Aliasing and AntiAliasing in Branch History Table Prediction

Aliasing and Anti-Aliasing in Branch History Table Prediction Kris Breen EE 710 A 2 Nov. 26, 2002

OUTLINE o o o Introduction Theory Experiment Results Conclusions & Future Work Aliasing and Anti-Aliasing in BHT Prediction

OUTLINE o o o Introduction Theory Experiment Results Conclusions & Future Work Aliasing and Anti-Aliasing in BHT Prediction

INTRODUCTION • • Branch history predictors have advantage of simplicity Prediction accuracy limited by table size, due to aliasing Aliasing and Anti-Aliasing in BHT Prediction

INTRODUCTION 2 n-entry BHT PC of branch Aliasing and Anti-Aliasing in BHT Prediction Hash Function n

INTRODUCTION 2 n-entry BHT PC of branch 1 PC of branch 2 Aliasing and Anti-Aliasing in BHT Prediction Hash Function n ALIASING

OUTLINE o o o Introduction Theory Experiment Results Conclusions & Future Work Aliasing and Anti-Aliasing in BHT Prediction

THEORY • • What performance loss occurs due to aliasing? Can reasonably assume that aliasing has a negative impact on branch prediction. Start by assuming that hash function simply uses low bits of address. Analyze static program properties first. Aliasing and Anti-Aliasing in BHT Prediction

THEORY • • m = total number of addresses in program size of predictor table pb = probability that an instruction is a branch p(0 branches hash to x) = (1 - pb)m p(2 or more branches hash to x) = 1 - p(0 branches hash to x)- p(1 branch hashes to x) = 1 - (1 - pb)m - pb(1 - pb)m-1 Aliasing and Anti-Aliasing in BHT Prediction

THEORY • • Assuming 1 in 7 instructions is a branch, and a program size 10 times larger than the table size, p(2 or more branches hash to x) = 71% For a 1024 entry table, this means that about 700 entries are experiencing aliasing. Further evaluation shows that of the remaining 300 entries, about 250 of them likely have no branches hashing to them. These can be used for anti-aliasing. Aliasing and Anti-Aliasing in BHT Prediction

OUTLINE o o o Introduction Theory Experiment Results Conclusions & Future Work Aliasing and Anti-Aliasing in BHT Prediction

EXPERIMENT • • • Used Simple. Scalar to analyze effects of aliasing in programs Based experiments on PISA architecture Used 2 -bit predictor and hash function based on low bits of address Aliasing and Anti-Aliasing in BHT Prediction

EXPERIMENT • • • Modified bpred. [ch] to remember entire branch PC when BHT lookup is done. On every BHT lookup, bpred. c compares the current branch PC to the stored branch PC. If these differ, then a lookup alias has occurred, and it is recorded. On BHT update, bpred. c checks if the prediction was correct. If it wasn’t, and it was from an aliased table entry, a miss due to aliasing is recorded. Aliasing and Anti-Aliasing in BHT Prediction

EXPERIMENT • • • Simulations were run using Todd Austin’s SPEC 2000 -derived benchmarks (gcc, anagram, compress, go) Branch history table sizes of 32, 256, 1024, and 8192 were each used for each benchmark. Percentage of aliased lookups and percentage of misses due to aliasing were recorded for each simulation. Aliasing and Anti-Aliasing in BHT Prediction

OUTLINE o o o Introduction Theory Experiment Results Conclusions & Future Work Aliasing and Anti-Aliasing in BHT Prediction

RESULTS • Sample simulation output (from gcc using a 1024 entry table): bpred_bimod. lookups bpred_bimod. dir_hits bpred_bimod. misses bpred_bimod. alias_rate bpred_bimod. alias_misses Aliasing and Anti-Aliasing in BHT Prediction 56329774 # total number of bpred lookups 49575889 # total number of direction-predicted hits 6753885 # total number of misses 3991026 # total number of aliases 0. 0709 # alias rate (aliases/lookups) 2614746 # number of mispredictions due to aliasing

RESULTS Aliasing and Anti-Aliasing in BHT Prediction

RESULTS • Miss rate when using aliased BHT entries ranged from 44% to 61%, which supports concept that aliased accesses provide effectively random information Aliasing and Anti-Aliasing in BHT Prediction

OUTLINE o o o Introduction Theory Experiment Results Conclusions & Future Work Aliasing and Anti-Aliasing in BHT Prediction

CONCLUSIONS • • Reduction of aliasing can provide a substantial performance gain at minimal cost when using BHT Worst case miss rate = predictor miss rate + alias rate For a predictor miss rate of 10% and an alias rate of 10%, halving the alias rate improves overall prediction by 25% Aliasing and Anti-Aliasing in BHT Prediction

FUTURE WORK • • Attempt to reduce loss due to aliasing by passing hash information to processor from program Determine program properties (pb, m) for which anti -aliasing will be effective Aliasing and Anti-Aliasing in BHT Prediction

fin Aliasing and Anti-Aliasing in BHT Prediction