[Lldb-commits] [lldb] [lldb] Change lldb's breakpoint handling behavior (PR #96260)

[Jason Molenda via lldb-commits]

jasonmolenda wrote:

The problem I now have is that debugserver doesn't do this nice behavior that lldb-server does, recognizing that a thread was instruction-stepping and rewriting the stop-reason as "trace" when it was actually "breakpoint-hit".  I can't update the debugserver on the armv7 watches, and we need to support debugging on them for a while to come.  I may need to make a generic code change where Thread::GetTemporaryResumeState can return both eStateStepping and its original pc it started the step from, only for the armv7 debugservers.  With that small fix to `NativeProcessLinux::MonitorBreakpoint` outlined above, it should report the correct stop reason all by itself in this case.

https://github.com/llvm/llvm-project/pull/96260