Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | Add benchmark suite for chttp2 | 2017-03-21 | |
| | |||
* | Add debug macros | 2017-02-09 | |
| | |||
* | Make combiners refcounted, to facilitate sharing | 2017-02-09 | |
| | |||
* | Revert "Revert "Metadata handling rewrite"" | 2017-01-23 | |
| | | | | This reverts commit 5e01e2ac977655aa074faf7fde0a74298f5e4c55. | ||
* | Revert "Metadata handling rewrite" | 2017-01-20 | |
| | |||
* | Uncover the badness | 2017-01-12 | |
| | |||
* | Changes to exec_ctx/closure/combiner/workqueue interfaces | 2016-12-28 | |
| | | | | | | | | - make closures know where they should be executed (eg, on a workqueue, or a combiner, or on an exec_ctx) - this allows removal of a large number of trampoline functions that were appearing whenever we used combiners, and should allow for a much easier interface to combiner locks | ||
* | Fix logic race in chttp2 write path | 2016-11-16 | |
| | | | | | | | | | | | | IF: - we schedule a write in chttp2 in response to some stream op (which will cause a write that's covered by a poller to be scheduled on the combiner lock) - AND then, under that same combiner lock, we process a RST_STREAM - then we'll remove the op that's being processed, consequently removing the polling coverage - and then, IF that is the last poll on said transport, the transport will never write, which CAN cause servers to fail to shutdown | ||
* | Better readability | 2016-10-10 | |
| | |||
* | Cleanup | 2016-10-10 | |
| | |||
* | Fix resource leak | 2016-09-12 | |
| | |||
* | Tie workqueue implementation to event engine | 2016-09-09 | |
| | |||
* | Minor perf improvements | 2016-09-08 | |
| | |||
* | Get write batching working again | 2016-09-07 | |
| | |||
* | fixes | 2016-09-02 | |
| | |||
* | Merge branch 'who-combines-the-combiners' into direct-calls | 2016-09-02 | |
|\ | |||
| * | Merge branch 'grand-unified-closures' into who-combines-the-combiners | 2016-09-02 | |
| |\ | |||
| | * | Address review comments | 2016-09-02 | |
| | | | |||
* | | | Reinstate RST_STREAM at EOS | 2016-09-01 | |
| | | | |||
* | | | Note polling coverage when taking combiner locks: resolves offload issues | 2016-08-31 | |
|/ / | |||
* | | Merge combiner and exec_ctx execution better | 2016-08-24 | |
| | | | | | | | | | | | | | | Allows exec_ctx callbacks to be called while a combiner is executing. Also allows guaranteeing direct execution of callbacks from combiners, which should allow reducing cpu burn for up/down stack interactions in the future. | ||
* | | Timing nuances | 2016-08-09 | |
| | | |||
* | | Fixup compilation | 2016-08-04 | |
|/ | |||
* | Fix tsan reported error in trace | 2016-08-04 | |
| | |||
* | Force queuing to workqueue instead of relying on exec_ctx | 2016-07-19 | |
| | |||
* | Get traces around everything | 2016-07-18 | |
| | |||
* | Fixes & debug | 2016-07-18 | |
| | |||
* | Add timing data | 2016-07-14 | |
| | |||
* | Refs | 2016-07-13 | |
| | |||
* | Fix some refcounting bugs | 2016-07-11 | |
| | |||
* | Debugging | 2016-07-11 | |
| | |||
* | Progress to converting chttp2 to combiner locks | 2016-07-11 | |
| | |||
* | Afford other threads the opportunity to insert into a combiner prior to the ↵ | 2016-07-10 | |
| | | | | final list being executed but after the async pause | ||
* | Progress on a finalization list | 2016-07-08 | |
| | |||
* | Rename async_execution_lock --> combiner, make it use closures, unroll ↵ | 2016-07-07 | |
previous changes for idleness |