| b004b6d59d7407a8b1b32adeaafbf2eb04034ba8 |
24-Oct-2014 |
Amit Mahajan <amitmahajan@google.com> |
Change to consider SIGNAL_LOST as permanenet failure only if not attached.
Bug: 18062475
Change-Id: I6c1b220fbb3cc9798419ec190e83d3eaea16ec69
ataConnection.java
cController.java
cTracker.java
cTrackerBase.java
|
| ac664378d32ebae13aacbc6ab5fe043168a03935 |
23-Oct-2014 |
xinhe <xinhe@google.com> |
Device mobile data is not working though mobile data status indicator shows
"H" with full signal
Fix: Do not merge the APN configures which have the same type element. If there
are multiple type elements in the confgures, they should be merge only if
there is no common set between them.
Bug:18080519
Change-Id: Ic3bf5979abdf1623ae57d7a9fc659f6c3005b869
cTracker.java
|
| ba314fddbd9dc88bed37aba86331e6de6cb3f588 |
15-Oct-2014 |
Paul Jensen <pauljensen@google.com> |
Show ProgressDialog when waiting for radio to turn on when provisioning.
Turning on the cellular radio when user selects the provisioning notification
can take 30-60s so a UI is needed to keep the user informed that the radio is
coming up. This is only for provisioning APN support (only AT&T tablets now).
bug:15990642
Change-Id: I1e77c3a1b1a24be7acc7cb3c925c002e5fa2d1a9
cTracker.java
|
| 61cdbcf2de392d043de0e4a533fbc615fc423000 |
21-Oct-2014 |
Amit Mahajan <amitmahajan@google.com> |
Change apn comparison (apnsSimilar()) to make sure same APNs are merged.
Bug: 18015631
Change-Id: I590bfb04600d6f6f6b0f552713b04616bc6a3330
cTracker.java
|
| 6395443719ec3ee0257085945e753d02f603886b |
10-Oct-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Fix enabling of FailFast mechanism.
If you use the wrong obtainMessage it's cast into an Integer
and stored in obj instead of arg1. This causes us to retry onto
the provisioning apn in 3 seconds instead of 20 and speeds up
the process.
bug:15990642
Change-Id: I3deb32d49c9fa77b16af9daf5c6ffb682363f6d1
cTracker.java
|
| 9f638ce6ea260b3139fb627d7d232e720e771e03 |
06-Oct-2014 |
Amit Mahajan <amitmahajan@google.com> |
Merge "fix issues to set data profiles" into lmp-dev
|
| dd5db09966b6d24e2ba0f4d6f5d26985d3303540 |
01-Oct-2014 |
Hui Wang <huiwang@motorola.com> |
fix issues to set data profiles
Bug: 17781567
Change-Id: I62fbe1179c3d71f50b50c70f493e0f86c66c2991
ataProfile.java
cTrackerBase.java
|
| d8e86a5eb1f8757244fad0f9b54d2ce8ceb3287b |
06-Oct-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Merge "Merge similar APNs together when loading All Apns." into lmp-dev
|
| d98d019d2a23bc7a8f96e8388021e5c87b71f158 |
02-Oct-2014 |
Gordon Gao <njh348@motorola.com> |
Optimize CDMA RATs TCP buffer sizes
1. Add CDMA 1xRTT buffer sizes
2. Add CDMA eHPRD buffer sizes
Bug: 17786212
Change-Id: I31633829576ced088d2bcf563205bbe216c648dd
ataConnection.java
|
| 29c6659c8767212d23d417f2b7f032b6c0d82119 |
03-Oct-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Merge similar APNs together when loading All Apns.
They should have been merged together in the DB. Do it now
at this convenient point so that we don't end up
making duplicate requests for the same apn. That can lead
to duplicate calls with diff call IDs and then netd gets confuse
and things don't work.
bug:17634414
Change-Id: Ia77e43f53456f8ecadc9e94c20301fe8a52338d7
cTracker.java
|
| 5f0f5c18a1ec4cc85ba737a9fce3dd489b46c54a |
01-Oct-2014 |
Amit Mahajan <amitmahajan@google.com> |
Merge "Treat SIGNAL_LOST error in dataCallResponse as permanent failure." into lmp-dev
|
| 2b3d92bf5b79eddc8299c1236a83b508b31f3432 |
01-Oct-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Merge "Support bearer-specific DUN APN on CDMA" into lmp-dev
|
| ad31aa03d7cab4d719d81c23df2db33e029c8ced |
29-Sep-2014 |
Amit Mahajan <amitmahajan@google.com> |
Treat SIGNAL_LOST error in dataCallResponse as permanent failure.
Bug: 17632875
Change-Id: Ic2dfcf27039206f9eb4ed29b0a035920b14e2890
cFailCause.java
|
| 1484bfe9c58cfd2ddf59a5b8dad7373a36de946d |
22-Sep-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Support bearer-specific DUN APN on CDMA
The DUN apn needs to be listed last, and this should only be used on networks
that don't show the apn editor.
bug:17571681
Change-Id: Iefa676d3c741ffb2eef201a8683429ed9dbe44d1
ataConnection.java
cTracker.java
|
| b449dc75ef4b9cb996c34a11e758f7e6ca193240 |
17-Sep-2014 |
Paul Jensen <pauljensen@google.com> |
Move mobile provisioning APN support into telephony and rework.
This code had to be reworked to cooperate with the new multi-network
logic. The provisioning flow when using a provisioning APN is:
1. DcTracker detects connecting to provisioning APN, puts up
notification, and turns off radio (as per b/13190133, b/10328264).
2. User selects notification. ProvisionNotificationBroadcastReceiver
receives intent, turns on radio, enables fail-fast mode,
sets up provisioning logic in DcTrackerBase.
3. When radio connects again, DcTracker.completeConnection() launches
browser to facilitate activation. Note that activation can be
performed via any internet connection so it's not a problem that
browser isn't multi-network aware.
bug:17324098
Change-Id: Ibd1c42b1a75795f90a6483d3d0a5a14f88b193d8
cTracker.java
|
| aaf87159f290c573863038a4feb990f509d29ba0 |
25-Sep-2014 |
Sreeram Ramachandran <sreeram@google.com> |
Change type name of mobile from "Cellular" to "MOBILE".
Bug: 17609932
Change-Id: I22331426e04a5b45e73f9e768b7d88c1d4282c78
ataConnection.java
|
| a055c6e35a8b4cc3ffb9e2c1c6c5de87155d42a2 |
23-Sep-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Filter Settings-based DUN apn by bearer.
bug:17330745
Change-Id: Icae2cbc6db10c9cebc20f445ca0d71ea4ed6be77
cTrackerBase.java
|
| 1ed331920d5a765da8b926e6aad2cb668203be53 |
19-Sep-2014 |
Hui Wang <huiwang@motorola.com> |
Control radio restart on PDP_FAIL_REGULAR_DEACTIVATION by a config variable
Bug: 17567651, 16308209
Change-Id: I6bb1519a795dcb7e99a94d366afdfaec21c43d84
cFailCause.java
|
| 4c268833fd406b13b95f77f6d52fda25cd3901dd |
11-Sep-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Merge "Don't advertise dun insecurely" into lmp-dev
|
| c149ab8bbbbf21b09b343878009d381bd4f2adf5 |
11-Sep-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Don't advertise dun insecurely
If there is a carrier specified apn from config.xml unless the current connection
is that setting, don't advertise dun.
If there is no carrier specified apn, the we're ok.
bug:17286525
Change-Id: I6c6f66e9b4dcc9a1dda6fa0fbfea7b7dcc155530
ataConnection.java
|
| 3f52153a5409861eaa6e9aee5c8560bf3447cf48 |
09-Sep-2014 |
Hui Wang <huiwang@motorola.com> |
setup data on connectable apns on apn changed even there have already
been some apns connected
If default apn setting is added after other apn types connected, deafult apn
may not be able to be connected
Bug: 17426028
Change-Id: I5022ebde6ed1f8448755d9076aa5583a84895e00
cTracker.java
|
| f54d520ee2eaeb081c208b3d33b0d1fb6ae49254 |
09-Sep-2014 |
Wink Saville <wink@google.com> |
Fix ArrayOutOfBounds bug, tighten parameter checking.
Also, make logging consistent.
Bug: 17424200
Change-Id: Ie5bef1f557bf88b01eeb97c117807a40689c28e4
ctController.java
|
| 231c3c6271a27b5f03444f4d24b52c41c46566ab |
09-Sep-2014 |
Etan Cohen <etancohen@google.com> |
Fix NPE due to access before checking for NULL of APN setting.
Change-Id: I7bac303a2a5d562f7f1b2569ecfa467160be52a1
cTracker.java
|
| 3262c21feb1eede2917450cfdfd3faa86ee70cab |
25-Aug-2014 |
Shishir Agrawal <shishir@google.com> |
Allow MVNO tethering APNS through gservices flags.
The change allows multiple APNs to be specified in
Settings.Global.TETHER_DUN_APN. It also adds mvnoType, and mvnoMatchData
support to the ApnSetting class.
Bug: 12977533
Change-Id: I424fc405c1e6270bf55eadb130ed839d50575cf1
pnSetting.java
cTracker.java
cTrackerBase.java
|
| 12fffcf0d8df6b8268806d9aa7cc7a662e73743b |
30-Jul-2014 |
Jing Zhao <zhaojing@motorola.com> |
Fix improper mobile data connecting event
mAutoAttachOnCreation keeps DataConnection retry connecting
while ApnContext could still be turned into IDLE state.
Make mAutoAttachOnCreation configuable and keep consistency
between DataConnection and ApnContext
Bug: 17131842
Change-Id: Iebd4886959f5458c1b10c34d2e7f96f29c71d527
ataConnection.java
cAsyncChannel.java
cTracker.java
cTrackerBase.java
|
| 4a4195a11b125e56e25117e92c4c92543b14a5cb |
21-Aug-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Handle case when switching phones.
If the DC exits the mApnContexts may be set to null - handle that case
and don't crash. We had nothing left to do anyway.
bug:17172374
Change-Id: I365c7903e91b9f7f6e95e23761ae57d138d57c0d
ataConnection.java
|
| 8fa68fc28795817ce90fdf2a09b859a2d8453ffb |
18-Aug-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Allow overriding Telephony tcp buffer sizes.
bug:16549611
Change-Id: Id2504b3f91d54eab459be11b61027394ea5dd80e
ataConnection.java
|
| 27176514b825175a0f402fe7fc2eb4e281ce8bdd |
07-Aug-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Set the tcp buffer sizes based on RAT.
Replicates the behavior from KK and before.
Also fixes the use of TelephonyManager constants instead
of RIL constants when setting bandwidth number.
bug: 16549611
Change-Id: I44cbd24974f1eaa0ac0c6a2310afb2e23ee94c8c
ataConnection.java
|
| e9701717e43cc5aacbcf624f77a53be92350662c |
08-Aug-2014 |
w19976 <w19976@motorola.com> |
Configure MTU based on network PCO MTU parameter
Add logic to obtain the mtu from the network PCO parameter and set it to kernel
when the mobile data connection is established. When there is no PCO mtu configured
from the network, the mtu size defined in the corresponding APN will be used. In case
no mtu size is defined for an APN used for data connection, the MCC/MNC based MTU
defined in the framework overaly will be applied.
bug:17046179
Change-Id: I6465d4b8f2076aaa380ae3617fb3f24adbe136d4
pnSetting.java
ataCallResponse.java
ataConnection.java
cTracker.java
|
| 9d4ec7d45061f1da05f16cd244eb0a798e7f36bb |
13-Aug-2014 |
Amit Mahajan <amitmahajan@google.com> |
Revert "Revert "add apn setting parameters, and set apn parameters to bp""
This reverts commit 01f27d2c616797e5ba821159f8aca7b75aa54ae1.
Change-Id: Id9e7cf180092f320f635084f805b97f6409d0f03
pnSetting.java
ataProfile.java
cTracker.java
cTrackerBase.java
|
| 01f27d2c616797e5ba821159f8aca7b75aa54ae1 |
12-Aug-2014 |
Amit Mahajan <amitmahajan@google.com> |
Revert "add apn setting parameters, and set apn parameters to bp"
This reverts commit f4b74430a8c22ab58c3fbbd6c9e7a594514a8615.
Change-Id: I1f300741d858ff39ae573e576479613b084e4616
pnSetting.java
ataProfile.java
cTracker.java
cTrackerBase.java
|
| f4b74430a8c22ab58c3fbbd6c9e7a594514a8615 |
10-Aug-2014 |
Hui Wang <huiwang@motorola.com> |
add apn setting parameters, and set apn parameters to bp
Change-Id: I395b499eb0730894620614b3261bb6a784d53f46
pnSetting.java
ataProfile.java
cTracker.java
cTrackerBase.java
|
| 3ae69aa0ef2e02f140a2229e6a31cfd291528295 |
30-Jul-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Fix DUN based Tethering.
Turns out there was a case issue where the DUN commonly found in the tethering
apn config (in config.xml) didn't match the dun in PhoneConstants.APN_TYPE_DUN
so the NetworkCapabilities generating code din't mark it as DUN capable.
Fixed in the APNSetting as this is constructed infrequently.
bug:16242255
Change-Id: I90504e36bf58ee9b4eb1a757e7daf4786d47a875
pnSetting.java
|
| 3317b0381e9b73176a7f139ce01c4414ef648382 |
25-Jul-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Merge "Notify the world when our LinkProperties change." into lmp-dev
|
| d781fb47e89bb978bae989d96928851c1197c7d2 |
25-Jul-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Notify the world when our LinkProperties change.
Change-Id: I19485346f0941e58c41ebdf5525a36f38800d081
ataConnection.java
|
| 76f43316a5a6082d601bffd4b6898d0bd81e11fc |
17-Jul-2014 |
ram <ram.pl@samsung.com> |
IMS: Adding emergency PDN handling for no sim case.
(Changes already reviewed on klp-wireless-dev with CL 171814)
Change-Id: I732ad4ec350256478bb2ac974722f3b9cbd0e483
Signed-off-by: ram <ram.pl@samsung.com>
cTracker.java
cTrackerBase.java
|
| bc78e2f9988f380a3b88d4cb4a9c0b80b8f44bee |
18-Jul-2014 |
ram <ram.pl@samsung.com> |
IMS: getPCSCF() API enhanced to support multiple APN types.
Change-Id: I5e60d4f153deff5f4cb658e0548a7d9bc399e8b7
Signed-off-by: ram <ram.pl@samsung.com>
cTracker.java
cTrackerBase.java
|
| cf5205f70eb1eac497164124187a088ecb03fff5 |
19-Jun-2014 |
Ram <ram.pl@samsung.com> |
IMS: Adding Support for Emergency APN in Telephony Framework (opt/telephony)
(NOTE: No SIM case will be covered in next patch)
Change-Id: I87e653e29674f0e4821484908f799d6b84659c72
Signed-off-by: Ram <ram.pl@samsung.com>
cTracker.java
cTrackerBase.java
|
| 3aed18eb1c0f0dc3ed9e5793d31995ead0a6afe7 |
19-Jun-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Comment out some subId code that failed
mPhone was giving subId == -1 and failing all retries even well after boot
during OMA DM process and killing the process. Needs to be resolved. This
is a workaround.
bug: 15563289
Change-Id: I43d7ac82a64375f319b5e16f9fff37a2aefa8325
cTrackerBase.java
|
| ed8ec7c629dccf60a2fc817f871bab879302fdbd |
18-Jun-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Merge "Remove Obsolete enable/disableApn functions."
|
| 305122cd621385652826cf7d8cd4e651dc6b5e9f |
17-Jun-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Don't tear down existing cell connections.
NetworkFactories signals indicate if we're interested in making new
connections, but should not be used to tear things down.
ConnectivityService will communicate disconnects directly with NetworkAgents
when it wants them to tear down.
This fixes broken lingers and also potential score ties.
bug:15612739
Change-Id: Ib39ddadaac84898d080764afc365f09f7dce4ffa
ataConnection.java
cTracker.java
|
| 58c7be5dc52e9847470112c95aaf6bbab28f9b92 |
17-Jun-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Remove Obsolete enable/disableApn functions.
ConnectivityService now uses NetworkAgents instead and apps should use
ConnectivityManager.
Change-Id: Ifc5c8f5be802183e70a5d92b224c819815de00e6
cSwitchState.java
cTracker.java
cTrackerBase.java
|
| 50a562c63bf42017fe260b0ea299c3419a06a46d |
13-Jun-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Merge "Listen for retry intents."
|
| d32b58c8e7cea693c98f49b2291455e917cd9301 |
13-Jun-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Listen for retry intents.
Re-adding code stripped out during the NetworkAgent changes. Going back
to having ApnContexts pre-initialized at startup since they are no longer
NetworkAgents. This also lets us preregister for each of their retry alarms.
Undo of damage done by: https://googleplex-android-review.googlesource.com/#/c/476836/
bug:15563289
Change-Id: I6f8855eb24d913c73fef647d4219ae7041216e2b
cTracker.java
|
| 91f8e19cdafe6382ec1469167e208e2e928123cc |
13-Jun-2014 |
Wink Saville <wink@google.com> |
Merge "Merge kwd to master"
|
| a8467dd0c524787104b1ccdddc5e8af10ba729ed |
13-Jun-2014 |
Wink Saville <wink@google.com> |
Merge kwd to master
Change-Id: Idb607c0aa32f80fe4fe1539aedea7a221e9e7f04
ataCallResponse.java
ataConnection.java
cAsyncChannel.java
cController.java
cSwitchAsyncChannel.java
cSwitchState.java
cTracker.java
cTrackerBase.java
ctController.java
|
| b36f67e82882badcf856451e1ffc78ed388c2252 |
12-Jun-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
am 781ada03: Merge "Make NetworkCapabilities publicly immutable." into lmp-preview-dev
* commit '781ada0374c0b111c5cf9488d171c046b3079ff3':
Make NetworkCapabilities publicly immutable.
|
| 781ada0374c0b111c5cf9488d171c046b3079ff3 |
12-Jun-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Merge "Make NetworkCapabilities publicly immutable." into lmp-preview-dev
|
| fd555a178c05e409917f22745743ed59402208e6 |
09-Jun-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Make NetworkCapabilities publicly immutable.
Applying API council comments.
bug: 15142362
Change-Id: Ie0bde68b72656a676d90c0343b9756fe9268d8d6
ataConnection.java
cTracker.java
|
| 5efbe26116b51f55d623d482c43c44abf998c71a |
07-Jun-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Apply new LinkProperties API change.
bug:15142362
Change-Id: I1457111da7d3bd09998f7e010febb8bb4c45c8bc
(cherry picked from commit d68ea30947e9b6cd0d7aa8da23d2373a03310e2b)
ataCallResponse.java
|
| 235dd822c4ff48409d974b4fa7e789ea1fdaa7aa |
10-Jun-2014 |
Amit Mahajan <amitmahajan@google.com> |
am 45a6dd79: am db480f2b: am 63dce7fe: Handle provisioning APN by turning off/on radio.
* commit '45a6dd79c6a54f90fd3737cdf9434fe68cf65f14':
Handle provisioning APN by turning off/on radio.
|
| 45a6dd79c6a54f90fd3737cdf9434fe68cf65f14 |
09-Jun-2014 |
Amit Mahajan <amitmahajan@google.com> |
am db480f2b: am 63dce7fe: Handle provisioning APN by turning off/on radio.
* commit 'db480f2b7bd5ae373458e78b267f939dbb95f5c6':
Handle provisioning APN by turning off/on radio.
|
| 63dce7fe9678c4610a410ebddd49518190f90386 |
07-Jun-2014 |
Amit Mahajan <amitmahajan@google.com> |
Handle provisioning APN by turning off/on radio.
The change is specific to AT&T as they want no signaling from device during provisioning.
I've tested following cases:
- expired AT&T SIM to make sure provisioning flow works as expected.
- airplane mode on/off with both active and expired AT&T SIM.
- wifi <-> mobile transitions work okay.
- LTE with Verizon SIM (basic sanity).
bug: 13190133
Change-Id: I215963174ae6000ae71d1dda693f95413f3d6e81
cTrackerBase.java
|
| d68ea30947e9b6cd0d7aa8da23d2373a03310e2b |
07-Jun-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Apply new LinkProperties API change.
bug:15142362
Change-Id: I1457111da7d3bd09998f7e010febb8bb4c45c8bc
ataCallResponse.java
|
| 2b967cc8c303674fce005ba8aba6dfa0d1d60cba |
07-Jun-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
am b372e784: Correctly set NetCap for "*" APNs
* commit 'b372e78452b9432b51cda43f6264240b04d62dd4':
Correctly set NetCap for "*" APNs
|
| b372e78452b9432b51cda43f6264240b04d62dd4 |
06-Jun-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Correctly set NetCap for "*" APNs
bug: 15455750
Change-Id: I933502b417d7b15aaca98ec16f8d3c60de1e142d
ataConnection.java
|
| f11a4953b12a2d15fe07ae361b47fb5b37954bd4 |
04-Jun-2014 |
Lorenzo Colitti <lorenzo@google.com> |
am 48579fcc: Make requests for restricted networks not require unrestricted access.
* commit '48579fcccf584fbddbff59eaae27d434316a4248':
Make requests for restricted networks not require unrestricted access.
|
| 48579fcccf584fbddbff59eaae27d434316a4248 |
03-Jun-2014 |
Lorenzo Colitti <lorenzo@google.com> |
Make requests for restricted networks not require unrestricted access.
Currently, calling startUsingNetworkFeature for a restricted APN
type (e.g., IMS or FOTA) will create a request that requires
NET_CAPABILITY_NOT_RESTRICTED. Because these APNs are restricted,
when we bring them up we conclude that it does not match the
unrestricted requirement, and we tear them down.
1. Clear the NET_CAPABILITY_NOT_RESTRICTED capability when
creating requests in startUsingNetworkFeature.
2. Refactor the code to a common function so this cannot happen
again.
Bug: 15191336
Change-Id: Id1ec79c58ff79b1a83457ffaecc57d50b61ed4e4
ataConnection.java
|
| 62c954a7a7c4199956c127fcb575ed9114d44491 |
27-May-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Apply refactored NetworkFactory.
Make NetworkFactory a concrete class and divide responsibilites between it and NetworkAgent.
Factory will track requests and by default give a single connect/disconnect api for ease
of use. Then NetworkAgent is created and destroyed as needed with very simple logic.
Change-Id: I401c14a6e5466f2fc63b04219b97ff85bb9af291
(cherry picked from commit 071b9f85ac559a35430ed37c03a66271977b9d17)
pnContext.java
ataConnection.java
cTracker.java
|
| f3ab6ab303e1da929ce26b7c5d63565bff136221 |
23-May-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Revert "Switching to NetworkAgent API"
This reverts commit bce3d2575122929bb27ec8a37d56e96da39a3ca2.
Change-Id: Ia7d1752d4ba9f74f55657c7dbe4d64d60064394f
(cherry picked from commit 608588e9af271f0f5640236809f27c626f9d98e9)
pnContext.java
ataConnection.java
cAsyncChannel.java
cTracker.java
cTrackerBase.java
|
| 071b9f85ac559a35430ed37c03a66271977b9d17 |
27-May-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Apply refactored NetworkFactory.
Make NetworkFactory a concrete class and divide responsibilites between it and NetworkAgent.
Factory will track requests and by default give a single connect/disconnect api for ease
of use. Then NetworkAgent is created and destroyed as needed with very simple logic.
Change-Id: I401c14a6e5466f2fc63b04219b97ff85bb9af291
pnContext.java
ataConnection.java
cTracker.java
|
| 608588e9af271f0f5640236809f27c626f9d98e9 |
23-May-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Revert "Switching to NetworkAgent API"
This reverts commit bce3d2575122929bb27ec8a37d56e96da39a3ca2.
Change-Id: Ia7d1752d4ba9f74f55657c7dbe4d64d60064394f
pnContext.java
ataConnection.java
cAsyncChannel.java
cTracker.java
cTrackerBase.java
|
| 220d43e0611edd0f3a2eb7a33bb38a008fff0868 |
29-May-2014 |
Wink Saville <wink@google.com> |
Revert "Fix build: Revert "am 9349bd96: am 7d604d74: am ccb1ae5a: Merge "When Data RAT changes the new RAT retry connectable APN\'s"""
This reverts commit a82185ccdc6df168fcd093bf49b9c2ada0e1c080.
The original commit used EVENT_DATA_RAT_CHANGED but it was defined
in frameworks/base/telephony/*/DctConstants.java by this AOSP
change:
fb61bfa4470332a7ccfca7a2df5e13baca49d9fa
And I'd forgetten to submit it. Now that EVENT_DATA_RAT_CHANGED
is defined I am reverting the revert.
Change-Id: Ied6665136cb6c36053c8fe03eee5e9f87b637f38
cTracker.java
|
| a82185ccdc6df168fcd093bf49b9c2ada0e1c080 |
29-May-2014 |
Ed Heyl <edheyl@google.com> |
Fix build: Revert "am 9349bd96: am 7d604d74: am ccb1ae5a: Merge "When Data RAT changes the new RAT retry connectable APN\'s""
This reverts commit a259f97accfbad749731e4e95152f2ef3f1115f7, reversing
changes made to cbb34dc8c740d77506d51bb1c9045402301da32e.
cTracker.java
|
| a259f97accfbad749731e4e95152f2ef3f1115f7 |
29-May-2014 |
Wink Saville <wink@google.com> |
am 9349bd96: am 7d604d74: am ccb1ae5a: Merge "When Data RAT changes the new RAT retry connectable APN\'s"
* commit '9349bd96eb30f992b76d2a5a84eb593a093fdeef':
When Data RAT changes the new RAT retry connectable APN's
|
| 7d604d741a387ef1c8ab782890384dc2d4190629 |
29-May-2014 |
Wink Saville <wink@google.com> |
am ccb1ae5a: Merge "When Data RAT changes the new RAT retry connectable APN\'s"
* commit 'ccb1ae5a613a0ad714623f1cbb246dbbd43976b4':
When Data RAT changes the new RAT retry connectable APN's
|
| 50d5d9678ef0e14b2426979eb908afff773d40ec |
01-Feb-2014 |
Zhang Wei-e7976c <e7976c@motorola.com> |
When Data RAT changes the new RAT retry connectable APN's
Issue: In the scenario where phone is in LTE and has data up and running, then user makes a voice
call. During voice call, devices moves out of LTE coverage and moves into 1X. Since 1X does not
support simultaneous voice and data, the data call will be dropped. When the device moves back
into LTE coverage later, device does not attempt to set up the data call again until the voice
call is ended. This causes the data stall.
Fix: register for EVENT_DATA_RAT_CHANGED and attempt SETUP_DATA_CALL upon receiving this event.
Change-Id: Ibe9e68295661f2f2ffaca3057d696df6a1db3f0c
cTracker.java
|
| 980e8ba8312c4afb96adfee97ef17a2704c059b0 |
22-May-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Move dis/enable of mobile data to Telephony
ConnectivityService doesn't do this anymore.
bug:15077247
Change-Id: I3208c91b2c0369b594987f39ca29da7478435513
(cherry picked from commit 3a6fa6e964a0cb361f79302afe18b8d7f5f9ceef)
cTrackerBase.java
|
| 3a6fa6e964a0cb361f79302afe18b8d7f5f9ceef |
22-May-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Move dis/enable of mobile data to Telephony
ConnectivityService doesn't do this anymore.
bug:15077247
Change-Id: I3208c91b2c0369b594987f39ca29da7478435513
cTrackerBase.java
|
| 4dfda5470a2582c0fb543ead6c79ccf598c580e0 |
16-May-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Add unregisterNetworkFactory
Telephony creates multiple network factories sometimes (Rat changes). Give
it the ability to unregister the prior ones first.
Also, pass a name for identification purposes.
bug: 15003007
Change-Id: I717d63363f25c446f8ecf38d933b1a35d744af6e
cTracker.java
|
| bce3d2575122929bb27ec8a37d56e96da39a3ca2 |
30-Apr-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Switching to NetworkAgent API
Change-Id: I7bca6f65ba1c616d821db50b0d24772520c787a7
pnContext.java
ataConnection.java
cAsyncChannel.java
cTracker.java
cTrackerBase.java
|
| 96cce86cf08e37e0f09ed5057b1196e26b302743 |
09-Apr-2014 |
Robert Greenwalt <rgreenwalt@google.com> |
Replace LinkCapabilities with NetworkCapabilities
bug:13885501
Change-Id: I4ebc184c29278c385364ab0993c8783dfd9086fb
ataConnection.java
cAsyncChannel.java
cTracker.java
cTrackerBase.java
|
| 9c180aedfc9f0d20525c0128487d3500e6c0a715 |
02-May-2014 |
Jason Monk <jmonk@google.com> |
Support renaming of ProxyProperties to ProxyInfo
Change-Id: I151bb50bfafac46ad375dec4908e0470f2823d8d
ataConnection.java
cAsyncChannel.java
cTracker.java
|
| 8758315661d65c14d0e10d354a14f80ec1f65db7 |
16-Apr-2014 |
Wink Saville <wink@google.com> |
Add notifyDataConnectionRealTimeInfo.
Change-Id: Iadcbf9765af57c125bcdf74e3fc652ed6c8f7d42
cController.java
|
| e03d37181d0e1846e259813d8c7ddeeb33d4d1df |
19-Mar-2014 |
Wink Saville <wink@google.com> |
am d5adcb04: am c187f0e9: Handle provisioning APN by turning off/on data.
* commit 'd5adcb0455a2253b4e62d72bd8d997081013b266':
Handle provisioning APN by turning off/on data.
|
| ff1e89fdd2cc535c3ce15f6a5bbc6c90a23bdb18 |
19-Mar-2014 |
Wink Saville <wink@google.com> |
am c187f0e9: Handle provisioning APN by turning off/on data.
* commit 'c187f0e9bfb897405d6c41f669fc2aadacb0ee1e':
Handle provisioning APN by turning off/on data.
|
| c187f0e9bfb897405d6c41f669fc2aadacb0ee1e |
13-Mar-2014 |
Wink Saville <wink@google.com> |
Handle provisioning APN by turning off/on data.
This is a start and two tests succeed:
Tested expired AT&T SIM and waiting 15min for alarm to fire.
Tested a provisioned Verizon SIM and works normally.
I've NOT tested AT&T where I've properly completed the provisioning.
I've NOT tested T-Mobile SIM either provisioned or not-provisioned.
I've NOT tested provisioning over WiFi.
I've NOT tested that WiFi <-> Mobile works
I've NOT tested voice calls, SMS, MMS
...
The current bug is below, but it is poorly named either it should be
renamed or a new bug created.
Bug: 13190133
Change-Id: I0a09f642614cd27a8655e9dae764b8999ce485b8
cTrackerBase.java
|
| a29f56db119b93cc5f66c1cf2091a70ef886b30c |
14-Feb-2014 |
Wink Saville <wink@google.com> |
Merge "Handle DATA_CALL_LIST in DCC and update dataActivity properly"
|
| b077af12dfe0abfcdd51ce5c9852940b89567e99 |
24-Aug-2013 |
Yashdev Singh <yashdevs@codeaurora.org> |
Handle DATA_CALL_LIST in DCC and update dataActivity properly
Update mActivity in DCC before notifyDataActivity. Without this change,
the notifyDataActivity was notifying incorrect data activity status
due to race condition between DCT when mActivity was updated and
DCC thread where notifyDataActivity was called.
Bug: 11596853
Change-Id: Iffb7bf369062264dd773338c732bc69318d90b73
cController.java
cTracker.java
cTrackerBase.java
|
| 0742246233c3f03f864d23e52d89e1845d994701 |
27-May-2013 |
Antonio Marín Cerezuela <amarin@swissms.ch> |
Telephony API extension v2
Added the following notifications:
public void notifyPreciseCallState(Phone sender);
public void notifyDisconnectCause(Connection.DisconnectCause cause, int preciseCause);
public void notifyPreciseDataConnectionFailed(Phone sender, String reason, String apnType, String apn, String failCause);
Change-Id: If3b88c679507c529b746046c4a17cf6d9974bd09
cTracker.java
|
| 449cbf85380bd1d6716b7ab002da1efb9493bb02 |
15-Nov-2013 |
Wink Saville <wink@google.com> |
if provisioning apn conditionally restart the radio.
Some radios need a swift kick in the butt after provisioning.
Bug: 11673637
Change-Id: Ie758a303fd751d204fd6baa60217e88df6376aae
cTracker.java
|
| ff51a1e0381b303e5bb7e4e0a8a6ad7d124437bd |
12-Nov-2013 |
Sungmin Choi <sungmin.choi@lge.com> |
Use settings DB only when mcc/mnc matches SIM
Bug: 11600521
Change-Id: Iaeb1f279bf454ad0e7b933b8ac97d6384d599b59
cTrackerBase.java
|
| 53a0a7fd4954c3fd140e1123fcb0ac65d9e6c6c3 |
12-Nov-2013 |
Robert Greenwalt <rgreenwalt@google.com> |
Notify all ApnContexts on a conn of Disconnect
Without this you can end up with default thinking
it's connected when it's actually not.
bug:10212445
Change-Id: Iffe3bf2064ac6c9f8e68ddb0d2b5115deaac2033
ataConnection.java
|
| 3fad7daba62dcb7aafc4adc7f8cc123726ed5a7c |
31-Oct-2013 |
Robert Greenwalt <rgreenwalt@google.com> |
Add single-pdn arbitration.
If a device or network can only support a single data connection
then we need to tear down lower priority connections in favor
of higher priority requests.
bug:10212445
Change-Id: I5ff38ae3be993aee51b6716045e6857b4f2942af
pnContext.java
cTracker.java
cTrackerBase.java
|
| c8dc0c8244aac9f3985a53bc94b8ec2e295db430 |
17-Oct-2013 |
Robert Greenwalt <rgreenwalt@google.com> |
Change how we use provisioning url so post works
Needed to do an http post instead of a get for one carrier.
Do this by putting an auto-submitting form in the data to be
interpreted as a html doc by the browser. The ACTION_VIEW
intent only works on http uri, but by specifying ACTION_MAIN/
CATEGORY_APP_BROWSER we could use data:text/html.
bug:11168810
Change-Id: Ifd33e1c3c7f9f40b6add39e446e6a7d7cde22549
cTracker.java
|
| 45df26444864daad60afdd4d121ab4043da3834b |
07-Sep-2013 |
Sungmin Choi <sungmin.choi@lge.com> |
telephony: Set profile DATA_PROFILE_TETHERED when DUN type
Bug: 11152852
Change-Id: I018e3dfb9c8183b11faedc83cc40be658173c4a3
Conflicts:
src/java/com/android/internal/telephony/dataconnection/DcTracker.java
cTracker.java
|
| 1ec85a29a26bfab4e21ad68000afc637e0cf171c |
27-Sep-2013 |
Wink Saville <wink@google.com> |
Handle SIGNAL_LOST(-3) as a retryable error.
Qualcomm reports SIGNAL_LOST when a signal is lost temporarily
and should be retired. If a radio technology changes then the
DataConnection will stop retrying and the DcTracker will get a
new configuration and setup a new data call. If the rat doesn't
change then DataConnection will reestablish the call on the next
retry attempt.
Bug: 10804691
Change-Id: I4d310e7f2ea2dc46c7bcdd2b8687d1ef6b901e01
cFailCause.java
|
| 9f332c9a4f464b36bac30555c5cfa6431e7211f8 |
19-Sep-2013 |
Wink Saville <wink@google.com> |
Fix potential NPE in ApnContext.isProvisioningApn
Bug: 10839016
Change-Id: Ifa0f974c0df69928a5d1e29c215b5bcf8c4a5fb1
pnContext.java
|
| 150b65fc3b52beab23f0d714e4c3166f90d3985e |
11-Sep-2013 |
Robert Greenwalt <rgreenwalt@google.com> |
Remove special code for fake CDMA APN
revert of https://android-review.googlesource.com/#/c/62943/2
Solved the need for sprint using bearer-specific APNs instead
as the phone type is not a reliable trigger.
bug: 9315957
Change-Id: I0e5545ab4706df2c5917c7534291644c5d8797b3
cTracker.java
|
| 79391c0b55b4dbff55c2963bf7c56af0351dd274 |
05-Sep-2013 |
Wink Saville <wink@google.com> |
Merge commit 'ebca826a' into manualmerge
* commit 'ebca826a':
Add provisioning apn timeout.
Conflicts:
src/java/com/android/internal/telephony/dataconnection/DcTrackerBase.java
Change-Id: Ic89a22167b91e05068160a6ef0a194215913bfe7
|
| ebca826a2b78a8542146ad1600df1cb4275a0823 |
05-Sep-2013 |
Wink Saville <wink@google.com> |
am e9063f69: Add provisioning apn timeout.
* commit 'e9063f6973039d66e12d0d930af03906d377969e':
Add provisioning apn timeout.
|
| e9063f6973039d66e12d0d930af03906d377969e |
05-Sep-2013 |
Wink Saville <wink@google.com> |
Add provisioning apn timeout.
Bug: 10609993
Change-Id: I2dd4fe23fa2b22ef8f4849c161c630d1a85cdb4b
pnContext.java
cTrackerBase.java
|
| dd6e67fb45fa3ce0d43f8d874cd7ffaeddf0926f |
04-Sep-2013 |
Lorenzo Colitti <lorenzo@google.com> |
Don't prefer APNs with no type for initial attach.
When choosing an APN for initial attach, we currently treat an
APN with no type the same as an APN that explicitly sets a type
of "ia". Such an APN will take priority over the preferred APN,
and will even take priority over APNs with the type explicitly
set to "ia".
This breaks attach on networks where there are multiple
MVNOs with the same MCC/MNC and one or more of them has an APN
with no type.
This seems to be unintentional, because the comments say that
APNs with a type of "ia" are preferred over default APNs. Fix it
by only considering an APN to be an initial attach APN if it
explicitly specifies a type of "ia".
Bug: 10614153
Change-Id: I8ed81ab3b58ae2aab2c76b644b653bb2c08da163
cTrackerBase.java
|
| a08decf911ab55f06a0972a2ecb1995c46d9ac1e |
30-Aug-2013 |
Wink Saville <wink@google.com> |
Merge "Add support for a provisioning apn." into klp-dev
|
| c9b81a0c05128694c617fcdd67e73821895822fe |
29-Aug-2013 |
Wink Saville <wink@google.com> |
Add support for a provisioning apn.
When there is a provisioning apn, such as lwaactivate from AT&T it is
not necessary nor desireble to try to make network connections as
they will be no TCP ACK's to SYN messages, except for a limited set
of operator white listed addresses.
This is implemented by adding two commands CMD_IS_PRVISIONING_APN
and CMD_ENABLE_MOBILE_PROVISIONING. These are used by ConnectivityService
CheckMp and to know when we are connected to a provisioning apn and to
turn actually bring up the connection when the user needs to provision.
Also, incorporate the sEnableFailFastRefCounter change from klp to make
a singleton and remove the RuntimeException.
Bug: 10328264
Change-Id: I3925004011bb1243793c4c1b963d923dc2b00cb5
pnContext.java
cTracker.java
cTrackerBase.java
|
| 25cebe4f95bfee61477d4048219a6644aa0aa03b |
29-Aug-2013 |
Wink Saville <wink@google.com> |
Merge commit '43c93591' into manualmerge
* commit '43c93591':
Add support for a provisioning apn.
Conflicts:
src/java/com/android/internal/telephony/dataconnection/DcTracker.java
Change-Id: If78b4af0c466c6ca6939ef475726ad7a9688386a
|
| 9894b3fb2f35e21d9cfd45f233ed093589e14c26 |
24-Aug-2013 |
sy.yun <sy.yun@lge.com> |
fix trysetupdata when mAutoAttachOnCreation true
mAutoAttachOnCreation=true means that device can attach PS network
automatically. To do this, DCT should setup data call even PS is detached.
but when mAttached is false (PS detached) It blocks to setup data call
regardless of mAutoAttachOnCreation value. So changed to allow setup
data call when mAutoAttachOnCreation is true and mAttached is false.
Bug: 10196651
Change-Id: Ia121eb75cbbe932b94efe4c09eb3cb26f1ca1c52
cTracker.java
|
| 43c93591d6261be6dad375979e98b0f6ad2771fd |
29-Aug-2013 |
Wink Saville <wink@google.com> |
Add support for a provisioning apn.
When there is a provisioning apn, such as lwaactivate from AT&T it is
not necessary nor desireble to try to make network connections as
they will be no TCP ACK's to SYN messages, except for a limited set
of operator white listed addresses.
This is implemented by adding two commands CMD_IS_PRVISIONING_APN
and CMD_ENABLE_MOBILE_PROVISIONING. These are used by ConnectivityService
CheckMp and to know when we are connected to a provisioning apn and to
turn actually bring up the connection when the user needs to provision.
Also, incorporate the sEnableFailFastRefCounter change from klp to make
a singleton and remove the RuntimeException.
Bug: 10328264
Change-Id: I3925004011bb1243793c4c1b963d923dc2b00cb5
pnContext.java
cTracker.java
cTrackerBase.java
|
| a6855ef9a8cfcbe9ee01a437d5b35804c29fa5b5 |
26-Aug-2013 |
Wink Saville <wink@google.com> |
am 876763e9: Make EnableFailFastRefCounter a singleton.
* commit '876763e93ad0fa8006d290cc122f12d5e614efa2':
Make EnableFailFastRefCounter a singleton.
|
| 876763e93ad0fa8006d290cc122f12d5e614efa2 |
26-Aug-2013 |
Wink Saville <wink@google.com> |
Make EnableFailFastRefCounter a singleton.
Since there is only one DCT the mEnableFailFastRefCounter can be a
static/singleton and thus properly refcount the
CMD_SET_ENABLE_FAIL_FAST_MOBILE_DATA. There is still one problem
and that is if the PhoneApp crashes the singleton will be reinstantiated
and we'll have the same problem, but the frequency of this is small.
Also, the consequence of this is the DCT will be behaving "normally"
in that it will be attempting to do data-stall recovery although we
might like it not to it is the "normal" behavior.
Bug: 10304904
Change-Id: I4197c4236a402aa247f775e480e747620fd189bd
cTrackerBase.java
|
| b256cf8dcb3a478bd0b47a74a5e355c6f4c03d85 |
23-Aug-2013 |
Robert Greenwalt <rgreenwalt@google.com> |
Merge "fix mvno_match_data OutOfBounds Exception" into klp-dev
|
| 59a71354e0169c7877ffd56d476ddd65ecf5a88d |
19-Aug-2013 |
Robert Greenwalt <rgreenwalt@google.com> |
fix mvno_match_data OutOfBounds Exception
When the length of mvno_match_data is greater than GID from IccRecord,
it will generate StringIndexOutOfBoundsException.
Threrefore, codes are changed to use startsWith() instead of subString().
Bug: 10171336
Change-Id: I827b1d3b861d74e736875bbec1ba5bd28c6c3941
cTracker.java
|
| 7ab10e4710bdb54c6d9a5ee01cd443a42a2689f5 |
14-Aug-2013 |
Sungmin Choi <sungmin.choi@lge.com> |
Change set position of mAttached
When re-attach network after detaching, mAttached.get() is always false.
Although network is attached in modem side and state is connected,
phone cannot know whether connection state or not.
Because nobody notifies to phone.
So before notifyDataConnection(Phone.REASON_DATA_ATTACHED), set mAttached true
Bug: 10070685
Change-Id: Id1356e1c7e188bd484d892dc4255c3aefc81a032
cTracker.java
|
| 16a01fe85457587ad5fd02ea06aa2882baf3535b |
14-Aug-2013 |
Wink Saville <wink@google.com> |
Make EnableFailFastRefCounter a singleton.
Since there is only one DCT the mEnableFailFastRefCounter can be a
static/singleton and thus properly refcount the
CMD_SET_ENABLE_FAIL_FAST_MOBILE_DATA. There is still one problem
and that is if the PhoneApp crashes the singleton will be reinstantiated
and we'll have the same problem, but the frequency of this is small.
Also, the consequence of this is the DCT will be behaving "normally"
in that it will be attempting to do data-stall recovery although we
might like it not to it is the "normal" behavior.
Bug: 10304904
Change-Id: I38ff37445788ec952ca26be45b0374bc0a7ae34d
cTrackerBase.java
|
| 5016947ebeeb99d2f9cc2d902b46e9850b8b747b |
09-Aug-2013 |
Wink Saville <wink@google.com> |
am d1eb383d: am a76fa595: Merge "Add mEnableFailFastRefCounter." into jb-mr2-dev
* commit 'd1eb383d9bd58140607f212e46db64559f408780':
Add mEnableFailFastRefCounter.
|
| 4dbcf752298487d3af0ab778bfa804462ea24271 |
09-Aug-2013 |
Wink Saville <wink@google.com> |
am bc42667e: am 97978662: Restart, cleanup or retry inactve connections.
* commit 'bc42667e465b7dfacfe4592390b14c5518a05e41':
Restart, cleanup or retry inactve connections.
|
| d1eb383d9bd58140607f212e46db64559f408780 |
09-Aug-2013 |
Wink Saville <wink@google.com> |
am a76fa595: Merge "Add mEnableFailFastRefCounter." into jb-mr2-dev
* commit 'a76fa59574734cd1528dff2937231b78ea96e8f3':
Add mEnableFailFastRefCounter.
|
| bc42667e465b7dfacfe4592390b14c5518a05e41 |
09-Aug-2013 |
Wink Saville <wink@google.com> |
am 97978662: Restart, cleanup or retry inactve connections.
* commit '979786625e7ceacf4a545a25704ef8a15338a854':
Restart, cleanup or retry inactve connections.
|
| 585e96097525e593c2ae7edb9fcf62526ab958ae |
09-Aug-2013 |
Wink Saville <wink@google.com> |
am 2ae42257: DataConnection should not retry permanent errors.
* commit '2ae42257ebafdbb0470cedeab41c2f1b418b285c':
DataConnection should not retry permanent errors.
|
| a76fa59574734cd1528dff2937231b78ea96e8f3 |
09-Aug-2013 |
Wink Saville <wink@google.com> |
Merge "Add mEnableFailFastRefCounter." into jb-mr2-dev
|
| d54f9906ce6218a0a9e7a274342082901e7d5d18 |
07-Aug-2013 |
Wink Saville <wink@google.com> |
Add mEnableFailFastRefCounter.
Use a reference counter to control enabling fail fast
as there are now two entities which control failing fast:
ConnectifityService.checkProvisioningUrl and
MobileDataStateTracker.captivePortalCheckComplete.
Bug: 9462512
Change-Id: I0dea0eee519f8ee7f94e79d40e82c18f30d7fe2e
cTrackerBase.java
|
| 57da4aa945252ebed6b505387145f99be8ad5d72 |
07-Aug-2013 |
Wink Saville <wink@google.com> |
Unregistering a null receiver is now fatal.
This causes user builds to break, so only unregister
if we are actually registered.
Bug: 9657468
Change-Id: Ie801e6db37962e9de7559c8451806de1972cf430
cTesterDeactivateAll.java
cTesterFailBringUpAll.java
|
| 979786625e7ceacf4a545a25704ef8a15338a854 |
06-Aug-2013 |
Wink Saville <wink@google.com> |
Restart, cleanup or retry inactve connections.
Instead of retrying all inactive connections, restart radio on very bad
errors, clean those that are permanent failures and retry
the others.
Bug: 10131736
Change-Id: I2527b18d6cda52c55fb94be58fbb1a7a86818ef9
ataConnection.java
cController.java
cFailCause.java
cTrackerBase.java
|
| 2ae42257ebafdbb0470cedeab41c2f1b418b285c |
21-May-2013 |
Wink Saville <wink@google.com> |
DataConnection should not retry permanent errors.
Update DcFailCause with the "no retry" info from ril.h
Bug: 9040015
Change-Id: Ib94af9996c79800322251a45349b117e1f913269
ataConnection.java
cFailCause.java
|
| 11a8314b443c1f1fefa0636f40127f98449846d7 |
02-Aug-2013 |
Wink Saville <wink@google.com> |
am 4ce0e815: Merge "Create ApnSetting if none are found in database."
* commit '4ce0e815f8b0a796685fbc931cd6760c27172a0f':
Create ApnSetting if none are found in database.
|
| f46817da20f9628b7e55d1bb6ff0493f82dd9d85 |
18-Jun-2013 |
Alex Yakavenka <ayakav@codeaurora.org> |
Create ApnSetting if none are found in database.
Bug: 9315957
Change-Id: I0cf0debb703f6a4cad35300315922a5566871f85
cTracker.java
|
| 1b5fe200e47f40f82f0e28502a5f40bce64a82e6 |
27-Jul-2013 |
Wink Saville <wink@google.com> |
Add support for an initial APN
Add PhoneConstants.APN_TYPE_IA, DctConstants.APN_IA_ID and enhance
setInitialAttachApn to use the new Initial Attach apn.
Bug: 8733613
Change-Id: Ia627ac0cf5715660b6d02bb13a83d46ec1727b87
cTracker.java
cTrackerBase.java
|
| bbcd0e4257ad545a6466697088fc4ddec3464547 |
23-Jul-2013 |
Wink Saville <wink@google.com> |
Hipri should not be enabled when we initialize.
Without this fix hipri starts at boot mobile data won't be shutdown
as refCount won't go to zero.
Bug: 9475300
Change-Id: If9587fa5360d777328fda59bbea61a2cbd9fe804
cTracker.java
|
| e056c8263473f67dc78630f5535c3664fabacd23 |
02-Jul-2013 |
Wink Saville <wink@google.com> |
Merge commit '1f95c96e' into fix-merge-conflict
* commit '1f95c96e':
Add CMD_SET_ENABLE_FAIL_FAST_MOBILE_DATA to DcTracker.
Conflicts:
src/java/com/android/internal/telephony/dataconnection/DcTracker.java
Change-Id: I7c7234d1b13647a0534a77937ea51d4051043a55
|
| ec107cb8a36181fa26e61ab51f423ba862deb0fa |
28-Jun-2013 |
Wink Saville <wink@google.com> |
Add CMD_SET_ENABLE_FAIL_FAST_MOBILE_DATA to DcTracker.
To fail fast we turn off data stall detection and do no retires.
Bug: 9279964
Change-Id: I42c326a21e05aa301e9d974ed9ac1d59472780ec
cTracker.java
cTrackerBase.java
|
| 46f190903e93c653b90b970c1a21159bb957575b |
18-Jun-2013 |
Wink Saville <wink@google.com> |
Fix NPE when exiting DcDefaultState.
Bug: 9263045
Change-Id: I62069114d5e9f2b0e90f892d2333b27fe693294d
ataConnection.java
|
| 017166876a51eda9ae6b3254119023604e249bc5 |
12-Jun-2013 |
Wink Saville <wink@google.com> |
Merge commit '6aebe10f' into fix-merge
* commit '6aebe10f':
Fail on permanent errors in DcTracker.
Conflicts:
src/java/com/android/internal/telephony/dataconnection/DcTracker.java
Change-Id: Icbee3d7c73fa170813e0683fb7c22dd78e08d6e8
|
| 4a65a2c9c5d5a46453064e7047582d9045e8addc |
11-Jun-2013 |
Wink Saville <wink@google.com> |
Fail on permanent errors in DcTracker.
This reverts the retry added when fixing bug 9240744:
https://googleplex-android-review.git.corp.google.com/#/c/315817/2
Bug: 9373372
Change-Id: I15d8dcc2c04174d54dd33f8a34813fd72764fc77
cTracker.java
|
| 185370a905597ab107a8f2b0455440017e661321 |
04-Jun-2013 |
Wink Saville <wink@google.com> |
Merge commit 'c93dfda4' into fix-mergeconflict
* commit 'c93dfda4':
Do not retry forever in DataConnection.
Change-Id: I58bed4576638a459537d6b5a4c217e1f2fdb3826
|
| fcc57e87d1620ab7dc877d65e7e85dca77132054 |
03-Jun-2013 |
Wink Saville <wink@google.com> |
Do not retry forever in DataConnection.
When we are initially to connect we limit the the number of
retries so as to go through the list of waiting apns. But if retry
forever is enabled, as it is with Verizon devices, we can burn the
battery down quickly because we're retrying with short initial
timeouts.
In this change I also increase the timeout between apn retries from
5 seconds to 20 seconds to further reduce battery drain when retrying.
And added a retry with a longer timeout if to keep the contract with
ConnectivityService that when enabled we will continue to retry.
Bug: 9240744
Change-Id: I45b9ec5300d602d838dbdeacdbd1fca0c4d65ee3
ataConnection.java
cTracker.java
cTrackerBase.java
|
| 203e588e3c42a81aa8a56f595119c181a63b12ca |
31-May-2013 |
Wink Saville <wink@google.com> |
Handling data registration and radio technology changing.
When the DataRegistrationState (aka DRS) or RilDataRadioTechnology changes
aka Radio Access Technology or RAT) while retrying the DataConnection
StateMachine will return the connection as lost immediately. This allows
the connection to be retried with the correct handling of the changed
DRS or RAT quickly, rather than having to wait until retires are exhausted.
Changes:
- Pass RAT in bringUp
- Add the capability to register for DRS/RAT changes with SST
- When DRS/RAT changes in DcRetryState return LOST_CONNECTION to DcTracker
Bug: 9069568
Change-Id: Id4f29a63421af65ce153d24d646c09f7a1e8269b
ataConnection.java
cAsyncChannel.java
cTracker.java
|
| 187a39f896f88eb6c5e4306d9595546654825976 |
29-May-2013 |
Wink Saville <wink@google.com> |
Change DcTracker to not connect or retry if detached.
If we're detached than there is no possibility of connecting
so fail early. Upon attaching enabled connections will be retried.
Bug: 9024828
Change-Id: I363cd4580aeea68d88eff0b48ce579d2f27a13f3
cTracker.java
|
| 4a9b3afeb2ec4d573eca335a3706392ecf9f281e |
28-May-2013 |
Wink Saville <wink@google.com> |
Change DcTracker construction so only initialization is preformed.
In initApnContexts applyNewState and onSetDependency met could cause
calls to trySetupData which can send notifications or even try to
bring up a connection. This can cause problems as not all objects are
initialized. In particular when I tried to change ApnContext#isReady
to be false it caused a call to mPhone.notifyDataConnection but mPhone
wasn't completely initialized and an NPE occurred.
Bug: 9022095
Change-Id: I6d1bf350f2803aa62fa21b6a6759685456bef6eb
cTracker.java
|
| 8f184cf35fa7fbc8e5ed887beaec0d090000939b |
24-May-2013 |
Wink Saville <wink@google.com> |
Merge "DataConnection should not retry permanent errors."
|
| 5d5eea6ed231163c225144316b0d1913d48678a4 |
26-Apr-2013 |
Sungmin Choi <sungmin.choi@lge.com> |
Set initial attach apn for LTE network
When mobile register LTE network, APN provided by operators is needed.
If mobile does not have apn information, it could not register LTE network.
To deliver APN information which is used during LTE attach to modem chipset,
setLteAttachProfile() and RIL_REQUEST_SET_LTE_ATTACH_PROFILE command is added.
When Telephony DB is loaded or Telephony DB is changed, follow steps.
1. If preferred default ApnContext exists, this is delivered to modem chipset.
2. If preferred default ApnContext is null, find other ApnContext which can
handle default type.
3. If other ApnContext which can handle default type does not exist, any other
ApnContext delivered to modem chipset.
Then modem register LTE network with APN information delivered from AP.
Bug: 8733613
Change-Id: If84c62cddccf96a531dae0527ed1a6592ae76201
Signed-off-by: Seonju Oh <seonju.oh@lge.com>
Signed-off-by: Hyejin Kim <hyejin.kim@lge.com>
Signed-off-by: Sungmin Choi <sungmin.choi@lge.com>
cTracker.java
cTrackerBase.java
|
| 943b5090958bd8a2c0877218b1c8dbc72c4f8758 |
21-May-2013 |
Wink Saville <wink@google.com> |
DataConnection should not retry permanent errors.
Update DcFailCause with the "no retry" info from ril.h
Bug: 9040015
Change-Id: Ib94af9996c79800322251a45349b117e1f913269
ataConnection.java
cFailCause.java
|
| da1dd2e1bccb6141109a65d4e40253f39c405537 |
21-May-2013 |
Wink Saville <wink@google.com> |
Retry even on permanent failures.
The contract with ConnectivityService is that we should continue
to try to connect as long as we're ready. With this change we continue
but length the delay if there were permanent failures.
Change-Id: I83acfd2b07740d3650756aa20dad67b3c53e40d6
cTracker.java
cTrackerBase.java
|
| 1db7da14111228a5079d2970d7d0ce34173000b5 |
20-May-2013 |
Wink Saville <wink@google.com> |
Remap NONE to UNKNOWN on GENERIC_FAILURES for SETUP_DATA_CALL.
Bug: 9040003
Change-Id: I49340b88fe8cfe2c7895c7950e7b9ba0f947b145
(cherry picked from commit 912a45a1c0740a33938ec9c081d9f5baa984d376)
ataConnection.java
|
| 74672e8ee972f12406b72551261b4cc7e0651933 |
18-May-2013 |
Wink Saville <wink@google.com> |
Fix setup of retry filters.
There were multiple registered filters for INTENT_RESTART_TRYSETUP_ALARM
this caused a retry not to fire and contributing to not recovering from
a RAT change error in identified by Samsung in post #22 of 8849653.
Bug: 8849653
Change-Id: I9c5cdb740d87a05dd69d72f8bc5596230b75d6f2
cTracker.java
cTrackerBase.java
|
| 2428693913ae731d4ace3414429f5e91af24ea36 |
14-May-2013 |
Wink Saville <wink@google.com> |
Do a disconnect if ApnContext is ready but retrying or failed.
This change allows enabling of a ready ApnContext to reconnect if the
ApnContext is IDLE, FAILED or RETRYING. Without this change toggling
wifi or otherwise enabling the connection would not cause the connection
to be retried. Instead, toggling airplane mode would be required.
Bug: 8849653
Change-Id: I83f44dbfd412f4b0122a2108e22f36db94c59d18
cTracker.java
|
| 9a480819c79acf31f328d254b1bb9fe6cfac2c34 |
03-May-2013 |
Wink Saville <wink@google.com> |
On EVENT_CONNECT in DcRetryingState do not go inactive.
Going to DcInactiveState from DcRetryingState on EVENT_CONNECT causes
us to forget we are retrying the original connection and if this is
a different connection we can disconnect completely when we shouldn't.
For example, when we're retrying the default connection and we get a
request to connect to supl or mms which are short lived connections
we'll become completely disconnected.
Bug: 8770139
Change-Id: Id271bc7f88d12e6d9fe9451b09f9c9412e6959f4
ataConnection.java
|
| 0f94d3aebcf0daf154bf0876164844d09067f904 |
30-Apr-2013 |
Wink Saville <wink@google.com> |
Add synchronized to toString.
If toString is called at the same time removeWaitingApn we can
get a ConncurrentModificationException.
Bug: 8727610
Change-Id: I95c2181a1a5abc1a28a9e4eb337b674e49847103
pnContext.java
|
| 4750c8c11836338b024e159f04f0cbd13c7444b9 |
29-Mar-2013 |
Wink Saville <wink@google.com> |
Fix TODO in ApnContext#setDataConnectionAc.
Callers that really need to tearDown should call it directly
rather than as a side effect of setDataConnectionAc. The fixes
a "DcActiveState ERROR no such apnContext=" message caused by
an unnecssary tearDown when disconnecting.
Change-Id: I11321d098444659d99dec97b59041cefebb3285e
pnContext.java
cTracker.java
cTrackerBase.java
|
| 1f8e6cacec886736414f3b7d750ad1fa973ab20d |
29-Mar-2013 |
Wink Saville <wink@google.com> |
Merge "Add fix and add some debugging." into jb-mr2-dev
|
| c3b7c73fd6a494483e941adddcb3c0c23f4f0a70 |
28-Mar-2013 |
Wink Saville <wink@google.com> |
Add fix and add some debugging.
Bug: 8492574
Change-Id: I9cb1ab4ec3efb8186a4dd83736781835e1847720
ataConnection.java
|
| c6bbea82bf74ebb492508199b6f3e172b7ce860a |
28-Mar-2013 |
Wink Saville <wink@google.com> |
Map DcConstants.RETRYING to PhoneConstants.State.DISCONNECTED.
Previously RETRYING was mapped to CONNECTING this means that
ConnectivityService will not remove the route and a subsequent
change in IP addresses won't work because an old route has not
been removed.
By mapping to DISCONNECTED ConnectivityService will remove the
route. Another alternative would be to add PhoneConstants.State.RETRYING
but this is a simpler change so we'll try it first.
Bug: 8486114
Change-Id: I1c9946a1e441feda83f13730e835445624a87218
cTrackerBase.java
|
| 1f2a2323572b2f3ab18234726cc1b048c670fbab |
21-Mar-2013 |
Robert Greenwalt <rgreenwalt@google.com> |
Use new mechanism to get messenger to StateTracker
bug:8445208
Change-Id: Ice7dfb0b50c9481d359aed14a51372878185171c
cTracker.java
cTrackerBase.java
|
| ebc8af719102a64a2ae17cc2a836e23e2dbc3838 |
25-Mar-2013 |
Lorenzo Colitti <lorenzo@google.com> |
Allow 0.0.0.0 and :: as default gateways
On point-to-point interfaces, some RILs return the unspecified
address (0.0.0.0 or ::) as the gateway. This is correct, as
routes pointing at point-to-point interfaces don't need
gateways. Currently, telephony ignores these routes, causing
loss of connectivity (for example, when using an IPv6 APN on
Galaxy Nexus). Allow these routes to be set.
Bug: 8475171
Bug: 8276725
Change-Id: I3a770628c9717bb6e451a523417401b56ba55763
ataCallResponse.java
|
| 2cc8c148fa4cb6cba5deac6b011268b4174a0b02 |
25-Mar-2013 |
Wink Saville <wink@google.com> |
Use Build.IS_DEBUGGABLE
Change-Id: I8d4a2e83b8cdcf86cafb622013868f117d0701ef
ataConnection.java
cController.java
cRetryAlarmController.java
cTesterDeactivateAll.java
cTesterFailBringUpAll.java
|
| 454b1dfd508844b42eb775e4ab2359be74d3672b |
23-Mar-2013 |
Wink Saville <wink@google.com> |
Rename a few files and variables.
Change-Id: I4e90dbf57797b9485920f943e24fa7a4c29d070b
pnContext.java
ataConnection.java
ataConnectionAc.java
ataConnectionTracker.java
ataConnectionTrackerBase.java
cAsyncChannel.java
cController.java
cTracker.java
cTrackerBase.java
|
| ff4e317d24f0d23bdc0f306d53ddc51f2f1ecf6a |
22-Mar-2013 |
Wink Saville <wink@google.com> |
Move retrying into DC.
This is the first step in refactoring for bug 4772191.
Bug: 4772191
Change-Id: Id54a20ab192783c63939158670faaf531a527640
pnContext.java
ataCallResponse.java
ataConnection.java
ataConnectionAc.java
ataConnectionBase.java
ataConnectionTracker.java
ataConnectionTrackerBase.java
cController.java
cFailBringUp.java
cFailCause.java
cRetryAlarmController.java
cTesterDeactivateAll.java
cTesterFailBringUpAll.java
|
| 2c74d79cab4eee87acf0e74c0d8bf76be8f0f4aa |
16-Mar-2013 |
Robert Greenwalt <rgreenwalt@google.com> |
am f7bf2056: am 63152045: Merge "add MVNO check routine"
* commit 'f7bf205684ca101326e57921d81963da7541e3c6':
add MVNO check routine
|
| 22d85a8e3a575a6d01d2c788587971657dfe20c6 |
23-Feb-2013 |
Wink Saville <wink@google.com> |
Clean up member variables.
Change-Id: Ib60f350131ade626aca682407ea0b4377b16f6c6
pnSetting.java
ataConnection.java
ataConnectionBase.java
ataConnectionTracker.java
ataConnectionTrackerBase.java
|
| cbaa45bbf2cab852b6c9c3a887e9f803d4e857ea |
23-Feb-2013 |
Wink Saville <wink@google.com> |
Clean up
- Add @Overrides where needed.
- Update javadoc comments
- Remove extra semi-colons
- Rename DataConnection.java to DataConnectionBase.java
- Rename GsmDataConnection.java to DataConnection.java
- Add defaults to switch statements
- Remove/fix most "if (false)" statements. Fixed by using a CONSTANT
- Fix hidden variables by hoisting to base class or renaming
- Tweak some debug output
Change-Id: If38de2fdeaacafbf40cdfd7f84dc5c52030ba2a3
pnContext.java
pnSetting.java
ataConnection.java
ataConnectionAc.java
ataConnectionBase.java
ataConnectionTracker.java
ataConnectionTrackerBase.java
smDataConnection.java
smDataConnectionTracker.java
|
| 4918296afe1c667e9523cdfc799f558f7ebc2bfb |
21-Feb-2013 |
Wink Saville <wink@google.com> |
Move DataConnection related classes to their own package.
Change-Id: I34ee9e87545e6b15ebf4d3ec101d4ed3d322f7b2
pnContext.java
pnSetting.java
ataConnection.java
ataConnectionAc.java
ataConnectionTracker.java
smDataConnection.java
smDataConnectionTracker.java
|