Free-Dap integration in cortex M55 based MCU

[AjjaiAravind]

Hi,

I've integrated the free-dap code with my MCU SDK. I m able to achieve debugging (gdb) using hid interface with openocd in linux OS. But when I try doing the same with a windows host, I m facing issues with my hid.

I m using hid from tinyusb stack. I m suspecting from the descriptor. But with the same descriptor Linux is working fine.
Could you please help with this if there is a known issue with windows on this.

Thanks,
Ajjai

[ataradov]

Different OSes have different requirements for the descriptors. You can check with any reference implementation (for example https://github.com/ataradov/free-dap/blob/master/platform/samd11/usb_descriptors.c), they contain a descriptor that works with all.

But note that there is a difference in a descriptor for FS and HS devices. What is your device exactly?

Also, it is not the same descriptor that is used for keyboard and mice. And Linux may not actually care about the descriptor in this case. If you have not modified it at all, it is likely wrong, since it is rare for standard stacks to have this exact descriptor.

[AjjaiAravind]

The descriptors look good. But it looks like the windows is not able to map the proper driver for the device. It is showing as composite device. But when I try to map it to the libusbk driver it shows as cmsis dap device (with some issue mark symbol) but still we face the issue (openocd says no cmsis dap device).

[AjjaiAravind]

My device is 480M HS. Missed to mention in the previous reply

[ataradov]

For HID version, it should not need any additional drivers, it should just be a standard HID device and Windows should know about it. What does your descriptor looks like?

[AjjaiAravind]

Bus 003 Device 010: ID cafe:4006 Synaptics, Inc SR100 CMSIS-DAP
Device Descriptor:
bLength 18
bDescriptorType 1
bcdUSB 2.00
bDeviceClass 0
bDeviceSubClass 2
bDeviceProtocol 1
bMaxPacketSize0 64
idVendor 0xcafe
idProduct 0x4006
bcdDevice 1.00
iManufacturer 1 Synaptics, Inc
iProduct 2 SR100 CMSIS-DAP
iSerial 3 01.00.00.00
bNumConfigurations 1
Configuration Descriptor:
bLength 9
bDescriptorType 2
wTotalLength 0x0029
bNumInterfaces 1
bConfigurationValue 1
iConfiguration 5 TinyUSB Device
bmAttributes 0xc0
Self Powered
MaxPower 500mA
Interface Descriptor:
bLength 9
bDescriptorType 4
bInterfaceNumber 0
bAlternateSetting 0
bNumEndpoints 2
bInterfaceClass 3 Human Interface Device
bInterfaceSubClass 0
bInterfaceProtocol 0
iInterface 4 CMSIS-DAP v1 Adapter
HID Device Descriptor:
bLength 9
bDescriptorType 33
bcdHID 1.11
bCountryCode 0 Not supported
bNumDescriptors 1
bDescriptorType 34 Report
wDescriptorLength 30
Report Descriptors:
** UNAVAILABLE **
Endpoint Descriptor:
bLength 7
bDescriptorType 5
bEndpointAddress 0x86 EP 6 IN
bmAttributes 3
Transfer Type Interrupt
Synch Type None
Usage Type Data
wMaxPacketSize 0x0200 1x 512 bytes
bInterval 1
Endpoint Descriptor:
bLength 7
bDescriptorType 5
bEndpointAddress 0x06 EP 6 OUT
bmAttributes 3
Transfer Type Interrupt
Synch Type None
Usage Type Data
wMaxPacketSize 0x0200 1x 512 bytes
bInterval 1
Device Status: 0x0001
Self Powered

This is my descriptor

[ataradov]

No, I mean report descriptor. lsusb always shows it as "** UNAVAILABLE **" for whatever reason, so jut look in the code what you are sending as a report descriptor.

It should be a byte array that looks like this

free-dap/platform/same70/usb_descriptors.c

Line 114 in 49a30aa

const alignas(4) uint8_t usb_hid_report_descriptor[30] =

And "Report Count" entries must match the actual endpoint size (512 in your case).

[AjjaiAravind]

yeah we see the report descriptor unavailable and already working on to add it. Will update once added

[ataradov]

Unavailable from lsusb does not mean you don't have any, it always shows it as unavailable. But you must have the descriptor that is basically exactly like the one I linked. Except with endpoint sizes fixed.

If you just used some keyboard or mouse code as a reference, it will not match at all.

[AjjaiAravind]

Actually in Linux, it ll send a request for the report descriptor and then will send a response with it. Will the same happen in windows or we should append the descriptor to the config descriptors that we are sending?

[ataradov]

HID report descriptor is sent as part of the HID implementation, it is not a part of any other descriptors. Host will ask for it and you need to respond with it. Host will do it automatically during enumeration. All HID implementations will have this descriptor.

[AjjaiAravind]

Ok then I think we are not getting to that part and it is failing for us even before. Right?

[ataradov]

What do you mean? It is failing during the device enumeration. This is where Windows would request the descriptor to know what the device is. This is how it knows if you plugged in a keyboard or a mouse and how many buttons that mouse has. In case of a DAP, the descriptor just says that it is a generic input/outout device.

[AjjaiAravind]

yeah it is failing in the enumeration

[ataradov]

Just search your code for all mentions of report descriptor and see where it is handled. It should be sent in response to (IN, INTERFACE, STANDARD, GET_DESCRIPTOR) request over a control endpoint.

[AjjaiAravind]

Sure I ll check on it

[AjjaiAravind]

When I took wireshark of my hid, I cannot see the GET DESCRIPTOR REQUEST HID REPORT packet in it. I see first it is requesting for the device (GET DESCRIPTOR REQUEST DEVICE) and receiving a response (GET DESCRIPTOR RESPONSE DEVICE). Then it is requesting CONFIGURATION for size 9 and receiving the response. Next it sends request for configuration of size: 41 where we send the interface/HID/endpoint descriptors. Then again instead of proceeding further it is again requesting for size 9 configuration. This goes on a loop and never gets connected.

[ataradov]

Your link leads nowhere. But if windows refuses to continue enumeration, it does not like something about your device or configuration descriptors. And the first thing I would try is change .bDeviceSubClass = 0x00, .bDeviceProtocol = 0x00.

[AjjaiAravind]

ok will try that. Also I see wmaxpacketsize is 512. It should be 64 right? will try to change and check that as well. One query is that at this point would we require to develop any windows driver INF and try to install it to our device to get the enumeration done? Should we think on that point of view as well?

[ataradov]

It depends on what your hardware is configured for. The values in the configuration descriptor, HID report descriptor and the actual hardware configuration must match.

HID devices don't need INF file, you just need to fix your descriptors.

[AjjaiAravind]

Pls see the attached image. Now we are able to make up to the report request and response place after making some changes to the device and config descriptor. The device is also list under hid in the device manager as Hid_compliant_device and usb_input_device. Still we see the usb gets unmounted in the probe and unable to proceed further.

[ataradov]

There is still no attachment. New listing from lsusb may be helpful too.

[AjjaiAravind]

This is the latest lsusb -v output,

Bus 003 Device 006: ID cafe:4006 Synaptics, Inc SR100 CMSIS-DAP

Couldn't open device, some information will be missing

Device Descriptor:

bLength 18

bDescriptorType 1

bcdUSB 2.00

bDeviceClass 0

bDeviceSubClass 0

bDeviceProtocol 0

bMaxPacketSize0 64

idVendor 0xcafe

idProduct 0x4006

bcdDevice 1.00

iManufacturer 1 Synaptics, Inc

iProduct 2 SR100 CMSIS-DAP

iSerial 3 01.00.00.00

bNumConfigurations 1

Configuration Descriptor:

bLength                 9

bDescriptorType         2

wTotalLength       0x0029

bNumInterfaces          1

bConfigurationValue     1

iConfiguration          5 

bmAttributes         0xc0

  Self Powered

MaxPower              500mA

Interface Descriptor:

  bLength                 9

  bDescriptorType         4

  bInterfaceNumber        2

  bAlternateSetting       0

  bNumEndpoints           2

  bInterfaceClass         3 Human Interface Device

  bInterfaceSubClass      0 

  bInterfaceProtocol      0 

  iInterface              4 

    HID Device Descriptor:

      bLength                 9

      bDescriptorType        33

      bcdHID               1.11

      bCountryCode            0 Not supported

      bNumDescriptors         1

      bDescriptorType        34 Report

      wDescriptorLength      28

     Report Descriptors: 

       ** UNAVAILABLE **

  Endpoint Descriptor:

    bLength                 7

    bDescriptorType         5

    bEndpointAddress     0x87  EP 7 IN

    bmAttributes            3

      Transfer Type            Interrupt

      Synch Type               None

      Usage Type               Data

    wMaxPacketSize     0x0040  1x 64 bytes

    bInterval              10

  Endpoint Descriptor:

    bLength                 7

    bDescriptorType         5

    bEndpointAddress     0x06  EP 6 OUT

    bmAttributes            3

      Transfer Type            Interrupt

      Synch Type               None

      Usage Type               Data

    wMaxPacketSize     0x0040  1x 64 bytes

    bInterval              10

[ataradov]

Ok, then an actual USB log is needed. This looks fine. Does it still work under Linux?

[AjjaiAravind]

Sorry didn't get you. Can you please rephrase it

[ataradov]

You are mentioning some logs. I see no logs. I any case, you need to figure out how to debug whatever USB system you are using, this is beyond the scope of Free-DAP and there is a limited amount of help I can provide here.

[AjjaiAravind]

No,I didn't attach any logs. "Failure during Port Change Request" this s the error statement I mentioned. This is what I m seeing with the logman debugging.

[ataradov]

I don't know what that is. I don't know much about windows.
I would recommend looking at the behavior and descriptors in the reference implementations and comparing them to yours. Take RP2040 version and get Pi Pico. Program the binary and observe how it behaves. Then make sure your code behaves the same way.

[AjjaiAravind]

Sure will check it. Thank you for the help with this