Mercurial > hg > graal-jvmci-8
view src/share/vm/gc_implementation/g1/g1MMUTracker.cpp @ 1886:72a161e62cc4
6991377: G1: race between concurrent refinement and humongous object allocation
Summary: There is a race between the concurrent refinement threads and the humongous object allocation that can cause the concurrent refinement threads to corrupt the part of the BOT that it is being initialized by the humongous object allocation operation. The solution is to do the humongous object allocation in careful steps to ensure that the concurrent refinement threads always have a consistent view over the BOT, region contents, and top. The fix includes some very minor tidying up in sparsePRT.
Reviewed-by: jcoomes, johnc, ysr
author | tonyp |
---|---|
date | Sat, 16 Oct 2010 17:12:19 -0400 |
parents | b9bc732be7c0 |
children | f95d63e2154a |
line wrap: on
line source
/* * Copyright (c) 2001, 2009, Oracle and/or its affiliates. All rights reserved. * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. * * This code is free software; you can redistribute it and/or modify it * under the terms of the GNU General Public License version 2 only, as * published by the Free Software Foundation. * * This code is distributed in the hope that it will be useful, but WITHOUT * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License * version 2 for more details (a copy is included in the LICENSE file that * accompanied this code). * * You should have received a copy of the GNU General Public License version * 2 along with this work; if not, write to the Free Software Foundation, * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA. * * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA * or visit www.oracle.com if you need additional information or have any * questions. * */ #include "incls/_precompiled.incl" #include "incls/_g1MMUTracker.cpp.incl" #define _DISABLE_MMU 0 // can't rely on comparing doubles with tolerating a small margin for error #define SMALL_MARGIN 0.0000001 #define is_double_leq_0(_value) ( (_value) < SMALL_MARGIN ) #define is_double_leq(_val1, _val2) is_double_leq_0((_val1) - (_val2)) #define is_double_geq(_val1, _val2) is_double_leq_0((_val2) - (_val1)) /***** ALL TIMES ARE IN SECS!!!!!!! *****/ G1MMUTracker::G1MMUTracker(double time_slice, double max_gc_time) : _time_slice(time_slice), _max_gc_time(max_gc_time) { } G1MMUTrackerQueue::G1MMUTrackerQueue(double time_slice, double max_gc_time) : G1MMUTracker(time_slice, max_gc_time), _head_index(0), _tail_index(trim_index(_head_index+1)), _no_entries(0) { } void G1MMUTrackerQueue::remove_expired_entries(double current_time) { double limit = current_time - _time_slice; while (_no_entries > 0) { if (is_double_geq(limit, _array[_tail_index].end_time())) { _tail_index = trim_index(_tail_index + 1); --_no_entries; } else return; } guarantee(_no_entries == 0, "should have no entries in the array"); } double G1MMUTrackerQueue::calculate_gc_time(double current_time) { double gc_time = 0.0; double limit = current_time - _time_slice; for (int i = 0; i < _no_entries; ++i) { int index = trim_index(_tail_index + i); G1MMUTrackerQueueElem *elem = &_array[index]; if (elem->end_time() > limit) { if (elem->start_time() > limit) gc_time += elem->duration(); else gc_time += elem->end_time() - limit; } } return gc_time; } void G1MMUTrackerQueue::add_pause(double start, double end, bool gc_thread) { double longest_allowed = longest_pause_internal(start); if (longest_allowed < 0.0) longest_allowed = 0.0; double duration = end - start; remove_expired_entries(end); if (_no_entries == QueueLength) { // OK, we've filled up the queue. There are a few ways // of dealing with this "gracefully" // increase the array size (:-) // remove the oldest entry (this might allow more GC time for // the time slice than what's allowed) - this is what we // currently do // consolidate the two entries with the minimum gap between them // (this might allow less GC time than what's allowed) // In the case where ScavengeALot is true, such overflow is not // uncommon; in such cases, we can, without much loss of precision // or performance (we are GC'ing most of the time anyway!), // simply overwrite the oldest entry in the tracker. if (G1PolicyVerbose > 1) { warning("MMU Tracker Queue overflow. Replacing earliest entry."); } _head_index = trim_index(_head_index + 1); assert(_head_index == _tail_index, "Because we have a full circular buffer"); _tail_index = trim_index(_tail_index + 1); } else { _head_index = trim_index(_head_index + 1); ++_no_entries; } _array[_head_index] = G1MMUTrackerQueueElem(start, end); } // basically the _internal call does not remove expired entries // this is for trying things out in the future and a couple // of other places (debugging) double G1MMUTrackerQueue::longest_pause(double current_time) { if (_DISABLE_MMU) return _max_gc_time; MutexLockerEx x(MMUTracker_lock, Mutex::_no_safepoint_check_flag); remove_expired_entries(current_time); return longest_pause_internal(current_time); } double G1MMUTrackerQueue::longest_pause_internal(double current_time) { double target_time = _max_gc_time; while( 1 ) { double gc_time = calculate_gc_time(current_time + target_time); double diff = target_time + gc_time - _max_gc_time; if (!is_double_leq_0(diff)) { target_time -= diff; if (is_double_leq_0(target_time)) { target_time = -1.0; break; } } else { break; } } return target_time; } // basically the _internal call does not remove expired entries // this is for trying things out in the future and a couple // of other places (debugging) double G1MMUTrackerQueue::when_sec(double current_time, double pause_time) { if (_DISABLE_MMU) return 0.0; MutexLockerEx x(MMUTracker_lock, Mutex::_no_safepoint_check_flag); remove_expired_entries(current_time); return when_internal(current_time, pause_time); } double G1MMUTrackerQueue::when_internal(double current_time, double pause_time) { // if the pause is over the maximum, just assume that it's the maximum double adjusted_pause_time = (pause_time > max_gc_time()) ? max_gc_time() : pause_time; double earliest_end = current_time + adjusted_pause_time; double limit = earliest_end - _time_slice; double gc_time = calculate_gc_time(earliest_end); double diff = gc_time + adjusted_pause_time - max_gc_time(); if (is_double_leq_0(diff)) return 0.0; int index = _tail_index; while ( 1 ) { G1MMUTrackerQueueElem *elem = &_array[index]; if (elem->end_time() > limit) { if (elem->start_time() > limit) diff -= elem->duration(); else diff -= elem->end_time() - limit; if (is_double_leq_0(diff)) return elem->end_time() + diff + _time_slice - adjusted_pause_time - current_time; } index = trim_index(index+1); guarantee(index != trim_index(_head_index + 1), "should not go past head"); } }