Mark,
Here’s the PDL code that produces the error that you’re seeing. The best that I can say is that the PDL saw an “input error”. Then it flushes the queue. Since nothing is queued to the engine and if the connection stays up then the await replies timeout should still happen.
proc read.error {info} {
keylget info type type
switch -exact — $type {
input-error {
hci_pd_report_exception 1 “device error (remote side probably shut down)”
hci_pd_ignore_input -all
}
no-match {
hci_pd_ignore_input 1
hci_pd_ignore_input -until xb
}
default {
hci_pd_report_exception 2 “unknown fail: $type”
hci_pd_ignore_input -all
}
}
}
