1. CPE 731 Advanced Computer Architecture ILP: Part II – Branch Prediction Dr. Gheith Abandah Adapted from the slides of Prof. David Patterson, University of California, Berkeley

2. 5/24/2015CPE 731, ILP2 2 Outline Static Branch Prediction Dynamic Branch Prediction Branch History Table Correlated Branch Prediction Tournament Predictors Branch Target Buffer

3. 5/24/2015CPE 731, ILP2 3 Static Branch Prediction We learned how to schedule code around delayed branch To reorder code around branches, need to predict branch statically when compile Simplest scheme is to predict a branch as taken –Average misprediction = untaken branch frequency = 34% SPEC More accurate scheme predicts branches using profile information collected from earlier runs, and modify prediction based on last run: IntegerFloating Point

4. 5/24/2015CPE 731, ILP2 4 Dynamic Branch Prediction Why does prediction work? –Underlying algorithm has regularities –Data that is being operated on has regularities –Instruction sequence has redundancies that are artifacts of way that humans/compilers think about problems Is dynamic branch prediction better than static branch prediction? –Seems to be –There are a small number of important branches in programs which have dynamic behavior Performance = ƒ(accuracy, cost of misprediction)

5. 5/24/2015CPE 731, ILP2 5 Branch History Table Use the lower k bits of the PC to address index the table Branch Address Branch History Table (BHT) 2 k entries Predict Taken or not Taken k

6. 5/24/2015CPE 731, ILP2 6 BHT, n=1 Problem: in a loop, 1-bit BHT will cause two mispredictions (avg is 9 iterations before exit): –End of loop case, when it exits instead of looping as before –First time through loop on next time through code, when it predicts exit instead of looping 0 1 Not taken Taken Not taken Taken Predict not taken Predict taken

7. 5/24/2015CPE 731, ILP2 7 Solution: 2-bit scheme where change prediction only if get misprediction twice Red: stop, not taken Green: go, taken Adds hysteresis to decision making process BHT, n=2 T T NT Predict Taken Predict Not Taken Predict Taken Predict Not Taken T NT T

8. 5/24/2015CPE 731, ILP2 8 BHT Accuracy Mispredict because either: –Wrong guess for that branch –Got branch history of wrong branch when index the table 4096 entry table: Integer Floating Point

9. 5/24/2015CPE 731, ILP2 9 Outline Static Branch Prediction Dynamic Branch Prediction Branch History Table Correlated Branch Prediction Tournament Predictors Branch Target Buffer

10. 5/24/2015CPE 731, ILP2 10 Correlated Branch Prediction Aka Two-Level Predictors Motivation 1.if (a==2) a=0; 2.if (b==2) b=0; 3.if (a != b) {…} Can predict that Condition (3) is not true if Conditions (1) and (2) are true.

11. 5/24/2015CPE 731, ILP2 11 Correlated Branch Prediction Idea: record m most recently executed branches as taken or not taken, and use that pattern to select the proper n-bit branch history table In general, (m,n) predictor means record last m branches to select between 2 m history tables, each with n-bit counters –Thus, old 2-bit BHT is a (0,2) predictor Global Branch History: m-bit shift register keeping T/NT status of last m branches. Each entry in table has m n-bit predictors.

12. 5/24/2015CPE 731, ILP2 12 Correlating Branch Prediction (2,2) predictor –Behavior of recent branches selects between four predictions of next branch, updating just that prediction Branch address 2-bits per branch predictor Prediction 2-bit global branch history 4

13. 5/24/2015CPE 731, ILP2 13 Correlated Branch Prediction Example: Find the size of (2, 2) predictor with k=10 Size = 2 m * n * 2 k = 4 * 2 * 1K = 8 Kbits

14. 5/24/2015CPE 731, ILP2 14 0% Frequency of Mispredictions 0% 1% 5% 6% 11% 4% 6% 5% 1% 2% 4% 6% 8% 10% 12% 14% 16% 18% 20% 4,096 entries: 2-bits per entry Unlimited entries: 2-bits/entry 1,024 entries (2,2) Accuracy of Different Schemes 4096 Entries 2-bit BHT Unlimited Entries 2-bit BHT 1024 Entries (2,2) BHT nasa7 matrix300doducdspicefppppgccexpressoeqntottlitomcatv

15. 5/24/2015CPE 731, ILP2 15 Outline Static Branch Prediction Dynamic Branch Prediction Branch History Table Correlated Branch Prediction Tournament Predictors Branch Target Buffer

16. 5/24/2015CPE 731, ILP2 16 Tournament Predictors Multilevel branch predictor Use n-bit saturating counter to choose between predictors Usual choice between global and local predictors

17. 5/24/2015CPE 731, ILP2 17 Tournament Predictors Tournament predictor using, say, 4K 2-bit counters indexed by local branch address. Chooses between: Global predictor –4K entries index by history of last 12 branches (2 12 = 4K) –Each entry is a standard 2-bit predictor Local predictor –Local history table: 1024 10-bit entries recording last 10 branches, index by branch address –The pattern of the last 10 occurrences of that particular branch used to index table of 1K entries with 3-bit saturating counters

18. 5/24/2015CPE 731, ILP2 18 Alpha 21264 Branch Predictor Selector Global Local BHR 4K addr 4K k=12 Prediction Select addr 1K 10 1K Size = 4K*2 + 4K*2 + 1K*10 + 1K*3 = 29 Kbits

19. 5/24/2015CPE 731, ILP2 19 Comparing Predictors (Fig. 2.8) Advantage of tournament predictor is ability to select the right predictor for a particular branch –Particularly crucial for integer benchmarks. –A typical tournament predictor will select the global predictor almost 40% of the time for the SPEC integer benchmarks and less than 15% of the time for the SPEC FP benchmarks

20. 5/24/2015CPE 731, ILP2 20 Outline Static Branch Prediction Dynamic Branch Prediction Branch History Table Correlated Branch Prediction Tournament Predictors Branch Target Buffer

21. Branch target calculation is costly and stalls the instruction fetch. BTB stores PCs the same way as caches The PC of a branch is sent to the BTB When a match is found the corresponding Predicted PC is returned If the branch was predicted taken, instruction fetch continues at the returned predicted PC Branch Target Buffers (BTB)

22. Branch Target Buffers

23. 5/24/2015CPE 731, ILP2 23 Pentium 4 Misprediction Rate (per 1000 instructions, not per branch) SPECint2000SPECfp2000  6% misprediction rate per branch SPECint (19% of INT instructions are branch)  2% misprediction rate per branch SPECfp (5% of FP instructions are branch)

24. 5/24/2015CPE 731, ILP2 24 Dynamic Branch Prediction Summary Prediction becoming important part of execution Branch History Table: 2 bits for loop accuracy Correlation: Recently executed branches correlated with next branch –Either different branches (GA) –Or different executions of same branches (PA) Tournament predictors take insight to next level, by using multiple predictors –usually one based on global information and one based on local information, and combining them with a selector –In 2006, tournament predictors using  30K bits are in processors like the Power5 and Pentium 4 Branch Target Buffer: include branch address & prediction