Import SamplingProfiler from http://code.google.com/p/dalvik/
cp ~/android/dalvik/profiler/src/main/java/dalvik/profiler/* ~/android/master/libcore/dalvik/src/main/java/dalvik/system/profiler/
cp ~/android/dalvik/profiler/src/test/java/dalvik/profiler/* ~/android/master/libcore/dalvik/src/test/java/dalvik/system/profiler/
perl -p -i -e 's/package dalvik.profiler;/package dalvik.system.profiler;/' ~/android/master/libcore/dalvik/src/*/java/dalvik/system/profiler/*.java
perl -p -i -e 's/import dalvik.profiler./import dalvik.system.profiler./' ~/android/master/libcore/dalvik/src/*/java/dalvik/system/profiler/*.java
Change-Id: Iae3f63327f4a9dde7f037e69c22057e2ea704aaa
diff --git a/Docs.mk b/Docs.mk
index 13eddec..9450139 100644
--- a/Docs.mk
+++ b/Docs.mk
@@ -1,13 +1,29 @@
# List of libcore directories to include in documentation.
# Shared between libcore and frameworks/base.
-libcore_to_document := \
- dalvik/src/main/java/dalvik \
- json/src/main/java \
- junit/src/main/java \
- luni/src/main/java/java \
- luni/src/main/java/javax \
- luni/src/main/java/org/xml/sax \
- luni/src/main/java/org/w3c \
- xml/src/main/java/org/xmlpull/v1 \
+define libcoredoc-all-java-files-under
+$(patsubst ./%,%, \
+ $(shell cd $(1) ; \
+ find $(2) -name "*.java" -and -not -name ".*") \
+ )
+endef
+
+# List of libcore javadoc source files
+#
+# Note dalvik/system is non-recursive to exclude dalvik.system.profiler
#
+# $(1): directory for search (to support use from frameworks/base)
+define libcore_to_document
+ $(call libcoredoc-all-java-files-under,$(1),
+ dalvik/src/main/java/dalvik/system/ -maxdepth 1) \
+ $(call libcoredoc-all-java-files-under,$(1),
+ dalvik/src/main/java/dalvik/annotation \
+ dalvik/src/main/java/dalvik/bytecode \
+ json/src/main/java \
+ junit/src/main/java \
+ luni/src/main/java/java \
+ luni/src/main/java/javax \
+ luni/src/main/java/org/xml/sax \
+ luni/src/main/java/org/w3c \
+ xml/src/main/java/org/xmlpull/v1)
+endef
diff --git a/JavaLibrary.mk b/JavaLibrary.mk
index c49d529..60e5b63 100644
--- a/JavaLibrary.mk
+++ b/JavaLibrary.mk
@@ -294,7 +294,7 @@
# for shared defintion of libcore_to_document
include $(LOCAL_PATH)/Docs.mk
-LOCAL_SRC_FILES:=$(call find-other-java-files, $(libcore_to_document))
+LOCAL_SRC_FILES:=$(call libcore_to_document,$(LOCAL_PATH))
# rerun doc generation without recompiling the java
LOCAL_JAVA_LIBRARIES:=
LOCAL_MODULE_CLASS:=JAVA_LIBRARIES
diff --git a/dalvik/src/main/java/dalvik/system/SamplingProfiler.java b/dalvik/src/main/java/dalvik/system/SamplingProfiler.java
deleted file mode 100644
index b2ee6fb..0000000
--- a/dalvik/src/main/java/dalvik/system/SamplingProfiler.java
+++ /dev/null
@@ -1,1728 +0,0 @@
-/*
- * Copyright (C) 2010 The Android Open Source Project
- *
- * Licensed under the Apache License, Version 2.0 (the "License");
- * you may not use this file except in compliance with the License.
- * You may obtain a copy of the License at
- *
- * http://www.apache.org/licenses/LICENSE-2.0
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-package dalvik.system;
-
-import java.io.BufferedInputStream;
-import java.io.DataInputStream;
-import java.io.DataOutputStream;
-import java.io.EOFException;
-import java.io.File;
-import java.io.FileInputStream;
-import java.io.IOException;
-import java.io.InputStream;
-import java.io.OutputStream;
-import java.io.PrintWriter;
-import java.util.ArrayList;
-import java.util.Arrays;
-import java.util.Collections;
-import java.util.Comparator;
-import java.util.Date;
-import java.util.HashMap;
-import java.util.HashSet;
-import java.util.List;
-import java.util.Map.Entry;
-import java.util.Map;
-import java.util.Set;
-import java.util.Timer;
-import java.util.TimerTask;
-import libcore.util.Objects;
-import libcore.io.IoUtils;
-
-/**
- * A sampling profiler. It currently is implemented without any
- * virtual machine support, relying solely on {@code
- * Thread.getStackTrace} to collect samples. As such, the overhead is
- * higher than a native approach and it does not provide insight into
- * where time is spent within native code, but it can still provide
- * useful insight into where a program is spending time.
- *
- * <h3>Usage Example</h3>
- *
- * The following example shows how to use the {@code
- * SamplingProfiler}. It samples the current thread's stack to a depth
- * of 12 stack frame elements over two different measurement periods
- * with samples taken every 100 milliseconds. In then prints the
- * results in hprof format to the standard output.
- *
- * <pre> {@code
- * ThreadSet threadSet = SamplingProfiler.newArrayThreadSet(Thread.currentThread());
- * SamplingProfiler profiler = new SamplingProfiler(12, threadSet);
- * profiler.start(100);
- * // period of measurement
- * profiler.stop();
- * // period of non-measurement
- * profiler.start(100);
- * // another period of measurement
- * profiler.stop();
- * profiler.shutdown();
- * HprofWriter writer = new AsciiHprofWriter(profiler.getHprofData(), System.out);
- * writer.write();
- * }</pre>
- *
- * @hide
- */
-public final class SamplingProfiler {
-
- /**
- * Represents sampling profiler data. Can be converted to ASCII or
- * binary hprof-style output using {@link AsciiHprofWriter} or
- * {@link BinaryHprofWriter}.
- * <p>
- * The data includes:
- * <ul>
- * <li>the start time of the last sampling period
- * <li>the history of thread start and end events
- * <li>stack traces with frequency counts
- * <ul>
- */
- public static final class HprofData {
-
- public static enum ThreadEventType { START, END };
-
- /**
- * ThreadEvent represents thread creation and death events for
- * reporting. It provides a record of the thread and thread group
- * names for tying samples back to their source thread.
- */
- public static final class ThreadEvent {
-
- public final ThreadEventType type;
- public final int objectId;
- public final int threadId;
- public final String threadName;
- public final String groupName;
- public final String parentGroupName;
-
- public static ThreadEvent start(int objectId, int threadId, String threadName,
- String groupName, String parentGroupName) {
- return new ThreadEvent(ThreadEventType.START, objectId, threadId,
- threadName, groupName, parentGroupName);
- }
-
- public static ThreadEvent end(int threadId) {
- return new ThreadEvent(ThreadEventType.END, threadId);
- }
-
- private ThreadEvent(ThreadEventType type, int objectId, int threadId,
- String threadName, String groupName, String parentGroupName) {
- if (threadName == null) {
- throw new NullPointerException("threadName == null");
- }
- this.type = ThreadEventType.START;
- this.objectId = objectId;
- this.threadId = threadId;
- this.threadName = threadName;
- this.groupName = groupName;
- this.parentGroupName = parentGroupName;
- }
-
- private ThreadEvent(ThreadEventType type, int threadId) {
- this.type = ThreadEventType.END;
- this.objectId = -1;
- this.threadId = threadId;
- this.threadName = null;
- this.groupName = null;
- this.parentGroupName = null;
- }
-
- @Override public int hashCode() {
- int result = 17;
- result = 31 * result + objectId;
- result = 31 * result + threadId;
- result = 31 * result + Objects.hashCode(threadName);
- result = 31 * result + Objects.hashCode(groupName);
- result = 31 * result + Objects.hashCode(parentGroupName);
- return result;
- }
-
- @Override public boolean equals(Object o) {
- if (!(o instanceof ThreadEvent)) {
- return false;
- }
- ThreadEvent event = (ThreadEvent) o;
- return (this.type == event.type
- && this.objectId == event.objectId
- && this.threadId == event.threadId
- && Objects.equal(this.threadName, event.threadName)
- && Objects.equal(this.groupName, event.groupName)
- && Objects.equal(this.parentGroupName, event.parentGroupName));
- }
-
- @Override public String toString() {
- switch (type) {
- case START:
- return String.format(
- "THREAD START (obj=%d, id = %d, name=\"%s\", group=\"%s\")",
- objectId, threadId, threadName, groupName);
- case END:
- return String.format("THREAD END (id = %d)", threadId);
- }
- throw new IllegalStateException(type.toString());
- }
- }
-
- /**
- * A unique stack trace for a specific thread.
- */
- public static final class StackTrace {
-
- public final int stackTraceId;
- private int threadId;
- private StackTraceElement[] stackFrames;
-
- private StackTrace() {
- this.stackTraceId = -1;
- }
-
- public StackTrace(int stackTraceId, int threadId, StackTraceElement[] stackFrames) {
- if (stackFrames == null) {
- throw new NullPointerException("stackFrames == null");
- }
- this.stackTraceId = stackTraceId;
- this.threadId = threadId;
- this.stackFrames = stackFrames;
- }
-
- public int getThreadId() {
- return threadId;
- }
-
- public StackTraceElement[] getStackFrames() {
- return stackFrames;
- }
-
- @Override public int hashCode() {
- int result = 17;
- result = 31 * result + threadId;
- result = 31 * result + Arrays.hashCode(stackFrames);
- return result;
- }
-
- @Override public boolean equals(Object o) {
- if (!(o instanceof StackTrace)) {
- return false;
- }
- StackTrace s = (StackTrace) o;
- return threadId == s.threadId && Arrays.equals(stackFrames, s.stackFrames);
- }
-
- @Override public String toString() {
- StringBuilder frames = new StringBuilder();
- if (stackFrames.length > 0) {
- frames.append('\n');
- for (StackTraceElement stackFrame : stackFrames) {
- frames.append("\t at ");
- frames.append(stackFrame);
- frames.append('\n');
- }
- } else {
- frames.append("<empty>");
- }
- return "StackTrace[stackTraceId=" + stackTraceId
- + ", threadId=" + threadId
- + ", frames=" + frames + "]";
-
- }
- }
-
- /**
- * A read only container combining a stack trace with its frequency.
- */
- public static final class Sample {
-
- public final StackTrace stackTrace;
- public final int count;
-
- private Sample(StackTrace stackTrace, int count) {
- if (stackTrace == null) {
- throw new NullPointerException("stackTrace == null");
- }
- if (count < 0) {
- throw new IllegalArgumentException("count < 0:" + count);
- }
- this.stackTrace = stackTrace;
- this.count = count;
- }
-
- @Override public int hashCode() {
- int result = 17;
- result = 31 * result + stackTrace.hashCode();
- result = 31 * result + count;
- return result;
- }
-
- @Override public boolean equals(Object o) {
- Sample s = (Sample) o;
- return stackTrace.equals(s.stackTrace) && count == s.count;
- }
-
- @Override public String toString() {
- return "Sample[count=" + count + " " + stackTrace + "]";
- }
-
- }
-
- /**
- * Start of last sampling period.
- */
- private long startMillis;
-
- /**
- * CONTROL_SETTINGS flags
- */
- private int flags;
-
- /**
- * stack sampling depth
- */
- private int depth;
-
- /**
- * List of thread creation and death events.
- */
- private final List<ThreadEvent> threadHistory = new ArrayList<ThreadEvent>();
-
- /**
- * Map of thread id to a start ThreadEvent
- */
- private final Map<Integer, ThreadEvent> threadIdToThreadEvent
- = new HashMap<Integer, ThreadEvent>();
-
- /**
- * Map of stack traces to a mutable sample count. The map is
- * provided by the creator of the HprofData so only have
- * mutable access to the int[] cells that contain the sample
- * count. Only an unmodifiable iterator view is available to
- * users of the HprofData.
- */
- private final Map<HprofData.StackTrace, int[]> stackTraces;
-
- public HprofData(Map<StackTrace, int[]> stackTraces) {
- if (stackTraces == null) {
- throw new NullPointerException("stackTraces == null");
- }
- this.stackTraces = stackTraces;
- }
-
- /**
- * The start time in milliseconds of the last profiling period.
- */
- public long getStartMillis() {
- return startMillis;
- }
-
- /**
- * Set the time for the start of the current sampling period.
- */
- public void setStartMillis(long startMillis) {
- this.startMillis = startMillis;
- }
-
- /**
- * Get the {@link ControlSettings} flags
- */
- public int getFlags() {
- return flags;
- }
-
- /**
- * Set the {@link ControlSettings} flags
- */
- public void setFlags(int flags) {
- this.flags = flags;
- }
-
- /**
- * Get the stack sampling depth
- */
- public int getDepth() {
- return depth;
- }
-
- /**
- * Set the stack sampling depth
- */
- public void setDepth(int depth) {
- this.depth = depth;
- }
-
- /**
- * Return an unmodifiable history of start and end thread events.
- */
- public List<ThreadEvent> getThreadHistory() {
- return Collections.unmodifiableList(threadHistory);
- }
-
- /**
- * Return a new set containing the current sample data.
- */
- public Set<Sample> getSamples() {
- Set<Sample> samples = new HashSet<Sample>(stackTraces.size());
- for (Entry<StackTrace, int[]> e : stackTraces.entrySet()) {
- StackTrace stackTrace = e.getKey();
- int countCell[] = e.getValue();
- int count = countCell[0];
- Sample sample = new Sample(stackTrace, count);
- samples.add(sample);
- }
- return samples;
- }
-
- /**
- * Record an event in the thread history.
- */
- public void addThreadEvent(ThreadEvent event) {
- if (event == null) {
- throw new NullPointerException("event == null");
- }
- ThreadEvent old = threadIdToThreadEvent.put(event.threadId, event);
- switch (event.type) {
- case START:
- if (old != null) {
- throw new IllegalArgumentException("ThreadEvent already registered for id "
- + event.threadId);
- }
- break;
- case END:
- // Do not assert that the END_THREAD matches a
- // START_THREAD unless in strict mode. While thhis
- // hold true in the binary hprof BinaryHprofWriter
- // produces, it is not true of hprof files created
- // by the RI. However, if there is an event
- // already registed for a thread id, it should be
- // the matching start, not a duplicate end.
- if (old != null && old.type == ThreadEventType.END) {
- throw new IllegalArgumentException("Duplicate ThreadEvent.end for id "
- + event.threadId);
- }
- break;
- }
- threadHistory.add(event);
- }
-
- /**
- * Record an stack trace and an associated int[] cell of
- * sample cound for the stack trace. The caller is allowed
- * retain a pointer to the cell to update the count. The
- * SamplingProfiler intentionally does not present a mutable
- * view of the count.
- */
- public void addStackTrace(StackTrace stackTrace, int[] countCell) {
- if (!threadIdToThreadEvent.containsKey(stackTrace.threadId)) {
- throw new IllegalArgumentException("Unknown thread id " + stackTrace.threadId);
- }
- int[] old = stackTraces.put(stackTrace, countCell);
- if (old != null) {
- throw new IllegalArgumentException("StackTrace already registered for id "
- + stackTrace.stackTraceId + ":\n" + stackTrace);
- }
- }
- }
-
- public static interface HprofWriter {
- public void write() throws IOException;
- }
-
- public static final class AsciiHprofWriter implements HprofWriter {
-
- private final HprofData data;
- private final PrintWriter out;
-
- public AsciiHprofWriter(HprofData data, OutputStream outputStream) {
- this.data = data;
- this.out = new PrintWriter(outputStream);
- }
-
- public void write() throws IOException {
- for (HprofData.ThreadEvent e : data.getThreadHistory()) {
- out.println(e);
- }
-
- List<HprofData.Sample> samples
- = new ArrayList<HprofData.Sample>(data.getSamples());
- Collections.sort(samples, SAMPLE_COMPARATOR);
- int total = 0;
- for (HprofData.Sample sample : samples) {
- HprofData.StackTrace stackTrace = sample.stackTrace;
- int count = sample.count;
- total += count;
- out.printf("TRACE %d: (thread=%d)\n",
- stackTrace.stackTraceId,
- stackTrace.threadId);
- for (StackTraceElement e : stackTrace.stackFrames) {
- out.printf("\t%s\n", e);
- }
- }
- Date now = new Date(data.getStartMillis());
- // "CPU SAMPLES BEGIN (total = 826) Wed Jul 21 12:03:46 2010"
- out.printf("CPU SAMPLES BEGIN (total = %d) %ta %tb %td %tT %tY\n",
- total, now, now, now, now, now);
- out.printf("rank self accum count trace method\n");
- int rank = 0;
- double accum = 0;
- for (HprofData.Sample sample : samples) {
- rank++;
- HprofData.StackTrace stackTrace = sample.stackTrace;
- int count = sample.count;
- double self = (double)count/(double)total;
- accum += self;
-
- // " 1 65.62% 65.62% 542 300302 java.lang.Long.parseLong"
- out.printf("% 4d% 6.2f%%% 6.2f%% % 7d % 5d %s.%s\n",
- rank, self*100, accum*100, count, stackTrace.stackTraceId,
- stackTrace.stackFrames[0].getClassName(),
- stackTrace.stackFrames[0].getMethodName());
- }
- out.printf("CPU SAMPLES END\n");
- out.flush();
- }
-
- private static final Comparator<HprofData.Sample> SAMPLE_COMPARATOR
- = new Comparator<HprofData.Sample>() {
- public int compare(HprofData.Sample s1, HprofData.Sample s2) {
- return s2.count - s1.count;
- }
- };
- }
-
-
- /**
- * Hprof binary format related constants shared between the
- * BinaryHprofReader and BinaryHprofWriter.
- */
- public static class BinaryHprof {
- /**
- * Currently code only supports 4 byte id size.
- */
- public static final int ID_SIZE = 4;
-
- public static enum Tag {
-
- STRING_IN_UTF8(0x01, -ID_SIZE),
- LOAD_CLASS(0x02, 4 + ID_SIZE + 4 + ID_SIZE),
- UNLOAD_CLASS(0x03, 4),
- STACK_FRAME(0x04, ID_SIZE + ID_SIZE + ID_SIZE + ID_SIZE + 4 + 4),
- STACK_TRACE(0x05, -(4 + 4 + 4)),
- ALLOC_SITES(0x06, -(2 + 4 + 4 + 4 + 8 + 8 + 4)),
- HEAP_SUMMARY(0x07, 4 + 4 + 8 + 8),
- START_THREAD(0x0a, 4 + ID_SIZE + 4 + ID_SIZE + ID_SIZE + ID_SIZE),
- END_THREAD(0x0b, 4),
- HEAP_DUMP(0x0c, -0),
- HEAP_DUMP_SEGMENT(0x1c, -0),
- HEAP_DUMP_END(0x2c, 0),
- CPU_SAMPLES(0x0d, -(4 + 4)),
- CONTROL_SETTINGS(0x0e, 4 + 2);
-
- public final byte tag;
-
- /**
- * Minimum size in bytes.
- */
- public final int minimumSize;
-
- /**
- * Maximum size in bytes. 0 mean no specific limit.
- */
- public final int maximumSize;
-
- private Tag(int tag, int size) {
- this.tag = (byte) tag;
- if (size > 0) {
- // fixed size, max and min the same
- this.minimumSize = size;
- this.maximumSize = size;
- } else {
- // only minimum bound
- this.minimumSize = -size;
- this.maximumSize = 0;
- }
- }
-
- private static final Map<Byte, Tag> BYTE_TO_TAG
- = new HashMap<Byte, Tag>();
-
- static {
- for (Tag v : Tag.values()) {
- BYTE_TO_TAG.put(v.tag, v);
- }
- }
-
- public static Tag get(byte tag) {
- return BYTE_TO_TAG.get(tag);
- }
-
- /**
- * Returns null if the actual size meets expectations, or a
- * String error message if not.
- */
- public String checkSize(int actual) {
- if (actual < minimumSize) {
- return "expected a minimial record size of " + minimumSize + " for " + this
- + " but received " + actual;
- }
- if (maximumSize == 0) {
- return null;
- }
- if (actual > maximumSize) {
- return "expected a maximum record size of " + maximumSize + " for " + this
- + " but received " + actual;
- }
- return null;
- }
- }
-
- public static enum ControlSettings {
- ALLOC_TRACES(0x01),
- CPU_SAMPLING(0x02);
-
- public final int bitmask;
-
- private ControlSettings(int bitmask) {
- this.bitmask = bitmask;
- }
- }
-
- }
-
- public static final class BinaryHprofWriter implements HprofWriter {
-
- private int nextStringId = 1; // id 0 => null
- private int nextClassId = 1;
- private int nextStackFrameId = 1;
- private final Map<String, Integer> stringToId = new HashMap<String, Integer>();
- private final Map<String, Integer> classNameToId = new HashMap<String, Integer>();
- private final Map<StackTraceElement, Integer> stackFrameToId
- = new HashMap<StackTraceElement, Integer>();
-
- private final HprofData data;
- private final DataOutputStream out;
-
- public BinaryHprofWriter(HprofData data, OutputStream outputStream) {
- this.data = data;
- this.out = new DataOutputStream(outputStream);
- }
-
- public void write() throws IOException {
- try {
- writeHeader(data.getStartMillis());
-
- writeControlSettings(data.getFlags(), data.getDepth());
-
- for (HprofData.ThreadEvent event : data.getThreadHistory()) {
- writeThreadEvent(event);
- }
-
- Set<HprofData.Sample> samples = data.getSamples();
- int total = 0;
- for (HprofData.Sample sample : samples) {
- total += sample.count;
- writeStackTrace(sample.stackTrace);
- }
- writeCpuSamples(total, samples);
-
- } finally {
- out.flush();
- }
- }
-
- private void writeHeader(long dumpTimeInMilliseconds) throws IOException {
- out.writeBytes("JAVA PROFILE 1.0.2");
- out.writeByte(0); // null terminated string
- out.writeInt(BinaryHprof.ID_SIZE);
- out.writeLong(dumpTimeInMilliseconds);
- }
-
- private void writeControlSettings(int flags, int depth) throws IOException {
- if (depth > Short.MAX_VALUE) {
- throw new IllegalArgumentException("depth too large for binary hprof: "
- + depth + " > " + Short.MAX_VALUE);
- }
- writeRecordHeader(BinaryHprof.Tag.CONTROL_SETTINGS,
- 0,
- BinaryHprof.Tag.CONTROL_SETTINGS.maximumSize);
- out.writeInt(flags);
- out.writeShort((short) depth);
- }
-
- private void writeThreadEvent(HprofData.ThreadEvent e) throws IOException {
- switch (e.type) {
- case START:
- writeStartThread(e);
- return;
- case END:
- writeStopThread(e);
- return;
- }
- throw new IllegalStateException(e.type.toString());
- }
-
- private void writeStartThread(HprofData.ThreadEvent e) throws IOException {
- int threadNameId = writeString(e.threadName);
- int groupNameId = writeString(e.groupName);
- int parentGroupNameId = writeString(e.parentGroupName);
- writeRecordHeader(BinaryHprof.Tag.START_THREAD,
- 0,
- BinaryHprof.Tag.START_THREAD.maximumSize);
- out.writeInt(e.threadId);
- writeId(e.objectId);
- out.writeInt(0); // stack trace where thread was started unavailable
- writeId(threadNameId);
- writeId(groupNameId);
- writeId(parentGroupNameId);
- }
-
- private void writeStopThread(HprofData.ThreadEvent e) throws IOException {
- writeRecordHeader(BinaryHprof.Tag.END_THREAD,
- 0,
- BinaryHprof.Tag.END_THREAD.maximumSize);
- out.writeInt(e.threadId);
- }
-
- private void writeRecordHeader(BinaryHprof.Tag hprofTag,
- int timeDeltaInMicroseconds,
- int recordLength) throws IOException {
- String error = hprofTag.checkSize(recordLength);
- if (error != null) {
- throw new AssertionError(error);
- }
- out.writeByte(hprofTag.tag);
- out.writeInt(timeDeltaInMicroseconds);
- out.writeInt(recordLength);
- }
-
- private void writeId(int id) throws IOException {
- out.writeInt(id);
- }
-
- /**
- * Ensures that a string has been writen to the out and
- * returns its ID. The ID of a null string is zero, and
- * doesn't actually result in any output. In a string has
- * already been written previously, the earlier ID will be
- * returned and no output will be written.
- */
- private int writeString(String string) throws IOException {
- if (string == null) {
- return 0;
- }
- Integer identifier = stringToId.get(string);
- if (identifier != null) {
- return identifier;
- }
-
- int id = nextStringId++;
- stringToId.put(string, id);
-
- byte[] bytes = string.getBytes("UTF-8");
- writeRecordHeader(BinaryHprof.Tag.STRING_IN_UTF8,
- 0,
- BinaryHprof.ID_SIZE + bytes.length);
- out.writeInt(id);
- out.write(bytes, 0, bytes.length);
-
- return id;
- }
-
- private void writeCpuSamples(int totalSamples, Set<HprofData.Sample> samples)
- throws IOException {
- int samplesCount = samples.size();
- if (samplesCount == 0) {
- return;
- }
- writeRecordHeader(BinaryHprof.Tag.CPU_SAMPLES, 0, 4 + 4 + (samplesCount * (4 + 4)));
- out.writeInt(totalSamples);
- out.writeInt(samplesCount);
- for (HprofData.Sample sample : samples) {
- out.writeInt(sample.count);
- out.writeInt(sample.stackTrace.stackTraceId);
- }
- }
-
- private void writeStackTrace(HprofData.StackTrace stackTrace) throws IOException {
- int frames = stackTrace.stackFrames.length;
- int[] stackFrameIds = new int[frames];
- for (int i = 0; i < frames; i++) {
- stackFrameIds[i] = writeStackFrame(stackTrace.stackFrames[i]);
- }
- writeRecordHeader(BinaryHprof.Tag.STACK_TRACE,
- 0,
- 4 + 4 + 4 + (frames * BinaryHprof.ID_SIZE));
- out.writeInt(stackTrace.stackTraceId);
- out.writeInt(stackTrace.threadId);
- out.writeInt(frames);
- for (int stackFrameId : stackFrameIds) {
- writeId(stackFrameId);
- }
- }
-
- private int writeLoadClass(String className) throws IOException {
- Integer identifier = classNameToId.get(className);
- if (identifier != null) {
- return identifier;
- }
- int id = nextClassId++;
- classNameToId.put(className, id);
-
- int classNameId = writeString(className);
- writeRecordHeader(BinaryHprof.Tag.LOAD_CLASS,
- 0,
- BinaryHprof.Tag.LOAD_CLASS.maximumSize);
- out.writeInt(id);
- writeId(0); // class object ID
- out.writeInt(0); // stack trace where class was loaded is unavailable
- writeId(classNameId);
-
- return id;
- }
-
- private int writeStackFrame(StackTraceElement stackFrame) throws IOException {
- Integer identifier = stackFrameToId.get(stackFrame);
- if (identifier != null) {
- return identifier;
- }
-
- int id = nextStackFrameId++;
- stackFrameToId.put(stackFrame, id);
-
- int classId = writeLoadClass(stackFrame.getClassName());
- int methodNameId = writeString(stackFrame.getMethodName());
- int sourceId = writeString(stackFrame.getFileName());
- writeRecordHeader(BinaryHprof.Tag.STACK_FRAME,
- 0,
- BinaryHprof.Tag.STACK_FRAME.maximumSize);
- writeId(id);
- writeId(methodNameId);
- writeId(0); // method signature is unavailable from StackTraceElement
- writeId(sourceId);
- out.writeInt(classId);
- out.writeInt(stackFrame.getLineNumber());
-
- return id;
- }
-
- public void close() throws IOException {
- out.close();
- }
- }
-
- /**
- * <pre> {@code
- * BinaryHprofReader reader = new BinaryHprofReader(new BufferedInputStream(inputStream));
- * reader.setStrict(false); // for RI compatability
- * reader.read();
- * inputStream.close();
- * reader.getVersion();
- * reader.getHprofData();
- * }</pre>
- */
- public static final class BinaryHprofReader {
-
- private static final boolean TRACE = false;
-
- private final DataInputStream in;
-
- /**
- * By default we try to strictly validate rules followed by
- * our HprofWriter. For example, every end thread is preceded
- * by a matching start thread.
- */
- private boolean strict = true;
-
- /**
- * version string from header after read has been performed,
- * otherwise null. nullness used to detect if callers try to
- * access data before read is called.
- */
- private String version;
-
- private final Map<HprofData.StackTrace, int[]> stackTraces
- = new HashMap<HprofData.StackTrace, int[]>();
-
- private final HprofData hprofData = new HprofData(stackTraces);
-
- private final Map<Integer, String> idToString = new HashMap<Integer, String>();
- private final Map<Integer, String> idToClassName = new HashMap<Integer, String>();
- private final Map<Integer, StackTraceElement> idToStackFrame
- = new HashMap<Integer, StackTraceElement>();
- private final Map<Integer, HprofData.StackTrace> idToStackTrace
- = new HashMap<Integer, HprofData.StackTrace>();
-
- /**
- * Creates a BinaryHprofReader around the specified {@code
- * inputStream}
- */
- public BinaryHprofReader(InputStream inputStream) throws IOException {
- this.in = new DataInputStream(inputStream);
- }
-
- public boolean getStrict () {
- return strict;
- }
-
- public void setStrict (boolean strict) {
- if (version != null) {
- throw new IllegalStateException("cannot set strict after read()");
- }
- this.strict = strict;
- }
-
- /**
- * throws IllegalStateException if read() has not been called.
- */
- private void checkRead() {
- if (version == null) {
- throw new IllegalStateException("data access before read()");
- }
- }
-
- public String getVersion() {
- checkRead();
- return version;
- }
-
- public HprofData getHprofData() {
- checkRead();
- return hprofData;
- }
-
- /**
- * Read the hprof header and records from the input
- */
- public void read() throws IOException {
- parseHeader();
- parseRecords();
- }
-
- private void parseHeader() throws IOException {
- if (TRACE) {
- System.out.println("hprofTag=HEADER");
- }
- parseVersion();
- parseIdSize();
- parseTime();
- }
-
- private void parseVersion() throws IOException {
- byte[] bytes = new byte[512];
- for (int i = 0; i < bytes.length; i++) {
- byte b = in.readByte();
- if (b == '\0') {
- String version = new String(bytes, 0, i, "UTF-8");
- if (TRACE) {
- System.out.println("\tversion=" + version);
- }
- if (!version.startsWith("JAVA PROFILE ")) {
- throw new MalformedHprofException("Unexpected version: " + version);
- }
- this.version = version;
- return;
- }
- bytes[i] = b;
- }
- throw new MalformedHprofException("Could not find HPROF version");
- }
-
- private void parseIdSize() throws IOException {
- int idSize = in.readInt();
- if (TRACE) {
- System.out.println("\tidSize=" + idSize);
- }
- if (idSize != BinaryHprof.ID_SIZE) {
- throw new MalformedHprofException("Unsupported identifier size: " + idSize);
- }
- }
-
- private void parseTime() throws IOException {
- long time = in.readLong();
- if (TRACE) {
- System.out.println("\ttime=" + Long.toHexString(time) + " " + new Date(time));
- }
- hprofData.setStartMillis(time);
- }
-
- private void parseRecords() throws IOException {
- while (parseRecord()) {
- ;
- }
- }
-
- /**
- * Read and process the next record. Returns true if a
- * record was handled, false on EOF.
- */
- private boolean parseRecord() throws IOException {
- int tagOrEOF = in.read();
- if (tagOrEOF == -1) {
- return false;
- }
- byte tag = (byte) tagOrEOF;
- int timeDeltaInMicroseconds = in.readInt();
- int recordLength = in.readInt();
- BinaryHprof.Tag hprofTag = BinaryHprof.Tag.get(tag);
- if (TRACE) {
- System.out.println("hprofTag=" + hprofTag);
- }
- if (hprofTag == null) {
- skipRecord(hprofTag, recordLength);
- return true;
- }
- String error = hprofTag.checkSize(recordLength);
- if (error != null) {
- throw new MalformedHprofException(error);
- }
- switch (hprofTag) {
- case CONTROL_SETTINGS:
- parseControlSettings();
- return true;
-
- case STRING_IN_UTF8:
- parseStringInUtf8(recordLength);
- return true;
-
- case START_THREAD:
- parseStartThread();
- return true;
- case END_THREAD:
- parseEndThread();
- return true;
-
- case LOAD_CLASS:
- parseLoadClass();
- return true;
- case STACK_FRAME:
- parseStackFrame();
- return true;
- case STACK_TRACE:
- parseStackTrace(recordLength);
- return true;
-
- case CPU_SAMPLES:
- parseCpuSamples(recordLength);
- return true;
-
- case UNLOAD_CLASS:
- case ALLOC_SITES:
- case HEAP_SUMMARY:
- case HEAP_DUMP:
- case HEAP_DUMP_SEGMENT:
- case HEAP_DUMP_END:
- default:
- skipRecord(hprofTag, recordLength);
- return true;
- }
- }
-
- private void skipRecord(BinaryHprof.Tag hprofTag, long recordLength) throws IOException {
- if (TRACE) {
- System.out.println("\tskipping recordLength=" + recordLength);
- }
- long skipped = in.skip(recordLength);
- if (skipped != recordLength) {
- throw new EOFException("Expected to skip " + recordLength
- + " bytes but only skipped " + skipped + " bytes");
- }
- }
-
- private void parseControlSettings() throws IOException {
- int flags = in.readInt();
- short depth = in.readShort();
- if (TRACE) {
- System.out.println("\tflags=" + Integer.toHexString(flags));
- System.out.println("\tdepth=" + depth);
- }
- hprofData.setFlags(flags);
- hprofData.setDepth(depth);
- }
-
- private void parseStringInUtf8(int recordLength) throws IOException {
- int stringId = in.readInt();
- byte[] bytes = new byte[recordLength - BinaryHprof.ID_SIZE];
- in.read(bytes);
- String string = new String(bytes, "UTF-8");
- if (TRACE) {
- System.out.println("\tstring=" + string);
- }
- String old = idToString.put(stringId, string);
- if (old != null) {
- throw new MalformedHprofException("Duplicate string id: " + stringId);
- }
- }
-
- private void parseLoadClass() throws IOException {
- int classId = in.readInt();
- int classObjectId = readId();
- // serial number apparently not a stack trace id. (int vs ID)
- // we don't use this field.
- int stackTraceSerialNumber = in.readInt();
- String className = readString();
- if (TRACE) {
- System.out.println("\tclassId=" + classId);
- System.out.println("\tclassObjectId=" + classObjectId);
- System.out.println("\tstackTraceSerialNumber=" + stackTraceSerialNumber);
- System.out.println("\tclassName=" + className);
- }
- String old = idToClassName.put(classId, className);
- if (old != null) {
- throw new MalformedHprofException("Duplicate class id: " + classId);
- }
- }
-
- private int readId() throws IOException {
- return in.readInt();
- }
-
- private String readString() throws IOException {
- int id = readId();
- if (id == 0) {
- return null;
- }
- String string = idToString.get(id);
- if (string == null) {
- throw new MalformedHprofException("Unknown string id " + id);
- }
- return string;
- }
-
- private String readClass() throws IOException {
- int id = readId();
- String string = idToClassName.get(id);
- if (string == null) {
- throw new MalformedHprofException("Unknown class id " + id);
- }
- return string;
- }
-
- private void parseStartThread() throws IOException {
- int threadId = in.readInt();
- int objectId = readId();
- // stack trace where thread was created.
- // serial number apparently not a stack trace id. (int vs ID)
- // we don't use this field.
- int stackTraceSerialNumber = in.readInt();
- String threadName = readString();
- String groupName = readString();
- String parentGroupName = readString();
- if (TRACE) {
- System.out.println("\tthreadId=" + threadId);
- System.out.println("\tobjectId=" + objectId);
- System.out.println("\tstackTraceSerialNumber=" + stackTraceSerialNumber);
- System.out.println("\tthreadName=" + threadName);
- System.out.println("\tgroupName=" + groupName);
- System.out.println("\tparentGroupName=" + parentGroupName);
- }
- HprofData.ThreadEvent event
- = HprofData.ThreadEvent.start(objectId, threadId,
- threadName, groupName, parentGroupName);
- hprofData.addThreadEvent(event);
- }
-
- private void parseEndThread() throws IOException {
- int threadId = in.readInt();
- if (TRACE) {
- System.out.println("\tthreadId=" + threadId);
- }
- HprofData.ThreadEvent event = HprofData.ThreadEvent.end(threadId);
- hprofData.addThreadEvent(event);
- }
-
- private void parseStackFrame() throws IOException {
- int stackFrameId = readId();
- String methodName = readString();
- String methodSignature = readString();
- String file = readString();
- String className = readClass();
- int line = in.readInt();
- if (TRACE) {
- System.out.println("\tstackFrameId=" + stackFrameId);
- System.out.println("\tclassName=" + className);
- System.out.println("\tmethodName=" + methodName);
- System.out.println("\tmethodSignature=" + methodSignature);
- System.out.println("\tfile=" + file);
- System.out.println("\tline=" + line);
- }
- StackTraceElement stackFrame = new StackTraceElement(className, methodName, file, line);
- StackTraceElement old = idToStackFrame.put(stackFrameId, stackFrame);
- if (old != null) {
- throw new MalformedHprofException("Duplicate stack frame id: " + stackFrameId);
- }
- }
-
- private void parseStackTrace(int recordLength) throws IOException {
- int stackTraceId = in.readInt();
- int threadId = in.readInt();
- int frames = in.readInt();
- if (TRACE) {
- System.out.println("\tstackTraceId=" + stackTraceId);
- System.out.println("\tthreadId=" + threadId);
- System.out.println("\tframes=" + frames);
- }
- int expectedLength = 4 + 4 + 4 + (frames * BinaryHprof.ID_SIZE);
- if (recordLength != expectedLength) {
- throw new MalformedHprofException("Expected stack trace record of size "
- + expectedLength
- + " based on number of frames but header "
- + "specified a length of " + recordLength);
- }
- StackTraceElement[] stackFrames = new StackTraceElement[frames];
- for (int i = 0; i < frames; i++) {
- int stackFrameId = readId();
- StackTraceElement stackFrame = idToStackFrame.get(stackFrameId);
- if (TRACE) {
- System.out.println("\tstackFrameId=" + stackFrameId);
- System.out.println("\tstackFrame=" + stackFrame);
- }
- if (stackFrame == null) {
- throw new MalformedHprofException("Unknown stack frame id " + stackFrameId);
- }
- stackFrames[i] = stackFrame;
- }
-
- HprofData.StackTrace stackTrace
- = new HprofData.StackTrace(stackTraceId, threadId, stackFrames);
- if (strict) {
- hprofData.addStackTrace(stackTrace, new int[1]);
- } else {
- // The RI can have duplicate stacks, presumably they
- // have a minor race if two samples with the same
- // stack are taken around the same time. if we have a
- // duplicate, just skip adding it to hprofData, but
- // register it locally in idToStackFrame. if it seen
- // in CPU_SAMPLES, we will find a StackTrace is equal
- // to the first, so they will share a countCell.
- int[] countCell = stackTraces.get(stackTrace);
- if (countCell == null) {
- hprofData.addStackTrace(stackTrace, new int[1]);
- }
- }
-
- HprofData.StackTrace old = idToStackTrace.put(stackTraceId, stackTrace);
- if (old != null) {
- throw new MalformedHprofException("Duplicate stack trace id: " + stackTraceId);
- }
-
- }
-
- private void parseCpuSamples(int recordLength) throws IOException {
- int totalSamples = in.readInt();
- int samplesCount = in.readInt();
- if (TRACE) {
- System.out.println("\ttotalSamples=" + totalSamples);
- System.out.println("\tsamplesCount=" + samplesCount);
- }
- int expectedLength = 4 + 4 + (samplesCount * (4 + 4));
- if (recordLength != expectedLength) {
- throw new MalformedHprofException("Expected CPU samples record of size "
- + expectedLength
- + " based on number of samples but header "
- + "specified a length of " + recordLength);
- }
- int total = 0;
- for (int i = 0; i < samplesCount; i++) {
- int count = in.readInt();
- int stackTraceId = in.readInt();
- if (TRACE) {
- System.out.println("\tcount=" + count);
- System.out.println("\tstackTraceId=" + stackTraceId);
- }
- HprofData.StackTrace stackTrace = idToStackTrace.get(stackTraceId);
- if (stackTrace == null) {
- throw new MalformedHprofException("Unknown stack trace id " + stackTraceId);
- }
- if (count == 0) {
- throw new MalformedHprofException("Zero sample count for stack trace "
- + stackTrace);
- }
- int[] countCell = stackTraces.get(stackTrace);
- if (strict) {
- if (countCell[0] != 0) {
- throw new MalformedHprofException("Setting sample count of stack trace "
- + stackTrace + " to " + count
- + " found it was already initialized to "
- + countCell[0]);
- }
- } else {
- // Coalesce counts from duplicate stack traces.
- // For more on this, see comments in parseStackTrace.
- count += countCell[0];
- }
- countCell[0] = count;
- total += count;
- }
- if (strict && totalSamples != total) {
- throw new MalformedHprofException("Expected a total of " + totalSamples
- + " samples but saw " + total);
- }
- }
- }
-
- public static final class MalformedHprofException extends IOException {
- private MalformedHprofException(String message) {
- super(message);
- }
- private MalformedHprofException(String message, Throwable cause) {
- super(message, cause);
- }
- private MalformedHprofException(Throwable cause) {
- super(cause);
- }
- }
-
- /**
- * Run on device with:
- * adb shell dalvikvm 'dalvik.system.SamplingProfiler\$HprofBinaryToAscii'
- *
- * Run on host with:
- * java -classpath out/target/common/obj/JAVA_LIBRARIES/core_intermediates/classes.jar
- */
- public static final class HprofBinaryToAscii {
-
- public static void main(String[] args) {
- System.exit(convert(args) ? 0 : 1);
- }
-
- public static boolean convert(String[] args) {
-
- if (args.length != 1) {
- usage("binary hprof file argument expected");
- return false;
- }
- File file = new File(args[0]);
- if (!file.exists()) {
- usage("file " + file + " does not exist");
- return false;
- }
-
- HprofData hprofData;
- InputStream inputStream = null;
- try {
- inputStream = new BufferedInputStream(new FileInputStream(file));
- BinaryHprofReader reader = new BinaryHprofReader(inputStream);
- reader.setStrict(false);
- reader.read();
- hprofData = reader.getHprofData();
- } catch (IOException e) {
- System.out.println("Problem reading binary hprof data from "
- + file + ": " + e.getMessage());
- return false;
- } finally {
- IoUtils.closeQuietly(inputStream);
- }
- try {
- HprofWriter writer = new AsciiHprofWriter(hprofData, System.out);
- writer.write();
- } catch (IOException e) {
- System.out.println("Problem writing ASCII hprof data: " + e.getMessage());
- return false;
- }
- return true;
- }
-
- private static void usage(String error) {
- System.out.print("ERROR: ");
- System.out.println(error);
- System.out.println();
- System.out.println("usage: HprofBinaryToAscii <binary-hprof-file>");
- System.out.println();
- System.out.println("Reads a binary hprof file and print it in ASCII format");
- }
- }
-
- /**
- * Map of stack traces to a mutable sample count.
- */
- private final Map<HprofData.StackTrace, int[]> stackTraces
- = new HashMap<HprofData.StackTrace, int[]>();
-
- /**
- * Data collected by the sampling profiler
- */
- private final HprofData hprofData = new HprofData(stackTraces);
-
- /**
- * Timer that is used for the lifetime of the profiler
- */
- // note that dalvik/vm/Thread.c depends on this name
- private final Timer timer = new Timer("SamplingProfiler", true);
-
- /**
- * A sampler is created every time profiling starts and cleared
- * everytime profiling stops because once a {@code TimerTask} is
- * canceled it cannot be reused.
- */
- private TimerTask sampler;
-
- /**
- * The maximum number of {@code StackTraceElements} to retain in
- * each stack.
- */
- private final int depth;
-
- /**
- * The {@code ThreadSet} that identifies which threads to sample.
- */
- private final ThreadSet threadSet;
-
- /*
- * Real hprof output examples don't start the thread and trace
- * identifiers at one but seem to start at these arbitrary
- * constants. It certainly seems useful to have relatively unique
- * identifers when manual searching hprof output.
- */
- private int nextThreadId = 200001;
- private int nextStackTraceId = 300001;
- private int nextObjectId = 1;
-
- /**
- * The threads currently known to the profiler for detecting
- * thread start and end events.
- */
- private Thread[] currentThreads = new Thread[0];
-
- /**
- * Map of currently active threads to their identifiers. When
- * threads disappear they are removed and only referenced by their
- * identifiers to prevent retaining garbage threads.
- */
- private final Map<Thread, Integer> threadIds = new HashMap<Thread, Integer>();
-
- /**
- * Mutable StackTrace that is used for probing stackTraces Map
- * without allocating a StackTrace. If addStackTrace needs to
- * be thread safe, this would need to be reconsidered.
- */
- private final HprofData.StackTrace mutableStackTrace = new HprofData.StackTrace();
-
- /**
- * Create a sampling profiler that collects stacks with the
- * specified depth from the threads specified by the specified
- * thread collector.
- *
- * @param depth The maximum stack depth to retain for each sample
- * similar to the hprof option of the same name. Any stack deeper
- * than this will be truncated to this depth. A good starting
- * value is 4 although it is not uncommon to need to raise this to
- * get enough context to understand program behavior. While
- * programs with extensive recursion may require a high value for
- * depth, simply passing in a value for Integer.MAX_VALUE is not
- * advised because of the significant memory need to retain such
- * stacks and runtime overhead to compare stacks.
- */
- public SamplingProfiler(int depth, ThreadSet threadSet) {
- this.depth = depth;
- this.threadSet = threadSet;
- hprofData.setFlags(BinaryHprof.ControlSettings.CPU_SAMPLING.bitmask);
- hprofData.setDepth(depth);
- }
-
- /**
- * A ThreadSet specifies the set of threads to sample.
- */
- public static interface ThreadSet {
- /**
- * Returns an array containing the threads to be sampled. The
- * array may be longer than the number of threads to be
- * sampled, in which case the extra elements must be null.
- */
- public Thread[] threads();
- }
-
- /**
- * Returns a ThreadSet for a fixed set of threads that will not
- * vary at runtime. This has less overhead than a dynamically
- * calculated set, such as {@link newThreadGroupTheadSet}, which has
- * to enumerate the threads each time profiler wants to collect
- * samples.
- */
- public static ThreadSet newArrayThreadSet(Thread... threads) {
- return new ArrayThreadSet(threads);
- }
-
- /**
- * An ArrayThreadSet samples a fixed set of threads that does not
- * vary over the life of the profiler.
- */
- private static class ArrayThreadSet implements ThreadSet {
- private final Thread[] threads;
- public ArrayThreadSet(Thread... threads) {
- if (threads == null) {
- throw new NullPointerException("threads == null");
- }
- this.threads = threads;
- }
- public Thread[] threads() {
- return threads;
- }
- }
-
- /**
- * Returns a ThreadSet that is dynamically computed based on the
- * threads found in the specified ThreadGroup and that
- * ThreadGroup's children.
- */
- public static ThreadSet newThreadGroupTheadSet(ThreadGroup threadGroup) {
- return new ThreadGroupThreadSet(threadGroup);
- }
-
- /**
- * An ThreadGroupThreadSet sample the threads from the specified
- * ThreadGroup and the ThreadGroup's children
- */
- private static class ThreadGroupThreadSet implements ThreadSet {
- private final ThreadGroup threadGroup;
- private Thread[] threads;
- private int lastThread;
-
- public ThreadGroupThreadSet(ThreadGroup threadGroup) {
- if (threadGroup == null) {
- throw new NullPointerException("threadGroup == null");
- }
- this.threadGroup = threadGroup;
- resize();
- }
-
- private void resize() {
- int count = threadGroup.activeCount();
- // we can only tell if we had enough room for all active
- // threads if we actually are larger than the the number of
- // active threads. making it larger also leaves us room to
- // tolerate additional threads without resizing.
- threads = new Thread[count*2];
- lastThread = 0;
- }
-
- public Thread[] threads() {
- int threadCount;
- while (true) {
- threadCount = threadGroup.enumerate(threads);
- if (threadCount == threads.length) {
- resize();
- } else {
- break;
- }
- }
- if (threadCount < lastThread) {
- // avoid retaining pointers to threads that have ended
- Arrays.fill(threads, threadCount, lastThread, null);
- }
- lastThread = threadCount;
- return threads;
- }
- }
-
- /**
- * Starts profiler sampling at the specified rate.
- *
- * @param interval The number of milliseconds between samples
- */
- public void start(int interval) {
- if (interval < 1) {
- throw new IllegalArgumentException("interval < 1");
- }
- if (sampler != null) {
- throw new IllegalStateException("profiling already started");
- }
- sampler = new Sampler();
- hprofData.setStartMillis(System.currentTimeMillis());
- timer.scheduleAtFixedRate(sampler, 0, interval);
- }
-
- /**
- * Stops profiler sampling. It can be restarted with {@link
- * #start(int)} to continue sampling.
- */
- public void stop() {
- if (sampler == null) {
- return;
- }
- sampler.cancel();
- sampler = null;
- }
-
- /**
- * Shuts down profiling after which it can not be restarted. It is
- * important to shut down profiling when done to free resources
- * used by the profiler. Shutting down the profiler also stops the
- * profiling if that has not already been done.
- */
- public void shutdown() {
- stop();
- timer.cancel();
- }
-
- /**
- * Returns the hprof data accumulated by the profiler since it was
- * created. The profiler needs to be stopped, but not necessarily
- * shut down, in order to access the data. If the profiler is
- * restarted, there is no thread safe way to access the data.
- */
- public HprofData getHprofData() {
- if (sampler != null) {
- throw new IllegalStateException("cannot access hprof data while sampling");
- }
- return hprofData;
- }
-
- /**
- * The Sampler does the real work of the profiler.
- *
- * At every sample time, it asks the thread set for the set
- * of threads to sample. It maintains a history of thread creation
- * and death events based on changes observed to the threads
- * returned by the {@code ThreadSet}.
- *
- * For each thread to be sampled, a stack is collected and used to
- * update the set of collected samples. Stacks are truncated to a
- * maximum depth. There is no way to tell if a stack has been truncated.
- */
- private class Sampler extends TimerTask {
-
- private Thread timerThread;
-
- public void run() {
- if (timerThread == null) {
- timerThread = Thread.currentThread();
- }
-
- // process thread creation and death first so that we
- // assign thread ids to any new threads before allocating
- // new stacks for them
- Thread[] newThreads = threadSet.threads();
- if (!Arrays.equals(currentThreads, newThreads)) {
- updateThreadHistory(currentThreads, newThreads);
- currentThreads = newThreads.clone();
- }
-
- for (Thread thread : currentThreads) {
- if (thread == null) {
- break;
- }
- if (thread == timerThread) {
- continue;
- }
-
- // TODO replace with a VMStack.getThreadStackTrace
- // variant to avoid allocating unneeded elements
- StackTraceElement[] stackFrames = thread.getStackTrace();
- if (stackFrames.length == 0) {
- continue;
- }
- if (stackFrames.length > depth) {
- stackFrames = Arrays.copyOfRange(stackFrames, 0, depth);
- }
- recordStackTrace(thread, stackFrames);
- }
- }
-
- /**
- * Record a new stack trace. The thread should have been
- * previously registered with addStartThread.
- */
- private void recordStackTrace(Thread thread, StackTraceElement[] stackFrames) {
- Integer threadId = threadIds.get(thread);
- if (threadId == null) {
- throw new IllegalArgumentException("Unknown thread " + thread);
- }
- mutableStackTrace.threadId = threadId;
- mutableStackTrace.stackFrames = stackFrames;
-
- int[] countCell = stackTraces.get(mutableStackTrace);
- if (countCell == null) {
- countCell = new int[1];
- HprofData.StackTrace stackTrace
- = new HprofData.StackTrace(nextStackTraceId++, threadId, stackFrames);
- hprofData.addStackTrace(stackTrace, countCell);
- }
- countCell[0]++;
- }
-
- private void updateThreadHistory(Thread[] oldThreads, Thread[] newThreads) {
- // thread start/stop shouldn't happen too often and
- // these aren't too big, so hopefully this approach
- // won't be too slow...
- Set<Thread> n = new HashSet<Thread>(Arrays.asList(newThreads));
- Set<Thread> o = new HashSet<Thread>(Arrays.asList(oldThreads));
-
- // added = new-old
- Set<Thread> added = new HashSet<Thread>(n);
- added.removeAll(o);
-
- // removed = old-new
- Set<Thread> removed = new HashSet<Thread>(o);
- removed.removeAll(n);
-
- for (Thread thread : added) {
- if (thread == null) {
- continue;
- }
- if (thread == timerThread) {
- continue;
- }
- addStartThread(thread);
- }
- for (Thread thread : removed) {
- if (thread == null) {
- continue;
- }
- if (thread == timerThread) {
- continue;
- }
- addEndThread(thread);
- }
- }
-
- /**
- * Record that a newly noticed thread.
- */
- private void addStartThread(Thread thread) {
- if (thread == null) {
- throw new NullPointerException("thread == null");
- }
- int threadId = nextThreadId++;
- Integer old = threadIds.put(thread, threadId);
- if (old != null) {
- throw new IllegalArgumentException("Thread already registered as " + old);
- }
-
- String threadName = thread.getName();
- // group will become null when thread is terminated
- ThreadGroup group = thread.getThreadGroup();
- String groupName = group == null ? null : group.getName();
- ThreadGroup parentGroup = group == null ? null : group.getParent();
- String parentGroupName = parentGroup == null ? null : parentGroup.getName();
-
- HprofData.ThreadEvent event
- = HprofData.ThreadEvent.start(nextObjectId++, threadId,
- threadName, groupName, parentGroupName);
- hprofData.addThreadEvent(event);
- }
-
- /**
- * Record that a thread has disappeared.
- */
- private void addEndThread(Thread thread) {
- if (thread == null) {
- throw new NullPointerException("thread == null");
- }
- Integer threadId = threadIds.remove(thread);
- if (threadId == null) {
- throw new IllegalArgumentException("Unknown thread " + thread);
- }
- HprofData.ThreadEvent event = HprofData.ThreadEvent.end(threadId);
- hprofData.addThreadEvent(event);
- }
- }
-}
diff --git a/dalvik/src/main/java/dalvik/system/profiler/AsciiHprofWriter.java b/dalvik/src/main/java/dalvik/system/profiler/AsciiHprofWriter.java
new file mode 100644
index 0000000..130d032
--- /dev/null
+++ b/dalvik/src/main/java/dalvik/system/profiler/AsciiHprofWriter.java
@@ -0,0 +1,88 @@
+/*
+ * Copyright (C) 2010 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package dalvik.system.profiler;
+
+import java.io.IOException;
+import java.io.OutputStream;
+import java.io.PrintWriter;
+import java.util.ArrayList;
+import java.util.Collections;
+import java.util.Comparator;
+import java.util.Date;
+import java.util.List;
+
+public final class AsciiHprofWriter implements HprofWriter {
+
+ private final HprofData data;
+ private final PrintWriter out;
+
+ public AsciiHprofWriter(HprofData data, OutputStream outputStream) {
+ this.data = data;
+ this.out = new PrintWriter(outputStream);
+ }
+
+ public void write() throws IOException {
+ for (HprofData.ThreadEvent e : data.getThreadHistory()) {
+ out.println(e);
+ }
+
+ List<HprofData.Sample> samples
+ = new ArrayList<HprofData.Sample>(data.getSamples());
+ Collections.sort(samples, SAMPLE_COMPARATOR);
+ int total = 0;
+ for (HprofData.Sample sample : samples) {
+ HprofData.StackTrace stackTrace = sample.stackTrace;
+ int count = sample.count;
+ total += count;
+ out.printf("TRACE %d: (thread=%d)\n",
+ stackTrace.stackTraceId,
+ stackTrace.threadId);
+ for (StackTraceElement e : stackTrace.stackFrames) {
+ out.printf("\t%s\n", e);
+ }
+ }
+ Date now = new Date(data.getStartMillis());
+ // "CPU SAMPLES BEGIN (total = 826) Wed Jul 21 12:03:46 2010"
+ out.printf("CPU SAMPLES BEGIN (total = %d) %ta %tb %td %tT %tY\n",
+ total, now, now, now, now, now);
+ out.printf("rank self accum count trace method\n");
+ int rank = 0;
+ double accum = 0;
+ for (HprofData.Sample sample : samples) {
+ rank++;
+ HprofData.StackTrace stackTrace = sample.stackTrace;
+ int count = sample.count;
+ double self = (double)count/(double)total;
+ accum += self;
+
+ // " 1 65.62% 65.62% 542 300302 java.lang.Long.parseLong"
+ out.printf("% 4d% 6.2f%%% 6.2f%% % 7d % 5d %s.%s\n",
+ rank, self*100, accum*100, count, stackTrace.stackTraceId,
+ stackTrace.stackFrames[0].getClassName(),
+ stackTrace.stackFrames[0].getMethodName());
+ }
+ out.printf("CPU SAMPLES END\n");
+ out.flush();
+ }
+
+ private static final Comparator<HprofData.Sample> SAMPLE_COMPARATOR
+ = new Comparator<HprofData.Sample>() {
+ public int compare(HprofData.Sample s1, HprofData.Sample s2) {
+ return s2.count - s1.count;
+ }
+ };
+}
diff --git a/dalvik/src/main/java/dalvik/system/profiler/BinaryHprof.java b/dalvik/src/main/java/dalvik/system/profiler/BinaryHprof.java
new file mode 100644
index 0000000..26eb290
--- /dev/null
+++ b/dalvik/src/main/java/dalvik/system/profiler/BinaryHprof.java
@@ -0,0 +1,118 @@
+/*
+ * Copyright (C) 2010 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package dalvik.system.profiler;
+
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * Hprof binary format related constants shared between the
+ * BinaryHprofReader and BinaryHprofWriter.
+ */
+public final class BinaryHprof {
+ /**
+ * Currently code only supports 4 byte id size.
+ */
+ public static final int ID_SIZE = 4;
+
+ public static enum Tag {
+
+ STRING_IN_UTF8(0x01, -ID_SIZE),
+ LOAD_CLASS(0x02, 4 + ID_SIZE + 4 + ID_SIZE),
+ UNLOAD_CLASS(0x03, 4),
+ STACK_FRAME(0x04, ID_SIZE + ID_SIZE + ID_SIZE + ID_SIZE + 4 + 4),
+ STACK_TRACE(0x05, -(4 + 4 + 4)),
+ ALLOC_SITES(0x06, -(2 + 4 + 4 + 4 + 8 + 8 + 4)),
+ HEAP_SUMMARY(0x07, 4 + 4 + 8 + 8),
+ START_THREAD(0x0a, 4 + ID_SIZE + 4 + ID_SIZE + ID_SIZE + ID_SIZE),
+ END_THREAD(0x0b, 4),
+ HEAP_DUMP(0x0c, -0),
+ HEAP_DUMP_SEGMENT(0x1c, -0),
+ HEAP_DUMP_END(0x2c, 0),
+ CPU_SAMPLES(0x0d, -(4 + 4)),
+ CONTROL_SETTINGS(0x0e, 4 + 2);
+
+ public final byte tag;
+
+ /**
+ * Minimum size in bytes.
+ */
+ public final int minimumSize;
+
+ /**
+ * Maximum size in bytes. 0 mean no specific limit.
+ */
+ public final int maximumSize;
+
+ private Tag(int tag, int size) {
+ this.tag = (byte) tag;
+ if (size > 0) {
+ // fixed size, max and min the same
+ this.minimumSize = size;
+ this.maximumSize = size;
+ } else {
+ // only minimum bound
+ this.minimumSize = -size;
+ this.maximumSize = 0;
+ }
+ }
+
+ private static final Map<Byte, Tag> BYTE_TO_TAG
+ = new HashMap<Byte, Tag>();
+
+ static {
+ for (Tag v : Tag.values()) {
+ BYTE_TO_TAG.put(v.tag, v);
+ }
+ }
+
+ public static Tag get(byte tag) {
+ return BYTE_TO_TAG.get(tag);
+ }
+
+ /**
+ * Returns null if the actual size meets expectations, or a
+ * String error message if not.
+ */
+ public String checkSize(int actual) {
+ if (actual < minimumSize) {
+ return "expected a minimial record size of " + minimumSize + " for " + this
+ + " but received " + actual;
+ }
+ if (maximumSize == 0) {
+ return null;
+ }
+ if (actual > maximumSize) {
+ return "expected a maximum record size of " + maximumSize + " for " + this
+ + " but received " + actual;
+ }
+ return null;
+ }
+ }
+
+ public static enum ControlSettings {
+ ALLOC_TRACES(0x01),
+ CPU_SAMPLING(0x02);
+
+ public final int bitmask;
+
+ private ControlSettings(int bitmask) {
+ this.bitmask = bitmask;
+ }
+ }
+
+}
diff --git a/dalvik/src/main/java/dalvik/system/profiler/BinaryHprofReader.java b/dalvik/src/main/java/dalvik/system/profiler/BinaryHprofReader.java
new file mode 100644
index 0000000..65e5178
--- /dev/null
+++ b/dalvik/src/main/java/dalvik/system/profiler/BinaryHprofReader.java
@@ -0,0 +1,489 @@
+/*
+ * Copyright (C) 2010 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package dalvik.system.profiler;
+
+import java.io.BufferedInputStream;
+import java.io.DataInputStream;
+import java.io.EOFException;
+import java.io.IOException;
+import java.io.InputStream;
+import java.util.Date;
+import java.util.HashMap;
+import java.util.Map;
+
+/**
+ * <pre> {@code
+ * BinaryHprofReader reader = new BinaryHprofReader(new BufferedInputStream(inputStream));
+ * reader.setStrict(false); // for RI compatability
+ * reader.read();
+ * inputStream.close();
+ * reader.getVersion();
+ * reader.getHprofData();
+ * }</pre>
+ */
+public final class BinaryHprofReader {
+
+ private static final boolean TRACE = false;
+
+ private final DataInputStream in;
+
+ /**
+ * By default we try to strictly validate rules followed by
+ * our HprofWriter. For example, every end thread is preceded
+ * by a matching start thread.
+ */
+ private boolean strict = true;
+
+ /**
+ * version string from header after read has been performed,
+ * otherwise null. nullness used to detect if callers try to
+ * access data before read is called.
+ */
+ private String version;
+
+ private final Map<HprofData.StackTrace, int[]> stackTraces
+ = new HashMap<HprofData.StackTrace, int[]>();
+
+ private final HprofData hprofData = new HprofData(stackTraces);
+
+ private final Map<Integer, String> idToString = new HashMap<Integer, String>();
+ private final Map<Integer, String> idToClassName = new HashMap<Integer, String>();
+ private final Map<Integer, StackTraceElement> idToStackFrame
+ = new HashMap<Integer, StackTraceElement>();
+ private final Map<Integer, HprofData.StackTrace> idToStackTrace
+ = new HashMap<Integer, HprofData.StackTrace>();
+
+ /**
+ * Creates a BinaryHprofReader around the specified {@code
+ * inputStream}
+ */
+ public BinaryHprofReader(InputStream inputStream) throws IOException {
+ this.in = new DataInputStream(inputStream);
+ }
+
+ public boolean getStrict () {
+ return strict;
+ }
+
+ public void setStrict (boolean strict) {
+ if (version != null) {
+ throw new IllegalStateException("cannot set strict after read()");
+ }
+ this.strict = strict;
+ }
+
+ /**
+ * throws IllegalStateException if read() has not been called.
+ */
+ private void checkRead() {
+ if (version == null) {
+ throw new IllegalStateException("data access before read()");
+ }
+ }
+
+ public String getVersion() {
+ checkRead();
+ return version;
+ }
+
+ public HprofData getHprofData() {
+ checkRead();
+ return hprofData;
+ }
+
+ /**
+ * Read the hprof header and records from the input
+ */
+ public void read() throws IOException {
+ parseHeader();
+ parseRecords();
+ }
+
+ private void parseHeader() throws IOException {
+ if (TRACE) {
+ System.out.println("hprofTag=HEADER");
+ }
+ parseVersion();
+ parseIdSize();
+ parseTime();
+ }
+
+ private void parseVersion() throws IOException {
+ byte[] bytes = new byte[512];
+ for (int i = 0; i < bytes.length; i++) {
+ byte b = in.readByte();
+ if (b == '\0') {
+ String version = new String(bytes, 0, i, "UTF-8");
+ if (TRACE) {
+ System.out.println("\tversion=" + version);
+ }
+ if (!version.startsWith("JAVA PROFILE ")) {
+ throw new MalformedHprofException("Unexpected version: " + version);
+ }
+ this.version = version;
+ return;
+ }
+ bytes[i] = b;
+ }
+ throw new MalformedHprofException("Could not find HPROF version");
+ }
+
+ private void parseIdSize() throws IOException {
+ int idSize = in.readInt();
+ if (TRACE) {
+ System.out.println("\tidSize=" + idSize);
+ }
+ if (idSize != BinaryHprof.ID_SIZE) {
+ throw new MalformedHprofException("Unsupported identifier size: " + idSize);
+ }
+ }
+
+ private void parseTime() throws IOException {
+ long time = in.readLong();
+ if (TRACE) {
+ System.out.println("\ttime=" + Long.toHexString(time) + " " + new Date(time));
+ }
+ hprofData.setStartMillis(time);
+ }
+
+ private void parseRecords() throws IOException {
+ while (parseRecord()) {
+ ;
+ }
+ }
+
+ /**
+ * Read and process the next record. Returns true if a
+ * record was handled, false on EOF.
+ */
+ private boolean parseRecord() throws IOException {
+ int tagOrEOF = in.read();
+ if (tagOrEOF == -1) {
+ return false;
+ }
+ byte tag = (byte) tagOrEOF;
+ int timeDeltaInMicroseconds = in.readInt();
+ int recordLength = in.readInt();
+ BinaryHprof.Tag hprofTag = BinaryHprof.Tag.get(tag);
+ if (TRACE) {
+ System.out.println("hprofTag=" + hprofTag);
+ }
+ if (hprofTag == null) {
+ skipRecord(hprofTag, recordLength);
+ return true;
+ }
+ String error = hprofTag.checkSize(recordLength);
+ if (error != null) {
+ throw new MalformedHprofException(error);
+ }
+ switch (hprofTag) {
+ case CONTROL_SETTINGS:
+ parseControlSettings();
+ return true;
+
+ case STRING_IN_UTF8:
+ parseStringInUtf8(recordLength);
+ return true;
+
+ case START_THREAD:
+ parseStartThread();
+ return true;
+ case END_THREAD:
+ parseEndThread();
+ return true;
+
+ case LOAD_CLASS:
+ parseLoadClass();
+ return true;
+ case STACK_FRAME:
+ parseStackFrame();
+ return true;
+ case STACK_TRACE:
+ parseStackTrace(recordLength);
+ return true;
+
+ case CPU_SAMPLES:
+ parseCpuSamples(recordLength);
+ return true;
+
+ case UNLOAD_CLASS:
+ case ALLOC_SITES:
+ case HEAP_SUMMARY:
+ case HEAP_DUMP:
+ case HEAP_DUMP_SEGMENT:
+ case HEAP_DUMP_END:
+ default:
+ skipRecord(hprofTag, recordLength);
+ return true;
+ }
+ }
+
+ private void skipRecord(BinaryHprof.Tag hprofTag, long recordLength) throws IOException {
+ if (TRACE) {
+ System.out.println("\tskipping recordLength=" + recordLength);
+ }
+ long skipped = in.skip(recordLength);
+ if (skipped != recordLength) {
+ throw new EOFException("Expected to skip " + recordLength
+ + " bytes but only skipped " + skipped + " bytes");
+ }
+ }
+
+ private void parseControlSettings() throws IOException {
+ int flags = in.readInt();
+ short depth = in.readShort();
+ if (TRACE) {
+ System.out.println("\tflags=" + Integer.toHexString(flags));
+ System.out.println("\tdepth=" + depth);
+ }
+ hprofData.setFlags(flags);
+ hprofData.setDepth(depth);
+ }
+
+ private void parseStringInUtf8(int recordLength) throws IOException {
+ int stringId = in.readInt();
+ byte[] bytes = new byte[recordLength - BinaryHprof.ID_SIZE];
+ readFully(in, bytes);
+ String string = new String(bytes, "UTF-8");
+ if (TRACE) {
+ System.out.println("\tstring=" + string);
+ }
+ String old = idToString.put(stringId, string);
+ if (old != null) {
+ throw new MalformedHprofException("Duplicate string id: " + stringId);
+ }
+ }
+
+ private static void readFully(InputStream in, byte[] dst) throws IOException {
+ int offset = 0;
+ int byteCount = dst.length;
+ while (byteCount > 0) {
+ int bytesRead = in.read(dst, offset, byteCount);
+ if (bytesRead < 0) {
+ throw new EOFException();
+ }
+ offset += bytesRead;
+ byteCount -= bytesRead;
+ }
+ }
+
+ private void parseLoadClass() throws IOException {
+ int classId = in.readInt();
+ int classObjectId = readId();
+ // serial number apparently not a stack trace id. (int vs ID)
+ // we don't use this field.
+ int stackTraceSerialNumber = in.readInt();
+ String className = readString();
+ if (TRACE) {
+ System.out.println("\tclassId=" + classId);
+ System.out.println("\tclassObjectId=" + classObjectId);
+ System.out.println("\tstackTraceSerialNumber=" + stackTraceSerialNumber);
+ System.out.println("\tclassName=" + className);
+ }
+ String old = idToClassName.put(classId, className);
+ if (old != null) {
+ throw new MalformedHprofException("Duplicate class id: " + classId);
+ }
+ }
+
+ private int readId() throws IOException {
+ return in.readInt();
+ }
+
+ private String readString() throws IOException {
+ int id = readId();
+ if (id == 0) {
+ return null;
+ }
+ String string = idToString.get(id);
+ if (string == null) {
+ throw new MalformedHprofException("Unknown string id " + id);
+ }
+ return string;
+ }
+
+ private String readClass() throws IOException {
+ int id = readId();
+ String string = idToClassName.get(id);
+ if (string == null) {
+ throw new MalformedHprofException("Unknown class id " + id);
+ }
+ return string;
+ }
+
+ private void parseStartThread() throws IOException {
+ int threadId = in.readInt();
+ int objectId = readId();
+ // stack trace where thread was created.
+ // serial number apparently not a stack trace id. (int vs ID)
+ // we don't use this field.
+ int stackTraceSerialNumber = in.readInt();
+ String threadName = readString();
+ String groupName = readString();
+ String parentGroupName = readString();
+ if (TRACE) {
+ System.out.println("\tthreadId=" + threadId);
+ System.out.println("\tobjectId=" + objectId);
+ System.out.println("\tstackTraceSerialNumber=" + stackTraceSerialNumber);
+ System.out.println("\tthreadName=" + threadName);
+ System.out.println("\tgroupName=" + groupName);
+ System.out.println("\tparentGroupName=" + parentGroupName);
+ }
+ HprofData.ThreadEvent event
+ = HprofData.ThreadEvent.start(objectId, threadId,
+ threadName, groupName, parentGroupName);
+ hprofData.addThreadEvent(event);
+ }
+
+ private void parseEndThread() throws IOException {
+ int threadId = in.readInt();
+ if (TRACE) {
+ System.out.println("\tthreadId=" + threadId);
+ }
+ HprofData.ThreadEvent event = HprofData.ThreadEvent.end(threadId);
+ hprofData.addThreadEvent(event);
+ }
+
+ private void parseStackFrame() throws IOException {
+ int stackFrameId = readId();
+ String methodName = readString();
+ String methodSignature = readString();
+ String file = readString();
+ String className = readClass();
+ int line = in.readInt();
+ if (TRACE) {
+ System.out.println("\tstackFrameId=" + stackFrameId);
+ System.out.println("\tclassName=" + className);
+ System.out.println("\tmethodName=" + methodName);
+ System.out.println("\tmethodSignature=" + methodSignature);
+ System.out.println("\tfile=" + file);
+ System.out.println("\tline=" + line);
+ }
+ StackTraceElement stackFrame = new StackTraceElement(className, methodName, file, line);
+ StackTraceElement old = idToStackFrame.put(stackFrameId, stackFrame);
+ if (old != null) {
+ throw new MalformedHprofException("Duplicate stack frame id: " + stackFrameId);
+ }
+ }
+
+ private void parseStackTrace(int recordLength) throws IOException {
+ int stackTraceId = in.readInt();
+ int threadId = in.readInt();
+ int frames = in.readInt();
+ if (TRACE) {
+ System.out.println("\tstackTraceId=" + stackTraceId);
+ System.out.println("\tthreadId=" + threadId);
+ System.out.println("\tframes=" + frames);
+ }
+ int expectedLength = 4 + 4 + 4 + (frames * BinaryHprof.ID_SIZE);
+ if (recordLength != expectedLength) {
+ throw new MalformedHprofException("Expected stack trace record of size "
+ + expectedLength
+ + " based on number of frames but header "
+ + "specified a length of " + recordLength);
+ }
+ StackTraceElement[] stackFrames = new StackTraceElement[frames];
+ for (int i = 0; i < frames; i++) {
+ int stackFrameId = readId();
+ StackTraceElement stackFrame = idToStackFrame.get(stackFrameId);
+ if (TRACE) {
+ System.out.println("\tstackFrameId=" + stackFrameId);
+ System.out.println("\tstackFrame=" + stackFrame);
+ }
+ if (stackFrame == null) {
+ throw new MalformedHprofException("Unknown stack frame id " + stackFrameId);
+ }
+ stackFrames[i] = stackFrame;
+ }
+
+ HprofData.StackTrace stackTrace
+ = new HprofData.StackTrace(stackTraceId, threadId, stackFrames);
+ if (strict) {
+ hprofData.addStackTrace(stackTrace, new int[1]);
+ } else {
+ // The RI can have duplicate stacks, presumably they
+ // have a minor race if two samples with the same
+ // stack are taken around the same time. if we have a
+ // duplicate, just skip adding it to hprofData, but
+ // register it locally in idToStackFrame. if it seen
+ // in CPU_SAMPLES, we will find a StackTrace is equal
+ // to the first, so they will share a countCell.
+ int[] countCell = stackTraces.get(stackTrace);
+ if (countCell == null) {
+ hprofData.addStackTrace(stackTrace, new int[1]);
+ }
+ }
+
+ HprofData.StackTrace old = idToStackTrace.put(stackTraceId, stackTrace);
+ if (old != null) {
+ throw new MalformedHprofException("Duplicate stack trace id: " + stackTraceId);
+ }
+
+ }
+
+ private void parseCpuSamples(int recordLength) throws IOException {
+ int totalSamples = in.readInt();
+ int samplesCount = in.readInt();
+ if (TRACE) {
+ System.out.println("\ttotalSamples=" + totalSamples);
+ System.out.println("\tsamplesCount=" + samplesCount);
+ }
+ int expectedLength = 4 + 4 + (samplesCount * (4 + 4));
+ if (recordLength != expectedLength) {
+ throw new MalformedHprofException("Expected CPU samples record of size "
+ + expectedLength
+ + " based on number of samples but header "
+ + "specified a length of " + recordLength);
+ }
+ int total = 0;
+ for (int i = 0; i < samplesCount; i++) {
+ int count = in.readInt();
+ int stackTraceId = in.readInt();
+ if (TRACE) {
+ System.out.println("\tcount=" + count);
+ System.out.println("\tstackTraceId=" + stackTraceId);
+ }
+ HprofData.StackTrace stackTrace = idToStackTrace.get(stackTraceId);
+ if (stackTrace == null) {
+ throw new MalformedHprofException("Unknown stack trace id " + stackTraceId);
+ }
+ if (count == 0) {
+ throw new MalformedHprofException("Zero sample count for stack trace "
+ + stackTrace);
+ }
+ int[] countCell = stackTraces.get(stackTrace);
+ if (strict) {
+ if (countCell[0] != 0) {
+ throw new MalformedHprofException("Setting sample count of stack trace "
+ + stackTrace + " to " + count
+ + " found it was already initialized to "
+ + countCell[0]);
+ }
+ } else {
+ // Coalesce counts from duplicate stack traces.
+ // For more on this, see comments in parseStackTrace.
+ count += countCell[0];
+ }
+ countCell[0] = count;
+ total += count;
+ }
+ if (strict && totalSamples != total) {
+ throw new MalformedHprofException("Expected a total of " + totalSamples
+ + " samples but saw " + total);
+ }
+ }
+}
diff --git a/dalvik/src/main/java/dalvik/system/profiler/BinaryHprofWriter.java b/dalvik/src/main/java/dalvik/system/profiler/BinaryHprofWriter.java
new file mode 100644
index 0000000..808572d
--- /dev/null
+++ b/dalvik/src/main/java/dalvik/system/profiler/BinaryHprofWriter.java
@@ -0,0 +1,245 @@
+/*
+ * Copyright (C) 2010 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package dalvik.system.profiler;
+
+import java.io.DataOutputStream;
+import java.io.IOException;
+import java.io.OutputStream;
+import java.util.HashMap;
+import java.util.Map;
+import java.util.Set;
+
+public final class BinaryHprofWriter implements HprofWriter {
+
+ private int nextStringId = 1; // id 0 => null
+ private int nextClassId = 1;
+ private int nextStackFrameId = 1;
+ private final Map<String, Integer> stringToId = new HashMap<String, Integer>();
+ private final Map<String, Integer> classNameToId = new HashMap<String, Integer>();
+ private final Map<StackTraceElement, Integer> stackFrameToId
+ = new HashMap<StackTraceElement, Integer>();
+
+ private final HprofData data;
+ private final DataOutputStream out;
+
+ public BinaryHprofWriter(HprofData data, OutputStream outputStream) {
+ this.data = data;
+ this.out = new DataOutputStream(outputStream);
+ }
+
+ public void write() throws IOException {
+ try {
+ writeHeader(data.getStartMillis());
+
+ writeControlSettings(data.getFlags(), data.getDepth());
+
+ for (HprofData.ThreadEvent event : data.getThreadHistory()) {
+ writeThreadEvent(event);
+ }
+
+ Set<HprofData.Sample> samples = data.getSamples();
+ int total = 0;
+ for (HprofData.Sample sample : samples) {
+ total += sample.count;
+ writeStackTrace(sample.stackTrace);
+ }
+ writeCpuSamples(total, samples);
+
+ } finally {
+ out.flush();
+ }
+ }
+
+ private void writeHeader(long dumpTimeInMilliseconds) throws IOException {
+ out.writeBytes("JAVA PROFILE 1.0.2");
+ out.writeByte(0); // null terminated string
+ out.writeInt(BinaryHprof.ID_SIZE);
+ out.writeLong(dumpTimeInMilliseconds);
+ }
+
+ private void writeControlSettings(int flags, int depth) throws IOException {
+ if (depth > Short.MAX_VALUE) {
+ throw new IllegalArgumentException("depth too large for binary hprof: "
+ + depth + " > " + Short.MAX_VALUE);
+ }
+ writeRecordHeader(BinaryHprof.Tag.CONTROL_SETTINGS,
+ 0,
+ BinaryHprof.Tag.CONTROL_SETTINGS.maximumSize);
+ out.writeInt(flags);
+ out.writeShort((short) depth);
+ }
+
+ private void writeThreadEvent(HprofData.ThreadEvent e) throws IOException {
+ switch (e.type) {
+ case START:
+ writeStartThread(e);
+ return;
+ case END:
+ writeStopThread(e);
+ return;
+ }
+ throw new IllegalStateException(e.type.toString());
+ }
+
+ private void writeStartThread(HprofData.ThreadEvent e) throws IOException {
+ int threadNameId = writeString(e.threadName);
+ int groupNameId = writeString(e.groupName);
+ int parentGroupNameId = writeString(e.parentGroupName);
+ writeRecordHeader(BinaryHprof.Tag.START_THREAD,
+ 0,
+ BinaryHprof.Tag.START_THREAD.maximumSize);
+ out.writeInt(e.threadId);
+ writeId(e.objectId);
+ out.writeInt(0); // stack trace where thread was started unavailable
+ writeId(threadNameId);
+ writeId(groupNameId);
+ writeId(parentGroupNameId);
+ }
+
+ private void writeStopThread(HprofData.ThreadEvent e) throws IOException {
+ writeRecordHeader(BinaryHprof.Tag.END_THREAD,
+ 0,
+ BinaryHprof.Tag.END_THREAD.maximumSize);
+ out.writeInt(e.threadId);
+ }
+
+ private void writeRecordHeader(BinaryHprof.Tag hprofTag,
+ int timeDeltaInMicroseconds,
+ int recordLength) throws IOException {
+ String error = hprofTag.checkSize(recordLength);
+ if (error != null) {
+ throw new AssertionError(error);
+ }
+ out.writeByte(hprofTag.tag);
+ out.writeInt(timeDeltaInMicroseconds);
+ out.writeInt(recordLength);
+ }
+
+ private void writeId(int id) throws IOException {
+ out.writeInt(id);
+ }
+
+ /**
+ * Ensures that a string has been writen to the out and
+ * returns its ID. The ID of a null string is zero, and
+ * doesn't actually result in any output. In a string has
+ * already been written previously, the earlier ID will be
+ * returned and no output will be written.
+ */
+ private int writeString(String string) throws IOException {
+ if (string == null) {
+ return 0;
+ }
+ Integer identifier = stringToId.get(string);
+ if (identifier != null) {
+ return identifier;
+ }
+
+ int id = nextStringId++;
+ stringToId.put(string, id);
+
+ byte[] bytes = string.getBytes("UTF-8");
+ writeRecordHeader(BinaryHprof.Tag.STRING_IN_UTF8,
+ 0,
+ BinaryHprof.ID_SIZE + bytes.length);
+ out.writeInt(id);
+ out.write(bytes, 0, bytes.length);
+
+ return id;
+ }
+
+ private void writeCpuSamples(int totalSamples, Set<HprofData.Sample> samples)
+ throws IOException {
+ int samplesCount = samples.size();
+ if (samplesCount == 0) {
+ return;
+ }
+ writeRecordHeader(BinaryHprof.Tag.CPU_SAMPLES, 0, 4 + 4 + (samplesCount * (4 + 4)));
+ out.writeInt(totalSamples);
+ out.writeInt(samplesCount);
+ for (HprofData.Sample sample : samples) {
+ out.writeInt(sample.count);
+ out.writeInt(sample.stackTrace.stackTraceId);
+ }
+ }
+
+ private void writeStackTrace(HprofData.StackTrace stackTrace) throws IOException {
+ int frames = stackTrace.stackFrames.length;
+ int[] stackFrameIds = new int[frames];
+ for (int i = 0; i < frames; i++) {
+ stackFrameIds[i] = writeStackFrame(stackTrace.stackFrames[i]);
+ }
+ writeRecordHeader(BinaryHprof.Tag.STACK_TRACE,
+ 0,
+ 4 + 4 + 4 + (frames * BinaryHprof.ID_SIZE));
+ out.writeInt(stackTrace.stackTraceId);
+ out.writeInt(stackTrace.threadId);
+ out.writeInt(frames);
+ for (int stackFrameId : stackFrameIds) {
+ writeId(stackFrameId);
+ }
+ }
+
+ private int writeLoadClass(String className) throws IOException {
+ Integer identifier = classNameToId.get(className);
+ if (identifier != null) {
+ return identifier;
+ }
+ int id = nextClassId++;
+ classNameToId.put(className, id);
+
+ int classNameId = writeString(className);
+ writeRecordHeader(BinaryHprof.Tag.LOAD_CLASS,
+ 0,
+ BinaryHprof.Tag.LOAD_CLASS.maximumSize);
+ out.writeInt(id);
+ writeId(0); // class object ID
+ out.writeInt(0); // stack trace where class was loaded is unavailable
+ writeId(classNameId);
+
+ return id;
+ }
+
+ private int writeStackFrame(StackTraceElement stackFrame) throws IOException {
+ Integer identifier = stackFrameToId.get(stackFrame);
+ if (identifier != null) {
+ return identifier;
+ }
+
+ int id = nextStackFrameId++;
+ stackFrameToId.put(stackFrame, id);
+
+ int classId = writeLoadClass(stackFrame.getClassName());
+ int methodNameId = writeString(stackFrame.getMethodName());
+ int sourceId = writeString(stackFrame.getFileName());
+ writeRecordHeader(BinaryHprof.Tag.STACK_FRAME,
+ 0,
+ BinaryHprof.Tag.STACK_FRAME.maximumSize);
+ writeId(id);
+ writeId(methodNameId);
+ writeId(0); // method signature is unavailable from StackTraceElement
+ writeId(sourceId);
+ out.writeInt(classId);
+ out.writeInt(stackFrame.getLineNumber());
+
+ return id;
+ }
+
+ public void close() throws IOException {
+ out.close();
+ }
+}
diff --git a/dalvik/src/main/java/dalvik/system/profiler/HprofBinaryToAscii.java b/dalvik/src/main/java/dalvik/system/profiler/HprofBinaryToAscii.java
new file mode 100644
index 0000000..2fed9d8
--- /dev/null
+++ b/dalvik/src/main/java/dalvik/system/profiler/HprofBinaryToAscii.java
@@ -0,0 +1,88 @@
+/*
+ * Copyright (C) 2010 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package dalvik.system.profiler;
+
+import java.io.BufferedInputStream;
+import java.io.File;
+import java.io.FileInputStream;
+import java.io.IOException;
+import java.io.InputStream;
+
+/**
+ * Run on device with:
+ * adb shell dalvikvm 'dalvik.system.SamplingProfiler\$HprofBinaryToAscii'
+ *
+ * Run on host with:
+ * java -classpath out/target/common/obj/JAVA_LIBRARIES/core_intermediates/classes.jar
+ */
+public final class HprofBinaryToAscii {
+
+ public static void main(String[] args) {
+ System.exit(convert(args) ? 0 : 1);
+ }
+
+ public static boolean convert(String[] args) {
+
+ if (args.length != 1) {
+ usage("binary hprof file argument expected");
+ return false;
+ }
+ File file = new File(args[0]);
+ if (!file.exists()) {
+ usage("file " + file + " does not exist");
+ return false;
+ }
+
+ HprofData hprofData;
+ InputStream inputStream = null;
+ try {
+ inputStream = new BufferedInputStream(new FileInputStream(file));
+ BinaryHprofReader reader = new BinaryHprofReader(inputStream);
+ reader.setStrict(false);
+ reader.read();
+ hprofData = reader.getHprofData();
+ } catch (IOException e) {
+ System.out.println("Problem reading binary hprof data from "
+ + file + ": " + e.getMessage());
+ return false;
+ } finally {
+ if (inputStream != null) {
+ try {
+ inputStream.close();
+ } catch (IOException ignored) {
+ }
+ }
+ }
+ try {
+ HprofWriter writer = new AsciiHprofWriter(hprofData, System.out);
+ writer.write();
+ } catch (IOException e) {
+ System.out.println("Problem writing ASCII hprof data: " + e.getMessage());
+ return false;
+ }
+ return true;
+ }
+
+ private static void usage(String error) {
+ System.out.print("ERROR: ");
+ System.out.println(error);
+ System.out.println();
+ System.out.println("usage: HprofBinaryToAscii <binary-hprof-file>");
+ System.out.println();
+ System.out.println("Reads a binary hprof file and print it in ASCII format");
+ }
+}
diff --git a/dalvik/src/main/java/dalvik/system/profiler/HprofData.java b/dalvik/src/main/java/dalvik/system/profiler/HprofData.java
new file mode 100644
index 0000000..7b44bb9
--- /dev/null
+++ b/dalvik/src/main/java/dalvik/system/profiler/HprofData.java
@@ -0,0 +1,394 @@
+/*
+ * Copyright (C) 2010 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package dalvik.system.profiler;
+
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.HashSet;
+import java.util.List;
+import java.util.Map.Entry;
+import java.util.Map;
+import java.util.Set;
+
+/**
+ * Represents sampling profiler data. Can be converted to ASCII or
+ * binary hprof-style output using {@link AsciiHprofWriter} or
+ * {@link BinaryHprofWriter}.
+ * <p>
+ * The data includes:
+ * <ul>
+ * <li>the start time of the last sampling period
+ * <li>the history of thread start and end events
+ * <li>stack traces with frequency counts
+ * <ul>
+ */
+public final class HprofData {
+
+ public static enum ThreadEventType { START, END };
+
+ /**
+ * ThreadEvent represents thread creation and death events for
+ * reporting. It provides a record of the thread and thread group
+ * names for tying samples back to their source thread.
+ */
+ public static final class ThreadEvent {
+
+ public final ThreadEventType type;
+ public final int objectId;
+ public final int threadId;
+ public final String threadName;
+ public final String groupName;
+ public final String parentGroupName;
+
+ public static ThreadEvent start(int objectId, int threadId, String threadName,
+ String groupName, String parentGroupName) {
+ return new ThreadEvent(ThreadEventType.START, objectId, threadId,
+ threadName, groupName, parentGroupName);
+ }
+
+ public static ThreadEvent end(int threadId) {
+ return new ThreadEvent(ThreadEventType.END, threadId);
+ }
+
+ private ThreadEvent(ThreadEventType type, int objectId, int threadId,
+ String threadName, String groupName, String parentGroupName) {
+ if (threadName == null) {
+ throw new NullPointerException("threadName == null");
+ }
+ this.type = ThreadEventType.START;
+ this.objectId = objectId;
+ this.threadId = threadId;
+ this.threadName = threadName;
+ this.groupName = groupName;
+ this.parentGroupName = parentGroupName;
+ }
+
+ private ThreadEvent(ThreadEventType type, int threadId) {
+ this.type = ThreadEventType.END;
+ this.objectId = -1;
+ this.threadId = threadId;
+ this.threadName = null;
+ this.groupName = null;
+ this.parentGroupName = null;
+ }
+
+ @Override public int hashCode() {
+ int result = 17;
+ result = 31 * result + objectId;
+ result = 31 * result + threadId;
+ result = 31 * result + hashCode(threadName);
+ result = 31 * result + hashCode(groupName);
+ result = 31 * result + hashCode(parentGroupName);
+ return result;
+ }
+
+ private static int hashCode(Object o) {
+ return (o == null) ? 0 : o.hashCode();
+ }
+
+ @Override public boolean equals(Object o) {
+ if (!(o instanceof ThreadEvent)) {
+ return false;
+ }
+ ThreadEvent event = (ThreadEvent) o;
+ return (this.type == event.type
+ && this.objectId == event.objectId
+ && this.threadId == event.threadId
+ && equal(this.threadName, event.threadName)
+ && equal(this.groupName, event.groupName)
+ && equal(this.parentGroupName, event.parentGroupName));
+ }
+
+ private static boolean equal(Object a, Object b) {
+ return a == b || (a != null && a.equals(b));
+ }
+
+ @Override public String toString() {
+ switch (type) {
+ case START:
+ return String.format(
+ "THREAD START (obj=%d, id = %d, name=\"%s\", group=\"%s\")",
+ objectId, threadId, threadName, groupName);
+ case END:
+ return String.format("THREAD END (id = %d)", threadId);
+ }
+ throw new IllegalStateException(type.toString());
+ }
+ }
+
+ /**
+ * A unique stack trace for a specific thread.
+ */
+ public static final class StackTrace {
+
+ public final int stackTraceId;
+ int threadId;
+ StackTraceElement[] stackFrames;
+
+ StackTrace() {
+ this.stackTraceId = -1;
+ }
+
+ public StackTrace(int stackTraceId, int threadId, StackTraceElement[] stackFrames) {
+ if (stackFrames == null) {
+ throw new NullPointerException("stackFrames == null");
+ }
+ this.stackTraceId = stackTraceId;
+ this.threadId = threadId;
+ this.stackFrames = stackFrames;
+ }
+
+ public int getThreadId() {
+ return threadId;
+ }
+
+ public StackTraceElement[] getStackFrames() {
+ return stackFrames;
+ }
+
+ @Override public int hashCode() {
+ int result = 17;
+ result = 31 * result + threadId;
+ result = 31 * result + Arrays.hashCode(stackFrames);
+ return result;
+ }
+
+ @Override public boolean equals(Object o) {
+ if (!(o instanceof StackTrace)) {
+ return false;
+ }
+ StackTrace s = (StackTrace) o;
+ return threadId == s.threadId && Arrays.equals(stackFrames, s.stackFrames);
+ }
+
+ @Override public String toString() {
+ StringBuilder frames = new StringBuilder();
+ if (stackFrames.length > 0) {
+ frames.append('\n');
+ for (StackTraceElement stackFrame : stackFrames) {
+ frames.append("\t at ");
+ frames.append(stackFrame);
+ frames.append('\n');
+ }
+ } else {
+ frames.append("<empty>");
+ }
+ return "StackTrace[stackTraceId=" + stackTraceId
+ + ", threadId=" + threadId
+ + ", frames=" + frames + "]";
+
+ }
+ }
+
+ /**
+ * A read only container combining a stack trace with its frequency.
+ */
+ public static final class Sample {
+
+ public final StackTrace stackTrace;
+ public final int count;
+
+ private Sample(StackTrace stackTrace, int count) {
+ if (stackTrace == null) {
+ throw new NullPointerException("stackTrace == null");
+ }
+ if (count < 0) {
+ throw new IllegalArgumentException("count < 0:" + count);
+ }
+ this.stackTrace = stackTrace;
+ this.count = count;
+ }
+
+ @Override public int hashCode() {
+ int result = 17;
+ result = 31 * result + stackTrace.hashCode();
+ result = 31 * result + count;
+ return result;
+ }
+
+ @Override public boolean equals(Object o) {
+ if (!(o instanceof Sample)) {
+ return false;
+ }
+ Sample s = (Sample) o;
+ return count == s.count && stackTrace.equals(s.stackTrace);
+ }
+
+ @Override public String toString() {
+ return "Sample[count=" + count + " " + stackTrace + "]";
+ }
+
+ }
+
+ /**
+ * Start of last sampling period.
+ */
+ private long startMillis;
+
+ /**
+ * CONTROL_SETTINGS flags
+ */
+ private int flags;
+
+ /**
+ * stack sampling depth
+ */
+ private int depth;
+
+ /**
+ * List of thread creation and death events.
+ */
+ private final List<ThreadEvent> threadHistory = new ArrayList<ThreadEvent>();
+
+ /**
+ * Map of thread id to a start ThreadEvent
+ */
+ private final Map<Integer, ThreadEvent> threadIdToThreadEvent
+ = new HashMap<Integer, ThreadEvent>();
+
+ /**
+ * Map of stack traces to a mutable sample count. The map is
+ * provided by the creator of the HprofData so only have
+ * mutable access to the int[] cells that contain the sample
+ * count. Only an unmodifiable iterator view is available to
+ * users of the HprofData.
+ */
+ private final Map<HprofData.StackTrace, int[]> stackTraces;
+
+ public HprofData(Map<StackTrace, int[]> stackTraces) {
+ if (stackTraces == null) {
+ throw new NullPointerException("stackTraces == null");
+ }
+ this.stackTraces = stackTraces;
+ }
+
+ /**
+ * The start time in milliseconds of the last profiling period.
+ */
+ public long getStartMillis() {
+ return startMillis;
+ }
+
+ /**
+ * Set the time for the start of the current sampling period.
+ */
+ public void setStartMillis(long startMillis) {
+ this.startMillis = startMillis;
+ }
+
+ /**
+ * Get the {@link BinaryHprof.ControlSettings} flags
+ */
+ public int getFlags() {
+ return flags;
+ }
+
+ /**
+ * Set the {@link BinaryHprof.ControlSettings} flags
+ */
+ public void setFlags(int flags) {
+ this.flags = flags;
+ }
+
+ /**
+ * Get the stack sampling depth
+ */
+ public int getDepth() {
+ return depth;
+ }
+
+ /**
+ * Set the stack sampling depth
+ */
+ public void setDepth(int depth) {
+ this.depth = depth;
+ }
+
+ /**
+ * Return an unmodifiable history of start and end thread events.
+ */
+ public List<ThreadEvent> getThreadHistory() {
+ return Collections.unmodifiableList(threadHistory);
+ }
+
+ /**
+ * Return a new set containing the current sample data.
+ */
+ public Set<Sample> getSamples() {
+ Set<Sample> samples = new HashSet<Sample>(stackTraces.size());
+ for (Entry<StackTrace, int[]> e : stackTraces.entrySet()) {
+ StackTrace stackTrace = e.getKey();
+ int countCell[] = e.getValue();
+ int count = countCell[0];
+ Sample sample = new Sample(stackTrace, count);
+ samples.add(sample);
+ }
+ return samples;
+ }
+
+ /**
+ * Record an event in the thread history.
+ */
+ public void addThreadEvent(ThreadEvent event) {
+ if (event == null) {
+ throw new NullPointerException("event == null");
+ }
+ ThreadEvent old = threadIdToThreadEvent.put(event.threadId, event);
+ switch (event.type) {
+ case START:
+ if (old != null) {
+ throw new IllegalArgumentException("ThreadEvent already registered for id "
+ + event.threadId);
+ }
+ break;
+ case END:
+ // Do not assert that the END_THREAD matches a
+ // START_THREAD unless in strict mode. While thhis
+ // hold true in the binary hprof BinaryHprofWriter
+ // produces, it is not true of hprof files created
+ // by the RI. However, if there is an event
+ // already registed for a thread id, it should be
+ // the matching start, not a duplicate end.
+ if (old != null && old.type == ThreadEventType.END) {
+ throw new IllegalArgumentException("Duplicate ThreadEvent.end for id "
+ + event.threadId);
+ }
+ break;
+ }
+ threadHistory.add(event);
+ }
+
+ /**
+ * Record an stack trace and an associated int[] cell of
+ * sample cound for the stack trace. The caller is allowed
+ * retain a pointer to the cell to update the count. The
+ * SamplingProfiler intentionally does not present a mutable
+ * view of the count.
+ */
+ public void addStackTrace(StackTrace stackTrace, int[] countCell) {
+ if (!threadIdToThreadEvent.containsKey(stackTrace.threadId)) {
+ throw new IllegalArgumentException("Unknown thread id " + stackTrace.threadId);
+ }
+ int[] old = stackTraces.put(stackTrace, countCell);
+ if (old != null) {
+ throw new IllegalArgumentException("StackTrace already registered for id "
+ + stackTrace.stackTraceId + ":\n" + stackTrace);
+ }
+ }
+}
diff --git a/dalvik/src/main/java/dalvik/system/profiler/HprofWriter.java b/dalvik/src/main/java/dalvik/system/profiler/HprofWriter.java
new file mode 100644
index 0000000..b1d5a9b
--- /dev/null
+++ b/dalvik/src/main/java/dalvik/system/profiler/HprofWriter.java
@@ -0,0 +1,23 @@
+/*
+ * Copyright (C) 2010 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package dalvik.system.profiler;
+
+import java.io.IOException;
+
+public interface HprofWriter {
+ public void write() throws IOException;
+}
diff --git a/dalvik/src/main/java/dalvik/system/profiler/MalformedHprofException.java b/dalvik/src/main/java/dalvik/system/profiler/MalformedHprofException.java
new file mode 100644
index 0000000..b4097b2
--- /dev/null
+++ b/dalvik/src/main/java/dalvik/system/profiler/MalformedHprofException.java
@@ -0,0 +1,35 @@
+/*
+ * Copyright (C) 2010 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package dalvik.system.profiler;
+
+import java.io.IOException;
+
+public final class MalformedHprofException extends IOException {
+
+ private static final long serialVersionUID = 8558990237047894213L;
+
+ MalformedHprofException(String message) {
+ super(message);
+ }
+ MalformedHprofException(String message, Throwable cause) {
+ super(message, cause);
+ }
+ MalformedHprofException(Throwable cause) {
+ super(cause);
+ }
+}
+
diff --git a/dalvik/src/main/java/dalvik/system/profiler/SamplingProfiler.java b/dalvik/src/main/java/dalvik/system/profiler/SamplingProfiler.java
new file mode 100644
index 0000000..13421ec
--- /dev/null
+++ b/dalvik/src/main/java/dalvik/system/profiler/SamplingProfiler.java
@@ -0,0 +1,443 @@
+/*
+ * Copyright (C) 2010 The Android Open Source Project
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package dalvik.system.profiler;
+
+import java.util.Arrays;
+import java.util.HashMap;
+import java.util.HashSet;
+import java.util.Map;
+import java.util.Set;
+import java.util.Timer;
+import java.util.TimerTask;
+
+/**
+ * A sampling profiler. It currently is implemented without any
+ * virtual machine support, relying solely on {@code
+ * Thread.getStackTrace} to collect samples. As such, the overhead is
+ * higher than a native approach and it does not provide insight into
+ * where time is spent within native code, but it can still provide
+ * useful insight into where a program is spending time.
+ *
+ * <h3>Usage Example</h3>
+ *
+ * The following example shows how to use the {@code
+ * SamplingProfiler}. It samples the current thread's stack to a depth
+ * of 12 stack frame elements over two different measurement periods
+ * with samples taken every 100 milliseconds. In then prints the
+ * results in hprof format to the standard output.
+ *
+ * <pre> {@code
+ * ThreadSet threadSet = SamplingProfiler.newArrayThreadSet(Thread.currentThread());
+ * SamplingProfiler profiler = new SamplingProfiler(12, threadSet);
+ * profiler.start(100);
+ * // period of measurement
+ * profiler.stop();
+ * // period of non-measurement
+ * profiler.start(100);
+ * // another period of measurement
+ * profiler.stop();
+ * profiler.shutdown();
+ * HprofWriter writer = new AsciiHprofWriter(profiler.getHprofData(), System.out);
+ * writer.write();
+ * }</pre>
+ */
+public final class SamplingProfiler {
+
+ /**
+ * Map of stack traces to a mutable sample count.
+ */
+ private final Map<HprofData.StackTrace, int[]> stackTraces
+ = new HashMap<HprofData.StackTrace, int[]>();
+
+ /**
+ * Data collected by the sampling profiler
+ */
+ private final HprofData hprofData = new HprofData(stackTraces);
+
+ /**
+ * Timer that is used for the lifetime of the profiler
+ */
+ private final Timer timer = new Timer("SamplingProfiler", true);
+
+ /**
+ * A sampler is created every time profiling starts and cleared
+ * everytime profiling stops because once a {@code TimerTask} is
+ * canceled it cannot be reused.
+ */
+ private TimerTask sampler;
+
+ /**
+ * The maximum number of {@code StackTraceElements} to retain in
+ * each stack.
+ */
+ private final int depth;
+
+ /**
+ * The {@code ThreadSet} that identifies which threads to sample.
+ */
+ private final ThreadSet threadSet;
+
+ /*
+ * Real hprof output examples don't start the thread and trace
+ * identifiers at one but seem to start at these arbitrary
+ * constants. It certainly seems useful to have relatively unique
+ * identifers when manual searching hprof output.
+ */
+ private int nextThreadId = 200001;
+ private int nextStackTraceId = 300001;
+ private int nextObjectId = 1;
+
+ /**
+ * The threads currently known to the profiler for detecting
+ * thread start and end events.
+ */
+ private Thread[] currentThreads = new Thread[0];
+
+ /**
+ * Map of currently active threads to their identifiers. When
+ * threads disappear they are removed and only referenced by their
+ * identifiers to prevent retaining garbage threads.
+ */
+ private final Map<Thread, Integer> threadIds = new HashMap<Thread, Integer>();
+
+ /**
+ * Mutable StackTrace that is used for probing stackTraces Map
+ * without allocating a StackTrace. If addStackTrace needs to
+ * be thread safe, this would need to be reconsidered.
+ */
+ private final HprofData.StackTrace mutableStackTrace = new HprofData.StackTrace();
+
+ /**
+ * Create a sampling profiler that collects stacks with the
+ * specified depth from the threads specified by the specified
+ * thread collector.
+ *
+ * @param depth The maximum stack depth to retain for each sample
+ * similar to the hprof option of the same name. Any stack deeper
+ * than this will be truncated to this depth. A good starting
+ * value is 4 although it is not uncommon to need to raise this to
+ * get enough context to understand program behavior. While
+ * programs with extensive recursion may require a high value for
+ * depth, simply passing in a value for Integer.MAX_VALUE is not
+ * advised because of the significant memory need to retain such
+ * stacks and runtime overhead to compare stacks.
+ */
+ public SamplingProfiler(int depth, ThreadSet threadSet) {
+ this.depth = depth;
+ this.threadSet = threadSet;
+ hprofData.setFlags(BinaryHprof.ControlSettings.CPU_SAMPLING.bitmask);
+ hprofData.setDepth(depth);
+ }
+
+ /**
+ * A ThreadSet specifies the set of threads to sample.
+ */
+ public static interface ThreadSet {
+ /**
+ * Returns an array containing the threads to be sampled. The
+ * array may be longer than the number of threads to be
+ * sampled, in which case the extra elements must be null.
+ */
+ public Thread[] threads();
+ }
+
+ /**
+ * Returns a ThreadSet for a fixed set of threads that will not
+ * vary at runtime. This has less overhead than a dynamically
+ * calculated set, such as {@link #newThreadGroupTheadSet}, which has
+ * to enumerate the threads each time profiler wants to collect
+ * samples.
+ */
+ public static ThreadSet newArrayThreadSet(Thread... threads) {
+ return new ArrayThreadSet(threads);
+ }
+
+ /**
+ * An ArrayThreadSet samples a fixed set of threads that does not
+ * vary over the life of the profiler.
+ */
+ private static class ArrayThreadSet implements ThreadSet {
+ private final Thread[] threads;
+ public ArrayThreadSet(Thread... threads) {
+ if (threads == null) {
+ throw new NullPointerException("threads == null");
+ }
+ this.threads = threads;
+ }
+ public Thread[] threads() {
+ return threads;
+ }
+ }
+
+ /**
+ * Returns a ThreadSet that is dynamically computed based on the
+ * threads found in the specified ThreadGroup and that
+ * ThreadGroup's children.
+ */
+ public static ThreadSet newThreadGroupTheadSet(ThreadGroup threadGroup) {
+ return new ThreadGroupThreadSet(threadGroup);
+ }
+
+ /**
+ * An ThreadGroupThreadSet sample the threads from the specified
+ * ThreadGroup and the ThreadGroup's children
+ */
+ private static class ThreadGroupThreadSet implements ThreadSet {
+ private final ThreadGroup threadGroup;
+ private Thread[] threads;
+ private int lastThread;
+
+ public ThreadGroupThreadSet(ThreadGroup threadGroup) {
+ if (threadGroup == null) {
+ throw new NullPointerException("threadGroup == null");
+ }
+ this.threadGroup = threadGroup;
+ resize();
+ }
+
+ private void resize() {
+ int count = threadGroup.activeCount();
+ // we can only tell if we had enough room for all active
+ // threads if we actually are larger than the the number of
+ // active threads. making it larger also leaves us room to
+ // tolerate additional threads without resizing.
+ threads = new Thread[count*2];
+ lastThread = 0;
+ }
+
+ public Thread[] threads() {
+ int threadCount;
+ while (true) {
+ threadCount = threadGroup.enumerate(threads);
+ if (threadCount == threads.length) {
+ resize();
+ } else {
+ break;
+ }
+ }
+ if (threadCount < lastThread) {
+ // avoid retaining pointers to threads that have ended
+ Arrays.fill(threads, threadCount, lastThread, null);
+ }
+ lastThread = threadCount;
+ return threads;
+ }
+ }
+
+ /**
+ * Starts profiler sampling at the specified rate.
+ *
+ * @param interval The number of milliseconds between samples
+ */
+ public void start(int interval) {
+ if (interval < 1) {
+ throw new IllegalArgumentException("interval < 1");
+ }
+ if (sampler != null) {
+ throw new IllegalStateException("profiling already started");
+ }
+ sampler = new Sampler();
+ hprofData.setStartMillis(System.currentTimeMillis());
+ timer.scheduleAtFixedRate(sampler, 0, interval);
+ }
+
+ /**
+ * Stops profiler sampling. It can be restarted with {@link
+ * #start(int)} to continue sampling.
+ */
+ public void stop() {
+ if (sampler == null) {
+ return;
+ }
+ sampler.cancel();
+ sampler = null;
+ }
+
+ /**
+ * Shuts down profiling after which it can not be restarted. It is
+ * important to shut down profiling when done to free resources
+ * used by the profiler. Shutting down the profiler also stops the
+ * profiling if that has not already been done.
+ */
+ public void shutdown() {
+ stop();
+ timer.cancel();
+ }
+
+ /**
+ * Returns the hprof data accumulated by the profiler since it was
+ * created. The profiler needs to be stopped, but not necessarily
+ * shut down, in order to access the data. If the profiler is
+ * restarted, there is no thread safe way to access the data.
+ */
+ public HprofData getHprofData() {
+ if (sampler != null) {
+ throw new IllegalStateException("cannot access hprof data while sampling");
+ }
+ return hprofData;
+ }
+
+ /**
+ * The Sampler does the real work of the profiler.
+ *
+ * At every sample time, it asks the thread set for the set
+ * of threads to sample. It maintains a history of thread creation
+ * and death events based on changes observed to the threads
+ * returned by the {@code ThreadSet}.
+ *
+ * For each thread to be sampled, a stack is collected and used to
+ * update the set of collected samples. Stacks are truncated to a
+ * maximum depth. There is no way to tell if a stack has been truncated.
+ */
+ private class Sampler extends TimerTask {
+
+ private Thread timerThread;
+
+ public void run() {
+ if (timerThread == null) {
+ timerThread = Thread.currentThread();
+ }
+
+ // process thread creation and death first so that we
+ // assign thread ids to any new threads before allocating
+ // new stacks for them
+ Thread[] newThreads = threadSet.threads();
+ if (!Arrays.equals(currentThreads, newThreads)) {
+ updateThreadHistory(currentThreads, newThreads);
+ currentThreads = newThreads.clone();
+ }
+
+ for (Thread thread : currentThreads) {
+ if (thread == null) {
+ break;
+ }
+ if (thread == timerThread) {
+ continue;
+ }
+
+ // TODO replace with a VMStack.getThreadStackTrace
+ // variant to avoid allocating unneeded elements
+ StackTraceElement[] stackFrames = thread.getStackTrace();
+ if (stackFrames.length == 0) {
+ continue;
+ }
+ if (stackFrames.length > depth) {
+ stackFrames = Arrays.copyOfRange(stackFrames, 0, depth);
+ }
+ recordStackTrace(thread, stackFrames);
+ }
+ }
+
+ /**
+ * Record a new stack trace. The thread should have been
+ * previously registered with addStartThread.
+ */
+ private void recordStackTrace(Thread thread, StackTraceElement[] stackFrames) {
+ Integer threadId = threadIds.get(thread);
+ if (threadId == null) {
+ throw new IllegalArgumentException("Unknown thread " + thread);
+ }
+ mutableStackTrace.threadId = threadId;
+ mutableStackTrace.stackFrames = stackFrames;
+
+ int[] countCell = stackTraces.get(mutableStackTrace);
+ if (countCell == null) {
+ countCell = new int[1];
+ HprofData.StackTrace stackTrace
+ = new HprofData.StackTrace(nextStackTraceId++, threadId, stackFrames);
+ hprofData.addStackTrace(stackTrace, countCell);
+ }
+ countCell[0]++;
+ }
+
+ private void updateThreadHistory(Thread[] oldThreads, Thread[] newThreads) {
+ // thread start/stop shouldn't happen too often and
+ // these aren't too big, so hopefully this approach
+ // won't be too slow...
+ Set<Thread> n = new HashSet<Thread>(Arrays.asList(newThreads));
+ Set<Thread> o = new HashSet<Thread>(Arrays.asList(oldThreads));
+
+ // added = new-old
+ Set<Thread> added = new HashSet<Thread>(n);
+ added.removeAll(o);
+
+ // removed = old-new
+ Set<Thread> removed = new HashSet<Thread>(o);
+ removed.removeAll(n);
+
+ for (Thread thread : added) {
+ if (thread == null) {
+ continue;
+ }
+ if (thread == timerThread) {
+ continue;
+ }
+ addStartThread(thread);
+ }
+ for (Thread thread : removed) {
+ if (thread == null) {
+ continue;
+ }
+ if (thread == timerThread) {
+ continue;
+ }
+ addEndThread(thread);
+ }
+ }
+
+ /**
+ * Record that a newly noticed thread.
+ */
+ private void addStartThread(Thread thread) {
+ if (thread == null) {
+ throw new NullPointerException("thread == null");
+ }
+ int threadId = nextThreadId++;
+ Integer old = threadIds.put(thread, threadId);
+ if (old != null) {
+ throw new IllegalArgumentException("Thread already registered as " + old);
+ }
+
+ String threadName = thread.getName();
+ // group will become null when thread is terminated
+ ThreadGroup group = thread.getThreadGroup();
+ String groupName = group == null ? null : group.getName();
+ ThreadGroup parentGroup = group == null ? null : group.getParent();
+ String parentGroupName = parentGroup == null ? null : parentGroup.getName();
+
+ HprofData.ThreadEvent event
+ = HprofData.ThreadEvent.start(nextObjectId++, threadId,
+ threadName, groupName, parentGroupName);
+ hprofData.addThreadEvent(event);
+ }
+
+ /**
+ * Record that a thread has disappeared.
+ */
+ private void addEndThread(Thread thread) {
+ if (thread == null) {
+ throw new NullPointerException("thread == null");
+ }
+ Integer threadId = threadIds.remove(thread);
+ if (threadId == null) {
+ throw new IllegalArgumentException("Unknown thread " + thread);
+ }
+ HprofData.ThreadEvent event = HprofData.ThreadEvent.end(threadId);
+ hprofData.addThreadEvent(event);
+ }
+ }
+}
diff --git a/dalvik/src/test/java/dalvik/system/SamplingProfilerTest.java b/dalvik/src/test/java/dalvik/system/profiler/SamplingProfilerTest.java
similarity index 94%
rename from dalvik/src/test/java/dalvik/system/SamplingProfilerTest.java
rename to dalvik/src/test/java/dalvik/system/profiler/SamplingProfilerTest.java
index c132d94..6a43aad 100644
--- a/dalvik/src/test/java/dalvik/system/SamplingProfilerTest.java
+++ b/dalvik/src/test/java/dalvik/system/profiler/SamplingProfilerTest.java
@@ -14,17 +14,17 @@
* limitations under the License.
*/
-package dalvik.system;
+package dalvik.system.profiler;
-import dalvik.system.SamplingProfiler.AsciiHprofWriter;
-import dalvik.system.SamplingProfiler.BinaryHprofReader;
-import dalvik.system.SamplingProfiler.BinaryHprofWriter;
-import dalvik.system.SamplingProfiler.HprofData.Sample;
-import dalvik.system.SamplingProfiler.HprofData.StackTrace;
-import dalvik.system.SamplingProfiler.HprofData.ThreadEvent;
-import dalvik.system.SamplingProfiler.HprofData;
-import dalvik.system.SamplingProfiler.HprofWriter;
-import dalvik.system.SamplingProfiler.ThreadSet;
+import dalvik.system.profiler.AsciiHprofWriter;
+import dalvik.system.profiler.BinaryHprofReader;
+import dalvik.system.profiler.BinaryHprofWriter;
+import dalvik.system.profiler.HprofData.Sample;
+import dalvik.system.profiler.HprofData.StackTrace;
+import dalvik.system.profiler.HprofData.ThreadEvent;
+import dalvik.system.profiler.HprofData;
+import dalvik.system.profiler.HprofWriter;
+import dalvik.system.profiler.SamplingProfiler.ThreadSet;
import java.io.ByteArrayInputStream;
import java.io.ByteArrayOutputStream;
import java.io.File;
@@ -221,8 +221,8 @@
private void assertHprofData(HprofData hprofData, boolean strict) throws Exception {
List<ThreadEvent> threadHistory = hprofData.getThreadHistory();
assertNotNull(threadHistory);
- Set<Integer> threadsSeen = new HashSet();
- Set<Integer> threadsActive = new HashSet();
+ Set<Integer> threadsSeen = new HashSet<Integer>();
+ Set<Integer> threadsActive = new HashSet<Integer>();
for (ThreadEvent event : threadHistory) {
assertNotNull(event);
assertNotNull(event.type);