Age | Commit message (Collapse) | Author |
|
When the EthernetNetworkAgent has already been stopped, a call to
unwanted will cause a crash due to a nullptr derefernce.
Bug: 197279734
Test: atest EthernetNetworkFactoryTest
Change-Id: Ice39c603bdd3e3139aa8ece7fb897bf7b567e74b
|
|
Test: atest EthernetNetworkFactoryTest
Change-Id: Ibe0190b3108dca644da2d67bd107954cb26f41ce
|
|
Test: atest EthernetNetworkFactoryTest
Bug: 191635995
Change-Id: I1c07bb6d30706c4e13002eb402fadfecb97b36d1
|
|
Its methods are all no-ops. Stop using them.
Test: m
Bug: 151052811
Change-Id: Iace3ba898bec2940ec3c3323c5bf8a13627d545f
|
|
Update the naming and usgae according to API review feedback.
Bug: 184735772
Test: atest EthernetTrackerTest
Change-Id: Ie8dc0bcdf46ceebda5d6062231b48c1f63b250f6
Merged-In: Ie8dc0bcdf46ceebda5d6062231b48c1f63b250f6
|
|
Bug: 167544279
Test: atest EthernetServiceTests
Change-Id: Ie64e3cff0ace413f14682736de7a1b65fa93e705
Merged-In: Ie64e3cff0ace413f14682736de7a1b65fa93e705
(Cherry-picked from ag/14031548)
|
|
Bug: 182963415
Test: m ethernet-service
Test: atest EthernetServiceTests
Change-Id: I4a66f34daa129b3158e4dd2fb720c175611eaf69
|
|
Replaced by request.canBesatisfiedBy() to avoid direct reference
of hidden member.
Bug: 182963415
Test: m ethernet-service
Test: atest EthernetServiceTests
Change-Id: I2a27410a6148c525e17a11cb1ba3570f04e60b5d
|
|
NetworkCapabilities#clearAll() is a hidden API, use clearAll()
of NetworkCapabilities.Builder instead.
Bug: 182963415
Test: m ethernet-service
Test: atest EthernetServiceTests
Change-Id: I87745b5f7a5817f95cdbc623b2313c181ff48f19
|
|
|
|
The connection service will become the mainline module.
So, need to remove the hidden API usage of NetworkRequest.
The NetworkRequest.type(TRACK_DEFAULT) is no longer sent
to the factories. So, remove the (TRACK_DEFAULT) check to
avoid using hidden API.
Bug: 170598012
Test: atest EthernetServiceTests
Change-Id: I5b5fed864318e9ac7e190c40d4f52076ab0eece8
|
|
The new specifier represents ethernet interfaces more
specifically and is part of the public API.
Bug: 179329291
Test: atest CtsNetTestCases
Merged-In: I6cba1709b3007a22d95849a1281237c77e1464a4
Change-Id: I6cba1709b3007a22d95849a1281237c77e1464a4
|
|
The connection service will become the mainline module. The mutable
NetworkCapabilities is deprecated, and the NetworkCapabilities should
be built through their Builder instead.
Bug: 170598012
Test: atest EthernetServiceTests
Change-Id: I31f12a39a924302beca980b8d920e3c5c18816d4
|
|
Test: FrameworksNetTests
Change-Id: Ib262871e4909ae65b5e6b2e52d2f25bf5a653f34
|
|
|
|
Both EthernetManager#getAvailableInterfaces() and the no-arg
version of EthernetManager#isAvailable() end up calling
EthernetServiceImpl#getAvailableInterfaces. This method attempts
to filter out restricted interfaces if the caller does not have
the CONNECTIVITY_USE_RESTRICTED_NETWORKS permission.
Unfortunately EthernetNetworkFactory's notion of a restricted
interface is the opposite of what it should be. This means that
while privileged callers can see all interfaces, unprivileged
callers can only see restricted interfaces which they cannot use.
Fix this by correcting NetworkInterfaceState#isRestricted(), so
that unprivileged callers can see only unrestricted interfaces.
Privileged callers are unaffected.
This bug probably went unnoticed because restricted Ethernet
interfaces are rare (likely used only on auto devices) and
because the callers that care about those restricted interfaces
are generally privileged.
Bug: 148824036
Test: atest android.net.EthernetTetheringTest#testPhysicalEthernet
Merged-In: Idff3b05a193a80c581d80d0e4fe9c2d3fc0d8a25
Change-Id: Idff3b05a193a80c581d80d0e4fe9c2d3fc0d8a25
|
|
Test: FrameworksNetTests NetworkStackTests
Bug: 152238712
Change-Id: I7c27f76c54a76abeae46dcdcddaeb3ca48afa358
|
|
This reverts the parts of the above commit that have to do
with NetworkScore. The parts that convert the factory to the
new NetworkAgent API are conserved.
Bug: 113554781
Test: FrameworksNetTests
Change-Id: If2c368e084e29f9f358f43dcd547d42e8c7c4846
|
|
Bug: 146583853
Test: FrameworksNetTests
Change-Id: Ie7b70750ef0e17141080f4266dea6155c3601569
|
|
|
|
When releaseNetworkFor is called, refCount should be equal to 1 for
network.stop() to be called. This is the same logic that is followed by
WifiNetworkFactory also.
In the current code, when refCount is 2 and releaseNetworkFor() is called,
network.stop() will be executed and will stop Ethernet
Bug: 146089778
Test: build and boot OK
Change-Id: Ib7d1b488a2943364a8ba4a89eec5de4c33cf1d5a
|
|
This change avoids calling start() on needNetworkFor() if ethernet
cable is unplugged. Fixed by adding isLinkUp validation.
Test: ./cts-tradefed run cts-dev -d -m CtsSecurityTestCases -t \
android.security.cts.ListeningPortsTest#testNoRemotelyAccessibleListeningUdpPorts
adb shell netstat -lnup
Change-Id: I2debd7367a20b665249f9c19d2d8a61737435803
|
|
See aosp/708790
Bug: 18637384
Bug: 29030667
Test: manual
cts
runtest framework-net
atest frameworks/opt/net/ethernet/tests
Change-Id: I5caa4cb89ca1441f3c84b5675ceede6ecacfd132
|
|
When this.mIpConfig is not initialized, it will cause a
java.lang.NullPointerException and put the device in a boot
loop with the following error:
FATAL EXCEPTION IN SYSTEM PROCESS: EthernetServiceThread
Test: passed pre-submit boot test.
Change-Id: I47df68071b4c07a4136c0abcbe69ee7ada7090e0
|
|
Bug: 119085164
Test: manual, test with AndroidTV device
Partner gerrit review CL:
https://partner-android-review.googlesource.com/c/platform/frameworks/opt/net/ethernet/+/1215654
[Analysis]
APP called EthernetManager.setConfiguration API for switch DHCP to
Static,
setConfiguration’s flow just only update IpConfig finally but have not
trigger EthernetNetworkFactory.start().
So Static IP Setting just only take effect when unplug/plug Ethernet
Cable or AC Off&On.
[Repeat Steps]
1. Connect Ethernet with DHCP and Network access is normal
2. Change IP settings from DHCP to Static and set Static IP.
3. After setting, you can not automatically obtain Static IP.
[Recovery Method]
Unplug the Ethernet and insert it or AC OFF / ON
Change-Id: Idbb6a24ccfa360582f48e4f9369def7fc729fb9f
Bug: 112171349
|
|
Bug: b/112588045
Test: atest EthernetServiceTests --verbose
Test: kitchensink, dumpsys ethernet, with RPi devices and USB to
ethernet adapters.
Change-Id: Iaa260aabbbd7aa4b7864eba32ecff7e3f84123c3
|
|
Test: atest EthernetServiceTests
Bug: b/112869080
Change-Id: I54f32d3031e63c7dea78936b6fe5e83e5b293c33
|
|
Bug: b/112588045
Test: atest EthernetServiceTests --verbose
Test: kitchensink with RPi devices connected by USB -> Ethernet
adapters. Edit config.xml to try different tranport
overrides manually. Use dumpsys ethernet to show final
final network scores.
Change-Id: I482e78a76d06c9c090aa6816db14bb346eb6528b
|
|
These should not affect network refcount and thus should not be
passed to the network transport.
Bug: 116034205
Test: successfully established a VPN when connected to USB ethernet
Change-Id: I66517d86f35b8cfc1e211cb189127187d2744b60
|
|
Bug: 111326363
Test: Manual + atest frameworks-net
Change-Id: Id262df4e41de5c74784637ae4a5459c3fab9fb74
|
|
Test: none needed
Change-Id: Ic6cfedf9cfe60984ce7fbd59bdaec55b0bba3001
|
|
Handle the case when IpClient wasn't created for network interface which
may happen for virtual device
Bug: 78482534
Test: dumpsys ethernet
Merged-In: I374a91265660e814b4f0f033b3970fd98921ca84
Merged-In: Id4ab7c9a8a5238491c8405ae260862e76a7671c1
Change-Id: I3405dd4f72d4a6444448f0d6ec0497aead07bade
(cherry picked from commit 2c002de63839a7ce1ed334af2c75176c00e1e248)
|
|
This makes sure that we never accidentally have two IpClients vying
for control of the same interface.
Test: as follows
- built
- flashed
- booted
Bug: 62476366
Change-Id: Iab4ca7c2445e70b6f8beaf5dc6921cb0c6cd56ed
|
|
This change avoids calling start() on needNetworkFor() if an IpClient
object has already been created. This is the case when the interface is
connected from boot, as is the case in Chrome OS.
Bug: 73396557
Test: Networking is 100% stable in Chrome OS
Change-Id: Iaa073d6477aadca2e4e5233b333d717ad82c1f3d
|
|
- reworked EthernetNetworkFactory to support multiple active Ethernet
interfaces
- allow vendors to specify network capabilities + ip config through XML
config overlay
Test: manual using hikey960 + multiple usb->eth adapters
Change-Id: Ie39bcb0d2a3f960f497222159c7bd5797accaa68
|
|
Can be used by a network to indicate when it's congested, meaning that
apps should defer network traffic that could be done at a later time.
Test: bit FrameworksNetTests:android.net.,com.android.server.net.,com.android.server.connectivity.VpnTest,com.android.server.ConnectivityServiceTest
Bug: 64133169
Change-Id: I97664d041920c8d6e3f9fb7c9e8638925b15c30e
|
|
The "roaming" state of a network really belongs on NetworkCapabilities
instead of being published through NetworkInfo.isRoaming(). One major
reason is to support developers creating NetworkRequests for a
non-roaming network.
Test: bit FrameworksNetTests:android.net.,com.android.server.net.,com.android.server.connectivity.
Bug: 68397798, 16207332
Change-Id: I0722d7b38d0478de3834b869552f48403af9d553
|
|
Ethernet interface.
add different logic to handle static ip configuration.
Bug: b/64914694 <http://b/issue?id=64914694&query=64914694>
Test: follow the following steps to switch ethernet config
1.Plug Ethernet cable ,Ethernet succeed to access internet with DHCP mode
2.Change Ethernet IP settings to Static
3.Input correct ip , gateway, prefix length, DNS
4.UI show Connected, but failed to access Internet.
Change-Id: I223bca85814e94745fbb8b50ca677eb7032e9bb3
(cherry picked from commit 44a107bc8f058c6ac1aab6fbfbf60bf8f3474711)
|
|
If the interface disappears between interfaceAdded() and
setInterfaceUp, we'll crash with an IllegalStateException in
NetworkManagementService#setInterfaceConfig.
Ignore the error instead. This should be safe because we don't
modify any state unless our calls succeed.
Bug: 62870779
Test: builds aosp_bullhead-eng
Change-Id: If8c2375f7fffe25a9fa79dc4f1981c745384a276
|
|
1. Remove the IP provisioning thread and just attempt
provisioning indefinitely whenever we have an interface.
2. Make all methods run on the passed-in handler thread. This
makes it easier to verify correctness by code inspection.
3. Remove the code that changes the factory score depending on
whether we're tracking an interface and have link. This is
unnecessary complexity, as there is no penalty to accepting a
request even if we don't have an interface.
4. Remove code duplication and only have one codepath for
stopping layer 3.
Tested the following are tested with this CL:
- Booting with an interface connected.
- Disconnecting/reconnecting the Ethernet cable repeatedly,
particularly at inconvenient times (e.g., during provisioning).
- Similarly, disconnecting/reconnecting USB Ethernet interfaces.
- Falling back to another Ethernet interface if the currently
tracked Ethernet interface is unplugged.
- Disconnecting and restarting provisioning when provisioning is
lost (e.g., if the default route is deleted).
- Crashing the system server causes Ethernet to reconnect on
restart.
- The above while running watch -n 0.1 adb shell dumpsys ethernet
(cherry picked from commit 3c1204c1f3f0ee4e4e530772053f4003b7d58e99)
Bug: 62308954
Test: tested on marlin with USB ethernet adapters, as described
Test: smoketested on aosp_bullhead
Change-Id: Ie3041f1d453d15c81eee81adfc8593269c499d79
|
|
This fixes a longstanding bug where after a DHCP timeout, we
would never restart the DHCP client and get an IP address until
the link bounced.
Also, two minor improvements:
1. Dump IpManager info when dump() is called.
2. When onLinkPropertiesChange is called, also update
mLinkProperties. We were already sending the updated
LinkProperties to the NetworkAgent, so this is really only
useful for dump(), but it's just one line and safe because
onLinkPropertiesChange already grabs the lock.
Bug: 17733693
Change-Id: I42c3319cb4bc151c547ed721baf5e83f97e23862
|
|
- WaitForProvisioningCallback
- setHttpProxy
- setTcpBufferSizes
Bug: 26991160
Change-Id: I5e02039a35006f0466fb40c43805f0443cd758c2
|
|
Bug: 17733693
Bug: 26991160
Change-Id: Idaee8c64bf452e58b93031d8d7ad666edb0c82ae
|
|
The intent of the code was to catch exceptions, but it wasn't
actually doing that because when a netd command fails we get an
IllegalStateException, not a RemoteException.
Bug: 23581852
Change-Id: I66c7c33bdc7b0d0de4fa5cd92dff054797cf860f
|
|
This was not necessary in L, but it is necessary after
https://partner-android-review.googlesource.com/#/c/205415/ .
Bug: 22936203
Change-Id: I24a96c33e5bb375f36217a1c2cec2076eb5ac8b4
|
|
* commit 'ef8e0b9491467cf8d49fa3af7d43f12e3bd524e4':
eth: make sure dhcp is stopped before bringing up interface.
|
|
Signed-off-by: Pierre Couillaud <pierre@broadcom.com>
Signed-off-by: Blair Prescott <blairp@broadcom.com>
|
|
Bug: 18045481
Change-Id: I3fb517c3e7fa807e2937fda92026301e312be986
|
|
Previously we did not stop DHCP or try to force a disconnect.
This made it impossible to reconnect by unplugging and
replugging an USB Ethernet adapter.
Bug: 17736828
Change-Id: Idd8576a96b063b5ca726ee9149800bdcb0f62bd1
|
|
Network Factories are allowed to go below, but networks need to be
constrained. Allowing the network to go below 0 meant that -1 could
sometimes leak through and foul the logic.
bug:17361330
Change-Id: Ife34ca0f9c233dd3c3df80f6fea580af43afcdeb
|