Reset JIT method counters to 0 instead of 1 when clearing.
When the counters need to be cleared, the count should reset to 1
instead of 0. The counters may be reset during startup, and startup
methods can be lost by resetting the counters to 0.
Bug: 62200509
Test: mm test-art-host
Change-Id: I8936631d3bab369c6a14acf76b20e7245de14789
diff --git a/runtime/jit/jit_code_cache.cc b/runtime/jit/jit_code_cache.cc
index 2377b5b..42d7653 100644
--- a/runtime/jit/jit_code_cache.cc
+++ b/runtime/jit/jit_code_cache.cc
@@ -597,11 +597,10 @@
bool single_impl_still_valid = true;
for (ArtMethod* single_impl : cha_single_implementation_list) {
if (!single_impl->HasSingleImplementation()) {
- // We simply discard the compiled code. Clear the
- // counter so that it may be recompiled later. Hopefully the
- // class hierarchy will be more stable when compilation is retried.
+ // Simply discard the compiled code. Clear the counter so that it may be recompiled later.
+ // Hopefully the class hierarchy will be more stable when compilation is retried.
single_impl_still_valid = false;
- method->ClearCounter();
+ method->SetCounter(1);
break;
}
}
@@ -1069,8 +1068,9 @@
if (info->GetSavedEntryPoint() != nullptr) {
info->SetSavedEntryPoint(nullptr);
// We are going to move this method back to interpreter. Clear the counter now to
- // give it a chance to be hot again.
- info->GetMethod()->ClearCounter();
+ // give it a chance to be hot again, but set it to 1 so that this method can still be
+ // considered a startup method in case it's not executed again.
+ info->GetMethod()->SetCounter(1);
}
}
} else if (kIsDebugBuild) {
@@ -1377,10 +1377,9 @@
ProfilingInfo* info = method->GetProfilingInfo(kRuntimePointerSize);
if (info == nullptr) {
VLOG(jit) << method->PrettyMethod() << " needs a ProfilingInfo to be compiled";
- // Because the counter is not atomic, there are some rare cases where we may not
- // hit the threshold for creating the ProfilingInfo. Reset the counter now to
- // "correct" this.
- method->ClearCounter();
+ // Because the counter is not atomic, there are some rare cases where we may not hit the
+ // threshold for creating the ProfilingInfo. Reset the counter now to "correct" this.
+ method->SetCounter(1);
return false;
}
@@ -1432,12 +1431,12 @@
}
if (method->GetEntryPointFromQuickCompiledCode() == header->GetEntryPoint()) {
- // The entrypoint is the one to invalidate, so we just update
- // it to the interpreter entry point and clear the counter to get the method
- // Jitted again.
+ // The entrypoint is the one to invalidate, so we just update it to the interpreter entry point
+ // and clear the counter to get the method Jitted again. We reset the counter to 1 to preserve
+ // it as a potential startup method.
Runtime::Current()->GetInstrumentation()->UpdateMethodsCode(
method, GetQuickToInterpreterBridge());
- method->ClearCounter();
+ method->SetCounter(1);
} else {
MutexLock mu(Thread::Current(), lock_);
auto it = osr_code_map_.find(method);
