Computer Architecture
CSCI-UA.0436
2011-12 Fall
Tues / Thurs 3:30-4:45
Room 317 CIWW
Allan Gottlieb

Start Lecture #1

Chapter 0: Administrivia

I start at Chapter 0 so that when we get to chapter 1, the numbering will agree with the text.

0.1: Contact Information

0.2: Course Web Page

There is a web site for the course. You can find it from my home page listed above.

0.3: Textbook

The course text is Hennessy and Patterson, Computer Organization and Design: The Hardware/Software Interface, 4th edition, which I will refer to as 4e.

0.4: Computer Accounts and Mailman Mailing List

0.5: Grades

Your grade will be a function of your midterm, final exam, and laboratory assignments (see below). I am not yet sure of the exact weightings, but each of the three will be important.

0.6: The Upper Left Board

I use the upper left board for lab/homework assignments and announcements. I should never erase that board. If you see me start to erase an announcement, please let me know.

I try very hard to remember to write all announcements on the upper left board and I am normally successful. If, during class, you see that I have forgotten to record something, please let me know. HOWEVER, if I forgot and no one reminds me, the assignment has still been given.

0.7: Homeworks and Labs

I make a distinction between homeworks and labs.

Labs are

Homeworks are

0.7.1: Homework Numbering

Homeworks are numbered by the class in which they are assigned. So any homework given today is homework #1. Even if I do not give homework today, the homework assigned next class will be homework #2. Unless I explicitly state otherwise, all homeworks assignments can be found in the class notes. So the homework present in the notes for lecture #n is homework #n (even if I inadvertently forgot to write it to the upper left board).

0.7.2: Doing Labs on non-NYU Systems

This course will have graphical labs so I expect you will work on your personal computers. However, you will send your labs via email from your nyu.edu email account.

0.7.3: Obtaining Help with the Labs

Good methods for obtaining help include

  1. Asking me during office hours (see web page for my hours).
  2. Asking the mailing list.
  3. Asking another student, but ...
    Your lab must be your own.
    That is, each student must submit a unique lab. Naturally, simply changing comments, variable names, etc. does not produce a unique lab.

0.7.4: Computer Language Used for Labs

Most if not all labs will be in logisim, a graphical language for drawing electronic circuits and simulating their behavior. I do not assume you know logisim now.

0.8: A Grade of Incomplete

The rules for incompletes and grade changes are set by the school and not the department or individual faculty member. The rules set by CAS can be found in http://cas.nyu.edu/object/bulletin0608.ug.academicpolicies.html state:

The grade of I (Incomplete) is a temporary grade that indicates that the student has, for good reason, not completed all of the course work but that there is the possibility that the student will eventually pass the course when all of the requirements have been completed. A student must ask the instructor for a grade of I, present documented evidence of illness or the equivalent, and clarify the remaining course requirements with the instructor.

The incomplete grade is not awarded automatically. It is not used when there is no possibility that the student will eventually pass the course. If the course work is not completed after the statutory time for making up incompletes has elapsed, the temporary grade of I shall become an F and will be computed in the student's grade point average.

All work missed in the fall term must be made up by the end of the following spring term. All work missed in the spring term or in a summer session must be made up by the end of the following fall term. Students who are out of attendance in the semester following the one in which the course was taken have one year to complete the work. Students should contact the College Advising Center for an Extension of Incomplete Form, which must be approved by the instructor. Extensions of these time limits are rarely granted.

Once a final (i.e., non-incomplete) grade has been submitted by the instructor and recorded on the transcript, the final grade cannot be changed by turning in additional course work.

0.9: Academic Integrity Policy

This email from the assistant director, describes the policy.

  Dear faculty,

  The vast majority of our students comply with the
  department's academic integrity policies; see

  www.cs.nyu.edu/web/Academic/Undergrad/academic_integrity.html
  www.cs.nyu.edu/web/Academic/Graduate/academic_integrity.html

  Unfortunately, every semester we discover incidents in
  which students copy programming assignments from those of
  other students, making minor modifications so that the
  submitted programs are extremely similar but not identical.

  To help in identifying inappropriate similarities, we
  suggest that you and your TAs consider using Moss, a
  system that automatically determines similarities between
  programs in several languages, including C, C++, and Java.
  For more information about Moss, see:

  http://theory.stanford.edu/~aiken/moss/

  Feel free to tell your students in advance that you will be
  using this software or any other system.  And please emphasize,
  preferably in class, the importance of academic integrity.

  Rosemary Amico
  Assistant Director, Computer Science
  Courant Institute of Mathematical Sciences

The university-wide policy is described here

Appendix C: Logic Design

Homework: Download logisim (first google hit) and play with it. The help button offers a tutorial, try it.

Remark: Appendix C is on the CD that comes with the book, but is not in the book itself. If anyone does not have convenient access to a printer, please let me know and I will print a black and white copy for you. The pdf on the CD is in color so downloading it to your computer for viewing in color is probably a good idea. If you have a color printer that is not terribly slow, you might want to print it in color—that's what I did.

Lab 1 part 1: (15 points), due 13 September. See the detailed assignment here.

  1. Download and install logisim
  2. Use logisim to draw one NAND, one NOR, one NOT, and one XOR.
  3. Draw wires so that
  4. Save the resulting .circ file.
  5. Mail it as an attachment to the TA, Elena Petrashen <eap367@cims.nyu.edu>. Send the email from your NYU.EDU account using subject arch-lab1-your-last-name.

C.1: Introduction

Homework: Read C.1

C.2: Gates, Truth Tables and Logic Equations

Homework: Read C.2

The word digital, when used in digital logic or digital computer means discrete. That is, the electrical values (i.e., voltages) of the signals in a circuit are treated as a integers (normally just 0 and 1).

The alternative is analog, where the electrical values are treated as real numbers.

To summarize, we will use only two voltages: high and low. A signal at the high voltage is referred to as 1 or true or set or asserted. A signal at the low voltage is referred to as 0 or false or unset or deasserted.

osc-traces

The assumption that at any time all signals are either 1 or 0 hides a great deal of engineering.

Since this is not an engineering course, we will ignore these issues and assume square waves.

In English, digit implies 10 (a digit is a finger), but not in computers.

Indeed, the word Bit is short for Binary digIT and binary means base 2 not 10.

0 and 1 are called complements of each other as are true and false (also asserted/deasserted; also set/unset)

A logic block can be thought of as a black box that takes in electrical signals puts out other electrical signals. There are two kinds of blocks

We shall study combinational blocks first and will will study sequential blocks later (in a few lectures).

Truth Tables

Since combinatorial logic has no memory, it is simply a (mathematical) function from its inputs to its outputs.

A common way to represent the function is using a Truth Table. A Truth Table has a column for each input and a column for each output. It has one row for each possible set of input values. So, if there are A inputs, there are 2A rows. In each of these rows the output columns have the output for that input.

Such a table is possible only because there are only a finite number of possible input values. Consider trying to produce a table for the mathematical function

  y = f(x) = x3 + 6 x2 - 12 x - 3.5

There would be only two columns (one for x and one for y) but there would need to be an infinite number of rows!

How many possible truth tables are there?

1-input, 1-output Truth Tables
1-in-1-out

Let's start with a really simple truth table, one corresponding to a logic block with one input and one output.

How many different truth tables are there for a one input one output logic block?

1-input, 1-output Truth Table
InOut
0?
1?

There are two columns (1+1) and two rows (21). Hence the truth table looks like the one on the right with the question marks filled in.

Since there are two question marks and each one can have one of two values there are just 22=4 possible truth tables. They are:

  1. The constant function 1, which has output 1 (i.e., true) for either input value.
  2. The constant function 0.
  3. The identity function, i.e., the function whose output equals its input. This logic block is sometimes called a buffer.
  4. An inverter. This function has output the opposite of the input.

We will see pictures for the last two possibilities very soon.

2-input, 1-output Truth Table
In1In2Out



00?
01?
10?
11?
2-input, 1-output Truth Tables

Three columns (2+1) and 4 rows (22).

How many are there? It is just the number ways can you fill in the output entries, i.e. the question marks. There are 4 output entries so the answer is 24=16.

Larger Truth Tables

How about 2 in and 8 out?

3 in and 8 out?

n in and k out?

Boolean algebra

We use a notation that looks like algebra to express logic functions and expressions involving them.

The notation is called Boolean algebra in honor of George Boole.

A Boolean value is a 1 or a 0.
A Boolean variable takes on Boolean values.
A Boolean function takes in boolean variables and produces boolean values.

Four Boolean functions are especially common.

  1. The (inclusive) OR Boolean function of two variables. Draw its truth table on the board. This function is written + (e.g. X+Y where X and Y are Boolean variables) and often called the logical sum. Three out of four output values in the truth table are the same as the result for a normal (mathematical) sum.
  2. AND. Draw its truth table on the board. AND is often called the logical product and written as a centered dot (like the normal product in regular algebra). So we would write A·B for A AND B. I sometimes write it as a period, because that is easier in html. As in regular algebra, when all the logical variables are just one character long, we indicate the product by juxtaposition, that is, AB represents the product of A and B (i.e., A AND B) when it is clear that it does not represent the two character symbol AB. All four truth table values are the same as the result for the normal (mathematical) product.
  3. NOT. Draw its truth table on the board. This is a unary operator (i.e., it has one argument, not two as above; functions with two inputs are called binary operators).
    NOT A is written A with a bar over it, which is hard to do in html so I instead write A'.
  4. Exclusive OR (XOR). Draw its truth table on the board. Written as ⊕, a + with a circle around it. True if exactly one input is true. In particular, remember that TRUE ⊕ TRUE = FALSE.

Homework: Draw the truth table of the Boolean function of 3 boolean variables that is true if and only if exactly 1 of the three variables is true.

Some manipulation laws

Remember this is Boolean Algebra.

How does one prove these laws??

Answer: It is simple, but tedious.

Write the truth tables for each side and see that the outputs are the same. Actually you write just one truth table with columns for all the inputs and for the outputs of both sides. You often write columns for intermediate outputs as well, but that is only a convenience. The key is that you have a column for the final value of the LHS (left hand side) and a column for the final value of the RHS and that these two columns have identical results.

Prove the first distributive law on the board. The following columns are required: the inputs A, B, C; the LHS A(B+C); and the RHS AB+AC. Beginners like us would also use columns for the intermediate results B+C, AB, and AC. (Note that I am now indicating product by simple juxtaposition.)

Homework: Prove the second distributive law.

Homework: Prove DeMorgan's laws.

Let's do (on the board) the example on page C-5.

Consider a logic function with three inputs A, B, and C; and three outputs D, E, and F defined as follows:

Compute the truth table and logic equations.

  1. Constructing the truth table.
    This is straightforward; simply fill in the 24 entries by looking at the definitions of D, E, and F.
  2. Producing the logic equations for D, E, and F can be done in two ways.
    1. Examine the column of the truth table for a given output and write one term for each entry that is a 1. This method requires constructing the truth table and might be called the method of perspiration.
    2. Look at the definition of D, E, and F and just figure it out. This might be called the method of inspiration.
      For D and F it is fairly clear. E requires some cleverness: the key idea is that exactly two are true is the same as (at least) two are true AND it is not the case that all three are true . So we have the AND of two expressions: the first is a three way OR and the second the negation of a three way AND.

Start Lecture #2

The first way we solved the previous example shows that any logic equation can be written using just AND, OR, and NOT. Indeed it shows more. Each entry in the output column of the truth table corresponds to the AND of three (because there are three inputs) literals.

A literal is either an input variable or the negation of an input variable.

In mathematical logic such a formula is said to be in disjunctive normal form because it is the disjunction (i.e., OR) of conjunctions (i.e., ANDs).

In computer architecture disjunctive normal form is often called two levels of logic because it shows that any formula can be computed by passing signals through only two logic functions, AND and then OR (assuming we are given the inputs and their compliments).

  1. First compute all the ANDs. There can be many, many of these, but they can all be computed at once using many, many AND gates.
  2. Compute the required ORs of the ANDs computed in step 1. There is only one OR for each output variable, but that OR can have many inputs.

With DM (DeMorgan's Laws) we can do quite a bit without resorting to truth tables.

For example one can ...

Homework: Show that the two expressions for E in the example above are equal.

Start to do the homework on the board.

Remark: You may ignore the references to Verilog in the text.

gates

Gates

Gates implement the basic logic functions: AND OR NOT XOR Equivalence. When drawing logic functions, we use the standard shapes shown to the right.

Note that none of the figures is input-output symmetric. That is, one can tell which lines are inputs and which are outputs without resorting to arrowheads and without the convention that inputs are on the left. Sometimes the figure is rotated 90 or 180 degrees.


Bubbles
bubbles

We often omit the inverters and draw little circles at the input or output of the other gates (e.g., AND OR). These little circles are sometimes called bubbles.

For example, the diagram on the right shows three ways a writing the same logic function.

This explains why the inverter is drawn as a buffer with an output bubble.

Show why the picture above for equivalence is correct. That is, show that equivalence is the negation of XOR. Specifically, show that AB + A'B' = (A ⊕ B)'.

  (A ⊕ B)' =
  (A'B+AB')' =
  (A'B)' (AB')' =
  (A''+B') (A'+B'') =
  (A + B') (A' + B) =
  (A + B') A'  +  (A + B') B =
  AA' + B'A' + AB + B'B =
  0   + B'A' + AB + 0 =
  AB + A'B'

Homework: C.4.

Homework: Recall the Boolean function E that is true if and only if exactly 1 of the three variables is true. We have already drawn the truth table.
Draw a logic diagram for E using AND OR NOT.
Draw a logic diagram for E using AND OR and bubbles.

Universal Gates

A set of gates is called universal if these gates are sufficient to generate all logic functions.

nand and nor

Definition: NOR (NOT OR) is true when OR is false.

Draw the truth table on the board.

Definition: NAND (NOT AND) true when AND is false.

Draw the truth table on the board.

We can draw both NAND and NOR in two ways as shown in the diagram on the right. The top pictures are from the definition; the bottom use DeMorgan's laws.

Theorem A 2-input NOR is universal and a 2-input NAND is universal.

Proof We will show that you can get A', A+B, and AB using just a two input NOR.

Draw the truth tables showing the last three statements. Also say why they are correct, i.e., we are now at the point where simple identities like these don't need truth tables.

Homework: Show that a 2-input NAND is universal.

If fact the above proof was overkill, it would have been enough to show that you can get A' and A+B.
Why?
Because we already know that the pair OR NOT is universal.
It would also have been enough to show that you can get A' and AB.
inverter

Sneaky way to see that NAND is universal.


Minimizing the Gate Count

We have seen how to implement any logic function given its truth table. Indeed, the natural implementation from the truth table uses just two levels of logic. But that implementation might not be the simplest possible. That is, we may have more gates than are necessary.

Trying to minimize the number of gates is decidedly NOT trivial. Many texts, including one by Mano, cover the topic of gate minimization in detail. We will not cover it in this course. It is mentioned and referenced, but not covered, in 4e. I actually like the topic, but it takes a few lectures to cover well and it is no longer used in practice since it is done automatically by CAD tools.

Minimization is not unique, i.e. there can be two or more minimal forms.

Given A'BC + ABC + ABC'
Combine first two to get BC + ABC'
Combine last two to get A'BC + AB

Don't Cares (preview)

Sometimes when building a circuit, you don't care what the output is for certain input values. For example, that input combination might be known not to occur. Another example occurs when, for some combination of input values, a later part of the circuit will ignore the output of this part. These are called don't care outputs. Making use of don't cares can reduce the number of gates needed.

One can also have don't care inputs when, for certain values of a subset of the inputs, the output is already determined and you don't have to look at the remaining inputs. We will see a case of this very soon when we do multiplexors.

An aside on theory

Putting a circuit in disjunctive normal form (i.e. two levels of logic) means that every path from the input to the output goes through very few gates. In fact only two, an OR and an AND. Maybe we should say three since the AND can have a NOT (bubble). Theoreticians call this number (2 or 3 in our case) the depth of the circuit. Se we see that every logic function can be implemented with small depth. But what about the width, i.e., the number of gates.

The news is bad. The parity function takes n inputs and gives TRUE if and only if the number of TRUE inputs is odd. If the depth is fixed (say limited to 3), the number of gates needed for parity is exponential in n.

C.3 Combinational Logic

Homework: Read C.3.

Generic Homework: Read sections in book corresponding to the lectures.

Decoders (and Encoders)

Imagine you are writing a program and have 32 flags, each of which can be either true or false. You could declare 32 variables, one per flag. If permitted by the programming language, you would declare each variable to be a bit. In a language like C, without bits, you might use a single 32-bit int and play with shifts and masks to store the 32 flags in this one word.

In either case, an architect would say that you have these flags fully decoded. That is, you can specify the values of each of the bits.

Now imagine that for some reason you know that, at all times, exactly one of the flags is true and the other are all false. Then, instead of storing 32 bits, you could store a 5-bit integer that specifies which of the 32 flags is true. This is called fully encoded. For an example, consider radio buttons on a web page.

A 5-to-32 decoder converts an encoded 5-bit signal into the decoded 32-bit signal having the one specified signal true.

A 32-to-5 encoder does the reverse operations. Note that the output of an encoder is defined only if exactly one input bit is set (recall set means true).

encoder-decoder

The the top diagram on the right shows a 3-to-8 decoder.

Similarly, the bottom diagram shows an 8-3 encoder.

Why do we use decoders and encoders?

Remark: Lab 1 part 2 assigned, due 15 September 2011.

Multiplexors

2-way mux

A multiplexor, often called a mux or a selector is used to select one (output) signal from a group of (input) signals based on the value of a group of (select) signals. In the 2-input mux shown on the right, the select line S is thought of as an integer 0..1. If the integer has value j then the jth input is sent to the output.

Construct on the board an equivalent circuit with ANDs and ORs in two ways:

  1. Construct a truth table with 8 rows (don't forget that, despite its name, the select line is an input) and write the sum of product form, one product for each row and a large 8-input OR. This is the canonical two-levels of logic solution. (Method of perspiration.)
  2. A simpler, more clever, two-levels of logic solution. Two ANDs, one per input (not including the selector). The selector goes to each AND, one with a bubble. The output from the two ANDs goes to a 2-input OR. (Method of inspiration.)

Start Lecture #3

Remark: You should all have accounts on i5.nyu.edu.

4-way mux

The diagram on the right shows a 4-input MUX.

Construct on the board an equivalent circuit with ANDs and ORs in three ways:

  1. Construct the truth table (64 rows!) and write the sum of products form, one product (6-input AND) for each row and a gigantic 64-way OR. Just start this, don't finish it. (Perspiration.)
  2. A simpler (more clever) two-level logic solution. Four ANDS (one per input), each gets one of the inputs and both select lines with appropriate bubbles. The four outputs go into a 4-way OR. (Inspiration.)
  3. Construct a 2-input mux (using the clever solution). Then construct a 4-input mux using a tree of three 2-input muxes. One select line is used for the two muxes at the base of the tree, the other is used at the root. (Hierarchical.)
  4. This last solution is our first illustration of the usefulness of the hierarchical feature of logisim.

All three of these methods generalize to a mux with 2k input lines, and k select lines.

A 2-way mux is the hardware analogue of if-then-else.

    if S=0
        M=A
    else
        M=B
    endif

A 4-way mux is an if-then-elif-elif-else

    if S1=0 and S2=0
        M=A
    elif S1=0 and S2=1
        M=B
    elif S1=1 and S2=0
        M=C
    else -- S1=1 and S2=1
        M=D
    endif

Don't Cares (again)

SIn0In1Out


00X0
01X1
1X00
1X11

Consider a 2-input mux. If the selector is 0, the output is I0 and the value of I1 is irrelevant. Thus, when the selector is 0, I1 is a don't care input. Similarly, when the selector is 1, I0 is a don't care input.

On the right we see the resulting truth table. Recall that without using don't cares the table would have 8 rows since there are three inputs; in this example the use of don't cares reduced the table size by a factor of 2.

The truth table for a 4-input mux has 64 rows, but the use of don't care inputs has a dramatic effect. When the selector is 01 (i.e, S0 is 0 and S1 is 1), the output equals the value of I1 and the other three I's are don't care. A corresponding result occurs for other values of the selector.

Homework: Draw the truth table for a 4-input mux making use of don't care inputs. What size reduction occurred with the don't cares?

Homework: C.13.
C.10. (Assume you have constant signals 1 and 0 as well.)

Recall that a don't care output occurs when for some input values (i.e., rows in the truth table), we don't care what the value is for certain outputs.

Powers of 2 NOT Required

How can one construct a 5-way mux?

Construct an 8-way mux and use it as follows.

Can do better by realizing the select lines equalling 5, 6, or 7 are don't cares and hence the 8-way can be customized and would use fewer gates than an 8-way mux.

PLAs—Programmable Logic Arrays (and PALs)

ABCDEF


000000
001100
010100
011110
100100
101110
110110
111101

The idea is to partially automate the algorithmic way you can produce a circuit diagram (in the sums of product form) from a given truth table. Since the form of the circuit is always a bunch of ANDs feeding into a bunch of ORs, we can manufacture all the gates in advance of knowing the desired logic functions and, when the functions are specified, we just need to make the necessary connections from the ANDs to the ORs. In essence all possible connections are configured but with switches that can be open or closed.

Actually, the words above better describe a PAL (Programmable Array Logic) than a PLA, as we shall soon see.

Consider the truth table on the upper right, which we have seen before. It has three inputs A, B, and C, and three outputs D, E, F.

pla1

To the right we see the corresponding logic diagram in sum of products form.

Recall how we construct this diagram from the truth table.

pla2

To the right, the above figure is redrawn in a more schematic style.

Finally, we can draw a PLA in the more abstract form shown on bottom right.

pla3

When a PLA is manufactured all the specified connections are made. That is, a manufactured PLA is specific for a given circuit. Hence the name Programmable Logic Array is somewhat of a misnomer since the device is not programmable by the user.

Homework: C.11 and C.12.

PAL (Programmable Array Logic)

A PAL can be thought of as a PLA in which the final dots are made by the user. The manufacturer produces a sea of gates. The user programs it to the desired logic function by adding the dots.

ROMs

One way to implement a mathematical function (or a java function without side effects) is to perform a table lookup.

A ROM (Read Only Memory) is the analogous way to implement a logic function.

Important: A ROM does not have state. It is another combinational circuit. That is, it does not represent memory. The reason is that once a ROM is manufactured, the output depends only on the input. I realize this sounds wrong, but it is right.

Indeed, we will shortly see that a ROM is like a PLA. Both are structures that can be used to implement a truth table.

The key property of combinational circuits is that the outputs depend only on the inputs. This property (having no state) is false for a RAM chip: The input to a RAM is (like the input to a ROM) an address and (unlike a ROM) an operation (read vs write). The RAM (given a read request) responds by presenting at its outputs the value CURRENTLY stored at that address. Thus knowing just the input (i.e., the address and the operation) is NOT sufficient for determining the output. Whereas; knowing the address supplied to a given ROM is sufficient for determining the output.

A PROM is a programmable ROM. That is, you buy the ROM with nothing in its memory and then before it is placed in the circuit you load the memory, and never change it. This is like a CD-R. Again, as with a ROM, when you are using a PROM in a circuit, the output is determined by the input (the address).

An EPROM is an erasable PROM. It costs more but if you decide to change its memory this is possible (but is slow). This is like a CD-RW.

Normal EPROMs are erased by some ultraviolet light process that is performed outside the circuit. But EEPROMs (electrically erasable PROMS) are not as slow and are done electronically. Since this is done inside the circuit you could consider it a RAM if you considered the erasing as a normal circuit operation.

Flash is a modern EEPROM that is reasonably fast.

All these EPROMS are erasable not writable, i.e. you can't just change one byte to an arbitrary value. (Some modern flash rams can nearly replace true RAM and perhaps should not be called EPROMS).

ROMs and PLAs

A ROM is similar to PLA

Full Truth Table
ABCDEF


000000
001101
010011
011110
100111
101110
110110
111110

Don't Cares (bigger example)

Sometimes not all the input and output entries in a truth table are needed. We indicate this with an X and it can result in a smaller truth table.

Truth Table with Output Don't Cares
ABCDEF


000000
001101
010011
01111X
10011X
10111X
11011X
11111X

The top diagram on the right is the full truth table for the following example (from the book) that we have considered before. Consider a logic function with three inputs A, B, and C, and three outputs D, E, and F.

The full truth table has 7 minterms (rows with at least one nonzero output).

The middle truth table has the output don't cares indicated.


Truth Table with Input and Output Don't Cares
ABCDEF


000000
001101
010011
X1111X
1XX11X

Now we do the input don't cares

The resulting truth table is also shown on the right.

These don't cares are important for logic minimization. Compare the number of gates needed for the full truth table and the reduced truth table. There are techniques for minimizing logic, but we will not cover them.

Arrays of Logic Elements

arrays

Often we want to consider signals that are wider than a single bit. An array of logic elements is used when each of the individual bits is treated similarly. As we will soon see, sometimes most of the bits are treated similarly, but there are a few exceptions. For example, a 32-bit structure might treat the lob (low order bit) and hob differently from the others. In such a case we would have an array 30 bits wide and two 1-bit structures.

C.4: Using a Hardware Description Language

Start Lecture #4

C.5: Constructing a Basic Arithmetic Logic Unit (ALU)

We will produce logic designs for the integer portion of the MIPS ALU. The floating point operations are more complicated and will not be implemented.

MIPS is a computer architecture widely used in embedded designs. In the 80s and early 90s, it was quite popular for desktop (or desk-side) computers. This was the era of the killer micros that decimated the market for minicomputers. (When I got a DECstation desktop with a MIPS R3000, I think that, for a short while, it was the fastest integer computer at NYU.)

Much of the design we will present (indeed, all of the beginning part) is generic. I will point out when we are tailoring it for MIPS.

A 1-bit ALU

Our first goal will be a 1-bit wide structure that computes the AND, OR, and SUM of two 1-bit quantities. For the sum there is actually a third input, CarryIn, and a 2nd output, CarryOut.

Since out basic logic toolkit already includes AND and OR gates, our first real task is a 1-bit adder.

Half Adder

If the final goal was a 1-bit ALU, then we would not have a CarryIn. For a multi-bit ALU, the CarryIn for each bit is the CarryOut of the preceding lower-order bit (e.g., the CarryIn for bit 3 is the CarryOut from bit 2). When we don't have a CarryIn, the structure is sometimes called a half adder. Don't treat the name too seriously; it is not half of an adder.

A half adder has the following inputs and outputs.

Draw the truth table.

Homework: Draw the logic diagram.

Full Adder

adder

Now we include the carry-in.

Homework:

Combining 1-bit AND, OR, and ADD

adder-symbol

We have implemented 1-bit versions of AND (a basic gate), OR (a basic gate), and SUM (the FA just constructed, which we henceforth draw as shown on the right). We now want a single structure that, given another input (the desired operation, a so called control line), produces as output the specified operation.

There is a general principle used to produce a structure that yields either X or Y depending on the value of operation.

  1. Implement a structure that always computes X.
  2. Implement another structure that always computes Y.
  3. Mux X and Y together using operation as the select line.

This mux, with an operation select line, gives a structure that sometimes produces one result and sometimes produces another. Internally both results are always produced.

alu-1bit

In our case we have three possible operations so we need a three way mux and the select line is a 2-bit wide bus. With a 2-bit select line we can specify 4 operations, for now we are using only three.

We show the diagram for this 1-bit ALU on the right.

In subsequent diagrams the Operation input will be shown in green to distinguish it as a control line rather than a data line. (Now is it drawn in blue to show that it is new in this diagram.) The goal is to produce two bits of result from 2 (AND, OR) or 3 (ADD) bits of data. The 2 bits of control tell what to do, rather than what data to do it on.

The extra data output (CarryOut) is always produced. Presumably if the operation was AND or OR, CarryOut is not used.

I believe the distinction between data and control will become quite clear as we encounter more examples. However, I wouldn't want to be challenged to give a (mathematically precise) definition.

Remarks:

  1. Show the wire bundle and splitter features in logisim
  2. Lab 2 assigned; due 22 September.

A 32-bit ALU

A 1-bit ALU is interesting, but we need a 32-bit ALU to implement the MIPS 32-bit operations, acting on 32-bit data values.

For AND and OR, there is almost nothing to do; a 32-bit AND is just 32 1-bit ANDs so we can simply use an array of logic elements.

However, ADD is a little more interesting since the bits are not quite independent: The CarryOut of one bit becomes the CarryIn of the next.

A 32-bit Adder

adder4

Let's start with a 4-bit adder.

How about a 32-bit adder, or even an an n-bit adder?

alu-32bit

Combining 32-bit AND, OR, and ADD

To obtain a 32-bit ALU, we put together the 1-bit ALUs in a manner similar to the way we constructed a 32-bit adder from 32 FAs. Specifically we proceed as follows and as shown in the figure on the right.

  1. Use an array of logic elements for the logic. The individual logic element is the 1-bit ALU.
  2. Use buses for A, B, and Result.
  3. Broadcast Operation to all of the internal 1-bit ALUs. This means wire the external Operation to the Operation input of each of the internal 1-bit ALUs.

Facts Concerning (4-bit) Two's Complement Arithmetic

Remark
This is one place were the our treatment must go a little out of order. Appendix C in the book assumes you have read the chapter on computer arithmetic; in particular it assumes that you know about two's complement arithmetic.

I do not assume you know this material (although I know many of you do). We will cover it (briefly) later, when we do that chapter. What I will do here is assert some facts about two's complement arithmetic that we will use to implement the circuit for SUB.
End of Remark.

For simplicity I will be presenting 4-bit arithmetic. We are really interested in 32-bit arithmetic, but the idea is the same and the 4-bit examples are much shorter (and hence less likely to contain typos).

4-bit Twos's Complement Numbers

With 4 bits, there are 16 numbers. Since twos complement notation has one representation for each number, there are 15 nonzero values. Since there are an odd number of nonzero values, there cannot be the same number of positive and negative values. In fact 4-bit two's complement notation has 8 negative values, and 7 positive values. (In one's complement notation there are the same number of positive and negative values, but there are two representations for zero, which is inconvenient.)

The high order bit (hob) on the left is the sign bit. The sign bit is zero for positive numbers and for the number zero; the sign bit is one for negative numbers.

Zero is written simply 0000.

1-7 are written 0001, 0010, 0011, 0100, 0101, 0110, 0111. That is, you set the sign bit zero and write 1-7 using the remaining three lob's. This last statement is also true for zero.

-1, -2, ..., -7 are written by taking the two's complement of the corresponding positive number. The two's complement is computed in two steps.

  1. Take the (ordinary) complement, i.e. turn ones to zeros and vice versa. This is sometimes called the one's complement.
    For example, the (4-bit) one's complement of 3 is 1100.
  2. Add 1.
    For example, the (4-bit) two's complement of 3 is 1101.

If you take the two's complement of -1, -2, ..., -7, you get back the corresponding positive number. Try it.

If you take the two's complement of zero you get zero. Try it.

What about the 8th negative number?
-8 is written 1000.
But if you take its (4-bit) two's complement, you must get the wrong number because the correct number (+8) cannot be expressed in 4-bit two's complement notation.

Two's Complement Addition and Subtraction

Amazingly easy (if you ignore overflows).

Implementing SUB (Together With AND, OR, and ADD)

alu-sub-1bit

No change is needed to our circuit above to handle two's complement numbers for AND/OR/ADD. That statement is not clear for ADD and will be shown true later in the course.

We wish to augment the ALU so that we can perform subtraction as well. As we stated above, A-B is obtained by taking the two's complement of B and adding. A 1-bit implementation is drawn on the right with the new structures in blue (I often use blue for this purpose). The enhancement consists of

  1. Using an inverter to get the one's complement of B.
  2. Using a mux with control line Binvert (in green) to select whether B or B' is fed to the adder.
  3. Using a clever trick to obtain the effect of adding B's two complement when we are actually adding B's one complement. Namely we set Cin, the carry-in to the lob, equal to 1 instead of 0. This trick increases the sum by one and, as a result, calculates (A+B')+1=A+(B'+1), which is A plus the two's complement of B, which is A-B.
  4. So, for the lob, CarryIn is kinda-sorta a data line used as a control line.
  5. As before, setting Operation to 00 and 01 gives AND and OR respectively, providing we de-assert Binvert. CarryIn is a don't care for AND and OR.
  6. To implement addition we use opcode 10 as before and de-assert both Binvert and CarryIn
  7. To implement subtraction we again use opcode 10 but we assert both Binvert and CarryIn
alu-sub-32bit
Extending to 32 Bits

A 32-bit version is simply a bunch of the 1-bit structures wired together as shown on the right. I use CarryIn and CarryOut when referring to the external carry signals of the entire 32-bit structure. Please do not confuse them with Cin and Cout, the corresponding signals to each individual 1-bit structure.

Tailoring the 32-bit ALU to MIPS

AND, OR, AND, and SUB are found in nearly all ALUs. In that sense, the construction up to this point has been generic. However, most real architectures have some extras. For MIPS they include.

  1. NOR, not very special and very easy.
  2. Overflow handling, common but not so easy.
  3. Set on less than (slt), not common and not so easy.
  4. Equality test, not very special and easy.

Implementing NOR

alu nor

We noted above that our ALU already gives us the ability to calculate AB', an uncommon logic function. A MIPS ALU needs NOR and, by DeMorgan's law

  A NOR B = (A + B)' = A'B'

which is rather close to AB', we need just invert A as well as B.

The diagram on the right shows the added structures: an inverter to get A', a mux to choose between A and A', and a control line for the mux.

NOR is obtained by asserting Ainvert and Binvert and setting Operation=00.

The other operations are done as before, with Ainvert de-asserted.

The 32-bit version is a straightforward ...

Homework: Draw the 32-bit ALU that supports AND, OR, ADD, SUB, and NOR.

Start Lecture #5

Remark: See if final date can be moved.

Overflows

Remark: As with two's complement arithmetic, I just present the bare boned facts here; they are explained later in the course.

alu overflow hob

The facts are trivial (although the explanation is not). Indeed there is just one fact.

  1. An overflow occurs for two's complement addition (which includes subtraction) if and only if the carry-in to the sign bit does not equal the carry out from the sign bit.

Only the hob portion of the ALU needs to be changed. We need to see if the carry-in is different from the carry-out, but that is exactly XOR. The simple modification to the hob structure is shown on the right.

Do on the board 4-bit twos complement addition of

  1. 1 + 1
  2. -1 + -1 Note that there is NO overflow despite a carry-out.
  3. 6 + 6
  4. -6 + -6

The 32-bit version is again a straightforward ...

Homework: Draw the 32-bit ALU that supports AND, OR, ADD, SUB, and NOR and that asserts an overflow line when appropriate.

Note that to ease the homework and, more importantly, the real design, we can use the enhanced 1-bit ALU for all bits and simply ignore the overflow output for all but the HOB.

Implementing Set on Less Than (SLT)

We are given two 32-bit, two's complement numbers A and B as input and seek a 32-bit result that is 1 if A<B and 0 otherwise. Note that only the lob of the result varies; the other 31 bits are always 0.

The implementation is fairly clever as we shall see.

Idea #1

The first idea is rather simple. The sign of A-B is 1 precisely when A<B. Thus, to implement slt, we need to set the LOB of the result equal to the sign bit of the subtraction A-B, and set the rest of the result bits to zero.

Idea #2
alu-less-1bit

Give the 4-way mux another (i.e., a fourth) input, called Less. This input is brought in from outside the bit cell. To generate slt, we make the select line to the mux equal to 11 so that the the output is the this new input. See the diagram on the right.

Idea #3.
slt overview

Use the settings just mentioned so that the adder computes A-B (and the mux throws it away). Modify the HOB logic as follows (again it might be easier to do this modification for all bits, but just use the result from the HOB).

Why didn't I show a detailed diagram for this method?
Because this method is not used.

Why isn't the method used?
Because it is wrong!

The problem with the above solution is that it ignores overflows. Consider the following 4-bit (instead of 32-bit) example.

alu set hob

The fix is to use the correct rule for less than rather than the incorrect rule the sign bit of A-B is 1, which ignores overflows and gives the wrong answer when an overflow occurs.

Homework: Figure out the correct rule, i.e. a non-pictorial version of problem C.24. Hint: When an overflow occurs, the sign bit is definitely wrong.

An even bigger hint is that the diagram on the right shows the correct calculation of Set in the HOB.

This is an example where explaining the bug is harder than fixing it.

The Final 1-bit Cell

alu-1-bit-external

On the right is a picture of the final 1-bit ALU that shows the external interface but hides the internal details.

Recall that our goal is a 32-bit ALU. It will contain 32 of the 1-bit ALUs we have just constructed. When drawing the larger structure, we want to hide the details of the individual 1-bit cells. Thus we draw the 1-bit structure as shown on the right.

In the pictures below, to save space, I sometimes omit the labels on the interfaces of the internal structures. I try to ensure that they are in the same order as in the picture on the right (let me know of any bugs you see) and try to have enough information in the picture so that you do not need to know the order.

alu-32bit-final

Equality Detection

To see if A = B we simply form A-B and test if the result is zero.

The Final Result

The final 32-bit ALU is shown below on the right. Note that all 32 1-bit cells are identical; it is only the inter-cell wiring that differs. This is important!

CarryIn Missing

Although each 1-bit cell has 4 inputs (Ainvert, Binvert, Cin, Operation), the entire 32-bit ALU has only 3 inputs (CarryIn is not present).

For all bits except the LOB, Cin is wired to Cout of the preceding bit.

For the LOB, Cin is the same as Binvert. So we define a single external line Bnegate, which is sent to Binvert for every 1-bit alu and is also sent to Cin of the LOB. Thus there is no CarryIn signal needed.

All Bits the Same Internally
alu

Again note that all the bits have the same circuit. The lob and hob have special external wiring; the other 30 bits are wired the same.

ALU Symbol

To the right we see the symbol used for an ALU. Note that we have combined the two 1-bit control lines (Ainvert and Bnegate) plus the 2-bit Operation control line into a single 4-bit control line called ALUOperation.

The book uses the label Zero for the middle output. I believe a better label would be Equal since the line gives the Boolean result A==B (which is computed as (A-B)==0)). I use the term Equal Zero, rather than Equal, to ease a comparison with the book.

Setting the Control Lines

function4-bit cntlAinvBnegOper
AND00000000
OR00010001
ADD00100010
SUB01100110
slt01110111
NOR11001100

The ALU can directly perform the following MIPS instructions by setting the control lines as indicated in the table on the right.

Remark: We have developed the logic needed to implement 6 machine instructions. The technical term is that we have developed the data path. That is one of three tasks needed for a full implementation. The other two are:

  1. Given the actual machine instruction, we need to determine the value for ALUOperation. This is not too hard and involves no new concepts, as we shall see.
  2. We need to implement the register file since the operands for all these instructions are registers. Although this is also not terribly difficult it definitely requires an important new concept: sequential logic, i.e, circuits with memory or state.

Before we either of these tasks, we will learn a much faster method for addition (and subtraction).

Defining the MIPS ALU in Verilog

C.6: Faster Addition: Carry Lookahead

The alu above is not used in practice since it is too slow. The fundamental problem is that calculating the ith bit of the result requires the carry out of the i-1th bit. For this reason the above alu is said to perform a ripple carry since the carry computation ripples along from the lob to the hob and. Thus, for a 64-bit addition, the hob will take a long time to compute.

The adder we will study next is much faster than the ripple adder we did before, especially for wide (i.e., many bit) addition. (Do to the joys of two's complement addition, any adder can subtract by complementing the bits and adding—as we did above.

Fast Carry Using Infinite Hardware

This is a simple (theoretical) result, but not practical.

  1. An adder is a combinatorial circuit hence it can be constructed with two (or three if you count the bubbles) levels of logic. Done.
  2. Consider 32-bit (or 64-bit, or 128-bit, or N-bit) addition, R=A+B.
  3. The above analysis could be applied to any logic function; here are the calculations specific for addition.
  4. So theoretical, but impractical, fast circuits are easy to derive.

Fast Carry Using the First Level of Abstraction: Propagate and Generate

At each bit position we have two input bits ai and bi as well as a CarryIn input. We now define two other bits called propagate (pi=ai+bi) and generate (gi=aibi), which have the following properties.

The reason for the name propagate is that p is true if the current bit will propagate a carry from its input to its output. More precisely:

  if (pi) then {
    if there is a carry in to bit i then
    there is a carry out from bit i
  }
  if NOT (pi) then {
    there is never a carry out from bit i
  }

The reason for the name generate is that g is true if the current bit will generate a carry out (independent of the carry in). More precisely:

  if (gi) then {
    there is definitely a carry out from bit i
  }
  if NOT (gi) then {
    there might not be a carry out from bit i
hp C.6.1 (F0422 2e)

These key formulas are quite simple, but are very useful. To repeat:
    Generate:     gi = ai·bi
    Propagate:    pi = ai+bi

The diagram on the right, from P&H, gives a plumbing analogue for generate and propagate. A full size version of the diagram (in pdf) is here. (The plumbing diagrams in these notes are from the 2e; the colors changed between additions, but the contents are the same.)

The point is that liquid enters the main pipe if either the initial CarryIn or one of the generates is true. The water exits the pipe at the lower left (i.e., there is a CarryOut for this bit position) if all the propagate valves are open from the lowest liquid entrance to the exit.

Given the generates and propagates, we can calculate all the carries for a 4-bit addition as follows (recall that c0=Cin is an input). These formulas correspond directly to the plumbing picture on the right. For simplicity, I will stop writing subscripts smaller and subtended.

  c1 = g0 + p0 c0
  c2 = g1 + p1 c1 = g1 + p1 g0 + p1 p0 c0
  c3 = g2 + p2 c2 = g2 + p2 g1 + p2 p1 g0 + p2 p1 p0 c0
  c4 = g3 + p3 c3 = g3 + p3 g2 + p3 p2 g1 + p3 p2 p1 g0 + p3 p2 p1 p0 c0

Thus we can calculate c1 ... c4 in just two additional gate delays given the p's and g's. (We assume one gate can accept up to 5 inputs). Since we get gi and pi after one gate delay, the total delay for calculating all the carries is 3 gate delays. This includes calculating c4=CarryOut.

Start Lecture #6

Note: The above formulas are for 4-bit arithmetic. A crucial point is that, when we use more bits, the formulas will NOT get bigger.

4bit cla

Each bit of the sum si can be calculated in 2 gate delays given ai, bi, and ci. Since, we just say that the ci's can be calculated in 3 gate delays, for 4-bit addition, 5 gate delays after we are given a, b and Carry-In, we have calculated s and Carry-Out.

We show this in the diagram on the right.

Thus, for 4-bit addition, 5 gate delays after we are given a, b and the Carry-In, we have calculated s and the Carry-Out using a modest amount of realistic (no more than 5-input) logic.

How does the speed of this carry-lookahead adder CLA compare to our original ripple-carry adder?

Fast Carry Using the Second Level of Abstraction

We have finished the design of a 4-bit CLA. Our next goal is a 16-bit fast adder. Let's consider, at varying levels of detail, five possibilities.

  1. Ripple carry. Simple, easy to do, but not fast (16*2=32 gate delays).

  2. General 2 levels of logic. Always applicable and very fast (2 gate delays). We know how to do it, but it is not practical.

  3. Extend the above 4-bit CLA design to 16 bits.
  4. cla-hybrid-16bit
  5. Put together four of the 4-bit CLAs. Shown in the diagram to the right is a schematic of our 4-bit CLA and a 16-bit adder constructed from four of them.
  6. Be more clever and put together the 4-bit CLAs in a carry-lookahead manner. One could call the result a 2-level CLA.
hpfig C.6.2

Super Propagate and Super Generate

We start the adventure by defining super propagate and super generate bits.

From these super propagates and super generates, we can calculate the super carries, i.e. the carries for the four 4-bit adders. We will use four of the 4-bit CLAs to form our 16-bit CLA but we need to calculate all the Carry-In's to the 4-bit CLAs at once NOT in a ripple-carry manner.

  C0 = c0
  C1 = G0 + P0 c0
  C2 = G1 + P1 C1 = G1 + P1 G0 + P1 P0 c0
  C3 = G2 + P2 C2 = G2 + P2 G1 + P2 P1 G0 + P2 P1 P0 c0
  C4 = G3 + P3 C3 = G3 + P3 G2 + P3 P2 G1 + P3 P2 P1 G0 + P3 P2 P1 P0 c0

But this looks terrific! These super carries are what we need to combine four 4-bit CLAs into a 16-bit CLA in a carry-lookhead manner. Recall that the hybrid approach suffered because the carries from one 4-bit CLA to the next (i.e., the super carries) were done in a ripple carry manner.

Since it is not completely clear how to combine the pieces so far presented to get a 16-bit, 2-level CLA, I will give a pictorial account very soon.

Before the pictures, let's assume the pieces can be put together and see how fast the 16-bit, 2-level CLA actually is. Recall that we have already seen two practical 16-bit adders: A ripple carry version taking 32 gate delays and a hybrid structure taking 14 gate delays. If the 2-level design isn't faster than 14 gate delays, we won't bother with the pictures.

Remember we are assuming 5-input gates. We use lower case p, g, and c for propagates, generates, and carries; and use capital P, G, and C for the super- versions.

  1. We calculate the p's and g's (lower case) in 1 gate delay (as with the 4-bit CLA).
  2. We calculate the P's one gate delay after we have the p's or 2 gate delays after we start.
  3. The G's are determined 2 gate delays after we have the g's and p's. So the G's are done 3 gate delays after we start.
  4. The C's are determined 2 gate delays after the P's and G's. So the C's are done 5 gate delays after we start.
  5. Now the C's are sent back to the 4-bit CLAs, which have already calculated the p's and g's. The c's are calculated in 2 more gate delays (7 total) and the s's 2 more after that (9 total).

Since 9<14, let the pictures begin!

Step 1: Minor Surgery on the 4-bit CLA
cla-pg 4-bit
Combining Four 4-bit CLA-PG's with a CL Block

Next we put four of these 4-bit CLAs together with a new structure called a Carry Lookahead Block (CL-Block) that calculates the carries needed by the 4-bit CLA-PGs using the P's, G's and Cin=C0. The result is a 16-bit CLA!

The formulas for the Cs are above but I repeat them here

  C0 = Cin
  C1 = G0 + P0 Cin
  C2 = G1 + P1 C1 = G1 + P1 G0 + P1 P0 Cin
  C3 = G2 + P2 C2 = G2 + P2 G1 + P2 P1 G0 + P2 P1 P0 Cin
  C4 = G3 + P3 C3 = G3 + P3 G2 + P3 P2 G1 + P3 P2 P1 G0 + P3 P2 P1 P0 Cin

Note that I do not call the new structure a 4-bit CL Block or a 16-bit CL block. More on this latter

cla clb

The diagram on the right shows the result, the color of the wires indicate when the values are calculated

  1. The blue lines are inputs.
  2. Then the red lines are calculated.
  3. Then the magenta.
  4. Finally the brown.

The last paragraph is stated sloppily. Gates are always calculating their outputs from their inputs. When we say something is calculated in k gate delays, we mean that the outputs are correct k gate delays after the inputs are correct. A more accurate version of the previous paragraph would be

  1. The blue lines are input, which are assumed to be valid when we start the addition.
  2. The red lines are valid 3 gate delay after the blue (actually the Ps needs only 2 gate delays, but we use the Ps and Gs together so need to wait for the Gs).
    Summary: the red lines are valid 3 gate delays after we start
  3. The magenta lines are valid 2 gate delays after the red; so they are valid 5 gate delays after the start.
  4. The brown lines are valid 4 gate delays after the magenta (2 gate delays to calculate the c's—note lower case, then two more for the Ss); so they are valid 9 gate delays after the start

Since the magenta lines flow right to left (then down), I drew their arrowheads. I typically do not draw arrowheads for lines that go to the right, go down, or go right and down.

Note that all the Ci's are calculated at once. For example, it is not the case that C3 depends on C2.

Start Lecture #7

Enhancing the CL Block

We are not done with the CL Block since our goal is to construct CLAs for any power-of-4 number of bits using this CL Block. Specifically, again assuming 5-input gates, we want the exact same CL Block to be used for a 4-bit (1-level) CLA; a 16-bit (2-level) CLA; a 64-bit (3-level) CLA; a 256-bit (4-level) CLA, etc.

In fact, we will go back further and construct a 1-bit (0-level) CLA, from which the 4-bit (1-level) CLA is built.

Moreover, when going from an 4n-bit (n-level) CLA to a 4n+1-bit (n+1-level) CLA, there should be no new logic that is needed. Specifically, we want a 64-bit (3-level) CLA to be composed of four 16-bit (2-level) CLAs, one additional CL Block (identical to those in the smaller constituent CLAs), some wires, and nothing else.

In the previous diagram we used a CL Block to assemble a 16-bit CLA from four 4-bit CLAs, but did not prepare for constructing a 64-bit CLA from four of these 16-bit CLAs. For that reason we did not have Pout and Gout (note that each 4-bit CLAs used did output a P and a G, which were used when constructing a 16-bit CLA).

In general, when constructing a CLA using the CL Block, there are actually three sizes of CLAs that are relevant (so far we have only dealt with two of the three).

  1. The previous size CLA, i.e., the size of the constituent CLAs (4-bit in the diagram above).
  2. The current size, i.e., the size being constructed (16-bit above).
  3. The next size, i.e., the size for which the CLA under construction will be a constituent. The diagram above did not support constructing the next size CLA (64-bit), a defect soon to be remedied.
cla clb

The full CL Block is drawn on the right and contains two outputs not shown or used previously, Pout and Gout.

This Block has the following 9 inputs.

It has the following 6 outputs

These outputs are calculated from the following, previously studied, formulas.

  C1 = Gin0 + PinO Cin
  C2 = Gin1 + Pin1 Gin0 + Pin1 Pin0 Cin
  C3 = Gin2 + Pin2 Gin1 + Pin2 Pin1 Gin0 + Pin2 Pin1 Pin0 Cin
  C4 = Gin3 + Pin3 Gin2 + Pin3 Pin2 Gin1 + Pin3 Pin2 Pin1 Gin0 + Pin3 Pin2 Pin1 Pin0 Cin
  Pout = Pin3 Pin2 Pin1 Pin0
  Gout = Gin3 + Pin3 Gin2 + Pin3 Pin2 Gin1 + Pin3 Pin2 Pin1 Gin0

Building CLAs Using the CL Block

It is now time to validate the claim that all (power of 4) sizes of PLAs can be built (recursively) using the CL Block.

1-bit CLA-PG

A 1-bit CLA is just a 1-bit adder. With only one bit there is no need for any lookahead since there is no ripple to try to avoid.

However, to enable us to build a 4-bit CLA from the 1-bit version, we actually need to build what we call a CLA-PG. The 1-bit CLA-PG has three inputs a, b, and cin. It produces 4 outputs s, cout, p, and g. We have given the logic formulas for all four outputs previously, but here they are again.

     s = a b cin + a b' cin' + a' b cin' + a' b' cin  odd number bits are 1
  cout = a b + a cin + b cin                          at least two bits are 1
     p = a + b
     g = a b
4-bit CLA-PG

A 4-bit CLA-PG is shown as the red portion in the figure to the right.

cla 4bit

It has nine inputs: 4 a's, b's, and cin and must produce seven outputs: 4 s's, cout, p, and g (recall that the last two were previously called the super propagate and super generate respectively).

The tall black box is our CL Block.

The question is, what must the ith ? box do in order for the entire (red) structure to be a 4-bit CLA-PG?.

So the ? box is just a 1-bit CLA-PG.
Wrong! The ? box is only a (large) subset of a 1-bit CLA-PG.
What is missing?
Ans. The ? box doesn't need to produce a carry out since the Cl-block produces all the carries.

cla 4bit pedantic

So, if we want to say that the 4-bit (1-level) CLA-PG is composed of four 1-bit (0-level) CLA-PGs together with a CL Block, we must draw the picture as on the right. The difference is that we explicitly show that the ? box produces cout, which is then not used.

This situation will occur for all sizes. For example, either picture on the right for a a 4-bit CLA-PG produces a carry out since all 4-bit full adders do so. However, a 16-bit CLA-PG, built from four of the 4-bit units and a CL Block, does not use the carry outs produced by the four 4-bit units.

We have several alternatives.

  1. Don't mention the problem of the unused cout. This is probably the most common solution, but too late for us since I already mentioned it.
  2. Draw the top version of the diagram (without the unused cout's) and delcare that a CLA-PG doesn't produce a carry out. It seems weird (to me, at least) that a CLA-PG doesn't have all the outputs of a full adder.
  3. Draw the top version of the diagram and admit that a level k CLA-PG doesn't really use four level k-1 CLA-PG's, but instead uses 4 copies of a large subset of a level k CLA-PG. I don't like this since it means you need to reverify the large subset.
  4. Draw the bottom version of the diagram. This seems reasonable to me.
  5. Draw the top version of the diagram, but view it as an abbreviation of the bottom version.

We choose the last alternative.

As another abbreviation, we will henceforth say CLA when we mean CLA-PG.

Remark: Hence the 4-bit CLA (meaning CLA-PG) is composed of

  1. Four 1-bit CLAs
  2. One CLA block
  3. Wires
  4. Nothing else
cla 16bit png
16-bit CLA-PG

Now take four of these 4-bit adders and use the identical CL Block to get a 16-bit adder.

The picture on the right shows one 4-bit adder (the red box) in detail. The other three 4-bit adders are just given schematically as small empty red boxes. The CL Block is also shown and is wired to all four 4-bit adders.

The complete (large) picture is shown here.

Remark: Hence the 16-bit CLA is composed of

  1. Four 4-bit CLAs
  2. One CLA block
  3. Wires
  4. Nothing else
64-bit CLA-PG

To construct a 64-bit CLA no new components are needed. That is, the only components needed have already been constructed. Specifically you need.

  1. Four magenta boxes, identical to the one just constructed.
  2. One additional CL Block, identical to the one just used to make the magenta box (which is also identical to the one used to make the red box.
  3. Wires to connect these five boxes.

Remark: Hence the 64-bit CLA (meaning CLA-PG) is composed of

  1. Four 16-bit CLAs
  2. One CLA block
  3. Wires
  4. Nothing else

When drawn (with a brown box) the 64-bit CLA-PG has 129 inputs (64+64+1) and 67 outputs (64+1+2).

256-bit CLA-PG
  1. Four brown boxes, identical to the one just constructed.
  2. One additional CL Block, identical to the one just used to make the brown box.
  3. Wires to connect these five boxes.

Remark: Hence the 256-bit CLA (meaning CLA-PG) is composed of

  1. Four 64-bit CLAs
  2. One CLA block
  3. Wires
  4. Nothing else
etc

Homework: How many gate delays are required for our 64-bit CLA-PG? How many gate delays are required for a 64-bit ripple carry adder (constructed from 1-bit full adders)?

A Logisim Illustration That All CLAs Look Alike

Start logisim and look at my circuits.

Remark: Lab 3 assigned; it is due 4 October 2011.

Summary

CLAs greatly speed up addition; the increase in speed grows with the size of the numbers to be added.

Remark: CLAs implement n-bit addition in O(log(n)) gate delays.

Shifters

MIPS (and most other) processors must execute shift (and rotate) instructions.

We could easily extend the ALU to do 1-bit shift/rotates (i.e., shift/rotate a 32-bit quantity by 1 bit), and then perform an n-bit shift/rotate as n 1-bit shift/rotates.

This is not done in practice. Instead a separate structure, called a barrel shifter is built outside the ALU.

Remark: Barrel shifters, like CLAs are of logarithmic complexity.

*** Big Change Coming ***

C.A Sequential Circuits, Memory, and State

Why do we need state?

C.7: Clocks

Assume you have a physical OR gate. Assume the two inputs are both zero for an hour. At time t one input becomes 1; the other one never changes. The output will oscillate for a while before settling on 1. We want to be sure we don't look at the answer before its ready.

This will require us to establish a clocking methodology, i.e., an approach to determining when data is valid.

First, however, we need some ...

Terminology

Micro-, Mega-, and Friends

Nano means one billionth, i.e., 10-9.
Micro means one millionth, i.e., 10-6.
Milli means one thousandth, i.e., 10-3.
Kilo means one thousand, i.e., 103.
Mega means one million, i.e., 106.
Giga means one billion, i.e., 109.

Frequency and period
clock

Consider the idealized waveform shown on the right. The horizontal axis is time and the vertical axis is (say) voltage.

If the waveform repeats itself indefinitely (as the one on the right does), it is called periodic.

The time required for one complete cycle, i.e., the time between two equivalent points in consecutive cycles, is called the period.

Since it is a time, period is measured in units such as seconds, days, nanoseconds, etc.

The rate at which cycles occur is called the frequency.

Since it is a rate, frequency is measured in units such as cycles per hour, cycles per second, kilocycles per micro-week, etc.

The modern (and less informative) name for cycles per second is Hertz, which is abbreviated Hz.

Prediction: At least one student will confuse frequency and periods on the midterm or final and hence mess up a gift question. Please, prove me wrong!.

Make absolutely sure you understand why

  1. A kilohertz clock is (a million times) faster than a millihertz clock.
  2. A clock with a kilosecond period is (a million times) slower than one with a millisecond period.
Edges

Look at the diagram above and note the rising edge and the falling edge.

We will use edge-triggered logic, which means that state changes (i.e., writes to memory) occur at a clock edge.

Each of our designs will either

The edge on which changes occur (either the rising or falling edge) is called the active edge. For us, choosing which edge is active is basically a coin flip.

In real designs the choice is governed by the technology used. Some designs permit both edges to be active. Examples include DDR (double data rate) memory and double-pumped register files. This permits a portion of the design to run at effectively twice the speed since state changes occur twice as often

Synchronous system

Now we are going to add state elements (memory) to the combinational circuits we have been using previously.

Remember that a combinational/combinatorial circuits has its outpus determined solely by its input, i.e. combinatorial circuits do not contain state.

sync-circuit

State elements include state (naturally).

bad-inverter

Combinatorial circuits can NOT contain loops. For example, imagine an inverter with its output connected to its input. So if the input is false, the output becomes true. But this output is wired to the input, which is now true. Thus the output becomes false, which is the new input. So the output becomes true ... .

Sequential circuits, however, can and often do contains loops.

sync-cycle

C.8: Memory Elements: Flip-Flops, Latches, and Registers

We will use only edge-triggered, clocked memory in our designs as they are the simplest memory to understand. So our current goal is to construct a 1-bit, edge-triggered, clocked memory cell. However, to get there we will proceed in three stages.

  1. We first show how to build unclocked memory.
  2. Then, using unclocked memory, we build level-sensitive clocked memory.
  3. Finally from level-sensitive clocked memory we build edge-triggered clocked memory.

Unclocked Memory

The only unclocked memory we will use is a so called S-R latch (S-R stands for Set-Reset).

When we define latch below to be a level-sensitive, clocked memory, we will see that the S-R latch is not really a latch.

s-r-latch

The circuit for an S-R latch is on the right, which is constructed from Cross-coupled nor gates.

Since it has two single-bit inputs, there are four possible input combinations.

  1. We never assert both S and R at the same time (the output is not defined in this case).
  2. When S is asserted (i.e., S=1 and R=0):
  3. When R is asserted:
  4. When neither one is asserted:

Clocked Memory: Flip-flops and latches

For both flip-flops and latches the output equals the value stored in the structure. Both have an input and an output (and the complemented output) and a clock input as well. The clock determines when the internal value is set to the current input. For a latch, the output can change whenever the clock is asserted (level sensitive). For a flip-flop, changes occur only at the active edge.

Unfortunately the terminology used is not perfect, the S-R latch defined above is unclocked memory.

D latch

The D stands for data.

d-latch

Note the following properties of the D latch circuit shown on the right.

d-latch-abbrev

The lower abbreviated diagram is how a D-latch is normally drawn.

A D latch is sometimes called a transparent latch since, whenever the clock is high, the output equals the input (the input passes right through the latch).

We won't use D latches in our designs, except right now to construct our workhorse, the master-slave flip-flop, an edge-triggered memory cell.

d-latch\operation
Traces

Note the following points illustrated by the traces to the right. We assume the stored value was initially low.

D or Master-Slave Flip-flop

This structure has been our goal. It is an edge-triggered, clocked memory. It is often referred to as a D-flop. Again the D stands for data.

d-flop

The circuit for a D flop is on the right has the following properties.

A D flop is sometimes called a master-slave flip-flop, with the left latch called the master and the right the slave.

Note that the substructures reuse the same letters as the main structure but have different meaning (similar to block structured languages in the algol style).

The left D latch is set during the time the clock is asserted. Remember that the latch is transparent, i.e. it follows its input when its clock is asserted. But the right latch is ignoring its input at this time. When the clock falls, the 2nd latch pays attention and the first latch keeps producing whatever D was at fall-time.

Actually D must remain constant for some time around the active edge.

Start Lecture #8

Remark: Forgot to mention shifters last time; do it now.

Homework: Move the inverter to the other latch. What has changed?

d-flop-operation
Traces

The picture on the right is for a master-slave flip-flop. As before we are assuming the output is initially low.

Note how much less wiggly the output is in this picture than before with the transparent latch.

Setup and Hold Times
setup-hold

The next picture shows the setup and hold times discussed above.

Homework: Which code better describes a flip-flop and which a latch?

  repeat {
     while (clock is low) {do nothing}
     Q=D
     while (clock is high) {do nothing}
  } until forever

or

  repeat {
     while (clock is high) {Q=D}
  } until forever
Flip-Flops in Verilog

Registers

A register is basically just an array of D flip-flops. For example a 32-bit register is an array of 32 D flops.

What if we don't want to write the register during a particular cycle (i.e. at the active edge of a particular cycle)?

As shown in the diagram on the right, we introduce another input, the write line, which is used to gate the clock.

register-orig

If the write line is high forever, the clock input to the register is passed right along to the D flop and hence the input to the register is stored in the D flop when the active edge occurs, which for us is the falling edge. That is, the register is written every cycle.

If the write line is low forever, the clock to the D flop is always low so has no edges. Thus the register is never written.

Now that we understand what happens if the write line is constant, either always high or always low, we must ask what happens if we change the write line from high to low or vice versa.

We do not change the write line when the (external) clock is high since that would cause extra edges to be passed to the D-flop. Instead we change the write line only when the clock is low.

This, however, is not so good!

register-active-low

Essentially the same affect can be achieved in another manner as well. Instead of having the register negate the write line, the register specifies that the input be a don't write line, i.e. W'. Such a register, which is depicted on the right is often call an active low register since it is active when its W' input is low.

register-3bit

Multibit Registers

To implement a multibit register, just use multiple D flops.


Register Files

register-file

A register file is just a set of registers, each one numbered.

read-ports

Reading From a Register File

To support reading a register we just need a (big) mux from the register file to select the correct register.

Writing a Register in a Register File

write-port

To support writing a register we use a decoder on the register number to determine which register to write. Note that errors in the book's figure were fixed.

  1. The decoder is log n to n (5 to 32 for MIPS).
  2. The decoder outputs are numbered 0 to n-1 (NOT n).

Note also that I show the clock explicitly.

Recall that the inputs to a register are W, the write line, D the data to write (if the write line is asserted), and the clock. We should perform a write to register r this cycle if the write line is asserted and the register number specified is r. The idea is to gate the write line with the output of the decoder.

Homework: C.36

Specifying Sequential Logic in Verilog

sram

C.9 Memory Elements: SRAMS and DRAMS

Note: There are other kinds of flip-flops T, J-K. Also one could learn about excitation tables for each. We will not cover this material (P&H doesn't either). If interested, see Mano.

C.10: Finite State Machines (FSMs)

More precisely, we are learning about deterministic finite state machines or deterministic finite automata (DSA). The alternative nondeterministic finite automata (NDA) are somewhat strange and, althought seemingly nonrealistic and of theoretical value only, form, together with DFAs, what I call the secret weapon used in the first stage of a compiler (the lexical analyzer).

We will do a different example from the one in the book (counters instead of traffic lights). The ideas are the same and the two generic pictures (below) apply to both examples.

state machine FSM

Counters

counter-1bit

A counter counts (naturally).

counter-1bit-states
The State Transition Diagram
The circuit diagram.
counter-1bit-circuit
Determining the combinatorial circuit
Truth Table for the Combinatorial Circuit
Current(Next A)
AIRDA


0000
1001
0101
1100
xx10

How do we determine the combinatorial circuit?

Start Lecture #9

A 2-bit Counter.

No new ideas are needed; just more work.

counter-2bit-circuit counter-2bit-states
Beginning of the Truth Table for a 2-bit Counter
Current Next
HLIR DHDL


xxx 100
001 001

To determine the combinatorial circuit we could precede as before. The beginning of the truth table is on the right.

This would work (do a few more rows on the board), but we can instead think about how a counter works and see that.

  DL = R'(L ⊕ I)
  DH = R'(H ⊕ LI)
counter-2bit

On the right is (a diagram depicting) the Logisim circuit for the 2-bit counter. The two 1-bit registers are on the right, the clock is near the middle and the combinatorial circuit is most of the left part. There are two 1-bit inputs, namely I and R.

If you want to play with this circuit the .circ file can be downloaded here.

A 3-bit Counter

Homework: C.39

C.11 Timing Methodologies

C.12 Field Programmable Devices

Simulating Combinatorial Circuits at the Gate Level

The idea is, given a circuit diagram, write a program that behaves the way the circuit does. This means more than getting the same answer. The program is to work the way the circuit does.

For each logic box, you write a procedure with the following properties.

Simulating a Full Adder

Remember that a full adder has three inputs and two outputs. Discuss FullAdder.c or perhaps FullAdder.java.

Simulating a 4-bit Adder

This implementation uses the full adder code above. Discuss FourBitAdder.c or perhaps FourBitAdder.java

Concluding Remarks

Read.

Chapter 1: Computer Abstractions and Technologies

1.1 Introduction

Read.

1.2 Below Your Program

Read.

1.3 Under The Covers

Read.

Homework: 1.1, a multi-part matching question.

1.4 Performance

Will be done later.

1.5 The Power Wall

1.6 The Sea Change: The Switch from Uniprocessors to Multiprocessors

1.7 Real Stuff: Manufacturing and Benchmarking the AMD Operton X4

Some of the benchmarking material will be done later.

1.8 Fallacies and Pitfalls

Some of the benchmarking material will be done later.

1.9 Concluding Remarks

Read this short section.

1.10 Historical Perspective and Further Reading

Chapter 2: Instructions: Language of the Machine

2.1 Introduction

Homework: Read section 2.1.

2.2 Operations of the Computer Hardware

Homework: Read section 2.2. For this course you do not have to worry much about how a program in C is translated into assembly language. However, it is an important concept (at least at the high level). It is touched upon in 201 (sometimes) and I believe will be covered more in future versions of that course.

Remark: The of chapter 2 contains much material on this same topic, which we are de-emphasizing. The book is quite well written and you may well find that material interesting. Were this a 2-semester course, we would certainly cover it.

2.3 Operands in the Computer Hardware

Registers

Many of the MIPS instructions operate of values stored in registers. The MIPS architecture we shall study has thirty-two 32-bit registers. There is another MIPS architecture that has thirty-two 64-bit registers.

A very serious task for a compiler is to make efficient use of this scarce resource; however, we shall not discuss that problem.

Register Names

The text, which emphasizes the correspondence between a C or Java program and assembly language much more that we shall, is very careful in distinguishing between those registers used for C-program variables, those used for temporary values, those used when one function calls another, those used for other purposes.

The actual hardware makes no such distinction: In machine instructions a register operand is simply a 5-bit number (from 0 to 31). The distinction between register types is just a convention used by software.

I shall try to follow the conventions, but remember that from a hardware perspective, a register is named simply by a 5-bit value.

Memory Operands

Of course computers can contain many more than thirty-two 32-bit values. Indeed, today even a modest laptop has a central memory at least ten million times larger.

In MIPS arithmetic is performed only on values located in registers. Thus, in addition to arithmetic instructions, MIPS (and essentially all other computers) need data transfer instructions to fetch values from memory to registers and to update memory with newly calculated values.

Constant or Immediate Operands

Often one operand in an arithmetic instruction is a constant, not a variable. MIPS supplies corresponding immediate instructions. For example, add naturally adds the contents of two registers, placing the result in a third; whereas addi (add immediate) adds a constant (contained in the instruction) to one register, placing the result in a second register.

2.4: Signed and Unsigned Numbers

MIPS uses 2s complement representation for signed numbers (as do all modern processors).

To form the 2s complement (of 0000 1111 0000 1010 0000 0000 1111 1100)

Unsigned Numbers

MIPS (like most computers) can also process 32-bit values as unsigned numbers, that is the hob is not a sign bit. It is instead the bit in the 232 place

As mentioned previously, addition/subtraction on signed numbers does not treat the sign bit specially so unsigned and signed addition/subtraction give the same answer if the operands are the same bit strings.

The differences are in overflow and comparisons.

Comments on Two's Complement

You could easily ask what does this funny notation have to do with negative numbers. Let me make a few comments.

  1. What does -1 mean?
  2. It is the unique number that when added to 1 gives zero.
  3. Our representation of -1 does do this (using regular binary addition and discarding the final carry-out) so we do have -1 correct.
  4. What does negative n mean?
    Ans: It is the unique number that, when added to n, gives zero.
  5. The 1s complement of n when added to n gives all 1s, which is -1.
  6. Thus the 2s complement, which is one larger, will give zero, as desired.

Homework: 2.7.1 and 2.7.2.

Homework: 2.7.4 and 2.7.6 (but change hexadecimal to binary.

2.5 Representing Instructions in the Computer (MIPS)

Hexadecimal (i.e., Base 16) Numbers

Converting base 2 values to/from base 10 is work, but converting base 2 to/from base 16 is easy. You simply group the base 2 number into groups of 4 or expand the base 16 number from right to left. The one question is how do you write the 6 digits past 9 and the answer is A, B, C, D, E, and F.

Do some examples on the board.

The Register File

We just learned how to build this structure. We need 2 read ports and 1 write port since MIPS instructions can read up to 2 register and write up to 1 register

Recall that, by software convention, registers are given two character names. The first character is a letter and indicates what the software conventionally uses the register for. For example $t2 is the third ($t0 is the first) register used for temporary values.

As stated above, MIPS has thirty-two 32-bit register. Some machines, notably the 32-bit Intel (PC) architecture, have a number of register classes, where only certain registers can be used for certain task. However, the MIPS treats registers 1-31 the same, only register 0 is special.

MIPS Fields

The fields of a MIPS instruction are fairly consistent. There are just a few classes of instruction formats and within each class the various bit positions of the instructions are used in the same way..

R-type Instructions (R for Register)

These instructions have three operands, each is a register number. All R-type instructions have the following fixed format.

  op    rs    rt    rd    shamt  funct   name of field
   6     5     5     5      5      6     number of bits

These fields are used consistently in R-type instructions.

Add and Subtract   op rd,rs,rt

Examples: add/sub $t1,$t2,$t3 (i.e., add/sub $9,$10,$11).

I-type (Immediate)

The I is for immediate.

Load Word and Store Word   op rt, immediate(rs)

Examples: lw/sw $t1,1000($s3) (i.e., lw/sw $9,1000($19))

RISC-like properties of the MIPS architecture.

Add Immediate  op rt, rs, immediate

Example: addi $t1,$t2,100 (i.e., addi $9,$10,100)

Homework: 2.6.5 and 2.6.6.

Homework: 2.10.1, 2.10.2, and 2.10.5. (officially assigned next class)

2.6: Logical Operations

These instructions deal with the bits within the word rather than treating the word as a unit. Such instructions are often called logical, presumably because they are often used if the logic of programs (conditionals and loops).

Shift Left/Right Logical   op rd, rt, shamt

Examples: sll/srl $t7,$t2,7 (i.e., sll/srl $15,$10,7)

Bitwise AND, OR, and NOR (includes NOT) With Immediates

Bitwise AND, OR, and NOR   op rd, rs, rt

Examples: and/or/nor $s2,$s1,$s0 (i.e., and/or/nor $18,$17,$16)

Bitwise AND and OR Immediate   op rt, rs, immediate

Examples andi/ori $s0,$t0,31 (i.e., andi/ori 16,8,31)

Homework: 2.13.1 and 2.14.1 (officially assigned next class)

2.7 Instructions for Making Decisions

Branch (Not) Equal op rt, rs, immediate

Examples: beq/bne $t1,$t2,123 (i.e., beq/bne $9,$10,123)

Start Lecture #10

Set Less Than     op rd, rs, rt

This is an old friend. Recall the extra effort we put into the alu a few weeks ago to implement this important MIPS instruction.

Example: slt $t1,$t2,$t3 (i.e., slt $9,$10,$11)

Set less Then Immediate   op rt, rs, immediate

Example: slti s1,s2,20000 (i.e., slti 17,18,20000)

Unsigned Comparisons

Recall that comparison is different for unsigned and signed numbers. For example: signed values with 1 in the hob are less than those with 0 in the hob (the first value is negative); but, if the values are unsigned, a 1 in the hob is greater than a 0 in the hob. For this reason, MIPS has in addition unsigned versions of slt and slti, that use the unsigned definition of less than.

The instructions are named sltu and sltiu as you would expect. Our MIPS subset implementation, will not include them.

Branch If Less Than

Example: blt $t5,$t7,123 (i.e., blt $13,$15,123)

Branch If Less Than Or Equal

Example: ble $t5,$t7,L (L a label to be calculated by the assembler.)

Branch If Greater Than

Example bgt $t5,$t7,L

Branch If Greater Than Or Equal

Example: bge $t5,$t7,L

Note: Please do not make the mistake of thinking that
    stl $t1,$t5,$t7
    beq $t1,$zero,L
is the same as
    stl $t1,$t7,$t5
    bne $t1,$zero,L

It is not true that the negation of X<Y is Y>X.
End of Note

J-type instructions (J for jump)

These have a different format, but again the opcode is the first 6 bits.

    op   address
    6    26

The effect is to jump to the specified (immediate) address. Note that there are no registers specified in this instruction and that the target address is not relative to (i.e. added to) the address of the current instruction as was done with branches.

Jump   op address

Example: j 10000

But MIPS is a 32-bit machine with 32-bit address and we have specified only 26 bits. What about the other 6 bits?

In detail the address of the next instruction is calculated via a multi-step process.

  1. The 26 bit address field is extracted from the instruction.
  2. This address is left shifted two bits. The result is a 28-bit address (call it A) that is always a multiple of 4, which makes sense since all instructions must begin on a multiple of 4 bytes.
  3. The high order 4 bits are extracted from the address of the current instruction (not the address in the current instruction). Call this 4-bit quantity B.
  4. The address of the next instruction is formed by concatenating B with A.

Homework: 2.16.1.

2.8 Supporting Procedures in Computer Hardware

Jump And Link   op address

Example: jal 40000

Jump Register   op rs

Important example: jr $ra (i.e., jr $31)

2.9 Communicating with People

2.10 MIPS Addressing for 32-bit Immediates and Addresses

How can we put a 32-bit value (say 2 billion) into register 6?

  1. Zero and add.
  2. Load the word
  3. Load shift add
  4. Load shift OR

Load Upper Immediate   op rt, immediate

Example: lui $t4,123 (i.e., lui $12,123)

Chapter 3: Arithmetic for Computers

Homework: Read 3.1-3-4

3.1 Introduction

I have nothing to add.

3.2 Addition and Subtraction

Recall that MIPS uses 2s complement (just like the intel chips)

To form the 2s complement (of 0000 1111 0000 1010 0000 0000 1111 1100)

To add two (signed) numbers just add them. That is, don't treat the sign bit special.

To subtract A-B, just take the 2s complement of B (forming -B) and add.

Overflows

An overflow occurs when the result of an operation cannot be represented with the available hardware. For MIPS this means when the result does not fit in a 32-bit word.

Recall that the operands each have 31 data bits and a sign bit; thus the result would definitely fit in 33 bits (32 data plus 1 sign).

  11111111111111111111111111111111   (32 ones is -1)
+ 11111111111111111111111111111111
----------------------------------
 111111111111111111111111111111110   Discard the carry out
  11111111111111111111111111111110   this is -2 as desired

As shown on the right the hardware simply discards the carry out of the high order (i.e., sign) bit, which might seem hopelessly naive, but is normally correct.

The bottom 31 bits are always correct.
Overflow occurs when the 32nd (sign) bit is set to a value and not the sign.

An overflow occurs in the following cases

  Operation  Operand A  Operand B  Result
  A+B         ≥ 0          ≥ 0       < 0
  A+B         < 0          < 0       ≥ 0
  A-B         ≥ 0          < 0       < 0
  A-B         < 0          ≥ 0       ≥ 0

These conditions are the same as
Carry-In to sign position != Carry-Out from sign position.

Homework: Prove this last statement (for fun only, do not hand in).

addu, subu, addiu

Since unsigned numbers are often used for address arithmetic where overflows should be ignored, these three instructions perform addition and subtraction the same way as do add and sub, but do not signal overflow.

shifter-trivial

Shifter

This is a sequential circuit.

It is just a string of D-flops; the output of one is input of the next.

shifter

We want more.

  1. Left and right shifting (with serial input/output).
  2. Parallel load.
  3. Parallel Output.
  4. Don't shift every cycle.

Parallel output is just wires.

Shifter has 4 modes (nop, left-shift, right-shift, load) so

We could modify our registers to be shifters (bigger mux), but ...
Our shifters are slow for big shifts; barrel shifters are faster and kept separate from the processor registers.

Homework: A 4-bit shift register initially contains 1101. It is shifted six times to the right with the serial input being 101101. What is the contents of the register after each shift.

Homework: Same register, same initial condition. For the first 6 cycles the opcodes are left, left, right, nop, left, right and the serial input is 101101. The next cycle the register is loaded (in parallel) with 1011. The final 6 cycles are the same as the first 6. What is the contents of the register after each cycle?

3.3 Multiplication

Of course we can do this with two levels of logic since multiplication is just a function of its inputs.

But just as with addition, would have a very big circuit and large fan in. Instead we use a sequential circuit that mimics the algorithm we all learned in grade school.

Recall how to do multiplication.

The First Attempt

Our first solution multiplies in essentially the same way.

We are doing binary arithmetic so each digit of the multiplier is 1 or zero. Hence multiplying the mulitplicand by a digit of the multiplier results in either

multiplier-1

Use an if appropriate bit of multiplier is 1 test. To get the appropriate bit:

Putting in the correct column means putting it one column further left than the last time. This is done by shifting the multiplicand left one bit each time (even if the multiplier bit is zero).

Instead of adding partial products at end, we keep a running sum.

This results in the following algorithm

    product ← 0
    for i = 0 to 31
        if LOB of multiplier = 1
            product = product + multiplicand
        shift multiplicand left 1 bit
        shift multiplier right 1 bit

Start Lecture #11

What about the control?

This works!

It clearly works if we test the LOB and write the product on one cycle and shift the next cycle (so two cycles per bit). With some more care you can do it all in one cycle, you just need to be sure you add the multiplicand before it is shifted and that you get the LOB before the multiplier is shifted.

The real weakness of the above solution, when compared to the improved versions to come, is that the first attempt is wasteful of resourses and hence is:

All these are bad.

Do on the board 4-bit multiplication (8-bit registers) 1100 x 1101. Since the result has (up to) 8 bits, this is often called a 4x4→8 multiply.

The diagrams are for a 32x32→64 multiplier.

An Improved Circuit

The product register must be 64 bits since the product can contain 64 bits.

Why is multiplicand register 64 bits?
Ans: So that we can shift it left, i.e., for our convenience.

By this I mean it is not required by the problem specification, but only by the solution method chosen.

Why is ALU 64-bits?

POOF!! ... as the smoke clears we see an idea.

multiplier-2

We can solve both problems at once.

Don't shift the multiplicand left

Instead shift the product right!

Add the high-order (HO) 32-bits of product register to the multiplicand and place the result back into HO 32-bits

  product <- 0
  for i = 0 to 31
      if LOB of multiplier = 1
         (serial_in, product[32-63]) <- product[32-63] + multiplicand
      shift product right 1 bit
      shift multiplier right 1 bit

What about control?

Redo the same example on the board.

multiplier-3
A final trick (gate bumming, like code bumming of 60s)

There is a still waste of registers, i.e. they are not fully utilized.

Timeshare the LO half of the product register.

The algorithm changes to:

    product[0-31] ← multiplier
    for i = 0 to 31
        if LOB of product = 1
            (serial_in, product[32-63]) ← product[32-63] + multiplicand
        shift product right 1 bit

Control again boring.

Redo the same example on the board.

Signed Multiplication

The above was for unsigned 32-bit multiplication. What about signed multiplication?

  1. Save the signs of the multiplier and multiplicand.
  2. Convert multiplier and multiplicand to non-negative numbers.
  3. Use above algorithm on the non-negative numbers.
  4. Only use 31 steps not 32 since there are only 31 multiplier bits (the HOB of the multiplier is the sign bit, not a bit used for multiplying).
  5. Compliment the final product if original signs were different.
Faster Multipliers

There are (asymptotically) faster multipliers, but we are not covering them.

3.4 Division

3.5 Floating Point

3.6 Parallelism and Computer Arithmetic: Associativity

3.7 Real Stuff: Floating Point in the IA-32

3.8 Fallacies and Pitfalls

Read for pleasure.

3.9 Concluding Remarks

Read for pleasure.

3.10 Historical Perspectives

Read for pleasure (located on CD).

Remark: End of material on midterm. I posted a practice midterm. Today is lecture 11. Exam is lecture 13, one week from today. Remind me anytime saturday, via email to the list, to post solutions. Advice: do NOT look at the answers until you have done the questions.

Chapter 4 The Processor

Homework: Start Reading Chapter 4.

4.1: Introduction

data-ath-abstract

We are going to build a basic MIPS processor.

Figure 4.1 redrawn below shows the main idea.

Note that the diagram shows the instruction including three register numbers, an immediate value to be added to a register, and an immediate value to be added to the PC.

No single instruction has all those components, but our datapath must include pathways for all possibilities. Eventually, we will add muxes to choose which possibilities are relevant for the given instruction.

We shall see how we arrange for only certain datapaths to be used for each instruction type.

Why are we doing arithmetic on the program counter?

4.2 Logic Design Convention

Done in appendix C.

4.3 Building a Datapath

Let's begin doing the pieces in more detail.

We draw buses in magenta (mostly 32 bits) and control lines in green.

Instruction Fetch

datapath-ifetch

We are ignoring branches and jumps for now.

The diagram on the right shows the main loop involving instruction fetch (i-fetch)

datapath-R-type

R-type instructions

We did the register file in appendix C. Recall the following points made when discussing the appendix.

The 32-bit bus with the instruction is divided into three 5-bit buses, one for each register number (plus other wires not shown).

Homework: What would happen if the RegWrite line had a stuck-at-0 fault (was always deasserted)? What would happen if the RegWrite line had a stuck-at-1 fault (was always asserted)?

Start Lecture #12

Remarks

For this week and next there are departmental meetings on thurs that will cut into my office hours.

For this week I will have an extra office hour on wednesday from 12:15-4:45. Architecture has priority, due to our midterm.

For next week I will have an extra office hour on monday from 12:15-4:45. OS has priority, due to their midterm.

End of Remarks

Notation for Register Numbers

In this chapter we are interested in building the processor, and not as interested in seeing how Java or C statements could be translated into machine instructions. As a result, I will refer to the registers in an instruction by their hardware names

datapath-load-store

Load and Store

The diagram on the right shows the structures used to implement load word and store word (lw and sw).

lw rt,disp(rs):

  1. Computes the effective address formed by adding the 16-bit immediate constant disp (displacement) to the contents of register rs.
  2. Fetches the value in data memory at this address.
  3. Writes this value into register rt.

sw rt,disp(rs):

  1. Computes the same effective address as lw rt,disp(rs).
  2. Stores the contents of register rt into this address.
Sign Extension
sign-extend

We have a 32-bit adder and more importantly have a 32-bit addend coming from the register file. Hence we need to extend the 16-bit immediate constant to 32 bits. That is we must replicate the HOB of the 16-bit immediate constant to produce an additional 16 HOBs all equal to the sign bit of the 16-bit immediate constant. This is called sign extending the constant.

Note that the Sign Extend oval consists of just wires, no gates at all.

On the right is a small example, a 4→8 sign extender.

Control Lines

What about the control lines?

Homework: What would happen if the RegWrite line had a stuck-at-0 fault (was always deasserted)?
What would happen if the RegWrite line had a stuck-at-1 fault (was always asserted)?
What would happen if the MemWrite line had a stuck-at-0 fault?
What would happen if the MemWrite line had a stuck-at-1 fault?

The Diagram is Wrong (Specifically, Incomplete)

The diagram cheats a little for clarity.

Branch on Equal (beq)

Compare two registers and branch if equal. The circuit on the right computes two values, the branch target address and a Boolean specifying whether or not to branch. Note the familiar pattern.

datapath-beq

Remember that this diagram is just for beq. If the instruction is not beq then the Equal line from the ALU is not relevant. This will be fixed up later when we do all the control.

Recall the following from appendix C, where we built the ALU, and from chapter 2, where we discussed beq.

To check if two registers equal we subtract one from the other and test the result for zero (our ALU subtracts if ALU Operation says to, and our ALU always checks if the result is 0). In this case we are not interested in the result itself (so we don't wire that output to anything), just whether it is zero.

The target of the branch on equal instruction
    beq rs,rt,disp
is the sum of

  1. The program counter PC after it has been incremented, that is, the address of the next sequential instruction. This address is always calculate by the ifetch unit above.
  2. The 16-bit immediate constant disp (treated as a signed number) left shifted 2 bits. The constant represents (32-bit) words and the address is specified in (8-bit) bytes). Since there are 4 bytes per word, we must multiple the word address by 4, which can be accomplished by a left shift of 2.
shift-left-2

The shift left 2 is not a shifter. It simply moves wires and includes two zero wires. We need a 32-bit version of the 5 bit version shown on the right.

Since the immediate constant is signed it must be sign extended. As mentioned and drawn previously this is just replicating the HOB.

The top alu labeled add is just an adder so does not need any control.

Homework: What would happen if the RegWrite line had a stuck-at-0 fault? What would happen if the RegWrite line had a stuck-at-1 fault?

Creating a Single Datapath

We will first put the pieces together in a way that the resulting single datapath is able to execute all of the above instructions (several R-type instructions including set-less-than, load and store word, and branch on equal).

This will require several multiplexors and their associated select lines. After we have the pieces assembled into a unified whole, we will discuss how to calculate the select lines (and other control lines).

We are not now worried about speed.

We are assuming that the instruction memory and data memory are separate. So we are not permitting self modifying code. We are not showing how either memory is connected to the outside world (i.e., we are ignoring I/O).

We must use the same register file for all the instruction types since when a load changes a register, a subsequent R-type instruction must see the change and when an R-type instruction makes a change, the lw/sw must see it (for storing or calculating the effective address).

We could use separate ALUs for each type of instruction so that several instructions could proceed at the same time, but we are not worried about speed so we will use the same ALU for all instruction types. We do have a separate adder for incrementing the PC.

Combining R-type and lw/sw

The problem is that some inputs can come from different sources depending the instruction type. We need to add muxes as shown on the right.

datapath-R-type-lw-sw
  1. For R-type instructions, both ALU operands are registers. For I-type instructions (lw/sw) the second operand is the (sign extended) immediate field. This is handled by the mux with control line named ALUSrc. The name was presumably chosen to indicate that this control line determines one of the sources (i.e., inputs) for the ALU.
  2. For R-type instructions, the write data comes from the ALU. For lw it comes from the memory. See the control line MemToReg (i.e., should the value from the Memory go to the Register).
  3. For R-type instructions, the write register comes from field rd, which is bits 15-11. For sw, the write register comes from field rt, which is bits 20-16. The diagram does not deal with this issue. We shall do it later using (surprise) another mux.

Including Instruction Fetch

datapath-R-type-lw-sw-ifetch

Adding instruction fetch is quite easy.

We simply attach the instruction fetch block done above to the left of the previous diagram.

The result is shown on the right, where the new material is in blue.

Not shown yet is how the 32-bit instruction leaving the instruction memory is divided into into the various 5-bit and 15-bit fields. This is not trivial since it is not true that the same bits always go to the same field.

Including beq

datapath-full

We need to have an if stmt for updating the PC corresponding to the two possiblities: the branch is taken and the branch is not taken.

This conditional assignment to the PC should be compared to the conditional expressions found in C and Java, for example

  y = (c==4) ? x : z;

As usual, in logic design the conditional assignment is done with a mux (and a control line, named PCSrc—what is the input to the PC register).

Homework: Extend the datapath just constructed to support the addi instruction as well as the instructions already supported.

Homework: Extend the datapath just constructed to support an R-type instruction that is a variation of the lw instruction where the memory address is computed by adding the contents of two registers (instead of using an immediate field) and the contents of that memory location is loaded into the third register. Continue to support all the instructions that the original datapath supported.

Homework: Can you support a hypothetical swap instruction that swaps the contents of two registers using the same building blocks that we have used to date?

Start Lecture #13

Midterm Exam

Start Lecture #14

Remarks on the Midterm Exam

  1. Median was 85.
  2. CLA question was supposed to be 10 points, but was listed as 5 on the Cream exam. I graded it out of 5 and gave everyone 5 additional points.
  3. When I said implement blt and ble, I meant for you to write the instructions. Since I didn't say so explicitly, I gave credit for an English explanation.

Review the Exam

4.4: A Simple Implementation Scheme

What is Left to Do (for our Datapath)?

There are basically two tasks remaining. We shall see they are related; the key is the instruction itself.

1: Dividing the Instruction

The diagram above has a 32-bit instruction magically dividing into various fields, three of 5-bits and one of 16 bits. Moreover we know that not all fields are relevant for all instructions: I-type instructions do not have a third register and R-type instructions do not have a 16-bit immediate field.

In addition, register rt is sometimes a read register and sometimes a write register.

We must figure which bits of the 32-bit instructions should go to each of the various fields in all possible circumstances.

2: Setting the Muxes and Other Control Lines

We have ignored the control signals. Each of our muxes has a 1-bit control line that appears to be created out of thin air. We need to determine the values of each of these lines for all cases. Similarly, our ALU takes a 4-bit control line, but we have not determined how to calculate those four bits.

Dividing the Instruction

The diagram below shows (in blue as usual) the additions needed to divide the instruction. One cost of this solution is yet another mux, with yet another to-be-calculated 1-bit control line (having yet another slightly cryptic name RegDst, meaning this line determines whether Register rt or rd should be the Destination register).

Also added is an unspecified logic block ALU Control (abbreviated in the diagram as ALU Cntl to save space), with an unspecified 2-bit control line ALUOp as input. The new control lines will be determined in the next section entitled The Control for the Datapath, and the new block will calculate the 4-bit ALU Operation from the new control lines and the funct bits of the instruction.

datapath-final

We write I:n-m to represent instruction bits n through m (inclusive). For example I:15-0 represents the low order 16 bits of the instruction, which we recall is the immediate field in an I-type instruction.

Bits I:31-26, the opcode, have not been used up to this point. We shall see that the opcode, will play a prominent role when we determine the control lines.

The Control for the Datapath

  AND     0 0 00
  OR      0 0 01
  Add     0 0 10
  Sub     0 1 10
  Set-LT  0 1 11
  NOR     1 1 00

Remark: Much of the material in this section is found in Appendix D section 2.

Now that we have added the one missing mux and shown how the instruction bits are divided, two related tasks remain.

  1. We need to generate the control lines used to set the muxes.
  2. We need to generate the four ALU control lines: 1-bit Anegate, 1-bit Bnegate and 2-bit OP. The appropriate values for each of our ALU instructions is shown in the table on the right.
    Don't forget that we use the ALU for instructions such as lw/sw as well.

Homework: What happens if we use 0 1 00 for the four ALU control lines?
What if we use 0 1 01?

What information can we use to decide on the muxes and alu control lines?

The instruction!

What must we calculate?

No problem, just do a truth table.

Start Lecture #15

A Two-Stage Approach

  ALUOp   Action needed by ALU

  00      Addition (for load and store)
  01      Subtraction (for beq)
  10      Determined by funct field (R-type instruction)
  11      Not used

We will let the main control (to be done later) summarize the opcode for us. From this summary and the 6-bit funct field, we shall determine the control lines for the ALU. Specifically, the main control will summarize the opcode as the 2-bit field ALUOp, whose meaning is shown on the right

Controlling the ALU Given the Summary

How many entries do we have now in the truth table?

Some Simplifications
opcodeALUOpoperation functALU actionALU cntl
LW00load word xxxxxxadd0010
SW00store word xxxxxxadd0010
BEQ01branch equal xxxxxxsubtract0110
R-type10add 100000add0010
R-type10subtract 100010subtract0110
R-type10AND 100100and0000
R-type10OR 100101or0001
R-type10SLT 101010set on less than0111
  1. We will not be doing NOR so
    1. We have only 8 MIPS instructions that use the ALU as shown in the table on the right.
    2. The left bit of ALU cntl (Anegate) is always 0 and hence is not shown in future tables.
  2. The first two rows of the table above are the same. Each says that when ALUOp=00, ALU cntl is 0010. Hence need need only one line.
  3. When funct is used, its two HOBs are always 10 so are don't care inputs.
  4. ALUOp=11 impossible and hence
    1. 01 = x1
    2. 10 = 1x
  ALUOp | Funct        ||  Bnegate:OP
  1 0   | 5 4 3 2 1 0  ||  B OP
  ------+--------------++------------
  0 0   | x x x x x x  ||  0 10
  x 1   | x x x x x x  ||  1 10
  1 x   | x x 0 0 0 0  ||  0 10
  1 x   | x x 0 0 1 0  ||  1 10
  1 x   | x x 0 1 0 0  ||  0 00
  1 x   | x x 0 1 0 1  ||  0 01
  1 x   | x x 1 0 1 0  ||  1 11

Applying these simplifications yields the truth table on the right

How should we implement this?
We will do it PLA style (disjunctive normal form, 2-levels of logic) for each of the three output bits separately.

Specifically, for each output, we will

  1. First find the minterms, i.e., the cases in which that output bit is asserted.
  2. Then produce a circuit for each miniterm (an AND of literals).
  3. Finally, construct an OR of these ANDs.

Only the first part requires any real work.

  ALUOp | Funct
  1 0   | 5 4 3 2 1 0
  ------+------------
  x 1   | x x x x x x
  1 x   | x x 0 0 1 0
  1 x   | x x 1 0 1 0
BNegate

When is Bnegate (called Op2 in book) asserted?
Ans: Those rows in the table above where its bit (the leftmost output bit) is 1. That is, rows 2, 4, and 7.

We show those three rows on the right.

  ALUOp | Funct
  1 0   | 5 4 3 2 1 0
  ------+-------------
  x 1   | x x x x x x
  1 x   | x x x x 1 x

Looking again at the full (7-row) table, we notice that, in the 5 rows with ALUOp=1x, F1=1 is enough to distinugish the two rows where Bnegate is asserted. This gives the last table for BNegate, again shown on the right.

Hence Bnegate is simply ALUOp0 + (ALUOp1 · F1).

  ALUOp | Funct
  1 0   | 5 4 3 2 1 0
  ------+------------
  1 x   | x x 0 1 0 1
  1 x   | x x 1 0 1 0
Op0

Now we apply the same technique to determine when OP0 is asserted and begin by listing on the right the rows in the full table where its bit (the rightmost output bit) is set.

As with BNegate, we look back at the full table and study all the rows where ALUOp=1x, and, within that group of rows, those rows where OP0 is asserted (the last two rows).

  ALUOp | Funct
  1 0   | 5 4 3 2 1 0
  ------+------------
  1 x   | x x x x x 1
  1 x   | x x 1 x x x

We see that the rows where OP0 is asserted are characterized by just two Function bits (3 and 0), which reduces the table to that on the right.

Hence OP0 is ALUOp1 · F0 + ALUOp1 · F3

  ALUOp | Funct
  1 0   | 5 4 3 2 1 0
  ------+------------
  0 0   | x x x x x x
  x 1   | x x x x x x
  1 x   | x x 0 0 0 0
  1 x   | x x 0 0 1 0
  1 x   | x x 1 0 1 0
Op1

Finally, we determine when is OP1 asserted using the same technique. However, we shall see that this bit is more inspiring than the first two. Once again the procedure begins by listing on the right those rows where the relevant bit (the middle output bit) is one.

Right away we get a hint that we have more work to do as five rows pop up.

As before we study the 5 rows in the original 7-row table that have ALUOp=1x, and, within that group, those rows where OP1 is asserted (rows 3, 4, and 7).

  ALUOp | Funct
  1 0   | 5 4 3 2 1 0
  ------+------------
  0 0   | x x x x x x
  x 1   | x x x x x x
  1 x   | x x x 0 x x

We again find that one Funct bit distinguishes when OP1 is asserted, namely Funct bit 2 (in this case OP1 is asserted when Funct bit 2 is false).

As a result the truth table for OP1 reduces to the 3-row version shown on the right.

Although this truth table would yield a fairly small circuit, we shall to simplify it further.

  ALUOp | Funct
  1 0   | 5 4 3 2 1 0
  ------+------------
  0 x   | x x x x x x
  1 x   | x x x 0 x x

Recall from the original table, that the x 1 in the second row is really 0 1.

Although x 1 gives us more freedom than 0 1 in implementing this row by itself, we are able to simply further by undoing the don't care and noting that, with 0 1 in the second row, rows 1 and 2 can be combined to give the table on the right.

  ALUOp | Funct
  1 0   | 5 4 3 2 1 0
  ------+------------
  0 x   | x x x x x x
  x x   | x x x 0 x x

Last, we can use the first row to enlarge the scope (and hence simplify the implementation) of the last row resulting in the final table on the right.

So OP1 = (ALUOp1)' + (Funct2)'

aluop-control
The Circuit

After all the simplification the circuit itself is very easy and is shown in the diagram below.

Indeed, the simplifications were so successful that we are lead to question whether this was due to

  1. Good luck.
  2. Well chosen Funct field values by the MIPS engineers.
  3. A wise selection of instructions to implement by Patterson and Hennessy.
  4. Some of each.
The Main Control

At long last we get to use the opcode (instruction bits 31-26). Shown in blue in the diagram below is the control unit, the logic block that calculates the green control lines that have appeared above but were floating, i.e., the started from nothing.

Specifically our task, illustrated in the diagram below, is to calculate the following nine bits. (Note that a smaller picture—without the control—is shown here).



datapath+control

All 9 bits are determined by the opcode. We show the logic diagram after we illustrate the operation of the control logic.

Note that the MIPS instruction set is fairly regular. Most of the fields we need are always in the same place in the instruction (independent of the instruction type).

MemRead: Memory delivers the value stored at the specified addr
MemWrite: Memory stores the specified value at the specified addr
ALUSrc: Second ALU operand comes from (reg-file / sign-ext-immediate)
RegDst: Number of reg to write comes from the (rt / rd) field
RegWrite: Reg-file stores the specified value in the specified register
PCSrc: New PC is Old PC+4 / Branch target
MemtoReg: Value written in reg-file comes from (alu / mem)

We have just seen how ALUOp is used to calculate the control bits for the ALU. The purpose of the the remaining 7 bits (recall that ALUOp is 2 bits) are described in the table to the right and their uses in controlling the datapath is shown in the picture above.

We are interested in four opcodes.

Do a stage play

The following figures illustrate the play. Bigger versions of the pictures are here.

R-type Instructions

execute-R-1



execute-R-2



execute-R-3



execute-R-4



lw Instruction

execute-lw-1



execute-lw-2



exeucte-lw-3



execute-lw-4



execute-lw-5

Start Lecture #16

Remark: Lab 5 assigned, due 8 November 2011.

Truth Tables

The following truth table shows, for each of the four opcodes we are studying, the values needed for each control line.

Recall that we have more than four instructions since we are implementing several R-type instructions all of which have the same opcode (opcode zero). As we have seen, these R-type instructions are distinguished by the 6 Funct bits.

Instruction Op5Op4Op3Op2Op1Op0 RegDstALUSrcMemtoRegRegWrite MemReadMemWriteBranchALUOp1 ALUOp0


R-type 000000 100100 010
lw 100011 011110 000
sw 101011 X1X001 000
beq 000100 X0X000 101
ControlSignal R-typelwswbeq
Inputs Op50110
Op40000
Op30010
Op20001
Op10110
Op00110
Outputs RegDst10XX
ALUSrc0110
MemtoReg01XX
RegWrite1100
MemRead0100
MemWrite0010
Branch0001
ALUOp11000
ALUOp00001




The numerous columns, many with wide labels clearly leads to an extremely wide, and hence awkward table.

To make the table easier to read, Patterson and Hennessy draw this particular table in a non-standard manner as shown on the right.

The key change is that what was previously the column headings is now the row headings.

Just for fun, I tried to keep the original format and arrived at the version shown below.



Instruc-
tion
Op5Op4Op3 Op2Op1Op0|
|
Reg
Dst
ALU
Src
Mem
toReg
Reg
Write
Mem
Read
Mem
Write
BranchALU
Opt1
ALU
Opt0

|
R-type 000000| 100100 010
lw 100011| 011110 000
sw 101011| X1X001 000
beq 000100| X0X000 101
control
The Circuit

As always, given a truth table, it is quite easy to produce logic equations and a logic diagram, both in PLA style (i.e., using 2-levels of logic.

The circuit, drawn in PLA style is shown on the right.

Homework: In a previous homework, you modified the datapath to support addi and a variant of lw. Determine the control needed for these instructions.

Homework: Can we eliminate MemtoReg and use MemRead instead?

Homework: Can any other control signals be eliminated?


Implementing a J-type Instruction, Unconditional Jump

Recall the jump instruction.

  opcode  addr
  31-26   25-0

Addr is a word address. Since the machine is byte addressable, we need to shift the address left 2 bits (filling the right with zeros).

The address in the instruction is 26 bits. When shifted and and 0 filled, the result is 28 bits. But the machine has 32-bit addresses. Where do the remaining 4 bits come from?
Answer: The high order 4 bits of the new address are set equal to the high order 4 bits of the previous instruction (after incrementing the latter by 4).

This is quite easy to implement as seen in the following diagram. Basically all that is added to the datapath is one mux and its associated control line (plus a few wires).

including-jump

What's Wrong, i.e., Why Isn't This Used?

Some instructions are likely slower than others and we must set the clock cycle time long enough for the slowest. The disparity between the cycle times needed for different instructions is quite significant when one considers implementing more difficult instructions, like divide and floating point ops.

Actually, if we considered cache misses, which result in references to external DRAM, the cycle time ratios would exceed 100.

Possible solutions

  1. Variable length cycle.
    How would this be implemented?
  2. Asynchronous logic
  3. Multi-cycle instructions.
  4. Pipeline the cycles.
  5. Multiple datapaths (superscalar).
  6. VLIW (Very Long Instruction Word).

4.5 An Overview of Pipelining

sandbagging

Patterson and Hennessy give a real-world example of pipeline based on doing multiple loads of laundry. For variety, I will present a different example, based on sandbagging a river to prevent (or at least minimize) flooding.

The Sandbagging Problem

We have a huge quantity of dirt in the western part of an old gray town and a river with rising water in the eastern part. Since we anticipated the possibility of the river rising, we stockpiled empty burlap bags near the dirt and we have a small loop of train tracks running between the dirt and the river. We purchased two bright red carts and placed them on track one at the dirt and one at the river.

A Single Cycle Solution

If we adopted the method of our single cycle MIPS implementation we would proceed as follows.

  1. We fill a bag with dirt and put it on a nearby cart.
  2. We send the cart east.
  3. When the cart arrives at the eastern part of the loop, we carry the bag of dirt to the river.
  4. We place the bag of dirt at the correct place.
  5. We send an empty cart west.
  6. Repeat.

If we make the simplifying assumption that each of the five steps takes the same time, say T minutes, then it takes 5T minutes to complete job for one bag of sand.

A Pipelined Solution

We can do better than the approach just given; we can pipeline the activities. That is, once we start the cart carrying bag 1 east (step 2 of bag 1 begins) we can immediately start to fill bag 2 (step 1 of bag 2 begins)

When carrying bag 1 to the river (step 3 of bag 1 begins), we can start the cart carrying bag 2 east (step 2 of bag 2 begins) and can start filling bag 3 (step 1 of bag 3 begins).

It gets better. When placing bag 1 at the correct place (step 4 of bag 1 begins), we can (i) carry bag 2 to the river (step 3 of bag 2 begins), (ii) send the cart carrying bag 3 east (step 2 of bag 3 begins, and (iii) start filling bag 4 (step 1 of bag 4 begins).

Finally, when sending an empty cart west for the first time (step 5 of bag 1 begins), we can (i) place bag 2 at the correct place (step 4 of bag 2 begins), (ii) carry bag 3 to the correct place (step 3 of bag 3 begins, (iii) start the cart carrying bag 3 east (step 2 of bag 4 begins), and (iv) start filling bag 5 (step 1 of bag 5 begins).

Comparison

The second solution seems much better: Instead of a sand bag being placed onces every 5T minutes, we now place one every T minutes, a fivefold improvement.

But the time for each sand bag is unchanged; it remains 5T. The improvement comes from the fact that we are working on several sand bags simultaneously. This is the gain in pipelining. The overall latency of each operation remains constant (actually it increases—i.e., gets worse—slightly), but the throughput increases—i.e. gets better—considerably.

Put another way we can say that pipeline improves performance by increasing throughput not by decreasing the time for one instruction.

Pipelining the MIPS Processor

The same idea used for sand bagging and laundry can be applied for executing computer instructions. For executing MIPS instructions the pipeline has 5 steps or stages.

  1. Fetch the instruction from memory.
  2. Read the registers and decode the instruction.
  3. Use the ALU to execute the instruction or compute an address.
  4. Access data memory.
  5. Write a register (with the result).

An Illustrative example

Figure 4.26: Patterson and Hennessy 4e
Approximate times for each pipeline stage.
Instruc-
tion
Instruc-
tion fetch
Register
read
ALU
Operation
Data
access
Register
write
Total
time
lw200 ps100 ps200 ps 200 ps100 ps800 ps
sw200 ps100 ps200 ps 200 ps700 ps
R-type200 ps100 ps200 ps 100 ps600 ps
beq200 ps100 ps200 ps 500 ps

The table on the right gives approximate times for each part of executing the MIPS instructions we have implemented.

Using our single cycle implementation would need make the clock cycle time 800ps, the time for the longest instruction.

Using a five-stage pipeline, we would need to make the cycle time 200ps, the time of the slowest stage. Since all instructions go through all 5 stages (even if nothing is done for that instruction during one or more stages), every instruction will take 1000ps=1ns from beginning to end.

This sounds worse!

Indeed, it is worse if you judge performance by the time for one instruction. But, as we mentioned before, the more relevant measure is the throughput, i.e., the number of instructions executed in one second.

Let's look at executing a three instruction program that adds value in register 3 to a location in memory.

pipeline-execution
  lw  $r1, 50($r2)   // uses all 5 stages
  add $r1, $r1, $r3  // no data access
  sw  $r1, 50($r2)   // no register write

Our single stage implementation requires 3 * 800ps = 2400ps to execute these three instructions. The instruction execution time is 800ps and the throughput is

  3 instructions / (2400 * 10-12 second) = 1.25 * 109 instructions/second

The pipelined execution requires 1400ps; the instruction execution time is 1000ps; and the throughput is

  3 instructions / (1400 * 10-12 second) ∼ 2.14 * 109 instructions/second

The result would get better if we used a bigger example. Indeed the asymptotic speedup is 4 since the single cycle implementation starts one instruction every 800ps and the pipelined implementation starts one instruction every 200ps.

Remember that real instructions execute (at least) billions of instructions so the value obtained for such programs would be extremely close to 4.

Designing Instruction Sets for Pipelining

The MIPS instruction set was designed to ease pipelining in the following ways.

  1. All instructions are the same length. This facilitates fetching instruction 2 before have really begun executing instruction 1. If you needed to decode the instruction to determine its length, I-fetch could not begin for an instruction until the previous instruction was decoded.
  2. Few formats with source register fields always in the same place. If the source fields depended on the opcode, we could not read the registers right after I-fetch, but would need another pipeline stage to determine which registers to read. Note that it is not a problem that we might actually read a register whose value is not needed.
  3. Memory operands appear only in loads and stores. If, for example, you could add a register to a memory location, we would need to use the ALU twice (once to calculate the memory address; a second time to do the actual addition). This would need another pipeline stage (and a second ALU).
  4. Operands are word aligned in memory. When we learn about memory, we will see that this permits a single 32-bit data-access.

Pipeline Hazards

The example shown above gives a slightly unrealistic view of pipeline since we have not discussed hazards that can delay an instruction because it cannot execute the next pipeline stage right away.

Structural Hazards

This occurs when the hardware cannot execute all the actions required during one cycle. MIPS was designed to minimize this possibility, but consider a different design that had one memory instead of the two as in MIPS. Then, if the first instruction was lw, it would be accessing this memory during cycle 4 to read the data. But at the same time, the 4th instruction needs to do an I-fetch. The resulting contention for the combined data-instruction memory is a structural hazard.

Data Hazards

  lw  $r1,40($r8)
  lw  $r2,44($r8)
  add $r3,$r1+$r2
  sw  $r3,48($r8)

Consider the familiar statement A = B+C; found in most languages. It would most likely be translated by the compiler into an instruction sequence like the one shown on the right.

Assume the sequence starts at cycle 1. Then during cycle 4, the add instruction reads registers $r1 and $r2.

But those registers are not written with the required values until cycles 5 and 6. Hence the third instruction encounters a data hazard at its 2nd cycle, which causes the pipeline to stall for three cycles so that at cycle 7, the add can perform its second step. All such stalls are often called bubbles in the pipeline.

Forwarding (Bypassing)
  add $r1,$r1,$r2
  add $r1,$r1,$r3

Consider the two instruction sequence on the right that replaces register r1 with the sum of the first three registers. The second instruction adds the third register to what should be the sum of the first two. Looking at a cycle-by-cycle picture of the pipeline (draw this on the board) we see that the second instruction reads register r1 during cycle 3 (its second cycle). But that sum will not appear in register r1 until the fifth cycle.

However, if we look again at the cycle-by-cycle picture, we see that the sum is calculated during cycle 3 (the 3 stage of the first instruction) and not actually used until cycle 4 (the third stage of the 2nd instruction).

As a result one could run a wire from the end of stage 3 to the beginning of stage 3 (and add a mux and some serious control logic) and get the value there in time.

We say the value has been forwarded from the first instruction to the second or that it has bypassed some of the steps.

In this case the solution was perfect, no bubble remains.

Homework: How would forwarding be used in the previous 4-instruction sequence? Do any bubbles remain?

Control Hazards

   beq $r1,$r2,L
   some instructions
L: some other instructions

Consider the conditional branch shown on the right. During cycle 2 we need to fetch the 2nd instruction to execute, but we don't know what that instruction is since we don't know yet if the branch will be taken. We won't know until the end of cycle 3, when the ALU has determined if registers r1 and r2 are equal.

We could guess that the branch will not be taken and start executing some instructions. If we guess wrong, we must throw out the work we did based on the guess.

This hazard has lead to a large field of study called branch prediction that uses sophisticated techniques to make a more intelligent (i.e., a more-likely-to-be-correct) guess as to whether or not the branch will be taken.

Pipeline Overview Summary

Pipelining is an important component in the processor designer's toolbox; all modern microprocessors use it. Pipelining permits the execution of consecutive instructions to be overlapped.

Although no instruction is itself sped up (indeed some are slowed down), the throughput is increased significantly.

Hazards can greatly decrease the potential improvement of pipelining; a well designed ISA can make hazards easier to deal with, but in any case hazards complicate the design of modern high-performance processors.

Start Lecture #17

Homework: 4.6.1, 4.6.2, 4.7.1, 4.7.2, 4.9.1-3.

4.6 Pipelined Datapath and Control

Remark: We only sketch some ideas in the rest of this chapter. For a complete treatment, read the book carefully.

The diagram below shows the datapath divided into the same 5 pipeline stages we just studied

. datapath-pipe-1

These stages are normally referred to as:

  1. IF: Instruction fetch.
  2. ID" Instruction decode / register read.
    For MIPS the former is quite easy; for other ISAs, less so.
  3. EX: Execute / Address Calculation.
  4. MEM: Memory Access.
  5. WB: Write Back.
    Write the result back into the register file.

The next step is to capture the state after each stage. This means that we need to replace the simple dotted red lines, which were just for our visualization by pipeline registers that hold all the values produced by each stage

datapath-pipe-2

Now you can do another stage play and see that, at the beginning of each stage, the pipeline registers are read and, at the end of each stage, they are written.

There are various subtleties that must be addressed.

For example, the register file is written during the fifth pipeline stage, but the register number is read from the instruction during the first stage. Hence, by the time the fifth stage is executed, the register number is from a later instruction.

This particular problem is fixed by moving the write register number from pipeline register to pipeline register as the instruction moves through the pipeline. At the fifth stage, the register number is sent from the last pipeline register to the write register input of the register file.

You might wonder why there are only 4 pipeline registers since there are five stages. The answer is that all the fifth stage does is write a register so this value is being saved for subsequent instructions and no pipeline register is needed at the end of stage 5.

Graphically Representing Pipelines

pipeline-execution

The last diagram above represents what is called a single-clock-cycle pipeline diagram. Diagrams such as the one on the right (which we have already discussed) are called multiple-clock-cycle pipeline diagrams. The later are easier to follow, but supply fewer details.

Pipelined Control

We already calculated the control lines. The trouble is that we calculate them at the beginning, but use them in subsequent stages. Hence them must be passed from pipeline register to pipeline register as the instruction moves along the pipeline.

4.7 Data Hazards: Forwarding versus Stalling

Data Hazards and Stalls

4.8 Control Hazards

Assume Branch Not Taken

Reducing the Delay of Branches

Dynamic Branch Prediction

Pipeline Summary

4.9 Exceptions

How Exceptions are Handled in the MIPS Architecture

Exceptions in a Pipelined Implementation

4.10 Parallelism and Advanced Instruction-Level Parallelism

The Concept of Speculation

Static Multiple Issue

An Example: Static Multiple Issue with the MIPS ISA

Dynamic Multiple-Issue Processors

Dynamic Pipeline Scheduling

Power Efficiency and Advanced Pipelining

4.11 Real Stuff: the AMD Opteron X4 (Barcelona) Pipeline

4.12 Advanced Topic: an Introduction to Digital Design Using a Hardware Design Language to Describe and Model a Pipeline and More Pipelining Illustrations

4.13 Fallacies and Pitfalls

Three points are made

  1. Fallacy: Pipelining is easy.
    Pipelining is easy conceptually but the details are challenging, to say the least, and have caused many bugs.
  2. Fallacy: Pipelining can be implemented independent of technology.
    Of course it can be implemented independent of technology, but the best implementation has changed several times since 1980 due to first more resources (transistors, memory) becoming available, and more recently due to power limitations.
  3. Pitfall: Failure to consider instruction set design can adversely affect pipelining.
    For example, varying length instructions make it hard to find the next instruction.

4.14 Concluding Remarks

We have seen how to design the datapath and control for a subset of the MIPS processor using a single-cycle strategy. Although successful this simple implementation is too slow so we investigated, to a limited extent, a more aggressive, pipelined implementation.

In addition to pipelining, modern designs are multiple-issue/superscalar. That is, they issue several instructions each cycle and hence have several instructions performing each pipeline stage and thus can have very many instructions active (in flight) at one time.

This, coupled with out-of-order execution, which we haven't discussed, complicates the design and increases the power usage considerably.

Design complexity may have already already hit and passed its peak. Power concerns have causes all the major players to cut back on the complexity of their designs.

4.15 Historical Perspective and Further Reading

Chapter 1 Computer Abstractions and Technology (Revisited)

1.1 Introduction

Read.

1.2 Below Your Program

Read.

1.3 Under the Covers

Read.

Problem 1.1 was assigned earlier in the course.

1.4 Performance

Defining Performance

Throughput measures the number of jobs per day/second/etc that can be accomplished.

Response time measures how long an individual job takes.

We define Performance as 1 / Execution time.

Relative Performance

We say that machine X is n times faster than machine Y or machine X has n times the performance of machine Y if the execution time of a given program on X = (1/n) * the execution time of the same program on Y.

But what program should be used for the comparison? Various suites have been proposed; some emphasizing CPU integer performance, others floating point performance, and still others I/O performance.

Measuring Performance

How should we measure execution time?

We mostly employ user-mode CPU time, but this does not mean the other metrics are worse.

Cycle time vs. Clock rate.

What is the cycle time for a 700MHz computer?

What is the clock rate for a machine with a 10ns cycle time?

The Classic CPU Performance Equation

The execution time for a given job on a given computer is

  (CPU) execution time = (#CPU clock cycles required) * (cycle time)
                       = (#CPU clock cycles required) / (clock rate)

Since the number of CPU clock cycles required equals the number of instructions executed times the average number of cycles in each instruction, we can write this equation in other equivalent forms.

  CPU Time (in seconds) =  #Instructions * CPI * Cycle_time (in seconds).
  CPU Time (in ns)      =  #Instructions * CPI * Cycle_time (in ns).
  CPU Rate (in seconds) =  #Instructions * CPI / Clock_Rate (in Hz).

What is the CPI?

In our single cycle implementation, the number of cycles required is just the number of instructions executed. That is, the CPI is 1.

Similarly, if every instruction took 5 cycles, the number of cycles required would be five times the number of instructions executed.

But real systems are more complicated than that!

After extensive measurements, one calculates for a given machine the average CPI (cycles per instruction).

We shall sometimes assume this average CPI actually applies to all instructions. Other times we shall say something like

Assume there are two classes of instructions. Class A instructions require 4 cycles to execute; class B instructions require 3 cycles to execute. Assume an execution of program P involves 30% class A instructions and 70% class B. What is the (average) CPI for this execution?

The number of instructions required for a given program depends on the instruction set. For example, assume we want to add the contents of register 1 to a location X in memory. Then MIPS would require three instructions; whereas, x86 needs only one.

  lw  $r3,X
  add $r3,$r1       add X,$r1
  sw  $r3,X

CPI is a good way to compare two implementations of the same instruction set (also called the same ISA instruction set architecture).

IF the clock cycle is unchanged, then the performance of a given ISA is inversely proportional to the CPI (e.g., halving the CPI doubles the performance).

Naturally, complicated instructions often take longer to execute. They require either more cycles or a longer cycle time. Older machines with complicated instructions (e.g., the Digital Equipment Corporation VAX, an important machine in the 1980s) had CPI>>1.

As we have seen, with pipelining we can have many cycles for each instruction but still achieve a CPI of nearly 1.

Modern superscalar machines often have a CPI less than one. As a result sometimes one speaks of the IPC or instructions per cycle for such machines. However, we won't use IPC.

Start Lecture #18

Remark: For midterm grades I computed

  0.6* exam + 0.4 * (lab1 + lab2 + lab3 + lab4)

I did not believe there was enough precision to give +- so only gave A,B,C,D.

Remark: I draft of lab6 is available. It will be assigned next class. I constructed is a demo of implementing a small RAM in logisim. The file name is ram.circ. It implements (a small version) of the Data Memory component of lab 6.

Do on the board the following example from pages 35-36.

A compiler designer is developing code sequences for a particular computer. The computer has three classes of instructions, A, B, and C, which have CPIs of 1, 2, and 3 respectively.

Note: This is shorthand for saying Class A instructions, on average, add one cycle to the execution time and similarly for classes B and C. It is not saying that executing one class A instruction takes one cycle from beginning to end. Again we see the difference between the latency of a single instruction and the throughput (instructions per second). Perhaps it would be better to say that the cost of a class A instruction is one cycle.

The compiler writer has a choice of two possible sequences of machine language instructions as a translation of a particular high-level language statement. The first sequence has 2 class A instructions, 1 class B, and 2 class C. The second sequence has 4 class A and 1 each class B and C.

Which sequence executes the most instructions? Which is faster? What is the CPI for each sequence.

Homework: Carefully go through and understand the example that I just did in class.

Homework: 1.3.1, 1.3.3, 1.3.4, 1.4.1, 1.4.2, 1.5.2a.

1.5 The Power Wall

1.6 The Sea Change: The Switch from Uniprocessors to Multiprocessors

1.7 Real Stuff: Manufacturing and Benchmarking the AMD Opteron X4

We will cover none of the manufacturing and only some of the benchmarking material

As mentioned in the previous section, some instructions take longer than others. Moreover, different ISAs perform better on certain instructions and other ISAs perform better on other instructions. Different application programs use different mixes of instructions.

It might be, for example, that computer A does great on programs that reference memory heavily, but poorly on programs dominated by floating point operations. In contrast computer B might excel on floating-point, but be sluggish on memory references.

As you can imagine, computer manufacturers would prefer that a customer evaluates the company's products by running programs on which the products do particularly well.

To standardize the measurements, many vendors agreed on certain sets of benchmarks on which they would provide performance evaluations. Perhaps the best know standard benchmarks are those sanctioned by SPEC (System Performance Evaluation Cooperative). SPEC actually contains several benchmark suites.

1.8 Fallacies and Pitfalls

Amhahl's Law

One fallacy is to assume that by fixing part of the problem, the entire problem is fixed to a very great extent.

For example, assume a simple system with two classes of instructions

A new wiz-bang floating-point unit is proposed that speeds floating-point instructions by a factor of 5 (new CPI is 2), has no effect on cycle time, and only doubles the cost of the machine.
Sounds great; a speedup of 5 at a cost of only 2!
But just how great is it?

Say the customer is primarily interested in a single application A. Measurements show that A executes N instructions, 20% of which are floating point.

To execute application A, the old system takes

  .2N * 10 cycles + .8N * 2 cycles  =  2N + 1.6N cycles  =  3.6N cycles

The new, improved system would take

  .2N * 2 cycles + .8N * 2 cycles  =  .4N + 1.6N cycles  =  2N cycles

Since the cycle time hasn't changed, execution time is proportional to the number of cycles. Thus the new system is 3.6N / 2N = 1.8 times faster for only twice the price.

No sale!

Speedup Using Multiple Computers (the Original Amdahl's Law)

NOAA has a new computer program that is predicting tomorrow's weather very well, but the computation takes a week, which makes the results useless. They need to reduct the 168 hours (one week) to 1 hour.

They know that the program spends 99% of its time doing material that can be partitioned evenly on up to a thousand processors.

Since they need a speedup of at least 168 and have the money, they decide to buy 1000 processors. How long does the program now take to run?

Answer: 1% of 168 hours is 1.68 hours; 99% of 168 hours is 166.32 hours. The 1000 processors cooperate on the second piece and reduce it to 166.32/1000=.16632 hours.

However, the small 1% piece isn't be sped up at all and still takes 1.68 hours. So the entire job takes 1.68+.16632 = 1.84632 hours, which is exceeds the 1 hour requirement.

The 1000 processors gave a speedup of only 168/1.84632 ∼ 91.

Homework: What would the speedup be if they purchased only 100 processors?

Power

MIPS vs Time

MIPS is an acronym abbreviating Millions of Instructions Per Second. It is a unit of rate or speed (like MHz); it is NOT a unit of time (like ns.).

As its full name suggests, MIPS is defined as

  how many million instructions were executed / how many seconds were required

This is the same as

  the number of instructions executed / the number of microseconds used

The instructions we have been studying (lw, R-type, etc) are those of the MIPS computer company. That usage of the word MIPS is different from the acronym above, but the company's founders certainly knew about the acronym above.

Indeed, the company started with a Stanford research project headed by Hennessy. This project was called MIPS standing for Microprocessor without Interlocked Pipeline Stages. However, the microprocessors produced by the MIPS company did have interlocked pipeline stages.

At roughly the same time, Patterson headed a research project called RISC standing for Reduced Instruction Set Computer. Sun Microsystems was to an extent based on this research project.

Problems with the MIPS Rating

MIPS only counts instructions and does not take into account that some ISAs require more instruction than other ISAs to solve the same problem. For example, we saw that adding a register to a memory location takes 1 instruction on an x86 ISA but three on a MIPS. If the single instruction took one microsecond, the x86 would achieve a 1 MIPS rating. If the three instructions took a total of 2 microseconds, the MIPS computer would achieve a 1.5 MIPS rating, much better than the x86 even though it required twice as long to accomplish the same task.

For this reason, the MIPS cannot be used (or at least should not be used) to compare systems with different ISAs.

Even with a fixed ISA there are difficulties with MIPS. As with many computer ratings, the program used is important. A program that has predominately fast instructions will achieve a higher MIPS rating that a program that has predominately slow instructions.

A new sophisticated compiler might be able to reduce the number of instructions needed to complete a program and also reduce the total execution time. Clearly a good thing. But, if the instructions eliminated were the fastest ones, then the MIPS rating would go down even though performance went up! To say it in reverse, it is often the case that if one padded a program with NOPs (which are very fast), the program would have exactly the same effect, would take longer, would execute more instructions, but yet would likely achieve a higher MIPS rating than the original.

MFLOPS (Megaflops)

For numerical calculations floating point operations are often the ones you are interested in; the others are overhead (a very rough approximation to reality). For this reason the MFLOPS (Millions of FLoating point OPerations per Second) was introduced.

It has similar problems to MIPS.

Benchmarks

Benchmarks are better than MIPS or MFLOPS, but still have difficulties.

1.9 Concluding Remarks

Homework: Read this (very short) section.

1.10 Historical Perspective and Further Reading

Chapter 5: Large and Fast: Exploiting the Memory Hierarchy

Homework: Read Chapter 5.

Remark: Perhaps the chapter should be entitled Large vs. Fast.

5.1: Introduction

An ideal memory is

Unable to achieve the impossible ideal we use a memory hierarchy consisting of

  1. Registers.
  2. Cache (really L1, L2, and maybe L3).
  3. (Central or Main) Memory.
  4. Disk.
  5. Archive (e.g. Tape).

... and try to satisfy most references in the small fast memories near the top of the hierarchy.

There is a capacity/performance/price gap between each pair of adjacent levels. We will study the cache-to-memory gap.

We observe empirically (and teach in OS).

Start Lecture #19

Remark: Lab 6 assigned; due 17 November 2011.

A cache is a small fast memory between the processor and the main memory. It contains a subset of the contents of the main memory.

A Cache is organized in units of blocks or lines. Common block sizes are 16, 32, and 64 bytes.

A block is the smallest unit we can move to/from a cache (some designs move subblocks, but we will not discuss such designs).

A hit occurs when a memory reference is found in the upper level (small, fast) of the memory hierarchy.

Definitions

Addressing Bytes, (4-byte) Word, and Blocks

Consider the following address (in binary). 10101010_11110000_00001111_11001010.
This is a 32-bit address. I used underscores to separated it into four 8-bit pieces just to make it easy to read; the underscores have no significance.

Machine addresses are non-negative (unsigned) so the address above is a large positive number (greater than 2 billion).

All the computers we shall discuss are byte addressed. Thus the 32-bit number references a byte. So far, so good.

The (4-Byte) Word Addressed and the Byte Offset

We will always assume that each word is four bytes. That is, we assume the computer has 32-bit words. This is not always true (many old machines had 16-bit, or smaller, words; and many new machines have 64-bit words), but to repeat, we will always assume 32-bit words.

Since 32 bits is 4 bytes, each word contains 4 bytes. We assume aligned accesses (as does the MIPS architecture we studied). This means that a word (a 4-byte quantity) must begin on a byte address that is a multiple of the word size, i.e., a multiple of 4. So word 0 includes bytes 0-3; word 1 includes bytes 4-7; word n includes bytes 4n, 4n+1, 4n+2 and 4n+3; and the four consecutive bytes 6-9 do NOT form a word.

What word includes the byte address given above, 10101010_11110000_00001111_11001010?
Answer: 10101010_11110000_00001111_110010, i.e, the address divide by 4.
What are the other bytes in this word?
Answer: 10101010_11110000_00001111_11001000,   10101010_11110000_00001111_11001001,   and 10101010_11110000_00001111_11001011

What is the byte offset of the original byte in its word?
Answer: 10 (i.e., two), the address mod 4..
What are the byte-offsets of the other three bytes in that same word?
Answer: 00, 01, and 11 (i.e, zero, one, and three).

The 32-Byte Block Addressed and the Word and Byte Offset

Blocks vary in size. We will not make any assumption about the size, other than that it is a power of two. For these examples (only), assume each block is 32 bytes.

Since we assume aligned accesses, each 32-byte block has a byte address that is a multiple of 32. So block 0 is bytes 0-31, which is words 0-7. Block n is bytes 32n, 32n+1, ..., 32n+31.

What block includes our byte address 10101010_11110000_00001111_11001010?
Answer: 10101010_11110000_00001111_110, i.e., the byte address divide by 32 (the number of bytes in the block) or the word address divided by 8 (the number of words in the block).

5.2 The Basics of Caches

We start with a very simple cache organization, one that was used on the Decstation 3100, a 1980s workstation. In this design cache lines (and hence memory blocks) are one word long.

Also in this design each memory block can only go in one specific cache line.

We shall assume that each memory references issued by the processor is for a single, complete word. This assumption holds for the MIPS subset we implemented since the only memory access were lw and sw. The full MIPS ISA, however, includes instructions that reference bytes and halfwords.

Accessing a Cache

On the right is a diagram representing a direct mapped cache with 4 blocks and a memory with 16 blocks.

How can we find a memory block in such a cache? This is actually two questions in one.

  1. Is the memory block present in the cache?
  2. Where in the cache is the memory block, assuming it is present?

The second question is the easier. Let C be the number of blocks in the cache. Then memory block number N can be found only in cache line number N mod C (it might not be present at all).

But many memory blocks are assigned to that same cache line. For example, in the diagram above all the green blocks in memory are assigned to the one green block in the cache.

So the first question reduces to: Is memory block N present in cache block N/C?

Referring to the diagram we note that, since only a green memory block can appear in the green cache block, we know that the last two digits of the memory block in the green cache block are 10 (the number of the green cache block). So to determine if a specific green memory block is in the green cache block we need the rest of the memory block number. Specifically is the memory block in the green cache block 0010, 0110, 1010, or 1110? It is also possible that the green cache block is empty (called invalid), i.e, it is possible that no memory block is in this cache block.

When the system is powered on, all the cache blocks are invalid so all the valid bits are off.

Addr(10)Addr(2)hit/missblock#
2210110miss110
2611010miss010
2210110hit110
2611010hit010
1610000miss000
300011miss011
1610000hit000
1810010miss010

On the right is an example from the book (page 460). It refers to figure 5, which is an enlarged version of the example diagram above. Figure 5 has C=8 (rather than 4) and M=32 (rather than 16).

In both the diagram above and the example from the book, we have M/C=4 memory blocks eligible to be stored in each cache block. Thus there are two tag bits for each cache block.

cache-contents

Cache Contents, Hits, and Misses

Shown on the right is a eight entry, direct-mapped cache with block size one word. As usual all references are for a single word. In order to make the diagram and arithmetic smaller the machine has only 10-bit addressing, instead of our usual 32-bit addressing. Above the cache we see a 10-bit address issued by the processor.

There are several points to go over.

  1. The valid bit. If this bit is not set, then the entire line is invalid. When the system is first powered on all the lines are invalid.
  2. This machine, like all the ones we will study is byte addressed and has 4-byte words. Since the cache only handles references to a word (in reality the cache must also handle single byte references, but we will not study these). Thus the rightmost two bits of the address from the processor, which specify the byte offset within the word, are ignored for cache access.
  3. Once we drop the byte-offset bits, the word address of the reference is 11010100. Since the block size is one word, the block number is also 11010100. Since the cache has eight entries, the cache line number is 11010100 mod 23 = 100 (the low order three bits).
  4. We see that the valid bit is on for entry 100 (i.e., entry 4) so the line is valid.
  5. However, the tags do not match. Hence the reference is a cache miss.
  6. Would a memory reference 1000001001 be a hit or miss?
  7. Would a memory reference 0000001001 be a hit or miss?
  8. Explain in class how we know that the data field in entry 2 contains the contents of word 130.
  9. Make sure you understand why the other data fields contain the contents indicated.

Circuitry Needed to Detect Hits and Misses

The circuitry needed for a simple cache (direct mapped, block size 1 word, all references to 1 word) is shown on the right. The only difference from the example above is size. This cache holds 1024 blocks (not just 8) and the memory holds 230∼1,000,000,000 blocks (not just 32). That is, the cache size is 4KB and the memory size is 4GB.

To determine if we have a hit or a miss, and to return the data in case of a hit is quite easy, as the circuitry indicates.

Make sure you understand the division of the 32 bit address into 20, 10, and 2 bits.

Calculate on the board the total number of bits in this cache and the number used to hold data.

Homework: 5.3.1. Calculate the total number of bits in the 5.3.1 cache and the number used to hold data.

Processing a Read for this Simple Cache

The action required for a hit is clear, namely return to the processor the data found in the cache.

For a miss, the best action is fairly clear, but requires some thought.

Handling Cache Misses in the Processor

We just need to note a few points.

Handling Writes

Processing a write for our simple cache (direct mapped with block size = reference size = 1 word).

We have 4 possibilities: For a write hit we must choose between Write through and Write back. For a write miss we must choose between write-allocate and write-no-allocate (also called store-allocate and store-no-allocate and other names).

Write Hits

Write through: Write the data to memory as well as to the cache.

With a write-through cache policy, both the memory and the cache are always up-to-date.

Write back: Don't write to memory now, do it later when this cache block is evicted.

With a write back policy, the cache is always up-to-date, but the memory can be stale (contain an out-of-date value).

The fact that an eviction must trigger a write to memory for write-back caches explains the comment above that the write hit policy effects the read miss policy.

Write Misses

Write-allocate: Allocate a slot and write the new data into the cache (recall we have a write miss). The handling of the eviction this allocation (probably) causes depends on the write hit policy.

  1. If the cache is write through, discard the old data (since it is in memory) and write the new data to memory (as well as in the cache).
  2. If the cache is write back, the old data must now be written back to memory, but the new data is not written to memory.

Write-no-allocate: Leave the cache alone and just write the new data to memory.

Write no-allocate is not normally as effective as write allocate due to temporal locality.

Start Lecture #20

Remark: I added a section Cache Contents, Hits, and Misses to lecture 19, which we will cover now.

The Decstation 3100 Write Policy

The simplest write policy is write-through, write-allocate. The decstation 3100 discussed above adopted this policy and performed the following actions for any write, hit or miss, (recall that, for the 3100, block size = reference size = 1 word and the cache is direct mapped).

  1. Index the cache using the correct LOBs (i.e., not the very lowest order bits as these give the byte offset).
  2. Write the data and the tag into the cache.
  3. Set Valid to true (it may already be true).
  4. Send request to main memory.

Although the above policy has the advantage of simplicity, it is out of favor due to its poor performance.

Improvement: Use a Write Buffer

Unified vs Split I and D (Instruction and Data) Caches

Given a fixed total size (in bytes) for the cache, is it better to have two caches, one for instructions and one for data; or is it better to have a single unified cache?

Improvement: Multiword Blocks

The setup we have described does not take any advantage of spatial locality. The idea of having a multiword block size is to bring into the cache words near the referenced word since, by spatial locality, they are likely to be referenced in the near future.

We continue to assume (for a while) that the cache is direct mapped and that all references are for one word.

The book's terminology for byte offset and block offset is inconsistent. The byte offset gives the offset of the byte within the word so the offset of the word within the block should be called the word offset, but alas it is called the block offset in the 2e, 3e, and 4e. I don't know if this is standard terminology or a long standing typo in all three editions. I wrote to Patterson, who basically agreed. I will try to use the longer but clearer term word-in-block for the offset of the word in the block.

cache-bigblock

The figure to the right shows a 64KB direct mapped cache with 4-word (16-byte) blocks.

What addresses in memory are in the block and where in the cache do they go?

Show from the diagram how this gives the red portion for the tag and the green portion for the index or cache block number.

Consider the cache shown in the diagram above and a reference to word 17003.

Summary: Memory word 17003 resides in word 3 of cache block 154 with tag 154 set to 1 and with the valid 154 true.

The cache size or cache capacity is the size of the data portion of the cache (normally measured in bytes).

For the caches we have see so far this is the block size times the number of entries. For the diagram above this is 64KB. For the simpler direct mapped caches block size = word size so the cache size is the word size times the number of entries.

Note that the total size of the cache includes all the bits. Everything except for the data portion is considered overhead since it is not part of the running program.

For the caches we have see so far the total size is
(block size + tag size + 1) * the number of entries

Let's compare the pictured cache with another one containing 64KB of data, but with one word blocks.

  1. Calculate on the board the total number of bits in each cache; this is not simply 8 times the cache size in bytes.
  2. If the references are strictly sequential the pictured cache has 75% hits; the simpler cache with one word blocks has no hits.

Homework: 5.3.1 and 5.3.2. Also calculate the total number of bits in each of the caches and the number used to hold data.

How do we process read/write hits/misses for a cache with multiword blocks?

Why not make block size enormous? For example, why not have the cache be one huge block.

Memory Support for Wider Blocks

Recall that our processor fetches one word at a time and our memory produces one word per request. With a large block size cache, the processor still requests one word and the cache still responds with one word. However the cache requests a multiword block from memory and to date our memory is only able to respond with a single word.

The question is, "Which pieces and buses should be narrow (one word) and which ones should be wide (a full block)?". The same question arises when the cache requests that the memory store a block and the answers are the same so we will only consider the case of reading the memory).

Since the processor is only requesting a single word, a wide bus between the cache and processor seems silly. The processor would then need a mux to discard the other words.

One could imagine a buffer to store the entire block acting as a kind of L0 cache, but this would not be so useful if the L1 cache was fast enough).

The question we want to consider is whether the memory should be wide. That is, should the memory have enough pins and the bus enough enough wires so that the entire block can be transferred at once.

We make the following timing assumptions assumptions.

  1. Only 1 cycle is required to send the address. This is valid since only one address is needed per access for all designs.
  2. 15 clocks are required for each memory access (independent of width). Today the number would likely be bigger than 15 (since processor cycle times have improved faster that memory access times), but it would remain independent of the width
  3. 1 Clock is required to transfer each busload of data.
memory-width

Consider the three designs shown on the right. The left one assumes the memory delivers one word at a time and the bus is 1-word wide. This is the most economical design.

The middle design has a wide memory that can deliver an entire (4-word) block at one time and has a block-wide bus that can deliver the entire block to the cache in one cycle.

The rightmost design has four word-wide memories that are interleaved and thus can together produce a 4-word block at one time. However, the bus can only deliver one word at a time to the cache.

The question is how long does it take to satisfy a read miss for the cache above and each of the three memory/bus systems.

Interleaving works great because in this case we are guaranteed to have sequential accesses.

Imagine a design between (a) and (b) with a 2-word wide datapath.
It takes 33 cycles and is more expensive to build than (c).

Homework: Assume the block size is 8 words. How long would an access take for a narrow, wide, and interleaved design? How long for a 2-word wide design and for a 4-word design.

5.3: Measuring and Improving Cache Performance

Do the following performance example on the board. It would be an appropriate final exam question.

A lower base (i.e. miss-free) CPI makes stalls appear more expensive since waiting a fixed amount of time for the memory corresponds to losing more instructions if the CPI is lower.

A faster CPU (i.e., a faster clock) makes stalls appear more expensive since waiting a fixed amount of time for the memory corresponds to more cycles if the clock is faster (and hence more instructions since the base CPI is the same).

Another performance example.

Homework: Consider a system that has a miss-free CPI of 2, a D-cache miss rate of 5%, an I-cache miss rate of 2%, has 1/3 of the instructions referencing memory, and has a memory that gives a miss penalty of 20 cycles. The clock speed stays the same throughout this problem.

  1. What is the CPI?
  2. What would be the CPI if the memory was double speed, but the CPU+caches remained the same as the original?
  3. What would be the CPI if the memory remained the same as the original but the CPU+cache were double speed.
  4. How fast would the CPU+cache have to be so that the system was twice as fast as the original, with the original memory?

Remark: Larger caches have longer hit times.

Reducing Cache Misses by More Flexible Placement of Blocks

Improvement: Associative Caches

Consider the following sad story. Jane has a cache that holds 1000 blocks and has a program that only references 4 (memory) blocks, namely 23, 1023, 123023, and 7023. In fact the references occur in order: 23, 1023, 123023, 7023, 23, 1023, 123023, 7023, 23, 1023, 123023, 7023, 23, 1023, 123023, 7023, etc. Referencing only 4 blocks and having room for 1000 in her cache, Jane expected an extremely high hit rate for her program. In fact, the hit rate was zero. She was so sad, she gave up her job as webmistress, went to medical school, and is now a brain surgeon at the mayo clinic in Rochester MN.

So far we have studied only direct mapped caches, i.e., those for which the location in the cache is determined by the address. Since there is only one possible location in the cache for any block, to check for a hit we compare one tag with the HOBs of the addr.

The other extreme is a fully associative cache.

Set Associative Caches

Most common for caches is an intermediate configuration called set associative or n-way associative (e.g., 4-way associative). The value of n is typically 2, 4, or 8.

If the cache has B blocks, we group them into B/n sets each of size n. Since an n-way associative cache has sets of size n blocks, it is often called a set size n cache. For example, you often hear of set size 4 caches.

In a set size n cache, memory block number K is stored in set K mod the number of sets, which equals K mod (B/n).

Determining the Set Number and the Tag.

Recall that for the a direct-mapped cache, the cache index gives the number of the block in the cache. For a set-associative cache, the cache index gives the number of the set.

Just as the line number for a direct-mapped cache is the memory block number mod the number of blocks in the cache, the set number equals the (memory) block number mod the number of sets.

Just as the tag for a direct mapped cache is the memory block number divided by the number of blocks, the tag for a set-associative cache is the memory block number divided by the number of sets.

Do NOT make the mistake of thinking that a set size 2 cache has 2 sets, it has B/2 sets each of size 2.

Ask in class.

Why is set associativity good? For example, why is 2-way set associativity better than direct mapped?

Locating a Block in the Cache

cache-4-way

How do we find a memory block in a set associative cache with block size 1 word?

Recall that a 1-way associative cache is a direct mapped cache and that an n-way associative cache for n the number of blocks in the cache is a fully associative.

The advantage of increased associativity is normally an increased hit ratio.

What are the disadvantages?
Answer: It is a slower and a little bigger due to the extra logic.

Start Lecture #22

Remark: Lab 7 is assigned. It is due in 1.5 NYU weeks, i.e., 3 lectures, 6 December 2011.

Remark: Last time, after doing the 2-way set associative example, I should have redone it in class for 4-way set associative. Let's do it now. It is right before Determining the Set Number and the Tag.

Combining Set-Associativity and Multiword Blocks

cache-4-way-big-block

This is a fairly simple combination of the two ideas and is illustrated by the diagram on the right.

  1. Start with the picture just above for a set-associative cache.
  2. Each blue portion of the cache is now a block, not just a word.
  3. Hence the data coming out of the original multiplexor at the bottom right is a block. In the diagram, the block is 4 words.
  4. As with direct-mapped caches having multi-word blocks, we again use the word-within-block bits to choose the proper word. In the diagram this is performed by the very bottom multiplexor, using the word-within-block bits as the selector line.
  5. Our description and picture of multi-word block, direct-mapped caches is here, and our description and picture of single-word block, set-associative caches is just above. It is useful to compare those two picture with the one on the right to see how the concepts are combined.
  6. Below we give a more detailed discussion of which bits of the memory address are used for which purpose.

Choosing Which Block to Replace

When an existing block must be replaced, which victim should we choose? We ask the exact same question (with different words) when we study demand paging in 202.

How Big Is a Cache?

There are two notions of size.

Definition: The cache size is the capacity of the cache.

Another size is is the total number of bits in the cache, which includes tags and valid bits. For the example above with a 4-way associative, 1-word block cache, this size is computed as follows.

For this cache, what fraction of the bits are user data?
Ans: 4KB / 55Kb = 32Kb / 55Kb = 32/55.

Calculate in class the equivalent fraction for the last diagrammed cache, having 4-word blocks (and still 4-way set associative).

Tag Size and Division of the Address Bits

We continue to assume a byte addressed machines with all references to a 4-byte word.

The 2 LOBs are not used (they specify the byte within the word, but all our references are for a complete word). We show these two bits in light blue. We continue to assume 32-bit addresses so there are 230 words in the address space.

Let us review various possible cache organizations and determine for each the tag size and how the various address bits are used. We will consider four configurations each a 16KB cache. That is the size of the data portion of the cache is 16KB = 4 kilowords = 212 words.

Direct Mapped, Block Size 1 (Word)
Direct Mapped, Block Size 8
4-Way Set Associative, Block Size 1
4-Way Set Associative, Block Size 8

On the board calculate, for each of the four caches, what is the overhead percentage.

Homework: Redo the four caches above with the size of the cache 64KB (instead of 16KB) determining the number of bits in each portion of the address as well as the overhead percentages.

Start Lecture #23

Reducing the Miss Penalty Using Multilevel Caches

Improvement: Multilevel caches

Modern high end PCs and workstations all have at least two levels of caches: A very fast, and hence not very big, first level (L1) cache together with a larger but slower L2 cache.

When a miss occurs in L1, L2 is examined and only if a miss occurs there is main memory referenced.

So the average miss penalty for an L1 miss is

    (L2 hit rate)*(L2 time) + (L2 miss rate)*(L2 time + memory time)
  

We are assuming that L2 time is the same for an L2 hit or L2 miss and that the main memory access doesn't begin until the L2 miss has occurred.

Example 1

Do this example on the board (a reasonably exam question, but a little long since it has so many parts).

Assume

Calculate

  1. How many instructions per second does this machine execute?
  2. How many instructions per second would this machine execute if the L2 cache was eliminated?
  3. How many instructions per second would this machine execute if all caches were eliminated?
  4. How many instructions per second would this machine execute if the L2 cache had a 0% miss rate (L1 as originally specified)?
  5. How many instructions per second would this machine execute if both L1 caches had a 0% miss rate?
Example 2

Our company's current product, has the following characteristics

  1. Calculate the MIPS rating of the current product.
  2. Marketing says we need to double the speed and engineering has decided to do this by adding a 20ns L2 cache (the current caches become L1). What L2 hit rate is needed.

Homework: Redo example 2 with a memory access time of 50ns.

5.4: Virtual Memory

I realize this material is covered in operating systems class (CSCI-UA.0202). I am just reviewing it here. The goal is to show the similarity to caching, which we just studied. Indeed, (the demand part of) demand paging is caching: In demand paging the memory serves as a cache for the disk, just as in caching the cache serves as a cache for the memory.

However, the names used are different and there are other differences as well.

Cache conceptDemand paging analogue
Memory blockPage
Cache linePage Frame (frame)
Block SizePage Size
TagNone (table lookup)
Word in blockPage offset
Valid bitValid bit
MissPage fault
HitNot a page fault
Miss ratePage fault rate
Hit rate1 - Page fault rate
Placement questionPlacement question
Replacement questionReplacement question
AssociativityNone (fully associative)

For both caching and demand paging, the placement question does not have serious performance implications since the items are fixed size (no first-fit, best-fit, buddy, etc) as are the slots into which they are placed.

The replacement question, in contrast, is quite important for performance. Indeed, we spend significant time discussing replacements strategies in 202. Approximations to LRU (least recently used) are popular for both caching and demand paging. However, cache approximations are very crude since miss processing must be very fast and cannot involve a long calculation.d

The cost of a page fault vastly exceeds the cost of a cache miss so it is worth while in paging to slow down hit processing to lower the miss rate. Hence demand paging is fully associative and uses a table to locate the frame in which the page is located.

The figure on the right and the one below it both indicate the translation of page numbers into frame numbers, the latter showing the involvement of the page table. Although both figures are worded in terms of demand paging they can be interpreted for caching as well by essentially changing the names of certain concepts and realizing that demand paging corresponds to the extreme of a fully-associative cache.

page-table

Write Through vs. Write Back

In this section, choose the first element of each parenthesized pair for caching, and choose the second for paging.

Question: On a (write hit / write to an in-memory page) should we write the new value through to (memory/disk) or just keep it in the (cache/memory) and write it back to (memory/disk) when the (cache-line/page) is replaced?

Translation Lookaside Buffer (TLB)

A TLB is a cache of the page table. It is there for the same reason as any cache, the page table is too big to access fast enough so we maintain a subset that can be accessed quickly and (we hope) has few misses.

Without a TLB, every memory reference in the program would require two memory references, one to read the page table and one to read the requested memory word.

This would be an unacceptable performance loss and hence a TLB is crucial for a system with paging.

For now, we ignore the cache and just look at the TLB, pages, and frames. The diagram on the right shows the three possibilities, color-coded to indicate their relative speeds.

Typical TLB parameter values

Integrating Virtual Memory (Paging), TLBs, and Caches

Real systems have TLBs, page tables, and caches. Since the caches are based on real memory addresses the cache can be accessed only after the TLB or page table has converted the virtual address (page number + offset) to the real address (frame number + offset). In some systems, caches are accessed by virtual address (page number + offset), but we will ignore this possibility.

The diagram on the right is based on the decstation 3100, which is perhaps the simplest possible design. The 3100 had the following parameter values.

Actions taken

  1. The page number is searched in the fully associative TLB
  2. If a TLB hit occurs, the frame number from the TLB together with the page offset gives the physical address. A TLB miss causes an exception to reload the TLB from the page table, which the figure does not show.
  3. The physical address is broken into a cache tag and cache index (plus a two bit byte offset that is not used for word references).
  4. If the reference is a write, just do it (i.e., update the cache and main memory) without checking for a cache hit. This simplification (doing the same action on a hit as on a miss) is possible because the cache is primitive as we discussed previously).
  5. For a read, if the tag located in the cache entry specified by the index matches the tag in the physical address, the referenced word has been found in the cache; i.e., we had a read hit.
  6. For a read miss, the cache entry specified by the index is fetched from memory, loaded in the cache (replacing whatever was in that slot, which is always clean due to the primitive nature of the cache), and returned to the processor to satisfy the request.

Hit/Miss possibilities


TLBPageCacheRemarks
hithithit Possible, but page table not checked on TLB hit, data from cache
hithitmiss Possible, but page table not checked, cache entry loaded from memory
hitmisshit Impossible, TLB references only in-memory pages
hitmissmiss Impossible, TLB references only in-memory pages
misshithit Possible, TLB entry loaded from page table, data from cache
misshitmiss Possible, TLB entry loaded from page table, cache entry loaded from memory
missmisshit Impossible, cache is a subset of memory
missmissmiss Possible, page fault brings in page, TLB entry loaded, cache loaded

Implementing Protection with Virtual Memory

Handling TLB Misses and Page Faults

Summary

Disk access are extremely expensive, which dictates many choices made for demand paging and explains why choices good for caching (where a miss costs a few 10s of a nanosecond), although valid choices for demand paging, are not good choices for the latter (where the miss penalty is several milliseconds). In particular, demand paging implementations make the following choices.

  1. The block size (called the page size for paging) is large (about 1000 words) to take advantage of spacial locality and thereby lower the all-important miss rate.
  2. Whereas caches have limited associativity (2-way to 8-way are common choices), demand paging is a fully-associative cache with thousands (or more) entries.
  3. Since a page fault is so expensive, elaborate techniques are used to decide which page to evict.
  4. Since disk writes are so expensive, dirty bits and write-back are employed, but not on the primitive 3100 shown above.

Matching

1. L1 cachea. Not a cache
2. L2 cacheb. A cache for a cache
3. Main memory     c. A cache for disks
4. TLBd. A cache for main memory
5. Page Tablee. A cache for page table entries

Do the following two problems in class.

  1. Match the elements on the left column with the phrase in the right column that best describes it.
  2. Consider a memory reference that is miss/miss/miss in the table above. In what order are the memory elements in the left column accessed?

5.5: A Common Framework for Memory Hierarchies

Typical Sizes and Costs


Feature Typical values
for L1 caches
Typical values
for L2 caches
Typical values
for demand paging
Typical values
for TLBs
Size16KB-64KB500KB-4MB 1GB-1TB256B-16KB
Block size 16B-64B64-128 4KB-64KB4B-32B
Miss penalty in clocks10-25100-1000 10M-100M10-1000
Miss rate2%-5%0.1%-2% 0.00001%-0.0001%0.01%-2%

Question 1: Where Can/Should the Block Be Placed?

This question has two parts.

  1. In what slot are we able to place the block.
  2. If several possible slots are available (i.e. unoccupied), which one should be used?

Question 2: How Is a Block Found?


AssociativityLocation methodComparisons Required
Direct mappedIndex1
Set AssociativeIndex the set, search among elements Degree of associativity
FullSearch all entries Number of entries
Separate lookup table0

The difference in sizes and costs for demand paging vs. caching, leads to different algorithms for finding the block. Demand paging always uses the bottom row with a separate table (page table) but caching never uses such a table.

Question 3: Which Block Should Be Replaced?

If no possible slots are available, which victim should be chosen?

I call this the replacement question and is much studied in demand paging.

Question 4: What Happens on a Write?

Write-Through

Write-Back

Write Miss Policy (Advanced)

The Three Cs: An Intuitive Model for Understanding the Behavior of Memory Hierarchies

5.6 Virtual Machines

5.7 Using a Finite-State machine to Control a Simple Cache

5.8 Parallelism and Memory Hierarchies: Cache Coherence

5.9 Advanced Material; Implementing Cache Controllers

5.10 Real Stuff: the AMD Opteron X4 (Barcelona) and Intel Nehalem Memory Hierarchies

Fallacies and Pitfalls

Concluding Remarks

Historical Perspective and Further Reading

Start Lecture #24

Remark: A second L2 cache example and a related homework problem were added. Do the example and assign the homework.

Chapter 6: Storage and Other I/O Topics.

6.1 Introduction

Peripherals are varied; indeed they vary widely in many dimensions, e.g., cost, physical size, purpose, capacity, transfer rate, response time, support for random access, connectors, and protocol.

Consider just transfer rate for the moment.

The text mentions three especially important characteristics which can be used to classify peripherals.

Performance Metrics

Probably the most important quality metric for I/O is not performance but how frequently is data irretrievably corrupted. We will soon discuss RAID, a technique to improve this metirc.

There are at least three ways to measure I/O performance

  1. How much data can be processed per second. This metric normally improves as the size of each request increases since there is normally a startup overhead for each request.
  2. How many operations can be performed per second. This metric normally degrades as the size of each request increases.
  3. How long does a single request take. This metric normally degrades as the size of each request increases.

Do not make the error of thinking that the 3rd metric is simply the reciprocal of the second. It takes the post office at least one day to deliver a letter from here to California, but I can send one every minute if I wish. This is another example of pipelining.

6.2 Dependability, Reliability, and Availability

A system alternates between two states of delivered service

  1. The service is delivered as specified.
  2. The service is not delivered as specified.

Transitioning from the first state to the second is called a failure. Transitioning from the second state to the first is called a restoration.

Reliability measures the length of time during which services is continuously delivered as expected.

An example reliability measure is mean time to failure (MTTF), which measures the average length of time that the system is delivering service as expected. Bigger values are better.

Another important measure is mean time to repair (MTTR), which measures how long the system is not delivering service as expected. Smaller values are better.

Finally we have mean time between failures (MTBF).
MTBF = MTTF + MTTR.

One might think that having a large MTBF is good, but that is not necessarily correct. Consider a system with a certain MTBF and simply have the repair center deliberately add an extra 1 hour to the repair time and poof the MTBF goes up by one hour!

6.3 Disk Storage

Devices are quite varied and their data rates vary enormously.

Show a real disk opened up and illustrate the components.

Disk Access Time

The time for a disk access has five components, of which we concentrate on the first three.

  1. Seek.
  2. Rotational latency.
  3. Transfer time.
  4. Controller overhead.
  5. Queuing delays.
Seek Time

Today seek times are typically 3-8ms on average. It takes longer to go all the way across the disk but it does not take twice as long to go twice as far (the head must accelerate, decelerate, and settle on the track).

How should we calculate the average?

Rotational Latency

Since disks have just one arm the average rotational latency is half the time of a revolution, and is thus determined by the RPM (revolutions per minute) of the disk.

Disks today spin at 5400-15,000 RPM; they used to all spin at 3600 RPM.

Calculate on the board the average rotational latency of a 3600 RPM disk.

Homework: What is the average rotational latency for a 5400 RPM disk, a 5400 RPM disk, a 10,000 RPM, and a 15,000 RPM disk.

Transfer Time

You might consider the other four times all overhead since it is the transfer time during which the data is actually being supplied.

The transfer rate is typically tens of MB per second, sometimes over 100MB/sec. Given the rate, which is determined by the disk in use, the transfer time is proportional to the length of the request.

Some manufacturers quote a much higher rate, but that is for cache hits. In addition to supplying data much sooner, the electronic cache can transfer data at a higher rate than the mechanical disk.

Consider a disk with a 5ns seek time, a transfer rate of 80MB/sec, and a rotational rate of 10,000 RPM. Calculate on the board how long it takes for a 1K block request to What overall transfer rate (bytes delivered / total time) was achieved.

Start Lecture #25

Remark: Assign the homework problem above about rotational latency and do section 6.2.

Homework: Consider a disk with a 6ns seek time, a transfer rate of 60MB/sec, and a rotational rate of 10,000 RPM. How long does a request for a 100K block require to complete? A 10MB block? What overall transfer rates (bytes delivered / total time) were achieved in each case.

Controler Time

Not much to say. It is typically small. We will use 0ms (i.e., ignore this time).

Queuing Delays

This can be the largest component, but we will ignore it since it is not a function of the architecture, but rather of the load and OS.

6.9 Parallelism and I/O: Redundant Arrays of Inexpensive Disks (RAID)

Remark: I am doing 6.9 now since it concerns disks.

The acronym RAID was coined by Patterson and his students to abbreviate for Redundant Array of Inexpensive Disks. Now it is often redefined as Redundant Array of Independent Disks.

RAID comes in several flavors often called levels.

Striping a.k.a. Interleaving

To increase performance, rather than reliability and availability, it is a good idea to stripe or interleave blocks across several disks. In this scheme block n is stored on disk n mod k, where k is the number of disks. The quotient n/k is called the stripe number. For example, if there are 4 disks, stripe number 0 (the first stripe) consists of block 0, which is stored on disk 0, block 1 stored on 1, block 2 stored on 2, and block 3 stored on 3. Stripe 1 (like all stripes in this example) also contains 4 blocks. The first one is block 4, which is stored on disk 0.

Striping is especially good if one is accessing full stripes in which case all the blocks in the stripe can be read or written concurrently.

No Redundancy (RAID 0)

The base, non-redundant, case from which the others are built. Typically the data is striped across several disks. Since RAID 0 has no redundancy, it offers no reliability advantage. It does permit large (multi-block) I/Os to use multiple disks and hence to finish faster.

Mirroring (RAID 1)

Two disks containing the same content.

Error Detecting and Correcting Code (RAID 2)

Often called ECC (error correcting code or error checking and correcting code). Widely used in RAM, not used in RAID.

Bit-Interleaved Parity (RAID 3)

Normally byte-interleaved or several-byte-interleaved. For most applications, RAID 4 is better.

Block-Interleaved Parity (RAID 4)

RAID 4 combines striping and parity. In addition to the k so-called data disks used in striping, one has a single parity disk that contains the parity of the stripe.

Consider all k data blocks in one stripe. Extend this stripe to k+1 blocks by including the corresponding block on the parity disk. The block on the parity disk is calculated as the bitwise exclusive OR of the k data blocks.

Thus a stripe contains k data blocks and one parity block, which is the exclusive OR of the data blocks.

The great news is that any block in the stripe, parity or data, is the exclusive OR of the other k. This means we can survive the failure of any one disk.

For example, let k=4 and let the data blocks be A, B, C, and D.

  1. If the parity disk fails, we can easily recreate it since, by definition, the parity block for this stripe is
          A ⊕ B ⊕ C ⊕ D
    which is the exclusive OR of the other blocks.
  2. If a data disk fails, we can again recreate it since, by the commutative and associative properties of XOR,
        A ⊕ B ⊕ C ⊕ the parity block = A ⊕ B ⊕ C ⊕ (A ⊕ B ⊕ C ⊕ D) = D
    and again the missing block is the exclusive OR of the remaining blocks.

Properties of RAID 4.

Distributed Block-Interleaved Parity RAID 5

Rotate the disk used for parity.

Again using our 4 data-disk example, we continue to put the parity for blocks 0-3 on disk 4 (the fifth disk) but rotate the assignment of which disk holds the parity block of different stripes. In more detail.

Raid 0, Raid 1, and Raid 5 are widely used.

P + Q Redundancy (RAID 6)

Gives more than single error correction at a higher storage overhead.

6.4 Flash Storage

Often called a solid-state disk, flash is the latest attempted gap-filler technology, i.e., a technology between RAM and conventional disks. Unlike most past efforts, this one has succeeded to some extent.

Flash is between DRAM and disks in both price and performance: it is cheaper and slower than DRAM, but more expensive and faster than disks. However, the minimal size disk is much larger than the minimal size flash and hence, for devices with a modest memory requirement, flash is cheaper than (as well as faster than) a disk.

Other advantages of flash over disks include lower power, smaller physical size, silence, and shock resistance. These are due to the semiconductor nature of flash implying that it has no moving parts.

Technically flash is a kind of EEPROM, an electrically erasable, programmable read-only memory. Like other EEPROM technologies, but unlike DRAM, flash retains the values stored when power is turned off, a crucial requirement for a disk replacement.

Another typical characteristic of EEPROMs shared by flash is a significantly limited lifetime with respect to writing. A given flash cell can be rewritten many thousands of times, but not millions of times. This is a serious limitation and solid state disks contain software that remap heavily used flash blocks to other flash cells, a technique called wear leveling.

There are two flavors of flash called NOR and NAND. The former is older technology, but has higher performance primarily because NAND can be read and written only in large blocks. NAND flash is increasingly popular due to is lower price ($4/GB in 2008, compared to $65/GB for NOR). Today (Dec, 2011) NAND is available for about $1/GB about 10 times the price per byte of a very large disk.

6.5: Connecting Processors, Memory, and I/O Devices

A bus is a shared communication link, using one set of wires to connect many subsystems.

Tri-state Drivers

Bus Basics

Synchronous vs. Asynchronous Buses

A synchronous bus is clocked.

An asynchronous bus is not clocked.

device-mem

Consider the situation pictured at right where a device receives an I/O request and then needs to retrieve some data from memory. We are using an asynchronous bus for the memory request and transfer. Recall that this means, neither the device nor the memory knows the speed of its partner and must be prepared for very long or essentially instantaneous responses.

Note that Ack is bidirectional. We must insure that both sides are never driving (outputting on) this line at the same time. You may think of Ack as two lines, one going in each direction but, in fact, one line is sufficient if tri-state drivers are used. A similar consideration applies to the Data Lines.

We describe below the protocol used between the device and the memory and illustrate on the right a finite state machine used to manage this interactions.

The system is initialized with the memory in the top right state and the device in the top left state. Ack is not asserted by either side. ReadReq, DataRdy, and NewReq are also deasserted.

At some point an external entity (likely the CPU) raises NewReq. Events then proceed as follows.

  1. The device makes a request (asserts ReadReq and puts the desired address on the data lines). The name data lines sounds odd since it is (now) being used for the address. It will also be used for the data itself in this design. Data lines should be contrasted with control lines (such as ReadReq).
  2. Memory, which has been waiting, sees ReadReq, records the address and asserts Ack.
  3. The device waits for the Ack; once seen, it drops the data lines and deasserts ReadReq.
  4. The memory waits for the request line to drop. Then it can drop Ack (which it knows the device has now seen). The memory now at its leasure puts the data on the data lines (which it knows the device is not driving) and then asserts DataRdy. (DataRdy has been deasserted until now).
  5. The device has been waiting for DataRdy. It detects DataRdy and records the data. It then asserts Ack indicating that the data has been read.
  6. The memory sees Ack and then deasserts DataRdy and releases the data lines.
  7. The device seeing DataRdy low deasserts Ack ending the show. Note that both sides are prepared for another performance.

The Buses and Networks of the Pentium III

For a realistic example, on the right is a diagram adapted from the 25 October 1999 issue of Microprocessor Reports on a then brand new Intel chip set, the so called 840.

Bus adaptors have a variety of names, e.g. host adapters, hubs, bridges. The memory controller hub is often call the north bridge and the I/O controller hub is often called the south bridge.

Bus lines (i.e., wires) include those for data, function codes, device addresses. Data and address are considered data and the function codes are considered control (remember our datapath for MIPS).

Address and data may be multiplexed on the same lines (i.e., first send one then the other) or may be given separate lines. One is cheaper (good) and the other has higher performance (also good). Which is which?
Ans: the multiplexed version is cheaper.

Improving Bus Performance

These improvements mostly come at the cost of increased expense and/or complexity.

Start Lecture #26

Obtaining bus access

Cost Performance Tradeoffs


OptionHigh performanceLow cost
bus widthseparate addr and data lines multiplex addr and data lines
data widthwidenarrow
transfer sizemultiple bus loads single bus loads
bus mastersmultiplesingle
clockingsynchronousasynchronous

Do on the board the following example. Given

Find

  1. Sustained bandwidth and latency for reading 256 words using 4 word transfers.
  2. Sustained bandwidth and latency for reading 256 words using 16 word transfers.
  3. How many bus transactions per sec for each (a transaction includes both address and data.

Solution with four word blocks.

Solution with sixteen word blocks

Homework: Redo the last example but do not permit transmitting data to overlap reading more data.

6.6: Interfacing I/O Devices to the Processor, Memory, and Operating System

This is an I/O issue and is taught in 202.

Giving commands to I/O Devices

This is really an OS issue. Must write/read to/from device registers, i.e. must communicate commands to the controller. Note that a controller normally contains a microprocessor, but when we say the processor, we mean the central processor not the one on the controller.

Communicating with the Processor

Should we check periodically or be told when there is something to do? Better yet can we get someone else to do it since we are not needed for the job?

Polling

Processor continually checks the device status to see if action is required.

Do on the board the example on pages 676-677

Interrupt driven I/O

Processor is told by the device when to look. The processor is interrupted by the device.

Do on the board the example on pages 681-682.

Direct Memory Access (DMA)

The processor initiates the I/O operation then something else takes care of it and notifies the processor when it is done (or if an error occurs).

More Sophisticated Controllers

Subtleties involving the memory system

Start Lecture #27

6.7 I/O Performace Measures: Examples from Disk and File Systems

Transaction Processing I/O Benchmarks

File System and Web I/O Benchmarks

I/O Performance versus Processor Performance

We do an example to illustrate the increasing impact of I/O time.

Assume

  1. A job currently takes 100 seconds of CPU time and 50 seconds of I/O time.
  2. The CPU and I/O times can not be overlapped. Thus the total time required is 150 seconds.
  3. The CPU speed increases at a rate of 40% per year. This implies that the CPU time required in year n+1 is (1/1.4) times the CPU time required in year n.
  4. The I/O speed increases at a rate of 10% per year.

Calculate

  1. The CPU, I/O, and overall time required after 1,2,5,10,20 years.
  2. The percentage of the job time that the CPU is active for each year.
  3. The CPU, I/O, and overall speedup for each year.

Homework: Redo the above example assuming that CPU and I/O activity can be overlapped, i.e., assume the overall time is MAX(CPU,I/O) rather than SUM(CPU,I/O)?

6.8: Designing an I/O system

buses

Recall the picture on the right. When we are dealing with disks, the bus adapters between the backplane bus and the various I/O buses are called disk controllers. On each of those I/O buses, one would find disks.

Assume a system with the following characteristics is executing a workload of 64KB reads with 100K instructions between reads..

Find

  1. The maximum I/O rate achievable.
  2. How many controllers are needed for this rate?
  3. How many disks are needed for this rate?

Solution

Remark: The above analysis was very simplistic. It assumed everything overlapped just right and the I/Os were not bursty and that the I/Os conveniently spread themselves accross the disks.

Homework: Redo the above with the following parameters (more reflective of 2011 technology). Parameters not mentioned should be given the values in the example and your work should make the same simplistic assumptions that were made in the analysis.

Start Lecture #28

6.9 Parallelism and I/O: Redundant Arrays of Inexpensive Disks (RAID)

Done right after 6.3

6.10 Real Stuff: Sun Fire x4150 Server

6.11 Advanced Topics: Networks

6.12 Fallacies and Pitfalls

6.13 Concluding Remarks

Remark: Review of practice final.
Good luck on the (real) final!