This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

EHCI USB Problem on DM3730

Other Parts Discussed in Thread: DM3730

Hi,

We've developed a machine that is using the DM3730 processor and it's already in production.  It's running WinCE6 (we are in the process of moving to WEC7). After having more and more of them in the field locations it turned out that on some of them once a while the USB devices stop working. We use the EHCI port and the USB hub. When it happens all the devices stop working (in case of USB hub with LEDs, all LEDs are off) and reconnection doesn’t help. The only think that helps is power cycle the processor (in our case – the entire machine).

The issue is very hard to reproduce and diagnose, but I captured the USB traces when it happened, and one of the last actions I’ve seen there was going to the lower speed. That, as far as I know, is a known issue and causes the USB port to freeze. And this is what we see.

I tried to recover from this state, and I wasn’t successful. One of the things I tried is to reset our USB chip (SMSC USB3320) using the GPIO reset line, but it turned out that this action not only doesn’t help, but even locks the USB by itself.  Basically, even if the USB is working correctly and I toggle that reset pin, it goes to the same state – it freezes.

Now, since we have our machine in production and we are selling it to many far locations (many locations have machines unattended), this issue is very critical for us.

Please see the attached caprured USB communication between the DM3730 processor and the USB hub with 4 devices.  It freezes at line 8417.

I've checked the DM3730 Errata, and there is a list of several USB issues, but I haven't found any issue that can cause our problems dirrectly (but it still possible that this is caused by some of them indirrectly).

Could you help me with this problem?

Some time ago I've seen the CE6 BSP Errata with listed all the USB problems that the BSP has, but I cannot find it any more.  Could you point me to it (or send it to me)?

Thank you!

USB_Freeze.xls
  • Hi Zack,

    I'm searching for but didn't find an errata which could be responsible for such kind of problem. I see form the logs that the USB is in suspend state in the end of log but this doesn't seem to be a problem.

    I recommend you to track the GPIO reset line for glitches when the problem appears. I have several questions in addition:

    Could you try to make suspend to retention and resume when the problem appears to verify is it solve the problem?

    Do you see some problems in the Win logs which could be related to this USB issue?

    Could you reproduce the problem with an EVM board and the same software? If the problem appears on the EVM board then the suspicions for hardware problem will be rejected.

    BR

    Tsvetolin Shulev