This reverts commit a7a650a74e.
Original PR breaks #38286 and I think the breakage is in the PR -- in a construct like:
```
TrySeq(
/*a*/ []() -> Promise<absl::Status>,
/*b*/ []() -> Promise<StatusFlag>,
/*c*/ []() -> Promise<absl::Status>)
```
the original seq_state code would try to promote the intermediate `StatusFlag` to an `absl::Status` which is great - however with the optimization, if `b` is instantaneous we try to demote the result of `a` to `StatusFlag`, losing information.
I'll need to consider how to change this optimization to preserve the final output type before rolling forward.
Closes#38748
COPYBARA_INTEGRATE_REVIEW=https://github.com/grpc/grpc/pull/38748 from ctiller:rb1 743eb33bdf
PiperOrigin-RevId: 727109120
Increasingly as we write sequences, we find ourselves writing some asynchronous parts interleaved with some synchronous parts. This PR seeks to provide a basic optimizer for sequences - it detects immediate returning promises (by the lack of a `Poll<T>` return type), and eliminates the state machinery required to evaluate them at compile time.
Closes#38445
COPYBARA_INTEGRATE_REVIEW=https://github.com/grpc/grpc/pull/38445 from ctiller:optseq 81c5025d76
PiperOrigin-RevId: 726641700
<!--
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#37601
COPYBARA_INTEGRATE_REVIEW=https://github.com/grpc/grpc/pull/37601 from yashykt:LogicalConnection 50aacc3dd1
PiperOrigin-RevId: 713079761
There's enough here that it's worth getting this reviewed and checkpointed - although certainly this is far from production ready.
About 50% of the retry e2e tests are failing right now, some justifiably, some less so.
That said, half are passing, the general structure seems sound, and there's some nice new tests.
Closes#37816
COPYBARA_INTEGRATE_REVIEW=https://github.com/grpc/grpc/pull/37816 from ctiller:and-again 9d8bc04e0c
PiperOrigin-RevId: 704309686
[PH2][NewFile][ClassStructure][Important] Add client and server class
1. New classes Http2ServerTransport and Http2ClientTransport
2. Similar to the classes in [Chaotic Good Client Transport](https://github.com/grpc/grpc/blob/master/src/core/ext/transport/chaotic_good/client_transport.h) and [Chaotic Good Server Transport](https://github.com/grpc/grpc/blob/master/src/core/ext/transport/chaotic_good/server_transport.h)
3. Added new Test files. For now, the 2 new tests just call the constructor of Http2ServerTransport and Http2ClientTransport.
Tested locally using
```
CC=cc bazel test --test_output=all -c dbg --config=asan --verbose_failures //test/core/transport/chttp2:http2_client_transport_test
```
```
CC=cc bazel test --test_output=all -c dbg --config=asan --verbose_failures //test/core/transport/chttp2:http2_server_transport_test
```
Closes#37840
COPYBARA_INTEGRATE_REVIEW=https://github.com/grpc/grpc/pull/37840 from tanvi-jagtap:ph2_add_client_server_class c6c3a0d5fb
PiperOrigin-RevId: 692824127
I'm continuing to look into some flakes here, but in the meantime these shouldn't halt submissions. Marking them flaky.
Closes#37880
COPYBARA_INTEGRATE_REVIEW=https://github.com/grpc/grpc/pull/37880 from ctiller:mark-flaky 27427c7978
PiperOrigin-RevId: 684526341
This is the last piece of gRFC A83 (https://github.com/grpc/proposal/pull/438).
Note that although this is the first use-case for this "blackboard" mechanism, we will also use it in the future for the xDS rate-limiting filter on the gRPC server side.
Closes#37646
COPYBARA_INTEGRATE_REVIEW=https://github.com/grpc/grpc/pull/37646 from markdroth:gcp_auth_filter_state 72d0d96c79
PiperOrigin-RevId: 679707134
Basic building block for retries, hedging: buffer outgoing messages & metadata, allow for replay whilst buffered (with a single reader able to read once buffering ends)
Closes#37448
COPYBARA_INTEGRATE_REVIEW=https://github.com/grpc/grpc/pull/37448 from ctiller:once-again-into-the-breach-my-friends 79cb121054
PiperOrigin-RevId: 677959212
The following files have been moved:
- src/core/lib/avl/*
- src/core/lib/backoff/*
- src/core/lib/debug/event_log*
- src/core/lib/iomgr/gethostname*
- src/core/lib/iomgr/grpc_if_nametoindex*
- src/core/lib/matchers/*
- src/core/lib/uri/* (renamed from uri_parser.* to uri.*)
- src/core/lib/gprpp/* (existing src/core/util/time.cc was renamed to gpr_time.cc to avoid conflict)
Closes#36792
COPYBARA_INTEGRATE_REVIEW=https://github.com/grpc/grpc/pull/36792 from markdroth:reorg_util d4e8996f48
PiperOrigin-RevId: 676947640
Final piece of gRFC A83 (https://github.com/grpc/proposal/pull/438): the GCP authentication filter itself.
Infrastructure changes include:
- Added a general-purpose LRU cache library that can be reused elsewhere.
- Fixed the client channel code to use the channel args returned by the resolver for the dynamic filters. This was necessary so that the GCP auth filter could access the `XdsConfig` object, which is passed via a channel arg.
- Unlike the other xDS HTTP filters we support, the GCP auth filter does not support config overrides, and its configuration includes a cache size parameter that we always need at the channel level, not per-call. As a result, I had to change the xDS HTTP filter API to give it the ability to set top-level fields in the service config, not just per-method fields. (We use the service config as a way of passing configuration down into xDS HTTP filters.) Note that for now, this works only on the client side, because we don't have machinery for a top-level service config on the server side.
- The GCP auth filter is also the first case where the filter needs to know its instance name from the xDS config, so I changed the xDS HTTP filter API to plumb that through.
- Fixed a bug in the HTTP client library that prevented the override functions from declining to override a particular request.
Closes#37550
COPYBARA_INTEGRATE_REVIEW=https://github.com/grpc/grpc/pull/37550 from markdroth:xds_gcp_auth_filter 19eaefb52f
PiperOrigin-RevId: 669371249
Add validation of the `Audience` cluster metadata type, as per gRFC A83 (https://github.com/grpc/proposal/pull/438).
I had previously changed the metadata to be represented as JSON in #37468. However, while working on the GCP Authentication filter implementation, I realized that that's not an ideal representation, because it would have required us to validate the JSON on a per-RPC basis, which would be bad for performance. So I've changed the representation of metadata to be an abstract type, and we now store the `Audience` metadata as a simple string. I've also moved metadata into its own type with its own validation code, so that in the future we can use it in places other than CDS (many xDS resource types have metadata fields).
While I was at it, I also add some helper functions for validating the `UInt32Value` and `UInt64Value` wrapper protos.
Closes#37566
PiperOrigin-RevId: 668281729
The microbenchmarks are tools for us, and it doesn't seem like we're buying much by maintaining them for different build systems.
Closes#37391
COPYBARA_INTEGRATE_REVIEW=https://github.com/grpc/grpc/pull/37391 from ctiller:pickthis 24184f2b44
PiperOrigin-RevId: 658861012
Two new benchmarks here-in.
Benchmark 1: `bm_picker`
------
Measures various load balancing policies pick performance. For now we cover `pick_first` and `weighted_round_robin` at 1, 10, 100, 1000, 10000, and 100000 backends.
Today's output:
```
------------------------------------------------------------------------------
Benchmark Time CPU Iterations
------------------------------------------------------------------------------
BM_Pick/pick_first/1 20.4 ns 20.4 ns 68285
BM_Pick/pick_first/10 20.6 ns 20.6 ns 68274
BM_Pick/pick_first/100 20.5 ns 20.5 ns 67817
BM_Pick/pick_first/1000 20.6 ns 20.6 ns 67347
BM_Pick/pick_first/10000 20.7 ns 20.7 ns 67317
BM_Pick/pick_first/100000 20.9 ns 20.9 ns 67385
BM_Pick/weighted_round_robin/1 54.7 ns 54.7 ns 26641
BM_Pick/weighted_round_robin/10 54.2 ns 54.2 ns 25828
BM_Pick/weighted_round_robin/100 55.2 ns 55.2 ns 26210
BM_Pick/weighted_round_robin/1000 54.1 ns 54.1 ns 25678
BM_Pick/weighted_round_robin/10000 77.3 ns 76.6 ns 15776
BM_Pick/weighted_round_robin/100000 148 ns 148 ns 9882
```
Benchmark 2: `bm_load_balanced_call_destination`
-----
This benchmark measures call performance when a call spine passes through a `LoadBalancedCallDestination`, and with `BM_LoadBalancedCallDestination` also the construction/destruction cost of this object.
We do not consider picker performance in this benchmark as it's separately covered by `bm_picker` above.
Today's output:
```
-----------------------------------------------------------------------------------------------------------------------------------------
Benchmark Time CPU Iterations
-----------------------------------------------------------------------------------------------------------------------------------------
BM_UnaryWithSpawnPerEnd<UnstartedCallDestinationFixture<LoadBalancedCallDestinationTraits>> 1255 ns 1255 ns 1076
BM_UnaryWithSpawnPerOp<UnstartedCallDestinationFixture<LoadBalancedCallDestinationTraits>> 1459 ns 1459 ns 939
BM_ClientToServerStreaming<UnstartedCallDestinationFixture<LoadBalancedCallDestinationTraits>> 209 ns 209 ns 6775
BM_LoadBalancedCallDestination 92.8 ns 92.8 ns 15063
```
Notes
------
There's some duplicated code between the benchmarks & tests -- this is ok -- as the tests evolve we'll likely want to add more checks to the fixtures, whereas as the benchmarks evolve we may well want to optimize the fixtures so that performance of the systems under test dominate more. That is, the duplicated code is expected to have different evolutionary tracks.
Closes#37052
COPYBARA_INTEGRATE_REVIEW=https://github.com/grpc/grpc/pull/37052 from ctiller:moar-benchy 30c7072d87
PiperOrigin-RevId: 658181731
- add a benchmark for various metadata creation styles
- add factory functions for status + message - these are 3-10x faster than going via absl::Status
- add a `MakePooledForOverwrite` function to Arena, use it everywhere -- this naming matches `std::make_unique_for_overwrite` in C++20, and avoids some language mandated initialization in `Table` (underlying `MetadataMap<>`) - speeding creation of metadata handles by 30%
For `bm_call_spine` we see before:
```
BM_UnaryWithSpawnPerEnd<CallSpineFixture>_median 745 ns 745 ns
```
and after:
```
BM_UnaryWithSpawnPerEnd<CallSpineFixture>_median 699 ns 699 ns
```
Closes#37111
COPYBARA_INTEGRATE_REVIEW=https://github.com/grpc/grpc/pull/37111 from ctiller:meta-magic-2 100464fc7e
PiperOrigin-RevId: 652900726
As we've learned what configuration is needed for our benchmarks the settings have been growing more and more bespoke for each binary. Try to consolidate that into some useful defaults.
Also ensure we always `linkstatic=1`. `cc_binary` defaults to this, so it's reasonable to assume that's the performance our customers see. It also deeply impacts performance for small microbenchmarks, and so enabling it gives us more apples:apples, and saves chasing things that don't matter.
Closes#37012
COPYBARA_INTEGRATE_REVIEW=https://github.com/grpc/grpc/pull/37012 from ctiller:benchmark-bin b84cab58c6
PiperOrigin-RevId: 645483183
Ensure arena accounting is working, and add a test that a constant call size results in a constant call size estimate.
Closes#36933
COPYBARA_INTEGRATE_REVIEW=https://github.com/grpc/grpc/pull/36933 from ctiller:arena-accounts 116c805633
PiperOrigin-RevId: 644102412
This change brings up the direct channel, and inproc promise based transports.
This work exposed a bug that was very difficult to fix with the current call_filters.cc implementation, so I've substantially revamped that - instead of having a pipe-like object per call element, we now have a big ol' combined state machine for the entire call. It's a touch more code, but substantially easier to reason about individual cases, so I much prefer this form (it's also a slight memory improvement: 12 bytes total to track call state, and 10 of those are wakeup bitmasks...).
Closes#36734
COPYBARA_INTEGRATE_REVIEW=https://github.com/grpc/grpc/pull/36734 from ctiller:transport-refs-9 3e2a80b40d
PiperOrigin-RevId: 644034593
All TraceFlags are now configured in `src/core/lib/debug/trace_flags.yaml`. The format is:
```
my_flag:
default: false # the default value; default=false
description: Some Description
debug_only: false # debug_only flags only work in debug builds; default=false
internal: false # internal flags will not show up in documentation; default=false
```
To regenerate the trace flag source code, run `tools/codegen/core/gen_trace_flags.py` (requires mako). This script is also run when sanity checking.
This PR also adds two new features:
### Glob-based flag configuration
Trace flag configuration now supports `?` (single wildcard character) and `*` (one or more wildcard characters). For example, using `GRPC_TRACE='event_engine*'` will enable all flags that match that glob. It expands to:
* event_engine
* event_engine_client_channel_resolver
* event_engine_dns
* event_engine_endpoint
* event_engine_endpoint_data
* event_engine_poller
### A cleaner trace-logging macro in abseil logging format
If your goal is only to add log statements when the `fault_injection_filter` trace flag is enabled, you can use the macro:
```
GRPC_TRACE_LOG(fault_injection, INFO) << "Filtered:" << 42;
```
When the trace flag is enabled, the the log will show something like this:
```
I0000 00:00:1715733657.430042 16 file.cc:174] Filtered:42
```
----
Note: just like with the gpr_log to abseil logging conversion, the pre-existing trace logging usages can be replaced with the new tracing macro across multiple PRs.
Closes#36576
PiperOrigin-RevId: 641295215
Not the fastest implementation possible, but it's a log helper so I'm not particularly fussed either -- but a useful utility that we can iterate on later to help debugging.
Closes#36554
COPYBARA_INTEGRATE_REVIEW=https://github.com/grpc/grpc/pull/36554 from ctiller:args 33b4802fcc
PiperOrigin-RevId: 632997320
Craig Tiller
Tanvi Jagtap - Google LLC
AJ Heller
Yash Tibrewal
Vignesh Babu
Mark D. Roth
erm-g