InlineTypeBaseNode::buffer currently adds a StoreStore barrier to not let stores that initialize this buffer be reordered with a subsequent store that would make this buffer accessible by other threads.
We should evaluate the impact of the barrier which is potentially blocking other optimizations and re-visit the fixes of the linked bugs.
We should also re-visit the barrier code in Parse::array_load.
We should evaluate the impact of the barrier which is potentially blocking other optimizations and re-visit the fixes of the linked bugs.
We should also re-visit the barrier code in Parse::array_load.
- relates to
-
JDK-8221647 [lworld] Performance regression due to the fact that check if array is array of values is not hoisted out of the loop.
- Resolved
-
JDK-8238780 [lworld] C2: Partially initialized value type buffer might be observed by other thread
- Resolved
-
JDK-8264414 [lworld] [AArch64] TestBufferTearing.java fails with C1
- Resolved
-
JDK-8270995 [lworld] G1BarrierSetC2::step_over_gc_barrier asserts with "bad barrier shape"
- Resolved