GkvJwa wrote: > I understand that you're working on improving the support for async SEH. > That's great, and it sounds like Chromium has code which will benefit. > > However, you didn't answer my main question: if `err_seh_object_unwinding` is > for diagnosing mixing SEH and C++ exception handling, why did this PR gate it > behind `getLangOpts().EHAsynch` and not `getLangOpts().CXXExceptions`?
I expressed that seh only approaches usability when eha is enabled. Therefore, I used ehasynch detection instead of cxxexceptions. Moreover, chromium disables exceptions. I think if the code that reported an error before is executed and only eha is enabled, Ilvm is likely to crash as well In addition, these switches may not necessarily be equivalent even in the presence of seh (based on the current code of Ilvm) https://github.com/llvm/llvm-project/pull/180108 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
