Craig Tiller
5fc3ff8203
grpc_millis -> Timestamp/Duration ( #28119 )
...
* wip
* Automated change: Fix sanity tests
* fixes
* progress
* progress
* grpc compiles
* Automated change: Fix sanity tests
* fixing tests
* x
* progress
* better code
* Automated change: Fix sanity tests
* progress
* progress
* windows fix
* Make Duration metadata trivial
* better message
* fix
* Automated change: Fix sanity tests
* fix
* fix
* fix
* fix
* Automated change: Fix sanity tests
* Automated change: Fix sanity tests
* fix
* progress
* fixes
* fix
* fix
* spam
* un-disable errantly disabled tests
* gain insight
* Automated change: Fix sanity tests
* fixes
* fixes
* fix
* debug
* tweak
* fix
* fix timeout
* fix comment
* fixes
* x
* better test
* tests
* Automated change: Fix sanity tests
* missed file
* fix
* x
* fix
* fix
* fix
* fix
* Automated change: Fix sanity tests
* fix
* merge
* Automated change: Fix sanity tests
* fix
Co-authored-by: ctiller <ctiller@users.noreply.github.com>
3 years ago
Esun Kim
342cb4457c
Added a temporary trap to prevent Abseil-Status breakages. ( #28766 )
...
* Added a new trap to prevent build errors with use_abseil_status enabled
* Fix build errors.
3 years ago
Vignesh Babu
0dda706907
fixing heap use after free bug due to mismanaged ref counting in passthru endpoint ( #28417 )
...
* fixing heap use after free bug due to mismanaged ref counting in passthru endpoint
* fix sanity checks
3 years ago
Craig Tiller
f6d392b623
Fix timeout bug in api fuzzer ( #28385 )
...
* Fix timeout bug in api fuzzer
* add example
3 years ago
Vignesh Babu
bb02562307
Fix use after free bug in API fuzzer ( #28230 )
3 years ago
Vignesh Babu
851ad13313
Filter zero length slices passed to grpc passthru endpoint ( #28205 )
...
* updating passthru endpoint to filter 0 length slices passed to endpoint write
* adding testcase which caused the failure
* fix minor bug in passthru endpoint to handle rare case where endpoint write is called with no data to be written
3 years ago
Vignesh Babu
4dd0bba12e
Revert "Revert "Api fuzzer extensions to support simulating traffic congestion ( #27820 )" ( #27973 )" ( #27974 )
...
* Revert "Revert "Api fuzzer extensions to support simulating traffic congestion (#27820 )" (#27973 )"
This reverts commit 879f97ef70
.
* updating passthru_endpoint file to fix windows breakages
* Automated change: Fix sanity tests
Co-authored-by: Vignesh2208 <Vignesh2208@users.noreply.github.com>
3 years ago
Craig Tiller
a629c9a03e
Reland resource quota work ( #28017 )
...
* Check if memory owner available prior to polling it
The transport may drop the memory owner during its destruction sequence
* tcp_fix
* Revert "Revert "New resource quota integration (#27643 )" (#28014 )"
This reverts commit 0ea2c37263
.
* clang-format
* fix-path
* fix
3 years ago
Craig Tiller
0ea2c37263
Revert "New resource quota integration ( #27643 )" ( #28014 )
...
This reverts commit 39f0877d25
.
3 years ago
Craig Tiller
39f0877d25
New resource quota integration ( #27643 )
...
* new resource quota integration
* Automated change: Fix sanity tests
* fix
* fix
* fixes
* fixes
* fixes
* Automated change: Fix sanity tests
* fixes
* fixes
* Automated change: Fix sanity tests
* fixes
* fix
* fixes
* windows-fix
* fixes
* fixes
* fix
* fix-asan
* banned
* banned
* fixes
* clang-tidy-fix
* Automated change: Fix sanity tests
* fix-cronet
* review feedback
* review feedback
* Automated change: Fix sanity tests
* fixes
* bug fix
* fixes
* compile fix
* exclude megabyte size payloads from 1byte tests
* windows fix
* start moving ios
* keep moving windows
* Get windows compilation working.
* Automated change: Fix sanity tests
* better
* fixes
* remove slice buffer from memory_allocator.h
* Revert "remove slice buffer from memory_allocator.h"
This reverts commit 234a63b6e3
.
* ugh
* #fixtests
* pthread tls fixes
* Automated change: Fix sanity tests
* fixfixfix
* xxx
* add reset
* review feedback
* fix
* fix
* fixes
* fix
* mac progress
* cpp-impl-of
* rename ptr
* Automated change: Fix sanity tests
* memory-owner-is-a-memory-allocator
* fixes
* fix
* fix from prod
* fix
* Fix issue leading to bad pointers being returned on Windows.
* Automated change: Fix sanity tests
* fix multislice bug
* argh
* hyrums law fixes
* hyrums law fixes
* clang-format
* hyrums law fixes
Co-authored-by: ctiller <ctiller@users.noreply.github.com>
3 years ago
Vignesh Babu
879f97ef70
Revert "Api fuzzer extensions to support simulating traffic congestion ( #27820 )" ( #27973 )
...
This reverts commit c1089d2964
.
3 years ago
Vignesh Babu
c1089d2964
Api fuzzer extensions to support simulating traffic congestion ( #27820 )
...
* adding api_fuzzer changes
* adding api fuzzer changes
* updating some typos
* updating api_fuzzer and corpus entries
* updating api_fuzzer to fix crash due to two successive receive_op batches
* adding some reverted fixes to api_fuzzer.cc
* updating api_fuzzer and corpus as per initial comments
* fix some typos
* fix memory leaks and timeout issues
* adding some comments and removing debug strings
* updating api_fuzzer to remove previous edits to always add recv initial metadata ops for client calls
* updating passthru endpoint to account for erroneous initialization when channel effects are not simulated
* tidying up code
3 years ago
Craig Tiller
ea389c00c2
Adjust include order per style guide ( #27175 )
...
Introduce clang-format configuration to sort includes closer to our rules.
3 years ago
AJ Heller
2d16865693
Delete libuv-iomgr implementation and GRPC_UV build option ( #27188 )
...
This has been unmaintained for years, last supported in gRPC-core v1.24.
3 years ago
AJ Heller
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
3 years ago
Craig Tiller
f5d3ed2db1
Revert "Move resource_user ownership into chttp2 transport/server/connector ( #26643 )" ( #27029 )
...
This reverts commit d1935a65a1
. Will be rolled forward with a fix.
3 years ago
AJ Heller
d1935a65a1
Move resource_user ownership into chttp2 transport/server/connector ( #26643 )
3 years ago
Esun Kim
ca945a58e9
Introduced grpc_error_handle ( #25902 )
...
- Define grpc_error_handle
- Replace grpc_error* with grpc_error_handle
4 years ago
AJ Heller
0e3a02e903
Revert "Remove the `urgent` argument from iomgr tcp read API ( #25494 )" ( #25592 )
...
This reverts commit a3398f9
. Justification: see b/181367644. tl;dr: assuming urgent==false does not hold in all situations.
4 years ago
AJ Heller
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 ).
4 years ago
Esun Kim
7d7c3cc711
Update by review
4 years ago
Esun Kim
e0d8c498a6
C++ cast for others
4 years ago
Esun Kim
47dd5fd78e
Fix by modernize-use-bool-literals
4 years ago
Michael Wang
c6586f087f
implemented get_local_address accessor for grpc_endpoint
4 years ago
Mark D. Roth
9404f66f70
Replace most uses of gpr_asprintf() with absl calls.
4 years ago
Yash Tibrewal
3a189d7e08
test changes
5 years ago
Vijay Pai
7aa662b97b
Remove unused-parameter warnings, round 2 (18 of 19)
5 years ago
Soheil Hassas Yeganeh
18b19105f2
Implement TCP_INQ for gRPC in Linux
...
TCP_INQ is a socket option we added to Linux to report pending bytes
on the socket as a control message.
Using TCP_INQ we can accurately decide whether to continue read or not.
Add an urgent parameter, when we do not want to wait for EPOLLIN.
This commit improves the latency of 1 RPC unary (minimal benchmark)
significantly:
Before:
l_50: 61.3584984733
l_90: 94.8328711277
l_99: 126.211351174
l_999: 158.722406029
After:
l_50: 51.3546011488 (-16%)
l_90: 72.3420731581 (-23%)
l_99: 103.280218974 (-18%)
l_999: 130.905689996 (-17%)
6 years ago
Yash Tibrewal
bab043e865
Cleanup
6 years ago
Yash Tibrewal
8d47cd4992
Revert "Revert "Fathom tcp changes""
6 years ago
Jan Tattermusch
97ba943f17
Revert "Fathom tcp changes"
6 years ago
Yash Tibrewal
9c5bde5e4e
More commits
6 years ago
Yash Tibrewal
f0397933b0
Fathom TCP level changes. TracedBuffer for keeping track of all buffers
...
to be traced. Adding tests for Fathom and TracedBuffer. A lot more.
Please read PR description.
6 years ago
Noah Eisen
4d20a66685
Run clang fmt
7 years ago
Noah Eisen
be82e64b3d
Autofix c casts to c++ casts
7 years ago
Jan Tattermusch
12ba4b1e05
make grpc_passthru_endpoint_stats refcounted
7 years ago
Yash Tibrewal
8cf1470a51
Revert "Revert "All instances of exec_ctx being passed around in src/core removed""
7 years ago
Yash Tibrewal
ad4d2dde00
Revert "All instances of exec_ctx being passed around in src/core removed"
7 years ago
Craig Tiller
4ac2b8e585
Enable clang-tidy as a sanity check, fix up all known failures
7 years ago
Yash Tibrewal
34a57d0346
rename all test core files to cc and a lot of C++ style conversions
7 years ago
Yash Tibrewal
0ee7574732
Removing instances of exec_ctx being passed around in functions in
...
src/core. exec_ctx is now a thread_local pointer of type ExecCtx instead of
grpc_exec_ctx which is initialized whenever ExecCtx is instantiated. ExecCtx
also keeps track of the previous exec_ctx so that nesting of exec_ctx is
allowed. This means that there is only one exec_ctx being used at any
time. Also, grpc_exec_ctx_finish is called in the destructor of the
object, and the previous exec_ctx is restored to avoid breaking current
functionality. The code still explicitly calls grpc_exec_ctx_finish
because removing all such instances causes the code to break.
7 years ago
Yuchen Zeng
ab74a8c86e
Add grpc_endpoint_delete_from_pollset_set
7 years ago
Yash Tibrewal
7cdd99c86e
Adding more pointer type conversions
7 years ago
ncteisen
969b46ef73
Add rich closure debug mode
8 years ago
ncteisen
274bbbe6a0
Add rich closure debug mode
8 years ago
Jan Tattermusch
7897ae9308
auto-fix most of licenses
8 years ago
Craig Tiller
ee4b145213
Remove workqueue, covered_by_poller as concepts, get Mac build up
8 years ago
Craig Tiller
339e421b29
Change endpoint interface to declare poller coveredness
8 years ago
Craig Tiller
e38948a1cd
clang-format
8 years ago
Craig Tiller
85bf34a4a8
Copy slices for in-process transports: its no longer safe to just ref
8 years ago