bosccTransform performance remedy problem

[ghost]

I turn on config.enableHeuristicBOSCC in PACXX and test it using aobench and rodinia/cfd. This transformation does not bypass the desired path, however, with hard-coded optimization, I could get more than 2X than the original RV version. I find several problems in this bosccTransform.

1. In the computeDispersion func of bosccTransform.cpp: line 528, GetEdgeProb(*start, *end) should not been used here. GetEdgeProb(*start, *end) computes all the paths from *start to *end, while AFAIU, we only need the direct path from *start to *end. GetEdgeProb(*start, Index) is more suitable here.
2. In bosccHeuristic func, I wonder why maxRatio and minScore is assgined to a specific number(0.14&17). The aforementioned two applications fail to meeting the requirement of <0.14.

[simoll]

1. Feel free to fix it (eg by skpping predecessors we have already seen (line 504). The current implementation gives you over-pessimistic results when a divergent switch jumps to the same block on multiple cases.
2. These are heuristic values. You can use the environment variables BOSCC_T and BOSCC_LIMIT to feed in your own thresholds. I'd like to look into this. Can you send me a mail with the details on aobench, rodinia/cfd?

[ghost]

Sure. You could find the source code in aobench folder and cfd folder in my sample repos (https://github.com/HuihuiSun/samples). A preliminary improve of this boscctransformation is updated in pacxx-rv.