Chiebot-Mirror/grpc - grpc - Gitea: Git with a cup of tea

Commit Graph

Author	SHA1	Message	Date
Jan Tattermusch	0da52ab133	Fix typo in flow control trace	6 years ago
Jan Tattermusch	e3f37b7f43	Merge pull request #17928 from jtattermusch/kokoro_v4_perf_image grpc-performance-kokoro-v4 kokoro perf image changes	6 years ago
apolcyn	9aee1731c9	Merge pull request #17598 from crwilcox/terminate-hostname Add period at end of metadata.google.internal to prevent unnecessary DNS lookups	6 years ago
Nicolas Noble	e06b73cab3	Merge pull request #17930 from grpc/boringssl-ubsan-test Removing BoringSSL-specific ubsan suppressions.	6 years ago
Nicolas Noble	9abc673def	Restore gsec* suppressions These aren't from BoringSSL	6 years ago
Eric Gribkoff	5e50a89502	Merge pull request #17929 from ericgribkoff/abort_test_flake fix flake in test_abort_does_not_leak_local_vars	6 years ago
Nicolas Noble	510fba2deb	Removing BoringSSL-specific ubsan suppressions. Let's see if #17791 is really fixed.	6 years ago
Eric Gribkoff	5a20b60cda	fix flake in test_abort_does_not_leak_local_vars	6 years ago
Jan Tattermusch	d6ca2c9c95	v4 kokoro perf image changes	6 years ago
apolcyn	7e57732a1f	Merge pull request #17912 from apolcyn/java_and_go Unskip google default creds for Go and Java in cloud to prod tests	6 years ago
apolcyn	879d3cb623	Merge pull request #17894 from apolcyn/disable_ares_ios Disable c-ares on iOS	6 years ago
rmstar	5cb668011b	Merge pull request #17643 from rmstar/bad_conn Bad connection test	6 years ago
Vijay Pai	588d33d1df	Merge pull request #17925 from vjpai/oops Fix a test broken for 3 years....	6 years ago
Noah Eisen	ed0a565583	Merge pull request #17907 from ncteisen/build-madness Fix Internal Build	6 years ago
Yash Tibrewal	e3dcfbe571	Merge pull request #17923 from grpc/revert-17916-revert-17896-17338fix Revert "Revert "Fix for 17338. Delay shutdown of buffer list till tcp_free to avoid races""	6 years ago
Vijay Pai	b23abe832c	GPR_ARRAY_SIZE is meant for arrays	6 years ago
Eric Gribkoff	96403bc640	Merge pull request #17899 from ericgribkoff/undead_server python: do not store raised exception in _Context.abort()	6 years ago
Yash Tibrewal	cc19a55338	Revert "Revert "Fix for 17338. Delay shutdown of buffer list till tcp_free to avoid races""	6 years ago
Soheil Hassas Yeganeh	872ca1f8a5	Merge pull request #17823 from soheilhy/fd-cache Track the pollsets of an FD in PO_MULTI mode for pollex.	6 years ago
Vijay Pai	8026f552fd	Merge pull request #17918 from yashykt/dummyshutdownunref Dummy Shutdown should still unref the error	6 years ago
Vijay Pai	8bf6f02e0e	Merge pull request #17917 from vjpai/fix_delist C++ callback server: request spawning limit is supposed to be per-server, not per-method	6 years ago
Soheil Hassas Yeganeh	e83e463b5a	Track the pollsets of an FD in PO_MULTI mode for pollex. Each pollset in pollex has a lock, grabbed upon adding an FD to the pollset. Since this is called on a per-call basis, there is a flat array caching the FDs of the pollset, to avoid unnecessarily calling epoll_ctl multiple times for the same FD. This has two problems: 1) When multiple threads add FDs to the same pollset, we will have contention on the pollset lock. 2) When we have many FDs we simply run out of cache storage, and call epoll_ctl(). This commit changes the caching strategy by simply storing the pollsets of an FD inside that FD, when we are in PO_MULTI mode. This results in address in both (1) and (2). Moreover, this commit fixes another performance bug. When we have a release FD callback, we do not call close(). That FD will remain in our epollset, until the new owner of the FD actually call close(). This results in a lot of spurious wake ups when we simply hand off gRPC FDs to other FDs.	6 years ago
Yash Tibrewal	f37e18b8fd	Dummy Shutdown should still unref the error	6 years ago
Vijay Pai	92d37b1273	SOFT_MAXIMUM is supposed to be per-server, not per-method	6 years ago
Vijay Pai	023a6b5c8e	Merge pull request #17916 from grpc/revert-17896-17338fix Revert "Fix for 17338. Delay shutdown of buffer list till tcp_free to avoid races"	6 years ago
Vijay Pai	95fe85f090	Revert "Fix for 17338. Delay shutdown of buffer list till tcp_free to avoid races"	6 years ago
Vijay Pai	977df7208a	Merge pull request #17909 from vjpai/delist Replace list of outstanding callback requests with count	6 years ago
Soheil Hassas Yeganeh	4617ecd879	Merge pull request #17908 from soheilhy/tcp-posix-tsan Use full_fetch_add for ref counting the backup poller.	6 years ago
yihuaz	ec7dfe41f7	Merge pull request #17664 from yihuazhang/SSL_ROOT_DUPLICATE Allow duplicates in ssl root certificate list.	6 years ago
Alexander Polcyn	c0125a7cd4	Unskip google default creds for Go and Java in cloud to prod tests	6 years ago
Prashant Jaikumar	6b19927bc4	Bad connection test	6 years ago
Eric Gribkoff	28252eb0dd	force gc in test	6 years ago
ncteisen	82171553cf	clang fmt	6 years ago
Alexander Polcyn	fd185cd1ea	Disable c-ares on iOS	6 years ago
Vijay Pai	8521c0394b	Address optional reviewer comments	6 years ago
Juanli Shen	37c7c1217d	Merge pull request #17883 from AspirinSJL/fix_hcc_segv Unref watcher after releasing lock	6 years ago
Yash Tibrewal	3db446f398	Merge pull request #17896 from yashykt/17338fix Fix for 17338. Delay shutdown of buffer list till tcp_free to avoid races	6 years ago
Soheil Hassas Yeganeh	bdd02e5852	Merge pull request #17439 from soheilhy/worktree-slice-align Reorder fields in slice to share the same bytes for length fields.	6 years ago
Soheil Hassas Yeganeh	3492539b32	Use full_fetch_add for ref counting the backup poller. There were spurious TSAN errors on PR #17823 because TSAN doesn't really understand how `g_uncovered_notifications_pending` works. It's odd in the sense that we destroy the backup poller, when the ref count reaches 1 (instead of 0 which is commonly used). Prior to PR #17823, TSAN doesn't complain because we (unnecessarily) always grab the pollset's lock, which TSAN understands. This commit uses full_fetch_add to explain the synchronization primitive to TSAN.	6 years ago
Yihua Zhang	09cd07cfa0	revision 1	6 years ago
Vijay Pai	e56c832c0d	Replace list of outstanding callback requests with count only	6 years ago
ncteisen	fa74259769	Reintroduce commented debugging tips	6 years ago
ncteisen	632aa8125f	reintroduce anon namespace	6 years ago
Soheil Hassas Yeganeh	af4687bd4d	Merge pull request #17824 from soheilhy/worktree-weak Introduce weak and nonline attribute.	6 years ago
ncteisen	c215e8e359	Fix internal build	6 years ago
Lidi Zheng	15bb25ae54	Merge pull request #17798 from lidizheng/bazel-install Install Bazel using its installation script	6 years ago
Eric Gribkoff	c5255e9a5e	python: do not store raised exception in _Context.abort() Python 3 exceptions include a `__traceback__` attribute that includes refs to all local variables. Saving the exception results in leaking references to the, among other things, the Cython grpc_call wrapper and prevents garbage collection and release of core resources, even after the server is shutdown. See https://www.python.org/dev/peps/pep-3134/#open-issue-garbage-collection	6 years ago
Muxi Yan	343c2286af	Merge pull request #17890 from grpc/jtattermusch-patch-2 Increase VM timeout for grpc_ios_binary_size job	6 years ago
Soheil Hassas Yeganeh	ef20840174	Reorder fields in slice to share the same bytes for length fields. Ref-counted and inlined slices both have a lenght value, but they are not in the same byte offset. As a result, we will have two different loads based on a branch. Instead move them to the same byteoffset, so that we will have one move and the same number of branches. Difference can be seen on: https://godbolt.org/z/kqFZcP	6 years ago
Vijay Pai	cc392e89bc	Merge pull request #17901 from vjpai/unlock_cq Remove an overly-conservative mutex from callback CQ implementation	6 years ago

1 2 3 4 5 ...

36826 Commits (0da52ab133b7fd6e9ae2fbd7518af25ce4d8e004) All Branches Search

36826 Commits (0da52ab133b7fd6e9ae2fbd7518af25ce4d8e004)

All Branches