Paul Edwards  wrote:   
   > On Tuesday, November 29, 2022 at 5:08:37 AM UTC+8, anti...@nos   
   icedham.math.uni.wroc.pl wrote:   
   >   
   > > Concerning debugging, integrated debug support is relatively   
   > > late, but there were external circuits that could single   
   > > step the processor, stop it on access to some specific   
   > > address and similar. High-end systems on this sort were   
   > > expensive, but established company could easily afford   
   > > them.   
   >   
   > By "integrated debug" I assume you mean things like   
   > the TF triggering an interrupt. That doesn't sound like   
   > something complicated. Is there a reason for it to be   
   > "relatively late"?   
      
   One thing is debug registers.  IIUC they were first in 386 and were   
   improved later.  Now it seems that $0.4 processors contain   
   equivalent or better functionality.  In 1975 typical   
   microprocessor needed external support chips, so there was   
   understandable resistance to putting something which was   
   not essential to running programs in processor chip.   
   And once we speak about external chips, folks who wanted   
   could add equivalent of debug registers using external   
   chips.  Mass produces machines skipped debugging support   
   for cost reasons.   
      
   Another is view of "internal" state of "processor".  In now   
   we have things like JTAG for this purpose.  I early days   
   this was less important because signals that are internal   
   now were visible on processors bus and could be seen/manipulated   
   by external chips.  Mainframes and minis typically contained   
   "operator console" which allowed view and modifications to   
   system bus and memory.  Apparently home computers mostly   
   skipped this for cost reasons, but "developement systems"   
   had them.   
      
   --   
                                 Waldek Hebisch   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)