This PR refactors the credentials types to remove Secure and Insecure Channel and Call credentials types. We standardize on a `c_creds()` accessor method for all credentials types, which can now be treated uniformly. This notably removes special-case handling of insecure credentials.
The special code-paths for insecure creds are no longer necessary in the wake of #25586
This removes two Executor::Run dependencies, and requires that all ServerCallbackCall implementations implement the new `RunAsync` method. There's one other known other implementation of ServerCallbackCall that will need to be updated.
We could also support an "inefficient" path that uses the default engine (not implemented here), for all subclasses that do not want to update. As far as anyone is aware, the ServerCallbackCall class was never intended to be subclassed externally.
Closes#36126
Remove rolled out experiments v3_compression_filter, v3_server_auth_filter; Extend WIP v3_channel_idle_filter for a bit, and extend work_serializer_clears_time_cache until OSS issues with work_serializer_dispatch are resolved.
Closes#36166
<!--
If you know who should review your pull request, please assign it to that
person, otherwise the pull request would get assigned randomly.
If your pull request is for a specific language, please add the appropriate
lang label.
-->
Closes#36070
These two things are different (event engine being a context has nothing to do with the ability to get a default event engine).
Lumping them into the same header means that things that want the context also need to want the default event engine declarations, which in turn pull in CoreConfiguration -- which ultimately means that nothing that is depended on by CoreConfiguration can consume the event engine context, which is undesirable (and causing dependency loops in current work).
Also includes an impressive number of cleanups to `grpc_base` because I ran `fix_build_deps` on that target as part of this.
Closes#36148
This commit does the following 3 things for Android bazel rules, which depends on each other
1. Update NDK to 26.2.11394342 (and a few other Android things)
2. Use rules_android_ndk instead of builtin `native.android_ndk_repository`
* `third_party/android/android_configure.bzl` was a workaround for users who don't set `$ANDROID_NDK_HOME` env var. Now with rules_android_ndk, we can declare repo without registering the toolchain. Instead users who need NDK toolchain should use `--extra_toolchains` to manually register the toolchain, as shown in README.md.
3. Migrate to platforms. See https://bazel.build/concepts/platforms
* Currently we declares android platforms that is needed for binder transport APK. Later gRPC repo can gradually migrate to platforms for other platforms.
* The value of `crosstool_top` will still be `//external:android/crosstool` (which is the default value), so existings android `config_settings` will still work. We should migrate them to match with `@platforms//os:android` constraint later.
The platforms migration needs Bazel 7.0 to work so we also override bazel version in testing scripts.
Closes#36116
<!--
If you know who should review your pull request, please assign it to that
person, otherwise the pull request would get assigned randomly.
If your pull request is for a specific language, please add the appropriate
lang label.
-->
Closes#36108
This adds the following new targets:
- `channel`: A virtual interface for a channel.
- `legacy_channel`: A channel implementation that supports the filter stack and call v2.
- `channel_create`: A standalone function to create a channel.
- `server_interface`: A base class with a few accessor methods used in surface/call.cc.
- `server`: The actual server implementation.
- `api_trace`, `call_tracer`, `server_call_tracer_filter`, `call_finalization`: These were split out of `grpc_base` to avoid various dependency problems.
- `compression`: This is a combination of the previously existing `compression_internal` target and the compression code that was part of `grpc_base`.
Closes#35924
There's a race with the iomgr/EventEngine shims on listener shutdown, specifically for the PosixEventEngine implementation. This showed up as TSAN failures, such as in https://source.cloud.google.com/results/invocations/539def09-0647-4508-960a-f22b76e76dae/targets/%2F%2Ftest%2Fcore%2Fend2end:retry_cancel_after_first_attempt_starts_test@poller%3Depoll1/log. This PR alters the lifetime of the grpc_tcp_server object and its constituent members, letting the EventEngine listener manage it when EE listeners are enabled. Previously, some member objects were destroyed before the tcp server itself was destroyed.
Notes:
* This is best reviewed once with whitespace diff turned off. Blocks was added to manage mutex lifetimes.
* BUILD file changes are result of `fix_build_deps.py :iomgr`
Closes#35990
<!--
If you know who should review your pull request, please assign it to that
person, otherwise the pull request would get assigned randomly.
If your pull request is for a specific language, please add the appropriate
lang label.
-->
Closes#35970
As per gRFC A78 (https://github.com/grpc/proposal/pull/419).
Note that these new metrics are populated only if the `wrr_delegate_to_pick_first` experiment is enabled, which is the case by default.
Closes#35977
, since I did not have permissions to add commits to it. That PR has been verified to work (see the top-level description). This PR just makes the gRPC tests pass (e.g. adding includes, clang formatting).
Closes#34874
)"
This reverts commit 310770d61d.
It breaks all portability tests.
```
[==========] Running 1 test from 1 test suite.
[----------] Global test environment set-up.
[----------] 1 test from CrlUtils
[ RUN ] CrlUtils.HasCrlSignBitExists
/[var/local/git/grpc/test/core/tsi/ssl_transport_security_utils_test.cc:566](https://cs.corp.google.com/piper///depot/google3/var/local/git/grpc/test/core/tsi/ssl_transport_security_utils_test.cc?l=566): Failure
Value of: HasCrlSignBit(root_ca_)
Actual: false
Expected: true
[ FAILED ] CrlUtils.HasCrlSignBitExists (0 ms)
[----------] 1 test from CrlUtils (0 ms total)
[----------] Global test environment tear-down
[==========] 1 test from 1 test suite ran. (0 ms total)
[ PASSED ] 0 tests.
[ FAILED ] 1 test, listed below:
[ FAILED ] CrlUtils.HasCrlSignBitExists
```
Closes#35962
This breaks the following pieces out of the `grpc_client_channel` BUILD target:
- backend_metric_parser
- oob_backend_metric
- child_policy_handler
- backup_poller
- service_config_channel_arg_filter
- client_channel_channelz
- client_channel_internal_header
- subchannel_connector
- subchannel_pool_interface
- config_selector
- client_channel_service_config_parser
- retry_service_config_parser
- retry_throttle
The code left in the `grpc_client_channel` target will need more work to pull apart.
Closes#35879
This breaks the following pieces out of the `grpc_base` BUILD target:
- channelz
- call_trace
- dynamic_annotations
- call_combiner
- resource_quota_api
- iomgr
More work is still needed to pull apart the remaining parts of `grpc_base`.
Closes#35846
This PR does 2 distinct things, I can unbundle them if desired
1) Add functions in `ssl_transport_security_utils` and associated tests that will eventually be used for additional Crl validation (the logic of actually doing this will be in a future PR), so other than the tests these fns are currently unused.
2) Remove the use of `X509_NAME_oneline` - it is not a guaranteed stable way to get the issuer name for lookups. Instead, use the DER encoding via `i2d_X509_NAME` - the results in a non-human readable string that is stable for lookup, and necessitated some change to the CrlProvider test code that previously used a human readable string for this value.
Neither should result in behavior changes.
Closes#35641
gRPC has been having the `wrap_memcpy` workaround to handle the breaking change of `memcpy` on glibc 2.14 ([ref1](https://www.win.tue.nl/~aeb/linux/misc/gcc-semibug.html), [ref2](https://stackoverflow.com/questions/35656696/explanation-of-memcpy-memmove-glibc-2-14-2-2-5)). This was necessary to build more portable artifacts which are expected to run on most linux distributions and we manged to have https://github.com/grpc/grpc/pull/5007 years ago for this.
Since we started to use manylinux2010-based docker images for artifacts, however, this became unnecessary since CentOS 6 has glibc 2.12 which is older than 2.14 so it doesn't have memcpy issue, which is another benefit of using manylinux2010.
Superseding https://github.com/grpc/grpc/pull/23385
Changes after the original PR was made; Ruby is using manylinux images (manylinux2014) enabling this change, no more old docker images in our test env.
Closes#35826
This new directory combines code from the following locations:
- src/core/ext/filters/client_channel/lb_policy
- src/core/lib/load_balancing
Closes#35786
This PR:
* adds FD extensions to the public headers
* Adds the query extension interface to EventEngine, Listener, and Endpoint, via a new `Extensible` interface
* Refactors the PosixEventEngine to use the Extensible interface.
Closes#35648
A call execution environment for the V3 runtime.
The `CallFilters` class will ultimately be a (private) member of `CallSpine`, and the `StackBuilder` component will be used by a channel when all of the filters it needs are known to allow the call spine to start processing a call.
This is accompanied by a reasonably extensive test suite.
I expect to fine tune semantics, implementation, and tests over the coming weeks/months as we iterate to bring up the rest of the pieces.
Closes#35533
We probably want to reconsider this types role in the system, but for now removing it would be a large job. Move it to somewhere isolated for now.
Closes#35504
(Required recursively splitting some other targets too)
Gets us closer to eliminating `grpc_base`, and (more importantly) sets up some pieces of transport to be split out so I can rebuild them in coming months.
Closes#35496
AJ Heller
Mark D. Roth
Craig Tiller
Yijie Ma
Ming-Chuan
Tanvi Jagtap
Alisha Nanda
Gregory Cooke
Matthew Stevenson
Esun Kim
Eugene Ostroukhov
Stanley Cheung
ericsalo
Yash Tibrewal