Information Technology Reference
In-Depth Information
Table 6 Area consumption of MicroBlaze processor synthesis
Con
guration
With optimization
synthesis
Without optimization
synthesis
MB + Units
LUTs
F-Fs
LUTs
F-Fs
1: Basic
1,210
1,452
1,657
1,693
2: BS
1,570
1,247
1,818
1,727
3: FPU
1,620
2,153
2,395
2,105
4: Mul
1,456
1,232
1,714
1,709
5: ID
1,581
1,326
1,801
1,805
6: MSRU
1,458
1,214
1,675
1,690
7: BS + mul + ID
1,727
1,380
1,964
1,867
8: BS + mul + FPU
2,307
1,674
2,511
2,162
9: BS + ID + FPU
2,433
1,769
2,668
2,258
10: BS + mul + MSRU
1,609
1,267
1,830
1,749
11: BS + ID + MSRU
1,734
1,365
1,966
1,846
12: BS + FPU + MSRU
2,313
1,659
2,533
2,142
13: mul + ID + FPU
2,358
1,755
2,575
2,241
14: mul + ID + MSRU
1,628
1,351
1,864
1,829
15: mul + FPU + MSRU
2,207
1,645
2,418
2,127
16: ID + MSRU + FPU
2,359
1,740
2,554
2,224
MSRU + FPU
2,202
1,625
2,417
2,107
Fig. 7 Quark benchmark execution time usage for different MicroBlaze configurations
In order to select the greatest hardware/software architecture (partitioning pro-
cess), we used the Integer Linear Programing (ILP) algorithm. Under the ILP
algorithm, Gains of execution time and resource consumption are computed
(as described in Table
7
) using these two formulas:
Search WWH ::
Custom Search