CS355 Sylabus

The Direct Memory Access device

The DMA is a simple special purpose (i.e., a specialized - can only do one specific task) device.
It can be manufactured very inexpensively.
It's sole purpose is to transfer data between the memory and an IO device.
You have to remember that the DMA is a helper processor for the CPU. Thus: the CPU must be able to tell the DMA what data to transfer. The only way that the CPU can do that is when the DMA is accessible on the system bus
It is therefore of no surprise that the DMA is connected onto the system bus....

The internal structure of a DMA is as follows:

The structure resembles somewhat the structure of an IO device (with the command, data and status registers), but it has two additional registers that allow the DMA to perform the data transfer operation.

The registers in the DMA are used as follows:

Status register: readable by the CPU to determine the status of the DMA device (idle, busy, etc)
Command register: writable by the CPU to issue a command to the DMA
Data register: readable and writable. It is the buffering place for data that is being transfered between the memory and the IO device.
Address register: contains the starting location of memory where from or where to the data will be transfered.
The Address register must be programmed by the CPU before issueing a "start" command to the DMA.
Count register: contains the number of bytes that need to be transfered.

The information in the command register, address register and in the count register combined specify exactly what the DMA will do:

If the command in the command register is write (= write to memory), then:

The DMA will transfer count number (in the Count register) of bytes from the IO device to the memory starting at memory address Address (in the Address register)

If the command in the command register is read (= read from memory), then:

The DMA will transfer count number (in the Count register) of bytes from the memory starting at memory address Address (in the Address register) to the IO device

The IO communication using a DMA proceeds as follows:

The CPU first programs the DMA's address register and count register

The CPU then write a read or a write command into the DMA's command register

Let's assume the CPU wants to perform an IO write operation that transfers data from the memory to the IO device....
(for an IO read operation (reading data from an IO device), the process is similar)

After the CPU has issued a write command, the DMA become active and will repeatedly execute the following steps:

Acquire (through the bus arbiter) the right to use the system bus
Send out a read request for address given by the value in the address register
Go through the system bus cycle (see click here) to obtain a word from memory into DMA's data register.
Write the data from the data register to the actual IO device
Increment the address register
Decrement the count register
Repeat as long as count register is not zero

After the CPU has programmed DMA with the necessary information to perform the transfer, and started the DMA, the DMA will perform the transfer for the CPU.
So you clearly see that the DMA is a helper device to the CPU.

Let me give you an example of IO communication using a DMA device first before going into the more difficult issues surrounding IO communication with an DMA device....

First, I have to make some assumptions on how the DMA is connected to the system bus:

I will again use M68000 to illustrate with memory mapped IO

These are the registers of the terminal0 DMA device in Emacsim:

Command register: address 1400 (Hex)
Count register: address 1404 (Hex)
Address register: address 1408 (Hex)
Status register: address 140C (Hex)
Data register: internal used only (for transfer by DMA)

The following is a assembler code fragment that will instruct the DMA to transfer 1000 bytes of data from memory at the variable buffer to the IO device:

(#buffer = starting address of the data in memory) move.l #buffer, $1408 // Setup the Address register of DMA move.l #1000, $1404 // Setup the Count register of DMA move.l #1, $1400 // Issue print command to IO device // operation code 1 means "print"

After these 3 simple move assembler instructions, the CPU has programmed the DMA and started the DMA to perform the IO transfer operation.

OK, now you know what must be done to get the DMA going.

The 6-million dollar (i.e., very important) question that remains is:

What is the CPU gonna do next ???

Answers:

Wait until the I/O operation has completed and then continue with the program execution
Continue with the program execution while the DMA transfer the data (in the I/O operation

Option 1: wait until the I/O operaton completes and then continue with the program execution

Example: the following program prints "Hello World" using the DMA and waits before it continues with the rest of the execution

xdef Start, Stop, End Start: * .... * .... Program before IO operation * .... * Print "Hello World" by programming the DMA move.l #-1, $140C ; Write a non-zero value in status register move.l #11, $1404 ; Program the count register in DMA move.l #Str,$1408 ; Program the address register in DMA move.l #1, $1400; ; Issue the Write command to DMA * Wait until the IO operation is completed wait: move.l $140C, d0 cmp.l #-1, d0 ; Test if status has changed to ready beq wait ; Repeat testing until status changes * .... * .... Program after IO operation * .... Stop: nop nop nop X: dc.b 'Hello World' End: end

Example Program: (Demo above code)

Program file: /home/cs355001/demo/Programmed-IO/m68000-print.s Compile with: /bin/cs355001/bin/as255 m68000-print Run with: /bin/cs355001/bin/m68000 (load m68000-print)

Fact:

Option 1 is a terrible solution

Reason:

We wanted to free up the CPU so that the CPU can do something useful
After starting the IO operation, the CPU is waiting and and does nothing useful !!!
(See the wait loop that the program executes ???)

Option 2: continue executing with the rest of the program while the DMA transfers the data

$64,000 question:

Answer:

No !!!
Often, a program must wait for the IO operation to complete before it can continue !!!

Example: program that read in 1000 grade values and computes its average:

main() { int grades[1000]; int i; double sum, average; read(grades, grade-file, 1000); // Start DMA to do data transfer // The DMA will fill the grades[ ] "slowly" // Suppose program continues execution WHILE DMA is performing the IO transfer.... sum = 0; for (i = 0; i < 1000; i++) sum = sum + grades[i]; // the grades[ ] value is not yet available !!! average = sum/1000; }

If the program preceeds without waiting with the program segment following the read() subroutine, it will use incorrect values (because the data has not yet been transfered into the "grades" array in memory) !!!

The simplest solution to the IO operation problem is busy waiting (see above program):

* Wait until the IO operation is completed wait: move.l $140C, d0 cmp.l #-1, d0 ; Test if status has changed to ready beq wait ; Repeat testing until status changes

This solution is clearly unattractive:

we are again making the very expensive CPU do nothing at all
(the loop accomplishes absolutely nothing useful).

In fact, the solution is dumb... why would you want to make an extra helper device (the DMA) for the CPU and then make the CPU wait for the DMA to finish ? You could just as well make the CPU do the transfer and you don't need an extra DMA device....

The real 6-million dollar question is:

How to to keep the CPU doing useful work ???

Clearly:

The CPU cannot continue with the currently running program !!!!

Answer:

There are other programs from other users in the computer !!!

More precisely:

Make the current program (that performed an IO operation) give up the CPU
Then make the CPU run another program
(This program must not be waiting on an IO operation to complete)
When the IO transfer operation is done , switch the CPU back to the original program.

To realise the solution, we need to solve 2 major problems:

How to make a program stop running in such a way that the program can be restarted at a later time from the same state as when it was stopped.
How can we make the stopped program start running again
I.e.: "revive" the stopped program so it will run again !!!
(The stooped program needs to reclaim the use of the CPU so that the CPU can execute its instructions)

From the point of view of the programs, the releasing and reclaiming of the CPU is the act of one program giving the control of the CPU to another program.

In Computer Science jargon, when two (or more) programs try to interact with one another, we say that the program must synchronize with one each other.

The technique to make them agree is called a synchronization method.