If we run an iterations test, TCP connections opened by previous tests
won't be closed immediately. They will stay in the TIME_WAIT state for
120 seconds before closed. Netfilter socket listens to both
NF_NETLINK_CONNTRACK_NEW and NF_NETLINK_CONNTRACK_DESTROY groups.
There is a tiny time gap between socket creation and sending dump
request to kernel. If an old TCP connection happened to be closed in
this time gap, listening socket would see a DESTROY message before
seeing the NEW messages triggered by dump request.
The current test only parsing the first contrack message in the read
buffer, so it will only find the DESTROY message and ignore all of the
following NEW messages. It does a netfilter socket dump, but does not
check whether the dump produces reasonable output. Now that we have
the ability to parse conntrack messages, do more in-depth checking.
Bug: 254608655
Test: atest ConntrackSocketTest --iteration 500
Change-Id: Icf95adb9d9390d598e9c6bdb4e245c5fce764e3e
Change the name to reflect the cronet class that is being tested
Test: atest CronetApiTest
Bug: 251012627
Change-Id: I8b9c6421393582b57043a140b400f6a1e8ace601
Current test triggers batterystats to update mobile stats by
switching default network and ConnectivityService will call
reportMobileRadioPowerState to have batterystats update the
mobile stats. But in current design, batterystats will
rate-limit the update if it is due to modem power state change.
So the modem stats might not be updated if it's rate-limited.
The other reasons (plug/unplug, battery level change,etc) will
be unaffected. Thus, dump battery stats in the test to trigger
a full sync of data.
Bug: 256087047
Test: CtsNetTestCases:BatteryStatsManagerTest
Change-Id: I936d51f0c205640eac94f80062ad66f55df9d0df
* It seems that an apostrophe is required for the command to be executed successfully
Test: update_results.sh && m cronet_aml_components_cronet_android_cronet
Change-Id: Ide44e8be9d7db6894cb5bf1bcf0c3e4d67c2a228
Mohannad Farragauthored and Motomu Utsumicommited 2022-12-05 12:09:49 +09:00
Now, the value in TextEntry would be a zero length byte array if
the text record response doesn't contain the value. The problem
is that the value could either be an empty string ("key="), or
there can be no value ("key"), as per rfc6763 6.4 "If there is
no '=' in a DNS-SD TXT record string, then it is a boolean
attribute, simply identified as being present, with no value.".
Bug: 254166302
Test: atest FramworksNetTests
Change-Id: I4eaa5e593ce9ddaa6b6cb2f069c1aef977910796
* This would help when creating a module for both java and c++. Where the java can get a different name depending on the Sanitizer used.
Test: update_results.sh
Change-Id: I9d1180b98a00c14d4c9851041801c6fd646d4739
* This is a copy of the JniGeneratorActionSanitizer class but supposed to be for the java version where it removes the .h files.
Test: update_results.sh
Change-Id: I9fd38e75b57a9a7a25cd2a9b6caea117f88a2f05
* Instead of going through creating `java_genrule` through `create_modules_from_target`. It's better to just create the genrules from `create_action_module` directly.
Test: update_results.sh
Change-Id: I35d1bf8854b5724390ea776c01e54452fcbaaba6
This shouldn't be exposed as it doesn't benefit the module in any way. The module can call `get_cmd` to generate a suitable `cmd`.
Test: update_results.sh
Change-Id: I671b176ca748fc2d956d738199f0d7390a29dea9
* Quotations escaping is a universal operation that would benefit every subclass. Moving this up in the BaseActionSanitizer to cover for the `java_genrule` errors.
Test: update_results.sh
Change-Id: If6a7296b85170294b596566da413332fe395913f
* jni_registration_generator action has extra cmds that needs to be appeneded to the original cmd like generation of the java files.
* Moved `NEWLINE` to the top of the file.
Test: update_results.sh
Change-Id: I92cec28786472ef31b514698422e100885714250
* Move the creation of cmd implementation to `BaseActionSanitizer`. This is important because we don't want to expose the target args as this will change the original target.
Test: update_results.sh
Change-Id: I4932408d8add345ed8e480b0dc2cd65e73d54b09
* This is already called in `create_action_module`. JniGenerator_Java will be added again in an upcoming CL.
[I think create_action_module shouldn't call that in the first place.]
Test: update_results.sh
Change-Id: I409fb37a135c2249f777d47b97c1565552d2b7ab
* This is a GN-specific flag that should live in gn_utils as it might be used in the future when converting to Bazel.
Test: update_results.sh
Change-Id: If870f757cf3ee57a00c571947a7ffea18a7db55d
* This was affecting the original target args as it was being executed after setting target.args . So this is a step in maintaining the original target as a read-only object.
Test: update_results.sh
Change-Id: I6852dbecbd1ce304af091f3e4d00e0d698cd53f8
* changes:
gn2bp: Move arg output from get_outputs to _sanitize_outputs
gn2bp: Move arg sanitizing from get_args to _sanitize_args
gn2bp: Add sanitize API to the BaseActionSanitizer
gn2bp: Update is_header_generated in the BaseActionSanitizer
gn2bp: Update is_header_generated of MakeDafsaSanitizer
gn2bp: Add MakeDafsaSanitizer
gn2bp: Add is_header_generated to the BaseActionSanitizer
gn2bp: Remove unrequired hack for action_with_pydeps
gn2bp: Use the result of get_outputs as module.out
gn2bp: Add get_outputs to JniGeneratorSanitizer
gn2bp: Fix to use set for target.outputs
gn2bp: Add get_outputs to the BaseActionSanitizer
gn2bp: Fix arg normalization and remove hack in the VersionSanitizer
gn2bp: Normalize args by for loop instead of comprehension
gn2bp: Update the comment for value args and list args
gn2bp: Update _is_value_arg
Mohannad Farragauthored and Gerrit Code Reviewcommited 2022-12-02 12:06:03 +00:00
Hungming Chen
commited 
Thiébaud Weksteen
authored and
Gerrit Code Review
commited 
KH Shi
authored and
Paul Hu
authored and
Chidera Olibie
authored and
Aaron Huang
authored and
Motomu Utsumi
commited 
Mohannad Farrag
authored and
Chiachang Wang
authored and
Treehugger Robot
authored and
Bill Yi
commited 
Cole Faust
authored and