CS 7960-4 Lecture 7

1
CS 7960-4 Lecture 7
Combining Branch Predictors Scott McFarling WRL
Tech. Report TN-36 1993
2
Bimodal Branch Prediction

• Identifies most popular prediction in recent
  past
• Updates happen during commit

1
0
PC
10-bit index
1024 entries
2-bit saturating counters
3
Results

• SPEC89 programs simulated for 10M instrs
• (modern studies use hard-to-predict programs)
• A larger predictor reduces contention for
  counters
• Prediction rates saturate at 93.5 (at 2K bytes)
• (Fig.3)

4
Local Predictors

• Two-Level predictor The first level has
  history,
• the second level has saturating counters
• History gets updated immediately

0
1
1
1
PC
1
0
10-bit index
16 entries
1024 entries
2-bit saturating counters
4-bit history table
5
Results

• For small predictors, there could be contention
• at both levels, resulting in inaccurate
  predictions
• Will also take longer to warm up after every
• context switch
• Does very well for large predictors saturates
  at
• 97.1

6
Global Predictors

• A single history register neighboring branches
• have correlated results
• However, the PC is not used

1
0
1024 entries
10-bit global history
2-bit saturating counters
7
Do We Need PC?

• Note that the global history reveals which
  branch
• is being examined
• Hence, it outdoes bimodal predictors when the
• transistor budget is large (Fig.7)
• Local predictor does better it is more
  important
• to identify the PC and local history than
  behavior
• of neighboring branches

8
Gselect

• Use a combination of PC and global history
• Bimodal and global prediction are special cases
• (Fig.9)

1
0
n
PC
/
nm
/
/
1024 entries
m
5-bit global history
2-bit saturating counters
9
GShare

• Xor-ing 10 history bits and 10 PC bits has more
• info than the concatenation of 5 bits of each
  and
• more info than each individual component

Branch Address Global History Gselect 4/4 Gshare 8/8
00000000 00000001 00000001 00000001
00000000 00000000 00000000 00000000
11111111 00000000 11110000 11111111
11111111 10000000 11110000 01111111
01111110
00000001
11100001
01111111
10
Terminology

• GAG Global history indexes into global array
• of saturating counters
• PAG Per-address history indexes into global
  array
• of saturating counters
• GAP Global history indexes into each PCs
  private
• array of counters (gselect)
• PAP Per-address history indexes into each PCs
• private array of counters

11
Trade-Offs

• Some predictors warm-up faster than others
• Some programs benefit from global history, some
• from local history
• Some programs have branches that interfere
• with each other
• Note that a 64KB local predictor has fewer
• saturating counters than a 64KB bimodal
  predictor
• the former wont be better for every program

12
Combining Predictors

• Use an array of saturating counters to pick the
• best available predictor for each PC

Predictor A
1
0
PC
1024 entries
Predictor B
2-bit saturating counters
13
Results

• The combination of local and gshare increases
• the prediction accuracy to 98.1 (Fig.16)
• For smaller transistor budgets, the combination
• of bimodal and gshare is better (gshare is
  twice
• the size to make sure the total is a power of
  two)
• A 1KB combined predictor does as well as a
• 16KB gselect predictor

14
Future Work

• Detect conflicts, correlations, and common
• predictions through profiling/compiler analysis
• Functions that compress information in history
• or PC
• Pipeline predictions predict two branches
  ahead
• Hierarchical predictors get a quick prediction
  in
• a cycle and a more accurate one two cycles later

15
Next Weeks Paper

• Design Trade-Offs for the Alpha EV8 Conditional
• Branch Predictor, Seznec et al., ISCA02

16
Title

• Bullet