smeenai added inline comments.
================
Comment at: test/CodeGen/exceptions-seh-finally.c:281
+// CHECK-LABEL: define internal void @"?fin$0@0@finally_with_func@@"({{[^)]*}})
+// CHECK: call void @cleanup_with_func(i8* getelementptr inbounds ([18 x i8],
[18 x i8]* @"??_C@_0BC@COAGBPGM@finally_with_func?$AA@", i32 0, i32 0))
+
----------------
How stable is the mangling here? I included the complete `call` instruction
because I wanted to ensure that the value of `__func__` inside a `__finally`
block reflects the parent function, not the outlined funclet, and both the `[18
x i8]` and the mangled string name are pretty good proxies for that. I can
generalize it more if it'll make the test more reliable though.
Repository:
rC Clang
https://reviews.llvm.org/D45523
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
