Closed mdhayter closed 1 year ago
On a mission to clean up old issues, scraping from stalest to freshest.
@vogelpi @mdhayter does @stefanlipunner's recent work in this area change the status of this issue? Related comment: we will initiate a conversation on how to move forward with USB soon.
The recent work was a good clean-up but doesn't really address the code being fragile to timing. I think both these are possible approaches to improvements.
@alees24 is currently undertaking significant work in this area. I think this issue is superseded by issues he has opened and work he's currently doing so closing.
The usbdpi interface currently includes the test stimulus in the main-line code and is quite fragile as regards to timing.
Improvements:
Read stimulus from a file (will require the DPI state machine to understand higher level constructs like packets)
Build in initialization of USB device and make the simulation available as a usbip device. This allows the simulated USB device to be "plugged in" to a virtual USB port on the host.