0e6ced8800
Revert "Move metadata setting into HPackParser ( #27545 )" ( #27686 )
...
This reverts commit f57a1f7d8c
2021-10-12 08:37:00 -07:00
f57a1f7d8c
Move metadata setting into HPackParser ( #27545 )
...
* forward work to get append, limit check into parser
* further cleanup
* Remove chttp2_incoming_metadata_buffer
* Automated change: Fix sanity tests
* test fixes
* fix cronet
* wip
* first pass mementofication of parsing - libgrpc compiles
* further progress
* fixes
* fixes
* fix leak
* Automated change: Fix sanity tests
* fix leak
* fixes
* x
* improve fuzzer
* init ordering fix
* Update hpack_parser_table.h
* Update hpack_parser.cc
* Update hpack_parser_test.cc
* Automated change: Fix sanity tests
* Automated change: Fix sanity tests
* i mustache myself why
* fix use after free
* Automated change: Fix sanity tests
* fixes
* eliminate second pass metadata - this is an illegal frame and fails the new parser
* handle mementos with size > sizeof(intptr_t)
* fix
* add parsed metadata header, test
* Automated change: Fix sanity tests
* fix includes
* phase1
* phase2
* Update parsed_metadata.h
* fix
Co-authored-by: ctiller <ctiller@users.noreply.github.com>
2021-10-11 20:02:29 -07:00
a145013d6d
Temporarily disable benchmarks ( #27629 )
...
benchmark 1.6.0 introduced API breaking changes so we temporarily disabled the benchmarks that are using the old API to avoid breaking.
2021-10-07 14:29:19 -07:00
d1e01ac139
Add a test for a (now-illegal) build file construct ( #27602 )
...
* Add a test for a (now-illegal) build file construct
* add to test suite!
* fix
* fix
* fix syntax
2021-10-06 07:43:59 -07:00
84d9e7e9fa
Speed up microbenchmark tests ( #27584 )
...
* Speed up microbenchmark tests
* Automated change: Fix sanity tests
* Update grpc_benchmark_config.bzl
Co-authored-by: ctiller <ctiller@users.noreply.github.com>
2021-10-05 07:09:02 -07:00
b2942f601c
Modify hpack benchmark scenarios ( #27567 )
...
It used to be that we could run the hpack deserializer into an arbitrary
sink - that's about to change and the hpack deserializer will buffer
directly into a map. That being the case, it's really hard to compare
performance of the current and the new system. With this PR, I'm
updating the relevant scenario to be more similar to the eventual
scenario that we plan, so that as the new work finishes up, we can get a
clearer idea of what changes we're really making.
2021-10-04 13:14:43 -07:00
1c6634ac44
Changed grpc_error_get|set_str to use std string instead of slice ( #27466 )
...
* Changed grpc_error_get|set_str to use std string
* Fix init order in tests with gtest
* Undo gtest-tify credentials_test
2021-10-01 18:58:47 -07:00
91a2db9bc2
Make deadline the first metadata trait ( #27468 )
2021-09-28 10:09:04 -07:00
dbade1f5e7
Revert "Revert "pass subchannel address separately instead of as a channel arg ( #27443 )" ( #27489 )" ( #27491 )
...
This reverts commit 20cc6e7414
2021-09-28 09:54:07 -07:00
9b41852a5e
channel_init --> core configuration system ( #27347 )
...
* channel_init --> core configuration system
* Automated change: Fix sanity tests
* fixes
* Automated change: Fix sanity tests
* fixes
* fixes
* eliminate grpc_base_c
* fixes
* fix merge
* fix
* update visibility
* fixes
* feedback
* fix comment
Co-authored-by: ctiller <ctiller@users.noreply.github.com>
2021-09-27 15:51:39 -07:00
20cc6e7414
Revert "pass subchannel address separately instead of as a channel arg ( #27443 )" ( #27489 )
...
This reverts commit 57982f0ec6
2021-09-27 10:42:16 -07:00
57982f0ec6
pass subchannel address separately instead of as a channel arg ( #27443 )
...
* pass subchannel address separately instead of as a channel arg
* fix bug
* fix passing address to connector
* clang-format
* fix build of bm_call_create
* revert to C-style semantics for grpc_resolved_address
* fix msan bugs
2021-09-25 07:59:47 -07:00
d4407b7e59
Revert "Remove dead code ( #27445 )" ( #27461 )
...
This reverts commit 693f36f91c
2021-09-23 12:56:40 -07:00
693f36f91c
Remove dead code ( #27445 )
...
* Remove dead code
* Automated change: Fix sanity tests
* build
* remove benchmark
Co-authored-by: ctiller <ctiller@users.noreply.github.com>
2021-09-22 19:31:00 -07:00
47586fb36e
Revert "Revert "Added absl::Status support to error_utils ( #27358 )" ( #27418 )" ( #27429 )
...
This reverts commit 933676c56c
2021-09-21 14:42:57 -07:00
933676c56c
Revert "Added absl::Status support to error_utils ( #27358 )" ( #27418 )
...
This reverts commit 9b3f75d322
2021-09-21 09:48:57 -07:00
b669a3c521
Revert "Revert "a useful change ( #27381 )" ( #27411 )" ( #27412 )
...
This reverts commit 5db17f7350
2021-09-20 19:46:11 -07:00
9b3f75d322
Added absl::Status support to error_utils ( #27358 )
2021-09-20 18:03:17 -07:00
9872da79a1
Eliminate grpc_metadata_batch_{init,destroy,move} ( #27349 )
...
* Eliminate grpc_metadata_batch_{init,destroy,move}
* fixes
* fixes
* fixes
* fixes
* portability fixes
* fixes
* fix
* fix
* fix
* be explicit about being explicit
2021-09-20 17:25:26 -07:00
5db17f7350
Revert "a useful change ( #27381 )" ( #27411 )
...
This reverts commit 11eb70bbed
2021-09-20 10:30:59 -07:00
11eb70bbed
a useful change ( #27381 )
...
* a useful change
* Automated change: Fix sanity tests
* fix
* fix
* QsortCompare it is!
* Automated change: Fix sanity tests
* add missing header
* clang-tidy fixes, tweak clamp to only use <
Co-authored-by: ctiller <ctiller@users.noreply.github.com>
2021-09-20 09:13:27 -07:00
fd233193ec
Encapsulate metadata aggregates ( #27262 )
...
* Encapsulate metadata aggregates
* Automated change: Fix sanity tests
* add const
* Automated change: Fix sanity tests
* add a comment about an awful api
Co-authored-by: ctiller <ctiller@users.noreply.github.com>
2021-09-14 09:45:38 -07:00
ea389c00c2
Adjust include order per style guide ( #27175 )
...
Introduce clang-format configuration to sort includes closer to our rules.
2021-09-08 12:14:44 -07:00
dc701787e2
Convert HPACK encoder to C++ ( #27226 )
...
* Rebuild HPACK encoder table as C++
* move comment
* incguards
* build
* Automated change: Fix sanity tests
* c++ initialization ftw
* Automated change: Fix sanity tests
* Add missing header
* Add missing header
* Begin converting HPACK encoder to c++
* First pass conversion to c++
* fixes
* Automated change: Fix sanity tests
Co-authored-by: ctiller <ctiller@users.noreply.github.com>
2021-09-07 14:28:06 -07:00
b062fcd5c6
Ensure consistency of include style for <grpc.* headers ( #27233 )
...
* Add a tool to ensure consistency in how grpc public headers are included
* Apply tool
* add to test suite
* isort
2021-09-02 11:42:33 -07:00
d10617edb5
Move resource_user ownership into chttp2 transport/server/connector v2 ( #27032 )
...
Reintroducing PR #26643 , which was reverted in #27029
Fixed a memory leak and added a test that would have caught it (ASAN build): ca0c8c4
2021-08-18 17:04:00 -07:00
f5d3ed2db1
Revert "Move resource_user ownership into chttp2 transport/server/connector ( #26643 )" ( #27029 )
...
This reverts commit d1935a65a1
2021-08-16 19:07:39 -07:00
d1935a65a1
Move resource_user ownership into chttp2 transport/server/connector ( #26643 )
2021-08-16 13:50:56 -07:00
bbea27de46
Reland HPACK parsing changes ( #26997 )
...
* Revert "Revert "HPACK Table --> C++ (#26851 )" (#26995 )"
This reverts commit 840bcce9c4
2021-08-13 10:57:42 -07:00
840bcce9c4
Revert "HPACK Table --> C++ ( #26851 )" ( #26995 )
...
This reverts commit 83bcb0cf2e
2021-08-12 11:20:19 -07:00
83bcb0cf2e
HPACK Table --> C++ ( #26851 )
...
* Buffer HPACK parsing until the end of a header boundary
HTTP2 headers are sent in (potentially) many frames, but all must be
sent sequentially with no traffic intervening.
This was not clear when I wrote the HPACK parser, and still indeed quite
contentious on the HTTP2 mailing lists.
Now that matter is well settled (years ago!) take advantage of the fact
by delaying parsing until all bytes are available.
A future change will leverage this to avoid having to store and verify
partial parse state, completely eliminating indirect calls within the
parser.
* maybe fixes
* xx
* fix boundary detection
* clang-format
* Revert "xx"
This reverts commit 258d712ed3#26700 )"
This reverts commit 8bab3e4bf4
2021-08-09 16:07:41 -07:00
613b90ba93
Revert "Buffer HPACK parsing until the end of a header boundary ( #26700 )" ( #26825 )
...
This reverts commit 8bab3e4bf4
2021-07-29 11:22:47 -07:00
8bab3e4bf4
Buffer HPACK parsing until the end of a header boundary ( #26700 )
...
HTTP2 headers are sent in (potentially) many frames, but all must be
sent sequentially with no traffic intervening.
This was not clear when I wrote the HPACK parser, and still indeed quite
contentious on the HTTP2 mailing lists.
Now that matter is well settled (years ago!) take advantage of the fact
by delaying parsing until all bytes are available.
A future change will leverage this to avoid having to store and verify
partial parse state, completely eliminating indirect calls within the
parser.
2021-07-28 15:36:51 -07:00
d3e5803cb2
Move HPACK parser to C++ ( #26689 )
...
This is a fairly low effort migration of the current codebase into a C++ class, instead of free standing C code.
It builds upon #26657 as a necessary first step.
I've tried to minimize any changes to semantics or logic in this change, except where required to get a minimal amount of encapsulation - which is the major aim of this change.
A future change in this series will buffer slices until all HPACK headers are in memory for a stream prior to decoding -- it's important to have an encapsulated API to the parser before doing so however (hence this CL).
The next change after that will be an almost complete rewrite of the parsing functionality -- since we'll have the total set of header bytes, we'll no longer need to support suspending decoding at arbitrary points. This will allow us to move to a simple recursive descent parser, eliminate a bunch of indirection in this code, and end up in a much more malleable place for when we start doing metadata API changes.
(we likely also end up with some good performance wins!)
2021-07-15 13:51:23 -07:00
ea4b68e7a1
Drop experimental tags from core callback API ( #26535 )
2021-06-24 17:20:43 -07:00
8aff94558c
De-experimentalize C++ callback API ( #25728 )
...
* De-experimentalize callback API
* Make FromServerContext based on ServerContextBase
* Fix lambda
* Fix headers
* De-experimentalize tests
* clang-format
* Fix consistency checks
* wip
* Fix const-ness of callback client read RPC requests
* Fix golden file
* Give full route_guide callback client example with Hold etc
* Complete example route-guide server
* De-experimentalize test services
* Omit unneeded using
* Remove some uses of non-experimental macro from test
* clang-format examples
* De-experimentalize async stub calls
* Remove experimental namespace use in qps, ubms
* De-experimentalize alarms, generic stubs, and context allocator
* De-experimentalize SetContextAllocator
* clang-format
* Fix conflicts
* Leave obsolete API in place until users can be migrated
2021-05-27 14:55:25 -07:00
0f80378a21
Resolve race between OnReadDone(ok=false) and IsCancelled ( #26245 )
...
* Resolve and test race between OnReadDone(ok=false) and IsCancelled
* Fix retry case
* Fix health check case
* Address reviewer comments.
* ADD TODO requested by markdroth
2021-05-19 14:05:46 -07:00
ca945a58e9
Introduced grpc_error_handle ( #25902 )
...
- Define grpc_error_handle
- Replace grpc_error* with grpc_error_handle
2021-04-22 12:00:28 -07:00
3f19333ced
Move retry code into its own filter in the DynamicFilter stack ( #25820 )
...
* rename ChannelData to ClientChannel
* make ClientChannel class definition public
* move retry code to its own filter
* move LB call factory method to ClientChannel class
* move dynamic termination filter out of ClientChannel class
* update comments
* remove retry parsing from client channel service config parser
* fix clang-tidy
* fix service_config_test
* clang-format
2021-03-31 15:39:14 -07:00
7021b72d1f
Revert Revert Xds Status Notifier ( #25718 )
...
* Revert "Revert "xDS status notifier (#25321 )" (#25702 )"
This reverts commit 3c9f3972e3
2021-03-15 14:29:06 -07:00
3c9f3972e3
Revert "xDS status notifier ( #25321 )" ( #25702 )
...
This reverts commit 81e90432e1
2021-03-12 08:54:34 -08:00
81e90432e1
xDS status notifier ( #25321 )
...
* Serving status notification for xds enabled servers
2021-03-08 21:47:13 -08:00
0901c9914b
Remove CAS loops in global subchannel pool and simplify subchannel refcounting ( #25485 )
2021-03-04 20:29:34 -08:00
0e3a02e903
Revert "Remove the `urgent` argument from iomgr tcp read API ( #25494 )" ( #25592 )
...
This reverts commit a3398f9
2021-03-01 22:04:22 -08:00
a3398f924c
Remove the `urgent` argument from iomgr tcp read API ( #25494 )
...
The urgent argument is a platform-specific flag that leaked into the (ideally) platform-independent HTTP/2 transport layer. In an effort to clean up the cross-platform API surface, it would be helpful if we can remove this argument from the TCP Read api without losing the performance optimization that was introduced along with it (see #18240 ).
2021-02-24 12:30:14 -08:00
af76d0321a
Restore callback microbenchmarks but make them manual for now ( #25491 )
2021-02-19 13:36:11 -08:00
0e6db8fbec
Reduce max size of bm_callback_streaming_ping_pong messages to match unary
2021-02-17 09:55:45 -08:00
dba8bb59a0
Implement callback CQ at C++ layer using shared thread pool if non-background iomgr
2021-02-16 16:02:23 -08:00
f4e55b85bc
Replace disrespectful term
2021-02-08 15:29:53 -08:00
9bd7c4917a
Fix unused-parameter warning
2021-01-19 18:18:09 -08:00
53ab235fb8
Fix google-explicit-constructor
2020-12-03 15:03:39 -08:00
e0d8c498a6
C++ cast for others
2020-11-30 17:36:21 -08:00
2ff84ed8c2
C++ cast for void*
2020-11-30 17:29:15 -08:00
ec0b88a5d7
Fix google-readability-casting
2020-11-30 17:27:51 -08:00
75e3e660fd
Fix google-readability-todo
2020-11-25 10:09:42 -08:00
131579d52d
Roll forward of commit 7883b51: Use abseil's flags in tests.
...
This replaces gflags. Added TODOs where use of `absl::Duration` or `absl::FlagSaver` might be preferred in follow-up cleanup. Fixes #24493 .
This reverts commit da66b7d14e5b43440
2020-11-06 16:07:19 +01:00
da66b7d14e
Revert "Abseil Flags"
2020-10-24 15:19:55 -07:00
7883b5133d
Use abseil's flags in tests.
...
This replaces gflags. Added TODOs where use of `absl::Duration` or `absl::FlagSaver` might be preferred in follow-up cleanup. Fixes #24493 .
2020-10-23 19:36:33 +02:00
5b375a3728
Fix by misc-definitions-in-headers
2020-10-19 21:56:17 -07:00
a62a70207c
Merge pull request #24448 from veblush/clang-format-8
...
Upgrade clang-format 8
2020-10-19 09:29:35 -07:00
84df61b31a
Merge pull request #24453 from veblush/tidy-modernize-make-unique
...
[Clang-Tidy] Enable modernize-make-unique
2020-10-16 16:35:08 -07:00
e7434d385e
Fix by tidy-modernize-use-override
2020-10-16 14:01:06 -07:00
6ae8463500
Fix by modernize-make-unique
2020-10-16 13:30:56 -07:00
34789b2d83
Fix by clang-format-8.
2020-10-16 11:37:18 -07:00
ece76ee1e8
Fix code by readability-container-size-empty
2020-10-16 09:48:20 -07:00
91ca04cf6f
Add TestGrpcScope
2020-09-17 12:41:14 -07:00
7f7a6916f6
Merge pull request #23581 from markdroth/server_c++
...
Convert grpc_server to idiomatic C++.
2020-08-03 14:45:53 -07:00
0b61aea1c5
Convert grpc_server to idiomatic C++.
2020-08-03 12:56:29 -07:00
cd20a80a1e
Merge pull request #23653 from vjpai/bm_cq
...
Properly follow callback CQ API guarantees rather than existing behavior in ubm
2020-08-03 12:39:17 -07:00
94a85b051d
Expand comment to address reviewer feedback
2020-07-31 14:26:17 -07:00
c6586f087f
implemented get_local_address accessor for grpc_endpoint
2020-07-29 19:39:22 +00:00
968e1b40a5
Properly follow callback API guarantees rather than existing behavior
2020-07-28 12:42:20 -07:00
9c5a39c6db
Revert "Merge pull request #23361 from vjpai/em_agnostic_core_callback_cq"
...
This reverts commit a46cb5e86ab5d42e75fb
2020-07-28 09:55:43 -07:00
9d79ca6058
C++ify core server
2020-07-20 03:15:18 -07:00
7ee2bc6db5
EM-agnostic callback completion queue
2020-07-08 16:12:23 -07:00
b92ca15356
Increase size of a test to match internal guidance
2020-07-06 09:51:05 -07:00
c736cf073c
Added call to grpc::testing::TestEnvironment in more tests
2020-07-01 13:27:08 -07:00
62ac3f075a
Added call to grpc::testing::TestEnvironment in tests
2020-07-01 11:37:57 -07:00
165ee5007a
Replaced grpc::string with std::string
2020-06-29 17:56:36 -07:00
e98eaa5052
Revert "Revert "Add message-size check before message decompression""
2020-06-23 12:19:07 -07:00
70465c92ec
Merge pull request #23180 from karthikravis/opencensus
...
Move OpenCensus back to ::grpc from ::grpc_impl
2020-06-23 11:31:43 -07:00
f0d2aff3c7
Merge branch 'master' into opencensus
2020-06-22 17:24:05 -07:00
c7f2956b67
Revert "Add message-size check before message decompression"
2020-06-22 16:18:36 -07:00
0497ad8bb0
Add missing message-size check before decompressing
...
Add and fix tests for limit check before decompression
Code restructuring to allow easy reuse of service config call data
Regenerate projects
Reviewer comments
2020-06-19 14:49:18 -07:00
7fa6826134
Merge branch 'master' into opencensus
2020-06-10 17:47:16 -07:00
89763a96fd
Move OpenCensus back to ::grpc from ::grpc_impl
...
Revert - Revert "Revert "Fold opencensus into grpc_impl namespace"" #18396
2020-06-10 17:36:58 -07:00
92d31d6f99
Remove the flaky flag from bm_opencensus_plugin
2020-05-04 11:41:09 -07:00
234b356e2f
Fix the flaky bm_opencensus_plugin
2020-04-29 09:35:57 -07:00
527509abc4
Unmark flaky on streaming_pump
2020-04-08 10:31:47 -07:00
f1979cece2
Disable SockPair fixture
2020-04-08 08:41:54 -07:00
a791e16983
mark more tests as flaky
2020-03-17 05:32:17 -04:00
880111fe47
Merge pull request #22355 from jtattermusch/bm_metadata_not_flaky
...
bm_metadata is not flaky anymore
2020-03-16 18:14:25 +01:00
b4f0cad4bd
bm_arena needs more time
...
Internal tests show that it needs a large timeout. This has passed 100x without timeout.
2020-03-16 08:12:15 -07:00
29528fc026
bm_metadata is not flaky anymore
2020-03-16 03:07:25 -04:00
b40b162cc9
Add missing grpc_slice_unref to bm_metadata
2020-03-13 18:28:16 -07:00
cbb0ef2717
try marking some offenders as flaky
2020-02-27 22:37:32 -05:00
543d8796f0
dont build some microbenchmarks on windows
2020-02-26 19:56:49 -05:00
b92c3a2879
Changed Closure to TestClosure to avoid name conflict
2020-02-18 17:25:55 -08:00
b18ae4b784
Add notap tag to tests that cause issues timeout issues internally
2020-02-18 12:31:46 -08:00
5e19408ef5
disable bm_chttp2_transport on msan
2020-02-18 03:20:18 -08:00
cb6d09ac21
add missing bm_pollset and bm_chttp2_transport targets to bazel build
2020-02-18 02:34:42 -08:00
f7d857a8e4
Make threadpool and trickle tests manual only (excessive time/resources)
2020-02-13 10:12:12 -08:00
1513f3a7b4
Reduce unneeded sizes and disable tests on mac for resource limits
2020-02-13 09:34:15 -08:00
0c0f3f0556
Declare microbenchmarks as tests rather than binaries, fix issues
2020-02-13 09:34:15 -08:00
2678afe310
Stop using core server shutdown in microbenchmarks
2020-01-21 14:24:42 -08:00
d5d6914751
Make buildifier sanity test fail-able
2019-12-18 16:16:49 -08:00
5d352ded8b
disable a few macos bazel tests
2019-11-25 13:35:26 +01:00
be6979202b
Revamp server-side C++ callback API, message allocator, and mocking
2019-11-20 16:56:18 -08:00
79aa21fa71
Reviewer comments
2019-11-13 14:14:02 -08:00
61020b55f9
Remove GRPC_CLOSURE_LIST_SCHED and remove scheduler field from grpc_closure
2019-11-12 18:23:29 -08:00
d60b60a837
Reviewer comments
2019-11-08 10:45:50 -08:00
fe88f14410
Merge branch 'master' into removrun
2019-11-07 15:13:03 -08:00
3a519a0b64
Replaced grpc_core New & Delete with C++ new & delete
2019-11-06 17:52:26 -08:00
c34bef4ca4
Format all Bazel files with buildifier
2019-11-05 14:03:52 -08:00
08ba7246c0
Remove GRPC_CLOSURE_RUN and replace with grpc_core::Closure::Run
2019-11-01 16:21:16 -07:00
3a189d7e08
test changes
2019-10-31 17:09:38 -07:00
e35c869aaf
Remove unused-parameter warnings, round 2 (19 of 19)
2019-10-21 09:16:35 -07:00
780d41224c
Squashed commit of the following:
...
commit 1547cb209a3edeee7ce922b343e4fbe8f78e7a5d22b343e4fb19e60dfe8f153bdcbc97feae38d3ab153bdcbc97c6da80bcce6a32264cdf98abc22f4c6a32264cdf6bbd3a1c3caaa04526a24266be13d514107957aa5643aa6cb3c5964494309f933903b93c3a7d0e9b56539cc448714ec01e4b54dcbd170d7a3388b077714e4c849f1d17ad7d443110c062c50e10bc17ea4a71a911e88610a64ec95f22055d0d4b1223f875a17bbbd8403a33ed476211058748fd3d363368ca3d363368ca
2019-10-18 15:07:54 -07:00
e8f78e7a5d
Merge pull request #20644 from vjpai/remove_warn_17
...
Remove unused parameter warnings (17 of 20)
2019-10-18 10:32:48 -07:00
82e185399d
clang-format
2019-10-17 09:08:21 -07:00
0cab2084e6
clang-format
2019-10-17 09:08:04 -07:00
6ca827a008
Remove unused parameter warning (18 of 20)
2019-10-17 08:52:05 -07:00
406b70629a
Remove unused parameter warning (17 of 20)
2019-10-17 08:52:00 -07:00
0a57e9cc69
Fix bm chttp2 transport
2019-10-02 16:14:43 -07:00
c9376b4e0b
Use cycle clock instead of clock monotonic to measure call latency.
...
This removes two more getttime syscalls from the hot path, when cycle
clock is enabled.
2019-09-09 10:37:48 -04:00
0b06676c9e
hpack encoder optimizations.
...
Removed some cycles and branches from hpack_enc for CH2.
Specifically:
1. Pushed certain metadata key/value length checks to
prepare_application_metadata() in src/core/lib/surface/call.cc.
This means that rather than check all key/val lengths for all metadata, we only
do so for custom added user metadata. Inside CH2, we change the length checks to
debug checks so we can catch if core/filter metadata fails to pass the check.
2. Changed various asserts to debug asserts when able.
3. Refactored some of the header emission code to remove duplicated code.
4. Un-inlined some logging methods.
This results in somewhat faster hpack_encoder performance:
BM_HpackEncoderInitDestroy
222ns ± 0% 221ns ± 0% -0.29% (p=0.000 n=34+34)
BM_HpackEncoderEncodeDeadline
[framing_bytes/iter:9 header_bytes/iter:6 ] 135ns ± 1%
124ns ± 0% -8.05% (p=0.000 n=39+38)
BM_HpackEncoderEncodeHeader<EmptyBatch>/0/16384
[framing_bytes/iter:9 header_bytes/iter:0 ] 34.2ns ± 0%
34.2ns ± 0% -0.01% (p=0.014 n=34+38)
BM_HpackEncoderEncodeHeader<EmptyBatch>/1/16384
[framing_bytes/iter:9 header_bytes/iter:0 ] 34.2ns ± 0%
34.2ns ± 0% -0.04% (p=0.004 n=34+37)
BM_HpackEncoderEncodeHeader<SingleStaticElem>/0/16384
[framing_bytes/iter:9 header_bytes/iter:1 ] 47.5ns ± 0%
45.9ns ± 0% -3.28% (p=0.000 n=28+38)
BM_HpackEncoderEncodeHeader<SingleInternedKeyElem>/0/16384
[framing_bytes/iter:9 header_bytes/iter:6 ] 77.0ns ± 1%
68.3ns ± 1% -11.33% (p=0.000 n=39+40)
BM_HpackEncoderEncodeHeader<SingleInternedElem>/0/16384
[framing_bytes/iter:9 header_bytes/iter:1 ] 47.7ns ± 1%
45.5ns ± 0% -4.63% (p=0.000 n=39+33)
BM_HpackEncoderEncodeHeader<SingleInternedBinaryElem<1, false>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:1 ] 47.2ns ± 0%
45.3ns ± 0% -3.96% (p=0.000 n=33+34)
BM_HpackEncoderEncodeHeader<SingleInternedBinaryElem<3, false>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:1 ] 47.7ns ± 0%
45.6ns ± 0% -4.54% (p=0.000 n=38+40)
BM_HpackEncoderEncodeHeader<SingleInternedBinaryElem<10, false>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:1 ] 47.7ns ± 0%
45.5ns ± 0% -4.63% (p=0.000 n=39+32)
BM_HpackEncoderEncodeHeader<SingleInternedBinaryElem<31, false>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:1 ] 47.8ns ± 0%
45.6ns ± 1% -4.59% (p=0.000 n=38+39)
BM_HpackEncoderEncodeHeader<SingleInternedBinaryElem<100, false>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:1 ] 47.8ns ± 0%
45.5ns ± 0% -4.64% (p=0.000 n=39+36)
BM_HpackEncoderEncodeHeader<SingleInternedBinaryElem<1, true>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:1 ] 47.3ns ± 0%
45.3ns ± 0% -4.09% (p=0.000 n=38+36)
BM_HpackEncoderEncodeHeader<SingleInternedBinaryElem<3, true>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:1 ] 47.8ns ± 1%
45.6ns ± 0% -4.71% (p=0.000 n=37+40)
BM_HpackEncoderEncodeHeader<SingleInternedBinaryElem<10, true>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:1 ] 47.7ns ± 0%
45.5ns ± 0% -4.66% (p=0.000 n=39+32)
BM_HpackEncoderEncodeHeader<SingleInternedBinaryElem<31, true>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:1 ] 47.8ns ± 1%
45.6ns ± 1% -4.62% (p=0.000 n=37+39)
BM_HpackEncoderEncodeHeader<SingleInternedBinaryElem<100, true>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:1 ] 47.7ns ± 0%
45.5ns ± 0% -4.67% (p=0.000 n=38+32)
BM_HpackEncoderEncodeHeader<SingleNonInternedElem>/0/16384
[framing_bytes/iter:9 header_bytes/iter:9 ] 80.5ns ± 1%
74.7ns ± 0% -7.16% (p=0.000 n=38+35)
BM_HpackEncoderEncodeHeader<SingleNonInternedBinaryElem<1, false>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:12 ] 105ns ± 1%
99ns ± 0% -5.91% (p=0.000 n=38+34)
BM_HpackEncoderEncodeHeader<SingleNonInternedBinaryElem<3, false>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:14 ] 111ns ± 1%
106ns ± 1% -4.86% (p=0.020 n=39+2)
BM_HpackEncoderEncodeHeader<SingleNonInternedBinaryElem<10, false>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:23 ] 135ns ± 0%
130ns ± 0% -3.45% (p=0.020 n=35+2)
BM_HpackEncoderEncodeHeader<SingleNonInternedBinaryElem<31, false>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:46 ] 225ns ± 1%
223ns ± 0% -0.91% (p=0.003 n=37+2)
BM_HpackEncoderEncodeHeader<SingleNonInternedBinaryElem<100, false>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:120 ] 467ns ± 0%
472ns ± 0% +1.09% (p=0.003 n=38+2)
BM_HpackEncoderEncodeHeader<SingleNonInternedBinaryElem<1, true>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:12 ] 81.6ns ± 1%
74.8ns ± 0% -8.40% (p=0.000 n=37+33)
BM_HpackEncoderEncodeHeader<SingleNonInternedBinaryElem<3, true>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:14 ] 82.0ns ± 1%
74.8ns ± 0% -8.80% (p=0.000 n=37+32)
BM_HpackEncoderEncodeHeader<SingleNonInternedBinaryElem<10, true>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:21 ] 82.1ns ± 1%
74.9ns ± 0% -8.86% (p=0.000 n=35+34)
BM_HpackEncoderEncodeHeader<SingleNonInternedBinaryElem<31, true>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:42 ] 97.6ns ± 2%
91.8ns ± 0% -5.95% (p=0.000 n=35+27)
BM_HpackEncoderEncodeHeader<SingleNonInternedBinaryElem<100, true>>/0/16384
[framing_bytes/iter:9 header_bytes/iter:111 ] 97.2ns ± 1%
91.2ns ± 2% -6.19% (p=0.000 n=37+38)
BM_HpackEncoderEncodeHeader<SingleNonInternedElem>/0/1
[framing_bytes/iter:54 header_bytes/iter:9 ] 230ns ± 0%
221ns ± 0% -3.91% (p=0.000 n=38+37)
BM_HpackEncoderEncodeHeader<MoreRepresentativeClientInitialMetadata>/0/16384
[framing_bytes/iter:9 header_bytes/iter:16 ] 206ns ± 2%
170ns ± 1% -17.51% (p=0.000 n=39+39)
BM_HpackEncoderEncodeHeader<RepresentativeServerInitialMetadata>/0/16384
[framing_bytes/iter:9 header_bytes/iter:3 ] 66.4ns ± 2%
62.5ns ± 1% -5.85% (p=0.000 n=34+39)
BM_HpackEncoderEncodeHeader<RepresentativeServerTrailingMetadata>/1/16384
[framing_bytes/iter:9 header_bytes/iter:1 ] 47.5ns ± 0%
45.9ns ± 1% -3.29% (p=0.000 n=26+38)
2019-09-06 13:43:29 -07:00
9bb1247bdb
Merge pull request #16327 from vjpai/keep_running
...
C++ microbenchmarks: update from KeepRunning to range-based for
2019-09-05 11:07:53 -07:00
6175294416
Merge pull request #20014 from arjunroy/on_hdr_nobranch
...
s/branch/tail_call/ for CH2 on_hdr().
2019-09-03 12:44:56 -07:00
32801fb5eb
Remove build target for microbenchmark
2019-08-29 14:11:05 -07:00
1077b3435c
Use range-based for on state rather than state.KeepRunning when possible
2019-08-29 11:27:00 -07:00
b46e3668d3
s/branch/tail_call/ for CH2 on_hdr().
...
on_hdr() checks if a void-return function pointer is null before jumping to it.
If it is null, it returns an error; else it executes that function and returns
success.
This change converts the void-returning function to one that returns a
grpc_error* and thus saves a branch in on_hdr() (since we're branching once by
following the function pointer anyways, we're effectively coalescing these two
branches).
2019-08-28 13:06:11 -07:00
6dfe27ab08
Fix race in bm_chttp2_transport
2019-08-28 10:07:38 -07:00
0bdeca47b6
Merge pull request #19038 from markdroth/lb_policy_create_channel_api_improvement
...
Remove CreateChannel() method from LB helper API.
2019-08-27 10:54:52 -07:00
59564ebd96
Fix warnings to unblock gcc8 support
2019-08-20 16:02:53 -07:00
a23e7ab188
Merge remote-tracking branch 'upstream/master' into lb_policy_create_channel_api_improvement
2019-08-20 07:58:38 -07:00
557446a11e
Added specializations for grpc_mdelem_create.
...
In several cases, we create grpc mdelem structures using known-static
metadata inputs. Furthermore, in several cases we create a slice on
the heap (e.g. grpc_slice_from_copied_buffer) where we know we are
transferring refcount ownership. In several cases, then, we can:
1) Avoid unnecessary ref/unref operations that are no-ops (for static
slices) or superfluous (if we're transferring ownership).
2) Avoid unnecessarily comprehensive calls to grpc_slice_eq (since
they'd only be called with static or interned slice arguments,
which by construction would have equal refcounts if they were
in fact equal.
3) Avoid unnecessary checks to see if a slice is interned (when we
know that they are).
To avoid polluting the internal API, we introduce the notion of
strongly-typed grpc_slice objects. We draw a distinction between
Internal (interned and static-storage) slices and Extern (inline and
non-statically allocated). We introduce overloads to
grpc_mdelem_create() and grpc_mdelem_from_slices() for the fastpath
cases identified above based on these slice types.
From the programmer's point of view, though, nothing changes - they
need only use grpc_mdelem_create() and grpc_mdelem_from_slices() as
before, and the appropriate fastpath will be picked based on type
inference. If no special knowledge exists for the slice type (i.e. we
pass in generic grpc_slice objects), the slowpath method will still
always return correct behaviour.
This is good for:
- Roughly 1-3% reduction in CPU time for several unary/streaming
ping pong fullstack microbenchmarks.
- Reduction of about 15-20% in CPU time for some hpack parser
microbenchmarks.
- 10-12% reduction of CPU time for metadata microbenchmarks involving
interned slice comparisons.
2019-07-30 17:07:44 -07:00
f14f1dd61d
Merge pull request #19695 from yunjiaw26/bm_threadpool
...
Add threadpool benchmark and build files
2019-07-29 18:38:22 -07:00
8318e578db
SpikyLoad: construct outside
2019-07-29 10:52:32 -07:00
c3c24d089d
Use Template
2019-07-29 10:30:09 -07:00
d87b5285ca
Fix comment
2019-07-24 13:33:24 -07:00
efd6946d21
Reformat
2019-07-23 15:35:50 -07:00
9242fe122d
AddSelf more scenarios
2019-07-23 09:47:26 -07:00
46f706c99b
Revert "Merge pull request #19686 from gnossen/revert_breakage"
...
This reverts commit 1f2398b0d599169d811c
2019-07-23 07:50:27 -07:00
8278d3e6a5
Resolving comments
2019-07-22 17:24:09 -07:00
847faf407f
Removes unused variable error
2019-07-19 12:10:52 -07:00
85314b3fcc
Re-format
2019-07-19 10:49:35 -07:00
c6bc2b1875
Add threadpool benchmark and build files
2019-07-19 10:23:22 -07:00
63b4f3d819
Revert "Merge pull request #19673 from markdroth/channel_grpc_init"
...
This reverts commit 4e2198071662b8a783fa
2019-07-18 13:00:24 -07:00
8cc5b8f680
Defer grpc shutdown until after channel destruction.
2019-07-17 16:35:14 -07:00
b1d73a01f1
Removed duplicate static table from hpack table. Removed an or instruction for
...
every usage of static grpc metadata. Inlined hpack table lookups for static
metadata.
This leads to faster hpack parser creation:
BM_HpackParserInitDestroy 5.32µs ± 1% 0.06µs ± 1% -98.91% (p=0.000 n=18+19)
And slightly faster parsing:
BM_HpackParserParseHeader<RepresentativeClientInitialMetadata, OnInitialHeader>
456ns ± 1% 435ns ± 1% -4.74% (p=0.000 n=18+19)
BM_HpackParserParseHeader<MoreRepresentativeClientInitialMetadata,
OnInitialHeader>
1.06µs ± 2% 1.04µs ± 2% -1.82% (p=0.000 n=19+20)
It also yields a slight (0.5 - 1.0 microsecond) reduction in CPU time for
fullstack unary pingpong:
BM_UnaryPingPong<TCP, NoOpMutator, NoOpMutator>/0/512
[polls/iter:3.0001 ] 23.9µs ± 2%
23.0µs ± 1% -3.63% (p=0.002 n=6+6)
BM_UnaryPingPong<TCP, NoOpMutator, NoOpMutator>/0/32768
[polls/iter:3.00015 ] 35.1µs ± 1%
34.2µs ± 1% -2.57% (p=0.036 n=5+3)
BM_UnaryPingPong<MinTCP, NoOpMutator, NoOpMutator>/8/0
[polls/iter:3.00011 ] 21.7µs ± 3%
21.2µs ± 2% -2.44% (p=0.017 n=6+5)
2019-07-15 15:39:31 -07:00
56a0153f16
Heap allocate the stream object for other benchmark cases too
2019-06-14 14:44:49 -07:00
cceca10a8a
Fix data race, heap use-after-free issue in bm_chttp2_transport
2019-06-13 17:05:51 -07:00
4f3a55b73d
Merge remote-tracking branch 'upstream/master' into lb_policy_create_channel_api_improvement
2019-06-05 13:56:06 -07:00
0f5834b88d
Merge remote-tracking branch 'upstream/master' into lb_policy_create_channel_api_improvement
2019-05-29 15:40:24 -07:00
196b0aa3a3
Revert "Revert "Start supporting a callback-based RPC under lock""
2019-05-29 13:15:47 -07:00
b790c24e5c
Revert "Start supporting a callback-based RPC under lock"
2019-05-29 11:20:36 -07:00
477ebef532
Remove CreateChannel() method from LB helper API.
2019-05-28 14:18:15 -07:00
b18faa6c95
Fix tsan error
2019-05-23 20:08:45 -07:00
d2c8eb94c9
Fix microbenchmark failures
2019-05-23 18:09:12 -07:00
3f05ac9584
Merge branch 'master' into callback-rpc-lock
2019-05-22 17:39:08 -07:00
e1f62278e3
Fix clang error
2019-05-22 13:53:10 -07:00
4f7f561564
Add synchronization to bm test
...
- since we made the callback run on another thread, add synchronization
in bm tests as well
2019-05-22 10:42:50 -07:00
40210d3b8a
Move Channel to grpc_impl
2019-05-17 09:12:33 -07:00
772a74aced
Revert changes to Channel
2019-05-15 14:24:17 -07:00
a02c76dfb9
Cancel predefine number of streaming
2019-05-13 13:29:06 -07:00
87d75d2a88
Add explicit and fix error
2019-05-10 11:23:24 -07:00
7a464a3bc9
Merge remote-tracking branch 'upstream/master'
2019-05-10 11:16:14 -07:00
476f7cd81f
Merge pull request #18730 from karthikravis/grpc_namespace_channel_new
...
Fold Channel into grpc_impl from grpc namespace
2019-05-09 14:09:21 -07:00
a2daa4ff08
Clean format'
2019-05-09 12:29:05 -07:00
1ea651aee3
Add assertion
2019-05-09 12:22:55 -07:00
9b50fbcaac
Solve conflict
2019-05-08 18:50:06 -07:00
762e58b574
Change client context allocation
2019-05-08 17:34:27 -07:00
070902b871
Merge bm_callback_cq to bm_cq
2019-05-08 15:51:58 -07:00
3fc702510f
Reuse reactor to send new RPC
2019-05-08 10:15:43 -07:00
2d5a9750a0
Manually add echo.proto to pass Portability build test
2019-05-06 16:20:59 -07:00
1ba5f5c701
Modify build file
2019-05-06 14:55:00 -07:00
32e10e618a
address the reference arguments
2019-05-06 10:41:37 -07:00
714e13b426
Delete log
2019-05-06 09:25:46 -07:00
c905f76a5b
Clang format
2019-05-06 09:18:29 -07:00
4198c4fcc6
Merge remote-tracking branch 'upstream/master'
2019-05-03 10:41:12 -07:00
2787dedd70
Modify dependency of callback_test_service
2019-05-03 10:40:49 -07:00
927c2f2c61
Change format
2019-05-02 17:24:54 -07:00
5748665bc5
Add callback completion queue and modify callback streaming ping pong
2019-05-02 17:21:55 -07:00
e18ed03c04
Made gRPC inialized after entering main function in microbenchmarks.
2019-05-02 16:58:37 -07:00
25128d18c1
Modify unary ping pong to send next rpc in callback function
2019-05-02 15:14:44 -07:00
7d3fdec445
Add microbenchmark for callback completion queue
2019-05-01 11:29:28 -07:00
e70d507abe
Changes based on comment
2019-04-30 16:32:31 -07:00
e3a0c85960
Merge branch 'master' into grpc_namespace_channel_new
2019-04-30 14:30:03 -07:00
b028141f01
Change streaming ping pong args and add comment
2019-04-30 14:24:40 -07:00
bc09c70799
Merge pull request #18454 from grpc/grpc_namespace_channel_arguments
...
Bring ChannelArguments to grpc_impl from grpc
2019-04-30 10:16:45 -07:00
8bf138d799
Add copyright
2019-04-26 14:29:57 -07:00
57bc124d3f
Merge remote-tracking branch 'upstream/master'
2019-04-26 12:03:01 -07:00
875d2df399
Modify build file
2019-04-26 10:52:21 -07:00
bbd4eb5028
Add microbenchmark for callback unary ping pong and bidistreaming ping pong
2019-04-26 09:52:12 -07:00
9345b7e276
s/atleast/at\ least
2019-04-25 17:15:30 -07:00
8ce42f67b2
Shrink arena size by 40 bytes and add additional
...
alignment options (for cache-alignment).
We shrink by:
1) Removing an unnecessary zone pointer.
2) Replacing gpr_mu (40 bytes when using pthread_mutex_t) with
std::atomic_flag.
We also header-inline the fastpath alloc (ie. when not doing a zone
alloc) and move the malloc() for a zone alloc outside of the mutex
critical zone, which allows us to replace the mutex with a spinlock.
We also cache-align created arenas.
2019-04-24 16:39:03 -07:00
ce3ff86763
Convert call_combiner to C++.
2019-04-23 15:27:06 -07:00
858b5cca20
Fix clang_format_code.sh issues and move the internal calls to new name
2019-04-23 11:45:11 -07:00
b9a279c030
Resolving ambiguous call to CreateCustomChannel.
2019-04-22 23:53:53 +02:00
973fa6950e
Fold Channel into grpc_impl from grpc namespace
2019-04-11 12:04:23 -07:00
Craig Tiller