Thursday, March 31, 2022

 This Blog Documents an alternate implementation of a PDP8e.  





The PDP8e is implemented in an ICE40hx8k device.  (Lattice Semiconductor)

Verilog has been used to implement the PDP8e.  Iverilog is the simulation tool used as well Gtkwave is used to view the results of the simulation.  Verilator has been used as a linting tool.  The code is compiled with yosys , the Icestorm tools and nextpnr place and route tools.

The implementation currently is the base CPU plus extended memory and the console serial port.  The implementation has 8k x 12 bits of memory, using almost all of the memory that is in the ICE40 8k part.

A clock of 60 MHz is used for the logic.  It take 4 clocks to execute a machine cycle, there are three types of machine cycles, fetch, defer and execute.  On the original PDP8e a amchine cycle took 1.2 microseconds.  Here a machine cycle is executed in 4/60 MHz = 67 nanoseconds.  This is 17.9 times as fast as the original.

Info: Device utilisation:
Info:         ICESTORM_LC:    1285/ 7680    16%
Info:         ICESTORM_RAM:    24/ 32        75%
Info:         SB_IO:                        60/ 256       23%
Info:         SB_GB:                        6/  8          75%
Info:         ICESTORM_PLL:       1/  2           50%
Info:         SB_WARMBOOT:       0/  1            0%

Verilog Code to build the FPGA image has been committed to the repository:

 https://github.com/armstrong0/PDP8e

Two other repositories hold code to generate the front panel graph and a layout tool for wiring.

https://github.com/armstrong0/PDP8_panel

https://github.com/armstrong0/FPGA_module


Updated the repo.  Changed from 4 states per machine cycle to 5 and was able to increase the clock speed to result in a 9 % increase is speed.  This code now runs about 20 times faster than the original PDP8e  Device utilization has dropped marginally, may just be the way the synthesizer worked.  Now uses 1282 ICESTORM_LC, the rest of the resources are the same.


July 20.  Updated the repo.  Memory extension and USER mode now working.  Some hooks in place for EAE



2 comments:

  1. Very impressive. Nicely done front panel as well. The PDP 8/I FPGA implementations seem a bit too 'easy'. Any chance there's enough gates still available to include the EAE option?

    ReplyDelete
    Replies
    1. Yes there is plenty of logic to implement EAE, I have it on my list of things to do. I expect that I will need to increase the state machine states from 16 to 32 to enable control of the EAE elements. I don't know when I will get to it though....

      Delete

 I just committed working code for the EAE implementation.  This includes both the A and B modes of operation.   The logic utilization is:  ...