1. Loop Unrolling & Predication CSE 820

2. Michigan State University Computer Science and Engineering Software Pipelining With software pipelining a reorganized loop contains instructions from different iterations of the original loop. Sometimes called symbolic loop unrolling.

3. Michigan State University Computer Science and Engineering Software Pipelined Loop

4. Michigan State University Computer Science and Engineering Unrolled Loop select subset of each iteration (bold) Iteration 1: L.D F0,0 (R1) ADD.D F4, F0, F2 S.D F4, 0 (R1) Iteration 2: L.D F0,0 (R1) ADD.D F4, F0, F2 S.D F4, 0 (R1) Iteration 3: L.D F0,0 (R1) ADD.D F4, F0, F2 S.D F4, 0 (R1)

5. Michigan State University Computer Science and Engineering Software Pipelining Loop: S.D F4, 16 (R1); stores into M[i] ADD.D F4, F0, F2 ; adds to M[i-1] L.D F0,0 (R1) ; loads M[i-2] DADDUI R1, R1, # -8 BNE R1, R2, Loop Requires start-up and clean-up.

6. Michigan State University Computer Science and Engineering Symbolic Loop Unrolling Software pipelining can be thought of as symbolic loop unrolling, but has the advantage of generating less code.

7. Michigan State University Computer Science and Engineering Software Pipelining has less overhead

8. Michigan State University Computer Science and Engineering Global Code Scheduling allows moving instructions across branches Most techniques concentrate on determining a Straight-line code segment representing the most frequently executed code

9. Michigan State University Computer Science and Engineering Trace Scheduling Concept 1.Guess the likely path through branches (called the trace) 2.Trace now contains long stretches of code without taken branches (predicted) 3.Schedule the trace allowing movement across branches Add code to off-the-trace to undo the effects of movement The increased ability to move across branches should improve scheduling

10. Michigan State University Computer Science and Engineering Movement + Undo Consider if (cond) then { x=x + 5; // likely } else // unlikely After Movement x = x + 5; if (cond) then { // likely} else { x = x – 5; // unlikely} // undo

11. Michigan State University Computer Science and Engineering Select a trace

12. Michigan State University Computer Science and Engineering Trace showing jumps off the trace

13. Michigan State University Computer Science and Engineering Superblocks Avoid the multiple entry and exits of traces. Superblock has one entry and multiple exits which makes scheduling easier. The one-entry-multiple-exit is achieved by duplicating code where the unlikely path exits the trace so that no reentry is needed.

14. Michigan State University Computer Science and Engineering Superblock: one entry and multiple exits

15. Michigan State University Computer Science and Engineering Predicated Instructions Requires –Hardware –ISA modification Predicated instructions eliminate branches, converting a control dependence into a data dependence. IA-64 has predicated instructions, but many existing ISA contain at least one (the conditional move).

16. Michigan State University Computer Science and Engineering Conditional Move if (R1 == 0) R2 = R3; Branch: BNEZ R1,L ADDU R2, R3, R0 L: Conditional Move: CMOVZ R2, R3, R1 In a pipeline, the control dependence at the beginning of the pipeline is transformed into a data dependence at the end of the pipeline.

17. Michigan State University Computer Science and Engineering Full Predication Every instruction has a predicate: if the predicate is false, it becomes a NOP. It is particularly useful for global scheduling since non-loop branches can be eliminated: the harder ones to schedule.

18. Michigan State University Computer Science and Engineering Exceptions & Predication A predicated instruction must not be allowed to generate an exception, if the predicate is false.

19. Michigan State University Computer Science and Engineering Implementation Although predicated instructions can be annulled early in the pipeline, annulling during commit delays annulment until later so data hazards have an opportunity to be resolved. The disadvantage is that resources such as functional units and registers (rename or other) are used.

20. Michigan State University Computer Science and Engineering Predication is good for… Short alternative control flow Eliminating some unpredictable branches Reducing the overhead of global scheduling But the precise rules for compilation are still being determined.

21. Michigan State University Computer Science and Engineering Limitations Annulled instructions waste resources: registers, functional units, cache & memory bandwidth If predicate condition cannot be separated from the instruction, a branch might have had better performance, if it could have been accurately predicted.

22. Michigan State University Computer Science and Engineering Limitations (con’t) Predication across multiple branches can complicate control and is undesirable unless hardware supports it (as in IA-64). Predicated instructions may have a speed penalty—not the case when all instructions are predicated.

23. Michigan State University Computer Science and Engineering Example if (A==0) A=B; else A= A+4; LDR1,0(R3);load A BNEZR1,L1;test A LDR1,0(R2);then clause JL2;skip else L1:DADDIR1,R1,#4 ;else clause L2:SDR1,0(R3);store A

24. Michigan State University Computer Science and Engineering Hoist Load if (A==0) A=B; else A= A+4; LDR1,0(R3) ;load A LDR14,0(R2) ;speculative load B BEQZR1,L3 ;other branch of if DADDIR14,R1,#4 ;else clause L3:SDR14,0(R3) ;store A What if speculative load raises an exception?

25. Michigan State University Computer Science and Engineering Guard if (A==0) A=B; else A= A+4; LDR1,0(R3);load A sLDR14,0(R2);speculative load BNEZR1,L1;test A SPECCK0(R2);speculative check JL2;skip else L1:DADDIR14,R1,#4 ;else clause L2:SDR14,0(R3);store A sLD does not raise certain exceptions; leaves them for SPECCK (IA-64).

26. Michigan State University Computer Science and Engineering Other exception techniques Poison bit: –applied to destination register. –set upon exception –raise exception upon access to poisoned register.

27. Michigan State University Computer Science and Engineering Hoist Load above Store If memory addresses are known, a load can be hoisted above a store. If not, … add a special instruction to check addresses before the loaded value is used. (It is similar to SPECCK shown earlier: IA-64)

28. Michigan State University Computer Science and Engineering Speculation: soft vs. hard must be able to disambiguate memory (to hoist loads past stores), but at compile time information is insufficient hardware works best when control flow is unpredictable and when hardware branch prediction is superior exception handling is easier in hardware trace techniques require compensation code compilers see further for better scheduling

29. Michigan State University Computer Science and Engineering IA-64