Reinstate memory pressure Oilpan GC check post V8 major GCs.

third_party/WebKit/Source/platform/heap/ThreadState.cpp

Index: third_party/WebKit/Source/platform/heap/ThreadState.cpp
diff --git a/third_party/WebKit/Source/platform/heap/ThreadState.cpp b/third_party/WebKit/Source/platform/heap/ThreadState.cpp
index 4a1efe48e8904b9f5bcf581155bcf3d28980ee82..92148179428fb84048c98097305c44fffdcfd20a 100644
--- a/third_party/WebKit/Source/platform/heap/ThreadState.cpp
+++ b/third_party/WebKit/Source/platform/heap/ThreadState.cpp
@@ -678,6 +678,13 @@ void ThreadState::scheduleV8FollowupGCIfNeeded(V8GCType gcType)
     if (isGCForbidden())
         return;
 
+    // If V8 has acted on a memory pressure signal and performed a major GC,
+    // follow up, if needed.
+    if (gcType == V8MajorGC && shouldForceMemoryPressureGC()) {

[haraken, 2015/09/26 02:04:37]

A better form would be:

if (shouldForceMemoryPressureGC()) {
  completeSweep();
  if (shouldForceMemoryPressureGC()) {
    Heap::collectGarbage(...);
    return;
  }
}

We may want to create a helper method like shouldForceMemoryPressureGCIfNeeded()
that does the work.

[haraken, 2015/09/26 02:04:37]

Is there any reason we want to limit this to V8MajorGC?

[sof, 2015/09/26 06:17:22]

v8 will force major GCs on detecting/being signalled memory pressure, so it
doesn't pay to be more vigilant here. That along with being mindful of the v8
perf regressions that were caused last time around when triggering too many
conservative GCs, https://code.google.com/p/chromium/issues/detail?id=499822

[sof, 2015/09/26 06:17:22]

We've been over that two-step formulation before in
https://codereview.chromium.org/1174123002/ . Adopted.

[sof, 2015/09/26 06:17:22]

Done.

+        Heap::collectGarbage(HeapPointersOnStack, GCWithoutSweep, Heap::ConservativeGC);
+        return;
+    }
+
     if (isSweepingInProgress())
         return;
     ASSERT(!sweepForbidden());