Reasons for failure of running the official routine USBFS-USB_Host-MSC_Host on GD32F105

千本樱大白

Reasons for failure of running the official routine USBFS-USB_Host-MSC_Host on GD32F105 [Copy link]

邀请：@yang377156216   参与回复

Has anyone tried running this GD example? When I run it, it prompts me to press the user key. After I simulate pressing it, no new information is printed. When debugging, I find that it is always stuck in the USB interrupt and cannot come out. I don’t see any enumeration information, but I can read the information of the USB disk. I can also see the disconnected information after disconnecting the USB disk.

lugl4313820

If you can enter the debugging, you can see where it is stuck. In theory, the routine should be verified. Or you can run another routine to rule out hardware problems.

千本樱大白

lugl4313820 posted on 2022-4-18 07:47 You can enter the debugger and see where it is stuck. In theory, the routine is verified. Or you can run another routine to rule out hardware problems.

The problem has been solved. It is a problem of clock priority.

ccbb294

We also found this problem during testing, but not all USB devices will get stuck. Some devices can recognize and read and write normally, but most will get stuck when pressing the button after displaying the information. At this time, there will be a prompt when unplugging the USB drive.

ccbb294

Posted by Senbenzakura Dabai on 2022-5-6 17:44 The problem has been solved, the problem is the clock priority

Can you explain it in more detail? I have also encountered this problem and have not found out the reason yet.

纪帅

I have also encountered the same problem, please advise me

ccbb294

My problem is solved

I ported the USBHS USB3300 used by the MSC HOST routine of the USB HOST in GD32450I_EVAL_Demo_Suites.

The problem is that some USB flash drives cannot be enumerated (they get stuck and keep sending INQUARY after pressing the button) while some USB flash drives can be recognized normally.

Just comment out part of the usbh_request_submit function in usbh_transc.c. I have pasted the code that needs to be commented out below. I share it for your reference.

// if (1U == udev->host.pipe[pp_num].do_ping) {
// (void)usb_pipe_ping (udev, (uint8_t)pp_num);
// return USB_OK;
// }

ccbb294

ccbb294 posted on 2022-6-10 17:34 My problem was solved. I transplanted the USBHS used by the MSC HOST routine of the USB HOST in GD32450I_EVAL_Demo_Suites ...

I don't know if this is a common problem or if it's just my board. I hope the GD32 engineers can check it out.