Waqar,

Your problem is due to the slow rise time of the signal on P2.0. This generates quantum uncertainty at the junction between the PCB and the leg of the IC - basically the electrons can't decide whether to cross through the solder joint or not. If the IC is socketed a similar effect can be observed, however in this case the electrons suffer vertigo after climbing up through the socket.

The way to tackle this problem is to use the undocumented instructions J2B and J2NB. Please realise that these instructions are only intended for expert users which is why they aren't mentioned in the datasheets. Experts just *know* they are there.

J2B and J2NB exploit the fact that electrons are very curious. The microcontroller uses quantum certainty (known as "fact" to engineers) to project a virtual positron through the IC package just beside the port pin during S2P1. Any electrons that are dithering around at the bottom of the pin will momentarily stop while they study the positron. At this instant the logic level at the pin is sampled - static electrons *always* show a pseudo voltage level of 2.5 volts. This is the third logic state which is known as "2". (Originally it was known as three, but Intel engineers didn't like the gap between one and three. Zero and one were renamed to one and two briefly during the late seventies but this didn't catch on).

So, to summarise:
J2B jumps if the logic level is 2, J2NB jumps if the logic level is not 2. Don't be confused by 2 being very close to 2.5 - the two things are unrelated by anything except arithmetic.

I hope this has helped.