jni: Do not create a managed frame for @CriticalNative.

Omit managed frame for @CriticalNative methods, do not check
for exceptions and and make a tail call when possible.
Pass the method pointer in a hidden argument to prepare for
implementing late binding for @CriticalNative methods.

This changes only the JNI compiler, Generic JNI shall be
updated in a separate change.

Performance improvements reported by Golem (art-opt-cc):
                                 x86 x86-64    arm  arm64
NativeDowncallStaticCritical6   +17%   +50%   +88%  +139%
NativeDowncallStaticCritical    +37%   +32%  +103%  +216%

Test: m test-art-host-gtest
Test: testrunner.py --host --optimizing
Test: aosp_taimen-userdebug boots.
Test: run-gtests.sh
Test: testrunner.py --target --optimizing
Bug: 112189621
Change-Id: I5758c8f478627f2eee8f615b4537a907c211b9f8

1 files changed, 3 insertions, 0 deletions

diff --git a/compiler/utils/jni_macro_assembler.h b/compiler/utils/jni_macro_assembler.h
index e6130cfc4c..bbe0f734ba 100644
--- a/compiler/utils/jni_macro_assembler.h
+++ b/compiler/utils/jni_macro_assembler.h
@@ -197,6 +197,9 @@ class JNIMacroAssembler : public DeletableArenaObject<kArenaAllocAssembler> {
   virtual void VerifyObject(ManagedRegister src, bool could_be_null) = 0;
   virtual void VerifyObject(FrameOffset src, bool could_be_null) = 0;
 
+  // Jump to address held at [base+offset] (used for tail calls).
+  virtual void Jump(ManagedRegister base, Offset offset, ManagedRegister scratch) = 0;
+
   // Call to address held at [base+offset]
   virtual void Call(ManagedRegister base, Offset offset, ManagedRegister scratch) = 0;
   virtual void Call(FrameOffset base, Offset offset, ManagedRegister scratch) = 0;