Revert "[lldb] Make CommandInterpreter's execution context the same as debugger's one."

This reverts commit a01b26fb51c710a3a8ef88cc83b0701461f5b9ab, because it
breaks the "finish" command in some way -- the command does not
terminate after it steps out, but continues running the target. The
exact blast radius is not clear, but it at least affects the usage of
the "finish" command in TestGuiBasicDebug.py. The error is *not*
gui-related, as the same issue can be reproduced by running the same
steps outside of the gui.

There is some kind of a race going on, as the test fails only 20% of the
time on the buildbot.

1 files changed, 10 insertions, 4 deletions

diff --git a/lldb/source/Commands/CommandObjectExpression.cpp b/lldb/source/Commands/CommandObjectExpression.cpp
index c7866f9..58eaa3f 100644
--- a/lldb/source/Commands/CommandObjectExpression.cpp
+++ b/lldb/source/Commands/CommandObjectExpression.cpp
@@ -292,12 +292,18 @@ void CommandObjectExpression::HandleCompletion(CompletionRequest &request) {
   options.SetAutoApplyFixIts(false);
   options.SetGenerateDebugInfo(false);
 
-  ExecutionContext exe_ctx(m_interpreter.GetExecutionContext());
-
-  // Get out before we start doing things that expect a valid frame pointer.
-  if (exe_ctx.GetFramePtr() == nullptr)
+  // We need a valid execution context with a frame pointer for this
+  // completion, so if we don't have one we should try to make a valid
+  // execution context.
+  if (m_interpreter.GetExecutionContext().GetFramePtr() == nullptr)
+    m_interpreter.UpdateExecutionContext(nullptr);
+
+  // This didn't work, so let's get out before we start doing things that
+  // expect a valid frame pointer.
+  if (m_interpreter.GetExecutionContext().GetFramePtr() == nullptr)
     return;
 
+  ExecutionContext exe_ctx(m_interpreter.GetExecutionContext());
   Target *exe_target = exe_ctx.GetTargetPtr();
   Target &target = exe_target ? *exe_target : GetDummyTarget();