summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
11 daysMerge "Revert "test_mappings: Add OWNERS."" into mainHEADmastermainYifan Hong
12 daysMerge "Delete test_mappings Bazel targets." into mainTreehugger Robot
13 daysAdd an error message to remind users to run from kernel directoryPatrick Rohr
This change will hopefully remind my future self to run these tests from the right directory. Test: run script from Android dir Change-Id: I717baa05e6103d0601644e2b13e96dbed676d4c2
2024-04-09Merge "Revert "net-test: xfrm_algorithm_test - reduce sleep from 0.1s to ↵Treehugger Robot
0.01s"" into main
2024-04-09Revert "test_mappings: Add OWNERS."Yifan Hong
This reverts commit 3f82c7cbaff0c422159f6967e4c6e0b2309886d5. Reason for revert: we are deleting test_mapping Bazel targets, so the build team / kernel test team no longer owns this directory. Change-Id: Id4564e27e300a979fb9d86778b94946913459992
2024-04-09Delete test_mappings Bazel targets.Yifan Hong
We no longer care about a central place for test mapping files. We only need TEST_MAPPING files from common/. Test: TH Bug: 326993647 Change-Id: I30c21f99de98c9c857c6ede0a090dd24e4e8c385
2024-04-09Revert "net-test: xfrm_algorithm_test - reduce sleep from 0.1s to 0.01s"Maciej Żenczykowski
This reverts commit ef3147da3d33e95807fc78e772e9306b2b41db17. Reason for revert: too flaky Bug: 333602696 Change-Id: Ic6c4f898f4cdc9955c14a26c8446c7a2bbcf6711 Signed-off-by: Maciej Żenczykowski <maze@google.com>
2024-04-09Drop //external/zlib:test_mappings.Yifan Hong
//external/xxx is not a valid package. Test: TH Bug: 333557250 Change-Id: I573cfdcfe41e620077e6d59c4fc5827aeda0c07f
2024-04-06net-test: gpylint ping6_test.pyMaciej Żenczykowski
Test: TreeHugger, manually on UML Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: I21a0f9d3ed66ef655bf3eaaf3e7c4fb1aef7851a
2024-04-06net-test: gpylint pf_key_test.pyMaciej Żenczykowski
Test: TreeHugger, manually on UML Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: I768ac776adfdec2a35fe01798cd3fa7bccfac5a6
2024-04-06net-test: xfrm_algorithm_test - reduce sleep from 0.1s to 0.01sMaciej Żenczykowski
hopefully will not cause flakiness to increase Bug: 332955989 Test: TreeHugger, manually on UML Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: Ice4553e7772fdbc9042b31e48e7198024929c9cb
2024-04-06net-test: namespace.py - gpylint fixesMaciej Żenczykowski
'gpylint namespace.py' now no longer complains Test: TreeHugger, manually on UML Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: I42c9b56e21dd138f81763dee8916f738636cf04d
2024-04-05net-test: namespace - remove stdout loggingMaciej Żenczykowski
all_tests.py always calls this, so this is actually spurious info Test: TreeHugger, manually on UML Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: I1f6e74e366ab8c225cdf1e748552b7c804855532
2024-04-05net-test: all_tests.py - move kver log to stderrMaciej Żenczykowski
I think the dashboard only shows stderr on error... Test: TreeHugger, manually no UML Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: I7d71b64f89e875608ba835eaff38fd1c4253e054
2024-04-06net-test: xfrm_algorithm_test - attempt to deflakeMaciej Żenczykowski
We don't need to use 'self.' before server_error if we mark it nonlocal when we assign to it from inner functions. Bump two of the delays as I've seen occasional failures on the timeouts (this may well not help, but shouldn't hurt...) Get rid of the shutdown(RDWR) before close(), it doesn't seem to help (initially it helped, because it always threw an exception, and thus prevented the close() from even happening, thus we would always leak the socket, furthermore due to lack of self. and nonlocal, the server_error assignment was inner function local and thus didn't propagate correctly (ie. the exception was silently hidden), but now that that is all fixed, we still have lots of flakes/failures...). Instead add a 0.1s delay before we close() the accepted socket. I hope this will deflake things. My guess is there is a real kernel bug of some sort here that this works around, but I've yet to figure it out. ACK 5.10-S works without the sleep/shutdown, ACK 5.10-T does not... even though they're the same LTS version of the kernel (but there are a few extra changes in tcp/xfrm) Test: TreeHugger, manually on UML Bug: 332955989 Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: I63007161bfad535b713e483a5a590a82c0ba04b2
2024-04-05net-test: all_tests.py - log kernel version informationMaciej Żenczykowski
This will be useful for debugging failures... New logs: # ./all_tests.py Running on Linux 5.10.209-00013-gf8dd8e9579a9 (5, 10, 209) x86_64-64bit Creating clean namespace... succeeded. testAnycastNetdeviceUnregister (anycast_test.AnycastTest) ... ok ... # ./all_tests_gki.py Running on Linux 5.10.209-00013-gf8dd8e9579a9 (5, 10, 209) x86_64-64bit GKI Creating clean namespace... succeeded. testAnycastNetdeviceUnregister (anycast_test.AnycastTest) ... ok ... Test: TreeHugger Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: I6140ccd014b580b5879df1e4c6244e9813e83308
2024-04-05net-test: leak_test - fix a missing close()Maciej Żenczykowski
*sometimes* (rarely) we see the following: /usr/lib/python3.9/threading.py:896: ResourceWarning: unclosed <socket.socket fd=3, family=AddressFamily.AF_INET6, type=SocketKind.SOCK_DGRAM, proto=0, laddr=('::1', 44185, 0, 0)> del self._target, self._args, self._kwargs Object allocated at (most recent call last): File "/usr/lib/python3.9/unittest/main.py", lineno 271 self.result = testRunner.run(self.test) File "/usr/lib/python3.9/unittest/runner.py", lineno 176 test(result) File "/usr/lib/python3.9/unittest/suite.py", lineno 84 return self.run(*args, **kwds) File "/usr/lib/python3.9/unittest/suite.py", lineno 122 test(result) File "/usr/lib/python3.9/unittest/suite.py", lineno 84 return self.run(*args, **kwds) File "/usr/lib/python3.9/unittest/suite.py", lineno 122 test(result) File "/usr/lib/python3.9/unittest/case.py", lineno 653 return self.run(*args, **kwds) File "/usr/lib/python3.9/unittest/case.py", lineno 593 self._callTestMethod(testMethod) File "/usr/lib/python3.9/unittest/case.py", lineno 550 method() File "/host/mnt/zoom/maze/aosp-tests/net/test/leak_test.py", lineno 30 s = socket(AF_INET6, SOCK_DGRAM, 0) hopefully, this will fix it... Test: TreeHugger Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: I534806aa8bff8d79eba3e05bb900d4664a52db0f
2024-04-05net-test: xfrm_algorithm_test - gpylint fixesMaciej Żenczykowski
'gpylint xfrm_algorithm_test.py' now passes Test: gpylint Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: Icdda9e541fcb669784f030bef50a99627f9579cf
2024-04-05net-test: fix use of non existing socket.socket.SHUT_RDWRMaciej Żenczykowski
socket is actually already socket.socket due to earlier import Test: TreeHugger, manually on UML Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: I0f88d155a7fc78701a1e50fea91feff8e56c413d
2024-04-05net-test: fix propagation of server_error exceptionMaciej Żenczykowski
Test: TreeHugger, manually Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: I51ea188b68747c69d985f16a9a3ead161608b13d
2024-04-05net-test: run_net_test.sh - rework CONFIG_FRAME_WARNMaciej Żenczykowski
The previous version isn't able to build ACK 5.10-T kernel. (Somehow even though it tries to disable it still ends up with 2048) Test: manually on ACK 5.10-T Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: I9292766a9ba8e675bd36b2390246f1703fc157ba
2024-04-04net-test: xfrm_algorithm_test - tentative fix!?!Maciej Żenczykowski
Test: TreeHugger, manually on UML Bug: 332955989 Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: Idc89edc4d10df1abf99f9275b1406a59ef9ce3c3
2024-04-04Merge changes I82db15c1,If838bab9 into mainTreehugger Robot
* changes: net-test: re-enable 'ignore::ResourceWarning' python warnings net-test: close even more leaking sockets
2024-04-04net-test: re-enable 'ignore::ResourceWarning' python warningsMaciej Żenczykowski
Test: TreeHugger, manually on UML with ACK 5.15U Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: I82db15c17d3945ce7868b30f55a8544372f032ca
2024-04-04net-test: close even more leaking socketsMaciej Żenczykowski
Test: TreeHugger, manually on UML 5.15U Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: If838bab9f8b056eb18e707bf9cefe1b2286cab96
2024-04-04Merge "net-test: close some leaking sockets" into mainTreehugger Robot
2024-04-04net-test: close some leaking socketsMaciej Żenczykowski
Test: TreeHugger, manually on UML 5.15U Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: I347c25f436f5d4f92cb73e3550782dfb645a5dfb
2024-04-03test_mappings: remove //common reference.Yifan Hong
We do not want to hardcode //common for anything part of Kleaf tooling, because ACK could be checked out elsewhere. Instead, create a pkg_filegroup so it can be the pkg_zip can refer to them in the //common package. Test: TH Bug: 327493370 Change-Id: Ic8dab284353b20befade6fe5c54f9813adde3e7c
2024-03-28Merge "test_mappings: drop prebuilts/rust/TEST_MAPPING." into mainYifan Hong
2024-03-28test_mappings: drop prebuilts/rust/TEST_MAPPING.Yifan Hong
Temporarily drop prebuilts/rust/TEST_MAPPING because - It doesn't contain useful content; the files it imports do not exist in the kernel tree as of 2024-03-28. - prebuilts/rust does not exist on android15-6.1 manifest. Bug: 330775243 Change-Id: Ic81f591c5000b9172e351285734fd6a30ad49585
2024-03-28test_mappings: Add OWNERS.Yifan Hong
Test: TH Bug: 330775243 Bug: 327493370 Change-Id: I022f9ce33470be88dbe1bf6276464a45789c71cf
2024-03-28test_mappings: fix typo.Yifan Hong
Test: TH Bug: 330775243 Bug: 327493370 Change-Id: I8e3589e556ab29eba7ed87087a17af14d1c15ffb
2024-03-28kleaf: generate test_mappings.zip with pkg_zip.Yifan Hong
... instead of an awkward test_mappings_dist rule. This is so that all inputs are clear. If more TEST_MAPPING files needs to be added, they must be explicitly added to the list. Test: TH Bug: 330775243 Bug: 327493370 Change-Id: Ibd5999cd5ad04e7eeb0be71170a9817d0534e324
2024-03-06net-test: add a new target for old gki branchesMaciej Żenczykowski
MinLTS passes on all GKI ACK branches at HEAD, but fails on old release branches... This change gives us a new test target to run on release branches, without affecting actual V VTS requirements. Test: TreeHugger, manually on UML, atest vts_kernel_net_tests gki_kernel_net_tests Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: I852dae8377b60857f42e702959305adae1254712
2024-03-06Merge changes I7a3642a5,Ieae2f64d,Ibb50fee3,I21fbc0b4 into mainTreehugger Robot
* changes: net-test: skip minLTS test on GSI net-test: minor refactor of kernel_feature_test net-test: print kernel version of kernel which fails test net-test: support /boot/config-$(uname -r)
2024-03-05net-test: skip minLTS test on GSIMaciej Żenczykowski
Testing on GSI basically doesn't make sense as we're testing new VTS + new GSI + old vendor (which includes kernel). As such it is not surprising if this test fails... and not much you can do about it. In practice this test is now more-or-less duplicated in VTS netd_integration_test's kernel_test.cpp. Test: TreeHugger Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: I7a3642a5d3ba79eb6ac51f0bbf0958f4ed2f76b3
2024-03-05net-test: minor refactor of kernel_feature_testMaciej Żenczykowski
to better support difference of built-in vs modular Test: TreeHugger Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: Ieae2f64d1ce43c529c754bd7d8cef0fa2806634e
2024-03-05net-test: print kernel version of kernel which fails testMaciej Żenczykowski
Test: TreeHugger Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: Ibb50fee37d63dce11b3552c97cdb3ab0ce84e54b
2024-03-05net-test: support /boot/config-$(uname -r)Maciej Żenczykowski
Not actually useful on Android itself which always includes /proc/config.gz, but trivial to do and useful elsewhere. Test: TreeHugger Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: I21fbc0b409313ddcb46bf36b1aaecad248520cba
2024-02-28Revert "net-test: kernel_feature_test - temporarily allow 6.1.25" am: a1482f9e92Maciej Żenczykowski
Original change: https://android-review.googlesource.com/c/kernel/tests/+/2973833 Change-Id: I16c2ac44468ee4c93dc3eceee3eaa494d546744d Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
2024-02-23net-test: kernel_feature_test - temporarily allow 6.1.25 am: 9ba9474aacMaciej Żenczykowski
Original change: https://android-review.googlesource.com/c/kernel/tests/+/2974891 Change-Id: Id4cec914e8276082a65e86bc0152267f61aeb733 Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
2024-02-23net-test: V VTS - enforce minimum non-buggy LTS kernel versions. am: 075e66dfa7Maciej Żenczykowski
Original change: https://android-review.googlesource.com/c/kernel/tests/+/2970616 Change-Id: Ia1e15336cadbac2bb8f9adea0d04cd4c78cce82a Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
2024-02-23Revert "net-test: kernel_feature_test - temporarily allow 6.1.25"Maciej Żenczykowski
This reverts commit 9ba9474aac1caa92010ded669d78aba19f8fa7ad. Reason for revert: cuttlefish has upgraded. Change-Id: I4b08f7bb1e3f6ed35b7a9b527ce35532af8c06fa
2024-02-23net-test: kernel_feature_test - temporarily allow 6.1.25Maciej Żenczykowski
cuttlefish is trying to upgrade to 6.6, but is running into some issues, and is thus still on 6.1.25 Test: TreeHugger Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: I6d5a7cd3f986ee749cb58c490c0505dccc782e79
2024-02-21net-test: rename RemovedFeatureTest to KernelFeatureTest am: d471d9f949Maciej Żenczykowski
Original change: https://android-review.googlesource.com/c/kernel/tests/+/2970552 Change-Id: I323e2aceefb6e290d4e6e892a8afb9b6c16cc3f4 Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
2024-02-21net-test: V VTS - enforce minimum non-buggy LTS kernel versions.Maciej Żenczykowski
This is driven by the desire to correctly support - XFRM_MIGRATE, which requires 4.19.236+ 5.4.186+ 5.10.107+ 5.15.30+ see xfrm_tunnel_test.py:55 HasXfrmMigrateFixes() - IPV6 UDP ENCAP, which requires 5.10.108+ 5.15.31+ see xfrm_test.py:488 testIPv6UDPEncapRecvTransport() & testIPv6UDPEncapRecvTunnel() - ACCEPT_MIN_RA_LFT, which requires 5.10.199+ 5.15.136+ 6.1.57+ see multinetwork_base.py:61 HAVE_ACCEPT_RA_MIN_LFT For now we leave those tests as is (with kernel version checks that are now effectively required to be true), as this will result in a cleaner failure message. Bug: 325828213 Test: TreeHugger, manually in UML on 6.1-V Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: I93a100d0032d2943c490b926614cfbb599dd43eb
2024-02-21net-test: rename RemovedFeatureTest to KernelFeatureTestMaciej Żenczykowski
(missed during the rename of the file) Test: TreeHugger Signed-off-by: Maciej Żenczykowski <maze@google.com> Change-Id: Ib50cbc04498ccb499cae81cabaf9543cdba96a8c
2024-02-21net-test: fix gpylint warnings in kernel_feature_test.py am: 6729e2d12eMaciej Żenczykowski
Original change: https://android-review.googlesource.com/c/kernel/tests/+/2970615 Change-Id: Ib746354571367c04f546d11d786d436dd7f1b83e Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
2024-02-21net-test: kernel_feature_test - simplify reject_skerr test am: 554fbe59d6Maciej Żenczykowski
Original change: https://android-review.googlesource.com/c/kernel/tests/+/2970614 Change-Id: I67d189b9af63a54aa0adc22a5316f45c85738266 Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
2024-02-21net-test: rename removed_feature_test to kernel_feature_test am: 1e76fc2ae0Maciej Żenczykowski
Original change: https://android-review.googlesource.com/c/kernel/tests/+/2970613 Change-Id: I267a4150b86acd60d849f9831a6eedabd7370fcd Signed-off-by: Automerger Merge Worker <android-build-automerger-merge-worker@system.gserviceaccount.com>
generated by cgit v1.2.3 (git 2.25.1) at 2024-04-28 11:08:55 +0000