6810672: Comment typos

I have collected some typos I have found while looking at the code.

Reviewed-by: kvn, never

hotspot/src/share/vm/gc_implementation/concurrentMarkSweep/cmsAdaptiveSizePolicy.hpp

@@ -393,7 +393,7 @@ class CMSAdaptiveSizePolicy : public AdaptiveSizePolicy {
    // Restarts the concurrent phases timer.
    void concurrent_phases_resume();
 
-   // Time begining and end of the marking phase for
+   // Time beginning and end of the marking phase for
    // a synchronous MS collection.  A MS collection
    // that finishes in the foreground can have started
    // in the background.  These methods capture the

hotspot/src/share/vm/gc_implementation/concurrentMarkSweep/cmsGCAdaptivePolicyCounters.hpp

@@ -69,7 +69,7 @@ class CMSGCAdaptivePolicyCounters : public GCAdaptivePolicyCounters {
   // end of the sweep of the tenured generation.
   PerfVariable* _avg_cms_free_counter;
   // Average of the free space in the tenured generation at the
-  // start of the sweep of the tenured genertion.
+  // start of the sweep of the tenured generation.
   PerfVariable* _avg_cms_free_at_sweep_counter;
   // Average of the free space in the tenured generation at the
   // after any resizing of the tenured generation at the end

hotspot/src/share/vm/gc_implementation/concurrentMarkSweep/concurrentMarkSweepGeneration.cpp

@@ -4178,7 +4178,7 @@ bool CMSCollector::do_marking_mt(bool asynch) {
     // and is deferred for now; see CR# TBF. 07252005YSR. XXX
     assert(!CMSAbortSemantics || tsk.aborted(), "Inconsistency");
     // If _restart_addr is non-NULL, a marking stack overflow
-    // occured; we need to do a fresh marking iteration from the
+    // occurred; we need to do a fresh marking iteration from the
     // indicated restart address.
     if (_foregroundGCIsActive && asynch) {
       // We may be running into repeated stack overflows, having
@@ -4221,7 +4221,7 @@ bool CMSCollector::do_marking_st(bool asynch) {
   // should be incremental with periodic yields.
   _markBitMap.iterate(&markFromRootsClosure);
   // If _restart_addr is non-NULL, a marking stack overflow
-  // occured; we need to do a fresh iteration from the
+  // occurred; we need to do a fresh iteration from the
   // indicated restart address.
   while (_restart_addr != NULL) {
     if (_foregroundGCIsActive && asynch) {

hotspot/src/share/vm/gc_implementation/g1/g1CollectedHeap.cpp

@@ -2513,7 +2513,7 @@ G1CollectedHeap::do_collection_pause_at_safepoint(HeapRegion* popular_region) {
     }
     save_marks();
 
-    // We must do this before any possible evacuation that should propogate
+    // We must do this before any possible evacuation that should propagate
     // marks, including evacuation of popular objects in a popular pause.
     if (mark_in_progress()) {
       double start_time_sec = os::elapsedTime();

hotspot/src/share/vm/gc_implementation/parallelScavenge/cardTableExtension.cpp

@@ -78,7 +78,7 @@ class CheckForUnmarkedObjects : public ObjectClosure {
   }
 
   // Card marks are not precise. The current system can leave us with
-  // a mismash of precise marks and begining of object marks. This means
+  // a mismash of precise marks and beginning of object marks. This means
   // we test for missing precise marks first. If any are found, we don't
   // fail unless the object head is also unmarked.
   virtual void do_object(oop obj) {
@@ -258,7 +258,7 @@ void CardTableExtension::scavenge_contents_parallel(ObjectStartArray* start_arra
     if (!start_array->object_starts_in_range(slice_start, slice_end)) {
       continue;
     }
-    // Update our begining addr
+    // Update our beginning addr
     HeapWord* first_object = start_array->object_start(slice_start);
     debug_only(oop* first_object_within_slice = (oop*) first_object;)
     if (first_object < slice_start) {

hotspot/src/share/vm/gc_implementation/parallelScavenge/objectStartArray.hpp

@@ -127,7 +127,7 @@ class ObjectStartArray : public CHeapObj {
   // Optimized for finding the first object that crosses into
   // a given block. The blocks contain the offset of the last
   // object in that block. Scroll backwards by one, and the first
-  // object hit should be at the begining of the block
+  // object hit should be at the beginning of the block
   HeapWord* object_start(HeapWord* addr) const {
     assert(_covered_region.contains(addr), "Must be in covered region");
     jbyte* block = block_for_addr(addr);

hotspot/src/share/vm/gc_implementation/parallelScavenge/prefetchQueue.hpp

@@ -26,7 +26,7 @@
 // PrefetchQueue is a FIFO queue of variable length (currently 8).
 //
 // We need to examine the performance penalty of variable lengths.
-// We may also want to split this into cpu dependant bits.
+// We may also want to split this into cpu dependent bits.
 //
 
 const int PREFETCH_QUEUE_SIZE  = 8;

hotspot/src/share/vm/gc_implementation/shared/mutableNUMASpace.cpp

@@ -74,7 +74,7 @@ void MutableNUMASpace::ensure_parsability() {
   for (int i = 0; i < lgrp_spaces()->length(); i++) {
     LGRPSpace *ls = lgrp_spaces()->at(i);
     MutableSpace *s = ls->space();
-    if (s->top() < top()) { // For all spaces preceeding the one containing top()
+    if (s->top() < top()) { // For all spaces preceding the one containing top()
       if (s->free_in_words() > 0) {
         size_t area_touched_words = pointer_delta(s->end(), s->top());
         CollectedHeap::fill_with_object(s->top(), area_touched_words);