This commit is contained in:
Bob Vandette 2017-11-16 09:50:49 -05:00
commit 68c5f06bc1
30 changed files with 1946 additions and 411 deletions

View file

@ -29,7 +29,7 @@
#include "classfile/moduleEntry.hpp"
#include "classfile/stringTable.hpp"
#include "classfile/symbolTable.hpp"
#include "gc/shared/cardTableRS.hpp"
#include "gc/shared/gcArguments.hpp"
#include "gc/shared/genCollectedHeap.hpp"
#include "gc/shared/referenceProcessor.hpp"
#include "gc/shared/taskqueue.hpp"
@ -61,11 +61,6 @@
#if INCLUDE_JVMCI
#include "jvmci/jvmciRuntime.hpp"
#endif
#if INCLUDE_ALL_GCS
#include "gc/cms/compactibleFreeListSpace.hpp"
#include "gc/g1/g1CollectedHeap.inline.hpp"
#include "gc/parallel/parallelScavengeHeap.hpp"
#endif // INCLUDE_ALL_GCS
// Note: This is a special bug reporting site for the JVM
#define DEFAULT_VENDOR_URL_BUG "http://bugreport.java.com/bugreport/crash.jsp"
@ -1508,161 +1503,6 @@ void Arguments::set_tiered_flags() {
}
}
#if INCLUDE_ALL_GCS
static void disable_adaptive_size_policy(const char* collector_name) {
if (UseAdaptiveSizePolicy) {
if (FLAG_IS_CMDLINE(UseAdaptiveSizePolicy)) {
warning("Disabling UseAdaptiveSizePolicy; it is incompatible with %s.",
collector_name);
}
FLAG_SET_DEFAULT(UseAdaptiveSizePolicy, false);
}
}
void Arguments::set_parnew_gc_flags() {
assert(!UseSerialGC && !UseParallelOldGC && !UseParallelGC && !UseG1GC,
"control point invariant");
assert(UseConcMarkSweepGC, "CMS is expected to be on here");
if (FLAG_IS_DEFAULT(ParallelGCThreads)) {
FLAG_SET_DEFAULT(ParallelGCThreads, Abstract_VM_Version::parallel_worker_threads());
assert(ParallelGCThreads > 0, "We should always have at least one thread by default");
} else if (ParallelGCThreads == 0) {
jio_fprintf(defaultStream::error_stream(),
"The ParNew GC can not be combined with -XX:ParallelGCThreads=0\n");
vm_exit(1);
}
// By default YoungPLABSize and OldPLABSize are set to 4096 and 1024 respectively,
// these settings are default for Parallel Scavenger. For ParNew+Tenured configuration
// we set them to 1024 and 1024.
// See CR 6362902.
if (FLAG_IS_DEFAULT(YoungPLABSize)) {
FLAG_SET_DEFAULT(YoungPLABSize, (intx)1024);
}
if (FLAG_IS_DEFAULT(OldPLABSize)) {
FLAG_SET_DEFAULT(OldPLABSize, (intx)1024);
}
// When using compressed oops, we use local overflow stacks,
// rather than using a global overflow list chained through
// the klass word of the object's pre-image.
if (UseCompressedOops && !ParGCUseLocalOverflow) {
if (!FLAG_IS_DEFAULT(ParGCUseLocalOverflow)) {
warning("Forcing +ParGCUseLocalOverflow: needed if using compressed references");
}
FLAG_SET_DEFAULT(ParGCUseLocalOverflow, true);
}
assert(ParGCUseLocalOverflow || !UseCompressedOops, "Error");
}
// Adjust some sizes to suit CMS and/or ParNew needs; these work well on
// sparc/solaris for certain applications, but would gain from
// further optimization and tuning efforts, and would almost
// certainly gain from analysis of platform and environment.
void Arguments::set_cms_and_parnew_gc_flags() {
assert(!UseSerialGC && !UseParallelOldGC && !UseParallelGC, "Error");
assert(UseConcMarkSweepGC, "CMS is expected to be on here");
// Turn off AdaptiveSizePolicy by default for cms until it is complete.
disable_adaptive_size_policy("UseConcMarkSweepGC");
set_parnew_gc_flags();
size_t max_heap = align_down(MaxHeapSize,
CardTableRS::ct_max_alignment_constraint());
// Now make adjustments for CMS
intx tenuring_default = (intx)6;
size_t young_gen_per_worker = CMSYoungGenPerWorker;
// Preferred young gen size for "short" pauses:
// upper bound depends on # of threads and NewRatio.
const size_t preferred_max_new_size_unaligned =
MIN2(max_heap/(NewRatio+1), ScaleForWordSize(young_gen_per_worker * ParallelGCThreads));
size_t preferred_max_new_size =
align_up(preferred_max_new_size_unaligned, os::vm_page_size());
// Unless explicitly requested otherwise, size young gen
// for "short" pauses ~ CMSYoungGenPerWorker*ParallelGCThreads
// If either MaxNewSize or NewRatio is set on the command line,
// assume the user is trying to set the size of the young gen.
if (FLAG_IS_DEFAULT(MaxNewSize) && FLAG_IS_DEFAULT(NewRatio)) {
// Set MaxNewSize to our calculated preferred_max_new_size unless
// NewSize was set on the command line and it is larger than
// preferred_max_new_size.
if (!FLAG_IS_DEFAULT(NewSize)) { // NewSize explicitly set at command-line
FLAG_SET_ERGO(size_t, MaxNewSize, MAX2(NewSize, preferred_max_new_size));
} else {
FLAG_SET_ERGO(size_t, MaxNewSize, preferred_max_new_size);
}
log_trace(gc, heap)("CMS ergo set MaxNewSize: " SIZE_FORMAT, MaxNewSize);
// Code along this path potentially sets NewSize and OldSize
log_trace(gc, heap)("CMS set min_heap_size: " SIZE_FORMAT " initial_heap_size: " SIZE_FORMAT " max_heap: " SIZE_FORMAT,
min_heap_size(), InitialHeapSize, max_heap);
size_t min_new = preferred_max_new_size;
if (FLAG_IS_CMDLINE(NewSize)) {
min_new = NewSize;
}
if (max_heap > min_new && min_heap_size() > min_new) {
// Unless explicitly requested otherwise, make young gen
// at least min_new, and at most preferred_max_new_size.
if (FLAG_IS_DEFAULT(NewSize)) {
FLAG_SET_ERGO(size_t, NewSize, MAX2(NewSize, min_new));
FLAG_SET_ERGO(size_t, NewSize, MIN2(preferred_max_new_size, NewSize));
log_trace(gc, heap)("CMS ergo set NewSize: " SIZE_FORMAT, NewSize);
}
// Unless explicitly requested otherwise, size old gen
// so it's NewRatio x of NewSize.
if (FLAG_IS_DEFAULT(OldSize)) {
if (max_heap > NewSize) {
FLAG_SET_ERGO(size_t, OldSize, MIN2(NewRatio*NewSize, max_heap - NewSize));
log_trace(gc, heap)("CMS ergo set OldSize: " SIZE_FORMAT, OldSize);
}
}
}
}
// Unless explicitly requested otherwise, definitely
// promote all objects surviving "tenuring_default" scavenges.
if (FLAG_IS_DEFAULT(MaxTenuringThreshold) &&
FLAG_IS_DEFAULT(SurvivorRatio)) {
FLAG_SET_ERGO(uintx, MaxTenuringThreshold, tenuring_default);
}
// If we decided above (or user explicitly requested)
// `promote all' (via MaxTenuringThreshold := 0),
// prefer minuscule survivor spaces so as not to waste
// space for (non-existent) survivors
if (FLAG_IS_DEFAULT(SurvivorRatio) && MaxTenuringThreshold == 0) {
FLAG_SET_ERGO(uintx, SurvivorRatio, MAX2((uintx)1024, SurvivorRatio));
}
// OldPLABSize is interpreted in CMS as not the size of the PLAB in words,
// but rather the number of free blocks of a given size that are used when
// replenishing the local per-worker free list caches.
if (FLAG_IS_DEFAULT(OldPLABSize)) {
if (!FLAG_IS_DEFAULT(ResizeOldPLAB) && !ResizeOldPLAB) {
// OldPLAB sizing manually turned off: Use a larger default setting,
// unless it was manually specified. This is because a too-low value
// will slow down scavenges.
FLAG_SET_ERGO(size_t, OldPLABSize, CompactibleFreeListSpaceLAB::_default_static_old_plab_size); // default value before 6631166
} else {
FLAG_SET_DEFAULT(OldPLABSize, CompactibleFreeListSpaceLAB::_default_dynamic_old_plab_size); // old CMSParPromoteBlocksToClaim default
}
}
// If either of the static initialization defaults have changed, note this
// modification.
if (!FLAG_IS_DEFAULT(OldPLABSize) || !FLAG_IS_DEFAULT(OldPLABWeight)) {
CompactibleFreeListSpaceLAB::modify_initialization(OldPLABSize, OldPLABWeight);
}
log_trace(gc)("MarkStackSize: %uk MarkStackSizeMax: %uk", (unsigned int) (MarkStackSize / K), (uint) (MarkStackSizeMax / K));
}
#endif // INCLUDE_ALL_GCS
void set_object_alignment() {
// Object alignment.
assert(is_power_of_2(ObjectAlignmentInBytes), "ObjectAlignmentInBytes must be power of 2");
@ -1681,11 +1521,6 @@ void set_object_alignment() {
if (SurvivorAlignmentInBytes == 0) {
SurvivorAlignmentInBytes = ObjectAlignmentInBytes;
}
#if INCLUDE_ALL_GCS
// Set CMS global values
CompactibleFreeListSpace::set_cms_values();
#endif // INCLUDE_ALL_GCS
}
size_t Arguments::max_heap_for_compressed_oops() {
@ -1761,28 +1596,13 @@ void Arguments::set_conservative_max_heap_alignment() {
// the alignments imposed by several sources: any requirements from the heap
// itself, the collector policy and the maximum page size we may run the VM
// with.
size_t heap_alignment = GenCollectedHeap::conservative_max_heap_alignment();
#if INCLUDE_ALL_GCS
if (UseParallelGC) {
heap_alignment = ParallelScavengeHeap::conservative_max_heap_alignment();
} else if (UseG1GC) {
heap_alignment = G1CollectedHeap::conservative_max_heap_alignment();
}
#endif // INCLUDE_ALL_GCS
size_t heap_alignment = GCArguments::arguments()->conservative_max_heap_alignment();
_conservative_max_heap_alignment = MAX4(heap_alignment,
(size_t)os::vm_allocation_granularity(),
os::max_page_size(),
CollectorPolicy::compute_heap_alignment());
}
bool Arguments::gc_selected() {
#if INCLUDE_ALL_GCS
return UseSerialGC || UseParallelGC || UseParallelOldGC || UseConcMarkSweepGC || UseG1GC;
#else
return UseSerialGC;
#endif // INCLUDE_ALL_GCS
}
#ifdef TIERED
bool Arguments::compilation_mode_selected() {
return !FLAG_IS_DEFAULT(TieredCompilation) || !FLAG_IS_DEFAULT(TieredStopAtLevel) ||
@ -1802,31 +1622,6 @@ void Arguments::select_compilation_mode_ergonomically() {
}
#endif //TIERED
void Arguments::select_gc_ergonomically() {
#if INCLUDE_ALL_GCS
if (os::is_server_class_machine()) {
FLAG_SET_ERGO_IF_DEFAULT(bool, UseG1GC, true);
} else {
FLAG_SET_ERGO_IF_DEFAULT(bool, UseSerialGC, true);
}
#else
UNSUPPORTED_OPTION(UseG1GC);
UNSUPPORTED_OPTION(UseParallelGC);
UNSUPPORTED_OPTION(UseParallelOldGC);
UNSUPPORTED_OPTION(UseConcMarkSweepGC);
FLAG_SET_ERGO_IF_DEFAULT(bool, UseSerialGC, true);
#endif // INCLUDE_ALL_GCS
}
void Arguments::select_gc() {
if (!gc_selected()) {
select_gc_ergonomically();
if (!gc_selected()) {
vm_exit_during_initialization("Garbage collector not selected (default collector explicitly disabled)", NULL);
}
}
}
#if INCLUDE_JVMCI
void Arguments::set_jvmci_specific_flags() {
if (UseJVMCICompiler) {
@ -1860,13 +1655,17 @@ void Arguments::set_jvmci_specific_flags() {
}
#endif
void Arguments::set_ergonomics_flags() {
jint Arguments::set_ergonomics_flags() {
#ifdef TIERED
if (!compilation_mode_selected()) {
select_compilation_mode_ergonomically();
}
#endif
select_gc();
jint gc_result = GCArguments::initialize();
if (gc_result != JNI_OK) {
return gc_result;
}
#if COMPILER2_OR_JVMCI
// Shared spaces work fine with other GCs but causes bytecode rewriting
@ -1895,145 +1694,12 @@ void Arguments::set_ergonomics_flags() {
#endif // _LP64
#endif // !ZERO
}
void Arguments::set_parallel_gc_flags() {
assert(UseParallelGC || UseParallelOldGC, "Error");
// Enable ParallelOld unless it was explicitly disabled (cmd line or rc file).
if (FLAG_IS_DEFAULT(UseParallelOldGC)) {
FLAG_SET_DEFAULT(UseParallelOldGC, true);
}
FLAG_SET_DEFAULT(UseParallelGC, true);
// If no heap maximum was requested explicitly, use some reasonable fraction
// of the physical memory, up to a maximum of 1GB.
FLAG_SET_DEFAULT(ParallelGCThreads,
Abstract_VM_Version::parallel_worker_threads());
if (ParallelGCThreads == 0) {
jio_fprintf(defaultStream::error_stream(),
"The Parallel GC can not be combined with -XX:ParallelGCThreads=0\n");
vm_exit(1);
}
if (UseAdaptiveSizePolicy) {
// We don't want to limit adaptive heap sizing's freedom to adjust the heap
// unless the user actually sets these flags.
if (FLAG_IS_DEFAULT(MinHeapFreeRatio)) {
FLAG_SET_DEFAULT(MinHeapFreeRatio, 0);
}
if (FLAG_IS_DEFAULT(MaxHeapFreeRatio)) {
FLAG_SET_DEFAULT(MaxHeapFreeRatio, 100);
}
}
// If InitialSurvivorRatio or MinSurvivorRatio were not specified, but the
// SurvivorRatio has been set, reset their default values to SurvivorRatio +
// 2. By doing this we make SurvivorRatio also work for Parallel Scavenger.
// See CR 6362902 for details.
if (!FLAG_IS_DEFAULT(SurvivorRatio)) {
if (FLAG_IS_DEFAULT(InitialSurvivorRatio)) {
FLAG_SET_DEFAULT(InitialSurvivorRatio, SurvivorRatio + 2);
}
if (FLAG_IS_DEFAULT(MinSurvivorRatio)) {
FLAG_SET_DEFAULT(MinSurvivorRatio, SurvivorRatio + 2);
}
}
if (UseParallelOldGC) {
// Par compact uses lower default values since they are treated as
// minimums. These are different defaults because of the different
// interpretation and are not ergonomically set.
if (FLAG_IS_DEFAULT(MarkSweepDeadRatio)) {
FLAG_SET_DEFAULT(MarkSweepDeadRatio, 1);
}
}
}
void Arguments::set_g1_gc_flags() {
assert(UseG1GC, "Error");
#if defined(COMPILER1) || INCLUDE_JVMCI
FastTLABRefill = false;
#endif
FLAG_SET_DEFAULT(ParallelGCThreads, Abstract_VM_Version::parallel_worker_threads());
if (ParallelGCThreads == 0) {
assert(!FLAG_IS_DEFAULT(ParallelGCThreads), "The default value for ParallelGCThreads should not be 0.");
vm_exit_during_initialization("The flag -XX:+UseG1GC can not be combined with -XX:ParallelGCThreads=0", NULL);
}
#if INCLUDE_ALL_GCS
if (FLAG_IS_DEFAULT(G1ConcRefinementThreads)) {
FLAG_SET_ERGO(uint, G1ConcRefinementThreads, ParallelGCThreads);
}
#endif
// MarkStackSize will be set (if it hasn't been set by the user)
// when concurrent marking is initialized.
// Its value will be based upon the number of parallel marking threads.
// But we do set the maximum mark stack size here.
if (FLAG_IS_DEFAULT(MarkStackSizeMax)) {
FLAG_SET_DEFAULT(MarkStackSizeMax, 128 * TASKQUEUE_SIZE);
}
if (FLAG_IS_DEFAULT(GCTimeRatio) || GCTimeRatio == 0) {
// In G1, we want the default GC overhead goal to be higher than
// it is for PS, or the heap might be expanded too aggressively.
// We set it here to ~8%.
FLAG_SET_DEFAULT(GCTimeRatio, 12);
}
// Below, we might need to calculate the pause time interval based on
// the pause target. When we do so we are going to give G1 maximum
// flexibility and allow it to do pauses when it needs to. So, we'll
// arrange that the pause interval to be pause time target + 1 to
// ensure that a) the pause time target is maximized with respect to
// the pause interval and b) we maintain the invariant that pause
// time target < pause interval. If the user does not want this
// maximum flexibility, they will have to set the pause interval
// explicitly.
if (FLAG_IS_DEFAULT(MaxGCPauseMillis)) {
// The default pause time target in G1 is 200ms
FLAG_SET_DEFAULT(MaxGCPauseMillis, 200);
}
// Then, if the interval parameter was not set, set it according to
// the pause time target (this will also deal with the case when the
// pause time target is the default value).
if (FLAG_IS_DEFAULT(GCPauseIntervalMillis)) {
FLAG_SET_DEFAULT(GCPauseIntervalMillis, MaxGCPauseMillis + 1);
}
log_trace(gc)("MarkStackSize: %uk MarkStackSizeMax: %uk", (unsigned int) (MarkStackSize / K), (uint) (MarkStackSizeMax / K));
return JNI_OK;
}
void Arguments::set_gc_specific_flags() {
#if INCLUDE_ALL_GCS
// Set per-collector flags
if (UseParallelGC || UseParallelOldGC) {
set_parallel_gc_flags();
} else if (UseConcMarkSweepGC) {
set_cms_and_parnew_gc_flags();
} else if (UseG1GC) {
set_g1_gc_flags();
}
if (AssumeMP && !UseSerialGC) {
if (FLAG_IS_DEFAULT(ParallelGCThreads) && ParallelGCThreads == 1) {
warning("If the number of processors is expected to increase from one, then"
" you should configure the number of parallel GC threads appropriately"
" using -XX:ParallelGCThreads=N");
}
}
if (MinHeapFreeRatio == 100) {
// Keeping the heap 100% free is hard ;-) so limit it to 99%.
FLAG_SET_ERGO(uintx, MinHeapFreeRatio, 99);
}
// If class unloading is disabled, also disable concurrent class unloading.
if (!ClassUnloading) {
FLAG_SET_CMDLINE(bool, CMSClassUnloadingEnabled, false);
FLAG_SET_CMDLINE(bool, ClassUnloadingWithConcurrentMark, false);
}
#endif // INCLUDE_ALL_GCS
// Set GC flags
GCArguments::arguments()->initialize_flags();
}
julong Arguments::limit_by_allocatable_memory(julong limit) {
@ -4491,7 +4157,8 @@ jint Arguments::parse(const JavaVMInitArgs* initial_cmd_args) {
jint Arguments::apply_ergo() {
// Set flags based on ergonomics.
set_ergonomics_flags();
jint result = set_ergonomics_flags();
if (result != JNI_OK) return result;
#if INCLUDE_JVMCI
set_jvmci_specific_flags();