Simmo Saan

It looks to me like this patch wants to use `hook_connect_gnutls_set_certificates` from the core and does that in an unacceptable (including core directly, not using the API) and broken (autotools...

I'm still interested in having this but haven't worked on this since #1483 is still open as well.

The upjumping check being bad is a known issue from the start (#1245), but we can keep this open because it shows other implications to more complex loops.

Yesterday I opened Signal after a while and was surprised that it's making me re-register (I previously successfully transferred Signal to this phone and never registered any other device). But...

> Yesterday I opened Signal after a while and was surprised that it's making me re-register (I previously successfully transferred Signal to this phone and never registered any other device)....

While looking at Concrat again, I found out that `pthread_cleanup_push` is a macro which doesn't actually store the cleanup routine is some magic place, but implements it using some flavor...

I learned today that GCC/Clang also have a `__cleanup__` attribute which gives C++ RAII-like features in C. And it's starting to be used in the Linux kernel: https://lwn.net/Articles/934679/. It's more...

> a first comparison on the svcomp benchmark revealed 10 new tasks that could be solved by a narrowing_gas value of 5 compared to a vanilla polyhedra analysis -- interestingly...

> I am not sure why the regression test fail, as it runs locally. Do you have a suggestion how to debug this? The CI artifact (https://github.com/goblint/analyzer/actions/runs/16374091305/artifacts/3564686204) should have the...

> * Is there a way to enforce configuration settings to be set to certain values in order for the analysis to work (i.e. similar to how analyses can be...