Downloaded 72 times
![34 This allows us to replace the loop above with the following code sequence, which makes possible overlapping of the iterations of the loop: a[1] = a[1] + b[1]; for (i=1; i<=99; i= i+1){ b[i+1] = c[i] + d[i]; a[i+1] = a[i+1] + b[i+1]; } b[101] = c[100] + d[100]; Example 3 for (i=1; i<=100; i= i+1){ a[i+1] = a[i] + c[i]; //S1 b[i+1] = b[i] + a[i+1]; //S2 } This loop is not parallel it has cycles in the dependencies, namely the statements S1 and S2 depend on themselves!](https://image.slidesharecdn.com/dealingwithexceptions-171231173625/75/Dealing-with-exceptions-Computer-Architecture-part-2-34-2048.jpg)
Uploaded byGaditek
Dealing with exceptions Computer Architecture part 2
The document discusses exceptions in computer architecture. It describes two types of exceptions - interrupts and traps. Interrupts are caused by external events like I/O requests, while traps are caused by internal events like arithmetic overflow. When an exception occurs, the pipeline must stop executing the offending instruction, save state like the program counter, and jump to an exception handler. Handling exceptions in a pipelined processor is challenging as it can disrupt instruction flow. The document outlines some of the techniques used to handle exceptions in a pipeline, like writing exception information to registers and flushing instructions after the exception.
Dealing with exceptions Computer Architecture part 2
- 1. Lecture 8 Dealing withExceptions Computer Architecture CNE-301 Lecturer: Irfan Ali 1
- 2.
- 3.
- 4.
- 5.
- 6.
- 7.
- 8. Two Types ofExceptions: Interrupts and Traps • Interrupts • Caused by external events: • Network, Keyboard, Disk I/O, Timer • Page fault - virtual memory • System call - user request for OS action • Asynchronous to program execution • May be handled between instructions • Simply suspend and resume user program • Traps • Caused by internal events • Exceptional conditions (overflow) • Undefined Instruction • Hardware malfunction • Usually Synchronous to program execution • Condition must be remedied by the handler • Instruction may be retried or simulated and program continued or program may be aborted 8
- 9. Synchronous vs Asynchronous •Definition: If the event occurs at the same place every time the program is executed with the same data and memory allocation, the event is synchronous. Otherwise asynchronous. • Except for hardware malfunctions, asynchronous events are caused by devices external to the CPU and memory. • Asynchronous events usually are easier to handled because asynchronous events can be handled after the completion of the current instruction. 9
- 10. Exceptions in Simplefive-stage pipeline • Instruction Fetch, & Memory stages • Page fault on instruction/data fetch • Misaligned memory access • Memory-protection violation • Instruction Decode stage • Undefined/illegal opcode • Execution stage • Arithmetic exception • Write-Back stage • No exceptions! 10
- 11. What happens duringan exception In The Hardware • The pipeline has to 1) stop executing the offending instruction in midstream, 2) let all preceding instructions complete, 3) flush all succeeding instructions, 4) set a register to show the cause of the exception, 5) save the address of the offending instruction, and 6) then jump to a prearranged address (the address of the exception handler code) In The Software • The software (OS) looks at the cause of the exception and “deals” with it • Normally OS kills the program 11
- 12. Exceptions Exception = non-programmedcontrol transfer • system takes action to handle the exception • must record the address of the offending instruction • record any other information necessary to return afterwards • returns control to user • must save & restore user state user program normal control flow: sequential, jumps, branches, calls, returns System Exception Handler Exception: return from exception 12 must record
- 13.
- 14.
- 15. 15 What Makes Pipelining Hard? Examplesof interrupts: • Power failing, • Arithmetic overflow, • I/O device request, • OS call, • Page fault Interrupts (also known as: faults, exceptions, traps) often require • surprise jump (to vectored address) • linking return address • saving of PSW-Program Status Word (including CCs- Condition Codes) • state change (e.g., to kernel mode) Interrupts cause great havoc! There are 5 instructions executing in 5 stage pipeline when an interrupt occurs: • How to stop the pipeline? • How to restart the pipeline? • Who caused the interrupt?
- 16. 16 What Makes Pipelining Hard? Interruptscause great havoc! What happens on interrupt while in delay slot ? • Next instruction is not sequential solution #1: save multiple PCs • Save current and next PC • Special return sequence, more complex hardware solution #2: single PC plus • Branch delay bit • PC points to branch instruction Stage Problem that causes the interrupt IF Page fault on instruction fetch; misaligned memory access; memory-protection violation ID Undefined or illegal opcode EX Arithmetic interrupt MEM Page fault on data fetch; misaligned memory access; memory-protection violation
- 17. 17 What Makes Pipelining Hard? •Simultaneous exceptions in more than one pipeline stage, e.g., • Load with data page fault in MEM stage • Add with instruction page fault in IF stage • Add fault will happen BEFORE load fault • Solution #1 • Interrupt status vector per instruction • Defer check until last stage, kill state update if exception • Solution #2 • Interrupt ASAP(as soon as possible) • Restart everything that is incomplete Another advantage for state update late in pipeline! Interrupts cause great havoc!
- 18.
- 19.
- 20. 20 What Makes Pipelining Hard? Here’swhat happens on a data page fault. 1 2 3 4 5 6 7 8 9 i F D X M W i+1 F D X M W < page fault i+2 F D X M W < squash i+3 F D X M W < squash i+4 F D X M W < squash i+5 trap > F D X M W i+6 trap handler > F D X M W Interrupts cause great havoc!
- 21. Exceptions - “StuffHappens” • Exceptions definition: “unexpected change in control flow” • Another form of control hazard. For example: add $1, $2, $1; causing an arithmetic overflow sw $3, 400($1); add $5, $1, $2; Invalid $1 contaminates other registers or memory locations! 21
- 22. Additions to MIPSISA to support Exceptions • EPC (Exceptional Program Counter) • A 32-bit register • Hold the address of the offending instruction • Cause • A 32-bit register in MIPS (some bits are unused currently.) • Record the cause of the exception • Status - interrupt mask and enable bits and determines what exceptions can occur. • Control signals to write EPC , Cause, and Status • Be able to write exception address into PC, increase mux set PC to exception address (MIPS uses 8000 00180hex ). • May have to undo PC = PC + 4, since want EPC to point to offending instruction (not its successor); PC = PC – 4 • What else? flush all succeeding instructions in pipeline 22
- 23. Flush instructions inBranch Hazard 36 sub $10, $4, $8 40 beq $1, $3, 7 # taget = 40 + 4 + 7*4 = 72 44 and $12, $2, $5 if($1==$2) 48 or $13, $2, $6 go(PC+4+PC*C) 52 …. …. 72 lw $4, 50($7) 23
- 24.
- 25. Pipelining in MIPS •MIPS architecture was designed to be pipelined • Simple instruction format (makes IF, ID easy) • Single-word instructions • Small number of instruction formats • Common fields in same place (e.g., rs, rt) in different formats • Memory operations only in lw, sw instructions (simplifies EX) • Memory operands aligned in memory (simplifies MEM) • Single value for Write-Back (limits forwarding) • Pipelining is harder in CISC architectures 25
- 26.
- 27. 27 Key observation: architectedstate only change in memory and register write stages.
- 28.
- 29.
- 30. Introduction to ILP •What is ILP? • Processor and Compiler design techniques that speed up execution by causing individual machine operations to execute in parallel • ILP is transparent to the user • Multiple operations executed in parallel even though the system is handed a single program written with a sequential processor in mind • Same execution hardware as a normal RISC machine • May be more than one of any given type of hardware
- 31.
- 32.
- 33.
- 34. 34 This allows usto replace the loop above with the following code sequence, which makes possible overlapping of the iterations of the loop: a[1] = a[1] + b[1]; for (i=1; i<=99; i= i+1){ b[i+1] = c[i] + d[i]; a[i+1] = a[i+1] + b[i+1]; } b[101] = c[100] + d[100]; Example 3 for (i=1; i<=100; i= i+1){ a[i+1] = a[i] + c[i]; //S1 b[i+1] = b[i] + a[i+1]; //S2 } This loop is not parallel it has cycles in the dependencies, namely the statements S1 and S2 depend on themselves!
- 35. 35 There are anumber of techniques for converting such loop-level parallelism into instruction-level parallelism. Basically, such techniques work by unrolling the loop. An important alternative method for exploiting loop-level parallelism is the use of vector instructions on a vector processor.