[Lldb-commits] [lldb] [lldb] Improve unwinding for discontinuous functions (PR #111409)

[Pavel Labath via lldb-commits]

labath wrote:

Thanks Jason. I have no doubt that the failure was triggered by this patch (the test tests unwinding, the patch touches unwinding), but the mechanism itself is very unclear to me. The functions in that patch have a valid eh_frame (and they don't have debug info), so I would expect that a change like this would be a noop in this case. I suspect there's something subtle going on, and that this could very well be a bug in some other lldb component, but it's hard tell what that is without more info. It could also easily depend on some system code that makes its way into the debugged process (process startup code and stuff). If you're unable to reproduce this on your machine, maybe we'll have to modify the test to dump loads of extra output to make sense of it.

https://github.com/llvm/llvm-project/pull/111409