Changed "whitelist" to "allowlist" in many Zygote files.
BYPASS_INCLUSIVE_LANGUAGE_REASON="compatability with vendor files"
Test: cts
Change-Id: Ia0a8548f2e254fc36a9dce7e50af8284ed0caa47
For testing purposes, system components such as mainline modules may
need to relax some security restrictions on debuggable builds only, in
cases where there is no testing strategy that would be reasonable on
user builds security-wise.
One example is to test interfaces between a mainline module and the
rest of the system, which typically require very privileged permissions.
There is no particular reason or permission check to keep the symbol
non-public so add it to the public API.
Test: m
Fixes: 163699557
Change-Id: I15c093af04f024eed5567151b241a4bc35c611d4
UidRange is a data class that is an implementation detail of
Connectivity. Move it to the connectivity boundary.
Remaining usages of UidRange outside of Connectivity (in VPN) should be
migrated to other classes, like Range<Integer> or UidRangeParcel.
Bug: 181512874
Test: m
Change-Id: I6f2e3685ad1c07171dd90480d1e546329de8732d
Previous "return 0" was not reachable, this makes intention
cleaner.
Bug: minor cleanup
Test: TH only
Change-Id: I50d7d1a2ed53b8f8fe245c6ac6aad4fb8b4ce6e3
Since FUSE in R, there are only 4 mount modes that are used
from IVold.aidl:
REMOUNT_MODE_NONE
REMOUNT_MODE_DEFAULT
REMOUNT_MODE_INSTALLER
REMOUNT_MODE_PASS_THROUGH
REMOUNT_MODE_ANDROID_WRITABLE
Bug: 148454884
Test: Manual
Change-Id: I62b66083f2099b429474872b6f9c7cffb9fb906d
Merged-In: I62b66083f2099b429474872b6f9c7cffb9fb906d
Update legecy route is replaced by calling INetd directly.
addLegacyRouteForNetId() is no longer needed. This method also
have dependency with the hidden method in RouteInfo which will
be a part of incoming connectivity module. Thus, remove the
addLegacyRouteForNetId() to resolve the dependency.
Bug: 172183305
Test: m ; atest FrameworksNetTests
Change-Id: Ifc561b6dfb45bc51dee244ae114029ae520d8556
NetworkManagementService that is outside the incoming
connectivity module should stop using hidden NETID_UNSET
in ConnectivityManager. The only usage in NMS is
setDnsForwarders() method which is previously used by CS.
The reference is now replaced by calling INetd directly.
The method could be removed now, so remove it to resolve
the dependency.
Bug: 172183305
Test: m ; atest FrameworksNetTests
Change-Id: I63137b240c4938b082c460bb1a25b447e7f29e80
Do not return to Java mode between consecutive fork operations.
This greatly reduces the Zygote overhead, since we no longer need to
stop and restart Java daemons.
By not switching back to Java mode, and being careful about what memory
we touch between forks, we also keep the Zygote heaps much more stable,
facilitating page sharing between the zygote and all its children.
Under normal operation we should no longer allocate any memory in the
zygote between forks. That applies to both the Java and C++ heap.
This makes the zygote behave much more like the mental model many
of us had assumed: It has nearly constant memory contents, which are
copy-on-right cloned at each fork. This does not apply to the initial
system server and webzygote forks, that are currently still handled
differently.
This includes
1. Add ZygoteCommandBuffer, and switch the argument parsing code to use it.
This slightly reduces allocation and enables (3).
2. Support process specialization in the child, even when the arguments
are already know, Leverages existing Usap code.
3. Add support for forking multiple child processes directly to the
ZygoteCommandBuffer data structure. This directly uses the buffer
internals, and avoids returning to Java so long as it can handle the
zygote commands it sees.
FUNCTIONALITY CHANGE:
We now limit the total size of the zygote command, rather than the
number of arguments.
Initial performance observations:
[ These are not perfect, since I'm comparing to numbers before I
started. There may have been other moving parts, but they should be
minor. ]
System-server-observed launch latency:
[Not the best metric, but easy to measure. In particular, this
does not represent a significant reduction in application launch
time.]
Based on measuring the last 10 launches in a lightly used cf AOSP
instance, the system server latency from requesting an app launch to
response with the pid (which does not require the child to execute
anything) went from an average of about 10.7(25) msecs to 6.8(9) and
7.9(16) in two tries with the CL. (The parenthetical numbers are
maxima from among the 10; the variance appears to have decreased
appreciably.)
Dirty pages:
The number of private dirty pages in the zygote itself appears to have
decreased from about 4000 to about 2200. The number of dalvik-main
private dirty pages went from about 1500 to nearly zero.
Initially ART benchmarking service claim -1.88% in PSS. But this is not
consistently repeatable.
Drive-by fix:
Call setAllowNetworkingForProcess on usap / native loop path.
Bug: 159631815
Bug: 174211442
Test: Boots AOSP
Change-Id: I90d2e381bada1b6c9857666d5e87372b6a4c1a70
* changes:
[IT06]Move INetworkActivityListener into connectivity module
[IT05] Remove the unused network activity logic out from NMS
[IT4.7] Add network activity info into dumpsys for debugging
[IT4.6] Unbundle NMS out from ConnectivityManager
[IT4.5] Update radio power from CS directly
If the device doesn't have a pin, the current RoR preparation will
go through; but the lskf capture event won't happen. In order not
to confuse the caller, throw an exception instead in the no-pin
case.
Bug: 170664917
Test: remove pin, request RoR with adb shell
Change-Id: Ib04c629234b71e3f6cb36bdfe47759a715a69146
- Expand BugreportManager class javadoc
- Explain what "privileged" means
- Highlight only one-BR-at-a-time behavior
- Add more detail to BugreportCallback javadocs
- Explain standard callback invocation flow
- More detail on when BUGREPORT_ERROR_INVALID_INPUT is used
- Add @WorkerThread to {start{,Connectivity},cancel}Bugreport
Bug: 180015324
Test: make
Change-Id: Id3400b1cdd271f47bdeaf8cdc1a0dea61aafc933
INetworkActivityListener is hidden and the only usage is inside
the connectivity module. Thus, move this into module scope.
Bug: 170598012
Test: atest FrameworksNetTests
Change-Id: I0a75c440c1daa773217bbd362b212fda4d07ec64
The network activity logic is moved to ConnectivityService.
CS is no longer setting the idle timer via NMS to netd. The
idle timer should only set from CS, so remove the network
activity logic out from NMS.
Bug: 170598012
Test: atest FrameworksNetTests
Change-Id: Ief9e892586aaee8e9542b9313088fa50d1d411e5
This patch adds the total size of DMA-BUFs exported from the DMA-BUF
heaps framework to dumpsys meminfo.
Bug: 167709539
Test: adb shell dumpsys meminfo
Change-Id: I153721a0dc91ee5084a446ba1ce4f8ec9c51c4f0
Merged-In: I153721a0dc91ee5084a446ba1ce4f8ec9c51c4f0
The DMA buffers mapped by GPU device(s) are already accounted for
in the total exported dmabuf size (ionHeap) size.
To remove GPU memory from lost RAM, provide an API to get only the
DMA buf memory mapped by the GPU device(s).
Bug: 176477627
Test: adb shell dumpsys meminfo
Change-Id: Ie72f46c8de026b143629406714f413677ce6693b
Merged-In: Ie72f46c8de026b143629406714f413677ce6693b
Currently getPssPid includes memtrack HAL reported values in the pss
calculation. This makes it impossible to get these values separately.
Extend this interface to include itemized memtrack values.
Bug: 165832481
Test: dumpsys meminfo
Signed-off-by: Suren Baghdasaryan <surenb@google.com>
Change-Id: Ieb7e1b79ad8ecae6023338d9992467952558e2bd
* changes:
Add total DMA-BUF heap pool size information to dumpsys meminfo
Add total size of DMA-BUFs exported to 'dumpsys meminfo'
Rename getIonMappedSizeKb to getDmabufMappedSizeKb
Some DMA-BUF heaps maintain pools of pre-zeroed memory for faster
allocations. Print the total size of all DMA-BUF heap pools as part of
dumpsys meminfo and use the same in LostRam calculation.
Test: dumpsys meminfo
Bug: 167709539
Change-Id: Ifb61fe596c1a8fab1bd536ec904285e532629b71
Merged-In: Ifb61fe596c1a8fab1bd536ec904285e532629b71
When ION support is not detected, print the total size of DMA-BUFs
exported as part of dumpsys meminfo.
Bug: 167709539
Test: dumpsys meminfo
Change-Id: I3439a8f00cf1aef37e5043b505ed53d47a8a4756
Merged-In: I3439a8f00cf1aef37e5043b505ed53d47a8a4756
The function is actually calculating the total mapped DMA-BUF memory.
Test: dumpsys meminfo
Bug: 167709539
Change-Id: I5c19dc604cf48c3d7ccbfd2d9a83ddd7b35f20d5
Merged-In: I5c19dc604cf48c3d7ccbfd2d9a83ddd7b35f20d5
SysetmConfig is internal service for frameworks only which cannot
use by Connectivity mainline module, but PermissionMonitor which
is part of the module needs to get uids that have been granted
INTERNET/UPDATE_DEVICE_STATS permission from SystemConfig.
Therefore, add a new API to SystemConfigManager that can get
these uids from SysetmConfig.
Bug: 177188455
Test: atest SystemConfigTest
Change-Id: I62bdd969b1b813631ed6223df97bda21a39aec2c
For the incoming connectivity mainline, it need to compute the
intersection of a set of uids and appIds for the permission
control. As @UserIdInt values are internal implementation details
of the OS, any mainline modules should use strongly-typed
UserHandle arguments. Thus, create a new API that accepts
UserHandle and appId to get the computed uid.
Bug: 170598012
Test: atest FrameworksNetTests
Change-Id: I046546d2c5c76eac7a251e63e4234776975bf2bc
Similar to how HIDL HALs are here, we don't know the commandline of the
service, so we can dump them based on service name prefixes.
Power stats and lights are added here, for parity w/ HIDL since these
hvae been converted to AIDL.
Fixes: 175322136
Test: after `adb shell am hang`, we can verify the light service is
dumped, when it wasn't before:
vsoc_x86_64:/data/anr # cat anr_2021-01-28-22-42-44-969 | grep light
Cmd line: /vendor/bin/hw/android.hardware.lights-service.example
Change-Id: I8c8b0cff0c102221875114015a5524c03cfb5b5c
Root cause:
1) a client thread calls "transact", and then, reads a java binder object from the java parcel object
2) the java binder object sends BC_ACQUIRE/BC_INCREFS, but still in the queue, not yet flushed to driver
3) the java parcel object is garbage-collect-ed, and, its finalize method may possibly sends BC_FREE_BUFFER
Because "BC_FREE_BUFFER" is from the java finalize thread, which is different from the "client thread", so, it is possible that "BC_FREE_BUFFER" will be flushed to driver before "BC_ACQUIRE/BC_INCREFS", which makes driver destroy the related "binder_ref" object prematurely.
Consequences of the issue:
The user space process might always hold the above java binder object, whose "handle value" is indeed invalid because the related "binder_ref" in driver is destroyed. This causes a lot of chaos inside the process:
<a> any binder call on the java binder object will be a failure, with kernel log complain: ...got transaction to invalid handle...
<b> afterward, any new bind object passed to this process will NOT create a brand new one for it, instead, it will be simply and incorrectly mapped to the above old biner object, because the new incoming one will use the same "handle value" as the old one. This will make a mess and many weird bugs.
First, it may break "binder object identity" compare based functionality. For example, after registering a listener to a service, all following registering may fail because the latter new listener object is mapped to the old one incorrectly. The service will reject them as "already registered".
Second, binder call on the old binder object is actually dispatched to the new remote binder object, it might even be a success if they are of the same class/interface, which is actually not we expected; and on the other hand, the new binder object may even be of different class/interface, and of course, the binder call may be a failure due to "interface descriptor check".
<c> the user space process cannot recover from the bug automatically unless restart.
Solution:
Hold a temporary reference to the parcel object until "BC_ACQUIRE/BC_INCREFS" flushed to driver.
Bug: 139327211
Test: monkey test for one day and one night
Co-authored-by: Steven Moreland <smoreland@google.com>
Signed-off-by: Jintao Zhu <zhujtcsieee@gmail.com>
Change-Id: I9345f443996b0bdef9d57ddaad119b86205e817f
