Simulating USB 2.0 device

We are using a low cost ($1 to $2) USB/LCD uC for a customer. Before committing to several thousand masked ROM dice, they want us to simulate all the codes. Simulating the LCD function is not a problem. Simulating the USB 2.0 device controller would not be easy, especially under Windows.

Should I just say that they are crazy and just go away (with the risk of losing the contract)? What should I charge for doing the virtual USB stacks/drivers under Win98/XP/Vista?

For LCD demo under Win98/XP/Vista, see:

formatting link

Reply to
linnix
Loading thread data ...

Hello,

Do you have some other way of demonstrating the correctness of your code ? Most reasonable customers would accept a set of test results from a development part and a paper trail to prove that the development part and the ROM part would perform in the same way.

If you are just expecting them to believe that your code is correct without testing or other proof I think that they have a point.

Finally, if I were the buyer, I would not accept simulation of USB where the application developer wrote the simulator because the risk of duplication of a mis-understanding of the USB spec is so high.

For what its worth, in a similar situation, I have asked the customer to provide a simulator that they will accept as good for sign off. Which has the virtue of getting them to think about the cost implications of the approach they demand.

Michael Kellett

Reply to
MK

I would have thought that there is much more chance (given the vagaries of the USB2.0 specification) that simulating will not produce the same results. Certainly coding it all up will be a mammoth effort (particularly if you have to interwork with the undocumented Windows USB API).

I take it that you can run your code on the real hardware? You could run USBCV tests on it and use that to say that your device is performing to specification (you'll need to do that in any event to get certification). If your device offers a standard device class (like mass storage) then that can also be covered by the USBCV tests.

If you have to try and plumb into the Windows USB stack then a great deal.

Jungo might be able to help as they have host and device drivers, and other simulations.

Andrew

Reply to
Andrew Jackson

I think that's the point. They want to have an independent QC/QA programmer to test with the simulator, but we have to provide the tools to do so.

They are really asking for a software assisted hardware ICE. I can image several target chips for these functions:

  1. Maintance LCD segment drives
  2. Handle USB endpoint pipes (or perhaps one chip per pipe per direction?)
  3. PWM output for SigmaDelta A2D (another for comparator input)

And somehow, all these chips communicate with the PC via a separate USB channel.

Reply to
linnix

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.