 9 0 0 . 1.0 ... . s 0 1 0 0*Root Hub Y 9 0 0 VIA Technologies Inc ÄÄÄÄ ÄÄÄÄ ÄÄÄÄ ÄÄÄ ÄÄÄ ÄÄÄ Ä ÄÄÄ ÄÄÄ Ä ÄÄÄÄ Ä Ä Ä ÄÄÄÄÄÄÄÄÄÄÄÄÄÄ Ä ÄÄÄ ÄÄÄ ÄÄÄ 2 D209 1501 0 0 0 . 2.0 1 1 500 1 0 0*Human Intf Dvc . 3 0 0 No Matching Record Found 1 0*Human Intf Dvc . 3 0 0 2 0*Human Intf Dvc . 3 0 0 3 0*Mouse . 3 1 2

Basically there is no response at all, so my questions:
i) To use the Joystick as a mouse, what configuration do I need.It is worth pointing out that the designer
of the interface made these comments.

Usage in DOS raises all kinds of questions. The DOS driver must be capable
of supporting a USB game controller device. I am not sure if such software
is available. The user program would require special API calls to
interface with the DOS driver, if such a thing exists.

When I asked if the controller could connect to the ps2 port via an adapter

There is no way to pass analog 12 but data packets over a PS/2 connection
and there is n such thing as a USB to PS/2 game controller converter.

I have tested the controller on XP, and it basically works.

2. usbmouse Hangs computer.
=========================
As I have said previously, the usbmouse works perfectly, However after leaving it running for a few hours it hangs the computer.
Initially I suspected other issues, (updated to latest FreeDos, updated FreeBasic compiler, etc) but after a process of careful elimination
the usb port or the usb programs are to blame. Simply disabling the usb mouse and plugging it into the ps2 port via an adapter
makes the problem go away. I have even tried not touching the mouse at all, still after an hour or so it hangs the computer.
At one stage there was an error message on the dos screen ??fatal exception in Ring0. Stopped.
Because this happens during the execution of my gui interface, I cant see what happens, so I will test again in the dos prompt.

Any suggestions appreciated.

Regards
UPDATE: After 1 hour or so of running the usbmouse in ps2mtest the following error:
Invalid opcode at 00A8 0006 3406 0353 004D 0600 0000 0000 0000 0000 0000 0000 4150
Will do another test with minimum drivers installed.
If DisableLegacySupport used,(default IRQ) the error happens within seconds.
If usbuhcil IRQ 11, lasted for 1.5 hours,(occasionally moved mouse) then same error as above.
UPDATE: After 3 hours and 50 minutes I had a failure with the following Autoexec.bat

C:\Dos32\cwsdpmi.exe
usbuhcil IRQ 10 DisableLegacySupport
usbprint
usbmouse
cd\
C:\Dos32\ctmouse
c:\usb\ps2mtest

Disabling Cwsdpmi made the computer lockup in minutes, but no error message showing.

by **Bret** » Fri Jul 31, 2009 3:25 am

Code: Select all
IRQ 1.00, (C) 2008, Bret E. Johnson. IRQ INT COMMON USE ENABLED ÄÄÄ ÄÄÄ ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ ÄÄÄÄÄÄÄ 00 08h *System Timer* Yes 01 09h *Keyboard* Yes 02 0Ah *IRQ Controller #2* Yes 03 0Bh COM2/COM4 04 0Ch COM1/COM3 05 0Dh LPT2/Sound Card 06 0Eh Floppy Disk 07 0Fh LPT1 ÄÄÄ ÄÄÄ ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ ÄÄÄÄÄÄÄ 08 70h *CMOS Clock* 09 71h n/a Yes 10 72h n/a Yes 11 73h n/a Yes 12 74h PS/2 Mouse Yes 13 75h *Math Coprocessor* Yes 14 76h Hard Disk Yes 15 77h IDE Controller #2 Yes

This is from an earlier e-mail, where you did an IRQ dump. This indicates that IRQ's 9-15 are already used, and the USB controller was trying to use IRQ 12 along with the PS2 mouse. Your use of IRQ's 10 and 11 in the tests could still be causing an IRQ conflict, since those IRQ's are being used by some other hardware devices inside the computer (though I have no idea what they are). Try using one of the IRQ's that are unused above (like IRQ 5 or 3) and see if you still have a problem. The problem may not be an IRQ conflict, but let's try to absolutely eliminate that as a possibility.

Also send the results of a USBUHCIL DESCR, USBUHCIL REPORT 2 0, USBUHCIL REPORT 2 1, USBUHCIL REPORT 2 2, and USBUHCIL REPORT 2 3. This will display the HID report descriptors that the U-HID is generating to see if it really looks like a joystick or not.

by **Dinosaur** » Fri Jul 31, 2009 12:30 pm

Hi Bret

The IRQ's on the industrial board are pre-assigned to various functions, ie: Com1:,Floppy etc.
When I enable all devices, the only IRQ's that will work are 9,10 & 11
The others, although they show as being enabled, create an error with usbuhcil.
I think the n/a actually means "Not Assigned", and therefore these are available for use.

I did the USBUHCIL DESCR, and got this, (with the Joystick)

Code: Select all: USBUHCIL 0.08, (C) 2007-2009, Bret E. Johnson. DOS Driver for a Universal HCI compatible USB Host Controller. LITE version (maximum 16 Devices, no Isochronous Transactions). DEVICE ADDRESS: 2 CONFIGURATION: 1 DEVICE DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 12h 18 Descr Type: 01h Device USB Release: 0200h 2.00 Device Class: 00h Look at Interface Device SubClass: 00h None Device Protocol: 00h None Max Packet Size EP0: 08h 8 Vendor ID: D209h No Matching Record Found Product ID: 1501h 5,377 Device Release: 0019h 0.19 Manufacturer String: 01h "Universal Herface Device" Product String: 01h "Univers Interface Device" Serial # String: 05h "9" # of Configs: 01h 1 CONFIGURATION DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 02h Configuration Descr Total Length: 006Dh 109 # of Interfaces: 04h 4 Configuration Value: 01h 1 Configuration String: 01h "Universal Human Interfaee D" Attributes: 80h Valid Historical Attribute 80h Bus Powered No Remote Wakeup Not Battery Powered Current Draw / 2: FAh 250 (500 mA) INTERFACE DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 04h Interface Interface Number: 00h 0 Alternate Setting: 00h 0 # of End Points: 01h 1 Interface Class: 03h HID (Human Interface Device) Interface SubClass: 00h None Interface Protocol: 00h None Interface String: 01h "Universal Human Interface Device" HUMAN INTERFACE DEVICE DESCR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 21h HID (Human Interface Device) HID Class Release: 0111h 1.11 Country Code: 00h None/Not Applicable # of Classes: 01h 1 Class Descr Type 1: 22h Report Class Descr Size 1: 0021h 33 END POINT DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 07h 7 Descr Type: 05h EndPoint End Point Number: 81h EndPoint = 1 Direction = In Attributes: 03h Interrupt NonSynchronous Data Max Packet Size: 0008h 8 Max Poll Interval: 0Ah 10 INTERFACE DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 04h Interface Interface Number: 01h 1 Alternate Setting: 00h 0 # of End Points: 01h 1 Interface Class: 03h HID (Human Interface Device) Interface SubClass: 00h None Interface Protocol: 00h None Interface String: 03h "UHID Gamepad Device #1" HUMAN INTERFACE DEVICE DESCR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 21h HID (Human Interface Device) HID Class Release: 0111h 1.11 Country Code: 00h None/Not Applicable # of Classes: 01h 1 Class Descr Type 1: 22h Report Class Descr Size 1: 0046h 70 END POINT DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 07h 7 Descr Type: 05h EndPoint End Point Number: 82h EndPoint = 2 Direction = In Attributes: 03h Interrupt NonSynchronous Data Max Packet Size: 0014h 20 Max Poll Interval: 0Ah 10 INTERFACE DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 04h Interface Interface Number: 02h 2 Alternate Setting: 00h 0 # of End Points: 01h 1 Interface Class: 03h HID (Human Interface Device) Interface SubClass: 00h None Interface Protocol: 00h None Interface String: 04h "UHID Keyboard Device" HUMAN INTERFACE DEVICE DESCR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 21h HID (Human Interface Device) HID Class Release: 0111h 1.11 Country Code: 00h None/Not Applicable # of Classes: 01h 1 Class Descr Type 1: 22h Report Class Descr Size 1: 003Fh 63 END POINT DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 07h 7 Descr Type: 05h EndPoint End Point Number: 83h EndPoint = 3 Direction = In Attributes: 03h Interrupt NonSynchronous Data Max Packet Size: 0010h 16 Max Poll Interval: 0Ah 10 INTERFACE DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 04h Interface Interface Number: 03h 3 Alternate Setting: 00h 0 # of End Points: 01h 1 Interface Class: 03h HID (Human Interface Device) Interface SubClass: 01h Boot Interface Protocol: 02h Mouse Interface String: 02h "UHID Mouse Device" HUMAN INTERFACE DEVICE DESCR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 21h HID (Human Interface Device) HID Class Release: 0111h 1.11 Country Code: 00h None/Not Applicable # of Classes: 01h 1 Class Descr Type 1: 22h Report Class Descr Size 1: 0036h 54 END POINT DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 07h 7 Descr Type: 05h EndPoint End Point Number: 84h EndPoint = 4 Direction = In Attributes: 03h Interrupt NonSynchronous Data Max Packet Size: 0008h 8 Max Poll Interval: 0Ah 10

However I have no idea what you mean with USBUHCIL REPORT 2 0, I have searched through the doc's and cant find any reference to these switches.
Typing in any configuration of USBUHCIL REPORT creates an error.

Regards

by **Bret** » Sat Aug 01, 2009 6:06 am

The IRQ's on the industrial board are pre-assigned to various functions, ie: Com1:,Floppy etc.
When I enable all devices, the only IRQ's that will work are 9,10 & 11
The others, although they show as being enabled, create an error with usbuhcil.
I think the n/a actually means "Not Assigned", and therefore these are available for use.

The n/a in the COMMON USE column means "not applicable", indicating that there is no "common use" for that IRQ. It is simply an "extra" IRQ that can be used for almost anything.

The column that determines whether the IRQ is actually doing anything at the moment is the ENABLED column. On your computer, IRQ's 9, 10, & 11 are enabled which means that there is already some kind of hardware already attached to that IRQ and the hardware is "enabled". If the ENABLED column is blank (does not say "Yes"), there could still be hardware attached to the IRQ, but it is not currently enabled or not currently using the IRQ. For example, many motherboards assign an IRQ to the video card, but it never gets enabled because the IRQ of the video card is almost never useful for anything. WIth COM (serial) ports, the IRQ is not enabled until there is a program that actually needs to use the COM port for something, and the program enables and disables the COM port while the program is running.

The only IRQ's that are SUPPOSED to be dedicated/hardwired are the ones with the "*" around the COMMON USE description. Apparently, that is not the case on your particular computer.

However I have no idea what you mean with USBUHCIL REPORT 2 0, I have searched through the doc's and cant find any reference to these switches.
Typing in any configuration of USBUHCIL REPORT creates an error.

I screwed up the documentation -- sorry about that. The REPORT option is discussed on page 79 of USBINTRO.DOC, but the two example lines near the top of the page that say:

USBUHCI [2-127} I#
USBUHCIL [2-17} I#

should actually say:

USBUHCI Report [2-127} I#
USBUHCIL Report [2-17} I#

I'll fix that and upload a new doc this weekend.

Nonetheless, "USBUHCIL Report 2 0" should work, as should the other incantations (2 1, 2 2, and 2 3). Exactly what kind of error message are you seeing when you do that? I will need to see those reports to figure out if the U-HID is issuing reports that USBJSTIK and USBMOUSE can understand. Also, I don't know what kind of rapport you have with the U-HID vendor, but you can tell them that the third Interface (Interface #2) should declare itself to be a keyboard rather than simply a Human Interface Device, or it will not be recognized by USBKEYB.

******************

There are a couple of more things we can try to further isolate the problem.

If you have a USB keyboard or "regular' USB joystick (something other than the experimental U-HID), try using those with USBKEYB or USBJSTIK and see what happens. This should help further isolate the problem to either USBMOUSE or USBUHCIL. If you're using USBKEYB or USBJSTIK instead of USBMOUSE, you should also be able to leave the IRQ at the default setting (#12).

Also, if you can, see what happens if you use another version of DOS (MS-DOS or PC-DOS). I don't normally use FreeDOS, but simply have it on a test system on which I do minimalistic tests (I don't do extensive testing on anything except MS-DOS, since it is the "standard" to which all other DOS's must attempt to comply). I use USB mice, keyboards, and disk drives for several hours at a time on my systems running MS-DOS, and never have problems like you're describing.

I apologize for this being so time-consuming on your part, but I hope you understand that you are actually on the "bleeding edge" of technology, at least in the DOS world. You're trying to do things that nobody else has ever done before, and which many people thought could never be done.

Bret

by **Dinosaur** » Sat Aug 01, 2009 10:54 am

Hi Bret

Just a quick reply
With only the mouse plugged in.
=============================
Your statement reads
USBUHCIL space REPORT space 2 point 0
so typing in
USBUHCIL REPORT 2.0
gets me
USBUHCIL REPORT 2.0
^ ERROR

If I simply type without 2.0 then I get
USBUHCIL REPORT
^ ERROR

With only Joystick Plugged in
========================
Same result as above.

Regards

Edit : the ^ is under the 2.0 or under the t of report
Same Result with USB Keyboard installed. I dont have any other usb devices (other than memory sticks)
Are you using any sort of dos extender such as CWSDPMI ?

Edit2: I have set up a test with keyboard only, so
Cwsdpmi
usbuhcil
usbprint
usbkeyb
After 1.5 Hour the same problem, so must be in cpu board or usbuhcil

by **Dinosaur** » Sat Aug 01, 2009 6:59 pm

Hi Bret

Also, I don't know what kind of rapport you have with the U-HID vendor, but you can tell them that the third Interface (Interface #2) should declare itself to be a keyboard rather than simply a Human Interface Device, or it will not be recognized by USBKEYB.

Are you saying that their code in the adapter is reporting the wrong name for #2 ?

I apologize for this being so time-consuming on your part, but I hope you understand that you are actually on the "bleeding edge" of technology, at least in the DOS world. You're trying to do things that nobody else has ever done before, and which many people thought could never be done.

No need to apologise, my interest are purely selfish. I want to see you succeed in providing the Dos community with a usb interface. I will do whatever it takes (Time permitting) to help achieve that.

I can write some test routines to test the operation, but you need to advise me.
One suspect is Memory bleed. So I wrote a test that printed the Free memory every minute, but with the keyboard only, it stayed constant
untill the error.

Regards

by **Bret** » Sun Aug 02, 2009 2:23 am

Your statement reads
USBUHCIL space REPORT space 2 point 0
so typing in
USBUHCIL REPORT 2.0
gets me
USBUHCIL REPORT 2.0
^ ERROR

Sorry if I did that. It should be:

USBUHCIL {space} REPORT {space} 2 {space} 0

The 2 is the Device Address, and the 0 is the Interface Number. These can both be gleaned from the USBUHCIL AddressData output. The U-HID has a total of four interfaces, numbered 0-3, and each has its own HID Report that can be downloaded. Sorry for the confusion.

Are you using any sort of dos extender such as CWSDPMI ?

No, this is all pure DOS. The only "memory extension" that is involved is EMM386, or it's equivalent.

Are you saying that their code in the adapter is reporting the wrong name for #2 ?

It's not the name that's wrong. The name that's provided is simply a descriptor string provided by the manufacturer, and means absolutely nothing when it comes to the device operation. The problem is in the rest of the Interface Descriptor:

Code: Select all
INTERFACE DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 04h Interface Interface Number: 02h 2 Alternate Setting: 00h 0 # of End Points: 01h 1 Interface Class: 03h HID (Human Interface Device) Interface SubClass: 00h None Interface Protocol: 00h None Interface String: 04h "UHID Keyboard Device"

For a keyboard, the Interface SubClass is supposed to be 1 (not 0), and the Interface Protocol is also supposed to be 1 (not 0).

One suspect is Memory bleed. So I wrote a test that printed the Free memory every minute, but with the keyboard only, it stayed constant untill the error.

I'm not all that familiar with FreeDOS, but getting an "invalid opcode" error definitely sounds like a memory leak or stack corruption problem. Again, if it's possible, I would like to see if this happens with a different version of DOS (MS-DOS, PC-DOS, DR-DOS, etc.). Also, try increasing your stacks and/or stack size with the STACKS statement in CONFIG.SYS. Try something like STACKS 9,384 and see if it helps.

by **Dinosaur** » Sun Aug 02, 2009 10:43 am

Hi Bret

Have started the same test with a Win95 Dos disk.
However, I have to say that if you dont test the drivers under cwsdpmi or something similar
you are going to severely restrict the user base.
I have noticed that when searching for Dos items, in the last year or so, there has been a huge
increase in the hits from people looking for "Ms-Dos" solutions to networking. BUT, these mainly refer
to a Dos console in Windows, not pure MS-Dos.
The next biggest user base appears to be protected mode dos users (me included).

Anyway, we will see. Perhaps the problems solved here will solve it for protected mode dos or FreeDos.
Here is result of report

Code: Select all: USB HID (HUMAN INTERFACE DEVICE) REPORT DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ 05 01 Usage Page: Generic Desktop 09 02 Usage: Mouse A1 01 Collection: Application 09 01 Usage: Pointer A1 00 Collection: Physical 05 09 Usage Page: Buttons 19 01 Usage Min: Button #1 (Primary) 29 03 Usage Max: Button #3 (Tertiary) 15 00 Logical Min: 00h 0 0 25 01 Logical Max: 01h 1 +1 75 01 Report Size: 1 95 03 Report Count: 3 81 02 INPUT: Data Variable Absolute No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field 75 05 Report Size: 5 95 01 Report Count: 1 81 01 INPUT: Constant Array Absolute No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field 05 01 Usage Page: Generic Desktop 09 30 Usage: X-axis 09 31 Usage: Y-axis 09 38 Usage: Wheel 15 81 Logical Min: 81h 129 -127 25 7F Logical Max: 7Fh 127 +127 75 08 Report Size: 8 95 03 Report Count: 3 81 06 INPUT: Data Variable Relative No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field C0 End Collection 05 FF Usage Page: Vendor Specific 09 02 Usage: 0002h 15 00 Logical Min: 00h 0 0 25 01 Logical Max: 01h 1 +1 75 01 Report Size: 1 95 01 Report Count: 1 B1 22 FEATURE: Data Variable Absolute No_Wrap Linear No_Preferred_State No_Null_State Non_Volatile Bit_Field 75 07 Report Size: 7 95 01 Report Count: 1 B1 01 FEATURE: Constant Array Absolute No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field C0 End Collection

Regards
PS. This report is for Mouse only, when it's behaviour is known, I will test with Joystick.

by **Dinosaur** » Sun Aug 02, 2009 4:13 pm

Hi Bret

Good News. Using the following config the system ran for 5 hours without a problem.
The dos is Win95.

cwsdpmi
cd\usb
usbuhcil
usbmouse
cd\
Filedemo

The filedemo basically prints the Free Mem every minute. The report on Joystick below
Report 2 0

Code: Select all: USB HID (HUMAN INTERFACE DEVICE) REPORT DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ 05 01 Usage Page: Generic Desktop 09 00 Usage: None/Unknown/Not Applicable A1 01 Collection: Application 05 09 Usage Page: Buttons 19 01 Usage Min: Button #1 (Primary) 29 20 Usage Max: 0020h 15 00 Logical Min: 00h 0 0 25 01 Logical Max: 01h 1 +1 95 20 Report Count: 32 75 01 Report Size: 1 81 02 INPUT: Data Variable Absolute No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field 05 08 Usage Page: LEDs (Light Emitting Diodes) 09 4B Usage: Generic Indicator 95 20 Report Count: 32 75 01 Report Size: 1 91 20 OUTPUT: Data Array Absolute No_Wrap Linear No_Preferred_State No_Null_State Non_Volatile Bit_Field C0 End Collection

Report 2 1

Code: Select all: USB HID (HUMAN INTERFACE DEVICE) REPORT DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ 05 01 Usage Page: Generic Desktop 09 04 Usage: Joystick A1 01 Collection: Application 09 01 Usage: Pointer A1 00 Collection: Physical 05 01 Usage Page: Generic Desktop 09 30 Usage: X-axis 09 31 Usage: Y-axis 09 32 Usage: Z-axis 09 33 Usage: Rotation around X-axis 09 34 Usage: Rotation around Y-axis 09 35 Usage: Rotation around Z-axis 09 36 Usage: Slider 09 37 Usage: Dial 16 00 F8 Logical Min: F800h 63,488 -2,048 26 FF 07 Logical Max: 07FFh 2,047 +2,047 36 00 F8 Physical Min: F800h 63,488 -2,048 46 FF 07 Physical Max: 07FFh 2,047 +2,047 65 00 Units: 00h 75 10 Report Size: 16 95 08 Report Count: 8 81 02 INPUT: Data Variable Absolute No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field C0 End Collection 05 09 Usage Page: Buttons 19 01 Usage Min: Button #1 (Primary) 29 20 Usage Max: 0020h 15 00 Logical Min: F800h 63,488 -2,048 25 01 Logical Max: 0701h 1,793 +1,793 35 00 Physical Min: F800h 63,488 -2,048 45 01 Physical Max: 0701h 1,793 +1,793 95 20 Report Count: 32 75 01 Report Size: 1 81 02 INPUT: Data Variable Absolute No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field C0 End Collection

Report 2 2

Code: Select all: USB HID (HUMAN INTERFACE DEVICE) REPORT DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ 05 01 Usage Page: Generic Desktop 09 06 Usage: Keyboard A1 01 Collection: Application 05 07 Usage Page: Keyboard/Keypad 19 E0 Usage Min: Left Control 29 E7 Usage Max: Right Windows (Right GUI) 15 00 Logical Min: 00h 0 0 25 01 Logical Max: 01h 1 +1 75 01 Report Size: 1 95 08 Report Count: 8 81 02 INPUT: Data Variable Absolute No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field 95 01 Report Count: 1 75 08 Report Size: 8 81 01 INPUT: Constant Array Absolute No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field 95 05 Report Count: 5 75 01 Report Size: 1 05 08 Usage Page: LEDs (Light Emitting Diodes) 19 01 Usage Min: Num Lock 29 05 Usage Max: Kana 91 02 OUTPUT: Data Variable Absolute No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field 95 01 Report Count: 1 75 03 Report Size: 3 91 01 OUTPUT: Constant Array Absolute No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field 95 0E Report Count: 14 75 08 Report Size: 8 15 00 Logical Min: 00h 0 0 25 65 Logical Max: 65h 101 +101 05 07 Usage Page: Keyboard/Keypad 19 00 Usage Min: None 29 65 Usage Max: Mouse Menu (Application) 81 00 INPUT: Data Array Absolute No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field C0 End Collection

Report 2 3

Code: Select all: USB HID (HUMAN INTERFACE DEVICE) REPORT DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ 05 01 Usage Page: Generic Desktop 09 02 Usage: Mouse A1 01 Collection: Application 09 01 Usage: Pointer A1 00 Collection: Physical 05 09 Usage Page: Buttons 19 01 Usage Min: Button #1 (Primary) 29 03 Usage Max: Button #3 (Tertiary) 15 00 Logical Min: 00h 0 0 25 01 Logical Max: 01h 1 +1 95 03 Report Count: 3 75 01 Report Size: 1 81 02 INPUT: Data Variable Absolute No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field 95 01 Report Count: 1 75 05 Report Size: 5 81 01 INPUT: Constant Array Absolute No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field 05 01 Usage Page: Generic Desktop 09 30 Usage: X-axis 09 31 Usage: Y-axis 09 32 Usage: Z-axis 16 00 80 Logical Min: 8000h 32,768 -32,768 26 FF 7F Logical Max: 7FFFh 32,767 +32,767 75 10 Report Size: 16 95 03 Report Count: 3 81 06 INPUT: Data Variable Relative No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field C0 End Collection C0 End Collection

Regards

by **Bret** » Mon Aug 03, 2009 3:28 am

However, I have to say that if you dont test the drivers under cwsdpmi or something similar you are going to severely restrict the user base. I have noticed that when searching for Dos items, in the last year or so, there has been a huge increase in the hits from people looking for "Ms-Dos" solutions to networking. BUT, these mainly refer to a Dos console in Windows, not pure MS-Dos. The next biggest user base appears to be protected mode dos users (me included).

Anyway, we will see. Perhaps the problems solved here will solve it for protected mode dos or FreeDos.

Although my programs do not directly use protected mode, they can work with programs running in protected mode (as you have successfully proven). The thing is, my API is a real mode API, and there is no protected mode API. So, as long as the protected mode program uses real-mode BIOS/DOS calls to access the hardware, it shouldn't be a problem. The issue is protected-mode programs that try to access hardware directly. A lot of programs do this because it's "easier" than trying to use the BIOS and DOS calls it's supposed to use, but doing so means the program will NEVER be able to use newer hardware. Although USB is still really popular, in a few years it wouldn't surprise me a bit if some new bus architecture came out that replaced, or at least augmented, USB (FireWire and Bluetooth are already out there). You program yourself into obsolescence if you don't use BIOS- or DOS-level calls when they are available.

For example, I've actually seen programmers requesting help on forums asking how to use BIOS or DOS calls to access a PS2 mouse from a protected mode program. The recommendations that usually come back are to not use DOS or BIOS calls, but to access the PS2 hardware directly because it's "easier" to code. That means the program will never be able to work correctly on a computer that has a serial mouse, bus mouse, USB mouse, etc. That is absolutely the WRONG thing to do, even though it's "easier".

Good News. Using the following config the system ran for 5 hours without a problem.
The dos is Win95.

I ran some tests over the weekend as well, with DR-DOS 7.03, PC-DOS 5.00, and PC-DOS 7.01, all on the exact same computer. I ran each of them for at least two hours, with no problems at all. I regularly use MS-DOS 6.20 and 7.10 for many hours at a time, so I didn't need to run special tests with those.

When I used FreeDOS, though, it would crash. It usually took less than a half hour, sometimes only a few minutes. Sometimes it would just stop, and sometimes it would generate an "invalid opcode" error. So, the problem appears to be very specific to FreeDOS, as your tests have confirmed as well.

Trying to isolate the problem a little further, I tried installing just USBUHCIL (not USBMOUSE) with FreeDOS, and it did not crash, nor did it crash with USBUHCIL & USBHUB. It only seemed to crash when USBUHCIL & USBMOUSE were installed at the same time. As a side note, this was occurring with NO memory manager installed (HIMEM/EMM386), so it's definitely something in the FreeDOS kernel or command interpreter. I think this is going to be an ugly, time-consuming problem to track down.

I'll look at the Report Descriptors and see what I discover.

by **Dinosaur** » Mon Aug 03, 2009 10:05 am

Hi Bret

Thank you for that thorough explanation. Luckily my combination of
FreeBasic, Cgui ,Allegro uses Bios / Dos calls.

I originally tried to test the FreeDos with no config.sys entries at all, (I never use them) however
to lay this baby to rest, I tried with the same settings as MS-Dos, but still it failed.

The FreeDos development community will be interested in these results, but I dont know if & what is needed for them to look into it.
I for one do not see any difference in performance between MS-Dos and FreeDos. And as I am legally allowed to make as many bootable
disks as I need I will continue to use Win95 Dos. So I wont follow this issue any further.

With reference to the joystick, I have made an agreement with the supplier, where he will create standard mouse emulation
in the U-HID Nano board.Googling for weeks did not show up any boards that allowed an industrial joystick to be interfaced to the usb port with mouse emulation.
So this is the best solution. I also asked him to read this thread and he replied, which is below. This is not intended as anything other then for your information.

The reason the interface is not defined as a boot interface subclass (1) is that the keyboard HID desccriptor does not follow the requirements for a boot device. The requirements for a boot device are a standard descriptor of one modifier byte, one padding byte and 6 keycode bytes.
The U-HID uses a longer format to enable more simultaneous keys to be pressed. Windows and MacOS are fine with this (we did extensive testing) but it departs from the specification of a boot device. Actually Windows makes no destinction between boot and non-boot devices as far as I am aware, but PC BIOS might have an issue with a device defined as a boot device but departs from the convention. We didnt test BIOS support extensively though and it might be no different if defined as a boot protocol device.

As of tomorrow, I will be away from the office and so no test rig within reach, however I will contact you when I get the new boards. Many thanks for your help.

Regards

by **Bret** » Mon Aug 03, 2009 11:33 am

The FreeDos development community will be interested in these results, but I dont know if & what is needed for them to look into it. I for one do not see any difference in performance between MS-Dos and FreeDos. And as I am legally allowed to make as many bootable disks as I need I will continue to use Win95 Dos. So I wont follow this issue any further.

FYI, I have e-mailed the FreeDOS kernel developers list, but they have not responded yet. As stated before, it will likely take awhile to even figure out where the problem is. It's really hard to troubleshoot when you can't figure out a way to "trigger" something to happen when you want it to. I have not posted anything on the FreeDOS users list, though that might be appropriate as well. I'll wait to see what happens with the kernel developers list first.

You can have the U-HID developers contact me if you want. They don't seem to understand how the boot protocol is supposed to work with the keyboard (which also makes me wonder if the boot protocol for the mouse really works like it should), and they also have the joystick Report Descriptor messed up. USBKEYB would not work with their U-HID device the way it is now (because of the simultaneous keypresses thing), but USBKEYB could be changed without too much difficulty to accommodate that if the need arose. Also FYI, the standard USB keyboard handler allows up to 6 "regular" keys to be pressed at the same time, in addition to any or all of the 8 "shift" keys (Shift, Alt, Control, and Windows). The U-HID allows up to 14 "regular" keys to be pressed in addition to the "Shift" keys.

by **Dinosaur** » Mon Aug 03, 2009 8:52 pm

Hi Bret

When the FreeDos community thinks it has isolated the culprit, I will test it again.
One side affect from MS-dos, is that the LFN doesnt work, so will want to go back to FreeDos.

I dont know if anyone else has this problem with your site, but it is painfully slow.
If I press the reply button, the screen changes fairly quickly, but the last 10% on the progress bar
in FireFox can take up to 30 sec's.

Regards

by **Dinosaur** » Tue Aug 04, 2009 9:57 am

Hi Bret

Can you send Andy at Ultimarc dot com , an email address that you two can communicate privately
instead of on this forum.Mention Jan van de Poll from Compu-Weigh and he will understand.

This way, there will be a board on the market that will allow an analog joystick to appear like a standard
mouse, and should work using your drivers.

Regards

by **Bret** » Wed Aug 05, 2009 2:13 am

I just sent Andy him an e-mail. Thanks, Jan.

by **Bret** » Sat Aug 08, 2009 1:41 pm

I just uploaded a new version of USBMOUSE which will hopefully solve the problem with FreeDOS crashing. I tested it on my computers, and was able to run for several hours with no problem. I also uploaded some updates to some of the other programs.

Please test it and let me know if you still have problems.

by **Bret** » Sun Aug 09, 2009 1:42 am

I dont know if anyone else has this problem with your site, but it is painfully slow. If I press the reply button, the screen changes fairly quickly, but the last 10% on the progress bar in FireFox can take up to 30 sec's.

Is that on the web site or in the forum? Nobody else has ever said anything, and I don't have that problem from where I'm at. If you don't have a high-speed internet connection, I think the forum probably could take a long time to come up, but I'm not sure what (if anything) could be done about that. The web site should be very fast, though -- it's just plain text with no fancy colors or backgrounds or pictures or java scripts or anything that would slow it down.

by **Dinosaur** » Fri Aug 14, 2009 9:29 am

Hi Bret

Sorry for the lack of response, I have been away from the ofice.
The slowdown on the forum is on my Desktop only, and still have not resolved it.
The same problem happened on the FreeBasic Forum, but no where else ????

I have had a response from the FreeDos forum by Eric Aur (spelling ?) but before responding,
thought I would ask you what responses (if any) you have had raising the issues we have discovered ?

Regards

by **Bret** » Fri Aug 14, 2009 9:50 am

I believe I received the same e-mail from Eric, and have already responded to him. He knows what the problem is. It turns out that the way FreeDOS implements STACKS (in CONFIG.SYS) is different than the way MS (and DR and IBM and ...) do it, and it causes problems with certain programs. Eric's solution is to set STACKS=0,0 in CONFIG.SYS, which effectively turns STACKS completely off. I've told him I think they should actually fix STACKS in FreeDOS so it works correctly, instead of forcing the user to change something in CONFIG.SYS. It's debatable whether STACKS really does anything useful or not, but in any case the onus should not be on the user to disable something that doesn't work like it should.

I have patched the latest versions of USBMOUSE & USBMOUSE so that they work with FreeDOS even if you don't set STACKS=0,0, so you should be OK either way. I don't know if they're going to "fix" FreeDOS or not.

by **Dinosaur** » Fri Aug 14, 2009 6:39 pm

Hi Bret

Will leave testing of FreeDos until I have MS-Dos under control.
Have today received my new interface board from Andy and have hooked it up.
So, I now have an Industrial Joystick that behaves like a mouse.
In Windows it works as I expected it to.
However in ms-dos I have no response all.
With my previous settings an ordinary Microsoft Mouse works.
I have downloaded the latest from your site, but to no avail.
Below the reports.

Code: Select all: USBDEVIC 0.05, (C) 2008, Bret E. Johnson. Program to display information about Devices attached to the USB Host(s). DEVICE ADDRESSES ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Host Index: 0 Host Type: UHCI Bus Type: PCI IRQ#: 10 Root Hub Ports: 2 Vendor: 1106h = VIA Technologies Inc Product: 3038h ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ DEVICES INTERFACES ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ L C I A O ADRS o P o n l w ÍÍÍÍ (hex) S o BUS n t t n Test VEND PROD Sub Pro p USB HUB r POWR f f I e Sub Pro RWak ID ID Cls Cls col d VER ADR t (mA) g c n DESCRIPTION d Cls Cls col ÍÍÍÍ ÍÍÍÍ ÍÍÍÍ ÍÍÍ ÍÍÍ ÍÍÍ Í ÍÍÍ ÍÍÍ Í ÍÍÍÍ Í Í Í ÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Í ÍÍÍ ÍÍÍ ÍÍÍ 1 1106 3038 9 0 0 . 1.0 ... . s 0 1 0 0*Root Hub Y 9 0 0 VIA Technologies Inc ÄÄÄÄ ÄÄÄÄ ÄÄÄÄ ÄÄÄ ÄÄÄ ÄÄÄ Ä ÄÄÄ ÄÄÄ Ä ÄÄÄÄ Ä Ä Ä ÄÄÄÄÄÄÄÄÄÄÄÄÄÄ Ä ÄÄÄ ÄÄÄ ÄÄÄ 2R 045E 0053 0 0 0 Y 1.1 1 1 100 1 0 0*Mouse Y 3 1 2 Microsoft Corp

Report 2 0

Code: Select all: USBUHCIL 0.09, (C) 2007-2009, Bret E. Johnson. DOS Driver for a Universal HCI compatible USB Host Controller. LITE version (maximum 16 Devices, no Isochronous Transactions). USB HID (HUMAN INTERFACE DEVICE) REPORT DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ 05 01 Usage Page: Generic Desktop 09 02 Usage: Mouse A1 01 Collection: Application 09 01 Usage: Pointer A1 00 Collection: Physical 05 09 Usage Page: Buttons 19 01 Usage Min: Button #1 (Primary) 29 03 Usage Max: Button #3 (Tertiary) 15 00 Logical Min: 00h 0 0 25 01 Logical Max: 01h 1 +1 75 01 Report Size: 1 95 03 Report Count: 3 81 02 INPUT: Data Variable Absolute No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field 75 05 Report Size: 5 95 01 Report Count: 1 81 01 INPUT: Constant Array Absolute No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field 05 01 Usage Page: Generic Desktop 09 30 Usage: X-axis 09 31 Usage: Y-axis 09 38 Usage: Wheel 15 81 Logical Min: 81h 129 -127 25 7F Logical Max: 7Fh 127 +127 75 08 Report Size: 8 95 03 Report Count: 3 81 06 INPUT: Data Variable Relative No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field C0 End Collection 05 FF Usage Page: Vendor Specific 09 02 Usage: 0002h 15 00 Logical Min: 00h 0 0 25 01 Logical Max: 01h 1 +1 75 01 Report Size: 1 95 01 Report Count: 1 B1 22 FEATURE: Data Variable Absolute No_Wrap Linear No_Preferred_State No_Null_State Non_Volatile Bit_Field 75 07 Report Size: 7 95 01 Report Count: 1 B1 01 FEATURE: Constant Array Absolute No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field C0 End Collection

Descr

Code: Select all: USBUHCIL 0.09, (C) 2007-2009, Bret E. Johnson. DOS Driver for a Universal HCI compatible USB Host Controller. LITE version (maximum 16 Devices, no Isochronous Transactions). DEVICE ADDRESS: 2 CONFIGURATION: 1 DEVICE DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 12h 18 Descr Type: 01h Device USB Release: 0200h 2.00 Device Class: 00h Look at Interface Device SubClass: 00h None Device Protocol: 00h None Max Packet Size EP0: 08h 8 Vendor ID: D209h Ultimarc? Product ID: 1501h 5,377 Device Release: 0020h 0.20 Manufacturer String: 01h "l Human Interfaee D" Product String: 01h "Universal Human Interface Device" Serial # String: 05h "9" # of Configs: 01h 1 CONFIGURATION DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 02h Configuration Descr Total Length: 006Dh 109 # of Interfaces: 04h 4 Configuration Value: 01h 1 Configuration String: 01h "Universal Human Interface Device" Attributes: 80h Valid Historical Attribute 80h Bus Powered No Remote Wakeup Not Battery Powered Current Draw / 2: FAh 250 (500 mA) INTERFACE DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 04h Interface Interface Number: 00h 0 Alternate Setting: 00h 0 # of End Points: 01h 1 Interface Class: 03h HID (Human Interface Device) Interface SubClass: 00h None Interface Protocol: 00h None Interface String: 01h "Universal Human Interface Device" HUMAN INTERFACE DEVICE DESCR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 21h HID (Human Interface Device) HID Class Release: 0111h 1.11 Country Code: 00h None/Not Applicable # of Classes: 01h 1 Class Descr Type 1: 22h Report Class Descr Size 1: 0021h 33 END POINT DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 07h 7 Descr Type: 05h EndPoint End Point Number: 81h EndPoint = 1 Direction = In Attributes: 03h Interrupt NonSynchronous Data Max Packet Size: 0008h 8 Max Poll Interval: 0Ah 10 INTERFACE DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 04h Interface Interface Number: 01h 1 Alternate Setting: 00h 0 # of End Points: 01h 1 Interface Class: 03h HID (Human Interface Device) Interface SubClass: 00h None Interface Protocol: 00h None Interface String: 03h "UHID Gamepad Device #1" HUMAN INTERFACE DEVICE DESCR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 21h HID (Human Interface Device) HID Class Release: 0111h 1.11 Country Code: 00h None/Not Applicable # of Classes: 01h 1 Class Descr Type 1: 22h Report Class Descr Size 1: 0046h 70 END POINT DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 07h 7 Descr Type: 05h EndPoint End Point Number: 82h EndPoint = 2 Direction = In Attributes: 03h Interrupt NonSynchronous Data Max Packet Size: 0014h 20 Max Poll Interval: 0Ah 10 INTERFACE DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 04h Interface Interface Number: 02h 2 Alternate Setting: 00h 0 # of End Points: 01h 1 Interface Class: 03h HID (Human Interface Device) Interface SubClass: 00h None Interface Protocol: 00h None Interface String: 04h "UHID Keyboard Device" HUMAN INTERFACE DEVICE DESCR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 21h HID (Human Interface Device) HID Class Release: 0111h 1.11 Country Code: 00h None/Not Applicable # of Classes: 01h 1 Class Descr Type 1: 22h Report Class Descr Size 1: 003Fh 63 END POINT DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 07h 7 Descr Type: 05h EndPoint End Point Number: 83h EndPoint = 3 Direction = In Attributes: 03h Interrupt NonSynchronous Data Max Packet Size: 0010h 16 Max Poll Interval: 0Ah 10 INTERFACE DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 04h Interface Interface Number: 03h 3 Alternate Setting: 00h 0 # of End Points: 01h 1 Interface Class: 03h HID (Human Interface Device) Interface SubClass: 01h Boot Interface Protocol: 02h Mouse Interface String: 02h "UHID Mouse Device" HUMAN INTERFACE DEVICE DESCR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 09h 9 Descr Type: 21h HID (Human Interface Device) HID Class Release: 0111h 1.11 Country Code: 00h None/Not Applicable # of Classes: 01h 1 Class Descr Type 1: 22h Report Class Descr Size 1: 0036h 54 END POINT DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Descr Length: 07h 7 Descr Type: 05h EndPoint End Point Number: 84h EndPoint = 4 Direction = In Attributes: 03h Interrupt NonSynchronous Data Max Packet Size: 0008h 8 Max Poll Interval: 0Ah 10

Please tell me what else I can do to identify the problem for you.(and me)

Regards

by **Bret** » Sat Aug 15, 2009 4:27 pm

The first two reports (USBDEVIC & USBUHCIL Report 2 0) look like they are for the Microsoft mouse, not the UHID mouse. I'm slightly confused by what you mean by "With my previous settings an ordinary Microsoft Mouse works." I assume the Microsoft mouse still works OK based on what I see in the USBDEVIC output (a "Y" in the Owned column).

Please send "USBDEVIC", "USBUHCIL Report 2 3", and "USBMOUSE Bad" outputs with the new U-HID device attached.

by **Dinosaur** » Sat Aug 15, 2009 4:54 pm

Hi Bret

Sorry, I had both connected.And yes the MOuse still works perfectly.
Device

Code: Select all: USBDEVIC 0.05, (C) 2008, Bret E. Johnson. Program to display information about Devices attached to the USB Host(s). DEVICE ADDRESSES ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Host Index: 0 Host Type: UHCI Bus Type: PCI IRQ#: 10 Root Hub Ports: 2 Vendor: 1106h = VIA Technologies Inc Product: 3038h ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ DEVICES INTERFACES ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ L C I A O ADRS o P o n l w ÍÍÍÍ (hex) S o BUS n t t n Test VEND PROD Sub Pro p USB HUB r POWR f f I e Sub Pro RWak ID ID Cls Cls col d VER ADR t (mA) g c n DESCRIPTION d Cls Cls col ÍÍÍÍ ÍÍÍÍ ÍÍÍÍ ÍÍÍ ÍÍÍ ÍÍÍ Í ÍÍÍ ÍÍÍ Í ÍÍÍÍ Í Í Í ÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Í ÍÍÍ ÍÍÍ ÍÍÍ 1 1106 3038 9 0 0 . 1.0 ... . s 0 1 0 0*Root Hub Y 9 0 0 VIA Technologies Inc ÄÄÄÄ ÄÄÄÄ ÄÄÄÄ ÄÄÄ ÄÄÄ ÄÄÄ Ä ÄÄÄ ÄÄÄ Ä ÄÄÄÄ Ä Ä Ä ÄÄÄÄÄÄÄÄÄÄÄÄÄÄ Ä ÄÄÄ ÄÄÄ ÄÄÄ 2 D209 1501 0 0 0 . 2.0 1 1 500 1 0 0*Human Intf Dvc . 3 0 0 No Matching Record Found 1 0*Human Intf Dvc . 3 0 0 2 0*Human Intf Dvc . 3 0 0 3 0*Mouse . 3 1 2

Report

Code: Select all: USBUHCIL 0.08, (C) 2007-2009, Bret E. Johnson. DOS Driver for a Universal HCI compatible USB Host Controller. LITE version (maximum 16 Devices, no Isochronous Transactions). USB HID (HUMAN INTERFACE DEVICE) REPORT DESCRIPTOR ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ 05 01 Usage Page: Generic Desktop 09 02 Usage: Mouse A1 01 Collection: Application 09 01 Usage: Pointer A1 00 Collection: Physical 05 09 Usage Page: Buttons 19 01 Usage Min: Button #1 (Primary) 29 03 Usage Max: Button #3 (Tertiary) 15 00 Logical Min: 00h 0 0 25 01 Logical Max: 01h 1 +1 95 03 Report Count: 3 75 01 Report Size: 1 81 02 INPUT: Data Variable Absolute No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field 95 01 Report Count: 1 75 05 Report Size: 5 81 01 INPUT: Constant Array Absolute No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field 05 01 Usage Page: Generic Desktop 09 30 Usage: X-axis 09 31 Usage: Y-axis 09 32 Usage: Z-axis 16 00 80 Logical Min: 8000h 32,768 -32,768 26 FF 7F Logical Max: 7FFFh 32,767 +32,767 75 10 Report Size: 16 95 03 Report Count: 3 81 06 INPUT: Data Variable Relative No_Wrap Linear Preferred_State No_Null_State Non_Volatile Bit_Field C0 End Collection C0 End Collection

Mouse Bad

Code: Select all: USBMOUSE 0.08, (C) 2007-2009, Bret E. Johnson. DOS Driver for up to 8 Generic USB Mice. Beep for Bad Mice: Yes BAD (INCOMPATIBLE) MICE ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ HOST DVC INTF STAGE ERROR INDX ADDR NUM NUM CODE ÄÄÄÄ ÄÄÄÄ ÄÄÄÄ ÄÄÄÄÄ ÄÄÄÄÄ 0 2 3 210 0000h

Now that I see the statement : Beep for bad Mice, I am getting a non standard beep when booting up.

Regards

PS: I am happy for you to send me a test version which gives more detail, if that will help.
Also note that the hardware I have allows upgrade of software in it, so if you feel that something is amiss
I can email Andy and ask him to change it.

by **Bret** » Sun Aug 16, 2009 12:52 pm

I just uploaded a new version of USBMOUSE which will hopefully work OK for you now.

FYI, the problem was that the U-HID mouse declares the third "axis" to be a Z-Axis instead of a Wheel, which is technically incorrect for a mouse.

Also, based on the last several descriptors you sent, it looks like you either have two different versions of the USB drivers (in different directories, perhaps?) or were running from two different machines. The Microsoft Mouse Descriptors were obtained using newer versions of the Drivers than the U-HID Descriptors. I know it can be confusing with all of the updates that have been happening recently, and can also understand if you want to keep old drivers around in case a new version screws something up that used to work in an old version (which can certainly happen every once-in-awhile).

by **Dinosaur** » Sun Aug 16, 2009 1:38 pm

Hi Bret

Have tested and works, GREAT.

I will now complete the tests.
It is also worth noting that I have left a FreeDos test running for 24 Hours and it hasn't failed.
I did put Stacks=0,0 in the config file, just for safety.

With reference to the older versions, it happened due to me saving my CFC's as Winimage .ima files
So I re-wrote the the disks without updating the driver.
I dont like keeping older versions laying around.

Will report to you when testing is finished, in the mean time many thanks for your efforts.

Regards

by **Dinosaur** » Sun Aug 16, 2009 5:21 pm

Hi Bret

Interesting observation.
Eric at FreeDos noted that the loading of cwsdpmi does not need to
be loaded seperately. The FreeBasic program will load it automatically.
With my Autoexec.bat like this:

Code: Select all: Path C:\dos;C:\usb; C:\Dos\Cwsdpmi C:\Dos\DosLFN.com cd\usb usbuhcil usbprint usbmouse cd\ C:\Dos\ctmouse

There are nor errors, no Beeps
If on the other hand I remove the cwsdpmi line, then there is a beep.
Device

Code: Select all: USBDEVIC 0.05, (C) 2008, Bret E. Johnson. Program to display information about Devices attached to the USB Host(s). DEVICE ADDRESSES ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Host Index: 0 Host Type: UHCI Bus Type: PCI IRQ#: 10 Root Hub Ports: 2 Vendor: 1106h = VIA Technologies Inc Product: 3038h ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ DEVICES INTERFACES ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ L C I A O ADRS o P o n l w ÍÍÍÍ (hex) S o BUS n t t n Test VEND PROD Sub Pro p USB HUB r POWR f f I e Sub Pro RWak ID ID Cls Cls col d VER ADR t (mA) g c n DESCRIPTION d Cls Cls col ÍÍÍÍ ÍÍÍÍ ÍÍÍÍ ÍÍÍ ÍÍÍ ÍÍÍ Í ÍÍÍ ÍÍÍ Í ÍÍÍÍ Í Í Í ÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Í ÍÍÍ ÍÍÍ ÍÍÍ 1 1106 3038 9 0 0 . 1.0 ... . s 0 1 0 0*Root Hub Y 9 0 0 VIA Technologies Inc ÄÄÄÄ ÄÄÄÄ ÄÄÄÄ ÄÄÄ ÄÄÄ ÄÄÄ Ä ÄÄÄ ÄÄÄ Ä ÄÄÄÄ Ä Ä Ä ÄÄÄÄÄÄÄÄÄÄÄÄÄÄ Ä ÄÄÄ ÄÄÄ ÄÄÄ 2 D209 1501 0 0 0 . 2.0 1 1 500 1 0 0*Human Intf Dvc . 3 0 0 Ultimarc? 1 0*Human Intf Dvc . 3 0 0 2 0*Human Intf Dvc . 3 0 0 3 0*Mouse Y 3 1 2

Descr

Code: Select all: USBDEVIC 0.05, (C) 2008, Bret E. Johnson. Program to display information about Devices attached to the USB Host(s). DEVICE ADDRESSES ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Host Index: 0 Host Type: UHCI Bus Type: PCI IRQ#: 10 Root Hub Ports: 2 Vendor: 1106h = VIA Technologies Inc Product: 3038h ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ DEVICES INTERFACES ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ L C I A O ADRS o P o n l w ÍÍÍÍ (hex) S o BUS n t t n Test VEND PROD Sub Pro p USB HUB r POWR f f I e Sub Pro RWak ID ID Cls Cls col d VER ADR t (mA) g c n DESCRIPTION d Cls Cls col ÍÍÍÍ ÍÍÍÍ ÍÍÍÍ ÍÍÍ ÍÍÍ ÍÍÍ Í ÍÍÍ ÍÍÍ Í ÍÍÍÍ Í Í Í ÍÍÍÍÍÍÍÍÍÍÍÍÍÍ Í ÍÍÍ ÍÍÍ ÍÍÍ 1 1106 3038 9 0 0 . 1.0 ... . s 0 1 0 0*Root Hub Y 9 0 0 VIA Technologies Inc ÄÄÄÄ ÄÄÄÄ ÄÄÄÄ ÄÄÄ ÄÄÄ ÄÄÄ Ä ÄÄÄ ÄÄÄ Ä ÄÄÄÄ Ä Ä Ä ÄÄÄÄÄÄÄÄÄÄÄÄÄÄ Ä ÄÄÄ ÄÄÄ ÄÄÄ 2 D209 1501 0 0 0 . 2.0 1 1 500 1 0 0*Human Intf Dvc . 3 0 0 Ultimarc? 1 0*Human Intf Dvc . 3 0 0 2 0*Human Intf Dvc . 3 0 0 3 0*Mouse Y 3 1 2

Mouse Bad

Code: Select all: USBMOUSE 0.10, (C) 2007-2009, Bret E. Johnson. DOS Driver for up to 8 Generic USB Mice. Beep for Bad Mice: Yes BAD (INCOMPATIBLE) MICE ÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ HOST DVC INTF STAGE ERROR INDX ADDR NUM NUM CODE ÄÄÄÄ ÄÄÄÄ ÄÄÄÄ ÄÄÄÄÄ ÄÄÄÄÄ 0 2 0 210 0000h 0 2 1 70 0010h Short Packet 0 2 2 210 0000h

Regards
Edit: If I use FreeDos the "Problem" is reversed. Loading the cwsdpmi independently anywhere causes the "Short Packet"
Not loading the cwsdpmi allows FreeDos to operate correctly.???

Edit2:It happens both in Ms-Dos & FreeDos with the Stage Num changing to 50
The Joystick/Mouse still works correctly.
Conclusion: If I plug an ps2 mouse in at the same time, I dont get the error on either ms-dos or freedos.
As soon as I unplug the ps2, it may happen after powering up the second time.
Hope this helps

http://bretjohnson.us/forum

Joystick / Mouse issues

Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Re: Joystick / Mouse issues

Who is online