On 7/6/2018 4:54 AM, Michał Pleban wrote: > Jim Brain wrote: > >> It looks like the test is not failing in the same spot all the time. It >> looks like a transient error. Any chance to extend your program to do a >> bunch of write and reads and report any issues? It would help to see if >> there is a stuck bit or something else. > Transient errors are hardest to debug... But I presume you made a trace > then, can you show it? Yes, and no. It is easier to set up the 8 port PC-based LA, so I added code in the CPLD to activate a pin when the address was reached, and then traced it and the port values and a signal indicating whether the port was showing the $0 or $1 value. I see that sometimes, the port changes to #$1 a number of times, and then it shows your address. The number of tests varies. > > I can extend the test to read or write a value several times (or to > repeat the whole test XX times in a row), but a live trace from the > KERNAL will be the best starting point. I don't have a way (that I know of ) to grab data from the HP 1650B (I usually take photos, so old skool). > > Regards, > Michau. > > -- Jim Brain brain@jbrain.com www.jbrain.comReceived on 2018-07-06 20:02:22
Archive generated by hypermail 2.2.0.