search for: cgexcept

...to start porting EH code to the new C++ API? Nope! :-) > Are there sample implementations? llvm.org/demo still uses the old > EH gizmos. should we wait for 3.0 release? If you're using the LLVM APIs, then looking at how it's implemented in clang is a good start: clang/lib/CodeGen/CGException.cpp If you're upgrading your bitcode files, then just (re)compile with top-of-tree clang, and you should get the new EH stuff automatically. -bw

Hi all, Is it too soon to start porting EH code to the new C++ API? Are there sample implementations? llvm.org/demo still uses the old EH gizmos. should we wait for 3.0 release? -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20110930/3bc09ea9/attachment.html>

...instance. The version I'm working with is LLVM 3.8.1, and the code I've created is based on the documentation http://llvm.org/docs/ExceptionHandling.html . I've also been reading up on the Exception toy/example code that's shipped with LLVM, as well as several mail discussions, the CGException class from clang and of course the ever useful 'clang --emit-llvm'. My aim here is to create a minimum working test case. My first attempt was to simply compile a piece of C++ code with an external function call and run that through LLC. The code is pretty straight-forward: -- extern &...

...g-Out-of-_Finally) The implementation can be seen here: https://github.com/tentzen/llvm-project/compare/SEH-LU-base...SEH-LU?expand=1 There are only 13 files changed that is much less complicated than previously when it’s combined with -EHa. Major code is surrounding at SehTryStmt and _Finally in CGexception.cpp that is the place to house Windows SEH specific code anyways. The implementation chose to lever Parser/Semantic phase (SemaStmt.cpp, JumpDiagnostics.cpp, scope.h, etc) to identify the right _Try scope for LU dispatching because that is the place “warn_jump_out_of_seh_finally” is diagnosed...

As stated in the design paragraph, this design does not intend to model precise CFG at instruction level since it’s complicated and unnecessary. As long as we comply C and C++ rules listed below, we achieve -EHa semantic. There is NO need to precisely model HW exception control flow at instruction-level. Your example about memcpy() is just a bug in current implementation. I will fix it so that