On Sat, Jun 19, 2021 at 11:05 AM Henry Bent <henry.r.bent@gmail.com> wrote:From what I can gather the only way to reasonably examine the disassembly of a program in the early days of Unix was adb. Is this true?From Research, yes - although sdb and later dbx could do it also I think.Was there a way to easily produce a full disassembly?Yes, look at the contents in the early USENIX (Harvard) tape. IIRC: Along with the macro-11 assembler and linker, there was also a disassembler -- I want to say it was done at Cooper Union, but it may have been someone else[The CU folks got the DEC PDP-10 BLISS binary to run on an emulator 'good enough' on their 11/45 to they could 'port' the DEC Fortran compiler to V6. They used/built up PDP-11 tools to support that project].
Wait, so it was easier to write an emulator for a PDP-10 binary than it would have been to port BLISS to the PDP-11? Given the different word sizes I would not have expected that.
I'll definitely look at those early USENIX tapes - I have copies of them and I did think of them when I was writing this, but (at least the early ones I looked at) don't appear to have a good index of any sort. Did they have a cover sheet or something equivalent that they came with? I'm having trouble imagining dealing with that much unindexed data on an early system.
There was an even better set of assembly/disassembly/link tools from 'down-under' on ??maybe? the Delaware tape. Plus, Purdue released a ton of microprocessor tools, which included PDP-11 support. All of them tried to use the symbol table to reconstruct things like jsr's and memory access. Somebody [IIRC it was Phil Karn but I don't remember] tried to do some instruction pattern matching / early AI stuff, to see if they could reconstruct some of the code with some human help. As I recall he could pick up pointers and if given some header files for sizes would try to match code snippets, but I have no idea how he got or what happened too it.
Fascinating. Thank you as always for the insight.
-Henry
ᐧ