* Precondition ChannelArgs with EventEngines
If an EventEngine is not explicitly provided to ChannelArgs, the default
EventEngine will be set when ChannelArgs are preconditioned.
* channel_idle_filter: EE from channel_args
* grpclb: EE from channel_args
* weighted_target: ee from channel_args
* sanitize
* xds cluster manager
* posix native resolver: own an EE ref from iomgr initialization
* reviewer feedback
* reviewer feedback
* iwyu
* iwyu
* change ownership and remove unneeded methods
* clang_format and use consistent engine naming
* store EE ref in channel_stack and use it in channel idle filter
* don't store a separate shared_ptr in NativeDNSResolver
* add GetEventEngine() method to LB policy helper interface
* stop holding refs to the EE instance in LB policies
* clang-format
* change channel stack to get EE instance from channel args
* update XdsWrrLocalityLb
* fix lb_policy_test
* precondition channel_args in ServerBuilder and microbenchmark fixtures
* add required engine to channel_stack test
* sanitize
* dep fix
* add EE to filter fuzzer
* precondition BM_IsolatedFilter channelargs
* fix
* remove unused using statement
* iwyu again??
* remove preconditioning from C++ surface API
* fix bm_call_create
* Automated change: Fix sanity tests
* iwyu
* rm this->
* rm unused deps
* add internal EE arg macro
* precondition filter_fuzzer
* Automated change: Fix sanity tests
* iwyu
* ChannelStackBuilder requires preconditioned ChannelArgs
* iwyu
* iwyu again?
* rm build.SetChannelArgs; rm unused declaration
* fix nullptr string creation
Co-authored-by: Mark D. Roth <roth@google.com>
Co-authored-by: drfloob <drfloob@users.noreply.github.com>
Additional changes:
- Use absl::variant<> to differentiate between API listeners and TCP listeners.
- Use absl::variant<> to differentiate between an RDS reference and an inline RouteConfig.
- Use common xDS extension code for transport socket extensions in both Listener and Cluster resources.
- Improve xDS HTTP filter registration mechanism.
- Use ValidationErrors for HTTP filter config validation.
* ThreadPool benchmarks
These are nearly identical to the EventEngine benchmarks at the moment. We can consider removing the redundant tests from the EventEngine code and focusing on EventEngine-specific things (e.g., timer cancellation)
* rm unused header
* rm leak
* fix: moved dependencies
* Refactor ThreadManager to leverage a shared ThreadPool
This is a step towards a work-stealing thread pool implementation:
unifying thread pools so that work stealing affects timer callbacks as
well.
A subsequent step will expose the timeout logic so that the timer wakeup
and check can be triggered externally (by pollers, in the common case).
* fix atomic uint64_t type missing on some platforms
* sanitize + platform fixes
* ->quiesce
* shut down the timer manager to release the main thread
* roll back atomics
* use a dedicated thread for timer_manager to prevent local execution (work stealing)
* drain the pools after timer manager tests; sanitize
* iwyu
* reintroduce fork handling
* sanitize
* fix
* remove remains of grpc_csharp_ext artifact build completely
* add back part of C# distribtests
* redirect C# linux distribtests
* comment out some distribtests
* switch updateversion to Grpc.tools
* fix .proto
* fix distribtest complile
* temporarily comment out service in .proto
* cleanup
* improvements
* reenable mac distribtests
* fixes
* fix: grpcio fails to build from source on Apple Silicon mac #(25082)
Boringssl currently does not support macOS arm64, so installation on these platform with default config will undoubtedly fail, that means the env var 'BUILD_WITH_SYSTEM_OPENSSL' must be set to true.
* Use system installation of openssl by default on macOS arm64
* Format code by yapf
AJ Heller
Cheng-Yu Chung
Craig Tiller
Mark D. Roth
Vignesh Babu
Hannah Shi
apolcyn
tony
Mikhail Lappo
Micah Paul Ramos
Yash Tibrewal
Eric Anderson
Alisha Nanda
sanjaypujare
Jan Tattermusch
Amchii
Sergii Tkachenko
nickylba