> Keith wrote: >So it looks like there's some problem with the way it works >out the jump address. I compared the object code and the only >changes are in the header and a couple of extra char(0)s at >the end of the program. If the logic is unchanged, why does >the addressing go wrong? > > OLD NEW > 3 147 162 version > 4 18 0 version > 8 0 16 machine type > 29 0 255 number of registers > 30 0 9 number of registers > 31 0 8 number of registers > 32 0 216 number of registers > 51 0 42 checksum > 52 0 72 checksum > 73 0 70 compilation date/time > 74 0 85 compilation date/time > 75 0 18 compilation date/time > 76 0 15 compilation date/time > 657 255 0 compiler noise > 658 255 0 compiler noise
Hello, Keith. I have included annotations to give you the context for the differences you found between the two instances of the object code. You figured out the version is different. The issue you are seeing is likely rooted in the machine type difference between the two object code instances. This indicates that fnuxi needs to be run on the object code before using VLIST. I have heard that VLIST will automatically convert between machine types, but I have not seen this work reliably. The other differences in the header are not significant. The number of registers seems to change based on compiler revisions and machine types. The checksum is different and the compilation date/time is new with 10.x. The extra char(0)s at the end of the program are the separators between the program name, copyright and comment fields, with the last two being null. Just another compiler difference. <AD> srs4uv.com recovers compilable source code for UniVerse BASIC object code with 100% accuracy. srs4uv.com has helped many organizations around the world deal with missing source code and differing object code instances. Please contact me directly or visit www.srs4uv.com for additional information. </AD> Best regards, Gyle Iverson ------- u2-users mailing list [email protected] To unsubscribe please visit http://listserver.u2ug.org/
