Retro/pcjs
Watch
1
0
Fork
You've already forked pcjs
 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Added notes on the question of whether addressing mode 5 (auto-decrement deferred) can trigger a stack overflow on any PDP-11 model

Browse source
This commit is contained in:
Jeff Parsons 2016-11-19 17:08:43 -08:00 • committed by Jeff Parsons
commit a6689e1458
2 changed files with 6 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

2
apps/pdp11/tapes/diags/README.md
View file

@ -347,7 +347,7 @@ stack overflow; eg:
SP=000376 PC=021734 PS=000340 IR=000000 SL=000377 T0 N0 Z0 V0 C0
021734: 057636 000000 BIS @000000(SP),@(SP)+
whereas mode 6 *does*; eg:
whereas addressing mode 6 *does*; eg:
R0=001437 R1=154112 R2=000040 R3=000376 R4=160316 R5=153412
SP=000376 PC=022074 PS=000340 IR=000000 SL=000377 T0 N0 Z0 V0 C0

6
modules/pdp11/lib/cpustate.js
View file

@ -1775,7 +1775,11 @@ CPUStatePDP11.prototype.pushWord = function(data, fRed)
* --------------------
* On the PDP-11/20, stack overflow traps occur when an address below 400 is referenced
* by SP in either mode 4 (auto-decrement) or 5 (auto-decrement deferred). The instruction
* is allowed to complete before the trap is issued.
* is allowed to complete before the trap is issued. NOTE: This information comes
* directly from the PDP-11/20 Handbook (1971), but the 11/20 diagnostics apparently only
* test mode 4, not mode 5, because when I removed stack limit checks for mode 5 on the
* 11/70, none of the 11/20 tests complained. So, I still need some independent confirmation
* as to whether ANY models check for stack overflow on mode 5 (auto-decrement deferred).
*
* On the PDP-11/70, the stack limit register (177774) allows a variable boundary for the
* kernel stack.