stan/jdk-24
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

8148945: JDK-8148481: Devirtualize Klass::vtable breaks Zero build

Browse Source 
Use Klass::method_at_vtable() instead of InstanceClass::start_of_vtable()[index]

Reviewed-by: mgerdin, coleenp
This commit is contained in:
Severin Gehwolf 2016-02-03 12:19:03 +01:00
parent bbbbfdbf3f
commit c647cd04b0
1 changed files with 6 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
hotspot/src/share/vm/interpreter/bytecodeInterpreter.cpp
View File

@ -2502,10 +2502,10 @@ run:
// Same comments as invokevirtual apply here.
oop rcvr = STACK_OBJECT(-parms);
VERIFY_OOP(rcvr);
InstanceKlass* rcvrKlass = (InstanceKlass*)rcvr->klass();
callee = (Method*) rcvrKlass->start_of_vtable()[ cache->f2_as_index()];
Klass* rcvrKlass = rcvr->klass();
callee = (Method*) rcvrKlass->method_at_vtable(cache->f2_as_index());
// Profile 'special case of invokeinterface' virtual call.
BI_PROFILE_UPDATE_VIRTUALCALL(rcvr->klass());
BI_PROFILE_UPDATE_VIRTUALCALL(rcvrKlass);
}
istate->set_callee(callee);
istate->set_callee_entry_point(callee->from_interpreted_entry());
@ -2594,7 +2594,7 @@ run:
// but this works
oop rcvr = STACK_OBJECT(-parms);
VERIFY_OOP(rcvr);
InstanceKlass* rcvrKlass = (InstanceKlass*)rcvr->klass();
Klass* rcvrKlass = rcvr->klass();
/*
Executing this code in java.lang.String:
public String(char value[]) {
@ -2611,9 +2611,9 @@ run:
However it seems to have a vtable in the right location. Huh?
Because vtables have the same offset for ArrayKlass and InstanceKlass.
*/
callee = (Method*) rcvrKlass->start_of_vtable()[ cache->f2_as_index()];
callee = (Method*) rcvrKlass->method_at_vtable(cache->f2_as_index());
// Profile virtual call.
BI_PROFILE_UPDATE_VIRTUALCALL(rcvr->klass());
BI_PROFILE_UPDATE_VIRTUALCALL(rcvrKlass);
}
} else {
if ((Bytecodes::Code)opcode == Bytecodes::_invokespecial) {