dumpsys is a tool that runs on Android devices and provides information about system services. You can call dumpsys from the command line using the Android Debug Bridge (ADB) to get diagnostic output for all system services running on a connected device. This output is typically more verbose than you may want, so use the command line options described below to get output for only the system services you‘re interested in. This page also describes how to use dumpsys to accomplish common tasks, such as inspecting input, RAM, battery, or network diagnostics.

Syntax

The general syntax for using dumpsys is as follows:

 adb shell dumpsys [-t timeout] [--help | -l | --skip services | service [arguments] | -c | -h]

To get a diagnostic output for all system services for your connected device, simply run adb shell dumpsys. However, this outputs far more information than you would typically want. For more manageable output, specify the service you want to examine by including it in the command. For example, the command below provides system data for input components, such as touchscreens or built-in keyboards:

adb shell dumpsys input

For a complete list of system services that you can use with dumpsys, use the following command:

adb shell dumpsys -l

Command line options

The following table lists the available options when using dumpsys.

adb shell dumpsys procstats -h
    

Inspect input diagnostics

Specifying the input service, as shown below, dumps the state of the system’s input devices, such as keyboards and touchscreens, and the processing of input events.

adb shell dumpsys input

The output varies depending on the version of Android running on the connected device. The sections below describe the type of information you typically see.

Event hub state

The following is a sample of what you might see when inspecting the Event Hub State of the input diagnostics:

INPUT MANAGER (dumpsys input)

Event Hub State:
  BuiltInKeyboardId: -2
  Devices:
    -1: Virtual
      Classes: 0x40000023
      Path: 
      Descriptor: a718a782d34bc767f4689c232d64d527998ea7fd
      Location:
      ControllerNumber: 0
      UniqueId: 
      Identifier: bus=0x0000, vendor=0x0000, product=0x0000, version=0x0000
      KeyLayoutFile: /system/usr/keylayout/Generic.kl
      KeyCharacterMapFile: /system/usr/keychars/Virtual.kcm
      ConfigurationFile:
      HaveKeyboardLayoutOverlay: false
    1: msm8974-taiko-mtp-snd-card Headset Jack
      Classes: 0x00000080
      Path: /dev/input/event5
      Descriptor: c8e3782483b4837ead6602e20483c46ff801112c
      Location: ALSA
      ControllerNumber: 0
      UniqueId:
      Identifier: bus=0x0000, vendor=0x0000, product=0x0000, version=0x0000
      KeyLayoutFile:
      KeyCharacterMapFile:
      ConfigurationFile:
      HaveKeyboardLayoutOverlay: false
    2: msm8974-taiko-mtp-snd-card Button Jack
      Classes: 0x00000001
      Path: /dev/input/event4
      Descriptor: 96fe62b244c555351ec576b282232e787fb42bab
      Location: ALSA
      ControllerNumber: 0
      UniqueId:
      Identifier: bus=0x0000, vendor=0x0000, product=0x0000, version=0x0000
      KeyLayoutFile: /system/usr/keylayout/msm8974-taiko-mtp-snd-card_Button_Jack.kl
      KeyCharacterMapFile: /system/usr/keychars/msm8974-taiko-mtp-snd-card_Button_Jack.kcm
      ConfigurationFile:
      HaveKeyboardLayoutOverlay: false
    3: hs_detect
      Classes: 0x00000081
      Path: /dev/input/event3
      Descriptor: 485d69228e24f5e46da1598745890b214130dbc4
      Location:
      ControllerNumber: 0
      UniqueId:
      Identifier: bus=0x0000, vendor=0x0001, product=0x0001, version=0x0001
      KeyLayoutFile: /system/usr/keylayout/hs_detect.kl
      KeyCharacterMapFile: /system/usr/keychars/hs_detect.kcm
      ConfigurationFile:
      HaveKeyboardLayoutOverlay: false
...

Input reader state

The InputReader is responsible for decoding input events from the kernel. Its state dump shows information about how each input device is configured and recent state changes that have occurred, such as key presses or touches on the touch screen.

The following sample shows the output for a touch screen. Note the information about the resolution of the device and the calibration parameters that were used.

Input Reader State
...
  Device 6: Melfas MMSxxx Touchscreen
      IsExternal: false
      Sources: 0x00001002
      KeyboardType: 0
      Motion Ranges:
        X: source=0x00001002, min=0.000, max=719.001, flat=0.000, fuzz=0.999
        Y: source=0x00001002, min=0.000, max=1279.001, flat=0.000, fuzz=0.999
        PRESSURE: source=0x00001002, min=0.000, max=1.000, flat=0.000, fuzz=0.000
        SIZE: source=0x00001002, min=0.000, max=1.000, flat=0.000, fuzz=0.000
        TOUCH_MAJOR: source=0x00001002, min=0.000, max=1468.605, flat=0.000, fuzz=0.000
        TOUCH_MINOR: source=0x00001002, min=0.000, max=1468.605, flat=0.000, fuzz=0.000
        TOOL_MAJOR: source=0x00001002, min=0.000, max=1468.605, flat=0.000, fuzz=0.000
        TOOL_MINOR: source=0x00001002, min=0.000, max=1468.605, flat=0.000, fuzz=0.000
      Touch Input Mapper:
        Parameters:
          GestureMode: spots
          DeviceType: touchScreen
          AssociatedDisplay: id=0, isExternal=false
          OrientationAware: true
        Raw Touch Axes:
          X: min=0, max=720, flat=0, fuzz=0, resolution=0
          Y: min=0, max=1280, flat=0, fuzz=0, resolution=0
          Pressure: min=0, max=255, flat=0, fuzz=0, resolution=0
          TouchMajor: min=0, max=30, flat=0, fuzz=0, resolution=0
          TouchMinor: unknown range
          ToolMajor: unknown range
          ToolMinor: unknown range
          Orientation: unknown range
          Distance: unknown range
          TiltX: unknown range
          TiltY: unknown range
          TrackingId: min=0, max=65535, flat=0, fuzz=0, resolution=0
          Slot: min=0, max=9, flat=0, fuzz=0, resolution=0
        Calibration:
          touch.size.calibration: diameter
          touch.size.scale: 10.000
          touch.size.bias: 0.000
          touch.size.isSummed: false
          touch.pressure.calibration: amplitude
          touch.pressure.scale: 0.005
          touch.orientation.calibration: none
          touch.distance.calibration: none
        SurfaceWidth: 720px
        SurfaceHeight: 1280px
        SurfaceOrientation: 0
        Translation and Scaling Factors:
          XScale: 0.999
          YScale: 0.999
          XPrecision: 1.001
          YPrecision: 1.001
          GeometricScale: 0.999
          PressureScale: 0.005
          SizeScale: 0.033
          OrientationCenter: 0.000
          OrientationScale: 0.000
          DistanceScale: 0.000
          HaveTilt: false
          TiltXCenter: 0.000
          TiltXScale: 0.000
          TiltYCenter: 0.000
          TiltYScale: 0.000
        Last Button State: 0x00000000
        Last Raw Touch: pointerCount=0
        Last Cooked Touch: pointerCount=0

At the end of the input reader state dump there is some information about global configuration parameters, such as the tap interval.

Configuration:
  ExcludedDeviceNames: []
  VirtualKeyQuietTime: 0.0ms
  PointerVelocityControlParameters: scale=1.000, lowThreshold=500.000, highThreshold=3000.000, acceleration=3.000
  WheelVelocityControlParameters: scale=1.000, lowThreshold=15.000, highThreshold=50.000, acceleration=4.000
  PointerGesture:
    Enabled: true
    QuietInterval: 100.0ms
    DragMinSwitchSpeed: 50.0px/s
    TapInterval: 150.0ms
    TapDragInterval: 300.0ms
    TapSlop: 20.0px
    MultitouchSettleInterval: 100.0ms
    MultitouchMinDistance: 15.0px
    SwipeTransitionAngleCosine: 0.3
    SwipeMaxWidthRatio: 0.2
    MovementSpeedRatio: 0.8
    ZoomSpeedRatio: 0.3

Input dispatcher state

The InputDispatcher is responsible for sending input events to applications. As shown in the sample output below, its state dump shows information about which window is being touched, the state of the input queue, whether an ANR is in progress, and so on.

Input Dispatcher State:
  DispatchEnabled: 1
  DispatchFrozen: 0
  FocusedApplication: <null>
  FocusedWindow: name=‘Window{3fb06dc3 u0 StatusBar}‘
  TouchStates: <no displays touched>
  Windows:
    0: name=‘Window{357bbbfe u0 SearchPanel}‘, displayId=0, paused=false, hasFocus=false, hasWallpaper=false, visible=false, canReceiveKeys=false, flags=0x01820100, type=0x000007e8, layer=211000, frame=[0,0][1080,1920], scale=1.000000, touchableRegion=[0,0][1080,1920], inputFeatures=0x00000000, ownerPid=22674, ownerUid=10020, dispatchingTimeout=5000.000ms
    1: name=‘Window{3b14c0ca u0 NavigationBar}‘, displayId=0, paused=false, hasFocus=false, hasWallpaper=false, visible=false, canReceiveKeys=false, flags=0x01840068, type=0x000007e3, layer=201000, frame=[0,1776][1080,1920], scale=1.000000, touchableRegion=[0,1776][1080,1920], inputFeatures=0x00000000, ownerPid=22674, ownerUid=10020, dispatchingTimeout=5000.000ms
    2: name=‘Window{2c7e849c u0 com.vito.lux}‘, displayId=0, paused=false, hasFocus=false, hasWallpaper=false, visible=true, canReceiveKeys=false, flags=0x0089031a, type=0x000007d6, layer=191000, frame=[-495,-147][1575,1923], scale=1.000000, touchableRegion=[-495,-147][1575,1923], inputFeatures=0x00000000, ownerPid=4697, ownerUid=10084, dispatchingTimeout=5000.000ms
    ...
  MonitoringChannels:
    0: ‘WindowManager (server)‘
  RecentQueue: length=10
    MotionEvent(deviceId=4, source=0x00001002, action=2, flags=0x00000000, metaState=0x00000000, buttonState=0x00000000, edgeFlags=0x00000000, xPrecision=1.0, yPrecision=1.0, displayId=0, pointers=[0: (335.0, 1465.0)]), policyFlags=0x62000000, age=217264.0ms
    MotionEvent(deviceId=4, source=0x00001002, action=1, flags=0x00000000, metaState=0x00000000, buttonState=0x00000000, edgeFlags=0x00000000, xPrecision=1.0, yPrecision=1.0, displayId=0, pointers=[0: (335.0, 1465.0)]), policyFlags=0x62000000, age=217255.7ms
    MotionEvent(deviceId=4, source=0x00001002, action=0, flags=0x00000000, metaState=0x00000000, buttonState=0x00000000, edgeFlags=0x00000000, xPrecision=1.0, yPrecision=1.0, displayId=0, pointers=[0: (330.0, 1283.0)]), policyFlags=0x62000000, age=216805.0ms
    ...
  PendingEvent: <none>
  InboundQueue: <empty>
  ReplacedKeys: <empty>
  Connections:
    0: channelName=‘WindowManager (server)‘, windowName=‘monitor‘, status=NORMAL, monitor=true, inputPublisherBlocked=false
      OutboundQueue: <empty>
      WaitQueue: <empty>
    1: channelName=‘278c1d65 KeyguardScrim (server)‘, windowName=‘Window{278c1d65 u0 KeyguardScrim}‘, status=NORMAL, monitor=false, inputPublisherBlocked=false
      OutboundQueue: <empty>
      WaitQueue: <empty>
    2: channelName=‘357bbbfe SearchPanel (server)‘, windowName=‘Window{357bbbfe u0 SearchPanel}‘, status=NORMAL, monitor=false, inputPublisherBlocked=false
      OutboundQueue: <empty>
      WaitQueue: <empty>
    ...
  AppSwitch: not pending
    7: channelName=‘2280455f com.google.android.gm/com.google.android.gm.ConversationListActivityGmail (server)‘, windowName=‘Window{2280455f u0 com.google.android.gm/com.google.android.gm.ConversationListActivityGmail}‘, status=NORMAL, monitor=false, inputPublisherBlocked=false
      OutboundQueue: <empty>
      WaitQueue: <empty>
    8: channelName=‘1a7be08a com.android.systemui/com.android.systemui.recents.RecentsActivity (server)‘, windowName=‘Window{1a7be08a u0 com.android.systemui/com.android.systemui.recents.RecentsActivity EXITING}‘, status=NORMAL, monitor=false, inputPublisherBlocked=false
      OutboundQueue: <empty>
      WaitQueue: <empty>
    9: channelName=‘3b14c0ca NavigationBar (server)‘, windowName=‘Window{3b14c0ca u0 NavigationBar}‘, status=NORMAL, monitor=false, inputPublisherBlocked=false
      OutboundQueue: <empty>
      WaitQueue: <empty>
    ...
  Configuration:
    KeyRepeatDelay: 50.0ms
    KeyRepeatTimeout: 500.0ms

Things to check for

The following is a list of things to consider when inspecting the various output for the input service:

Event hub state:

• All of the input devices you expect are present.

• Each input device has an appropriate key layout file, key character map file, and input device configuration file. If the files are missing or contain syntax errors, then they will not be loaded.

• Each input device is classified correctly. The bits in the Classes field correspond to flags in EventHub.h, such as INPUT_DEVICE_CLASS_TOUCH_MT.

• The BuiltInKeyboardId is correct. If the device does not have a built-in keyboard, then the id must be -2. Otherwise, it should be the id of the built-in keyboard.
• If you observe that the BuiltInKeyboardId is not -2 but it should be, then you are missing a key character map file for a special function keypad somewhere. Special function keypad devices should have key character map files that contain just the line type SPECIAL_FUNCTION (that‘s what in the tuna-gpio-keykad.kcm file we see mentioned above).

Input reader state:

• All of the expected input devices are present.
• Each input device is configured correctly. In particular, check that the touch screen and joystick axes are correct.

Input dispatcher state:

• All input events are processed as expected.
• After touching the touch screen and running dumpsys at the same time, the TouchStates line correctly identifies the window that you are touching.

Test UI performance

Specifying the gfxinfo service provides output with performance information relating to frames of animation that are occurring during the recording phase. The following command uses gfxinfo to gather UI performance data for a specified package name:

adb shell dumpsys gfxinfo package-name

You can also include the framestats option to provide even more detailed frame timing information from recent frames, so that you can track down and debug problems more accurately, shown below:

adb shell dumpsys gfxinfo package-name framestats

To learn more about using gfxinfo and framestats to integrate UI performance measurements into your testing practices, go to Testing UI performance.

Inspect network diagnostics

Specifying the netstats service provides network usage statistics collected since the previous device booted up. To output additional information, such as detailed unique user ID (UID) information, include the detailoption, as follows:

adb shell dumpsys netstats detail

The output varies depending on the version of Android running on the connected device. The sections below describe the type of information you typically see.

Active interfaces and active UID interfaces

The following sample output lists the active interfaces and active UID interfaces of the connected device. In most cases, the information for active interfaces and active UID interfaces is the same.

Active interfaces:
  iface=wlan0 ident=[{type=WIFI, subType=COMBINED, networkId="Guest"}]
Active UID interfaces:
  iface=wlan0 ident=[{type=WIFI, subType=COMBINED, networkId="Guest"}]

‘Dev‘ and ‘Xt‘ statistics

The following is a sample output for the Dev statistics section:

Dev stats:
  Pending bytes: 1798112
  History since boot:
  ident=[{type=WIFI, subType=COMBINED, networkId="Guest", metered=false}] uid=-1 set=ALL tag=0x0
    NetworkStatsHistory: bucketDuration=3600
      st=1497891600 rb=1220280 rp=1573 tb=309870 tp=1271 op=0
      st=1497895200 rb=29733 rp=145 tb=85354 tp=185 op=0
      st=1497898800 rb=46784 rp=162 tb=42531 tp=192 op=0
      st=1497902400 rb=27570 rp=111 tb=35990 tp=121 op=0
Xt stats:
  Pending bytes: 1771782
  History since boot:
  ident=[{type=WIFI, subType=COMBINED, networkId="Guest", metered=false}] uid=-1 set=ALL tag=0x0
    NetworkStatsHistory: bucketDuration=3600
      st=1497891600 rb=1219598 rp=1557 tb=291628 tp=1255 op=0
      st=1497895200 rb=29623 rp=142 tb=82699 tp=182 op=0
      st=1497898800 rb=46684 rp=160 tb=39756 tp=191 op=0
      st=1497902400 rb=27528 rp=110 tb=34266 tp=120 op=0

UID stats

The following is a sample of detailed statistics of each UID.

UID stats:
  Pending bytes: 744
  Complete history:
  ident=[[type=MOBILE_SUPL, subType=COMBINED, subscriberId=311111...], [type=MOBILE, subType=COMBINED, subscriberId=311111...]] uid=10007  set=DEFAULT tag=0x0
    NetworkStatsHistory: bucketDuration=7200000
      bucketStart=1406167200000 activeTime=7200000 rxBytes=4666 rxPackets=7 txBytes=1597 txPackets=10 operations=0
  ident=[[type=WIFI, subType=COMBINED, networkId="MySSID"]] uid=10007  set=DEFAULT tag=0x0
    NetworkStatsHistory: bucketDuration=7200000
      bucketStart=1406138400000 activeTime=7200000 rxBytes=17086802 rxPackets=15387 txBytes=1214969 txPackets=8036 operations=28
      bucketStart=1406145600000 activeTime=7200000 rxBytes=2396424 rxPackets=2946 txBytes=464372 txPackets=2609 operations=70
      bucketStart=1406152800000 activeTime=7200000 rxBytes=200907 rxPackets=606 txBytes=187418 txPackets=739 operations=0
      bucketStart=1406160000000 activeTime=7200000 rxBytes=826017 rxPackets=1126 txBytes=267342 txPackets=1175 operations=35

To find the UID for your app, run this command: adb shell dumpsys package your-package-name. Then look for the line labeled userId.

For example, to find network usage for the app ‘com.example.myapp‘, run the following command:

adb shell dumpsys package com.example.myapp | grep userId

the output should be similar to the following:

    userId=10007 gids=[3003, 1028, 1015]

Using the sample dump above, look for lines that have uid=10007. Two such lines exist—the first indicates a mobile connection and the second indicates a Wi-Fi connection. Below each line, you can see the following information for each two-hour window (which bucketDuration specifies in milliseconds):

• set=DEFAULT indicates foreground network usage, while set=BACKGROUND indicates background usage. set=ALL implies both.
• tag=0x0 indicates the socket tag associated with the traffic.
• rxBytes and rxPackets represent received bytes and received packets in the corresponding time interval.
• txBytes and txPackets represent sent (transmitted) bytes and sent packets in the corresponding time interval.

Inspect battery diagnostics

Specifying the batterystats service generates interesting statistical data about battery usage on a device, organized by unique user ID (UID). To learn how to use dumpsys to test your app for Doze and App Standby, go toTesting with Doze and App Standby.

The command for batterystats is as follows:

adb shell dumpsys batterystats options

To see a list of additional options available to batterystats, include the -h option. The example below outputs battery usage statistics for a specified app package since the device was last charged:

adb shell dumpsys batterystats --charged package-name

The output typically includes the following:

• History of battery-related events
• Global statistics for the device
• Approximate power use per UID and system component
• Per-app mobile milliseconds per packet
• System UID aggregated statistics
• App UID aggregated statistics

To learn more about using batterystats and generating an HTML visualization of the output, which makes it easier to understand and diagnose battery-related issues, read Profile battery usage with Batterystats and Battery Historian.

Inspecting machine-friendly output

You can generate batterystats output in machine-readable CSV format by using the following command:

adb shell dumpsys batterystats --checkin

The following is an example of the output you should see:

9,0,i,vers,11,116,K,L
9,0,i,uid,1000,android
9,0,i,uid,1000,com.android.providers.settings
9,0,i,uid,1000,com.android.inputdevices
9,0,i,uid,1000,com.android.server.telecom
...
9,0,i,dsd,1820451,97,s-,p-
9,0,i,dsd,3517481,98,s-,p-
9,0,l,bt,0,8548446,1000983,8566645,1019182,1418672206045,8541652,994188
9,0,l,gn,0,0,666932,495312,0,0,2104,1444
9,0,l,m,6794,0,8548446,8548446,0,0,0,666932,495312,0,697728,0,0,0,5797,0,0
...

Battery-usage observations may be per-UID or system-level; data is selected for inclusion based on its usefulness in analyzing battery performance. Each row represents an observation with the following elements:

• A dummy integer
• The user ID associated with the observation
• The aggregation mode:
  • "i" for information not tied to charged/uncharged status.
  • "l" for --charged (usage since last charge).
  • "u" for --unplugged (usage since last unplugged). Deprecated in Android 5.1.1.
• Section identifier, which determines how to interpret subsequent values in the line.

The table below describes the various section identifiers you may see:

Note: Prior to Android 6.0, power use for Bluetooth radio, cellular radio, and Wi-Fi was tracked in the m (Misc) section category. In Android 6.0 and higher, power use for these components is tracked in the pwi (Power Use Item) section with individual labels (wifi, blue, cell) for each component.

View memory allocations

You can inspect your app‘s memory usage in one of two ways: over a period of time using procstats or at a particular snapshot in time using meminfo. The sections below show you how to use either method.

procstats

procstats makes it possible to see how your app is behaving over time—including how long it runs in the background and how much memory it uses during that time. It helps you quickly find inefficiencies and misbehaviors in your app, such as memory leaks, that can affect how it performs, especially when running on low-memory devices. Its state dump displays statistics about every application’s runtime, proportional set size (PSS) and unique set size (USS).

To get application memory usage stats over the last three hours, in human-readable format, run the following command:

adb shell dumpsys procstats --hours 3

As can be seen in the example below, the output displays what percentage of time the application was running, and the PSS and USS as minPSS-avgPSS-maxPSS/minUSS-avgUSS-maxUSS over the number of samples.

AGGREGATED OVER LAST 3 HOURS:
  * com.android.systemui / u0a20 / v22:
           TOTAL: 100% (109MB-126MB-159MB/108MB-125MB-157MB over 18)
      Persistent: 100% (109MB-126MB-159MB/108MB-125MB-157MB over 18)
  * com.android.nfc / 1027 / v22:
           TOTAL: 100% (17MB-17MB-17MB/16MB-16MB-16MB over 18)
      Persistent: 100% (17MB-17MB-17MB/16MB-16MB-16MB over 18)
  * android.process.acore / u0a4 / v22:
           TOTAL: 100% (14MB-15MB-15MB/14MB-14MB-14MB over 20)
          Imp Fg: 100% (14MB-15MB-15MB/14MB-14MB-14MB over 20)
  ...
  * com.coulombtech / u0a106 / v26:
           TOTAL: 0.01%
        Receiver: 0.01%
        (Cached): 21% (4.9MB-5.0MB-5.2MB/3.8MB-3.9MB-4.1MB over 2)
  * com.softcoil.mms / u0a86 / v32:
           TOTAL: 0.01%
        (Cached): 0.25%
  * com.udemy.android / u0a91 / v38:
           TOTAL: 0.01%
        Receiver: 0.01%
        (Cached): 0.75% (9.8MB-9.8MB-9.8MB/8.5MB-8.5MB-8.5MB over 1)
  ...
Run time Stats:
  SOff/Norm: +32m52s226ms
  SOn /Norm: +2h10m8s364ms
       Mod : +17s930ms
      TOTAL: +2h43m18s520ms

Memory usage:
  Kernel : 265MB (38 samples)
  Native : 73MB (38 samples)
  Persist: 262MB (90 samples)
  Top    : 190MB (325 samples)
  ImpFg  : 204MB (569 samples)
  ImpBg  : 754KB (345 samples)
  Service: 93MB (1912 samples)
  Receivr: 227KB (1169 samples)
  Home   : 66MB (12 samples)
  LastAct: 30MB (255 samples)
  CchAct : 220MB (450 samples)
  CchCAct: 193MB (71 samples)
  CchEmty: 182MB (652 samples)
  Cached : 58MB (38 samples)
  Free   : 60MB (38 samples)
  TOTAL  : 1.9GB
  ServRst: 50KB (278 samples)

          Start time: 2015-04-08 13:44:18
  Total elapsed time: +2h43m18s521ms (partial) libart.so

meminfo

You can record a snapshot of how your app‘s memory is divided between different types of RAM allocation with the following command:

adb shell dumpsys meminfo package_name|pid [-d]

The -d flag prints more info related to Dalvik and ART memory usage.

The output lists all of your app‘s current allocations, measured in kilobytes.

When inspecting this information, you should be familiar with the following types of allocation:

Private (Clean and Dirty) RAM

This is memory that is being used by only your process. This is the bulk of the RAM that the system can reclaim when your app’s process is destroyed. Generally, the most important portion of this is private dirtyRAM, which is the most expensive because it is used by only your process and its contents exist only in RAM so can’t be paged to storage (because Android does not use swap). All Dalvik and native heap allocations you make will be private dirty RAM; Dalvik and native allocations you share with the Zygote process are shared dirty RAM.

Proportional Set Size (PSS)

This is a measurement of your app’s RAM use that takes into account sharing pages across processes. Any RAM pages that are unique to your process directly contribute to its PSS value, while pages that are shared with other processes contribute to the PSS value only in proportion to the amount of sharing. For example, a page that is shared between two processes will contribute half of its size to the PSS of each process.

A nice characteristic of the PSS measurement is that you can add up the PSS across all processes to determine the actual memory being used by all processes. This means PSS is a good measure for the actual RAM weight of a process and for comparison against the RAM use of other processes and the total available RAM.

For example, below is the output for Map’s process on a Nexus 5 device. There is a lot of information here, but key points for discussion are listed below.

adb shell dumpsys meminfo com.google.android.apps.maps -d

Note: The information you see might vary slightly from what is shown here, because some details of the output differ across platform versions.

** MEMINFO in pid 18227 [com.google.android.apps.maps] **
                   Pss  Private  Private  Swapped     Heap     Heap     Heap
                 Total    Dirty    Clean    Dirty     Size    Alloc     Free
                ------   ------   ------   ------   ------   ------   ------
  Native Heap    10468    10408        0        0    20480    14462     6017
  Dalvik Heap    34340    33816        0        0    62436    53883     8553
 Dalvik Other      972      972        0        0
        Stack     1144     1144        0        0
      Gfx dev    35300    35300        0        0
    Other dev        5        0        4        0
     .so mmap     1943      504      188        0
    .apk mmap      598        0      136        0
    .ttf mmap      134        0       68        0
    .dex mmap     3908        0     3904        0
    .oat mmap     1344        0       56        0
    .art mmap     2037     1784       28        0
   Other mmap       30        4        0        0
   EGL mtrack    73072    73072        0        0
    GL mtrack    51044    51044        0        0
      Unknown      185      184        0        0
        TOTAL   216524   208232     4384        0    82916    68345    14570

 Dalvik Details
        .Heap     6568     6568        0        0
         .LOS    24771    24404        0        0
          .GC      500      500        0        0
    .JITCache      428      428        0        0
      .Zygote     1093      936        0        0
   .NonMoving     1908     1908        0        0
 .IndirectRef       44       44        0        0

 Objects
               Views:       90         ViewRootImpl:        1
         AppContexts:        4           Activities:        1
              Assets:        2        AssetManagers:        2
       Local Binders:       21        Proxy Binders:       28
       Parcel memory:       18         Parcel count:       74
    Death Recipients:        2      OpenSSL Sockets:        2

Here is an older dumpsys on Dalvik of the gmail app:

** MEMINFO in pid 9953 [com.google.android.gm] **
                 Pss     Pss  Shared Private  Shared Private    Heap    Heap    Heap
               Total   Clean   Dirty   Dirty   Clean   Clean    Size   Alloc    Free
              ------  ------  ------  ------  ------  ------  ------  ------  ------
  Native Heap      0       0       0       0       0       0    7800    7637(6)  126
  Dalvik Heap   5110(3)    0    4136    4988(3)    0       0    9168    8958(6)  210
 Dalvik Other   2850       0    2684    2772       0       0
        Stack     36       0       8      36       0       0
       Cursor    136       0       0     136       0       0
       Ashmem     12       0      28       0       0       0
    Other dev    380       0      24     376       0       4
     .so mmap   5443(5) 1996    2584    2664(5) 5788    1996(5)
    .apk mmap    235      32       0       0    1252      32
    .ttf mmap     36      12       0       0      88      12
    .dex mmap   3019(5) 2148       0       0    8936    2148(5)
   Other mmap    107       0       8       8     324      68
      Unknown   6994(4)    0     252    6992(4)    0       0
        TOTAL  24358(1) 4188    9724   17972(2)16388    4260(2)16968   16595     336

 Objects
               Views:    426         ViewRootImpl:        3(8)
         AppContexts:      6(7)        Activities:        2(7)
              Assets:      2        AssetManagers:        2
       Local Binders:     64        Proxy Binders:       34
    Death Recipients:      0
     OpenSSL Sockets:      1

 SQL
         MEMORY_USED:   1739
  PAGECACHE_OVERFLOW:   1164          MALLOC_SIZE:       62

In general, be concerned with only the Pss Total and Private Dirty columns. In some cases, the Private Clean and Heap Alloc columns also offer interesting data. More information about the different memory allocations (the rows) you should observe follows:

Dalvik Heap

The RAM used by Dalvik allocations in your app. The Pss Total includes all Zygote allocations (weighted by their sharing across processes, as described in the PSS definition above). The Private Dirty number is the actual RAM committed to only your app’s heap, composed of your own allocations and any Zygote allocation pages that have been modified since forking your app’s process from Zygote.

Note: On newer platform versions that have the Dalvik Other section, the Pss Total and Private Dirtynumbers for Dalvik Heap do not include Dalvik overhead such as the just-in-time compilation (JIT) and GC bookkeeping, whereas older versions list it all combined under Dalvik.

The Heap Alloc is the amount of memory that the Dalvik and native heap allocators keep track of for your app. This value is larger than Pss Total and Private Dirty because your process was forked from Zygote and it includes allocations that your process shares with all the others.

.so mmap and .dex mmap

The RAM being used for mapped .so (native) and .dex (Dalvik or ART) code. The Pss Total number includes platform code shared across apps; the Private Clean is your app’s own code. Generally, the actual mapped size will be much larger—the RAM here is only what currently needs to be in RAM for code that has been executed by the app. However, the .so mmap has a large private dirty, which is due to fix-ups to the native code when it was loaded into its final address.

.oat mmap

This is the amount of RAM used by the code image which is based off of the preloaded classes which are commonly used by multiple apps. This image is shared across all apps and is unaffected by particular apps.

.art mmap

This is the amount of RAM used by the heap image which is based off of the preloaded classes which are commonly used by multiple apps. This image is shared across all apps and is unaffected by particular apps. Even though the ART image contains Object instances, it does not count towards your heap size.

.Heap (only with -d flag)

This is the amount of heap memory for your app. This excludes objects in the image and large object spaces, but includes the zygote space and non-moving space.

.LOS (only with -d flag)

This is the amount of RAM used by the ART large object space. This includes zygote large objects. Large objects are all primitive array allocations larger than 12KB.

.GC (only with -d flag)

This is the overhead cost for garbage collection. There is not really any way to reduce this overhead.

.JITCache (only with -d flag)

This is the amount of memory used by the JIT data and code caches. Typically, this is zero since all of the apps will be compiled at installed time.

.Zygote (only with -d flag)

This is the amount of memory used by the zygote space. The zygote space is created during device startup and is never allocated into.

.NonMoving (only with -d flag)

This is the amount of RAM used by the ART non-moving space. The non-moving space contains special non-movable objects such as fields and methods. You can reduce this section by using fewer fields and methods in your app.

.IndirectRef (only with -d flag)

This is the amount of RAM used by the ART indirect reference tables. Usually this amount is small, but if it is too high, it might be possible to reduce it by reducing the number of local and global JNI references used.

Unknown

Any RAM pages that the system could not classify into one of the other more specific items. Currently, this contains mostly native allocations, which cannot be identified by the tool when collecting this data due to Address Space Layout Randomization (ASLR). Like the Dalvik heap, the Pss Total for Unknown takes into account sharing with Zygote, and Private Dirty is unknown RAM dedicated to only your app.

TOTAL

The total Proportional Set Size (PSS) RAM used by your process. This is the sum of all PSS fields above it. It indicates the overall memory weight of your process, which can be directly compared with other processes and the total available RAM.

The Private Dirty and Private Clean are the total allocations within your process, which are not shared with other processes. Together (especially Private Dirty), this is the amount of RAM that will be released back to the system when your process is destroyed. Dirty RAM is pages that have been modified and so must stay committed to RAM (because there is no swap); clean RAM is pages that have been mapped from a persistent file (such as code being executed) and so can be paged out if not used for a while.

 

ViewRootImpl

The number of root views that are active in your process. Each root view is associated with a window, so this can help you identify memory leaks involving dialogs or other windows.

AppContexts and Activities

The number of app Context and Activity objects that currently live in your process. This can help you to quickly identify leaked Activity objects that can’t be garbage collected due to static references on them, which is common. These objects often have many other allocations associated with them, which makes them a good way to track large memory leaks.

Note: A View or Drawable object also holds a reference to the Activity that it‘s from, so holding a View orDrawable object can also lead to your app leaking an Activity.