grpc 1.67.0 → 1.72.0

data/src/core/lib/experiments/experiments.cc

@@ -17,36 +17,76 @@
 #include "src/core/lib/experiments/experiments.h"
 
 #include <grpc/support/port_platform.h>
+#include <stdint.h>
 
 #ifndef GRPC_EXPERIMENTS_ARE_FINAL
 
 #if defined(GRPC_CFSTREAM)
 namespace {
+const char* const description_backoff_cap_initial_at_max =
+    "Backoff library applies max_backoff even on initial_backoff.";
+const char* const additional_constraints_backoff_cap_initial_at_max = "{}";
 const char* const description_call_tracer_in_transport =
     "Transport directly passes byte counts to CallTracer.";
 const char* const additional_constraints_call_tracer_in_transport = "{}";
-const char* const description_canary_client_privacy =
-    "If set, canary client privacy";
-const char* const additional_constraints_canary_client_privacy = "{}";
-const char* const description_client_privacy = "If set, client privacy";
-const char* const additional_constraints_client_privacy = "{}";
-const char* const description_event_engine_application_callbacks =
-    "Run application callbacks in EventEngine threads, instead of on the "
-    "thread-local ApplicationCallbackExecCtx";
-const char* const additional_constraints_event_engine_application_callbacks =
-    "{}";
+const char* const description_call_tracer_transport_fix =
+    "Use the correct call tracer in transport";
+const char* const additional_constraints_call_tracer_transport_fix = "{}";
+const char* const description_callv3_client_auth_filter =
+    "Use the CallV3 client auth filter.";
+const char* const additional_constraints_callv3_client_auth_filter = "{}";
+const char* const description_chaotic_good_framing_layer =
+    "Enable the chaotic good framing layer.";
+const char* const additional_constraints_chaotic_good_framing_layer = "{}";
+const char* const description_error_flatten =
+    "Flatten errors to ordinary absl::Status form.";
+const char* const additional_constraints_error_flatten = "{}";
 const char* const description_event_engine_client =
     "Use EventEngine clients instead of iomgr's grpc_tcp_client";
 const char* const additional_constraints_event_engine_client = "{}";
 const char* const description_event_engine_dns =
     "If set, use EventEngine DNSResolver for client channel resolution";
 const char* const additional_constraints_event_engine_dns = "{}";
+const char* const description_event_engine_dns_non_client_channel =
+    "If set, use EventEngine DNSResolver in other places besides client "
+    "channel.";
+const char* const additional_constraints_event_engine_dns_non_client_channel =
+    "{}";
+const char* const description_event_engine_fork =
+    "Enables event engine fork handling, including onfork events and file "
+    "descriptor generations";
+const char* const additional_constraints_event_engine_fork = "{}";
 const char* const description_event_engine_listener =
     "Use EventEngine listeners instead of iomgr's grpc_tcp_server";
 const char* const additional_constraints_event_engine_listener = "{}";
+const char* const description_event_engine_callback_cq =
+    "Use EventEngine instead of the CallbackAlternativeCQ.";
+const char* const additional_constraints_event_engine_callback_cq = "{}";
+const uint8_t required_experiments_event_engine_callback_cq[] = {
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineClient),
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineListener)};
+const char* const description_event_engine_for_all_other_endpoints =
+    "Use EventEngine endpoints for all call sites, including direct uses of "
+    "grpc_tcp_create.";
+const char* const additional_constraints_event_engine_for_all_other_endpoints =
+    "{}";
+const uint8_t required_experiments_event_engine_for_all_other_endpoints[] = {
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineClient),
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineDns),
+    static_cast<uint8_t>(
+        grpc_core::kExperimentIdEventEngineDnsNonClientChannel),
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineListener)};
 const char* const description_free_large_allocator =
     "If set, return all free bytes from a \042big\042 allocator";
 const char* const additional_constraints_free_large_allocator = "{}";
+const char* const description_keep_alive_ping_timer_batch =
+    "Avoid explicitly cancelling the keepalive timer. Instead adjust the "
+    "callback to re-schedule itself to the next ping interval.";
+const char* const additional_constraints_keep_alive_ping_timer_batch = "{}";
+const char* const description_local_connector_secure =
+    "Local security connector uses TSI_SECURITY_NONE for LOCAL_TCP "
+    "connections.";
+const char* const additional_constraints_local_connector_secure = "{}";
 const char* const description_max_pings_wo_data_throttle =
     "Experiment to throttle pings to a period of 1 min when "
     "GRPC_ARG_HTTP2_MAX_PINGS_WITHOUT_DATA limit has reached (instead of "
@@ -58,18 +98,54 @@ const char* const additional_constraints_monitoring_experiment = "{}";
 const char* const description_multiping =
     "Allow more than one ping to be in flight at a time by default.";
 const char* const additional_constraints_multiping = "{}";
-const char* const description_pick_first_new =
-    "New pick_first impl with memory reduction.";
-const char* const additional_constraints_pick_first_new = "{}";
+const char* const description_pollset_alternative =
+    "Code outside iomgr that relies directly on pollsets will use non-pollset "
+    "alternatives when enabled.";
+const char* const additional_constraints_pollset_alternative = "{}";
+const uint8_t required_experiments_pollset_alternative[] = {
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineClient),
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineListener)};
+const char* const description_posix_ee_skip_grpc_init =
+    "Prevent the PosixEventEngine from calling grpc_init & grpc_shutdown on "
+    "creation and destruction.";
+const char* const additional_constraints_posix_ee_skip_grpc_init = "{}";
+const char* const description_promise_based_http2_client_transport =
+    "Use promises for the http2 client transport. We have kept client and "
+    "server transport experiments separate to help with smoother roll outs and "
+    "also help with interop testing.";
+const char* const additional_constraints_promise_based_http2_client_transport =
+    "{}";
+const char* const description_promise_based_http2_server_transport =
+    "Use promises for the http2 server transport. We have kept client and "
+    "server transport experiments separate to help with smoother roll outs and "
+    "also help with interop testing.";
+const char* const additional_constraints_promise_based_http2_server_transport =
+    "{}";
 const char* const description_promise_based_inproc_transport =
     "Use promises for the in-process transport.";
 const char* const additional_constraints_promise_based_inproc_transport = "{}";
+const char* const description_retry_in_callv3 = "Support retries with call-v3";
+const char* const additional_constraints_retry_in_callv3 = "{}";
+const char* const description_rq_fast_reject =
+    "Resource quota rejects requests immediately (before allocating the "
+    "request structure) under very high memory pressure.";
+const char* const additional_constraints_rq_fast_reject = "{}";
+const char* const description_rst_stream_fix =
+    "Fix for RST_STREAM - do not send for idle streams "
+    "(https://github.com/grpc/grpc/issues/38758)";
+const char* const additional_constraints_rst_stream_fix = "{}";
 const char* const description_schedule_cancellation_over_write =
     "Allow cancellation op to be scheduled over a write";
 const char* const additional_constraints_schedule_cancellation_over_write =
     "{}";
-const char* const description_server_privacy = "If set, server privacy";
-const char* const additional_constraints_server_privacy = "{}";
+const char* const description_server_global_callbacks_ownership =
+    "If set, server global callbacks ownership is fixed to not be owned by "
+    "gRPC.";
+const char* const additional_constraints_server_global_callbacks_ownership =
+    "{}";
+const char* const description_server_listener =
+    "If set, the new server listener classes are used.";
+const char* const additional_constraints_server_listener = "{}";
 const char* const description_tcp_frame_size_tuning =
     "If set, enables TCP to use RPC size estimation made by higher layers. TCP "
     "would not indicate completion of a read operation until a specified "
@@ -79,117 +155,170 @@ const char* const additional_constraints_tcp_frame_size_tuning = "{}";
 const char* const description_tcp_rcv_lowat =
     "Use SO_RCVLOWAT to avoid wakeups on the read path.";
 const char* const additional_constraints_tcp_rcv_lowat = "{}";
-const char* const description_time_caching_in_party =
-    "Disable time caching in exec_ctx, and enable it only in a single party "
-    "execution.";
-const char* const additional_constraints_time_caching_in_party = "{}";
-const char* const description_trace_record_callops =
-    "Enables tracing of call batch initiation and completion.";
-const char* const additional_constraints_trace_record_callops = "{}";
 const char* const description_unconstrained_max_quota_buffer_size =
     "Discard the cap on the max free pool size for one memory allocator";
 const char* const additional_constraints_unconstrained_max_quota_buffer_size =
     "{}";
-const char* const description_work_serializer_clears_time_cache =
-    "Have the work serializer clear the time cache when it dispatches work.";
-const char* const additional_constraints_work_serializer_clears_time_cache =
-    "{}";
-const char* const description_work_serializer_dispatch =
-    "Have the work serializer dispatch work to event engine for every "
-    "callback, instead of running things inline in the first thread that "
-    "successfully enqueues work.";
-const char* const additional_constraints_work_serializer_dispatch = "{}";
 }  // namespace
 
 namespace grpc_core {
 
 const ExperimentMetadata g_experiment_metadata[] = {
+    {"backoff_cap_initial_at_max", description_backoff_cap_initial_at_max,
+     additional_constraints_backoff_cap_initial_at_max, nullptr, 0, true, true},
     {"call_tracer_in_transport", description_call_tracer_in_transport,
-     additional_constraints_call_tracer_in_transport, nullptr, 0, true, true},
-    {"canary_client_privacy", description_canary_client_privacy,
-     additional_constraints_canary_client_privacy, nullptr, 0, false, false},
-    {"client_privacy", description_client_privacy,
-     additional_constraints_client_privacy, nullptr, 0, false, false},
-    {"event_engine_application_callbacks",
-     description_event_engine_application_callbacks,
-     additional_constraints_event_engine_application_callbacks, nullptr, 0,
-     true, true},
+     additional_constraints_call_tracer_in_transport, nullptr, 0, true, false},
+    {"call_tracer_transport_fix", description_call_tracer_transport_fix,
+     additional_constraints_call_tracer_transport_fix, nullptr, 0, true, true},
+    {"callv3_client_auth_filter", description_callv3_client_auth_filter,
+     additional_constraints_callv3_client_auth_filter, nullptr, 0, false, true},
+    {"chaotic_good_framing_layer", description_chaotic_good_framing_layer,
+     additional_constraints_chaotic_good_framing_layer, nullptr, 0, true, true},
+    {"error_flatten", description_error_flatten,
+     additional_constraints_error_flatten, nullptr, 0, false, false},
     {"event_engine_client", description_event_engine_client,
-     additional_constraints_event_engine_client, nullptr, 0, false, true},
+     additional_constraints_event_engine_client, nullptr, 0, false, false},
     {"event_engine_dns", description_event_engine_dns,
      additional_constraints_event_engine_dns, nullptr, 0, false, false},
+    {"event_engine_dns_non_client_channel",
+     description_event_engine_dns_non_client_channel,
+     additional_constraints_event_engine_dns_non_client_channel, nullptr, 0,
+     false, false},
+    {"event_engine_fork", description_event_engine_fork,
+     additional_constraints_event_engine_fork, nullptr, 0, false, false},
     {"event_engine_listener", description_event_engine_listener,
-     additional_constraints_event_engine_listener, nullptr, 0, false, true},
+     additional_constraints_event_engine_listener, nullptr, 0, false, false},
+    {"event_engine_callback_cq", description_event_engine_callback_cq,
+     additional_constraints_event_engine_callback_cq,
+     required_experiments_event_engine_callback_cq, 2, true, true},
+    {"event_engine_for_all_other_endpoints",
+     description_event_engine_for_all_other_endpoints,
+     additional_constraints_event_engine_for_all_other_endpoints,
+     required_experiments_event_engine_for_all_other_endpoints, 4, true, false},
     {"free_large_allocator", description_free_large_allocator,
      additional_constraints_free_large_allocator, nullptr, 0, false, true},
-    {"max_pings_wo_data_throttle", description_max_pings_wo_data_throttle,
-     additional_constraints_max_pings_wo_data_throttle, nullptr, 0, false,
+    {"keep_alive_ping_timer_batch", description_keep_alive_ping_timer_batch,
+     additional_constraints_keep_alive_ping_timer_batch, nullptr, 0, false,
      true},
+    {"local_connector_secure", description_local_connector_secure,
+     additional_constraints_local_connector_secure, nullptr, 0, false, true},
+    {"max_pings_wo_data_throttle", description_max_pings_wo_data_throttle,
+     additional_constraints_max_pings_wo_data_throttle, nullptr, 0, true, true},
     {"monitoring_experiment", description_monitoring_experiment,
      additional_constraints_monitoring_experiment, nullptr, 0, true, true},
     {"multiping", description_multiping, additional_constraints_multiping,
      nullptr, 0, false, true},
-    {"pick_first_new", description_pick_first_new,
-     additional_constraints_pick_first_new, nullptr, 0, true, true},
+    {"pollset_alternative", description_pollset_alternative,
+     additional_constraints_pollset_alternative,
+     required_experiments_pollset_alternative, 2, false, false},
+    {"posix_ee_skip_grpc_init", description_posix_ee_skip_grpc_init,
+     additional_constraints_posix_ee_skip_grpc_init, nullptr, 0, true, true},
+    {"promise_based_http2_client_transport",
+     description_promise_based_http2_client_transport,
+     additional_constraints_promise_based_http2_client_transport, nullptr, 0,
+     false, true},
+    {"promise_based_http2_server_transport",
+     description_promise_based_http2_server_transport,
+     additional_constraints_promise_based_http2_server_transport, nullptr, 0,
+     false, true},
     {"promise_based_inproc_transport",
      description_promise_based_inproc_transport,
      additional_constraints_promise_based_inproc_transport, nullptr, 0, false,
      false},
+    {"retry_in_callv3", description_retry_in_callv3,
+     additional_constraints_retry_in_callv3, nullptr, 0, false, true},
+    {"rq_fast_reject", description_rq_fast_reject,
+     additional_constraints_rq_fast_reject, nullptr, 0, false, true},
+    {"rst_stream_fix", description_rst_stream_fix,
+     additional_constraints_rst_stream_fix, nullptr, 0, true, true},
     {"schedule_cancellation_over_write",
      description_schedule_cancellation_over_write,
      additional_constraints_schedule_cancellation_over_write, nullptr, 0, false,
      true},
-    {"server_privacy", description_server_privacy,
-     additional_constraints_server_privacy, nullptr, 0, false, false},
+    {"server_global_callbacks_ownership",
+     description_server_global_callbacks_ownership,
+     additional_constraints_server_global_callbacks_ownership, nullptr, 0,
+     false, true},
+    {"server_listener", description_server_listener,
+     additional_constraints_server_listener, nullptr, 0, true, true},
     {"tcp_frame_size_tuning", description_tcp_frame_size_tuning,
      additional_constraints_tcp_frame_size_tuning, nullptr, 0, false, true},
     {"tcp_rcv_lowat", description_tcp_rcv_lowat,
      additional_constraints_tcp_rcv_lowat, nullptr, 0, false, true},
-    {"time_caching_in_party", description_time_caching_in_party,
-     additional_constraints_time_caching_in_party, nullptr, 0, true, true},
-    {"trace_record_callops", description_trace_record_callops,
-     additional_constraints_trace_record_callops, nullptr, 0, true, true},
     {"unconstrained_max_quota_buffer_size",
      description_unconstrained_max_quota_buffer_size,
      additional_constraints_unconstrained_max_quota_buffer_size, nullptr, 0,
      false, true},
-    {"work_serializer_clears_time_cache",
-     description_work_serializer_clears_time_cache,
-     additional_constraints_work_serializer_clears_time_cache, nullptr, 0, true,
-     true},
-    {"work_serializer_dispatch", description_work_serializer_dispatch,
-     additional_constraints_work_serializer_dispatch, nullptr, 0, false, true},
 };
 
 }  // namespace grpc_core
 
 #elif defined(GPR_WINDOWS)
 namespace {
+const char* const description_backoff_cap_initial_at_max =
+    "Backoff library applies max_backoff even on initial_backoff.";
+const char* const additional_constraints_backoff_cap_initial_at_max = "{}";
 const char* const description_call_tracer_in_transport =
     "Transport directly passes byte counts to CallTracer.";
 const char* const additional_constraints_call_tracer_in_transport = "{}";
-const char* const description_canary_client_privacy =
-    "If set, canary client privacy";
-const char* const additional_constraints_canary_client_privacy = "{}";
-const char* const description_client_privacy = "If set, client privacy";
-const char* const additional_constraints_client_privacy = "{}";
-const char* const description_event_engine_application_callbacks =
-    "Run application callbacks in EventEngine threads, instead of on the "
-    "thread-local ApplicationCallbackExecCtx";
-const char* const additional_constraints_event_engine_application_callbacks =
-    "{}";
+const char* const description_call_tracer_transport_fix =
+    "Use the correct call tracer in transport";
+const char* const additional_constraints_call_tracer_transport_fix = "{}";
+const char* const description_callv3_client_auth_filter =
+    "Use the CallV3 client auth filter.";
+const char* const additional_constraints_callv3_client_auth_filter = "{}";
+const char* const description_chaotic_good_framing_layer =
+    "Enable the chaotic good framing layer.";
+const char* const additional_constraints_chaotic_good_framing_layer = "{}";
+const char* const description_error_flatten =
+    "Flatten errors to ordinary absl::Status form.";
+const char* const additional_constraints_error_flatten = "{}";
 const char* const description_event_engine_client =
     "Use EventEngine clients instead of iomgr's grpc_tcp_client";
 const char* const additional_constraints_event_engine_client = "{}";
 const char* const description_event_engine_dns =
     "If set, use EventEngine DNSResolver for client channel resolution";
 const char* const additional_constraints_event_engine_dns = "{}";
+const char* const description_event_engine_dns_non_client_channel =
+    "If set, use EventEngine DNSResolver in other places besides client "
+    "channel.";
+const char* const additional_constraints_event_engine_dns_non_client_channel =
+    "{}";
+const char* const description_event_engine_fork =
+    "Enables event engine fork handling, including onfork events and file "
+    "descriptor generations";
+const char* const additional_constraints_event_engine_fork = "{}";
 const char* const description_event_engine_listener =
     "Use EventEngine listeners instead of iomgr's grpc_tcp_server";
 const char* const additional_constraints_event_engine_listener = "{}";
+const char* const description_event_engine_callback_cq =
+    "Use EventEngine instead of the CallbackAlternativeCQ.";
+const char* const additional_constraints_event_engine_callback_cq = "{}";
+const uint8_t required_experiments_event_engine_callback_cq[] = {
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineClient),
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineListener)};
+const char* const description_event_engine_for_all_other_endpoints =
+    "Use EventEngine endpoints for all call sites, including direct uses of "
+    "grpc_tcp_create.";
+const char* const additional_constraints_event_engine_for_all_other_endpoints =
+    "{}";
+const uint8_t required_experiments_event_engine_for_all_other_endpoints[] = {
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineClient),
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineDns),
+    static_cast<uint8_t>(
+        grpc_core::kExperimentIdEventEngineDnsNonClientChannel),
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineListener)};
 const char* const description_free_large_allocator =
     "If set, return all free bytes from a \042big\042 allocator";
 const char* const additional_constraints_free_large_allocator = "{}";
+const char* const description_keep_alive_ping_timer_batch =
+    "Avoid explicitly cancelling the keepalive timer. Instead adjust the "
+    "callback to re-schedule itself to the next ping interval.";
+const char* const additional_constraints_keep_alive_ping_timer_batch = "{}";
+const char* const description_local_connector_secure =
+    "Local security connector uses TSI_SECURITY_NONE for LOCAL_TCP "
+    "connections.";
+const char* const additional_constraints_local_connector_secure = "{}";
 const char* const description_max_pings_wo_data_throttle =
     "Experiment to throttle pings to a period of 1 min when "
     "GRPC_ARG_HTTP2_MAX_PINGS_WITHOUT_DATA limit has reached (instead of "
@@ -201,18 +330,54 @@ const char* const additional_constraints_monitoring_experiment = "{}";
 const char* const description_multiping =
     "Allow more than one ping to be in flight at a time by default.";
 const char* const additional_constraints_multiping = "{}";
-const char* const description_pick_first_new =
-    "New pick_first impl with memory reduction.";
-const char* const additional_constraints_pick_first_new = "{}";
+const char* const description_pollset_alternative =
+    "Code outside iomgr that relies directly on pollsets will use non-pollset "
+    "alternatives when enabled.";
+const char* const additional_constraints_pollset_alternative = "{}";
+const uint8_t required_experiments_pollset_alternative[] = {
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineClient),
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineListener)};
+const char* const description_posix_ee_skip_grpc_init =
+    "Prevent the PosixEventEngine from calling grpc_init & grpc_shutdown on "
+    "creation and destruction.";
+const char* const additional_constraints_posix_ee_skip_grpc_init = "{}";
+const char* const description_promise_based_http2_client_transport =
+    "Use promises for the http2 client transport. We have kept client and "
+    "server transport experiments separate to help with smoother roll outs and "
+    "also help with interop testing.";
+const char* const additional_constraints_promise_based_http2_client_transport =
+    "{}";
+const char* const description_promise_based_http2_server_transport =
+    "Use promises for the http2 server transport. We have kept client and "
+    "server transport experiments separate to help with smoother roll outs and "
+    "also help with interop testing.";
+const char* const additional_constraints_promise_based_http2_server_transport =
+    "{}";
 const char* const description_promise_based_inproc_transport =
     "Use promises for the in-process transport.";
 const char* const additional_constraints_promise_based_inproc_transport = "{}";
+const char* const description_retry_in_callv3 = "Support retries with call-v3";
+const char* const additional_constraints_retry_in_callv3 = "{}";
+const char* const description_rq_fast_reject =
+    "Resource quota rejects requests immediately (before allocating the "
+    "request structure) under very high memory pressure.";
+const char* const additional_constraints_rq_fast_reject = "{}";
+const char* const description_rst_stream_fix =
+    "Fix for RST_STREAM - do not send for idle streams "
+    "(https://github.com/grpc/grpc/issues/38758)";
+const char* const additional_constraints_rst_stream_fix = "{}";
 const char* const description_schedule_cancellation_over_write =
     "Allow cancellation op to be scheduled over a write";
 const char* const additional_constraints_schedule_cancellation_over_write =
     "{}";
-const char* const description_server_privacy = "If set, server privacy";
-const char* const additional_constraints_server_privacy = "{}";
+const char* const description_server_global_callbacks_ownership =
+    "If set, server global callbacks ownership is fixed to not be owned by "
+    "gRPC.";
+const char* const additional_constraints_server_global_callbacks_ownership =
+    "{}";
+const char* const description_server_listener =
+    "If set, the new server listener classes are used.";
+const char* const additional_constraints_server_listener = "{}";
 const char* const description_tcp_frame_size_tuning =
     "If set, enables TCP to use RPC size estimation made by higher layers. TCP "
     "would not indicate completion of a read operation until a specified "
@@ -222,117 +387,170 @@ const char* const additional_constraints_tcp_frame_size_tuning = "{}";
 const char* const description_tcp_rcv_lowat =
     "Use SO_RCVLOWAT to avoid wakeups on the read path.";
 const char* const additional_constraints_tcp_rcv_lowat = "{}";
-const char* const description_time_caching_in_party =
-    "Disable time caching in exec_ctx, and enable it only in a single party "
-    "execution.";
-const char* const additional_constraints_time_caching_in_party = "{}";
-const char* const description_trace_record_callops =
-    "Enables tracing of call batch initiation and completion.";
-const char* const additional_constraints_trace_record_callops = "{}";
 const char* const description_unconstrained_max_quota_buffer_size =
     "Discard the cap on the max free pool size for one memory allocator";
 const char* const additional_constraints_unconstrained_max_quota_buffer_size =
     "{}";
-const char* const description_work_serializer_clears_time_cache =
-    "Have the work serializer clear the time cache when it dispatches work.";
-const char* const additional_constraints_work_serializer_clears_time_cache =
-    "{}";
-const char* const description_work_serializer_dispatch =
-    "Have the work serializer dispatch work to event engine for every "
-    "callback, instead of running things inline in the first thread that "
-    "successfully enqueues work.";
-const char* const additional_constraints_work_serializer_dispatch = "{}";
 }  // namespace
 
 namespace grpc_core {
 
 const ExperimentMetadata g_experiment_metadata[] = {
+    {"backoff_cap_initial_at_max", description_backoff_cap_initial_at_max,
+     additional_constraints_backoff_cap_initial_at_max, nullptr, 0, true, true},
     {"call_tracer_in_transport", description_call_tracer_in_transport,
-     additional_constraints_call_tracer_in_transport, nullptr, 0, true, true},
-    {"canary_client_privacy", description_canary_client_privacy,
-     additional_constraints_canary_client_privacy, nullptr, 0, false, false},
-    {"client_privacy", description_client_privacy,
-     additional_constraints_client_privacy, nullptr, 0, false, false},
-    {"event_engine_application_callbacks",
-     description_event_engine_application_callbacks,
-     additional_constraints_event_engine_application_callbacks, nullptr, 0,
-     true, true},
+     additional_constraints_call_tracer_in_transport, nullptr, 0, true, false},
+    {"call_tracer_transport_fix", description_call_tracer_transport_fix,
+     additional_constraints_call_tracer_transport_fix, nullptr, 0, true, true},
+    {"callv3_client_auth_filter", description_callv3_client_auth_filter,
+     additional_constraints_callv3_client_auth_filter, nullptr, 0, false, true},
+    {"chaotic_good_framing_layer", description_chaotic_good_framing_layer,
+     additional_constraints_chaotic_good_framing_layer, nullptr, 0, true, true},
+    {"error_flatten", description_error_flatten,
+     additional_constraints_error_flatten, nullptr, 0, false, false},
     {"event_engine_client", description_event_engine_client,
-     additional_constraints_event_engine_client, nullptr, 0, true, true},
+     additional_constraints_event_engine_client, nullptr, 0, true, false},
     {"event_engine_dns", description_event_engine_dns,
      additional_constraints_event_engine_dns, nullptr, 0, true, false},
+    {"event_engine_dns_non_client_channel",
+     description_event_engine_dns_non_client_channel,
+     additional_constraints_event_engine_dns_non_client_channel, nullptr, 0,
+     false, false},
+    {"event_engine_fork", description_event_engine_fork,
+     additional_constraints_event_engine_fork, nullptr, 0, false, false},
     {"event_engine_listener", description_event_engine_listener,
-     additional_constraints_event_engine_listener, nullptr, 0, true, true},
+     additional_constraints_event_engine_listener, nullptr, 0, true, false},
+    {"event_engine_callback_cq", description_event_engine_callback_cq,
+     additional_constraints_event_engine_callback_cq,
+     required_experiments_event_engine_callback_cq, 2, true, true},
+    {"event_engine_for_all_other_endpoints",
+     description_event_engine_for_all_other_endpoints,
+     additional_constraints_event_engine_for_all_other_endpoints,
+     required_experiments_event_engine_for_all_other_endpoints, 4, true, false},
     {"free_large_allocator", description_free_large_allocator,
      additional_constraints_free_large_allocator, nullptr, 0, false, true},
-    {"max_pings_wo_data_throttle", description_max_pings_wo_data_throttle,
-     additional_constraints_max_pings_wo_data_throttle, nullptr, 0, false,
+    {"keep_alive_ping_timer_batch", description_keep_alive_ping_timer_batch,
+     additional_constraints_keep_alive_ping_timer_batch, nullptr, 0, false,
      true},
+    {"local_connector_secure", description_local_connector_secure,
+     additional_constraints_local_connector_secure, nullptr, 0, false, true},
+    {"max_pings_wo_data_throttle", description_max_pings_wo_data_throttle,
+     additional_constraints_max_pings_wo_data_throttle, nullptr, 0, true, true},
     {"monitoring_experiment", description_monitoring_experiment,
      additional_constraints_monitoring_experiment, nullptr, 0, true, true},
     {"multiping", description_multiping, additional_constraints_multiping,
      nullptr, 0, false, true},
-    {"pick_first_new", description_pick_first_new,
-     additional_constraints_pick_first_new, nullptr, 0, true, true},
+    {"pollset_alternative", description_pollset_alternative,
+     additional_constraints_pollset_alternative,
+     required_experiments_pollset_alternative, 2, false, false},
+    {"posix_ee_skip_grpc_init", description_posix_ee_skip_grpc_init,
+     additional_constraints_posix_ee_skip_grpc_init, nullptr, 0, true, true},
+    {"promise_based_http2_client_transport",
+     description_promise_based_http2_client_transport,
+     additional_constraints_promise_based_http2_client_transport, nullptr, 0,
+     false, true},
+    {"promise_based_http2_server_transport",
+     description_promise_based_http2_server_transport,
+     additional_constraints_promise_based_http2_server_transport, nullptr, 0,
+     false, true},
     {"promise_based_inproc_transport",
      description_promise_based_inproc_transport,
      additional_constraints_promise_based_inproc_transport, nullptr, 0, false,
      false},
+    {"retry_in_callv3", description_retry_in_callv3,
+     additional_constraints_retry_in_callv3, nullptr, 0, false, true},
+    {"rq_fast_reject", description_rq_fast_reject,
+     additional_constraints_rq_fast_reject, nullptr, 0, false, true},
+    {"rst_stream_fix", description_rst_stream_fix,
+     additional_constraints_rst_stream_fix, nullptr, 0, true, true},
     {"schedule_cancellation_over_write",
      description_schedule_cancellation_over_write,
      additional_constraints_schedule_cancellation_over_write, nullptr, 0, false,
      true},
-    {"server_privacy", description_server_privacy,
-     additional_constraints_server_privacy, nullptr, 0, false, false},
+    {"server_global_callbacks_ownership",
+     description_server_global_callbacks_ownership,
+     additional_constraints_server_global_callbacks_ownership, nullptr, 0,
+     false, true},
+    {"server_listener", description_server_listener,
+     additional_constraints_server_listener, nullptr, 0, true, true},
     {"tcp_frame_size_tuning", description_tcp_frame_size_tuning,
      additional_constraints_tcp_frame_size_tuning, nullptr, 0, false, true},
     {"tcp_rcv_lowat", description_tcp_rcv_lowat,
      additional_constraints_tcp_rcv_lowat, nullptr, 0, false, true},
-    {"time_caching_in_party", description_time_caching_in_party,
-     additional_constraints_time_caching_in_party, nullptr, 0, true, true},
-    {"trace_record_callops", description_trace_record_callops,
-     additional_constraints_trace_record_callops, nullptr, 0, true, true},
     {"unconstrained_max_quota_buffer_size",
      description_unconstrained_max_quota_buffer_size,
      additional_constraints_unconstrained_max_quota_buffer_size, nullptr, 0,
      false, true},
-    {"work_serializer_clears_time_cache",
-     description_work_serializer_clears_time_cache,
-     additional_constraints_work_serializer_clears_time_cache, nullptr, 0, true,
-     true},
-    {"work_serializer_dispatch", description_work_serializer_dispatch,
-     additional_constraints_work_serializer_dispatch, nullptr, 0, false, true},
 };
 
 }  // namespace grpc_core
 
 #else
 namespace {
+const char* const description_backoff_cap_initial_at_max =
+    "Backoff library applies max_backoff even on initial_backoff.";
+const char* const additional_constraints_backoff_cap_initial_at_max = "{}";
 const char* const description_call_tracer_in_transport =
     "Transport directly passes byte counts to CallTracer.";
 const char* const additional_constraints_call_tracer_in_transport = "{}";
-const char* const description_canary_client_privacy =
-    "If set, canary client privacy";
-const char* const additional_constraints_canary_client_privacy = "{}";
-const char* const description_client_privacy = "If set, client privacy";
-const char* const additional_constraints_client_privacy = "{}";
-const char* const description_event_engine_application_callbacks =
-    "Run application callbacks in EventEngine threads, instead of on the "
-    "thread-local ApplicationCallbackExecCtx";
-const char* const additional_constraints_event_engine_application_callbacks =
-    "{}";
+const char* const description_call_tracer_transport_fix =
+    "Use the correct call tracer in transport";
+const char* const additional_constraints_call_tracer_transport_fix = "{}";
+const char* const description_callv3_client_auth_filter =
+    "Use the CallV3 client auth filter.";
+const char* const additional_constraints_callv3_client_auth_filter = "{}";
+const char* const description_chaotic_good_framing_layer =
+    "Enable the chaotic good framing layer.";
+const char* const additional_constraints_chaotic_good_framing_layer = "{}";
+const char* const description_error_flatten =
+    "Flatten errors to ordinary absl::Status form.";
+const char* const additional_constraints_error_flatten = "{}";
 const char* const description_event_engine_client =
     "Use EventEngine clients instead of iomgr's grpc_tcp_client";
 const char* const additional_constraints_event_engine_client = "{}";
 const char* const description_event_engine_dns =
     "If set, use EventEngine DNSResolver for client channel resolution";
 const char* const additional_constraints_event_engine_dns = "{}";
+const char* const description_event_engine_dns_non_client_channel =
+    "If set, use EventEngine DNSResolver in other places besides client "
+    "channel.";
+const char* const additional_constraints_event_engine_dns_non_client_channel =
+    "{}";
+const char* const description_event_engine_fork =
+    "Enables event engine fork handling, including onfork events and file "
+    "descriptor generations";
+const char* const additional_constraints_event_engine_fork = "{}";
 const char* const description_event_engine_listener =
     "Use EventEngine listeners instead of iomgr's grpc_tcp_server";
 const char* const additional_constraints_event_engine_listener = "{}";
+const char* const description_event_engine_callback_cq =
+    "Use EventEngine instead of the CallbackAlternativeCQ.";
+const char* const additional_constraints_event_engine_callback_cq = "{}";
+const uint8_t required_experiments_event_engine_callback_cq[] = {
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineClient),
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineListener)};
+const char* const description_event_engine_for_all_other_endpoints =
+    "Use EventEngine endpoints for all call sites, including direct uses of "
+    "grpc_tcp_create.";
+const char* const additional_constraints_event_engine_for_all_other_endpoints =
+    "{}";
+const uint8_t required_experiments_event_engine_for_all_other_endpoints[] = {
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineClient),
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineDns),
+    static_cast<uint8_t>(
+        grpc_core::kExperimentIdEventEngineDnsNonClientChannel),
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineListener)};
 const char* const description_free_large_allocator =
     "If set, return all free bytes from a \042big\042 allocator";
 const char* const additional_constraints_free_large_allocator = "{}";
+const char* const description_keep_alive_ping_timer_batch =
+    "Avoid explicitly cancelling the keepalive timer. Instead adjust the "
+    "callback to re-schedule itself to the next ping interval.";
+const char* const additional_constraints_keep_alive_ping_timer_batch = "{}";
+const char* const description_local_connector_secure =
+    "Local security connector uses TSI_SECURITY_NONE for LOCAL_TCP "
+    "connections.";
+const char* const additional_constraints_local_connector_secure = "{}";
 const char* const description_max_pings_wo_data_throttle =
     "Experiment to throttle pings to a period of 1 min when "
     "GRPC_ARG_HTTP2_MAX_PINGS_WITHOUT_DATA limit has reached (instead of "
@@ -344,18 +562,54 @@ const char* const additional_constraints_monitoring_experiment = "{}";
 const char* const description_multiping =
     "Allow more than one ping to be in flight at a time by default.";
 const char* const additional_constraints_multiping = "{}";
-const char* const description_pick_first_new =
-    "New pick_first impl with memory reduction.";
-const char* const additional_constraints_pick_first_new = "{}";
+const char* const description_pollset_alternative =
+    "Code outside iomgr that relies directly on pollsets will use non-pollset "
+    "alternatives when enabled.";
+const char* const additional_constraints_pollset_alternative = "{}";
+const uint8_t required_experiments_pollset_alternative[] = {
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineClient),
+    static_cast<uint8_t>(grpc_core::kExperimentIdEventEngineListener)};
+const char* const description_posix_ee_skip_grpc_init =
+    "Prevent the PosixEventEngine from calling grpc_init & grpc_shutdown on "
+    "creation and destruction.";
+const char* const additional_constraints_posix_ee_skip_grpc_init = "{}";
+const char* const description_promise_based_http2_client_transport =
+    "Use promises for the http2 client transport. We have kept client and "
+    "server transport experiments separate to help with smoother roll outs and "
+    "also help with interop testing.";
+const char* const additional_constraints_promise_based_http2_client_transport =
+    "{}";
+const char* const description_promise_based_http2_server_transport =
+    "Use promises for the http2 server transport. We have kept client and "
+    "server transport experiments separate to help with smoother roll outs and "
+    "also help with interop testing.";
+const char* const additional_constraints_promise_based_http2_server_transport =
+    "{}";
 const char* const description_promise_based_inproc_transport =
     "Use promises for the in-process transport.";
 const char* const additional_constraints_promise_based_inproc_transport = "{}";
+const char* const description_retry_in_callv3 = "Support retries with call-v3";
+const char* const additional_constraints_retry_in_callv3 = "{}";
+const char* const description_rq_fast_reject =
+    "Resource quota rejects requests immediately (before allocating the "
+    "request structure) under very high memory pressure.";
+const char* const additional_constraints_rq_fast_reject = "{}";
+const char* const description_rst_stream_fix =
+    "Fix for RST_STREAM - do not send for idle streams "
+    "(https://github.com/grpc/grpc/issues/38758)";
+const char* const additional_constraints_rst_stream_fix = "{}";
 const char* const description_schedule_cancellation_over_write =
     "Allow cancellation op to be scheduled over a write";
 const char* const additional_constraints_schedule_cancellation_over_write =
     "{}";
-const char* const description_server_privacy = "If set, server privacy";
-const char* const additional_constraints_server_privacy = "{}";
+const char* const description_server_global_callbacks_ownership =
+    "If set, server global callbacks ownership is fixed to not be owned by "
+    "gRPC.";
+const char* const additional_constraints_server_global_callbacks_ownership =
+    "{}";
+const char* const description_server_listener =
+    "If set, the new server listener classes are used.";
+const char* const additional_constraints_server_listener = "{}";
 const char* const description_tcp_frame_size_tuning =
     "If set, enables TCP to use RPC size estimation made by higher layers. TCP "
     "would not indicate completion of a read operation until a specified "
@@ -365,86 +619,100 @@ const char* const additional_constraints_tcp_frame_size_tuning = "{}";
 const char* const description_tcp_rcv_lowat =
     "Use SO_RCVLOWAT to avoid wakeups on the read path.";
 const char* const additional_constraints_tcp_rcv_lowat = "{}";
-const char* const description_time_caching_in_party =
-    "Disable time caching in exec_ctx, and enable it only in a single party "
-    "execution.";
-const char* const additional_constraints_time_caching_in_party = "{}";
-const char* const description_trace_record_callops =
-    "Enables tracing of call batch initiation and completion.";
-const char* const additional_constraints_trace_record_callops = "{}";
 const char* const description_unconstrained_max_quota_buffer_size =
     "Discard the cap on the max free pool size for one memory allocator";
 const char* const additional_constraints_unconstrained_max_quota_buffer_size =
     "{}";
-const char* const description_work_serializer_clears_time_cache =
-    "Have the work serializer clear the time cache when it dispatches work.";
-const char* const additional_constraints_work_serializer_clears_time_cache =
-    "{}";
-const char* const description_work_serializer_dispatch =
-    "Have the work serializer dispatch work to event engine for every "
-    "callback, instead of running things inline in the first thread that "
-    "successfully enqueues work.";
-const char* const additional_constraints_work_serializer_dispatch = "{}";
 }  // namespace
 
 namespace grpc_core {
 
 const ExperimentMetadata g_experiment_metadata[] = {
+    {"backoff_cap_initial_at_max", description_backoff_cap_initial_at_max,
+     additional_constraints_backoff_cap_initial_at_max, nullptr, 0, true, true},
     {"call_tracer_in_transport", description_call_tracer_in_transport,
-     additional_constraints_call_tracer_in_transport, nullptr, 0, true, true},
-    {"canary_client_privacy", description_canary_client_privacy,
-     additional_constraints_canary_client_privacy, nullptr, 0, false, false},
-    {"client_privacy", description_client_privacy,
-     additional_constraints_client_privacy, nullptr, 0, false, false},
-    {"event_engine_application_callbacks",
-     description_event_engine_application_callbacks,
-     additional_constraints_event_engine_application_callbacks, nullptr, 0,
-     true, true},
+     additional_constraints_call_tracer_in_transport, nullptr, 0, true, false},
+    {"call_tracer_transport_fix", description_call_tracer_transport_fix,
+     additional_constraints_call_tracer_transport_fix, nullptr, 0, true, true},
+    {"callv3_client_auth_filter", description_callv3_client_auth_filter,
+     additional_constraints_callv3_client_auth_filter, nullptr, 0, false, true},
+    {"chaotic_good_framing_layer", description_chaotic_good_framing_layer,
+     additional_constraints_chaotic_good_framing_layer, nullptr, 0, true, true},
+    {"error_flatten", description_error_flatten,
+     additional_constraints_error_flatten, nullptr, 0, false, false},
     {"event_engine_client", description_event_engine_client,
-     additional_constraints_event_engine_client, nullptr, 0, false, true},
+     additional_constraints_event_engine_client, nullptr, 0, true, false},
     {"event_engine_dns", description_event_engine_dns,
      additional_constraints_event_engine_dns, nullptr, 0, true, false},
+    {"event_engine_dns_non_client_channel",
+     description_event_engine_dns_non_client_channel,
+     additional_constraints_event_engine_dns_non_client_channel, nullptr, 0,
+     false, false},
+    {"event_engine_fork", description_event_engine_fork,
+     additional_constraints_event_engine_fork, nullptr, 0, false, false},
     {"event_engine_listener", description_event_engine_listener,
-     additional_constraints_event_engine_listener, nullptr, 0, true, true},
+     additional_constraints_event_engine_listener, nullptr, 0, true, false},
+    {"event_engine_callback_cq", description_event_engine_callback_cq,
+     additional_constraints_event_engine_callback_cq,
+     required_experiments_event_engine_callback_cq, 2, true, true},
+    {"event_engine_for_all_other_endpoints",
+     description_event_engine_for_all_other_endpoints,
+     additional_constraints_event_engine_for_all_other_endpoints,
+     required_experiments_event_engine_for_all_other_endpoints, 4, true, false},
     {"free_large_allocator", description_free_large_allocator,
      additional_constraints_free_large_allocator, nullptr, 0, false, true},
-    {"max_pings_wo_data_throttle", description_max_pings_wo_data_throttle,
-     additional_constraints_max_pings_wo_data_throttle, nullptr, 0, false,
+    {"keep_alive_ping_timer_batch", description_keep_alive_ping_timer_batch,
+     additional_constraints_keep_alive_ping_timer_batch, nullptr, 0, false,
      true},
+    {"local_connector_secure", description_local_connector_secure,
+     additional_constraints_local_connector_secure, nullptr, 0, false, true},
+    {"max_pings_wo_data_throttle", description_max_pings_wo_data_throttle,
+     additional_constraints_max_pings_wo_data_throttle, nullptr, 0, true, true},
     {"monitoring_experiment", description_monitoring_experiment,
      additional_constraints_monitoring_experiment, nullptr, 0, true, true},
     {"multiping", description_multiping, additional_constraints_multiping,
      nullptr, 0, false, true},
-    {"pick_first_new", description_pick_first_new,
-     additional_constraints_pick_first_new, nullptr, 0, true, true},
+    {"pollset_alternative", description_pollset_alternative,
+     additional_constraints_pollset_alternative,
+     required_experiments_pollset_alternative, 2, false, false},
+    {"posix_ee_skip_grpc_init", description_posix_ee_skip_grpc_init,
+     additional_constraints_posix_ee_skip_grpc_init, nullptr, 0, true, true},
+    {"promise_based_http2_client_transport",
+     description_promise_based_http2_client_transport,
+     additional_constraints_promise_based_http2_client_transport, nullptr, 0,
+     false, true},
+    {"promise_based_http2_server_transport",
+     description_promise_based_http2_server_transport,
+     additional_constraints_promise_based_http2_server_transport, nullptr, 0,
+     false, true},
     {"promise_based_inproc_transport",
      description_promise_based_inproc_transport,
      additional_constraints_promise_based_inproc_transport, nullptr, 0, false,
      false},
+    {"retry_in_callv3", description_retry_in_callv3,
+     additional_constraints_retry_in_callv3, nullptr, 0, false, true},
+    {"rq_fast_reject", description_rq_fast_reject,
+     additional_constraints_rq_fast_reject, nullptr, 0, false, true},
+    {"rst_stream_fix", description_rst_stream_fix,
+     additional_constraints_rst_stream_fix, nullptr, 0, true, true},
     {"schedule_cancellation_over_write",
      description_schedule_cancellation_over_write,
      additional_constraints_schedule_cancellation_over_write, nullptr, 0, false,
      true},
-    {"server_privacy", description_server_privacy,
-     additional_constraints_server_privacy, nullptr, 0, false, false},
+    {"server_global_callbacks_ownership",
+     description_server_global_callbacks_ownership,
+     additional_constraints_server_global_callbacks_ownership, nullptr, 0,
+     false, true},
+    {"server_listener", description_server_listener,
+     additional_constraints_server_listener, nullptr, 0, true, true},
     {"tcp_frame_size_tuning", description_tcp_frame_size_tuning,
      additional_constraints_tcp_frame_size_tuning, nullptr, 0, false, true},
     {"tcp_rcv_lowat", description_tcp_rcv_lowat,
      additional_constraints_tcp_rcv_lowat, nullptr, 0, false, true},
-    {"time_caching_in_party", description_time_caching_in_party,
-     additional_constraints_time_caching_in_party, nullptr, 0, true, true},
-    {"trace_record_callops", description_trace_record_callops,
-     additional_constraints_trace_record_callops, nullptr, 0, true, true},
     {"unconstrained_max_quota_buffer_size",
      description_unconstrained_max_quota_buffer_size,
      additional_constraints_unconstrained_max_quota_buffer_size, nullptr, 0,
      false, true},
-    {"work_serializer_clears_time_cache",
-     description_work_serializer_clears_time_cache,
-     additional_constraints_work_serializer_clears_time_cache, nullptr, 0, true,
-     true},
-    {"work_serializer_dispatch", description_work_serializer_dispatch,
-     additional_constraints_work_serializer_dispatch, nullptr, 0, false, true},
 };
 
 }  // namespace grpc_core