-
-
I don't know if it actually works or not, we should try first.
-
I'll need to take a look at exactly what the problem is here.
It could be a performance issue, but it probably can be handled so that this issue is resolved.
It may also have something to do with in…
-
This might be a longer-term plan. First step would be to have some partial support where at least we can parse and don't give wrong results (or harnesses can filter them).
-
-
SV-COMP and BenchExec can give the task time limit to the tool. Goblint doesn't use this information and just keeps running after the BenchExec soft time limit until it is killed at the hard time limi…
-
Our race analysis can output "proofs" as success messages with `allglobs` enabled. This is useful for inspecting why Goblint believes the program to be correct.
We could do something similar for te…
-
@brcfarias: Can I ask you to evaluate your implementations of the ub-shift-check over the HW benchmarks in SV-COMP 2023?
https://gitlab.com/sosy-lab/benchmarking/sv-benchmarks/-/tree/main/c/hardwar…
-
In https://github.com/esbmc/esbmc/pull/602#issuecomment-1030013156 the decision was made to move the operation model of pthreads to types provided by our own headers. It turns out that these types are…
-
Currently, `crux-llvm` has overrides for the following SV-COMP–related `__VERIFIER_nondet_*` functions:
https://github.com/GaloisInc/crucible/blob/c96a4041ec425a058c20660a99b8975f537d894f/crux-llvm…