HP Forum Archive 15

[ Return to Index | Top of Index ]

Weird, Unintentional "XEQ" out to PPC ROM (HP-41CX) Message #1 Posted by James F. Chumbley on 23 Oct 2005, 3:14 a.m. OK, It was my fault, and I know what I did to cause it. I pressed XEQ ALPHA CK ALPHA, and the -41CX must have executed a "jump" out to the PPC ROM and begun execution of the "CK" function (clear key assignments). What I really wanted to do was to execute the internal CLKT function, but I pressed the wrong keys by mistake. The problem is that I don't know how to reliably stop execution of the PPC's CK function (the -41CX just kept on running and running and running somehow) and eventually terminated with the "Memory Lost" error message! Well this certainly stopped execution of CK all right, but it seems to me that the program pointer was still pointing to the PPC ROM. Here's the question: Assuming that I go back in, reload from magnetic card the now-erased program (which was erased as a result of the "Memory Lost" condition), and reassign the also now-erased user key assignment, which was only one function assigned to the "A" or 11 key, when I now press the "A" key in User mode, will the -41CX begin searching for the executable global label from within Main Memory or from within the PPC ROM? Also, is this documented anywhere in the -41CX's manual set or within the PPC's very extensive manual? I have not quite yet completely memorized all 492 pages of the PPC manual, but I want to assure everyone that I do take it to bed with me every night. (Sure I do.) Thanks, Jim Chumbley

Re: Weird, Unintentional "XEQ" out to PPC ROM (HP-41CX) Message #2 Posted by Eric Smith on 23 Oct 2005, 9:36 p.m., in response to message #1 by James F. Chumbley If you got a "MEMORY LOST", the program pointer should be left pointing into the (now empty) RAM. AFAIK, the search order for alpha labels is not different based on whether the program pointer is in ROM or RAM. The key assignment functions in the PPC ROM are incompatible with any use of time module/41CX alarms. The routines were written before the time module was introduced, and the user community had never seen "buffers" in the 41 memory before that, so it was assumed that the key assignments would always start at address 0xc0, which unfortunately is not true if there are any buffers such as alarms.

[ Return to Index | Top of Index ]

Go back to the main exhibit hall