https://github.com/Xazax-hun commented:

Overall looks great, some nits inline.

I think it might make sense to have a general facility at some point that 
matches on some APIs (like methods) and replaces the API calls with a sequence 
of hard coded facts.

This is a way to hard code summaries. The static analyzer already has some 
convenience facilities to match APIs. I am wondering if those could be reused.

That being said, I think we can wait with all that until we see multiple 
functions being directly modeled in this analysis. So far, only for `release` 
these generalizations might not carry their weight. 

https://github.com/llvm/llvm-project/pull/180230
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Reply via email to