Fix instrumentation stack tracking
We were incorrectly tracking the instrumentation stack, bailing out of
installation early. This could lead to missing instrumentation events,
failing CHECKs or (potentially) corrupt thread stacks.
Test: ./test.py --host --trace --ntrace --stream -j50
Test: Use gapid.
Bug: 67384421
Bug: 74386428
Change-Id: I91686ac4f44550d89b9d7a5f716df88f505d91b1
diff --git a/runtime/instrumentation.cc b/runtime/instrumentation.cc
index b079598..84a148f 100644
--- a/runtime/instrumentation.cc
+++ b/runtime/instrumentation.cc
@@ -256,7 +256,7 @@
}
// We've reached a frame which has already been installed with instrumentation exit stub.
- // We should have already installed instrumentation on previous frames.
+ // We should have already installed instrumentation or be interpreter on previous frames.
reached_existing_instrumentation_frames_ = true;
const InstrumentationStackFrame& frame =
@@ -269,7 +269,9 @@
}
} else {
CHECK_NE(return_pc, 0U);
- if (UNLIKELY(reached_existing_instrumentation_frames_)) {
+ if (UNLIKELY(reached_existing_instrumentation_frames_ && !m->IsRuntimeMethod())) {
+ // We already saw an existing instrumentation frame so this should be a runtime-method
+ // inserted by the interpreter or runtime.
std::string thread_name;
GetThread()->GetThreadName(thread_name);
uint32_t dex_pc = dex::kDexNoIndex;
@@ -277,7 +279,8 @@
GetCurrentOatQuickMethodHeader() != nullptr) {
dex_pc = GetCurrentOatQuickMethodHeader()->ToDexPc(m, last_return_pc_);
}
- LOG(FATAL) << "While walking " << thread_name << " found existing instrumentation frames."
+ LOG(FATAL) << "While walking " << thread_name << " found unexpected non-runtime method"
+ << " without instrumentation exit return or interpreter frame."
<< " method is " << GetMethod()->PrettyMethod()
<< " return_pc is " << std::hex << return_pc
<< " dex pc: " << dex_pc;