Chromium Code Reviews
chromiumcodereview-hr@appspot.gserviceaccount.com (chromiumcodereview-hr) | Please choose your nickname with Settings | Help | Chromium Project | Gerrit Changes | Sign out
(300)

Unified Diff: test/cctest/heap/test-heap.cc

Issue 2289493002: Revert of "[heap] Switch to 500k pages" (Closed)
Patch Set: Created 4 years, 4 months ago
Use n/p to move between diff chunks; N/P to move between comments. Draft comments are only viewable by you.
Jump to:
View side-by-side diff with in-line comments
Download patch
« no previous file with comments | « test/cctest/heap/heap-utils.cc ('k') | test/cctest/heap/test-spaces.cc » ('j') | no next file with comments »
Expand Comments ('e') | Collapse Comments ('c') | Show Comments Hide Comments ('s')
Index: test/cctest/heap/test-heap.cc
diff --git a/test/cctest/heap/test-heap.cc b/test/cctest/heap/test-heap.cc
index bebd00c6a57dface409fee3d55eb5fdc533393a1..80c4cc7c3434912d9258101959828a0b48704b06 100644
--- a/test/cctest/heap/test-heap.cc
+++ b/test/cctest/heap/test-heap.cc
@@ -2218,18 +2218,6 @@
// Test the case where allocation must be done from the free list, so filler
// may precede or follow the object.
TEST(TestAlignedOverAllocation) {
- Heap* heap = CcTest::heap();
- // Test checks for fillers before and behind objects and requires a fresh
- // page and empty free list.
- heap::AbandonCurrentlyFreeMemory(heap->old_space());
- // Allocate a dummy object to properly set up the linear allocation info.
- AllocationResult dummy =
- heap->old_space()->AllocateRawUnaligned(kPointerSize);
- CHECK(!dummy.IsRetry());
- heap->CreateFillerObjectAt(
- HeapObject::cast(dummy.ToObjectChecked())->address(), kPointerSize,
- ClearRecordedSlots::kNo);
-
// Double misalignment is 4 on 32-bit platforms, 0 on 64-bit ones.
const intptr_t double_misalignment = kDoubleSize - kPointerSize;
Address start;
@@ -2371,9 +2359,6 @@
TEST(GrowAndShrinkNewSpace) {
CcTest::InitializeVM();
- // Avoid shrinking new space in GC epilogue. This can happen if allocation
- // throughput samples have been taken while executing the benchmark.
- i::FLAG_predictable = true;
Heap* heap = CcTest::heap();
NewSpace* new_space = heap->new_space();
@@ -3616,12 +3601,8 @@
i::FLAG_page_promotion = false;
CcTest::InitializeVM();
Isolate* isolate = CcTest::i_isolate();
- // If there's snapshot available, we don't know whether 20 small arrays will
- // fit on the initial pages.
- if (!isolate->snapshot_available()) return;
Factory* factory = isolate->factory();
Heap* heap = isolate->heap();
-
v8::HandleScope scope(CcTest::isolate());
static const int number_of_test_pages = 20;
@@ -3651,8 +3632,14 @@
"triggered by test 2");
CHECK_GE(overall_page_count, old_space->CountTotalPages() * 2);
+ // Triggering a last-resort GC should cause all pages to be released to the
+ // OS so that other processes can seize the memory. If we get a failure here
+ // where there are 2 pages left instead of 1, then we should increase the
+ // size of the first page a little in SizeOfFirstPage in spaces.cc. The
+ // first page should be small in order to reduce memory used when the VM
+ // boots, but if the 20 small arrays don't fit on the first page then that's
+ // an indication that it is too small.
heap->CollectAllAvailableGarbage("triggered really hard");
- // Triggering a last-resort GC should release all additional pages.
CHECK_EQ(initial_page_count, old_space->CountTotalPages());
}
« no previous file with comments | « test/cctest/heap/heap-utils.cc ('k') | test/cctest/heap/test-spaces.cc » ('j') | no next file with comments »

Powered by Google App Engine
This is Rietveld 408576698