aboutsummaryrefslogtreecommitdiffhomepage
path: root/src/core/lib/transport
Commit message (Collapse)AuthorAge
* * Fixed issue(17563) "Freeing heap block containing an active critical ↵Gravatar xtao2018-12-22
| | | | section." reported by Application Verifier on Windows.
* Merge pull request #17313 from yashykt/padstreamsizeGravatar Yash Tibrewal2018-11-28
|\ | | | | Stream size needs to be rounded up to alignment
| * Add required headerGravatar Yash Tibrewal2018-11-28
| |
| * Stream size needs to be rounded up to alignmentGravatar Yash Tibrewal2018-11-27
| |
* | Add traced information to stream opGravatar Yash Tibrewal2018-11-27
|/
* Run the tools/codegen/core/gen_static_metadata.py and generate theGravatar Vishal Powar2018-11-16
| | | | required files
* Remeve memset(0) from arena allocated memory.Gravatar Soheil Hassas Yeganeh2018-11-05
| | | | | | | Callers are updated to properly initialize the memory. This behavior can be overridden using GRPC_ARENA_INIT_STRATEGY environment variable.
* Client-side health checking support.Gravatar Mark D. Roth2018-10-24
|
* Optimize error handling for special cases.Gravatar Soheil Hassas Yeganeh2018-10-18
| | | | | | | | | | | | | | This commit contains a few improvements: 1. Using a consequetive range of [0..4], will allow us to merge all branches of error_is_special into one comparison. 2. With (1), we can remove the for loops to find entries in error_status_map with a single O(1) lookup. 3. grpc_error_is_special() code paths should be inlined for ref and unref to avoid callq for the majority of cases where speical error is used. 4. grpc_error_get_int() should never accept a nullptr argument to avoid an expensive branch in the hot path. Callers should all allocate a dummy int on the stack when calling.
* Merge pull request #16801 from soheilhy/worktree-sliceGravatar Soheil Hassas Yeganeh2018-10-15
|\ | | | | Avoid unnecessary ref/unref calls to get mdelem from slices.
| * Use grpc_mdelem_create() directly.Gravatar Soheil Hassas Yeganeh2018-10-14
| | | | | | | | | | Remove grpc_mdelem_from_slices_no_unref() since it's a wrapper around grpc_mdelem_create().
| * Avoid unnecessary ref/unref calls to get mdelem from slices.Gravatar Soheil Hassas Yeganeh2018-10-07
| | | | | | | | | | | | | | | | | | grpc_mdelem_from_slices() unref's the key and value. As a result, in quite a few cases on the hot path, we first ref slice, so that grpc_mdelem_from_slices() can unref them. Add grpc_mdelem_from_slices_no_unref() which does not unref() the input slices. This cuts 0.5% - 1.0% across app benchmarks.
* | Make link_head consistent with link_tail.Gravatar Soheil Hassas Yeganeh2018-10-06
|/ | | | | | This is a tiny change I note while removing memset() calls. link_head doesn't set the reserved pointer while link_tail (correctly) does.
* Delete hpack lookup tableGravatar Hope Casey-Allen2018-09-28
|
* Merge pull request #16663 from hcaseyal/delete_unecessary_callout_checksGravatar hcaseyal2018-09-20
|\ | | | | Remove unnecessary callout_is_default checks
| * Add in port_platform.h. The permanent fix for this is currently in a pending ↵Gravatar Hope Casey-Allen2018-09-19
| | | | | | | | PR that should get merged soon
| * Remove unnecessary callout_is_default checksGravatar Hope Casey-Allen2018-09-19
| |
* | Some reorganizing to respond to code review commentsGravatar Hope Casey-Allen2018-09-19
| |
* | Even cleaner implementation that does not modify the grpc_mdelem struct.Gravatar Hope Casey-Allen2018-09-12
| |
* | Clang tidyGravatar Hope Casey-Allen2018-09-11
| |
* | Small formatting and style changesGravatar Hope Casey-Allen2018-09-11
| |
* | Fix codegen script to include port_platformGravatar Hope Casey-Allen2018-09-10
| |
* | Much cleaner approach that should address code review comments as wellGravatar Hope Casey-Allen2018-09-10
| |
* | WIP. Modifying grpc_mdelem to store the static hpack table idnexGravatar Hope Casey-Allen2018-09-10
| |
* | Responding to code review commentsGravatar Hope Casey-Allen2018-09-10
| |
* | Temporary workaround for memory leak in debugging functionGravatar Hope Casey-Allen2018-09-09
| |
* | Fix memory leak in debugging functionGravatar Hope Casey-Allen2018-09-09
| |
* | Clang tidyGravatar Hope Casey-Allen2018-09-08
| |
* | Merge branch 'hpack_static_table' of github.com:hcaseyal/grpc into ↵Gravatar Hope Casey-Allen2018-09-08
|\ \ | | | | | | | | | hpack_static_table
* | | Fix objective c unit test and debugging functionsGravatar Hope Casey-Allen2018-09-08
| | |
| * | Clang tidyGravatar Hope Casey-Allen2018-09-08
|/ /
* | WIP. Setting md_index to 0 on grpc_linked_mdelem creationGravatar Hope Casey-Allen2018-09-08
| |
* | Fix assertion in debugging codeGravatar Hope Casey-Allen2018-09-06
| |
* | Fix internal_data to be the same size as the new grpc_linked_mdelemGravatar Hope Casey-Allen2018-09-06
| |
* | WIP. Handling error conditions such as duplicate metadata, metadata size ↵Gravatar Hope Casey-Allen2018-09-06
| | | | | | | | limit, and default count. Clang tidy.
* | WIP. Created static data for size of metadata and callouts indexGravatar Hope Casey-Allen2018-09-04
| |
* | Add ability to use static hpack table on the encode pathGravatar Hope Casey-Allen2018-08-28
|/
* size_t shouldn't have the value -1; switch to intGravatar Vijay Pai2018-08-21
|
* Add experimental API for resetting connection backoff.Gravatar Mark D. Roth2018-08-06
|
* Revert "Merge pull request #15746 from ↵Gravatar Mark D. Roth2018-06-14
| | | | | | | grpc/revert-15709-recv_trailing_metadata_ready2" This reverts commit 3f9308ce1f8cb42c96901c1700f0b9dbb531f186, reversing changes made to 92a0ae0b1081840d2c5a488f66bf6550c1a492f4.
* Revert "Second attempt: move recv_trailing_metadata into its own callback, ↵Gravatar Mark D. Roth2018-06-13
| | | | don't use on_complete for recv_ops"
* Revert "Revert "move recv_trailing_metadata into its own callback, don't use ↵Gravatar Mark D. Roth2018-06-11
| | | | | | on_complete for recv_ops"" This reverts commit f3715134458cb14efd855d948f229dc2661b4028.
* Revert "move recv_trailing_metadata into its own callback, don't use ↵Gravatar Mark D. Roth2018-06-08
| | | | on_complete for recv_ops"
* Move recv_trailing_metadata into its own callback. Don't useGravatar Mark D. Roth2018-06-08
| | | | on_complete for recv_ops.
* Merge pull request #15649 from markdroth/avoid_peer_string_allocGravatar Mark D. Roth2018-06-07
|\ | | | | Avoid allocating the peer string returned from the transport.
* \ Merge pull request #15069 from muxi/prototype-cfstreamGravatar Muxi Yan2018-06-07
|\ \ | | | | | | Implementation of CFStream
| | * Avoid allocating the peer string returned from the transport.Gravatar Mark D. Roth2018-06-05
| |/ |/|
| * Address comments on build systemsGravatar Muxi Yan2018-05-31
| |
| * Address comments on names and commentsGravatar Muxi Yan2018-05-30
| |
* | Merge pull request #14894 from sreecha/fix-timeGravatar Sree Kuchibhotla2018-05-22
|\ \ | | | | | | Fix grpc_millis type (timers are broken on 32-bit systems otherwise)
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-29 21:03:48 +0000