Casper S. Jensen

I'm debugging it now. My options object contains a lot of wrong (and impossible) settings, so something in Artemis is writing to some memory when it should not.

My problem is now fixed in 9e232e5181e604598d65c1f3d2aae68afdb41a36 and I did not observe the original problem (at least not up until iteration 150...). Christian, can you check if your problem is...

I'm able to run the full 500 iterations. === Statistics === FormCrawl::Entrypoints: 1000 timers::registered: 2838 WebKit::coverage::covered-unique: 217 timers::fired: 1407 InputGenerator::added-configurations: 4827 Concolic::Solver::ConstraintsNotWritten: 1 WebKit::coverage::covered: 1731725 ajax::fired: 2154

The only downside I can think of is a slight increase in the API size between the two parts and a small increase in the responsibilities of the execution listener....

I have done a bit of profiling using the webpage provided in issue #38. Right now there seems to be two major causes to the slowdown encountered on large pages:...

I have committed a couple of performance related changes. However, bytecode coverage (and line coverage) is still causing some slowdown.