OpenCores
URL https://opencores.org/ocsvn/zipcpu/zipcpu/trunk

Subversion Repositories zipcpu

[/] [zipcpu/] [trunk/] [doc/] [src/] [spec.tex] - Diff between revs 33 and 36

Go to most recent revision | Only display areas with differences | Details | Blame | View Log

Rev 33 Rev 36
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%
%%
%% Filename:    spec.tex
%% Filename:    spec.tex
%%
%%
%% Project:     Zip CPU -- a small, lightweight, RISC CPU soft core
%% Project:     Zip CPU -- a small, lightweight, RISC CPU soft core
%%
%%
%% Purpose:     This LaTeX file contains all of the documentation/description
%% Purpose:     This LaTeX file contains all of the documentation/description
%%              currently provided with this Zip CPU soft core.  It supersedes
%%              currently provided with this Zip CPU soft core.  It supersedes
%%              any information about the instruction set or CPUs found
%%              any information about the instruction set or CPUs found
%%              elsewhere.  It's not nearly as interesting, though, as the PDF
%%              elsewhere.  It's not nearly as interesting, though, as the PDF
%%              file it creates, so I'd recommend reading that before diving
%%              file it creates, so I'd recommend reading that before diving
%%              into this file.  You should be able to find the PDF file in
%%              into this file.  You should be able to find the PDF file in
%%              the SVN distribution together with this PDF file and a copy of
%%              the SVN distribution together with this PDF file and a copy of
%%              the GPL-3.0 license this file is distributed under.  If not,
%%              the GPL-3.0 license this file is distributed under.  If not,
%%              just type 'make' in the doc directory and it (should) build
%%              just type 'make' in the doc directory and it (should) build
%%              without a problem.
%%              without a problem.
%%
%%
%%
%%
%% Creator:     Dan Gisselquist
%% Creator:     Dan Gisselquist
%%              Gisselquist Technology, LLC
%%              Gisselquist Technology, LLC
%%
%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%
%%
%% Copyright (C) 2015, Gisselquist Technology, LLC
%% Copyright (C) 2015, Gisselquist Technology, LLC
%%
%%
%% This program is free software (firmware): you can redistribute it and/or
%% This program is free software (firmware): you can redistribute it and/or
%% modify it under the terms of  the GNU General Public License as published
%% modify it under the terms of  the GNU General Public License as published
%% by the Free Software Foundation, either version 3 of the License, or (at
%% by the Free Software Foundation, either version 3 of the License, or (at
%% your option) any later version.
%% your option) any later version.
%%
%%
%% This program is distributed in the hope that it will be useful, but WITHOUT
%% This program is distributed in the hope that it will be useful, but WITHOUT
%% ANY WARRANTY; without even the implied warranty of MERCHANTIBILITY or
%% ANY WARRANTY; without even the implied warranty of MERCHANTIBILITY or
%% FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
%% FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
%% for more details.
%% for more details.
%%
%%
%% You should have received a copy of the GNU General Public License along
%% You should have received a copy of the GNU General Public License along
%% with this program.  (It's in the $(ROOT)/doc directory, run make with no
%% with this program.  (It's in the $(ROOT)/doc directory, run make with no
%% target there if the PDF file isn't present.)  If not, see
%% target there if the PDF file isn't present.)  If not, see
%% <http://www.gnu.org/licenses/> for a copy.
%% <http://www.gnu.org/licenses/> for a copy.
%%
%%
%% License:     GPL, v3, as defined and found on www.gnu.org,
%% License:     GPL, v3, as defined and found on www.gnu.org,
%%              http://www.gnu.org/licenses/gpl.html
%%              http://www.gnu.org/licenses/gpl.html
%%
%%
%%
%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\documentclass{gqtekspec}
\documentclass{gqtekspec}
\project{Zip CPU}
\project{Zip CPU}
\title{Specification}
\title{Specification}
\author{Dan Gisselquist, Ph.D.}
\author{Dan Gisselquist, Ph.D.}
\email{dgisselq (at) opencores.org}
\email{dgisselq (at) opencores.org}
\revision{Rev.~0.3}
\revision{Rev.~0.4}
 
\definecolor{webred}{rgb}{0.2,0,0}
 
\definecolor{webgreen}{rgb}{0,0.2,0}
 
\usepackage[dvips,ps2pdf,colorlinks=true,
 
        anchorcolor=black,pagecolor=webgreen,pdfpagelabels,hypertexnames,
 
        pdfauthor={Dan Gisselquist},
 
        pdfsubject={Zip CPU}]{hyperref}
\begin{document}
\begin{document}
\pagestyle{gqtekspecplain}
\pagestyle{gqtekspecplain}
\titlepage
\titlepage
\begin{license}
\begin{license}
Copyright (C) \theyear\today, Gisselquist Technology, LLC
Copyright (C) \theyear\today, Gisselquist Technology, LLC
 
 
This project is free software (firmware): you can redistribute it and/or
This project is free software (firmware): you can redistribute it and/or
modify it under the terms of  the GNU General Public License as published
modify it under the terms of  the GNU General Public License as published
by the Free Software Foundation, either version 3 of the License, or (at
by the Free Software Foundation, either version 3 of the License, or (at
your option) any later version.
your option) any later version.
 
 
This program is distributed in the hope that it will be useful, but WITHOUT
This program is distributed in the hope that it will be useful, but WITHOUT
ANY WARRANTY; without even the implied warranty of MERCHANTIBILITY or
ANY WARRANTY; without even the implied warranty of MERCHANTIBILITY or
FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
for more details.
for more details.
 
 
You should have received a copy of the GNU General Public License along
You should have received a copy of the GNU General Public License along
with this program.  If not, see \hbox{<http://www.gnu.org/licenses/>} for a
with this program.  If not, see \hbox{<http://www.gnu.org/licenses/>} for a
copy.
copy.
\end{license}
\end{license}
\begin{revisionhistory}
\begin{revisionhistory}
 
0.4 & 9/19/2015 & Gisselquist & Added DMA controller, improved stall information, and self--assessment info.\\\hline
0.3 & 8/22/2015 & Gisselquist & First completed draft\\\hline
0.3 & 8/22/2015 & Gisselquist & First completed draft\\\hline
0.2 & 8/19/2015 & Gisselquist & Still Draft, more complete \\\hline
0.2 & 8/19/2015 & Gisselquist & Still Draft, more complete \\\hline
0.1 & 8/17/2015 & Gisselquist & Incomplete First Draft \\\hline
0.1 & 8/17/2015 & Gisselquist & Incomplete First Draft \\\hline
\end{revisionhistory}
\end{revisionhistory}
% Revision History
% Revision History
% Table of Contents, named Contents
% Table of Contents, named Contents
\tableofcontents
\tableofcontents
\listoffigures
\listoffigures
\listoftables
\listoftables
\begin{preface}
\begin{preface}
Many people have asked me why I am building the Zip CPU. ARM processors are
Many people have asked me why I am building the Zip CPU. ARM processors are
good and effective. Xilinx makes and markets Microblaze, Altera Nios, and both
good and effective. Xilinx makes and markets Microblaze, Altera Nios, and both
have better toolsets than the Zip CPU will ever have. OpenRISC is also
have better toolsets than the Zip CPU will ever have. OpenRISC is also
available, RISC--V may be replacing it. Why build a new processor?
available, RISC--V may be replacing it. Why build a new processor?
 
 
The easiest, most obvious answer is the simple one: Because I can.
The easiest, most obvious answer is the simple one: Because I can.
 
 
There's more to it, though. There's a lot that I would like to do with a
There's more to it, though. There's a lot that I would like to do with a
processor, and I want to be able to do it in a vendor independent fashion.
processor, and I want to be able to do it in a vendor independent fashion.
I would like to be able to generate Verilog code that can run equivalently
First, I would like to be able to place this processor inside an FPGA.  Without
on both Xilinx and Altera chips, and that can be easily ported from one
paying royalties, ARM is out of the question.  I would then like to be able to
manufacturer's chipsets to another. Even more, before purchasing a chip or a
generate Verilog code, both for the processor and the system it sits within,
board, I would like to know that my soft core works. I would like to build a test
that can run equivalently on both Xilinx and Altera chips, and that can be
bench to test components with, and Verilator is my chosen test bench. This
easily ported from one manufacturer's chipsets to another. Even more, before
forces me to use all Verilog, and it prevents me from using any proprietary
purchasing a chip or a board, I would like to know that my soft core works. I
cores. For this reason, Microblaze and Nios are out of the question.
would like to build a test bench to test components with, and Verilator is my
 
chosen test bench. This forces me to use all Verilog, and it prevents me from
 
using any proprietary cores. For this reason, Microblaze and Nios are out of
 
the question.
 
 
Why not OpenRISC? That's a hard question. The OpenRISC team has done some
Why not OpenRISC? That's a hard question. The OpenRISC team has done some
wonderful work on an amazing processor, and I'll have to admit that I am
wonderful work on an amazing processor, and I'll have to admit that I am
envious of what they've accomplished. I would like to port binutils to the
envious of what they've accomplished. I would like to port binutils to the
Zip CPU, as I would like to port GCC and GDB. They are way ahead of me. The
Zip CPU, as I would like to port GCC and GDB. They are way ahead of me. The
OpenRISC processor, however, is complex and hefty at about 4,500 LUTs. It has
OpenRISC processor, however, is complex and hefty at about 4,500 LUTs. It has
a lot of features of modern CPUs within it that ... well, let's just say it's
a lot of features of modern CPUs within it that ... well, let's just say it's
not the little guy on the block. The Zip CPU is lighter weight, costing only
not the little guy on the block. The Zip CPU is lighter weight, costing only
about 2,300 LUTs with no peripherals, and 3,200 LUTs with some very basic
about 2,300 LUTs with no peripherals, and 3,200 LUTs with some very basic
peripherals.
peripherals.
 
 
My final reason is that I'm building the Zip CPU as a learning experience. The
My final reason is that I'm building the Zip CPU as a learning experience. The
Zip CPU has allowed me to learn a lot about how CPUs work on a very micro
Zip CPU has allowed me to learn a lot about how CPUs work on a very micro
level. For the first time, I am beginning to understand many of the Computer
level. For the first time, I am beginning to understand many of the Computer
Architecture lessons from years ago.
Architecture lessons from years ago.
 
 
To summarize: Because I can, because it is open source, because it is light
To summarize: Because I can, because it is open source, because it is light
weight, and as an exercise in learning.
weight, and as an exercise in learning.
 
 
\end{preface}
\end{preface}
 
 
\chapter{Introduction}
\chapter{Introduction}
\pagenumbering{arabic}
\pagenumbering{arabic}
\setcounter{page}{1}
\setcounter{page}{1}
 
 
 
 
The original goal of the ZIP CPU was to be a very simple CPU.   You might
The original goal of the Zip CPU was to be a very simple CPU.   You might
think of it as a poor man's alternative to the OpenRISC architecture.
think of it as a poor man's alternative to the OpenRISC architecture.
For this reason, all instructions have been designed to be as simple as
For this reason, all instructions have been designed to be as simple as
possible, and are all designed to be executed in one instruction cycle per
possible, and are all designed to be executed in one instruction cycle per
instruction, barring pipeline stalls.  Indeed, even the bus has been simplified
instruction, barring pipeline stalls.  Indeed, even the bus has been simplified
to a constant 32-bit width, with no option for more or less.  This has
to a constant 32-bit width, with no option for more or less.  This has
resulted in the choice to drop push and pop instructions, pre-increment and
resulted in the choice to drop push and pop instructions, pre-increment and
post-decrement addressing modes, and more.
post-decrement addressing modes, and more.
 
 
For those who like buzz words, the Zip CPU is:
For those who like buzz words, the Zip CPU is:
\begin{itemize}
\begin{itemize}
\item A 32-bit CPU: All registers are 32-bits, addresses are 32-bits,
\item A 32-bit CPU: All registers are 32-bits, addresses are 32-bits,
                instructions are 32-bits wide, etc.
                instructions are 32-bits wide, etc.
\item A RISC CPU.  There is no microcode for executing instructions.  All
\item A RISC CPU.  There is no microcode for executing instructions.  All
        instructions are designed to be completed in one clock cycle.
        instructions are designed to be completed in one clock cycle.
\item A Load/Store architecture.  (Only load and store instructions
\item A Load/Store architecture.  (Only load and store instructions
                can access memory.)
                can access memory.)
\item Wishbone compliant.  All peripherals are accessed just like
\item Wishbone compliant.  All peripherals are accessed just like
                memory across this bus.
                memory across this bus.
\item A Von-Neumann architecture.  (The instructions and data share a
\item A Von-Neumann architecture.  (The instructions and data share a
                common bus.)
                common bus.)
\item A pipelined architecture, having stages for {\bf Prefetch},
\item A pipelined architecture, having stages for {\bf Prefetch},
                {\bf Decode}, {\bf Read-Operand}, the {\bf ALU/Memory}
                {\bf Decode}, {\bf Read-Operand}, the {\bf ALU/Memory}
                unit, and {\bf Write-back}.  See Fig.~\ref{fig:cpu}
                unit, and {\bf Write-back}.  See Fig.~\ref{fig:cpu}
\begin{figure}\begin{center}
\begin{figure}\begin{center}
\includegraphics[width=3.5in]{../gfx/cpu.eps}
\includegraphics[width=3.5in]{../gfx/cpu.eps}
\caption{Zip CPU internal pipeline architecture}\label{fig:cpu}
\caption{Zip CPU internal pipeline architecture}\label{fig:cpu}
\end{center}\end{figure}
\end{center}\end{figure}
                for a diagram of this structure.
                for a diagram of this structure.
\item Completely open source, licensed under the GPL.\footnote{Should you
\item Completely open source, licensed under the GPL.\footnote{Should you
        need a copy of the Zip CPU licensed under other terms, please
        need a copy of the Zip CPU licensed under other terms, please
        contact me.}
        contact me.}
\end{itemize}
\end{itemize}
 
 
Now, however, that I've worked on the Zip CPU for a while, it is not nearly
Now, however, that I've worked on the Zip CPU for a while, it is not nearly
as simple as I originally hoped.  Worse, I've had to adjust to create
as simple as I originally hoped.  Worse, I've had to adjust to create
capabilities that I was never expecting to need.  These include:
capabilities that I was never expecting to need.  These include:
\begin{itemize}
\begin{itemize}
\item {\bf External Debug:} Once placed upon an FPGA, some external means is
\item {\bf External Debug:} Once placed upon an FPGA, some external means is
        still necessary to debug this CPU.  That means that there needs to be
        still necessary to debug this CPU.  That means that there needs to be
        an external register that can control the CPU: reset it, halt it, step
        an external register that can control the CPU: reset it, halt it, step
        it, and tell whether it is running or not.  My chosen interface
        it, and tell whether it is running or not.  My chosen interface
        includes a second register similar to this control register.  This
        includes a second register similar to this control register.  This
        second register allows the external controller or debugger to examine
        second register allows the external controller or debugger to examine
        registers internal to the CPU.
        registers internal to the CPU.
 
 
\item {\bf Internal Debug:} Being able to run a debugger from within
\item {\bf Internal Debug:} Being able to run a debugger from within
        a user process requires an ability to step a user process from
        a user process requires an ability to step a user process from
        within a debugger.  It also requires a break instruction that can
        within a debugger.  It also requires a break instruction that can
        be substituted for any other instruction, and substituted back.
        be substituted for any other instruction, and substituted back.
        The break is actually difficult: the break instruction cannot be
        The break is actually difficult: the break instruction cannot be
        allowed to execute.  That way, upon a break, the debugger should
        allowed to execute.  That way, upon a break, the debugger should
        be able to jump back into the user process to step the instruction
        be able to jump back into the user process to step the instruction
        that would've been at the break point initially, and then to
        that would've been at the break point initially, and then to
        replace the break after passing it.
        replace the break after passing it.
 
 
        Incidentally, this break messes with the prefetch cache and the
        Incidentally, this break messes with the prefetch cache and the
        pipeline: if you change an instruction partially through the pipeline,
        pipeline: if you change an instruction partially through the pipeline,
        the whole pipeline needs to be cleansed.  Likewise if you change
        the whole pipeline needs to be cleansed.  Likewise if you change
        an instruction in memory, you need to make sure the cache is reloaded
        an instruction in memory, you need to make sure the cache is reloaded
        with the new instruction.
        with the new instruction.
 
 
\item {\bf Prefetch Cache:} My original implementation had a very
\item {\bf Prefetch Cache:} My original implementation had a very
        simple prefetch stage.  Any time the PC changed the prefetch would go
        simple prefetch stage.  Any time the PC changed the prefetch would go
        and fetch the new instruction.  While this was perhaps this simplest
        and fetch the new instruction.  While this was perhaps this simplest
        approach, it cost roughly five clocks for every instruction.  This
        approach, it cost roughly five clocks for every instruction.  This
        was deemed unacceptable, as I wanted a CPU that could execute
        was deemed unacceptable, as I wanted a CPU that could execute
        instructions in one cycle.  I therefore have a prefetch cache that
        instructions in one cycle.  I therefore have a prefetch cache that
        issues pipelined wishbone accesses to memory and then pushes
        issues pipelined wishbone accesses to memory and then pushes
        instructions at the CPU.  Sadly, this accounts for about 20\% of the
        instructions at the CPU.  Sadly, this accounts for about 20\% of the
        logic in the entire CPU, or 15\% of the logic in the entire system.
        logic in the entire CPU, or 15\% of the logic in the entire system.
 
 
 
 
\item {\bf Operating System:} In order to support an operating system,
\item {\bf Operating System:} In order to support an operating system,
        interrupts and so forth, the CPU needs to support supervisor and
        interrupts and so forth, the CPU needs to support supervisor and
        user modes, as well as a means of switching between them.  For example,
        user modes, as well as a means of switching between them.  For example,
        the user needs a means of executing a system call.  This is the
        the user needs a means of executing a system call.  This is the
        purpose of the {\bf `trap'} instruction.  This instruction needs to
        purpose of the {\bf `trap'} instruction.  This instruction needs to
        place the CPU into supervisor mode (here equivalent to disabling
        place the CPU into supervisor mode (here equivalent to disabling
        interrupts), as well as handing it a parameter such as identifying
        interrupts), as well as handing it a parameter such as identifying
        which O/S function was called.
        which O/S function was called.
 
 
My initial approach to building a trap instruction was to create an external
My initial approach to building a trap instruction was to create an external
peripheral which, when written to, would generate an interrupt and could
peripheral which, when written to, would generate an interrupt and could
return the last value written to it.  In practice, this approach didn't work
return the last value written to it.  In practice, this approach didn't work
at all: the CPU executed two instructions while waiting for the
at all: the CPU executed two instructions while waiting for the
trap interrupt to take place.  Since then, I've decided to keep the rest of
trap interrupt to take place.  Since then, I've decided to keep the rest of
the CC register for that purpose so that a write to the CC register, with the
the CC register for that purpose so that a write to the CC register, with the
GIE bit cleared, could be used to execute a trap.  This has other problems,
GIE bit cleared, could be used to execute a trap.  This has other problems,
though, primarily in the limitation of the uses of the CC register.  In
though, primarily in the limitation of the uses of the CC register.  In
particular, the CC register is the best place to put CPU state information and
particular, the CC register is the best place to put CPU state information and
to ``announce'' special CPU features (floating point, etc).  So the trap
to ``announce'' special CPU features (floating point, etc).  So the trap
instruction still switches to interrupt mode, but the CC register is not
instruction still switches to interrupt mode, but the CC register is not
nearly as useful for telling the supervisor mode processor what trap is being
nearly as useful for telling the supervisor mode processor what trap is being
executed.
executed.
 
 
Modern timesharing systems also depend upon a {\bf Timer} interrupt
Modern timesharing systems also depend upon a {\bf Timer} interrupt
to handle task swapping.  For the Zip CPU, this interrupt is handled
to handle task swapping.  For the Zip CPU, this interrupt is handled
external to the CPU as part of the CPU System, found in {\tt zipsystem.v}.
external to the CPU as part of the CPU System, found in {\tt zipsystem.v}.
The timer module itself is found in {\tt ziptimer.v}.
The timer module itself is found in {\tt ziptimer.v}.
 
 
\item {\bf Pipeline Stalls:} My original plan was to not support pipeline
\item {\bf Pipeline Stalls:} My original plan was to not support pipeline
        stalls at all, but rather to require the compiler to properly schedule
        stalls at all, but rather to require the compiler to properly schedule
        all instructions so that stalls would never be necessary.  After trying
        all instructions so that stalls would never be necessary.  After trying
        to build such an architecture, I gave up, having learned some things:
        to build such an architecture, I gave up, having learned some things:
 
 
        For example, in  order to facilitate interrupt handling and debug
        For example, in  order to facilitate interrupt handling and debug
        stepping, the CPU needs to know what instructions have finished, and
        stepping, the CPU needs to know what instructions have finished, and
        which have not.  In other words, it needs to know where it can restart
        which have not.  In other words, it needs to know where it can restart
        the pipeline from.  Once restarted, it must act as though it had
        the pipeline from.  Once restarted, it must act as though it had
        never stopped.  This killed my idea of delayed branching, since what
        never stopped.  This killed my idea of delayed branching, since what
        would be the appropriate program counter to restart at?  The one the
        would be the appropriate program counter to restart at?  The one the
        CPU was going to branch to, or the ones in the delay slots?  This
        CPU was going to branch to, or the ones in the delay slots?  This
        also makes the idea of compressed instruction codes difficult, since,
        also makes the idea of compressed instruction codes difficult, since,
        again, where do you restart on interrupt?
        again, where do you restart on interrupt?
 
 
        So I switched to a model of discrete execution: Once an instruction
        So I switched to a model of discrete execution: Once an instruction
        enters into either the ALU or memory unit, the instruction is
        enters into either the ALU or memory unit, the instruction is
        guaranteed to complete.  If the logic recognizes a branch or a
        guaranteed to complete.  If the logic recognizes a branch or a
        condition that would render the instruction entering into this stage
        condition that would render the instruction entering into this stage
        possibly inappropriate (i.e. a conditional branch preceding a store
        possibly inappropriate (i.e. a conditional branch preceding a store
        instruction for example), then the pipeline stalls for one cycle
        instruction for example), then the pipeline stalls for one cycle
        until the conditional branch completes.  Then, if it generates a new
        until the conditional branch completes.  Then, if it generates a new
        PC address, the stages preceding are all wiped clean.
        PC address, the stages preceding are all wiped clean.
 
 
        The discrete execution model allows such things as sleeping: if the
        The discrete execution model allows such things as sleeping: if the
        CPU is put to ``sleep,'' the ALU and memory stages stall and back up
        CPU is put to ``sleep,'' the ALU and memory stages stall and back up
        everything before them.  Likewise, anything that has entered the ALU
        everything before them.  Likewise, anything that has entered the ALU
        or memory stage when the CPU is placed to sleep continues to completion.
        or memory stage when the CPU is placed to sleep continues to completion.
        To handle this logic, each pipeline stage has three control signals:
        To handle this logic, each pipeline stage has three control signals:
        a valid signal, a stall signal, and a clock enable signal.  In
        a valid signal, a stall signal, and a clock enable signal.  In
        general, a stage stalls if it's contents are valid and the next step
        general, a stage stalls if it's contents are valid and the next step
        is stalled.  This allows the pipeline to fill any time a later stage
        is stalled.  This allows the pipeline to fill any time a later stage
        stalls.
        stalls.
 
 
        This approach is also different from other pipeline approaches.  Instead
        This approach is also different from other pipeline approaches.  Instead
        of keeping the entire pipeline filled, each stage is treated
        of keeping the entire pipeline filled, each stage is treated
        independently.  Therefore, individual stages may move forward as long
        independently.  Therefore, individual stages may move forward as long
        as the subsequent stage is available, regardless of whether the stage
        as the subsequent stage is available, regardless of whether the stage
        behind it is filled.
        behind it is filled.
 
 
\item {\bf Verilog Modules:} When examining how other processors worked
\item {\bf Verilog Modules:} When examining how other processors worked
        here on open cores, many of them had one separate module per pipeline
        here on open cores, many of them had one separate module per pipeline
        stage.  While this appeared to me to be a fascinating and commendable
        stage.  While this appeared to me to be a fascinating and commendable
        idea, my own implementation didn't work out quite so nicely.
        idea, my own implementation didn't work out quite so nicely.
 
 
        As an example, the decode module produces a {\em lot} of
        As an example, the decode module produces a {\em lot} of
        control wires and registers.  Creating a module out of this, with
        control wires and registers.  Creating a module out of this, with
        only the simplest of logic within it, seemed to be more a lesson
        only the simplest of logic within it, seemed to be more a lesson
        in passing wires around, rather than encapsulating logic.
        in passing wires around, rather than encapsulating logic.
 
 
        Another example was the register writeback section.  I would love
        Another example was the register writeback section.  I would love
        this section to be a module in its own right, and many have made them
        this section to be a module in its own right, and many have made them
        such.  However, other modules depend upon writeback results other
        such.  However, other modules depend upon writeback results other
        than just what's placed in the register (i.e., the control wires).
        than just what's placed in the register (i.e., the control wires).
        For these reasons, I didn't manage to fit this section into it's
        For these reasons, I didn't manage to fit this section into it's
        own module.
        own module.
 
 
        The result is that the majority of the CPU code can be found in
        The result is that the majority of the CPU code can be found in
        the {\tt zipcpu.v} file.
        the {\tt zipcpu.v} file.
\end{itemize}
\end{itemize}
 
 
With that introduction out of the way, let's move on to the instruction
With that introduction out of the way, let's move on to the instruction
set.
set.
 
 
\chapter{CPU Architecture}\label{chap:arch}
\chapter{CPU Architecture}\label{chap:arch}
 
 
The Zip CPU supports a set of two operand instructions, where the second operand
The Zip CPU supports a set of two operand instructions, where the second operand
(always a register) is the result.  The only exception is the store instruction,
(always a register) is the result.  The only exception is the store instruction,
where the first operand (always a register) is the source of the data to be
where the first operand (always a register) is the source of the data to be
stored.
stored.
 
 
\section{Simplified Bus}
\section{Simplified Bus}
The bus architecture of the Zip CPU is that of a simplified WISHBONE bus.
The bus architecture of the Zip CPU is that of a simplified WISHBONE bus.
It has been simplified in this fashion: all operations are 32--bit operations.
It has been simplified in this fashion: all operations are 32--bit operations.
The bus is neither little endian nor bit endian.  For this reason, all words
The bus is neither little endian nor big endian.  For this reason, all words
are 32--bits.  All instructions are also 32--bits wide.  Everything has been
are 32--bits.  All instructions are also 32--bits wide.  Everything has been
built around the 32--bit word.
built around the 32--bit word.
 
 
\section{Register Set}
\section{Register Set}
The Zip CPU supports two sets of sixteen 32-bit registers, a supervisor
The Zip CPU supports two sets of sixteen 32-bit registers, a supervisor
and a user set as shown in Fig.~\ref{fig:regset}.
and a user set as shown in Fig.~\ref{fig:regset}.
\begin{figure}\begin{center}
\begin{figure}\begin{center}
\includegraphics[width=3.5in]{../gfx/regset.eps}
\includegraphics[width=3.5in]{../gfx/regset.eps}
\caption{Zip CPU Register File}\label{fig:regset}
\caption{Zip CPU Register File}\label{fig:regset}
\end{center}\end{figure}
\end{center}\end{figure}
The supervisor set is used in interrupt mode when interrupts are disabled,
The supervisor set is used in interrupt mode when interrupts are disabled,
whereas the user set is used otherwise.  Of this register set, the Program
whereas the user set is used otherwise.  Of this register set, the Program
Counter (PC) is register 15, whereas the status register (SR) or condition
Counter (PC) is register 15, whereas the status register (SR) or condition
code register
code register
(CC) is register 14.  By convention, the stack pointer will be register 13 and
(CC) is register 14.  By convention, the stack pointer will be register 13 and
noted as (SP)--although there is nothing special about this register other
noted as (SP)--although there is nothing special about this register other
than this convention.
than this convention.
The CPU can access both register sets via move instructions from the
The CPU can access both register sets via move instructions from the
supervisor state, whereas the user state can only access the user registers.
supervisor state, whereas the user state can only access the user registers.
 
 
The status register is special, and bears further mention.  The lower
The status register is special, and bears further mention.  As shown in
10 bits of the status register form a set of CPU state and condition codes.
Fig.~\ref{tbl:cc-register},
Writes to other bits of this register are preserved.
\begin{table}\begin{center}
 
\begin{bitlist}
 
31\ldots 11 & R/W & Reserved for future uses\\\hline
 
10 & R & (Reserved for) Bus-Error Flag\\\hline
 
9 & R & Trap, or user interrupt, Flag.  Cleared on return to userspace.\\\hline
 
8 & R & (Reserved for) Illegal Instruction Flag\\\hline
 
7 & R/W & Break--Enable\\\hline
 
6 & R/W & Step\\\hline
 
5 & R/W & Global Interrupt Enable (GIE)\\\hline
 
4 & R/W & Sleep.  When GIE is also set, the CPU waits for an interrupt.\\\hline
 
3 & R/W & Overflow\\\hline
 
2 & R/W & Negative.  The sign bit was set as a result of the last ALU instruction.\\\hline
 
1 & R/W & Carry\\\hline
 
0 & R/W & Zero.  The last ALU operation produced a zero.\\\hline
 
\end{bitlist}
 
\caption{Condition Code Register Bit Assignment}\label{tbl:cc-register}
 
\end{center}\end{table}
 
the lower 11~bits of the status register form
 
a set of CPU state and condition codes.  Writes to other bits of this register
 
are preserved.
 
 
Of the condition codes, the bottom four bits are the current flags:
Of the condition codes, the bottom four bits are the current flags:
                Zero (Z),
                Zero (Z),
                Carry (C),
                Carry (C),
                Negative (N),
                Negative (N),
                and Overflow (V).
                and Overflow (V).
 
 
The next bit is a clock enable (0 to enable) or sleep bit (1 to put
The next bit is a clock enable (0 to enable) or sleep bit (1 to put
        the CPU to sleep).  Setting this bit will cause the CPU to
        the CPU to sleep).  Setting this bit will cause the CPU to
        wait for an interrupt (if interrupts are enabled), or to
        wait for an interrupt (if interrupts are enabled), or to
        completely halt (if interrupts are disabled).
        completely halt (if interrupts are disabled).
 
 
The sixth bit is a global interrupt enable bit (GIE).  When this
The sixth bit is a global interrupt enable bit (GIE).  When this
        sixth bit is a `1' interrupts will be enabled, else disabled.  When
        sixth bit is a `1' interrupts will be enabled, else disabled.  When
        interrupts are disabled, the CPU will be in supervisor mode, otherwise
        interrupts are disabled, the CPU will be in supervisor mode, otherwise
        it is in user mode.  Thus, to execute a context switch, one only
        it is in user mode.  Thus, to execute a context switch, one only
        need enable or disable interrupts.  (When an interrupt line goes
        need enable or disable interrupts.  (When an interrupt line goes
        high, interrupts will automatically be disabled, as the CPU goes
        high, interrupts will automatically be disabled, as the CPU goes
        and deals with its context switch.)  Special logic has been added to
        and deals with its context switch.)  Special logic has been added to
        keep the user mode from setting the sleep register and clearing the
        keep the user mode from setting the sleep register and clearing the
        GIE register at the same time, with clearing the GIE register taking
        GIE register at the same time, with clearing the GIE register taking
        precedence.
        precedence.
 
 
The seventh bit is a step bit.  This bit can be
The seventh bit is a step bit.  This bit can be
        set from supervisor mode only.  After setting this bit, should
        set from supervisor mode only.  After setting this bit, should
        the supervisor mode process switch to user mode, it would then
        the supervisor mode process switch to user mode, it would then
        accomplish one instruction in user mode before returning to supervisor
        accomplish one instruction in user mode before returning to supervisor
        mode.  Then, upon return to supervisor mode, this bit will
        mode.  Then, upon return to supervisor mode, this bit will
        be automatically cleared.  This bit has no effect on the CPU while in
        be automatically cleared.  This bit has no effect on the CPU while in
        supervisor mode.
        supervisor mode.
 
 
        This functionality was added to enable a userspace debugger
        This functionality was added to enable a userspace debugger
        functionality on a user process, working through supervisor mode
        functionality on a user process, working through supervisor mode
        of course.
        of course.
 
 
 
 
The eighth bit is a break enable bit.  This controls whether a break
The eighth bit is a break enable bit.  This controls whether a break
instruction in user mode will halt the processor for an external debugger
instruction in user mode will halt the processor for an external debugger
(break enabled), or whether the break instruction will simply send send the
(break enabled), or whether the break instruction will simply send send the
CPU into interrupt mode.  Encountering a break in supervisor mode will
CPU into interrupt mode.  Encountering a break in supervisor mode will
halt the CPU independent of the break enable bit.  This bit can only be set
halt the CPU independent of the break enable bit.  This bit can only be set
within supervisor mode.
within supervisor mode.
 
 
% Should break enable be a supervisor mode bit, while the break enable bit
% Should break enable be a supervisor mode bit, while the break enable bit
% in user mode is a break has taken place bit?
% in user mode is a break has taken place bit?
%
%
 
 
This functionality was added to enable an external debugger to
This functionality was added to enable an external debugger to
        set and manage breakpoints.
        set and manage breakpoints.
 
 
The ninth bit is reserved for a floating point enable bit.  When set, the
The ninth bit is reserved for an illegal instruction bit.  When the CPU
arithmetic for the next instruction will be sent to a floating point unit.
tries to execute either a non-existant instruction, or an instruction from
Such a unit may later be added as an extension to the Zip CPU.  If the
an address that produces a bus error, the CPU will (once implemented) switch
CPU does not support floating point instructions, this bit will never be set.
to supervisor mode while setting this bit.  The bit will automatically be
The instruction set could also be simply extended to allow other data types
cleared upon any return to user mode.
in this fashion, such as two by 16--bit vector operations or four by 8--bit
 
vector operations.
 
 
 
The tenth bit is a trap bit.  It is set whenever the user requests a soft
The tenth bit is a trap bit.  It is set whenever the user requests a soft
interrupt, and cleared on any return to userspace command.  This allows the
interrupt, and cleared on any return to userspace command.  This allows the
supervisor, in supervisor mode, to determine whether it got to supervisor
supervisor, in supervisor mode, to determine whether it got to supervisor
mode from a trap or from an external interrupt or both.
mode from a trap or from an external interrupt or both.
 
 
These status register bits are summarized in Tbl.~\ref{tbl:ccbits}.
These status register bits are summarized in Tbl.~\ref{tbl:ccbits}.
\begin{table}
\begin{table}
\begin{center}
\begin{center}
\begin{tabular}{l|l}
\begin{tabular}{l|l}
Bit & Meaning \\\hline
Bit & Meaning \\\hline
9 & Soft trap, set on a trap from user mode, cleared when returning to user mode\\\hline
9 & Soft trap, set on a trap from user mode, cleared when returning to user mode\\\hline
8 & (Reserved for) Floating point enable \\\hline
8 & (Reserved for) Floating point enable \\\hline
7 & Halt on break, to support an external debugger \\\hline
7 & Halt on break, to support an external debugger \\\hline
6 & Step, single step the CPU in user mode\\\hline
6 & Step, single step the CPU in user mode\\\hline
5 & GIE, or Global Interrupt Enable \\\hline
5 & GIE, or Global Interrupt Enable \\\hline
4 & Sleep \\\hline
4 & Sleep \\\hline
3 & V, or overflow bit.\\\hline
3 & V, or overflow bit.\\\hline
2 & N, or negative bit.\\\hline
2 & N, or negative bit.\\\hline
1 & C, or carry bit.\\\hline
1 & C, or carry bit.\\\hline
0 & Z, or zero bit. \\\hline
0 & Z, or zero bit. \\\hline
\end{tabular}
\end{tabular}
\caption{Condition Code / Status Register Bits}\label{tbl:ccbits}
\caption{Condition Code / Status Register Bits}\label{tbl:ccbits}
\end{center}\end{table}
\end{center}\end{table}
 
 
\section{Conditional Instructions}
\section{Conditional Instructions}
Most, although not quite all, instructions are conditionally executed.  From
Most, although not quite all, instructions may be conditionally executed.  From
the four condition code flags, eight conditions are defined.  These are shown
the four condition code flags, eight conditions are defined.  These are shown
in Tbl.~\ref{tbl:conditions}.
in Tbl.~\ref{tbl:conditions}.
\begin{table}
\begin{table}
\begin{center}
\begin{center}
\begin{tabular}{l|l|l}
\begin{tabular}{l|l|l}
Code & Mneumonic & Condition \\\hline
Code & Mneumonic & Condition \\\hline
3'h0 & None & Always execute the instruction \\
3'h0 & None & Always execute the instruction \\
3'h1 & {\tt .Z} & Only execute when 'Z' is set \\
3'h1 & {\tt .Z} & Only execute when 'Z' is set \\
3'h2 & {\tt .NE} & Only execute when 'Z' is not set \\
3'h2 & {\tt .NE} & Only execute when 'Z' is not set \\
3'h3 & {\tt .GE} & Greater than or equal ('N' not set, 'Z' irrelevant) \\
3'h3 & {\tt .GE} & Greater than or equal ('N' not set, 'Z' irrelevant) \\
3'h4 & {\tt .GT} & Greater than ('N' not set, 'Z' not set) \\
3'h4 & {\tt .GT} & Greater than ('N' not set, 'Z' not set) \\
3'h5 & {\tt .LT} & Less than ('N' set) \\
3'h5 & {\tt .LT} & Less than ('N' set) \\
3'h6 & {\tt .C} & Carry set\\
3'h6 & {\tt .C} & Carry set\\
3'h7 & {\tt .V} & Overflow set\\
3'h7 & {\tt .V} & Overflow set\\
\end{tabular}
\end{tabular}
\caption{Conditions for conditional operand execution}\label{tbl:conditions}
\caption{Conditions for conditional operand execution}\label{tbl:conditions}
\end{center}
\end{center}
\end{table}
\end{table}
There is no condition code for less than or equal, not C or not V.  Sorry,
There is no condition code for less than or equal, not C or not V.  Sorry,
I ran out of space in 3--bits.  Using these conditions will take an extra
I ran out of space in 3--bits.  Conditioning on a non--supported condition
instruction and a pipeline stall.  (Ex: \hbox{\em (Stall)}; \hbox{\tt TST \$4,CC;} \hbox{\tt STO.NZ R0,(R1)})
is still possible, but it will take an extra instruction and a pipeline stall.  (Ex: \hbox{\em (Stall)}; \hbox{\tt TST \$4,CC;} \hbox{\tt STO.NZ R0,(R1)})
 
 
 
Conditionally executed ALU instructions will not further adjust the
 
condition codes.
 
 
\section{Operand B}
\section{Operand B}
Many instruction forms have a 21-bit source ``Operand B'' associated with them.
Many instruction forms have a 21-bit source ``Operand B'' associated with them.
This Operand B is either equal to a register plus a signed immediate offset,
This Operand B is either equal to a register plus a signed immediate offset,
or an immediate offset by itself.  This value is encoded as shown in
or an immediate offset by itself.  This value is encoded as shown in
Tbl.~\ref{tbl:opb}.
Tbl.~\ref{tbl:opb}.
\begin{table}\begin{center}
\begin{table}\begin{center}
\begin{tabular}{|l|l|l|}\hline
\begin{tabular}{|l|l|l|}\hline
Bit 20 & 19 \ldots 16 & 15 \ldots 0 \\\hline
Bit 20 & 19 \ldots 16 & 15 \ldots 0 \\\hline
1'b0 & \multicolumn{2}{l|}{20--bit Signed Immediate value} \\\hline
1'b0 & \multicolumn{2}{l|}{20--bit Signed Immediate value} \\\hline
1'b1 & 4-bit Register & 16--bit Signed immediate offset \\\hline
1'b1 & 4-bit Register & 16--bit Signed immediate offset \\\hline
\end{tabular}
\end{tabular}
\caption{Bit allocation for Operand B}\label{tbl:opb}
\caption{Bit allocation for Operand B}\label{tbl:opb}
\end{center}\end{table}
\end{center}\end{table}
 
 
Sixteen and twenty bit immediate values don't make sense for all instructions.
Sixteen and twenty bit immediate values don't make sense for all instructions.
For example, what is the point of a 20--bit immediate when executing a 16--bit
For example, what is the point of a 20--bit immediate when executing a 16--bit
multiply?  Likewise, why have a 16--bit immediate when adding to a logical
multiply?  Likewise, why have a 16--bit immediate when adding to a logical
or arithmetic shift?  In these cases, the extra bits are reserved for future
or arithmetic shift?  In these cases, the extra bits are reserved for future
instruction possibilities.
instruction possibilities.
 
 
\section{Address Modes}
\section{Address Modes}
The ZIP CPU supports two addressing modes: register plus immediate, and
The Zip CPU supports two addressing modes: register plus immediate, and
immediate address.  Addresses are therefore encoded in the same fashion as
immediate address.  Addresses are therefore encoded in the same fashion as
Operand B's, shown above.
Operand B's, shown above.
 
 
A lot of long hard thought was put into whether to allow pre/post increment
A lot of long hard thought was put into whether to allow pre/post increment
and decrement addressing modes.  Finding no way to use these operators without
and decrement addressing modes.  Finding no way to use these operators without
taking two or more clocks per instruction,\footnote{The two clocks figure
taking two or more clocks per instruction,\footnote{The two clocks figure
comes from the design of the register set, allowing only one write per clock.
comes from the design of the register set, allowing only one write per clock.
That write is either from the memory unit or the ALU, but never both.} these
That write is either from the memory unit or the ALU, but never both.} these
addressing modes have been
addressing modes have been
removed from the realm of possibilities.  This means that the Zip CPU has no
removed from the realm of possibilities.  This means that the Zip CPU has no
native way of executing push, pop, return, or jump to subroutine operations.
native way of executing push, pop, return, or jump to subroutine operations.
Each of these instructions can be emulated with a set of instructions from the
Each of these instructions can be emulated with a set of instructions from the
existing set.
existing set.
 
 
\section{Move Operands}
\section{Move Operands}
The previous set of operands would be perfect and complete, save only that
The previous set of operands would be perfect and complete, save only that
the CPU needs access to non--supervisory registers while in supervisory mode.
the CPU needs access to non--supervisory registers while in supervisory mode.
Therefore, the MOV instruction is special and offers access to these registers
Therefore, the MOV instruction is special and offers access to these registers
\ldots when in supervisory mode.  To keep the compiler simple, the extra bits
\ldots when in supervisory mode.  To keep the compiler simple, the extra bits
are ignored in non-supervisory mode (as though they didn't exist), rather than
are ignored in non-supervisory mode (as though they didn't exist), rather than
being mapped to new instructions or additional capabilities.  The bits
being mapped to new instructions or additional capabilities.  The bits
indicating which register set each register lies within are the A-Usr and
indicating which register set each register lies within are the A-Usr and
B-Usr bits.  When set to a one, these refer to a user mode register.  When set
B-Usr bits.  When set to a one, these refer to a user mode register.  When set
to a zero, these refer to a register in the current mode, whether user or
to a zero, these refer to a register in the current mode, whether user or
supervisor.  Further, because a load immediate instruction exists, there is no
supervisor.  Further, because a load immediate instruction exists, there is no
move capability between an immediate and a register: all moves come from either
move capability between an immediate and a register: all moves come from either
a register or a register plus an offset.
a register or a register plus an offset.
 
 
This actually leads to a bit of a problem: since the MOV instruction encodes
This actually leads to a bit of a problem: since the MOV instruction encodes
which register set each register is coming from or moving to, how shall a
which register set each register is coming from or moving to, how shall a
compiler or assembler know how to compile a MOV instruction without knowing
compiler or assembler know how to compile a MOV instruction without knowing
the mode of the CPU at the time?  For this reason, the compiler will assume
the mode of the CPU at the time?  For this reason, the compiler will assume
all MOV registers are supervisor registers, and display them as normal.
all MOV registers are supervisor registers, and display them as normal.
Anything with the user bit set will be treated as a user register.  The CPU
Anything with the user bit set will be treated as a user register.  The CPU
will quietly ignore the supervisor bits while in user mode, and anything
will quietly ignore the supervisor bits while in user mode, and anything
marked as a user register will always be valid.  (Did I just say that in the
marked as a user register will always be valid.
last paragraph?)
 
 
 
\section{Multiply Operations}
\section{Multiply Operations}
The Zip CPU supports two Multiply operations, a
The Zip CPU supports two Multiply operations, a 16x16 bit signed multiply
16x16 bit signed multiply (MPYS) and the same but unsigned (MPYU).  In both
({\tt MPYS}) and a 16x16 bit unsigned multiply ({\tt MPYU}).  In both
cases, the operand is a register plus a 16-bit immediate, subject to the
cases, the operand is a register plus a 16-bit immediate, subject to the
rule that the register cannot be the PC or CC registers.  The PC register
rule that the register cannot be the PC or CC registers.  The PC register
field has been stolen to create a multiply by immediate instruction.  The
field has been stolen to create a multiply by immediate instruction.  The
CC register field is reserved.
CC register field is reserved.
 
 
\section{Floating Point}
\section{Floating Point}
The ZIP CPU does not support floating point operations.  However, the
The Zip CPU does not (yet) support floating point operations.  However, the
instruction set reserves two possibilities for future floating point
instruction set reserves two possibilities for future floating point
operations.
operations.
 
 
The first floating point operation hole in the instruction set involves
The first floating point operation hole in the instruction set involves
setting the floating point bit in the CC register.  The next instruction
setting a proposed (but non-existent) floating point bit in the CC register.
will simply interpret its operands as floating point instructions.
The next instruction
 
would then simply interpret its operands as floating point instructions.
Not all instructions, however, have floating point equivalents.  Further, the
Not all instructions, however, have floating point equivalents.  Further, the
immediate fields do not apply in floating point mode, and must be set to
immediate fields do not apply in floating point mode, and must be set to
zero.  Not all instructions make sense as floating point operations.
zero.  Not all instructions make sense as floating point operations.
Therefore, only the CMP, SUB, ADD, and MPY instructions may be issued as
Therefore, only the CMP, SUB, ADD, and MPY instructions may be issued as
floating point instructions.  Other instructions allow the examining of the
floating point instructions.  Other instructions allow the examining of the
floating point bit in the CC register.  In all cases, the floating point bit
floating point bit in the CC register.  In all cases, the floating point bit
is cleared one instruction after it is set.
is cleared one instruction after it is set.
 
 
The other possibility for floating point operations involves exploiting the
The other possibility for floating point operations involves exploiting the
hole in the instruction set that the NOOP and BREAK instructions reside within.
hole in the instruction set that the NOOP and BREAK instructions reside within.
These two instructions use 24--bits of address space.  A simple adjustment
These two instructions use 24--bits of address space, when only a single bit
to this space could create instructions with 4--bit register addresses for
is necessary.  A simple adjustment to this space could create instructions
each register, a 3--bit field for conditional execution, and a 2--bit field
with 4--bit register addresses for each register, a 3--bit field for
for which operation.  In this fashion, such a floating point capability would
conditional execution, and a 2--bit field for which operation.
only fill 13--bits of the 24--bit field, still leaving lots of room for
In this fashion, such a floating point capability would only fill 13--bits of
expansion.
the 24--bit field, still leaving lots of room for expansion.
 
 
In both cases, the Zip CPU would support 32--bit single precision floats
In both cases, the Zip CPU would support 32--bit single precision floats
only.
only, since other choices would complicate the pipeline.
 
 
The current architecture does not support a floating point not-implemented
The current architecture does not support a floating point not-implemented
interrupt.  Any soft floating point emulation must be done deliberately.
interrupt.  Any soft floating point emulation must be done deliberately.
 
 
\section{Native Instructions}
\section{Native Instructions}
The instruction set for the Zip CPU is summarized in
The instruction set for the Zip CPU is summarized in
Tbl.~\ref{tbl:zip-instructions}.
Tbl.~\ref{tbl:zip-instructions}.
\begin{table}\begin{center}
\begin{table}\begin{center}
\begin{tabular}{|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|c|}\hline
\begin{tabular}{|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|l|c|}\hline
 
\rowcolor[gray]{0.85}
Op Code & \multicolumn{8}{c|}{31\ldots24} & \multicolumn{8}{c|}{23\ldots 16}
Op Code & \multicolumn{8}{c|}{31\ldots24} & \multicolumn{8}{c|}{23\ldots 16}
        & \multicolumn{8}{c|}{15\ldots 8} & \multicolumn{8}{c|}{7\ldots 0}
        & \multicolumn{8}{c|}{15\ldots 8} & \multicolumn{8}{c|}{7\ldots 0}
        & Sets CC? \\\hline
        & Sets CC? \\\hline\hline
CMP(Sub) & \multicolumn{4}{l|}{4'h0}
CMP(Sub) & \multicolumn{4}{l|}{4'h0}
                & \multicolumn{4}{l|}{D. Reg}
                & \multicolumn{4}{l|}{D. Reg}
                & \multicolumn{3}{l|}{Cond.}
                & \multicolumn{3}{l|}{Cond.}
                & \multicolumn{21}{l|}{Operand B}
                & \multicolumn{21}{l|}{Operand B}
                & Yes \\\hline
                & Yes \\\hline
TST(And) & \multicolumn{4}{l|}{4'h1}
TST(And) & \multicolumn{4}{l|}{4'h1}
                & \multicolumn{4}{l|}{D. Reg}
                & \multicolumn{4}{l|}{D. Reg}
                & \multicolumn{3}{l|}{Cond.}
                & \multicolumn{3}{l|}{Cond.}
                & \multicolumn{21}{l|}{Operand B}
                & \multicolumn{21}{l|}{Operand B}
        & Yes \\\hline
        & Yes \\\hline
MOV & \multicolumn{4}{l|}{4'h2}
MOV & \multicolumn{4}{l|}{4'h2}
                & \multicolumn{4}{l|}{D. Reg}
                & \multicolumn{4}{l|}{D. Reg}
                & \multicolumn{3}{l|}{Cond.}
                & \multicolumn{3}{l|}{Cond.}
                & A-Usr
                & A-Usr
                & \multicolumn{4}{l|}{B-Reg}
                & \multicolumn{4}{l|}{B-Reg}
                & B-Usr
                & B-Usr
                & \multicolumn{15}{l|}{15'bit signed offset}
                & \multicolumn{15}{l|}{15'bit signed offset}
                & \\\hline
                & \\\hline
LODI & \multicolumn{4}{l|}{4'h3}
LODI & \multicolumn{4}{l|}{4'h3}
                & \multicolumn{4}{l|}{R. Reg}
                & \multicolumn{4}{l|}{R. Reg}
                & \multicolumn{24}{l|}{24'bit Signed Immediate}
                & \multicolumn{24}{l|}{24'bit Signed Immediate}
                & \\\hline
                & \\\hline
NOOP & \multicolumn{4}{l|}{4'h4}
NOOP & \multicolumn{4}{l|}{4'h4}
                & \multicolumn{4}{l|}{4'he}
                & \multicolumn{4}{l|}{4'he}
                & \multicolumn{24}{l|}{24'h00}
                & \multicolumn{24}{l|}{24'h00}
                & \\\hline
                & \\\hline
BREAK & \multicolumn{4}{l|}{4'h4}
BREAK & \multicolumn{4}{l|}{4'h4}
                & \multicolumn{4}{l|}{4'he}
                & \multicolumn{4}{l|}{4'he}
                & \multicolumn{24}{l|}{24'h01}
                & \multicolumn{24}{l|}{24'h01}
                & \\\hline
                & \\\hline
{\em Rsrd} & \multicolumn{4}{l|}{4'h4}
{\em Reserved} & \multicolumn{4}{l|}{4'h4}
                & \multicolumn{4}{l|}{4'he}
                & \multicolumn{4}{l|}{4'he}
                & \multicolumn{24}{l|}{24'bits, but not 0 or 1.}
                & \multicolumn{24}{l|}{24'bits, but not 0 or 1.}
                & \\\hline
                & \\\hline
LODIHI & \multicolumn{4}{l|}{4'h4}
LODIHI & \multicolumn{4}{l|}{4'h4}
                & \multicolumn{4}{l|}{4'hf}
                & \multicolumn{4}{l|}{4'hf}
                & \multicolumn{3}{l|}{Cond.}
                & \multicolumn{3}{l|}{Cond.}
                & 1'b1
                & 1'b1
                & \multicolumn{4}{l|}{R. Reg}
                & \multicolumn{4}{l|}{R. Reg}
                & \multicolumn{16}{l|}{16-bit Immediate}
                & \multicolumn{16}{l|}{16-bit Immediate}
                & \\\hline
                & \\\hline
LODILO & \multicolumn{4}{l|}{4'h4}
LODILO & \multicolumn{4}{l|}{4'h4}
                & \multicolumn{4}{l|}{4'hf}
                & \multicolumn{4}{l|}{4'hf}
                & \multicolumn{3}{l|}{Cond.}
                & \multicolumn{3}{l|}{Cond.}
                & 1'b0
                & 1'b0
                & \multicolumn{4}{l|}{R. Reg}
                & \multicolumn{4}{l|}{R. Reg}
                & \multicolumn{16}{l|}{16-bit Immediate}
                & \multicolumn{16}{l|}{16-bit Immediate}
                & \\\hline
                & \\\hline
16-b MPYU & \multicolumn{4}{l|}{4'h4}
16-b MPYU & \multicolumn{4}{l|}{4'h4}
                & \multicolumn{4}{l|}{R. Reg}
                & \multicolumn{4}{l|}{R. Reg}
                & \multicolumn{3}{l|}{Cond.}
                & \multicolumn{3}{l|}{Cond.}
                & 1'b0 & \multicolumn{4}{l|}{Reg}
                & 1'b0 & \multicolumn{4}{l|}{Reg}
                & \multicolumn{16}{l|}{16-bit Offset}
                & \multicolumn{16}{l|}{16-bit Offset}
                & Yes \\\hline
                & Yes \\\hline
16-b MPYU(I) & \multicolumn{4}{l|}{4'h4}
16-b MPYU(I) & \multicolumn{4}{l|}{4'h4}
                & \multicolumn{4}{l|}{R. Reg}
                & \multicolumn{4}{l|}{R. Reg}
                & \multicolumn{3}{l|}{Cond.}
                & \multicolumn{3}{l|}{Cond.}
                & 1'b0 & \multicolumn{4}{l|}{4'hf}
                & 1'b0 & \multicolumn{4}{l|}{4'hf}
                & \multicolumn{16}{l|}{16-bit Offset}
                & \multicolumn{16}{l|}{16-bit Offset}
                & Yes \\\hline
                & Yes \\\hline
16-b MPYS & \multicolumn{4}{l|}{4'h4}
16-b MPYS & \multicolumn{4}{l|}{4'h4}
                & \multicolumn{4}{l|}{R. Reg}
                & \multicolumn{4}{l|}{R. Reg}
                & \multicolumn{3}{l|}{Cond.}
                & \multicolumn{3}{l|}{Cond.}
                & 1'b1 & \multicolumn{4}{l|}{Reg}
                & 1'b1 & \multicolumn{4}{l|}{Reg}
                & \multicolumn{16}{l|}{16-bit Offset}
                & \multicolumn{16}{l|}{16-bit Offset}
                & Yes \\\hline
                & Yes \\\hline
16-b MPYS(I) & \multicolumn{4}{l|}{4'h4}
16-b MPYS(I) & \multicolumn{4}{l|}{4'h4}
                & \multicolumn{4}{l|}{R. Reg}
                & \multicolumn{4}{l|}{R. Reg}
                & \multicolumn{3}{l|}{Cond.}
                & \multicolumn{3}{l|}{Cond.}
                & 1'b1 & \multicolumn{4}{l|}{4'hf}
                & 1'b1 & \multicolumn{4}{l|}{4'hf}
                & \multicolumn{16}{l|}{16-bit Offset}
                & \multicolumn{16}{l|}{16-bit Offset}
                & Yes \\\hline
                & Yes \\\hline
ROL & \multicolumn{4}{l|}{4'h5}
ROL & \multicolumn{4}{l|}{4'h5}
                & \multicolumn{4}{l|}{R. Reg}
                & \multicolumn{4}{l|}{R. Reg}
                & \multicolumn{3}{l|}{Cond.}
                & \multicolumn{3}{l|}{Cond.}
                & \multicolumn{21}{l|}{Operand B, truncated to low order 5 bits}
                & \multicolumn{21}{l|}{Operand B, truncated to low order 5 bits}
                & \\\hline
                & \\\hline
LOD & \multicolumn{4}{l|}{4'h6}
LOD & \multicolumn{4}{l|}{4'h6}
                & \multicolumn{4}{l|}{R. Reg}
                & \multicolumn{4}{l|}{R. Reg}
                & \multicolumn{3}{l|}{Cond.}
                & \multicolumn{3}{l|}{Cond.}
                & \multicolumn{21}{l|}{Operand B address}
                & \multicolumn{21}{l|}{Operand B address}
                & \\\hline
                & \\\hline
STO & \multicolumn{4}{l|}{4'h7}
STO & \multicolumn{4}{l|}{4'h7}
                & \multicolumn{4}{l|}{D. Reg}
                & \multicolumn{4}{l|}{D. Reg}
                & \multicolumn{3}{l|}{Cond.}
                & \multicolumn{3}{l|}{Cond.}
                & \multicolumn{21}{l|}{Operand B address}
                & \multicolumn{21}{l|}{Operand B address}
                & \\\hline
                & \\\hline
SUB & \multicolumn{4}{l|}{4'h8}
SUB & \multicolumn{4}{l|}{4'h8}
        &       \multicolumn{4}{l|}{R. Reg}
        &       \multicolumn{4}{l|}{R. Reg}
        &       \multicolumn{3}{l|}{Cond.}
        &       \multicolumn{3}{l|}{Cond.}
        &       \multicolumn{21}{l|}{Operand B}
        &       \multicolumn{21}{l|}{Operand B}
        & Yes \\\hline
        & Yes \\\hline
AND & \multicolumn{4}{l|}{4'h9}
AND & \multicolumn{4}{l|}{4'h9}
        &       \multicolumn{4}{l|}{R. Reg}
        &       \multicolumn{4}{l|}{R. Reg}
        &       \multicolumn{3}{l|}{Cond.}
        &       \multicolumn{3}{l|}{Cond.}
        &       \multicolumn{21}{l|}{Operand B}
        &       \multicolumn{21}{l|}{Operand B}
        & Yes \\\hline
        & Yes \\\hline
ADD & \multicolumn{4}{l|}{4'ha}
ADD & \multicolumn{4}{l|}{4'ha}
        &       \multicolumn{4}{l|}{R. Reg}
        &       \multicolumn{4}{l|}{R. Reg}
        &       \multicolumn{3}{l|}{Cond.}
        &       \multicolumn{3}{l|}{Cond.}
        &       \multicolumn{21}{l|}{Operand B}
        &       \multicolumn{21}{l|}{Operand B}
        & Yes \\\hline
        & Yes \\\hline
OR & \multicolumn{4}{l|}{4'hb}
OR & \multicolumn{4}{l|}{4'hb}
        &       \multicolumn{4}{l|}{R. Reg}
        &       \multicolumn{4}{l|}{R. Reg}
        &       \multicolumn{3}{l|}{Cond.}
        &       \multicolumn{3}{l|}{Cond.}
        &       \multicolumn{21}{l|}{Operand B}
        &       \multicolumn{21}{l|}{Operand B}
        & Yes \\\hline
        & Yes \\\hline
XOR & \multicolumn{4}{l|}{4'hc}
XOR & \multicolumn{4}{l|}{4'hc}
        &       \multicolumn{4}{l|}{R. Reg}
        &       \multicolumn{4}{l|}{R. Reg}
        &       \multicolumn{3}{l|}{Cond.}
        &       \multicolumn{3}{l|}{Cond.}
        &       \multicolumn{21}{l|}{Operand B}
        &       \multicolumn{21}{l|}{Operand B}
        & Yes \\\hline
        & Yes \\\hline
LSL/ASL & \multicolumn{4}{l|}{4'hd}
LSL/ASL & \multicolumn{4}{l|}{4'hd}
        &       \multicolumn{4}{l|}{R. Reg}
        &       \multicolumn{4}{l|}{R. Reg}
        &       \multicolumn{3}{l|}{Cond.}
        &       \multicolumn{3}{l|}{Cond.}
        &       \multicolumn{21}{l|}{Operand B, imm. truncated to 6 bits}
        &       \multicolumn{21}{l|}{Operand B, imm. truncated to 6 bits}
        & Yes \\\hline
        & Yes \\\hline
ASR & \multicolumn{4}{l|}{4'he}
ASR & \multicolumn{4}{l|}{4'he}
        &       \multicolumn{4}{l|}{R. Reg}
        &       \multicolumn{4}{l|}{R. Reg}
        &       \multicolumn{3}{l|}{Cond.}
        &       \multicolumn{3}{l|}{Cond.}
        &       \multicolumn{21}{l|}{Operand B, imm. truncated to 6 bits}
        &       \multicolumn{21}{l|}{Operand B, imm. truncated to 6 bits}
        & Yes \\\hline
        & Yes \\\hline
LSR & \multicolumn{4}{l|}{4'hf}
LSR & \multicolumn{4}{l|}{4'hf}
        &       \multicolumn{4}{l|}{R. Reg}
        &       \multicolumn{4}{l|}{R. Reg}
        &       \multicolumn{3}{l|}{Cond.}
        &       \multicolumn{3}{l|}{Cond.}
        &       \multicolumn{21}{l|}{Operand B, imm. truncated to 6 bits}
        &       \multicolumn{21}{l|}{Operand B, imm. truncated to 6 bits}
        & Yes \\\hline
        & Yes \\\hline
\end{tabular}
\end{tabular}
\caption{Zip CPU Instruction Set}\label{tbl:zip-instructions}
\caption{Zip CPU Instruction Set}\label{tbl:zip-instructions}
\end{center}\end{table}
\end{center}\end{table}
 
 
As you can see, there's lots of room for instruction set expansion.  The
As you can see, there's lots of room for instruction set expansion.  The
NOOP and BREAK instructions are the only instructions within one particular
NOOP and BREAK instructions are the only instructions within one particular
24--bit hole.  This spaces are reserved for future enhancements.  For example,
24--bit hole.  The rest of this space is reserved for future enhancements.
floating point operations, consisting of a 3-bit floating point operation,
 
two 4-bit registers, no immediate offset, and a 3-bit condition would fit
 
nicely into 14--bits of this address space--making it so that the floating
 
point bit in the CC register need not be used.
 
 
 
\section{Derived Instructions}
\section{Derived Instructions}
The ZIP CPU supports many other common instructions, but not all of them
The Zip CPU supports many other common instructions, but not all of them
are single cycle instructions.  The derived instruction tables,
are single cycle instructions.  The derived instruction tables,
Tbls.~\ref{tbl:derived-1}, \ref{tbl:derived-2}, and~\ref{tbl:derived-3},
Tbls.~\ref{tbl:derived-1}, \ref{tbl:derived-2}, \ref{tbl:derived-3}
 
and~\ref{tbl:derived-4},
help to capture some of how these other instructions may be implemented on
help to capture some of how these other instructions may be implemented on
the ZIP CPU.  Many of these instructions will have assembly equivalents,
the Zip CPU.  Many of these instructions will have assembly equivalents,
such as the branch instructions, to facilitate working with the CPU.
such as the branch instructions, to facilitate working with the CPU.
\begin{table}\begin{center}
\begin{table}\begin{center}
\begin{tabular}{p{1.4in}p{1.5in}p{3in}}\\\hline
\begin{tabular}{p{1.4in}p{1.5in}p{3in}}\\\hline
Mapped & Actual  & Notes \\\hline
Mapped & Actual  & Notes \\\hline
 
ABS Rx
 
        & \parbox[t]{1.5in}{TST -1,Rx\\NEG.LT Rx}
 
        & Absolute value, depends upon derived NEG.\\\hline
\parbox[t]{1.4in}{ADD Ra,Rx\\ADDC Rb,Ry}
\parbox[t]{1.4in}{ADD Ra,Rx\\ADDC Rb,Ry}
        & \parbox[t]{1.5in}{Add Ra,Rx\\ADD.C \$1,Ry\\Add Rb,Ry}
        & \parbox[t]{1.5in}{Add Ra,Rx\\ADD.C \$1,Ry\\Add Rb,Ry}
        & Add with carry \\\hline
        & Add with carry \\\hline
BRA.Cond +/-\$Addr
BRA.Cond +/-\$Addr
        & \hbox{MOV.cond \$Addr+PC,PC}
        & \hbox{MOV.cond \$Addr+PC,PC}
        & Branch or jump on condition.  Works for 15--bit
        & Branch or jump on condition.  Works for 15--bit
                signed address offsets.\\\hline
                signed address offsets.\\\hline
BRA.Cond +/-\$Addr
BRA.Cond +/-\$Addr
        & \parbox[t]{1.5in}{LDI \$Addr,Rx \\ ADD.cond Rx,PC}
        & \parbox[t]{1.5in}{LDI \$Addr,Rx \\ ADD.cond Rx,PC}
        & Branch/jump on condition.  Works for
        & Branch/jump on condition.  Works for
        23 bit address offsets, but costs a register, an extra instruction,
        23 bit address offsets, but costs a register, an extra instruction,
        and sets the flags. \\\hline
        and sets the flags. \\\hline
BNC PC+\$Addr
BNC PC+\$Addr
        & \parbox[t]{1.5in}{Test \$Carry,CC \\ MOV.Z PC+\$Addr,PC}
        & \parbox[t]{1.5in}{Test \$Carry,CC \\ MOV.Z PC+\$Addr,PC}
        & Example of a branch on an unsupported
        & Example of a branch on an unsupported
                condition, in this case a branch on not carry \\\hline
                condition, in this case a branch on not carry \\\hline
BUSY & MOV \$-1(PC),PC & Execute an infinite loop \\\hline
BUSY & MOV \$-1(PC),PC & Execute an infinite loop \\\hline
CLRF.NZ Rx
CLRF.NZ Rx
        & XOR.NZ Rx,Rx
        & XOR.NZ Rx,Rx
        & Clear Rx, and flags, if the Z-bit is not set \\\hline
        & Clear Rx, and flags, if the Z-bit is not set \\\hline
CLR Rx
CLR Rx
        & LDI \$0,Rx
        & LDI \$0,Rx
        & Clears Rx, leaves flags untouched.  This instruction cannot be
        & Clears Rx, leaves flags untouched.  This instruction cannot be
                conditional. \\\hline
                conditional. \\\hline
EXCH.W Rx
EXCH.W Rx
        & ROL \$16,Rx
        & ROL \$16,Rx
        & Exchanges the top and bottom 16'bit words of Rx \\\hline
        & Exchanges the top and bottom 16'bit words of Rx \\\hline
HALT
HALT
        & Or \$SLEEP,CC
        & Or \$SLEEP,CC
        & Executed while in interrupt mode.  In user mode this is simply a
        & Executed while in interrupt mode.  In user mode this is simply a
        wait until interrupt instruction. \\\hline
        wait until interrupt instruction. \\\hline
INT & LDI \$0,CC
INT & LDI \$0,CC
        & Since we're using the CC register as a trap vector as well, this
        & Since we're using the CC register as a trap vector as well, this
        executes TRAP \#0. \\\hline
        executes TRAP \#0. \\\hline
IRET
IRET
        & OR \$GIE,CC
        & OR \$GIE,CC
        & Also an RTU instruction (Return to Userspace) \\\hline
        & Also an RTU instruction (Return to Userspace) \\\hline
JMP R6+\$Addr
JMP R6+\$Addr
        & MOV \$Addr(R6),PC
        & MOV \$Addr(R6),PC
        & \\\hline
        & \\\hline
JSR PC+\$Addr
JSR PC+\$Addr
        & \parbox[t]{1.5in}{SUB \$1,SP \\\
        & \parbox[t]{1.5in}{SUB \$1,SP \\\
        MOV \$3+PC,R0 \\
        MOV \$3+PC,R0 \\
        STO R0,1(SP) \\
        STO R0,1(SP) \\
        MOV \$Addr+PC,PC \\
        MOV \$Addr+PC,PC \\
        ADD \$1,SP}
        ADD \$1,SP}
        & Jump to Subroutine. Note the required cleanup instruction after
        & Jump to Subroutine. Note the required cleanup instruction after
        returning. \\\hline
        returning.  This could easily be turned into a three instruction
 
        operand, removing the preliminary stack instruction before and
 
        the cleanup after, by adjusting how any stack frame was built for
 
        this routine to include space at the top of the stack for the PC.
 
        \\\hline
JSR PC+\$Addr
JSR PC+\$Addr
        & \parbox[t]{1.5in}{MOV \$3+PC,R12 \\ MOV \$addr+PC,PC}
        & \parbox[t]{1.5in}{MOV \$3+PC,R12 \\ MOV \$addr+PC,PC}
        &This is the high speed
        &This is the high speed
        version of a subroutine call, necessitating a register to hold the
        version of a subroutine call, necessitating a register to hold the
        last PC address.  In its favor, this method doesn't suffer the
        last PC address.  In its favor, this method doesn't suffer the
        mandatory memory access of the other approach. \\\hline
        mandatory memory access of the other approach. \\\hline
LDI.l \$val,Rx
LDI.l \$val,Rx
        & \parbox[t]{1.5in}{LDIHI (\$val$>>$16)\&0x0ffff, Rx \\
        & \parbox[t]{1.5in}{LDIHI (\$val$>>$16)\&0x0ffff, Rx \\
                        LDILO (\$val \& 0x0ffff)}
                        LDILO (\$val \& 0x0ffff)}
        & Sadly, there's not enough instruction
        & Sadly, there's not enough instruction
                space to load a complete immediate value into any register.
                space to load a complete immediate value into any register.
                Therefore, fully loading any register takes two cycles.
                Therefore, fully loading any register takes two cycles.
                The LDIHI (load immediate high) and LDILO (load immediate low)
                The LDIHI (load immediate high) and LDILO (load immediate low)
                instructions have been created to facilitate this. \\\hline
                instructions have been created to facilitate this. \\\hline
\end{tabular}
\end{tabular}
\caption{Derived Instructions}\label{tbl:derived-1}
\caption{Derived Instructions}\label{tbl:derived-1}
\end{center}\end{table}
\end{center}\end{table}
\begin{table}\begin{center}
\begin{table}\begin{center}
\begin{tabular}{p{1.4in}p{1.5in}p{3in}}\\\hline
\begin{tabular}{p{1.4in}p{1.5in}p{3in}}\\\hline
Mapped & Actual  & Notes \\\hline
Mapped & Actual  & Notes \\\hline
LOD.b \$addr,Rx
LOD.b \$addr,Rx
        & \parbox[t]{1.5in}{%
        & \parbox[t]{1.5in}{%
        LDI     \$addr,Ra \\
        LDI     \$addr,Ra \\
        LDI     \$addr,Rb \\
        LDI     \$addr,Rb \\
        LSR     \$2,Ra \\
        LSR     \$2,Ra \\
        AND     \$3,Rb \\
        AND     \$3,Rb \\
        LOD     (Ra),Rx \\
        LOD     (Ra),Rx \\
        LSL     \$3,Rb \\
        LSL     \$3,Rb \\
        SUB     \$32,Rb \\
        SUB     \$32,Rb \\
        ROL     Rb,Rx \\
        ROL     Rb,Rx \\
        AND \$0ffh,Rx}
        AND \$0ffh,Rx}
        & \parbox[t]{3in}{This CPU is designed for 32'bit word
        & \parbox[t]{3in}{This CPU is designed for 32'bit word
        length instructions.  Byte addressing is not supported by the CPU or
        length instructions.  Byte addressing is not supported by the CPU or
        the bus, so it therefore takes more work to do.
        the bus, so it therefore takes more work to do.
 
 
        Note also that in this example, \$Addr is a byte-wise address, where
        Note also that in this example, \$Addr is a byte-wise address, where
        all other addresses in this document are 32-bit wordlength addresses.
        all other addresses in this document are 32-bit wordlength addresses.
        For this reason,
        For this reason,
        we needed to drop the bottom two bits.  This also limits the address
        we needed to drop the bottom two bits.  This also limits the address
        space of character accesses using this method from 16 MB down to 4MB.}
        space of character accesses using this method from 16 MB down to 4MB.}
                \\\hline
                \\\hline
\parbox[t]{1.5in}{LSL \$1,Rx\\ LSLC \$1,Ry}
\parbox[t]{1.5in}{LSL \$1,Rx\\ LSLC \$1,Ry}
        & \parbox[t]{1.5in}{LSL \$1,Ry \\
        & \parbox[t]{1.5in}{LSL \$1,Ry \\
        LSL \$1,Rx \\
        LSL \$1,Rx \\
        OR.C \$1,Ry}
        OR.C \$1,Ry}
        & Logical shift left with carry.  Note that the
        & Logical shift left with carry.  Note that the
        instruction order is now backwards, to keep the conditions valid.
        instruction order is now backwards, to keep the conditions valid.
        That is, LSL sets the carry flag, so if we did this the other way
        That is, LSL sets the carry flag, so if we did this the other way
        with Rx before Ry, then the condition flag wouldn't have been right
        with Rx before Ry, then the condition flag wouldn't have been right
        for an OR correction at the end. \\\hline
        for an OR correction at the end. \\\hline
\parbox[t]{1.5in}{LSR \$1,Rx \\ LSRC \$1,Ry}
\parbox[t]{1.5in}{LSR \$1,Rx \\ LSRC \$1,Ry}
        & \parbox[t]{1.5in}{CLR Rz \\
        & \parbox[t]{1.5in}{CLR Rz \\
        LSR \$1,Ry \\
        LSR \$1,Ry \\
        LDIHI.C \$8000h,Rz \\
        LDIHI.C \$8000h,Rz \\
        LSR \$1,Rx \\
        LSR \$1,Rx \\
        OR Rz,Rx}
        OR Rz,Rx}
        & Logical shift right with carry \\\hline
        & Logical shift right with carry \\\hline
NEG Rx & \parbox[t]{1.5in}{XOR \$-1,Rx \\ ADD \$1,Rx} & \\\hline
NEG Rx & \parbox[t]{1.5in}{XOR \$-1,Rx \\ ADD \$1,Rx} & \\\hline
 
NEG.C Rx & \parbox[t]{1.5in}{MOV.C \$-1+Rx,Rx\\XOR.C \$-1,Rx} & \\\hline
NOOP & NOOP & While there are many
NOOP & NOOP & While there are many
        operations that do nothing, such as MOV Rx,Rx, or OR \$0,Rx, these
        operations that do nothing, such as MOV Rx,Rx, or OR \$0,Rx, these
        operations have consequences in that they might stall the bus if
        operations have consequences in that they might stall the bus if
        Rx isn't ready yet.  For this reason, we have a dedicated NOOP
        Rx isn't ready yet.  For this reason, we have a dedicated NOOP
        instruction. \\\hline
        instruction. \\\hline
NOT Rx & XOR \$-1,Rx & \\\hline
NOT Rx & XOR \$-1,Rx & \\\hline
POP Rx
POP Rx
        & \parbox[t]{1.5in}{LOD \$-1(SP),Rx \\ ADD \$1,SP}
        & \parbox[t]{1.5in}{LOD \$-1(SP),Rx \\ ADD \$1,SP}
        & Note
        & Note
        that for interrupt purposes, one can never depend upon the value at
        that for interrupt purposes, one can never depend upon the value at
        (SP).  Hence you read from it, then increment it, lest having
        (SP).  Hence you read from it, then increment it, lest having
        incremented it first something then comes along and writes to that
        incremented it first something then comes along and writes to that
        value before you can read the result. \\\hline
        value before you can read the result. \\\hline
 
\end{tabular}
 
\caption{Derived Instructions, continued}\label{tbl:derived-2}
 
\end{center}\end{table}
 
\begin{table}\begin{center}
 
\begin{tabular}{p{1.4in}p{1.5in}p{3in}}\\\hline
PUSH Rx
PUSH Rx
        & \parbox[t]{1.5in}{SUB \$1,SP \\
        & \parbox[t]{1.5in}{SUB \$1,SP \\
        STO Rx,\$1(SP)}
        STO Rx,\$1(SP)}
        & \\\hline
        & \\\hline
 
PUSH Rx-Ry
 
        & \parbox[t]{1.5in}{SUB \$n,SP \\
 
        STO Rx,\$n(SP)
 
        \ldots \\
 
        STO Ry,\$1(SP)}
 
        & Multiple pushes at once only need the single subtract from the
 
        stack pointer.  This derived instruction is analogous to a similar one
 
        on the Motoroloa 68k architecture, although the Zip Assembler
 
        does not support this instruction (yet).\\\hline
RESET
RESET
        & \parbox[t]{1in}{STO \$1,\$watchdog(R12)\\NOOP\\NOOP}
        & \parbox[t]{1in}{STO \$1,\$watchdog(R12)\\NOOP\\NOOP}
        & \parbox[t]{3in}{This depends upon the peripheral base address being
        & \parbox[t]{3in}{This depends upon the peripheral base address being
        in R12.
        in R12.
 
 
        Another opportunity might be to jump to the reset address from within
        Another opportunity might be to jump to the reset address from within
        supervisor mode.}\\\hline
        supervisor mode.}\\\hline
RET & \parbox[t]{1.5in}{LOD \$-1(SP),PC}
RET & \parbox[t]{1.5in}{LOD \$1(SP),PC}
        & Note that this depends upon the calling context to clean up the
        & Note that this depends upon the calling context to clean up the
        stack, as outlined for the JSR instruction.  \\\hline
        stack, as outlined for the JSR instruction.  \\\hline
\end{tabular}
 
\caption{Derived Instructions, continued}\label{tbl:derived-2}
 
\end{center}\end{table}
 
\begin{table}\begin{center}
 
\begin{tabular}{p{1.4in}p{1.5in}p{3in}}\\\hline
 
RET & MOV R12,PC
RET & MOV R12,PC
        & This is the high(er) speed version, that doesn't touch the stack.
        & This is the high(er) speed version, that doesn't touch the stack.
        As such, it doesn't suffer a stall on memory read/write to the stack.
        As such, it doesn't suffer a stall on memory read/write to the stack.
        \\\hline
        \\\hline
STEP Rr,Rt
STEP Rr,Rt
        & \parbox[t]{1.5in}{LSR \$1,Rr \\ XOR.C Rt,Rr}
        & \parbox[t]{1.5in}{LSR \$1,Rr \\ XOR.C Rt,Rr}
        & Step a Galois implementation of a Linear Feedback Shift Register, Rr,
        & Step a Galois implementation of a Linear Feedback Shift Register, Rr,
                using taps Rt \\\hline
                using taps Rt \\\hline
STO.b Rx,\$addr
STO.b Rx,\$addr
        & \parbox[t]{1.5in}{%
        & \parbox[t]{1.5in}{%
        LDI \$addr,Ra \\
        LDI \$addr,Ra \\
        LDI \$addr,Rb \\
        LDI \$addr,Rb \\
        LSR \$2,Ra \\
        LSR \$2,Ra \\
        AND \$3,Rb \\
        AND \$3,Rb \\
        SUB \$32,Rb \\
        SUB \$32,Rb \\
        LOD (Ra),Ry \\
        LOD (Ra),Ry \\
        AND \$0ffh,Rx \\
        AND \$0ffh,Rx \\
        AND \$-0ffh,Ry \\
        AND \$-0ffh,Ry \\
        ROL Rb,Rx \\
        ROL Rb,Rx \\
        OR Rx,Ry \\
        OR Rx,Ry \\
        STO Ry,(Ra) }
        STO Ry,(Ra) }
        & \parbox[t]{3in}{This CPU and it's bus are {\em not} optimized
        & \parbox[t]{3in}{This CPU and it's bus are {\em not} optimized
        for byte-wise operations.
        for byte-wise operations.
 
 
        Note that in this example, \$addr is a
        Note that in this example, \$addr is a
        byte-wise address, whereas in all of our other examples it is a
        byte-wise address, whereas in all of our other examples it is a
        32-bit word address. This also limits the address space
        32-bit word address. This also limits the address space
        of character accesses from 16 MB down to 4MB.F
        of character accesses from 16 MB down to 4MB.F
        Further, this instruction implies a byte ordering,
        Further, this instruction implies a byte ordering,
        such as big or little endian.} \\\hline
        such as big or little endian.} \\\hline
SWAP Rx,Ry
SWAP Rx,Ry
        & \parbox[t]{1.5in}{
        & \parbox[t]{1.5in}{
        XOR Ry,Rx \\
        XOR Ry,Rx \\
        XOR Rx,Ry \\
        XOR Rx,Ry \\
        XOR Ry,Rx}
        XOR Ry,Rx}
        & While no extra registers are needed, this example
        & While no extra registers are needed, this example
        does take 3-clocks. \\\hline
        does take 3-clocks. \\\hline
TRAP \#X
TRAP \#X
        & LDILO \$x,CC
        & \parbox[t]{1.5in}{LDI \$x,R0 \\ AND ~\$GIE,CC }
        & This approach uses the unused bits of the CC register as a TRAP
        & This works because whenever a user lowers the \$GIE flag, it sets
        address.  The user will need to make certain
        a TRAP bit within the CC register.  Therefore, upon entering the
        that the SLEEP and GIE bits are not set in \$x.  LDI would also work,
        supervisor state, the CPU only need check this bit to know that it
        however using LDILO permits the use of conditional traps.  (i.e.,
        got there via a TRAP.  The trap could be made conditional by making
        trap if the zero flag is set.)  Should you wish to trap off of a
        the LDI and the AND conditional.  In that case, the assembler would
        register value, you could equivalently load \$x into the register and
        quietly turn the LDI instruction into an LDILO and LDIHI pair,
        then MOV it into the CC register. \\\hline
        but the effectt would be the same. \\\hline
 
\end{tabular}
 
\caption{Derived Instructions, continued}\label{tbl:derived-3}
 
\end{center}\end{table}
 
\begin{table}\begin{center}
 
\begin{tabular}{p{1.4in}p{1.5in}p{3in}}\\\hline
TST Rx
TST Rx
        & TST \$-1,Rx
        & TST \$-1,Rx
        & Set the condition codes based upon Rx.  Could also do a CMP \$0,Rx,
        & Set the condition codes based upon Rx.  Could also do a CMP \$0,Rx,
        ADD \$0,Rx, SUB \$0,Rx, etc, AND \$-1,Rx, etc.  The TST and CMP
        ADD \$0,Rx, SUB \$0,Rx, etc, AND \$-1,Rx, etc.  The TST and CMP
        approaches won't stall future pipeline stages looking for the value
        approaches won't stall future pipeline stages looking for the value
        of Rx. \\\hline
        of Rx. \\\hline
WAIT
WAIT
        & Or \$SLEEP,CC
        & Or \$SLEEP,CC
        & Wait 'til interrupt.  In an interrupts disabled context, this
        & Wait 'til interrupt.  In an interrupts disabled context, this
        becomes a HALT instruction.
        becomes a HALT instruction.
\end{tabular}
\end{tabular}
\caption{Derived Instructions, continued}\label{tbl:derived-3}
\caption{Derived Instructions, continued}\label{tbl:derived-4}
\end{center}\end{table}
\end{center}\end{table}
\iffalse
 
\fi
 
\section{Pipeline Stages}
\section{Pipeline Stages}
As mentioned in the introduction, and highlighted in Fig.~\ref{fig:cpu},
As mentioned in the introduction, and highlighted in Fig.~\ref{fig:cpu},
the Zip CPU supports a five stage pipeline.
the Zip CPU supports a five stage pipeline.
\begin{enumerate}
\begin{enumerate}
\item {\bf Prefetch}: Read instruction from memory (cache if possible).  This
\item {\bf Prefetch}: Reads instruction from memory and into a cache, if so
 
        configured.  This
        stage is actually pipelined itself, and so it will stall if the PC
        stage is actually pipelined itself, and so it will stall if the PC
        ever changes.  Stalls are also created here if the instruction isn't
        ever changes.  Stalls are also created here if the instruction isn't
        in the prefetch cache.
        in the prefetch cache.
\item {\bf Decode}: Decode instruction into op code, register(s) to read, and
 
        immediate offset.  This stage also determines whether the flags will
        The Zip CPU supports one of two prefetch methods, depending upon a flag
 
        set at build time within the {\tt zipcpu.v} file.  The simplest is a
 
        non--cached implementation of a prefetch.  This implementation is
 
        fairly small, and ideal for
 
        users of the Zip CPU who need the extra space on the FPGA fabric.
 
        However, because this non--cached version has no cache, the maximum
 
        number of instructions per clock is limited to about one per five.
 
 
 
        The second prefetch module is a pipelined prefetch with a cache.  This
 
        module tries to keep the instruction address within a window of valid
 
        instruction addresses.  While effective, it is not a traditional
 
        cache implementation.  One unique feature of this cache implementation,
 
        however, is that it can be cleared in a single clock.  A disappointing
 
        feature, though, was that it needs an extra internal pipeline stage
 
        to be implemented.
 
 
 
\item {\bf Decode}: Decodes an instruction into op code, register(s) to read,
 
        and immediate offset.  This stage also determines whether the flags will
        be set or whether the result will be written back.
        be set or whether the result will be written back.
\item {\bf Read Operands}: Read registers and apply any immediate values to
\item {\bf Read Operands}: Read registers and apply any immediate values to
        them.  There is no means of detecting or flagging arithmetic overflow
        them.  There is no means of detecting or flagging arithmetic overflow
        or carry when adding the immediate to the operand.  This stage will
        or carry when adding the immediate to the operand.  This stage will
        stall if any source operand is pending.
        stall if any source operand is pending.
\item Split into two tracks: An {\bf ALU} which will accomplish a simple
\item Split into two tracks: An {\bf ALU} which will accomplish a simple
        instruction, and the {\bf MemOps} stage which accomplishes memory
        instruction, and the {\bf MemOps} stage which handles {\tt LOD} (load)
        read/write.
        and {\tt STO} (store) instructions.
        \begin{itemize}
        \begin{itemize}
        \item Loads stall instructions that access the register until it is
        \item Loads will stall the entire pipeline until complete.
                written to the register set.
        \item Condition codes are available upon completion of the ALU stage
        \item Condition codes are available upon completion
        \item Issuing an instruction to the memory unit while the memory unit
        \item Issuing an instruction to the memory while the memory is busy will
                is busy will stall the entire pipeline.  If the bus deadlocks,
                stall the entire pipeline.  If the bus deadlocks, only a reset
                only a reset will release the CPU.  (Watchdog timer, anyone?)
                will release the CPU.  (Watchdog timer, anyone?)
 
        \item The Zip CPU currently has no means of reading and acting on any
        \item The Zip CPU currently has no means of reading and acting on any
        error conditions on the bus.
        error conditions on the bus.
        \end{itemize}
        \end{itemize}
\item {\bf Write-Back}: Conditionally write back the result to the register
\item {\bf Write-Back}: Conditionally write back the result to the register
        set, applying the condition.  This routine is bi-re-entrant: either the
        set, applying the condition.  This routine is bi-entrant: either the
        memory or the simple instruction may request a register write.
        memory or the simple instruction may request a register write.
\end{enumerate}
\end{enumerate}
 
 
The Zip CPU does not support out of order execution.  Therefore, if the memory
The Zip CPU does not support out of order execution.  Therefore, if the memory
unit stalls, every other instruction stalls.  Memory stores, however, can take
unit stalls, every other instruction stalls.  Memory stores, however, can take
place concurrently with ALU operations, although memory reads cannot.
place concurrently with ALU operations, although memory reads (loads) cannot.
 
 
\iffalse
 
 
 
\section{Pipeline Logic}
 
How the CPU handles some instruction combinations can be telling when
 
determining what happens in the pipeline.  The following lists some examples:
 
\begin{itemize}
 
\item {\bf Delayed Branching}
 
 
 
        I had originally hoped to implement delayed branching.   My goal
 
        was that the compiler would handle any pipeline stall conditions so
 
        that the pipeline logic could be simpler within the CPU.  I ran into
 
        two problems with this.
 
 
 
        The first problem has to deal with debug mode.  When the debugger
 
        single steps an instruction, that instruction goes to completion.
 
        This means that if the instruction moves a value to the PC register,
 
        the PC register would now contain that value, indicating that the
 
        next instruction would be on the other side of the branch.  There's
 
        just no easy way around this: the entire CPU state must be captured
 
        by the registers, to include the program counter.  What value should
 
        the program counter be equal to?  The branch?  Fine.  The address
 
        you are branching to?  Fine.  The address of the delay slot?  Problem.
 
 
 
        The second problem with delayed branching is the idea of suspending
 
        processing for an interrupt.  Which address should the CPU return
 
        to upon completing the interrupt processing?  The branch?  Good.  The
 
        address after the branch?  Also good.  The address of the delay slot?
 
        Not so good.
 
 
 
        If you then add into this mess the idea that, if the CPU is running
 
        from a really slow memory such as the flash, the delay slot may never
 
        be filled before the branch is determined, then this makes even less
 
        sense.
 
 
 
        For all of these reasons, this CPU does not support delayed branching.
 
 
 
\item {\bf Register Result:} {\tt MOV R0,R1; MOV R1,R2 }
 
 
 
        What value does R2 get, the value of R1 before the first move or the
 
        value of R0?  The Zip CPU has been optimized so that neither of these
 
        instructions require a pipeline stall--unless an immediate were to
 
        be added to R1 in the second instruction.
 
 
 
        The ZIP CPU architecture requires that R2 must equal R0 at the end of
 
        this operation.  Even better, such combinations do not (normally)
 
        stall the pipeline.
 
 
 
\item {\bf Condition Codes Result:} {\tt CMP R0,R1;} {\tt MOV.EQ \$x,PC}
 
 
 
        At issue is the same item as above, save that the CMP instruction
 
        updates the flags that the MOV instruction depends upon.
 
 
 
        The Zip CPU architecture requires that condition codes must be updated
 
        and available immediately for the next instruction without stalling the
 
        pipeline.
 
 
 
\item {\bf Condition Codes Register Result:} {\tt CMP R0,R1; MOV CC,R2}
 
 
 
        At issue is the
 
        fact that the logic supporting the CC register is more complicated than
 
        the logic supporting any other register.
 
 
 
        The ZIP CPU will stall for a cycle cycle on this instruction.
 
\item {\bf Condition Codes Register Operand:} {\tt MOV R0,R1; MOV CC,R2}
 
 
 
        Unlike the previous case, this move prior to reading the {\tt CC}
 
        register does not impact the {\tt CC} register.  Therefore, this
 
        does not stall the bus, whereas the previous one would.
 
\end{itemize}
 
 
 
As I've studied  this, I find several approaches to handling pipeline
 
        issues.  These approaches (and their consequences) are listed below.
 
 
 
\begin{itemize}
 
\item {\bf All issued instructions complete, stages stall individually}
 
 
 
        What about a slow pre-fetch?
 
 
 
        Nominally, this works well: any issued instruction
 
        just runs to completion.  If there are four issued instructions in the
 
        pipeline, with the writeback instruction being a write-to-PC
 
        instruction, the other three instructions naturally finish.
 
 
 
        This approach fails when reading instructions from the flash,
 
        since such reads require N clocks to clocks to complete.  Thus
 
        there may be only one instruction in the pipeline if reading from flash,
 
        or a full pipeline if reading from cache.  Each of these approaches
 
        would produce a different response.
 
 
 
        For this reason, the Zip CPU works off of a different basis: All
 
        instructions that enter either the ALU or the memory unit will
 
        complete.  Stages still stall individually.
 
 
 
\item {\bf Issued instructions may be canceled}
 
 
 
        The problem here is that
 
        memory operations cannot be canceled: even reads may have side effects
 
        on peripherals that cannot be canceled later.  Further, in the case of
 
        an interrupt, it's difficult to know what to cancel.  What happens in
 
        a \hbox{\tt MOV.C \$x,PC} followed by a \hbox{\tt MOV \$y,PC}
 
        instruction?  Which get canceled?
 
 
 
        Because it isn't clear what would need to be canceled, the Zip CPU
 
        will not permit this combination.  A MOV to the PC register will be
 
        followed by a stall, and possibly many stalls, so that the second
 
        move to PC will never be executed.
 
 
 
\item {\bf All issued instructions complete.}
 
 
 
        In this example, we try all issued instructions complete, but the
 
        entire pipeline stalls if one stage is not filled.  In this approach,
 
        though, we again struggle with the problems associated with
 
        delayed branching.  Upon attempting to restart the processor, where
 
        do you restart it from?
 
 
 
\item {\bf Memory instructions must complete}
 
 
 
        All instructions that enter into the memory module {\em must}
 
        complete.  Issued instructions from the prefetch, decode, or operand
 
        read stages may or may not complete.  Jumps into code must be valid,
 
        so that interrupt returns may be valid.  All instructions entering the
 
        ALU complete.
 
 
 
        This looks to be the simplest approach.
 
        While the logic may be difficult, this appears to be the only
 
        re-entrant approach.
 
 
 
        A {\tt new\_pc} flag will be high anytime the PC changes in an
 
        unpredictable way (i.e., it doesn't increment).  This includes jumps
 
        as well as interrupts and interrupt returns.  Whenever this flag may
 
        go high, memory operations and ALU operations will stall until the
 
        result is known.  When the flag does go high, anything in the prefetch,
 
        decode, and read-op stage will be invalidated.
 
 
 
\end{itemize}
 
\fi
 
 
 
\section{Pipeline Stalls}
\section{Pipeline Stalls}
The processing pipeline can and will stall for a variety of reasons.  Some of
The processing pipeline can and will stall for a variety of reasons.  Some of
these are obvious, some less so.  These reasons are listed below:
these are obvious, some less so.  These reasons are listed below:
\begin{itemize}
\begin{itemize}
\item When the prefetch cache is exhausted
\item When the prefetch cache is exhausted
 
 
This should be obvious.  If the prefetch cache doesn't have the instruction
This reason should be obvious.  If the prefetch cache doesn't have the
in memory, the entire pipeline must stall until enough of the prefetch cache
instruction in memory, the entire pipeline must stall until enough of the
is loaded to support the next instruction.
prefetch cache is loaded to support the next instruction.
 
 
\item While waiting for the pipeline to load following any taken branch, jump,
\item While waiting for the pipeline to load following any taken branch, jump,
        return from interrupt or switch to interrupt context (6 clocks)
        return from interrupt or switch to interrupt context (5 stall cycles)
 
 
If the PC suddenly changes, the pipeline is subsequently cleared and needs to
If the PC suddenly changes, the pipeline is subsequently cleared and needs to
be reloaded.  Given that there are five stages to the pipeline, that accounts
be reloaded.  Given that there are five stages to the pipeline, that accounts
for five of the six delay clocks.  The last clock is lost in the prefetch
for four of the five stalls.  The stall cycle is lost in the pipelined prefetch
stage which needs at least one clock with a valid PC before it can produce
stage which needs at least one clock with a valid PC before it can produce
a new output.  Hence, six clocks will always be lost anytime the pipeline needs
a new output.
to be cleared.
 
 
The Zip CPU handles {\tt MOV \$X(PC),PC}, {\tt ADD \$X,PC}, and
 
{\tt LDI \$X,PC} instructions specially, however.  These instructions, when
 
not conditioned on the flags, can execute with only 3~stall cycles.
 
 
\item When reading from a prior register while also adding an immediate offset
\item When reading from a prior register while also adding an immediate offset
\begin{enumerate}
\begin{enumerate}
\item\ {\tt OPCODE ?,RA}
\item\ {\tt OPCODE ?,RA}
\item\ {\em (stall)}
\item\ {\em (stall)}
\item\ {\tt OPCODE I+RA,RB}
\item\ {\tt OPCODE I+RA,RB}
\end{enumerate}
\end{enumerate}
 
 
Since the addition of the immediate register within OpB decoding gets applied
Since the addition of the immediate register within OpB decoding gets applied
during the read operand stage so that it can be nicely settled before the ALU,
during the read operand stage so that it can be nicely settled before the ALU,
any instruction that will write back an operand must be separated from the
any instruction that will write back an operand must be separated from the
opcode that will read and apply an immediate offset by one instruction.  The
opcode that will read and apply an immediate offset by one instruction.  The
good news is that this stall can easily be mitigated by proper scheduling.
good news is that this stall can easily be mitigated by proper scheduling.
 
That is, any instruction that does not add an immediate to {\tt RA} may be
 
scheduled into the stall slot.
 
 
\item When writing to the CC or PC Register
\item When any write to either the CC or PC Register is followed by a memory
 
        operation
\begin{enumerate}
\begin{enumerate}
\item\ {\tt OPCODE RA,PC} {\em Ex: a branch opcode}
\item\ {\tt OPCODE RA,PC} {\em Ex: a branch opcode}
\item\ {\em (stall, even if jump not taken)}
\item\ {\em (stall, even if jump not taken)}
\item\ {\tt OPCODE RA,RB}
\item\ {\tt LOD \$X(RA),RB}
\end{enumerate}
\end{enumerate}
Since branches take place in the writeback stage, the Zip CPU will stall the
Since branches take place in the writeback stage, the Zip CPU will stall the
pipeline for one clock anytime there may be a possible jump.  This prevents
pipeline for one clock anytime there may be a possible jump.  This prevents
an instruction from executing a memory access after the jump but before the
an instruction from executing a memory access after the jump but before the
jump is recognized.
jump is recognized.
 
 
This stall cannot be mitigated through scheduling.
This stall may be mitigated by shuffling the operations immediately following
 
a potential branch so that an ALU operation follows the branch instead of a
 
memory operation.
 
 
\item When reading from the CC register after setting the flags
\item When reading from the CC register after setting the flags
\begin{enumerate}
\begin{enumerate}
\item\ {\tt ALUOP RA,RB}
\item\ {\tt ALUOP RA,RB} {\em Ex: a compare opcode}
\item\ {\em (stall}
\item\ {\em (stall)}
\item\ {\tt TST sys.ccv,CC}
\item\ {\tt TST sys.ccv,CC}
\item\ {\tt BZ somewhere}
\item\ {\tt BZ somewhere}
\end{enumerate}
\end{enumerate}
 
 
The reason for this stall is simply performance.  Many of the flags are
The reason for this stall is simply performance.  Many of the flags are
determined via combinatorial logic after the writeback instruction is
determined via combinatorial logic after the writeback instruction is
determined.  Trying to then place these into the input for one of the operands
determined.  Trying to then place these into the input for one of the operands
created a time delay loop that would no longer execute in a single 100~MHz
created a time delay loop that would no longer execute in a single 100~MHz
clock cycle.  (The time delay of the multiply within the ALU wasn't helping
clock cycle.  (The time delay of the multiply within the ALU wasn't helping
either \ldots).
either \ldots).
 
 
This stall may be eliminated via proper scheduling, by placing an instruction
This stall may be eliminated via proper scheduling, by placing an instruction
that does not set flags in between the ALU operation and the instruction
that does not set flags in between the ALU operation and the instruction
that references the CC register.  For example, {\tt MOV \$addr+PC,uPC}
that references the CC register.  For example, {\tt MOV \$addr+PC,uPC}
followed by an {\tt RTU} ({\tt OR \$GIE,CC}) instruction will not incur
followed by an {\tt RTU} ({\tt OR \$GIE,CC}) instruction will not incur
this stall, whereas an {\tt OR \$BREAKEN,CC} followed by an {\tt OR \$STEP,CC}
this stall, whereas an {\tt OR \$BREAKEN,CC} followed by an {\tt OR \$STEP,CC}
will incur the stall.
will incur the stall, while a {\tt LDI \$BREAKEN|\$STEP,CC} will not.
 
 
\item When waiting for a memory read operation to complete
\item When waiting for a memory read operation to complete
\begin{enumerate}
\begin{enumerate}
\item\ {\tt LOD address,RA}
\item\ {\tt LOD address,RA}
\item\ {\em (multiple stalls, bus dependent, 7 clocks best)}
\item\ {\em (multiple stalls, bus dependent, 4 clocks best)}
\item\ {\tt OPCODE I+RA,RB}
\item\ {\tt OPCODE I+RA,RB}
\end{enumerate}
\end{enumerate}
 
 
Remember, the ZIP CPU does not support out of order execution.  Therefore,
Remember, the Zip CPU does not support out of order execution.  Therefore,
anytime the memory unit becomes busy both the memory unit and the ALU must
anytime the memory unit becomes busy both the memory unit and the ALU must
stall until the memory unit is cleared.  This is especially true of a load
stall until the memory unit is cleared.  This is especially true of a load
instruction, which must still write its operand back to the register file.
instruction, which must still write its operand back to the register file.
Store instructions are different, since they can be busy with no impact on
Store instructions are different, since they can be busy with no impact on
later ALU write back operations.  Hence, only loads stall the pipeline.
later ALU write back operations.  Hence, only loads stall the pipeline.
 
 
This also assumes that the memory being accessed is a single cycle memory.
This also assumes that the memory being accessed is a single cycle memory.
Slower memories, such as the Quad SPI flash, will take longer--perhaps even
Slower memories, such as the Quad SPI flash, will take longer--perhaps even
as long as forty clocks.   During this time the CPU and the external bus
as long as forty clocks.   During this time the CPU and the external bus
will be busy, and unable to do anything else.
will be busy, and unable to do anything else.
 
 
\item Memory operation followed by a memory operation
\item Memory operation followed by a memory operation
\begin{enumerate}
\begin{enumerate}
\item\ {\tt STO address,RA}
\item\ {\tt STO address,RA}
\item\ {\em (multiple stalls, bus dependent, 7 clocks best)}
\item\ {\em (multiple stalls, bus dependent, 4 clocks best)}
\item\ {\tt LOD address,RB}
\item\ {\tt LOD address,RB}
\item\ {\em (multiple stalls, bus dependent, 7 clocks best)}
\item\ {\em (multiple stalls, bus dependent, 4 clocks best)}
\end{enumerate}
\end{enumerate}
 
 
In this case, the LOD instruction cannot start until the STALL is finished.
In this case, the LOD instruction cannot start until the STO is finished.
With proper scheduling, it is possible to do something in the ALU while the
With proper scheduling, it is possible to do something in the ALU while the
STO is busy, but otherwise this pipeline will stall waiting for it to complete.
memory unit is busy with the STO instruction, but otherwise this pipeline will
 
stall waiting for it to complete.
 
 
Note that even though the Wishbone bus can support pipelined accesses at
Note that even though the Wishbone bus can support pipelined accesses at
one access per clock, only the prefetch stage can take advantage of this.
one access per clock, only the prefetch stage can take advantage of this.
Load and Store instructions are stuck at one wishbone cycle per instruction.
Load and Store instructions are stuck at one wishbone cycle per instruction.
 
 
 
\item When waiting for a conditional memory read operation to complete
 
\begin{enumerate}
 
\item\ {\tt LOD.Z address,RA}
 
\item\ {\em (multiple stalls, bus dependent, 7 clocks best)}
 
\item\ {\tt OPCODE I+RA,RB}
 
\end{enumerate}
 
 
 
In this case, the Zip CPU doesn't warn the prefetch cache to get off the bus
 
two cycles before using the bus, so there's a potential for an extra three
 
cycle cost due to bus contention between the prefetch and the CPU.
 
 
 
This is true for both the LOD and the STO instructions, with the exception that
 
the STO instruction will continue in parallel with any ALU instructions that
 
follow it.
 
 
\end{itemize}
\end{itemize}
 
 
 
 
\chapter{Peripherals}\label{chap:periph}
\chapter{Peripherals}\label{chap:periph}
 
 
While the previous chapter describes a CPU in isolation, the Zip System
While the previous chapter describes a CPU in isolation, the Zip System
includes a minimum set of peripherals as well.  These peripherals are shown
includes a minimum set of peripherals as well.  These peripherals are shown
in Fig.~\ref{fig:zipsystem}
in Fig.~\ref{fig:zipsystem}
\begin{figure}\begin{center}
\begin{figure}\begin{center}
\includegraphics[width=3.5in]{../gfx/system.eps}
\includegraphics[width=3.5in]{../gfx/system.eps}
\caption{Zip System Peripherals}\label{fig:zipsystem}
\caption{Zip System Peripherals}\label{fig:zipsystem}
\end{center}\end{figure}
\end{center}\end{figure}
and described here.  They are designed to make
and described here.  They are designed to make
the Zip CPU more useful in an Embedded Operating System environment.
the Zip CPU more useful in an Embedded Operating System environment.
 
 
\section{Interrupt Controller}
\section{Interrupt Controller}
 
 
Perhaps the most important peripheral within the Zip System is the interrupt
Perhaps the most important peripheral within the Zip System is the interrupt
controller.  While the Zip CPU itself can only handle one interrupt, and has
controller.  While the Zip CPU itself can only handle one interrupt, and has
only the one interrupt state: disabled or enabled, the interrupt controller
only the one interrupt state: disabled or enabled, the interrupt controller
can make things more interesting.
can make things more interesting.
 
 
The Zip System interrupt controller module supports up to 15 interrupts, all
The Zip System interrupt controller module supports up to 15 interrupts, all
controlled from one register.  Bit~31 of the interrupt controller controls
controlled from one register.  Bit~31 of the interrupt controller controls
overall whether interrupts are enabled (1'b1) or disabled (1'b0).  Bits~16--30
overall whether interrupts are enabled (1'b1) or disabled (1'b0).  Bits~16--30
control whether individual interrupts are enabled (1'b0) or disabled (1'b0).
control whether individual interrupts are enabled (1'b0) or disabled (1'b0).
Bit~15 is an indicator showing whether or not any interrupt is active, and
Bit~15 is an indicator showing whether or not any interrupt is active, and
bits~0--15 indicate whether or not an individual interrupt is active.
bits~0--15 indicate whether or not an individual interrupt is active.
 
 
The interrupt controller has been designed so that bits can be controlled
The interrupt controller has been designed so that bits can be controlled
individually without having any knowledge of the rest of the controller
individually without having any knowledge of the rest of the controller
setting.  To enable an interrupt, write to the register with the high order
setting.  To enable an interrupt, write to the register with the high order
global enable bit set and the respective interrupt enable bit set.  No other
global enable bit set and the respective interrupt enable bit set.  No other
bits will be affected.  To disable an interrupt, write to the register with
bits will be affected.  To disable an interrupt, write to the register with
the high order global enable bit cleared and the respective interrupt enable
the high order global enable bit cleared and the respective interrupt enable
bit set.  To clear an interrupt, write a `1' to that interrupts status pin.
bit set.  To clear an interrupt, write a `1' to that interrupts status pin.
Zero's written to the register have no affect, save that a zero written to the
Zero's written to the register have no affect, save that a zero written to the
master enable will disable all interrupts.
master enable will disable all interrupts.
 
 
As an example, suppose you wished to enable interrupt \#4.  You would then
As an example, suppose you wished to enable interrupt \#4.  You would then
write to the register a {\tt 0x80100010} to enable interrupt \#4 and to clear
write to the register a {\tt 0x80100010} to enable interrupt \#4 and to clear
any past active state.  When you later wish to disable this interrupt, you would
any past active state.  When you later wish to disable this interrupt, you would
write a {\tt 0x00100010} to the register.  As before, this both disables the
write a {\tt 0x00100010} to the register.  As before, this both disables the
interrupt and clears the active indicator.  This also has the side effect of
interrupt and clears the active indicator.  This also has the side effect of
disabling all interrupts, so a second write of {\tt 0x80000000} may be necessary
disabling all interrupts, so a second write of {\tt 0x80000000} may be necessary
to re-enable any other interrupts.
to re-enable any other interrupts.
 
 
The Zip System currently hosts two interrupt controllers, a primary and a
The Zip System currently hosts two interrupt controllers, a primary and a
secondary.  The primary interrupt controller has one interrupt line which may
secondary.  The primary interrupt controller has one interrupt line which may
come from an external interrupt controller, and one interrupt line from the
come from an external interrupt controller, and one interrupt line from the
secondary controller.  Other primary interrupts include the system timers,
secondary controller.  Other primary interrupts include the system timers,
the jiffies interrupt, and the manual cache interrupt.  The secondary interrupt
the jiffies interrupt, and the manual cache interrupt.  The secondary interrupt
controller maintains an interrupt state for all of the processor accounting
controller maintains an interrupt state for all of the processor accounting
counters.
counters.
 
 
\section{Counter}
\section{Counter}
 
 
The Zip Counter is a very simple counter: it just counts.  It cannot be
The Zip Counter is a very simple counter: it just counts.  It cannot be
halted.  When it rolls over, it issues an interrupt.  Writing a value to the
halted.  When it rolls over, it issues an interrupt.  Writing a value to the
counter just sets the current value, and it starts counting again from that
counter just sets the current value, and it starts counting again from that
value.
value.
 
 
Eight counters are implemented in the Zip System for process accounting.
Eight counters are implemented in the Zip System for process accounting.
This may change in the future, as nothing as yet uses these counters.
This may change in the future, as nothing as yet uses these counters.
 
 
\section{Timer}
\section{Timer}
 
 
The Zip Timer is also very simple: it simply counts down to zero.  When it
The Zip Timer is also very simple: it simply counts down to zero.  When it
transitions from a one to a zero it creates an interrupt.
transitions from a one to a zero it creates an interrupt.
 
 
Writing any non-zero value to the timer starts the timer.  If the high order
Writing any non-zero value to the timer starts the timer.  If the high order
bit is set when writing to the timer, the timer becomes an interval timer and
bit is set when writing to the timer, the timer becomes an interval timer and
reloads its last start time on any interrupt.  Hence, to mark seconds, one
reloads its last start time on any interrupt.  Hence, to mark seconds, one
might set the timer to 100~million (the number of clocks per second), and
might set the timer to 100~million (the number of clocks per second), and
set the high bit.  Ever after, the timer will interrupt the CPU once per
set the high bit.  Ever after, the timer will interrupt the CPU once per
second (assuming a 100~MHz clock).  This reload capability also limits the
second (assuming a 100~MHz clock).  This reload capability also limits the
maximum timer value to $2^{31}-1$, rather than $2^{32}-1$.
maximum timer value to $2^{31}-1$, rather than $2^{32}-1$.
 
 
\section{Watchdog Timer}
\section{Watchdog Timer}
 
 
The watchdog timer is no different from any of the other timers, save for one
The watchdog timer is no different from any of the other timers, save for one
critical difference: the interrupt line from the watchdog
critical difference: the interrupt line from the watchdog
timer is tied to the reset line of the CPU.  Hence writing a `1' to the
timer is tied to the reset line of the CPU.  Hence writing a `1' to the
watchdog timer will always reset the CPU.
watchdog timer will always reset the CPU.
To stop the Watchdog timer, write a `0' to it.  To start it,
To stop the Watchdog timer, write a `0' to it.  To start it,
write any other number to it---as with the other timers.
write any other number to it---as with the other timers.
 
 
While the watchdog timer supports interval mode, it doesn't make as much sense
While the watchdog timer supports interval mode, it doesn't make as much sense
as it did with the other timers.
as it did with the other timers.
 
 
\section{Jiffies}
\section{Jiffies}
 
 
This peripheral is motivated by the Linux use of `jiffies' whereby a process
This peripheral is motivated by the Linux use of `jiffies' whereby a process
can request to be put to sleep until a certain number of `jiffies' have
can request to be put to sleep until a certain number of `jiffies' have
elapsed.  Using this interface, the CPU can read the number of `jiffies'
elapsed.  Using this interface, the CPU can read the number of `jiffies'
from the peripheral (it only has the one location in address space), add the
from the peripheral (it only has the one location in address space), add the
sleep length to it, and write the result back to the peripheral.  The zipjiffies
sleep length to it, and write the result back to the peripheral.  The zipjiffies
peripheral will record the value written to it only if it is nearer the current
peripheral will record the value written to it only if it is nearer the current
counter value than the last current waiting interrupt time.  If no other
counter value than the last current waiting interrupt time.  If no other
interrupts are waiting, and this time is in the future, it will be enabled.
interrupts are waiting, and this time is in the future, it will be enabled.
(There is currently no way to disable a jiffie interrupt once set, other
(There is currently no way to disable a jiffie interrupt once set, other
than to disable the interrupt line in the interrupt controller.)  The processor
than to disable the interrupt line in the interrupt controller.)  The processor
may then place this sleep request into a list among other sleep requests.
may then place this sleep request into a list among other sleep requests.
Once the timer expires, it would write the next Jiffy request to the peripheral
Once the timer expires, it would write the next Jiffy request to the peripheral
and wake up the process whose timer had expired.
and wake up the process whose timer had expired.
 
 
Indeed, the Jiffies register is nothing more than a glorified counter with
Indeed, the Jiffies register is nothing more than a glorified counter with
an interrupt.  Unlike the other counters, the Jiffies register cannot be set.
an interrupt.  Unlike the other counters, the Jiffies register cannot be set.
Writes to the jiffies register create an interrupt time.  When the Jiffies
Writes to the jiffies register create an interrupt time.  When the Jiffies
register later equals the value written to it, an interrupt will be asserted
register later equals the value written to it, an interrupt will be asserted
and the register then continues counting as though no interrupt had taken
and the register then continues counting as though no interrupt had taken
place.
place.
 
 
The purpose of this register is to support alarm times within a CPU.  To
The purpose of this register is to support alarm times within a CPU.  To
set an alarm for a particular process $N$ clocks in advance, read the current
set an alarm for a particular process $N$ clocks in advance, read the current
Jiffies value, and $N$, and write it back to the Jiffies register.  The
Jiffies value, and $N$, and write it back to the Jiffies register.  The
O/S must also keep track of values written to the Jiffies register.  Thus,
O/S must also keep track of values written to the Jiffies register.  Thus,
when an `alarm' trips, it should be removed from the list of alarms, the list
when an `alarm' trips, it should be removed from the list of alarms, the list
should be sorted, and the next alarm in terms of Jiffies should be written
should be sorted, and the next alarm in terms of Jiffies should be written
to the register.
to the register.
 
 
\section{Manual Cache}
\section{Direct Memory Access Controller}
 
 
 
The Direct Memory Access (DMA) controller can be used to either move memory
 
from one location to another, to read from a peripheral into memory, or to
 
write from a peripheral into memory all without CPU intervention.  Further,
 
since the DMA controller can issue (and does issue) pipeline wishbone accesses,
 
any DMA memory move will by nature be faster than a corresponding program
 
accomplishing the same move.  To put this to numbers, it may take a program
 
18~clocks per word transferred, whereas this DMA controller can move one
 
word in two clocks--provided it has bus access.  (The CPU gets priority over the
 
bus.)
 
 
 
When copying memory from one location to another, the DMA controller will
 
copy in units of a given transfer length--up to 1024 words at a time.  It will
 
read that transfer length into its internal buffer, and then write to the
 
destination address from that buffer.  If the CPU interrupts a DMA transfer,
 
it will release the bus, let the CPU complete whatever it needs to do, and then
 
restart its transfer by writing the contents of its internal buffer and then
 
re-entering its read cycle again.
 
 
 
When coupled with a peripheral, the DMA controller can be configured to start
 
a memory copy on an interrupt line going high.  Further, the controller can be
 
configured to issue reads from (or two) the same address instead of incrementing
 
the address at each clock.  The DMA completes once the total number of items
 
specified (not the transfer length) have been transferred.
 
 
 
In each case, once the transfer is complete and the DMA unit returns to
 
idle, the DMA will issue an interrupt.
 
 
The manual cache is an experimental setting that may not remain with the Zip
 
CPU for very long.  It is designed to facilitate running from FLASH or ROM
 
memory, although the pipeline prefetch cache really makes this need obsolete.
 
The manual
 
cache works by copying data from a wishbone address (range) into the cache
 
register, and then by making that memory available as memory to the Zip System.
 
It is a {\em manual cache} because the processor must first specify what
 
memory to copy, and then once copied the processor can only access the cache
 
memory by the cache memory location.  There is no transparency.  It is perhaps
 
best described as a combination DMA controller and local memory.
 
 
 
Worse, this cache is likely going to be removed from the ZipSystem.  Having used
 
the ZipSystem now for some time, I have yet to find a need or use for the manual
 
cache.  I will likely replace this peripheral with a proper DMA controller.
 
 
 
\chapter{Operation}\label{chap:ops}
\chapter{Operation}\label{chap:ops}
 
 
The Zip CPU, and even the Zip System, is not a System on a Chip (SoC).  It
The Zip CPU, and even the Zip System, is not a System on a Chip (SoC).  It
needs to be connected to its operational environment in order to be used.
needs to be connected to its operational environment in order to be used.
Specifically, some per system adjustments need to be made:
Specifically, some per system adjustments need to be made:
\begin{enumerate}
\begin{enumerate}
\item The Zip System depends upon an external 32-bit Wishbone bus.  This
\item The Zip System depends upon an external 32-bit Wishbone bus.  This
        must exist, and must be connected to the Zip CPU for it to work.
        must exist, and must be connected to the Zip CPU for it to work.
\item The Zip System needs to be told of its {\tt RESET\_ADDRESS}.  This is
\item The Zip System needs to be told of its {\tt RESET\_ADDRESS}.  This is
        the program counter of the first instruction following a reset.
        the program counter of the first instruction following a reset.
\item If you want the Zip System to start up on its own, you will need to
\item If you want the Zip System to start up on its own, you will need to
        set the {\tt START\_HALTED} parameter to zero.  Otherwise, if you
        set the {\tt START\_HALTED} parameter to zero.  Otherwise, if you
        wish to manually start the CPU, that is if upon reset you want the
        wish to manually start the CPU, that is if upon reset you want the
        CPU start start in its halted, reset state, then set this parameter to
        CPU start start in its halted, reset state, then set this parameter to
        one.
        one.
\item The third parameter to set is the number of interrupts you will be
\item The third parameter to set is the number of interrupts you will be
        providing from external to the CPU.  This can be anything from one
        providing from external to the CPU.  This can be anything from one
        to nine, but it cannot be zero.  (Wire this line to a 1'b0 if you
        to nine, but it cannot be zero.  (Wire this line to a 1'b0 if you
        do not wish to support any external interrupts.)
        do not wish to support any external interrupts.)
\item Finally, you need to place into some wishbone accessible address, whether
\item Finally, you need to place into some wishbone accessible address, whether
        RAM or (more likely) ROM, the initial instructions for the CPU.
        RAM or (more likely) ROM, the initial instructions for the CPU.
\end{enumerate}
\end{enumerate}
If you have enabled your CPU to start automatically, then upon power up the
If you have enabled your CPU to start automatically, then upon power up the
CPU will immediately start executing your instructions.
CPU will immediately start executing your instructions.
 
 
This is, however, not how I have used the Zip CPU.  I have instead used the
This is, however, not how I have used the Zip CPU.  I have instead used the
ZIP CPU in a more controlled environment.  For me, the CPU starts in a
Zip CPU in a more controlled environment.  For me, the CPU starts in a
halted state, and waits to be told to start.  Further, the RESET address is a
halted state, and waits to be told to start.  Further, the RESET address is a
location in RAM.  After bringing up the board I am using, and further the
location in RAM.  After bringing up the board I am using, and further the
bus that is on it, the RAM memory is then loaded externally with the program
bus that is on it, the RAM memory is then loaded externally with the program
I wish the Zip System to run.  Once the RAM is loaded, I release the CPU.
I wish the Zip System to run.  Once the RAM is loaded, I release the CPU.
The CPU then runs until its halt condition, at which point its task is
The CPU then runs until its halt condition, at which point its task is
complete.
complete.
 
 
Eventually, I intend to place an operating system onto the ZipSystem, I'm
Eventually, I intend to place an operating system onto the ZipSystem, I'm
just not there yet.
just not there yet.
 
 
 
The rest of this chapter examines some common programming constructs, and
 
how they might be applied to the Zip System.
 
 
 
\section{Example: Idle Task}
 
One task every operating system needs is the idle task, the task that takes
 
place when nothing else can run.  On the Zip CPU, this task is quite simple,
 
and it is shown in assemble in Tbl.~\ref{tbl:idle-asm}.
 
\begin{table}\begin{center}
 
\begin{tabular}{ll}
 
{\tt idle\_task:} \\
 
&        {\em ; Wait for the next interrupt, then switch to supervisor task} \\
 
&        {\tt WAIT} \\
 
&        {\em ; When we come back, it's because the supervisor wishes to} \\
 
&        {\em ; wait for an interrupt again, so go back to the top.} \\
 
&        {\tt BRA idle\_task} \\
 
\end{tabular}
 
\caption{Example Idle Loop}\label{tbl:idle-asm}
 
\end{center}\end{table}
 
When this task runs, the CPU will fill up all of the pipeline stages up the
 
ALU.  The {\tt WAIT} instruction, upon leaving the ALU, places the CPU into
 
a sleep state where nothing more moves.  Sure, there may be some more settling,
 
the pipe cache continue to read until full, other instructions may issue until
 
the pipeline fills, but then everything will stall.  Then, once an interrupt
 
takes place, control passes to the supervisor task to handle the interrupt.
 
When control passes back to this task, it will be on the next instruction.
 
Since that next instruction sends us back to the top of the task, the idle
 
task thus does nothing but wait for an interrupt.
 
 
 
This should be the lowest priority task, the task that runs when nothing else
 
can.  It will help lower the FPGA power usage overall---at least its dynamic
 
power usage.
 
 
 
\section{Example: Memory Copy}
 
One common operation is that of a memory move or copy.  Consider the C code
 
shown in Tbl.~\ref{tbl:memcp-c}.
 
\begin{table}\begin{center}
 
\parbox{4in}{\begin{tabbing}
 
{\tt void} \= {\tt memcp(void *dest, void *src, int len) \{} \\
 
        \> {\tt for(int i=0; i<len; i++)} \\
 
        \> \hspace{0.2in} {\tt *dest++ = *src++;} \\
 
\}
 
\end{tabbing}}
 
\caption{Example Memory Copy code in C}\label{tbl:memcp-c}
 
\end{center}\end{table}
 
This same code can be translated in Zip Assembly as shown in
 
Tbl.~\ref{tbl:memcp-asm}.
 
\begin{table}\begin{center}
 
\begin{tabular}{ll}
 
memcp: \\
 
&        {\em ; R0 = *dest, R1 = *src, R2 = LEN} \\
 
&        {\em ; The following will operate in 17 clocks per word minus one clock} \\
 
&        {\tt CMP 0,R2} \\
 
&        {\tt LOD.Z -1(SP),PC} {\em ; A conditional return }\\
 
&        {\em ; (One stall on potentially writing to PC)} \\
 
&        {\tt LOD (R1),R3} \\
 
&        {\em ; (4 stalls, cannot be scheduled away)} \\
 
&        {\tt STO R3,(R2)} {\em ; (4 schedulable stalls, has no impact now)} \\
 
&        {\tt ADD 1,R1} \\
 
&        {\tt SUB 1,R2} \\
 
&        {\tt BNZ loop} \\
 
&        {\em ; (5 stalls, if branch taken, to clear and refill the pipeline)} \\
 
&        {\tt RET} \\
 
\end{tabular}
 
\caption{Example Memory Copy code in Zip Assembly}\label{tbl:memcp-asm}
 
\end{center}\end{table}
 
This example points out several things associated with the Zip CPU.  First,
 
a straightforward implementation of a for loop is not the fastest loop
 
structure.  For this reason, we have placed the test to continue at the
 
end.  Second, all pointers are {\tt void} pointers to arbitrary 32--bit
 
data types.  The Zip CPU does not have explicit support for smaller or larger
 
data types, and so this memory copy cannot be applied at a byte level.
 
Third, we've optimized the conditional jump to a return instruction into a
 
conditional return instruction.
 
 
 
\section{Context Switch}
 
 
 
Fundamental to any multiprocessing system is the ability to switch from one
 
task to the next.  In the ZipSystem, this is accomplished in one of a couple
 
ways.  The first step is that an interrupt happens.  Anytime an interrupt
 
happens, the CPU needs to execute the following tasks in supervisor mode:
 
\begin{enumerate}
 
\item Check for a trap instruction.  That  is, if the user task requested a
 
        trap, we may not wish to adjust the context, check interrupts, or call
 
        the scheduler.  Tbl.~\ref{tbl:trap-check}
 
\begin{table}\begin{center}
 
\begin{tabular}{ll}
 
{\tt return\_to\_user:} \\
 
&       {\em; The instruction before the context switch processing must} \\
 
&       {\em; be the RTU instruction that enacted user mode in the first} \\
 
&       {\em; place.  We show it here just for reference.} \\
 
&       {\tt RTU} \\
 
{\tt trap\_check:} \\
 
&       {\tt MOV uCC,R0} \\
 
&       {\tt TST \$TRAP,R0} \\
 
&       {\tt BNZ swap\_out} \\
 
&       {; \em Do something here to execute the trap} \\
 
&       {; \em Don't need to call the scheduler, so we can just return} \\
 
&       {\tt BRA return\_to\_user} \\
 
\end{tabular}
 
\caption{Checking for whether the user issued a TRAP instruction}\label{tbl:trap-check}
 
\end{center}\end{table}
 
        shows the rudiments of this code, while showing nothing of how the
 
        actual trap would be implemented.
 
 
 
You may also wish to note that the instruction before the first instruction
 
in our context swap {\em must be} a return to userspace instruction.
 
Remember, the supervisor process is re--entered where it left off.  This is
 
different from many other processors that enter interrupt mode at some vector
 
or other.  In this case, we always enter supervisor mode right where we last
 
left.\footnote{The one exception to this rule is upon reset where supervisor
 
mode is entered at a pre--programmed wishbone memory address.}
 
 
 
\item Capture user counters.  If the operating system is keeping track of
 
        system usage via the accounting counters, those counters need to be
 
        copied and accumulated into some master counter at this point.
 
 
 
\item Preserve the old context.  This involves pushing all the user registers
 
        onto the user stack and then copying the resulting stack address
 
        into the tasks task structure, as shown in Tbl.~\ref{tbl:context-out}.
 
\begin{table}\begin{center}
 
\begin{tabular}{ll}
 
{\tt swap\_out:} \\
 
&        {\tt MOV -15(uSP),R1} \\
 
&        {\tt STO R1,stack(R12)} \\
 
&        {\tt MOV uPC,R0} \\
 
&        {\tt STO R0,15(R1)} \\
 
&        {\tt MOV uCC,R0} \\
 
&        {\tt STO R0,14(R1)} \\
 
&       {\em ; We can skip storing the stack, uSP, since it'll be stored}\\
 
&       {\em ; elsewhere (in the task structure) }\\
 
&        {\tt MOV uR13,R0} \\
 
&        {\tt STO R0,13(R1)} \\
 
        & \ldots {\em ; Need to repeat for all user registers} \\
 
&        {\tt MOV uR0,R0} \\
 
&        {\tt STO R0,1(R1)} \\
 
\end{tabular}
 
\caption{Example Storing User Task Context}\label{tbl:context-out}
 
\end{center}\end{table}
 
For the sake of discussion, we assume the supervisor maintains a
 
pointer to the current task's structure in supervisor register
 
{\tt R12}, and that {\tt stack} is an offset to the beginning of this
 
structure indicating where the stack pointer is to be kept within it.
 
 
 
        For those who are still interested, the full code for this context
 
        save can be found as an assembler macro within the assembler
 
        include file, {\tt sys.i}.
 
 
 
\item Reset the watchdog timer.  If you are using the watchdog timer, it should
 
        be reset on a context swap, to know that things are still working.
 
        Example code for this is shown in Tbl.~\ref{tbl:reset-watchdog}.
 
\begin{table}\begin{center}
 
\begin{tabular}{ll}
 
\multicolumn{2}{l}{{\tt `define WATCHDOG\_ADDRESS 32'hc000\_0002}}\\
 
\multicolumn{2}{l}{{\tt `define WATCHDOG\_TICKS 32'd1\_000\_000} {; \em = 10 ms}}\\
 
&       {\tt LDI WATCHDOG\_ADDRESS,R0} \\
 
&       {\tt LDI WATCHDOG\_TICKS,R1} \\
 
&       {\tt STO R1,(R0)}
 
\end{tabular}
 
\caption{Example Watchdog Reset}\label{tbl:reset-watchdog}
 
\end{center}\end{table}
 
 
 
\item Interrupt handling.  An interrupt handler within the Zip System is nothing
 
        more than a task.  At context swap time, the supervisor needs to
 
        disable all of the interrupts that have tripped, and then enable
 
        all of the tasks that would deal with each of these interrupts.
 
        These can be user tasks, run at higher priority than any other user
 
        tasks.  Either way, they will need to re--enable their own interrupt
 
        themselves, if the interrupt is still relevant.
 
 
 
        An example of this master interrut handling is shown in
 
        Tbl.~\ref{tbl:pre-handler}.
 
\begin{table}\begin{center}
 
\begin{tabular}{ll}
 
{\tt pre\_handler:} \\
 
&       {\tt LDI PIC\_ADDRESS,R0 } \\
 
&       {\em ; Start by grabbing the interrupt state from the interrupt}\\
 
&       {\em ; controller.  We'll store this into the register R7 so that }\\
 
&       {\em ; we can keep and preserve this information for the scheduler}\\
 
&       {\em ; to use later. }\\
 
&       {\tt LOD (R0),R1} \\
 
&       {\tt MOV R1,R7 } \\
 
&       {\em ; As a next step, we need to acknowledge and disable all active}\\
 
&       {\em ; interrupts. We'll start by calculating all of our active}\\
 
&       {\em ; interrupts.}\\
 
&       {\tt AND 0x07fff,R1 } \\
 
&       {\em ; Put the active interrupts into the upper half of R1} \\
 
&       {\tt ROL 16,R1 } \\
 
&       {\tt LDILO 0x0ffff,R1   } \\
 
&       {\tt AND R7,R1}\\
 
&       {\em ; Acknowledge and disable active interrupts}\\
 
&       {\em ; This also disables all interrupts from the controller, so}\\
 
&       {\em ; we'll need to re-enable interrupts in general shortly } \\
 
&       {\tt STO R1,(R0) } \\
 
&       {\em ; We leave our active interrupt mask in R7 so the scheduler can}\\
 
&       {\em ; release any tasks that depended upon them. } \\
 
\end{tabular}
 
\caption{Example checking for active interrupts}\label{tbl:pre-handler}
 
\end{center}\end{table}
 
 
 
\item Calling the scheduler.  This needs to be done to pick the next task
 
        to switch to.  It may be an interrupt handler, or it may  be a normal
 
        user task.  From a priority standpoint, it would make sense that the
 
        interrupt handlers all have a higher priority than the user tasks,
 
        and that once they have been called the user tasks may then be called
 
        again.  If no task is ready to run, run the idle task to wait for an
 
        interrupt.
 
 
 
        This suggests a minimum of four task priorities:
 
        \begin{enumerate}
 
        \item Interrupt handlers, executed with their interrupts disabled
 
        \item Device drivers, executed with interrupts re-enabled
 
        \item User tasks
 
        \item The idle task, executed when nothing else is able to execute
 
        \end{enumerate}
 
 
 
        For our purposes here, we'll just assume that a pointer to the current
 
        task is maintained in {\tt R12}, that a {\tt JSR scheduler} is
 
        called, and that the next current task is likewise placed into
 
        {\tt R12}.
 
 
 
\item Restore the new tasks context.  Given that the scheduler has returned a
 
        task that can be run at this time, the stack pointer needs to be
 
        pulled out of the tasks task structure, placed into the user
 
        register, and then the rest of the user registers need to be popped
 
        back off of the stack to run this task.  An example of this is
 
        shown in Tbl.~\ref{tbl:context-in},
 
\begin{table}\begin{center}
 
\begin{tabular}{ll}
 
{\tt swap\_in:} \\
 
&       {\tt LOD stack(R12),R1} \\
 
&       {\tt MOV 15(R1),uSP} \\
 
&       {\tt LOD 15(R1),R0} \\
 
&       {\tt MOV R0,uPC} \\
 
&       {\tt LOD 14(R1),R0} \\
 
&       {\tt MOV R0,uCC} \\
 
&       {\tt LOD 13(R1),R0} \\
 
&       {\tt MOV R0,uR12} \\
 
        & \ldots {\em ; Need to repeat for all user registers} \\
 
&       {\tt LOD 1(R1),R0} \\
 
&       {\tt MOV R0,uR0} \\
 
&       {\tt BRA return\_to\_user} \\
 
\end{tabular}
 
\caption{Example Restoring User Task Context}\label{tbl:context-in}
 
\end{center}\end{table}
 
        assuming as before that the task
 
        pointer is found in supervisor register {\tt R12}.
 
        As with storing the user context, the full code associated with
 
        restoring the user context can be found in the assembler include
 
        file, {\tt sys.i}.
 
 
 
\item Clear the userspace accounting registers.  In order to keep track of
 
        per process system usage, these registers need to be cleared before
 
        reactivating the userspace process.  That way, upon the next
 
        interrupt, we'll know how many clocks the userspace program has
 
        encountered, and how many instructions it was able to issue in
 
        those many clocks.
 
 
 
\item Jump back to the instruction just before saving the last tasks context,
 
        because that location in memory contains the return from interrupt
 
        command that we are going to need to execute, in order to guarantee
 
        that we return back here again.
 
\end{enumerate}
 
 
\chapter{Registers}\label{chap:regs}
\chapter{Registers}\label{chap:regs}
 
 
The ZipSystem registers fall into two categories, ZipSystem internal registers
The ZipSystem registers fall into two categories, ZipSystem internal registers
accessed via the ZipCPU shown in Tbl.~\ref{tbl:zpregs},
accessed via the ZipCPU shown in Tbl.~\ref{tbl:zpregs},
\begin{table}[htbp]
\begin{table}[htbp]
\begin{center}\begin{reglist}
\begin{center}\begin{reglist}
PIC   & \scalebox{0.8}{\tt 0xc0000000} & 32 & R/W & Primary Interrupt Controller \\\hline
PIC   & \scalebox{0.8}{\tt 0xc0000000} & 32 & R/W & Primary Interrupt Controller \\\hline
WDT   & \scalebox{0.8}{\tt 0xc0000001} & 32 & R/W & Watchdog Timer \\\hline
WDT   & \scalebox{0.8}{\tt 0xc0000001} & 32 & R/W & Watchdog Timer \\\hline
CCHE  & \scalebox{0.8}{\tt 0xc0000002} & 32 & R/W & Manual Cache Controller \\\hline
  & \scalebox{0.8}{\tt 0xc0000002} & 32 & R/W & {\em (Reserved for future use)} \\\hline
CTRIC & \scalebox{0.8}{\tt 0xc0000003} & 32 & R/W & Secondary Interrupt Controller \\\hline
CTRIC & \scalebox{0.8}{\tt 0xc0000003} & 32 & R/W & Secondary Interrupt Controller \\\hline
TMRA  & \scalebox{0.8}{\tt 0xc0000004} & 32 & R/W & Timer A\\\hline
TMRA  & \scalebox{0.8}{\tt 0xc0000004} & 32 & R/W & Timer A\\\hline
TMRB  & \scalebox{0.8}{\tt 0xc0000005} & 32 & R/W & Timer B\\\hline
TMRB  & \scalebox{0.8}{\tt 0xc0000005} & 32 & R/W & Timer B\\\hline
TMRC  & \scalebox{0.8}{\tt 0xc0000006} & 32 & R/W & Timer C\\\hline
TMRC  & \scalebox{0.8}{\tt 0xc0000006} & 32 & R/W & Timer C\\\hline
JIFF  & \scalebox{0.8}{\tt 0xc0000007} & 32 & R/W & Jiffies \\\hline
JIFF  & \scalebox{0.8}{\tt 0xc0000007} & 32 & R/W & Jiffies \\\hline
MTASK  & \scalebox{0.8}{\tt 0xc0000008} & 32 & R/W & Master Task Clock Counter \\\hline
MTASK  & \scalebox{0.8}{\tt 0xc0000008} & 32 & R/W & Master Task Clock Counter \\\hline
MMSTL  & \scalebox{0.8}{\tt 0xc0000009} & 32 & R/W & Master Stall Counter \\\hline
MMSTL  & \scalebox{0.8}{\tt 0xc0000009} & 32 & R/W & Master Stall Counter \\\hline
MPSTL  & \scalebox{0.8}{\tt 0xc000000a} & 32 & R/W & Master Pre--Fetch Stall Counter \\\hline
MPSTL  & \scalebox{0.8}{\tt 0xc000000a} & 32 & R/W & Master Pre--Fetch Stall Counter \\\hline
MICNT  & \scalebox{0.8}{\tt 0xc000000b} & 32 & R/W & Master Instruction Counter\\\hline
MICNT  & \scalebox{0.8}{\tt 0xc000000b} & 32 & R/W & Master Instruction Counter\\\hline
UTASK  & \scalebox{0.8}{\tt 0xc000000c} & 32 & R/W & User Task Clock Counter \\\hline
UTASK  & \scalebox{0.8}{\tt 0xc000000c} & 32 & R/W & User Task Clock Counter \\\hline
UMSTL  & \scalebox{0.8}{\tt 0xc000000d} & 32 & R/W & User Stall Counter \\\hline
UMSTL  & \scalebox{0.8}{\tt 0xc000000d} & 32 & R/W & User Stall Counter \\\hline
UPSTL  & \scalebox{0.8}{\tt 0xc000000e} & 32 & R/W & User Pre--Fetch Stall Counter \\\hline
UPSTL  & \scalebox{0.8}{\tt 0xc000000e} & 32 & R/W & User Pre--Fetch Stall Counter \\\hline
UICNT  & \scalebox{0.8}{\tt 0xc000000f} & 32 & R/W & User Instruction Counter\\\hline
UICNT  & \scalebox{0.8}{\tt 0xc000000f} & 32 & R/W & User Instruction Counter\\\hline
 
DMACTRL  & \scalebox{0.8}{\tt 0xc0000010} & 32 & R/W & DMA Control Register\\\hline
 
DMALEN  & \scalebox{0.8}{\tt 0xc0000011} & 32 & R/W & DMA total transfer length\\\hline
 
DMASRC  & \scalebox{0.8}{\tt 0xc0000012} & 32 & R/W & DMA source address\\\hline
 
DMADST  & \scalebox{0.8}{\tt 0xc0000013} & 32 & R/W & DMA destination address\\\hline
% Cache  & \scalebox{0.8}{\tt 0xc0100000} & & & Base address of the Cache memory\\\hline
% Cache  & \scalebox{0.8}{\tt 0xc0100000} & & & Base address of the Cache memory\\\hline
\end{reglist}
\end{reglist}
\caption{Zip System Internal/Peripheral Registers}\label{tbl:zpregs}
\caption{Zip System Internal/Peripheral Registers}\label{tbl:zpregs}
\end{center}\end{table}
\end{center}\end{table}
and the two debug registers shown in Tbl.~\ref{tbl:dbgregs}.
and the two debug registers shown in Tbl.~\ref{tbl:dbgregs}.
\begin{table}[htbp]
\begin{table}[htbp]
\begin{center}\begin{reglist}
\begin{center}\begin{reglist}
ZIPCTRL & 0 & 32 & R/W & Debug Control Register \\\hline
ZIPCTRL & 0 & 32 & R/W & Debug Control Register \\\hline
ZIPDATA & 1 & 32 & R/W & Debug Data Register \\\hline
ZIPDATA & 1 & 32 & R/W & Debug Data Register \\\hline
\end{reglist}
\end{reglist}
\caption{Zip System Debug Registers}\label{tbl:dbgregs}
\caption{Zip System Debug Registers}\label{tbl:dbgregs}
\end{center}\end{table}
\end{center}\end{table}
 
 
\section{Peripheral Registers}
\section{Peripheral Registers}
The peripheral registers, listed in Tbl.~\ref{tbl:zpregs}, are shown in the
The peripheral registers, listed in Tbl.~\ref{tbl:zpregs}, are shown in the
CPU's address space.  These may be accessed by the CPU at these addresses,
CPU's address space.  These may be accessed by the CPU at these addresses,
and when so accessed will respond as described in Chapt.~\ref{chap:periph}.
and when so accessed will respond as described in Chapt.~\ref{chap:periph}.
These registers will be discussed briefly again here.
These registers will be discussed briefly again here.
 
 
The Zip CPU Interrupt controller has four different types of bits, as shown in
The Zip CPU Interrupt controller has four different types of bits, as shown in
Tbl.~\ref{tbl:picbits}.
Tbl.~\ref{tbl:picbits}.
\begin{table}\begin{center}
\begin{table}\begin{center}
\begin{bitlist}
\begin{bitlist}
31 & R/W & Master Interrupt Enable\\\hline
31 & R/W & Master Interrupt Enable\\\hline
30\ldots 16 & R/W & Interrupt Enables, write '1' to change\\\hline
30\ldots 16 & R/W & Interrupt Enables, write '1' to change\\\hline
15 & R & Current Master Interrupt State\\\hline
15 & R & Current Master Interrupt State\\\hline
15\ldots 0 & R/W & Input Interrupt states, write '1' to clear\\\hline
15\ldots 0 & R/W & Input Interrupt states, write '1' to clear\\\hline
\end{bitlist}
\end{bitlist}
\caption{Interrupt Controller Register Bits}\label{tbl:picbits}
\caption{Interrupt Controller Register Bits}\label{tbl:picbits}
\end{center}\end{table}
\end{center}\end{table}
The high order bit, or bit--31, is the master interrupt enable bit.  When this
The high order bit, or bit--31, is the master interrupt enable bit.  When this
bit is set, then any time an interrupt occurs the CPU will be interrupted and
bit is set, then any time an interrupt occurs the CPU will be interrupted and
will switch to supervisor mode, etc.
will switch to supervisor mode, etc.
 
 
Bits 30~\ldots 16 are interrupt enable bits.  Should the interrupt line go
Bits 30~\ldots 16 are interrupt enable bits.  Should the interrupt line go
ghile while enabled, an interrupt will be generated.  To set an interrupt enable
ghile while enabled, an interrupt will be generated.  To set an interrupt enable
bit, one needs to write the master interrupt enable while writing a `1' to this
bit, one needs to write the master interrupt enable while writing a `1' to this
the bit.  To clear, one need only write a `0' to the master interrupt enable,
the bit.  To clear, one need only write a `0' to the master interrupt enable,
while leaving this line high.
while leaving this line high.
 
 
Bits 15\ldots 0 are the current state of the interrupt vector.  Interrupt lines
Bits 15\ldots 0 are the current state of the interrupt vector.  Interrupt lines
trip when they go high, and remain tripped until they are acknowledged.  If
trip when they go high, and remain tripped until they are acknowledged.  If
the interrupt goes high for longer than one pulse, it may be high when a clear
the interrupt goes high for longer than one pulse, it may be high when a clear
is requested.  If so, the interrupt will not clear.  The line must go low
is requested.  If so, the interrupt will not clear.  The line must go low
again before the status bit can be cleared.
again before the status bit can be cleared.
 
 
As an example, consider the following scenario where the Zip CPU supports four
As an example, consider the following scenario where the Zip CPU supports four
interrupts, 3\ldots0.
interrupts, 3\ldots0.
\begin{enumerate}
\begin{enumerate}
\item The Supervisor will first, while in the interrupts disabled mode,
\item The Supervisor will first, while in the interrupts disabled mode,
        write a {\tt 32'h800f000f} to the controller.  The supervisor may then
        write a {\tt 32'h800f000f} to the controller.  The supervisor may then
        switch to the user state with interrupts enabled.
        switch to the user state with interrupts enabled.
\item When an interrupt occurs, the supervisor will switch to the interrupt
\item When an interrupt occurs, the supervisor will switch to the interrupt
        state.  It will then cycle through the interrupt bits to learn which
        state.  It will then cycle through the interrupt bits to learn which
        interrupt handler to call.
        interrupt handler to call.
\item If the interrupt handler expects more interrupts, it will clear its
\item If the interrupt handler expects more interrupts, it will clear its
        current interrupt when it is done handling the interrupt in question.
        current interrupt when it is done handling the interrupt in question.
        To do this, it will write a '1' to the low order interrupt mask,
        To do this, it will write a '1' to the low order interrupt mask,
        such as writing a {\tt 32'h80000001}.
        such as writing a {\tt 32'h80000001}.
\item If the interrupt handler does not expect any more interrupts, it will
\item If the interrupt handler does not expect any more interrupts, it will
        instead clear the interrupt from the controller by writing a
        instead clear the interrupt from the controller by writing a
        {\tt 32'h00010001} to the controller.
        {\tt 32'h00010001} to the controller.
\item Once all interrupts have been handled, the supervisor will write a
\item Once all interrupts have been handled, the supervisor will write a
        {\tt 32'h80000000} to the interrupt register to re-enable interrupt
        {\tt 32'h80000000} to the interrupt register to re-enable interrupt
        generation.
        generation.
\item The supervisor should also check the user trap bit, and possible soft
\item The supervisor should also check the user trap bit, and possible soft
        interrupt bits here, but this action has nothing to do with the
        interrupt bits here, but this action has nothing to do with the
        interrupt control register.
        interrupt control register.
\item The supervisor will then leave interrupt mode, possibly adjusting
\item The supervisor will then leave interrupt mode, possibly adjusting
        whichever task is running, by executing a return from interrupt
        whichever task is running, by executing a return from interrupt
        command.
        command.
\end{enumerate}
\end{enumerate}
 
 
Leaving the interrupt controller, we show the timer registers bit definitions
Leaving the interrupt controller, we show the timer registers bit definitions
in Tbl.~\ref{tbl:tmrbits}.
in Tbl.~\ref{tbl:tmrbits}.
\begin{table}\begin{center}
\begin{table}\begin{center}
\begin{bitlist}
\begin{bitlist}
31 & R/W & Auto-Reload\\\hline
31 & R/W & Auto-Reload\\\hline
30\ldots 0 & R/W & Current timer value\\\hline
30\ldots 0 & R/W & Current timer value\\\hline
\end{bitlist}
\end{bitlist}
\caption{Timer Register Bits}\label{tbl:tmrbits}
\caption{Timer Register Bits}\label{tbl:tmrbits}
\end{center}\end{table}
\end{center}\end{table}
As you may recall, the timer just counts down to zero and then trips an
As you may recall, the timer just counts down to zero and then trips an
interrupt.  Writing to the current timer value sets that value, and reading
interrupt.  Writing to the current timer value sets that value, and reading
from it returns that value.  Writing to the current timer value while also
from it returns that value.  Writing to the current timer value while also
setting the auto--reload bit will send the timer into an auto--reload mode.
setting the auto--reload bit will send the timer into an auto--reload mode.
In this mode, upon setting its interrupt bit for one cycle, the timer will
In this mode, upon setting its interrupt bit for one cycle, the timer will
also reset itself back to the value of the timer that was written to it when
also reset itself back to the value of the timer that was written to it when
the auto--reload option was written to it.  To clear and stop the timer,
the auto--reload option was written to it.  To clear and stop the timer,
just simply write a `32'h00' to this register.
just simply write a `32'h00' to this register.
 
 
The Jiffies register is somewhat similar in that the register always changes.
The Jiffies register is somewhat similar in that the register always changes.
In this case, the register counts up, whereas the timer always counted down.
In this case, the register counts up, whereas the timer always counted down.
Reads from this register, as shown in Tbl.~\ref{tbl:jiffybits},
Reads from this register, as shown in Tbl.~\ref{tbl:jiffybits},
\begin{table}\begin{center}
\begin{table}\begin{center}
\begin{bitlist}
\begin{bitlist}
31\ldots 0 & R & Current jiffy value\\\hline
31\ldots 0 & R & Current jiffy value\\\hline
31\ldots 0 & W & Value/time of next interrupt\\\hline
31\ldots 0 & W & Value/time of next interrupt\\\hline
\end{bitlist}
\end{bitlist}
\caption{Jiffies Register Bits}\label{tbl:jiffybits}
\caption{Jiffies Register Bits}\label{tbl:jiffybits}
\end{center}\end{table}
\end{center}\end{table}
always return the time value contained in the register.  Writes greater than
always return the time value contained in the register.  Writes greater than
the current Jiffy value, that is where the new value minus the old value is
the current Jiffy value, that is where the new value minus the old value is
greater than zero while ignoring truncation, will set a new Jiffy interrupt
greater than zero while ignoring truncation, will set a new Jiffy interrupt
time.  At that time, the Jiffy vector will clear, and another interrupt time
time.  At that time, the Jiffy vector will clear, and another interrupt time
may either be written to it, or it will just continue counting without
may either be written to it, or it will just continue counting without
activating any more interrupts.
activating any more interrupts.
 
 
The Zip CPU also supports several counter peripherals, mostly in the way of
The Zip CPU also supports several counter peripherals, mostly in the way of
process accounting.  This peripherals have a single register associated with
process accounting.  This peripherals have a single register associated with
them, shown in Tbl.~\ref{tbl:ctrbits}.
them, shown in Tbl.~\ref{tbl:ctrbits}.
\begin{table}\begin{center}
\begin{table}\begin{center}
\begin{bitlist}
\begin{bitlist}
31\ldots 0 & R/W & Current counter value\\\hline
31\ldots 0 & R/W & Current counter value\\\hline
\end{bitlist}
\end{bitlist}
\caption{Counter Register Bits}\label{tbl:ctrbits}
\caption{Counter Register Bits}\label{tbl:ctrbits}
\end{center}\end{table}
\end{center}\end{table}
Writes to this register set the new counter value.  Reads read the current
Writes to this register set the new counter value.  Reads read the current
counter value.
counter value.
 
 
The current design operation of these counters is that of performance counting.
The current design operation of these counters is that of performance counting.
Two sets of four registers are available for keeping track of performance.
Two sets of four registers are available for keeping track of performance.
The first is a task counter.  This just counts clock ticks.  The second
The first is a task counter.  This just counts clock ticks.  The second
counter is a prefetch stall counter, then an master stall counter.  These
counter is a prefetch stall counter, then an master stall counter.  These
allow the CPU to be evaluated as to how efficient it is.  The fourth and
allow the CPU to be evaluated as to how efficient it is.  The fourth and
final counter is an instruction counter, which counts how many instructions the
final counter is an instruction counter, which counts how many instructions the
CPU has issued.
CPU has issued.
 
 
It is envisioned that these counters will be used as follows: First, every time
It is envisioned that these counters will be used as follows: First, every time
a master counter rolls over, the supervisor (Operating System) will record
a master counter rolls over, the supervisor (Operating System) will record
the fact.  Second, whenever activating a user task, the Operating System will
the fact.  Second, whenever activating a user task, the Operating System will
set the four user counters to zero.  When the user task has completed, the
set the four user counters to zero.  When the user task has completed, the
Operating System will read the timers back off, to determine how much of the
Operating System will read the timers back off, to determine how much of the
CPU the process had consumed.
CPU the process had consumed.
 
 
 
The final peripheral to discuss is the DMA controller.  This controller
 
has four registers.  Of these four, the length, source and destination address
 
registers should need no further explanation.  They are full 32--bit registers
 
specifying the entire transfer length, the starting address to read from, and
 
the starting address to write to.  The registers can be written to when the
 
DMA is idle, and read at any time.  The control register, however, will need
 
some more explanation.
 
 
 
The bit allocation of the control register is shown in Tbl.~\ref{tbl:dmacbits}.
 
\begin{table}\begin{center}
 
\begin{bitlist}
 
31 & R & DMA Active\\\hline
 
30 & R & Wishbone error, transaction aborted (cleared on any write)\\\hline
 
29 & R/W & Set to '1' to prevent the controller from incrementing the source address, '0' for normal memory copy. \\\hline
 
28 & R/W & Set to '0' to prevent the controller from incrementing the
 
        destination address, '0' for normal memory copy. \\\hline
 
27 \ldots 16 & W & The DMA Key.  Write a 12'hfed to these bits to start the
 
        activate any DMA transfer.  \\\hline
 
27 & R & Always reads '0', to force the deliberate writing of the key. \\\hline
 
26 \ldots 16 & R & Indicates the number of items in the transfer buffer that
 
        have yet to be written. \\\hline
 
15 & R/W & Set to '1' to trigger on an interrupt, or '0' to start immediately
 
        upon receiving a valid key.\\\hline
 
14\ldots 10 & R/W & Select among one of 32~possible interrupt lines.\\\hline
 
9\ldots 0 & R/W & Intermediate transfer length minus one.  Thus, to transfer
 
        one item at a time set this value to 0. To transfer 1024 at a time,
 
        set it to 1024.\\\hline
 
\end{bitlist}
 
\caption{DMA Control Register Bits}\label{tbl:dmacbits}
 
\end{center}\end{table}
 
This control register has been designed so that the common case of memory
 
access need only set the key and the transfer length.  Hence, writing a
 
\hbox{32'h0fed03ff} to the control register will start any memory transfer.
 
On the other hand, if you wished to read from a serial port (constant address)
 
and put the result into a buffer every time a word was available, you
 
might wish to write \hbox{32'h2fed8000}--this assumes, of course, that you
 
have a serial port wired to the zero bit of this interrupt control.  (The
 
DMA controller does not use the interrupt controller, and cannot clear
 
interrupts.)  As a third example, if you wished to write to an external
 
FIFO anytime it was less than half full (had fewer than 512 items), and
 
interrupt line 2 indicated this condition, you might wish to issue a
 
\hbox{32'h1fed8dff} to this port.
 
 
\section{Debug Port Registers}
\section{Debug Port Registers}
Accessing the Zip System via the debug port isn't as straight forward as
Accessing the Zip System via the debug port isn't as straight forward as
accessing the system via the wishbone bus.  The debug port itself has been
accessing the system via the wishbone bus.  The debug port itself has been
reduced to two addresses, as outlined earlier in Tbl.~\ref{tbl:dbgregs}.
reduced to two addresses, as outlined earlier in Tbl.~\ref{tbl:dbgregs}.
Access to the Zip System begins with the Debug Control register, shown in
Access to the Zip System begins with the Debug Control register, shown in
Tbl.~\ref{tbl:dbgctrl}.
Tbl.~\ref{tbl:dbgctrl}.
\begin{table}\begin{center}
\begin{table}\begin{center}
\begin{bitlist}
\begin{bitlist}
31\ldots 14 & R & Reserved\\\hline
31\ldots 14 & R & Reserved\\\hline
13 & R & CPU GIE setting\\\hline
13 & R & CPU GIE setting\\\hline
12 & R & CPU is sleeping\\\hline
12 & R & CPU is sleeping\\\hline
11 & W & Command clear PF cache\\\hline
11 & W & Command clear PF cache\\\hline
10 & R/W & Command HALT, Set to '1' to halt the CPU\\\hline
10 & R/W & Command HALT, Set to '1' to halt the CPU\\\hline
9 & R & Stall Status, '1' if CPU is busy\\\hline
9 & R & Stall Status, '1' if CPU is busy\\\hline
8 & R/W & Step Command, set to '1' to step the CPU\\\hline
8 & R/W & Step Command, set to '1' to step the CPU, also sets the halt bit\\\hline
7 & R & Interrupt Request \\\hline
7 & R & Interrupt Request \\\hline
6 & R/W & Command RESET \\\hline
6 & R/W & Command RESET \\\hline
5\ldots 0 & R/W & Debug Register Address \\\hline
5\ldots 0 & R/W & Debug Register Address \\\hline
\end{bitlist}
\end{bitlist}
\caption{Debug Control Register Bits}\label{tbl:dbgctrl}
\caption{Debug Control Register Bits}\label{tbl:dbgctrl}
\end{center}\end{table}
\end{center}\end{table}
 
 
The first step in debugging access is to determine whether or not the CPU
The first step in debugging access is to determine whether or not the CPU
is halted, and to halt it if not.  To do this, first write a '1' to the
is halted, and to halt it if not.  To do this, first write a '1' to the
Command HALT bit.  This will halt the CPU and place it into debug mode.
Command HALT bit.  This will halt the CPU and place it into debug mode.
Once the CPU is halted, the stall status bit will drop to zero.  Thus,
Once the CPU is halted, the stall status bit will drop to zero.  Thus,
if bit 10 is high and bit 9 low, the debug port is open to examine the
if bit 10 is high and bit 9 low, the debug port is open to examine the
internal state of the CPU.
internal state of the CPU.
 
 
At this point, the external debugger may examine internal state information
At this point, the external debugger may examine internal state information
from within the CPU.  To do this, first write again to the command register
from within the CPU.  To do this, first write again to the command register
a value (with command halt still high) containing the address of an internal
a value (with command halt still high) containing the address of an internal
register of interest in the bottom 6~bits.  Internal registers that may be
register of interest in the bottom 6~bits.  Internal registers that may be
accessed this way are listed in Tbl.~\ref{tbl:dbgaddrs}.
accessed this way are listed in Tbl.~\ref{tbl:dbgaddrs}.
\begin{table}\begin{center}
\begin{table}\begin{center}
\begin{reglist}
\begin{reglist}
sR0 & 0 & 32 & R/W & Supervisor Register R0 \\\hline
sR0 & 0 & 32 & R/W & Supervisor Register R0 \\\hline
sR1 & 0 & 32 & R/W & Supervisor Register R1 \\\hline
sR1 & 0 & 32 & R/W & Supervisor Register R1 \\\hline
sSP & 13 & 32 & R/W & Supervisor Stack Pointer\\\hline
sSP & 13 & 32 & R/W & Supervisor Stack Pointer\\\hline
sCC & 14 & 32 & R/W & Supervisor Condition Code Register \\\hline
sCC & 14 & 32 & R/W & Supervisor Condition Code Register \\\hline
sPC & 15 & 32 & R/W & Supervisor Program Counter\\\hline
sPC & 15 & 32 & R/W & Supervisor Program Counter\\\hline
uR0 & 16 & 32 & R/W & User Register R0 \\\hline
uR0 & 16 & 32 & R/W & User Register R0 \\\hline
uR1 & 17 & 32 & R/W & User Register R1 \\\hline
uR1 & 17 & 32 & R/W & User Register R1 \\\hline
uSP & 29 & 32 & R/W & User Stack Pointer\\\hline
uSP & 29 & 32 & R/W & User Stack Pointer\\\hline
uCC & 30 & 32 & R/W & User Condition Code Register \\\hline
uCC & 30 & 32 & R/W & User Condition Code Register \\\hline
uPC & 31 & 32 & R/W & User Program Counter\\\hline
uPC & 31 & 32 & R/W & User Program Counter\\\hline
PIC & 32 & 32 & R/W & Primary Interrupt Controller \\\hline
PIC & 32 & 32 & R/W & Primary Interrupt Controller \\\hline
WDT & 33 & 32 & R/W & Watchdog Timer\\\hline
WDT & 33 & 32 & R/W & Watchdog Timer\\\hline
CCHE & 34 & 32 & R/W & Manual Cache Controller\\\hline
CCHE & 34 & 32 & R/W & Manual Cache Controller\\\hline
CTRIC & 35 & 32 & R/W & Secondary Interrupt Controller\\\hline
CTRIC & 35 & 32 & R/W & Secondary Interrupt Controller\\\hline
TMRA & 36 & 32 & R/W & Timer A\\\hline
TMRA & 36 & 32 & R/W & Timer A\\\hline
TMRB & 37 & 32 & R/W & Timer B\\\hline
TMRB & 37 & 32 & R/W & Timer B\\\hline
TMRC & 38 & 32 & R/W & Timer C\\\hline
TMRC & 38 & 32 & R/W & Timer C\\\hline
JIFF & 39 & 32 & R/W & Jiffies peripheral\\\hline
JIFF & 39 & 32 & R/W & Jiffies peripheral\\\hline
MTASK & 40 & 32 & R/W & Master task clock counter\\\hline
MTASK & 40 & 32 & R/W & Master task clock counter\\\hline
MMSTL & 41 & 32 & R/W & Master memory stall counter\\\hline
MMSTL & 41 & 32 & R/W & Master memory stall counter\\\hline
MPSTL & 42 & 32 & R/W & Master Pre-Fetch Stall counter\\\hline
MPSTL & 42 & 32 & R/W & Master Pre-Fetch Stall counter\\\hline
MICNT & 43 & 32 & R/W & Master instruction counter\\\hline
MICNT & 43 & 32 & R/W & Master instruction counter\\\hline
UTASK & 44 & 32 & R/W & User task clock counter\\\hline
UTASK & 44 & 32 & R/W & User task clock counter\\\hline
UMSTL & 45 & 32 & R/W & User memory stall counter\\\hline
UMSTL & 45 & 32 & R/W & User memory stall counter\\\hline
UPSTL & 46 & 32 & R/W & User Pre-Fetch Stall counter\\\hline
UPSTL & 46 & 32 & R/W & User Pre-Fetch Stall counter\\\hline
UICNT & 47 & 32 & R/W & User instruction counter\\\hline
UICNT & 47 & 32 & R/W & User instruction counter\\\hline
\end{reglist}
\end{reglist}
\caption{Debug Register Addresses}\label{tbl:dbgaddrs}
\caption{Debug Register Addresses}\label{tbl:dbgaddrs}
\end{center}\end{table}
\end{center}\end{table}
Primarily, these ``registers'' include access to the entire CPU register
Primarily, these ``registers'' include access to the entire CPU register
set, as well as the 16~internal peripherals.  To read one of these registers
set, as well as the internal peripherals.  To read one of these registers
once the address is set, simply issue a read from the data port.  To write
once the address is set, simply issue a read from the data port.  To write
one of these registers or peripheral ports, simply write to the data port
one of these registers or peripheral ports, simply write to the data port
after setting the proper address.
after setting the proper address.
 
 
In this manner, all of the CPU's internal state may be read and adjusted.
In this manner, all of the CPU's internal state may be read and adjusted.
 
 
As an example of how to use this, consider what would happen in the case
As an example of how to use this, consider what would happen in the case
of an external break point.  If and when the CPU hits a break point that
of an external break point.  If and when the CPU hits a break point that
causes it to halt, the Command HALT bit will activate on its own, the CPU
causes it to halt, the Command HALT bit will activate on its own, the CPU
will then raise an external interrupt line and wait for a debugger to examine
will then raise an external interrupt line and wait for a debugger to examine
its state.  After examining the state, the debugger will need to remove
its state.  After examining the state, the debugger will need to remove
the breakpoint by writing a different instruction into memory and by writing
the breakpoint by writing a different instruction into memory and by writing
to the command register while holding the clear cache, command halt, and
to the command register while holding the clear cache, command halt, and
step CPU bits high, (32'hd00).  The debugger may then replace the breakpoint
step CPU bits high, (32'hd00).  The debugger may then replace the breakpoint
now that the CPU has gone beyond it, and clear the cache again (32'h500).
now that the CPU has gone beyond it, and clear the cache again (32'h500).
 
 
To leave this debug mode, simply write a `32'h0' value to the command register.
To leave this debug mode, simply write a `32'h0' value to the command register.
 
 
\chapter{Wishbone Datasheets}\label{chap:wishbone}
\chapter{Wishbone Datasheets}\label{chap:wishbone}
The Zip System supports two wishbone ports, a slave debug port and a master
The Zip System supports two wishbone ports, a slave debug port and a master
port for the system itself.  These are shown in Tbl.~\ref{tbl:wishbone-slave}
port for the system itself.  These are shown in Tbl.~\ref{tbl:wishbone-slave}
\begin{table}[htbp]
\begin{table}[htbp]
\begin{center}
\begin{center}
\begin{wishboneds}
\begin{wishboneds}
Revision level of wishbone & WB B4 spec \\\hline
Revision level of wishbone & WB B4 spec \\\hline
Type of interface & Slave, Read/Write, single words only \\\hline
Type of interface & Slave, Read/Write, single words only \\\hline
Address Width & 1--bit \\\hline
Address Width & 1--bit \\\hline
Port size & 32--bit \\\hline
Port size & 32--bit \\\hline
Port granularity & 32--bit \\\hline
Port granularity & 32--bit \\\hline
Maximum Operand Size & 32--bit \\\hline
Maximum Operand Size & 32--bit \\\hline
Data transfer ordering & (Irrelevant) \\\hline
Data transfer ordering & (Irrelevant) \\\hline
Clock constraints & Works at 100~MHz on a Basys--3 board\\\hline
Clock constraints & Works at 100~MHz on a Basys--3 board\\\hline
Signal Names & \begin{tabular}{ll}
Signal Names & \begin{tabular}{ll}
                Signal Name & Wishbone Equivalent \\\hline
                Signal Name & Wishbone Equivalent \\\hline
                {\tt i\_clk} & {\tt CLK\_I} \\
                {\tt i\_clk} & {\tt CLK\_I} \\
                {\tt i\_dbg\_cyc} & {\tt CYC\_I} \\
                {\tt i\_dbg\_cyc} & {\tt CYC\_I} \\
                {\tt i\_dbg\_stb} & {\tt STB\_I} \\
                {\tt i\_dbg\_stb} & {\tt STB\_I} \\
                {\tt i\_dbg\_we} & {\tt WE\_I} \\
                {\tt i\_dbg\_we} & {\tt WE\_I} \\
                {\tt i\_dbg\_addr} & {\tt ADR\_I} \\
                {\tt i\_dbg\_addr} & {\tt ADR\_I} \\
                {\tt i\_dbg\_data} & {\tt DAT\_I} \\
                {\tt i\_dbg\_data} & {\tt DAT\_I} \\
                {\tt o\_dbg\_ack} & {\tt ACK\_O} \\
                {\tt o\_dbg\_ack} & {\tt ACK\_O} \\
                {\tt o\_dbg\_stall} & {\tt STALL\_O} \\
                {\tt o\_dbg\_stall} & {\tt STALL\_O} \\
                {\tt o\_dbg\_data} & {\tt DAT\_O}
                {\tt o\_dbg\_data} & {\tt DAT\_O}
                \end{tabular}\\\hline
                \end{tabular}\\\hline
\end{wishboneds}
\end{wishboneds}
\caption{Wishbone Datasheet for the Debug Interface}\label{tbl:wishbone-slave}
\caption{Wishbone Datasheet for the Debug Interface}\label{tbl:wishbone-slave}
\end{center}\end{table}
\end{center}\end{table}
and Tbl.~\ref{tbl:wishbone-master} respectively.
and Tbl.~\ref{tbl:wishbone-master} respectively.
\begin{table}[htbp]
\begin{table}[htbp]
\begin{center}
\begin{center}
\begin{wishboneds}
\begin{wishboneds}
Revision level of wishbone & WB B4 spec \\\hline
Revision level of wishbone & WB B4 spec \\\hline
Type of interface & Master, Read/Write, single cycle or pipelined\\\hline
Type of interface & Master, Read/Write, single cycle or pipelined\\\hline
Address Width & 32--bit bits \\\hline
Address Width & 32--bit bits \\\hline
Port size & 32--bit \\\hline
Port size & 32--bit \\\hline
Port granularity & 32--bit \\\hline
Port granularity & 32--bit \\\hline
Maximum Operand Size & 32--bit \\\hline
Maximum Operand Size & 32--bit \\\hline
Data transfer ordering & (Irrelevant) \\\hline
Data transfer ordering & (Irrelevant) \\\hline
Clock constraints & Works at 100~MHz on a Basys--3 board\\\hline
Clock constraints & Works at 100~MHz on a Basys--3 board\\\hline
Signal Names & \begin{tabular}{ll}
Signal Names & \begin{tabular}{ll}
                Signal Name & Wishbone Equivalent \\\hline
                Signal Name & Wishbone Equivalent \\\hline
                {\tt i\_clk} & {\tt CLK\_O} \\
                {\tt i\_clk} & {\tt CLK\_O} \\
                {\tt o\_wb\_cyc} & {\tt CYC\_O} \\
                {\tt o\_wb\_cyc} & {\tt CYC\_O} \\
                {\tt o\_wb\_stb} & {\tt STB\_O} \\
                {\tt o\_wb\_stb} & {\tt STB\_O} \\
                {\tt o\_wb\_we} & {\tt WE\_O} \\
                {\tt o\_wb\_we} & {\tt WE\_O} \\
                {\tt o\_wb\_addr} & {\tt ADR\_O} \\
                {\tt o\_wb\_addr} & {\tt ADR\_O} \\
                {\tt o\_wb\_data} & {\tt DAT\_O} \\
                {\tt o\_wb\_data} & {\tt DAT\_O} \\
                {\tt i\_wb\_ack} & {\tt ACK\_I} \\
                {\tt i\_wb\_ack} & {\tt ACK\_I} \\
                {\tt i\_wb\_stall} & {\tt STALL\_I} \\
                {\tt i\_wb\_stall} & {\tt STALL\_I} \\
                {\tt i\_wb\_data} & {\tt DAT\_I}
                {\tt i\_wb\_data} & {\tt DAT\_I}
                \end{tabular}\\\hline
                \end{tabular}\\\hline
\end{wishboneds}
\end{wishboneds}
\caption{Wishbone Datasheet for the CPU as Master}\label{tbl:wishbone-master}
\caption{Wishbone Datasheet for the CPU as Master}\label{tbl:wishbone-master}
\end{center}\end{table}
\end{center}\end{table}
I do not recommend that you connect these together through the interconnect.
I do not recommend that you connect these together through the interconnect.
Rather, the debug port of the CPU should be accessible regardless of the state
Rather, the debug port of the CPU should be accessible regardless of the state
of the master bus.
of the master bus.
 
 
You may wish to notice that neither the {\tt ERR} nor the {\tt RETRY} wires
You may wish to notice that neither the {\tt ERR} nor the {\tt RETRY} wires
have been implemented.  What this means is that the CPU is currently unable
have been implemented.  What this means is that the CPU is currently unable
to detect a bus error condition, and so may stall indefinitely (hang) should
to detect a bus error condition, and so may stall indefinitely (hang) should
it choose to access a value not on the bus, or a peripheral that is not
it choose to access a value not on the bus, or a peripheral that is not
yet properly configured.
yet properly configured.
 
 
\chapter{Clocks}\label{chap:clocks}
\chapter{Clocks}\label{chap:clocks}
 
 
This core is based upon the Basys--3 development board sold by Digilent.
This core is based upon the Basys--3 development board sold by Digilent.
The Basys--3 development board contains one external 100~MHz clock, which is
The Basys--3 development board contains one external 100~MHz clock, which is
sufficient to run the ZIP CPU core.
sufficient to run the Zip CPU core.
\begin{table}[htbp]
\begin{table}[htbp]
\begin{center}
\begin{center}
\begin{clocklist}
\begin{clocklist}
i\_clk & External & 100~MHz & 100~MHz & System clock.\\\hline
i\_clk & External & 100~MHz & 100~MHz & System clock.\\\hline
\end{clocklist}
\end{clocklist}
\caption{List of Clocks}\label{tbl:clocks}
\caption{List of Clocks}\label{tbl:clocks}
\end{center}\end{table}
\end{center}\end{table}
I hesitate to suggest that the core can run faster than 100~MHz, since I have
I hesitate to suggest that the core can run faster than 100~MHz, since I have
had struggled with various timing violations to keep it at 100~MHz.  So, for
had struggled with various timing violations to keep it at 100~MHz.  So, for
now, I will only state that it can run at 100~MHz.
now, I will only state that it can run at 100~MHz.
 
 
 
 
\chapter{I/O Ports}\label{chap:ioports}
\chapter{I/O Ports}\label{chap:ioports}
The I/O ports to the Zip CPU may be grouped into three categories.  The first
The I/O ports to the Zip CPU may be grouped into three categories.  The first
is that of the master wishbone used by the CPU, then the slave wishbone used
is that of the master wishbone used by the CPU, then the slave wishbone used
to command the CPU via a debugger, and then the rest.  The first two of these
to command the CPU via a debugger, and then the rest.  The first two of these
were already discussed in the wishbone chapter.  They are listed here
were already discussed in the wishbone chapter.  They are listed here
for completeness in Tbl.~\ref{tbl:iowb-master}
for completeness in Tbl.~\ref{tbl:iowb-master}
\begin{table}
\begin{table}
\begin{center}\begin{portlist}
\begin{center}\begin{portlist}
{\tt o\_wb\_cyc}   &  1 & Output & Indicates an active Wishbone cycle\\\hline
{\tt o\_wb\_cyc}   &  1 & Output & Indicates an active Wishbone cycle\\\hline
{\tt o\_wb\_stb}   &  1 & Output & WB Strobe signal\\\hline
{\tt o\_wb\_stb}   &  1 & Output & WB Strobe signal\\\hline
{\tt o\_wb\_we}    &  1 & Output & Write enable\\\hline
{\tt o\_wb\_we}    &  1 & Output & Write enable\\\hline
{\tt o\_wb\_addr}  & 32 & Output & Bus address \\\hline
{\tt o\_wb\_addr}  & 32 & Output & Bus address \\\hline
{\tt o\_wb\_data}  & 32 & Output & Data on WB write\\\hline
{\tt o\_wb\_data}  & 32 & Output & Data on WB write\\\hline
{\tt i\_wb\_ack}   &  1 & Input  & Slave has completed a R/W cycle\\\hline
{\tt i\_wb\_ack}   &  1 & Input  & Slave has completed a R/W cycle\\\hline
{\tt i\_wb\_stall} &  1 & Input  & WB bus slave not ready\\\hline
{\tt i\_wb\_stall} &  1 & Input  & WB bus slave not ready\\\hline
{\tt i\_wb\_data}  & 32 & Input  & Incoming bus data\\\hline
{\tt i\_wb\_data}  & 32 & Input  & Incoming bus data\\\hline
\end{portlist}\caption{CPU Master Wishbone I/O Ports}\label{tbl:iowb-master}\end{center}\end{table}
\end{portlist}\caption{CPU Master Wishbone I/O Ports}\label{tbl:iowb-master}\end{center}\end{table}
and~\ref{tbl:iowb-slave} respectively.
and~\ref{tbl:iowb-slave} respectively.
\begin{table}
\begin{table}
\begin{center}\begin{portlist}
\begin{center}\begin{portlist}
{\tt i\_wb\_cyc}   &  1 & Input & Indicates an active Wishbone cycle\\\hline
{\tt i\_wb\_cyc}   &  1 & Input & Indicates an active Wishbone cycle\\\hline
{\tt i\_wb\_stb}   &  1 & Input & WB Strobe signal\\\hline
{\tt i\_wb\_stb}   &  1 & Input & WB Strobe signal\\\hline
{\tt i\_wb\_we}    &  1 & Input & Write enable\\\hline
{\tt i\_wb\_we}    &  1 & Input & Write enable\\\hline
{\tt i\_wb\_addr}  &  1 & Input & Bus address, command or data port \\\hline
{\tt i\_wb\_addr}  &  1 & Input & Bus address, command or data port \\\hline
{\tt i\_wb\_data}  & 32 & Input & Data on WB write\\\hline
{\tt i\_wb\_data}  & 32 & Input & Data on WB write\\\hline
{\tt o\_wb\_ack}   &  1 & Output  & Slave has completed a R/W cycle\\\hline
{\tt o\_wb\_ack}   &  1 & Output  & Slave has completed a R/W cycle\\\hline
{\tt o\_wb\_stall} &  1 & Output  & WB bus slave not ready\\\hline
{\tt o\_wb\_stall} &  1 & Output  & WB bus slave not ready\\\hline
{\tt o\_wb\_data}  & 32 & Output  & Incoming bus data\\\hline
{\tt o\_wb\_data}  & 32 & Output  & Incoming bus data\\\hline
\end{portlist}\caption{CPU Debug Wishbone I/O Ports}\label{tbl:iowb-slave}\end{center}\end{table}
\end{portlist}\caption{CPU Debug Wishbone I/O Ports}\label{tbl:iowb-slave}\end{center}\end{table}
 
 
There are only four other lines to the CPU: the external clock, external
There are only four other lines to the CPU: the external clock, external
reset, incoming external interrupt line(s), and the outgoing debug interrupt
reset, incoming external interrupt line(s), and the outgoing debug interrupt
line.  These are shown in Tbl.~\ref{tbl:ioports}.
line.  These are shown in Tbl.~\ref{tbl:ioports}.
\begin{table}
\begin{table}
\begin{center}\begin{portlist}
\begin{center}\begin{portlist}
{\tt i\_clk} & 1 & Input & The master CPU clock \\\hline
{\tt i\_clk} & 1 & Input & The master CPU clock \\\hline
{\tt i\_rst} & 1 & Input &  Active high reset line \\\hline
{\tt i\_rst} & 1 & Input &  Active high reset line \\\hline
{\tt i\_ext\_int} & 1\ldots 6 & Input &  Incoming external interrupts \\\hline
{\tt i\_ext\_int} & 1\ldots 6 & Input &  Incoming external interrupts \\\hline
{\tt o\_ext\_int} & 1 & Output & CPU Halted interrupt \\\hline
{\tt o\_ext\_int} & 1 & Output & CPU Halted interrupt \\\hline
\end{portlist}\caption{I/O Ports}\label{tbl:ioports}\end{center}\end{table}
\end{portlist}\caption{I/O Ports}\label{tbl:ioports}\end{center}\end{table}
The clock line was discussed briefly in Chapt.~\ref{chap:clocks}.  We
The clock line was discussed briefly in Chapt.~\ref{chap:clocks}.  We
typically run it at 100~MHz.  The reset line is an active high reset.  When
typically run it at 100~MHz.  The reset line is an active high reset.  When
asserted, the CPU will start running again from its reset address in
asserted, the CPU will start running again from its reset address in
memory.  Further, depending upon how the CPU is configured and specifically on
memory.  Further, depending upon how the CPU is configured and specifically on
the {\tt START\_HALTED} parameter, it may or may not start running
the {\tt START\_HALTED} parameter, it may or may not start running
automatically.  The {\tt i\_ext\_int} line is for an external interrupt.  This
automatically.  The {\tt i\_ext\_int} line is for an external interrupt.  This
line may be as wide as 6~external interrupts, depending upon the setting of
line may be as wide as 6~external interrupts, depending upon the setting of
the {\tt EXTERNAL\_INTERRUPTS} line.  As currently configured, the ZipSystem
the {\tt EXTERNAL\_INTERRUPTS} line.  As currently configured, the ZipSystem
only supports one such interrupt line by default.  For us, this line is the
only supports one such interrupt line by default.  For us, this line is the
output of another interrupt controller, but that's a board specific setup
output of another interrupt controller, but that's a board specific setup
detail.  Finally, the Zip System produces one external interrupt whenever
detail.  Finally, the Zip System produces one external interrupt whenever
the CPU halts to wait for the debugger.
the CPU halts to wait for the debugger.
 
 
 
\chapter{Initial Assessment}\label{chap:assessment}
 
 
 
Having now worked with the Zip CPU for a while, it is worth offering an
 
honest assessment of how well it works and how well it was designed. At the
 
end of this assessment, I will propose some changes that may take place in a
 
later version of this Zip CPU to make it better.
 
 
 
\section{The Good}
 
\begin{itemize}
 
\item The Zip CPU is light weight and fully featured as it exists today. For
 
        anyone who wishes to build a general purpose CPU and then to
 
        experiment with building and adding particular features, the Zip CPU
 
        makes a good starting point--it is fairly simple. Modifications should
 
        be simple enough.
 
\item As an estimate of the ``weight'' of this implementation, the CPU has
 
        cost me less than 150 hours to implement from its inception.
 
\item The Zip CPU was designed to be an implementable soft core that could be
 
        placed within an FPGA, controlling actions internal to the FPGA. It
 
        fits this role rather nicely. It does not fit the role of a system on
 
        a chip very well, but then it was never intended to be a system on a
 
        chip but rather a system within a chip.
 
\item The extremely simplified instruction set of the Zip CPU was a good
 
        choice. Although it does not have many of the commonly used
 
        instructions, PUSH, POP, JSR, and RET among them, the simplified
 
        instruction set has demonstrated an amazing versatility. I will contend
 
        therefore and for anyone who will listen, that this instruction set
 
        offers a full and complete capability for whatever a user might wish
 
        to do with two exceptions: bytewise character access and accelerated
 
        floating-point support.
 
\item This simplified instruction set is easy to decode.
 
\item The simplified bus transactions (32-bit words only) were also very easy
 
        to implement.
 
\item The novel approach of having a single interrupt vector, which just
 
        brings the CPU back to the instruction it left off at within the last
 
        interrupt context doesn't appear to have been that much of a problem.
 
        If most modern systems handle interrupt vectoring in software anyway,
 
        why maintain hardware support for it?
 
\item My goal of a high rate of instructions per clock may not be the proper
 
        measure. For example, if instructions are being read from a SPI flash
 
        device, such as is common among FPGA implementations, these same
 
        instructions may suffer stalls of between 64 and 128 cycles per
 
        instruction just to read the instruction from the flash. Executing the
 
        instruction in a single clock cycle is no longer the appropriate
 
        measure. At the same time, it should be possible to use the DMA
 
        peripheral to copy instructions from the FLASH to a temporary memory
 
        location, after which they may be executed at a single instruction
 
        cycle per access again.
 
\end{itemize}
 
 
 
\section{The Not so Good}
 
\begin{itemize}
 
\item While one of the stated goals was to use a small amount of logic,
 
        3k~LUTs isn't that impressively small. Indeed, it's really much
 
        too expensive when compared against other 8 and 16-bit CPUs that have
 
        less than 1k LUTs.
 
 
 
        Still, \ldots it's not bad, it's just not astonishingly good.
 
 
 
\item The fact that the instruction width equals the bus width means that the
 
        instruction fetch cycle will always be interfering with any load or
 
        store memory operation, with the only exception being if the
 
        instruction is already in the cache.  {\em This has become the
 
        fundamental limit on the speed and performance of the CPU!}
 
        Those familiar with the Von--Neumann approach of sharing a bus
 
        between data and instructions will not be surprised by this assessment.
 
 
 
        This could be fixed in one of three ways: the instruction set
 
        architecture could be modified to handle Very Long Instruction Words
 
        (VLIW) so that each 32--bit word would encode two or more instructions,
 
        the instruction fetch bus width could be increased from 32--bits to
 
        64--bits or more, or the instruction bus could be separated from the
 
        data bus.  Any and all of these approaches would increase the overall
 
        LUT count.
 
 
 
\item The (non-existant) floating point unit was an after-thought, isn't even
 
        built as a potential option, and most likely won't support the full
 
        IEEE standard set of FPU instructions--even for single point precision.
 
        This (non-existant) capability would benefit the most from an
 
        out-of-order execution capability, which the Zip CPU does not have.
 
 
 
        Still, sharing FPU registers with the main register set was a good
 
        idea and worth preserving, as it simplifies context swapping.
 
 
 
        Perhaps this really isn't a problem, but rather a feature.  By not
 
        implementing FPU instructions, the Zip CPU maintains a lower LUT count
 
        than it would have if it did implement these instructions.
 
 
 
\item The CPU has no character support. This is both good and bad.
 
        Realistically, the CPU works just fine without it. Characters can be
 
        supported as subsets of 32-bit words without any problem. Practically,
 
        though, it will make compiling non-Zip CPU code difficult--especially
 
        anything that assumes sizeof(int)=4*sizeof(char), or that tries to
 
        create unions with characters and integers and then attempts to
 
        reference the address of the characters within that union.
 
 
 
\item The Zip CPU does not support a data cache. One can still be built
 
        externally, but this is a limitation of the CPU proper as built.
 
        Further, under the theory of the Zip CPU design (that of an embedded
 
        soft-core processor within an FPGA, where any ``address'' may reference
 
        either memory or a peripheral that may have side-effects), any data
 
        cache would need to be based upon an initial knowledge of whether or
 
        not it is supporting memory (cachable) or peripherals. This knowledge
 
        must exist somewhere, and that somewhere is currently (and by design)
 
        external to the CPU.
 
 
 
        This may also be written off as a ``feature'' of the Zip CPU, since
 
        the addition of a data cache can greatly increase the LUT count of
 
        a soft core.
 
 
 
\item Many other instruction sets offer three operand instructions, whereas
 
        the Zip CPU only offers two operand instructions. This means that it
 
        takes the Zip CPU more instructions to do many of the same operations.
 
        The good part of this is that it gives the Zip CPU a greater amount of
 
        flexibility in its immediate operand mode, although that increased
 
        flexibility isn't necessarily as valuable as one might like.
 
 
 
\item The Zip CPU does not currently detect and trap on either illegal
 
        instructions or bus errors.  Attempts to access non--existent
 
        memory quietly return erroneous results, rather than halting the
 
        process (user mode) or halting or resetting the CPU (supervisor mode).
 
 
 
\item The Zip CPU doesn't support out of order execution. I suppose it could
 
        be modified to do so, but then it would no longer be the ``simple''
 
        and low LUT count CPU it was designed to be. The two primary results
 
        are that 1) loads may unnecessarily stall the CPU, even if other
 
        things could be done while waiting for the load to complete, 2)
 
        bus errors on stores will never be caught at the point of the error,
 
        and 3) branch prediction becomes more difficult.
 
 
 
\item Although switching to an interrupt context in the Zip CPU design doesn't
 
        require a tremendous swapping of registers, in reality it still
 
        does--since any task swap still requires saving and restoring all
 
        16~user registers. That's a lot of memory movement just to service
 
        an interrupt.
 
 
 
\item The Zip CPU is by no means generic: it will never handle addresses
 
        larger than 32-bits (16GB) without a complete and total redesign.
 
        This may limit its utility as a generic CPU in the future, although
 
        as an embedded CPU within an FPGA this isn't really much of a limit
 
        or restriction.
 
 
 
\item While the Zip CPU has its own assembler, it has no linker and does not
 
        (yet) support a compiler. The standard C library is an even longer
 
        shot. My dream of having binutils and gcc support has not been
 
        realized and at this rate may not be realized. (I've been intimidated
 
        by the challenge everytime I've looked through those codes.)
 
 
 
\item While the Wishbone Bus (B4) supports a pipelined mode with single cycle
 
        execution, the Zip CPU is unable to exploit this parallelism. Instead,
 
        apart from the DMA and the pipelined prefetch, all loads and stores
 
        are single wishbone bus operations requiring a minimum of 3 clocks.
 
        (In practice, this has turned into 7-clocks.)
 
 
 
\iffalse
 
\item There is no control over whether or not an instruction sets the
 
        condition codes--certain instructions always set the condition codes,
 
        other instructions never set them. This effectively limits conditional
 
        instructions to a single instruction only (with two or more
 
        instructions as an exception), as the first instruction that sets
 
        condition codes will break the condition code chain.
 
 
 
        {\em (A proposed change below address this.)}
 
 
 
\item Using the CC register as a trap address was a bad idea--it limits the CC
 
        registers ability to be used in future expansion, such as by adding
 
        exception indication flags: bus error, floating point exception, etc.
 
 
 
        {\em (This can be changed by a different O/S implementation of the trap
 
        instruction.)}
 
\item The current implementation suffers from too many stalls on any
 
        branch--even if the branch is known early on.
 
 
 
        {\em (This is addressed in proposals below.)}
 
        % Addressed, 20150918
 
 
 
\item In a similar fashion, a switch to interrupt context forces the pipeline
 
        to be cleared, whereas it might make more sense to just continue
 
        executing the instructions already in the pipeline while the prefetch
 
        stage is working on switching to the interrupt context.
 
 
 
        {\em (Also addressed in proposals below.)}
 
        % This should happen so rarely that it is not really a problem
 
\fi
 
 
 
\end{itemize}
 
 
 
\section{The Next Generation}
 
This section could also be labeled as my ``To do'' list.
 
 
 
Given the feedback listed above, perhaps its time to consider what changes could be made to improve the Zip CPU in the future. I offer the following as proposals:
 
 
 
\begin{itemize}
 
\item {\bf Remove the low LUT goal.} It wasn't really achieved, and the
 
        proposals below will only increase the amount of logic the Zip CPU
 
        requires.  While I expect that the Zip CPU will always be somewhat
 
        of a light weight, it will never be the smallest kid on the block.
 
 
 
        I'm actually struggling with this idea.  The whole goal of the Zip
 
        CPU was to be light weight.  Wouldn't it make more sense to create and
 
        maintain options whereby it would remain lightweight?  For example, if
 
        the process accounting registers are anything but light weight, why
 
        keep them?  Why not instead make some compile flags that just turn them
 
        off, keeping the CPU lightweight?  The same holds for the prefetch
 
        cache.
 
 
 
\iffalse
 
\item {\bf Adjust the Zip CPU so that conditional instructions do not set
 
        flags}, although they may explicitly set condition codes if writing
 
        to the CC register.
 
 
 
        This is a simple change to the core, and may show up in new releases.
 
        % Fixed, 20150918
 
\fi
 
 
 
\item The `{\tt .V}' condition was never used in any code other than my test
 
        code.  Suggest changing it to a `{\tt .LE}' condition, which seems
 
        to be more useful.
 
 
 
\iffalse
 
\item Add in an {\bf unpredictable branch delay slot}, so that on any branch
 
        the delay slot may or may not be executed before the branch.
 
        Instructions that do not depend upon the branch, and that should be
 
        executed were the branch not taken, could be placed into the delay
 
        slot. Thus, if the branch isn't taken, we wouldn't suffer the stall,
 
        whereas it wouldn't affect the timing of the branch if taken. It would
 
        just do something irrelevant.
 
 
 
        % Changes made, 20150918, make this option no longer relevant
 
 
 
\item {\bf Re-engineer Branch Processing.}  There's no reason why a {\tt BRA}
 
        instruction should create five stall cycles.  The decode stage, plus
 
        the prefetch engine, should be able to drop this number of stalls via
 
        better branch handling.
 
 
 
        Indeed, this could turn into a simple means of branch prediction:
 
        if {\tt BRA} suffered a single stall only, whereas {\tt BRA.C}
 
        suffered five stalls, then {\tt BRA.!C} followed by {\tt BRA} would
 
        be faster than a {\tt BRA.C} instruction.  This would then allow a
 
        compiler to do explicit branch optimizations.
 
 
 
        Of course, longer branches using {\tt ADD X,PC} would still not be
 
        optimized.
 
 
 
        % DONE: 20150918 -- to include the ADD X,PC instructions
 
 
 
\item {\bf Request bus access for Load/Store two cycles earlier.}  The problem
 
        here is the contention for the bus between the memory unit and the
 
        prefetch unit.  Currently, the memory unit must ask the prefetch
 
        unit to release the bus if it is in the middle of a bus cycle.  At this
 
        point, the prefetch drops the {\tt STB} line on the next clock and must
 
        then wait for the last {\tt ACK} before releasing the bus.  If the
 
        request takes one clock, dropping the strobe line the next, waiting
 
        for an acknowledgement takes another, and then the bus must be idle
 
        for one cycle before starting again, these extra cycles add up.
 
        It should be possible to tell the prefetch stage to give up the bus
 
        as soon as the decoder knows the instruction will need the bus.
 
        Indeed, if done in the decode stage, this might drop the seven cycle
 
        access down by two cycles.
 
 
 
        % FIXED: 20150918
 
\fi
 
 
 
\item {\bf Consider a more traditional Instruction Cache.}  The current
 
        pipelined instruction cache just reads a window of memory into
 
        its cache.  If the CPU leaves that window, the entire cache is
 
        invalidated.  A more traditional cache, however, might allow
 
        common subroutines to stay within the cache without invalidating the
 
        entire cache structure.
 
 
 
\iffalse
 
\item {\bf Very Long Instruction Word (VLIW).}  Now, to speed up operation, I
 
        propose that the Zip CPU instruction set be modified towards a Very
 
        Long Instruction Word (VLIW) implementation. In this implementation,
 
        an instruction word may contain either one or two separate
 
        instructions. The first instruction would take up the high order bits,
 
        the second optional instruction the lower 16-bits. Further, I propose
 
        that any of the ALU instructions (SUB through LSR) automatically have
 
        a second instruction whenever their operand `B' is a register, and use
 
        the full 20-bit immediate if not. This will effectively eliminate the
 
        register plus immediate mode for all of these instructions.
 
 
 
        This is the minimal required change to increase the number of
 
        instructions per clock cycle. Other changes would need to take place
 
        as well to support this. These include:
 
        \begin{itemize}
 
        \item Instruction words containing two instructions would take two
 
                clocks to complete, while requiring only a single cycle
 
                instruction fetch.
 
        \item Instructions preceded by a label in the asseembler must always
 
                start in the high order word.
 
        \item VLIW's, once started, must always execute to completion. The
 
                upper word may set the PC, the lower word may not. Regardless
 
                of whether the upper word sets the PC, the lower word must
 
                still be guaranteed to complete before the PC changes. On any
 
                switch to (or from) interrupt context, both instructions must
 
                complete or none of the instructions in the word shall
 
                complete prior to the switch.
 
        \item STEP commands and BREAK instructions will only take place after
 
                the entire word is executed.
 
        \end{itemize}
 
 
 
        If done well, the assembler should be able to handle these changes
 
        with the biggest impacts to the user being increased performance and
 
        a loss of the register plus immediate ALU modes. (These weren't really
 
        relevant for the XOR, OR, AND, etc. operations anyway.) Machine code
 
        compatibility will not be maintained.
 
 
 
        A proposed secondary instruction set might consist of: a four bit
 
        operand (any of the prior instructions would be supported, with some
 
        exceptions such as moves to and from user registers while in
 
        supervisor mode not being supported), a 4-bit register result (PC not
 
        allowed), a 3-bit conditional (identical to the conditional for the
 
        upper word), a single bit for whether or not an immediate is present
 
        or not, followed by either a 4-bit register or a 4-bit signed
 
        immediate. The multiply instruction would steal the immediate flag to
 
        be used as a sign indication, forcing both operands to be registers
 
        without any immediate offsets.
 
 
 
        {\em Initial conversion of several library functions to this secondary
 
        instruction set has demonstrated little to no gain.   The problem was
 
        that the new instruction set was made by joining a rarely used
 
        instruction (ALU with register and not immediate) with a more common
 
        instruction.  The utility was then limited by the utility of the rare
 
        instrction, which limited the impact of the entire approach.  }
 
\else
 
\item {\bf Very Long Instruction Word (VLIW).}  The goal here would be to
 
        create a new instruction set whereby two instructions would be encoded
 
        in each 32--bit word.  While this may speed up
 
        CPU operation, it would necessitate an instruction redesign.
 
\fi
 
 
 
\end{itemize}
 
 
 
 
% Appendices
% Appendices
% Index
% Index
\end{document}
\end{document}
 
 
 
 
 
 

powered by: WebSVN 2.1.0

© copyright 1999-2022 OpenCores.org, equivalent to Oliscience, all rights reserved. OpenCores®, registered trademark.