ILP: Software Approaches

1. ENGS 116 Lecture 10 ILP: Software Approaches Vincent H. Berk October 12th Reading for today: 3.7-3.9, 4.1 Reading for Friday: 4.2 – 4.6 Homework #2: due Friday 14th, 2.8, A.2, A.13, 3.6a&b, 3.10, 4.5, 4.8, (4.13 optional)

2. ENGS 116 Lecture 10 Basic Loop Unrolling for (i=1000; i>0; i=i-1) x[i] = x[i] + s; Loop: LD F0, 0(R1) ; F0=array element ADDD F4, F0, F2 ; add scalar in F2 SD 0 (R1), F4 ; store result SUBI R1, R1, #8 ; decrement pointer 8 bytes (DW) BNEZ R1, Loop ; branch R1! = zero NOP ; delayed branch slot

3. ENGS 116 Lecture 10 FP Loop Hazards Loop: LD F0, 0(R1) ; F0=vector element ADDD F4, F0, F2 ; add scalar in F2 SD 0 (R1), F4 ; store result SUBI R1, R1, #8 ; decrement pointer 8 bytes (DW) BNEZ R1, Loop ; branch R1! = zero NOP ; delayed branch slot Where are the stalls?

4. ENGS 116 Lecture 10 FP Loop Showing Stalls Rewrite code to minimize stalls?

5. ENGS 116 Lecture 10 Revised FP Loop Minimizing Stalls Can we unroll the loop to make it faster?

6. ENGS 116 Lecture 10 Loop Unrolling • Short loop minimizes parallelism, induces significant overhead • Branches per instruction is high • Replicate the loop body several times and adjust the loop termination code for (i = 0; i < 100; i = i + 4) { x[i] = x[i] + y[i]; x[i + 1] = x[i + 1] + y[i + 1]; x[i + 2] = x[i + 2] + y[i + 2]; x[i + 3] = x[i + 3] + y[i + 3]; } • Improves scheduling since instructions from different iterations can be scheduled together • This is done very early in the compilation process • All dependences have to be found beforehand • Need to use different registers for each iteration

7. ENGS 116 Lecture 10 Where are the control dependences? 1 Loop: LD F0, 0 (R1) 2 ADDD F4, F0, F2 3 SD 0 (R1), F4 4 SUBI R1, R1, #8 5 BEQZ R1, exit 6 LD F0, 0 (R1) 7 ADDD F4, F0, F2 8 SD 0 (R1), F4 9 SUBI R1, R1, #8 10 BEQZ R1, exit 11 LD F0, 0 (R1) 12 ADDD F4, F0, F2 13 SD 0 (R1), F4 14 SUBI R1, R1, #8 15 BEQZ R1, exit ....

8. ENGS 116 Lecture 10 Data Dependences 1 Loop: LD F0, 0 (R1) 2 ADDD F4, F0, F2 3 SD 0 (R1), F4 ; drop SUBI & BNEZ 4 LD F0, –8 (R1) 2 ADDD F4, F0, F2 3 SD –8 (R1), F4 ; drop SUBI & BNEZ 7 LD F0, –16 (R1) 8 ADDD F4, F0, F2 9 SD –16 (R1), F4 ; drop SUBI & BNEZ 10 LD F0, –24 (R1) 11 ADDD F4, F0, F2 12 SD –24 (R1), F4 13 SUBI R1, R1, #32 ; alter to 4*8 14 BNEZ R1, LOOP 15 NOP

9. ENGS 116 Lecture 10 Name Dependences 1 Loop: LD F0, 0 (R1) 2 ADDD F4, F0, F2 3 SD 0 (R1), F4 ; drop SUBI & BNEZ 4 LD F6, –8 (R1) 5 ADDD F8, F6, F2 6 SD –8 (R1), F8 ; drop SUBI & BNEZ 7 LD F10, –16 (R1) 8 ADDD F12, F10, F2 9 SD –16 (R1), F12 ; drop SUBI & BNEZ 10 LD F14, –24 (R1) 11 ADDD F16, F14, F2 12 SD –24 (R1), F16 13 SUBI R1, R1, #32 ; alter to 4*8 14 BNEZ R1, LOOP 15 NOP Register renaming

10. ENGS 116 Lecture 10 Unroll Loop Four Times Rewrite loop to minimize stalls? 15 + 4  (1+2) +1 = 28 clock cycles to initiate, or 7 per iteration Assumes R1 is multiple of 4

11. ENGS 116 Lecture 10 Unrolled Loop That Minimizes Stalls • What assumptions were made when we moved code? • OK to move store past SUBI even though SUBI changes the register • OK to move loads before stores: get right data? • When is it safe for compiler to do such changes? Can we eliminate the remaining stall? 14+1=15 clock cycles, or 3.75 per iteration

12. ENGS 116 Lecture 10 Compiler Loop Unrolling • Most important: Code Correctness • Unrolling produces larger code that might interfere with cache: • Code sequence no longer fits in L1 cache • Cache to memory bandwidth might not be wide enough • Compiler must understand hardware: • Enough registers must be available OR • Compiler must rely on hardware register renaming • Compiler must understand the code: • Determine that loop iterations are independent • Eliminate branch instructions while preserving correctness • Determine that the LD and SD are independent over the loop • Rescheduling of instructions and adjusting the offsets

13. ENGS 116 Lecture 10 Superscalar Example • Superscalar: • Our system can issue one floating point and one other (non-floating point) instruction per cycle. • Instructions are dynamically scheduled from the window • Unroll the loop 5 times and reschedule to minimize cycles per iteration. (WHY?) • While Integer/FP split is simple for the HW, get CPI of 0.5 only for programs with: • Exactly 50% FP operations • No hazards • If more instructions issued at same time, greater difficulty in decode and issue • Even 2-way scalar  examine 2 opcodes, 6 register specifiers, & decide if 1 or 2 instructions can issue

14. ENGS 116 Lecture 10 Loop Unrolling in Superscalar Integer instruction FP instruction Clock cycle Loop: LD F0, 0 (R1) 1 LD F6, –8 (R1) 2 LD F10, –16 (R1) ADDD F4, F0, F2 3 LD F14, –24 (R1) ADDD F8, F6, F2 4 LD F18, –32 (R1) ADDD F12, F10, F2 5 SD 0 (R1), F4 ADDD F16, F14, F2 6 SD –8 (R1), F8 ADDD F20, F18, F2 7 SD –16 (R1), F12 8 SUBI R1, R1, #40 9 SD 16 (R1), F16 10 BNEZ R1, Loop 11 SD 8 (R1), F20 12 • Unrolled 5 times to avoid delays (+ 1 due to SS) • 12 clocks to initiate, or 2.4 clocks per iteration

15. ENGS 116 Lecture 10 VLIW Example • VLIW: • 5 instructions in one very long instruction word. • 2 FP, 2 Memory, 1 branch/integer • Compiler avoids hazards • Not all slots are always full • VLIW: tradeoff instruction space for simple decoding • The long instruction word has room for many operations • By definition, all the operations the compiler puts in the long instruction word are independent  execute in parallel • E.g., 2 integer operations, 2 FP ops, 2 memory refs, 1 branch  16 to 24 bits per field  7*16 or 112 bits to 7*24 or 168 bits wide • Need compiling technique that schedules across several branches

16. ENGS 116 Lecture 10 Loop Unrolling in VLIW Memory Memory FP FP Int. op/ Clockreference 1 reference 2 operation 1 op. 2 branch LD F0, 0 (R1) LD F6, –8 (R1) 1 LD F10, –16 (R1) LD F14, –24 (R1) 2 LD F18, –32 (R1) LD F22, –40 (R1) ADDD F4, F0, F2 ADDD F8, F6, F2 3 LD F26, –48 (R1) ADDD F12, F10, F2 ADDD F16, F14, F2 4 ADDD F20, F18, F2 ADDD F24, F22, F2 5 SD 0 (R1), F4 SD –8 (R1), F8 ADDD F28, F26, F2 6 SD –16 (R1), F12 SD –24 (R1), F16 7 SD –32 (R1), F20 SD –40 (R1), F24 SUBI R1, R1, #48 8 SD 0 (R1), F28 BNEZ R1, LOOP 9 Unrolled 7 times to avoid delays 9 clocks to initiate, or 1.3 clocks per iteration Average: 2.5 ops per clock, 50% efficiency Note: Need more registers in VLIW (15 vs. 6 in SS)

17. ENGS 116 Lecture 10 Limits to Multi-Issue Machines • Inherent limitations of instruction-level parallelism • 1 branch in 5: How to keep a 5-way VLIW busy? • Latencies of units: many operations must be scheduled • Easy: More instruction bandwidth • Easy: Duplicate functional units to get parallel execution • Hard: Increase ports to register file (bandwidth) • VLIW example needs 7 reads and 3 writes for integer registers & 5 reads and 3 writes for FP registers • Harder: Increase ports to memory (bandwidth) • Pipelines in lockstep: • One pipeline stall, stalls all others to avoid hazards

18. ENGS 116 Lecture 10 Limits to Multi-Issue Machines • Limitations specific to either superscalar or VLIW implementation • Decode issue in superscalar: how wide is practical? • VLIW code size: unroll loops + wasted fields in VLIW • IA-64 compresses dependent instructions, but still larger • VLIW lock step  1 hazard & all instructions stall • IA-64 not lock step? Dynamic pipeline? • VLIW & binary compatibility: IA-64 promises binary compatibility

19. ENGS 116 Lecture 10 Dependences • Two instructions are parallel if they can execute simultaneously in a pipeline without causing any stalls (assuming no structural hazards) and can be reordered (depending on code semantics) • Two instructions that are dependent are not parallel and cannot be reordered • Types of dependences • Data dependences • Name dependences • Control dependences • Dependences are properties of programs • Hazards are properties of the pipeline organization • Dependence indicates the potential for a hazard

20. ENGS 116 Lecture 10 Compiler Perspectives on Code Movement • Hard for memory accesses • Does 100(R4) = 20 (R6)? • From different loop iterations, does 20(R6) = 20(R6)? • Our example required compiler to know that if R1 doesn’t change then: 0(R1)  -8 (R1)  -16 (R1)  -24 (R1) There were no dependences between some loads and stores so they could be moved by each other

21. ENGS 116 Lecture 10 Detecting Loop Level Dependences for (i=1; i<=100; i=i+1) { A[i] = A[i] + B[i]; /* S1 */ B[i+1] = C[i] + D[i]; /* S2 */ } • Loop carried dependence: • S1 relies on the S2 of the previous iteration • There is no dependence between S1 and S2, consider: 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];

22. ENGS 116 Lecture 10 Dependence Distance for (i=6; i<=100; i=i+1) Y[i] = Y[i-5] + Y[i]; • Loop carried dependence in the form of a recurrence of Y • Dependence distance of 5 • Higher dependence distance allows for more ILP

23. ENGS 116 Lecture 10 Greatest Common Divisor test • Affine array indices: • All array indices DIRECTLY depend on loop variable i • Assume the code properties: • for loop runs from n to m with index i • loop has an access pattern: X [a * i +b] = X [c * i +d] … • two values for i: j and k both between n and m • store indexed by j and a load later on index by k with: a*j+b = c*k+d • A loop carried dependence exists if GCD (c,a) must divide (d-b) • a=2, b=3, c=2, d=0 GDC(a,c) = 2 and d-b = -3 • There is no loop dependence since 2 does not divide -3 for (i=1; i<=100; i=i+1) X[2*i+3] = X[2*i] * 5.0;

24. ENGS 116 Lecture 10 Problem Cases • Reference by pointers instead of array indices • partly eliminated by strict type checking • Sparse arrays with indexing through other arrays (similar to pointers) • When a dependence exists for values of the indices but those values are never reached • The loop-carried dependence has a distance far greater than what loop-unrolling would cover

25. Iteration 0 Iteration Iteration 1 Iteration 2 Iteration 3 4 pipelined ENGS 116 Lecture 10 Software Pipelining • Observation: if iterations from loops are independent, then can get more ILP by taking instructions from different iterations • Software pipelining: reorganizes loops so that each iteration is made from instructions chosen from different iterations of the original loop Software- iteration

26. Before: Unrolled 3 times After: Software Pipelined ENGS 116 Lecture 10 SW Pipelining Example 1 LD F0, 0 (R1) LD F0, 0 (R1) 2 ADDD F4, F0, F2 ADDD F4, F0, F2 3 SD 0 (R1), F4 LD F0, –8 (R1) 4 LD F6, –8 (R1) 1 SD 0 (R1), F4 Stores M[i] 5 ADDD F8, F6, F2 2 ADDD F4, F0, F2 Adds to M[i-1] 6 SD –8, (R1), F8 3 LD F0, –16 (R1) Loads M[i-2] 7 LD F10, –16 (R1) 4 SUBI R1, R1, #8 8 ADDD F12, F10, F2 5 BNEZ R1, LOOP 9 SD –16 (R1), F12 SD 0 (R1), F4 10 SUBI R1, R1, #24 ADDD F4, F0, F2 11 BNEZ R1, LOOP SD –8 (R1), F4 Read F4Read F0 SD IF ID EX Mem WB Write F4 ADD IF ID EX Mem WB LD IF ID EX Mem WB Write F0

27. ENGS 116 Lecture 10 SW Pipelining Example Symbolic Loop Unrolling • Smaller code space • Overhead paid only once vs. each iteration in loop unrolling 100 iterations = 25 loops with 4 unrolled iterations each Software Pipelining Number of overlapped operations (a) Software pipelining Time Loop Unrolling Number of overlapped operations Time (b) Loop unrolling