Ah -- if it was adb you redid, no doubt of its power.  I used adb for a long time -- PDP-11/VAX/68K but as you said, you could learn a lot about your system.   FWIW:  we embedded adb into RTU, calling it kdb.   We didn't have no fancy VMs to run the system under, when it halted, it halted.   On a personal machine that was not a problem and adb/kdb was very cool.


On Sat, Jun 19, 2021 at 5:50 PM Rob Pike <robpike@gmail.com> wrote:
Although upon reflection, I think what I did was fix 'adb' and call it 'db'. Haven't had my coffee yet this morning.


On Sun, Jun 20, 2021 at 7:49 AM Rob Pike <robpike@gmail.com> wrote:
For v8 or thereabouts, I spent some time fixing some fundamental bugs in db and found that it was arcane but remarkably powerful. Since it was lower level, it avoided the endemic debugging problem of misleading you about your program: All it could do was tell you what the machine was doing. (Cdb, sdb, and adb were, at least in my experience, always lying to you.) I may be the only person who appreciated db fully. Once the bugs were gone you really could use it to good effect, as long as you understood the CPU.

But it was buggy and arcane, no question about that.


On Sun, Jun 20, 2021 at 6:46 AM Richard Salz <rich.salz@gmail.com> wrote:
I remember compiling and playing Langston's "empire" that I was told came from a decompiled executable. This was in the 4.2 days.