Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | Don't assert if we're not able to set affinity for some reason | Vijay Pai | 2016-07-30 |
| | |||
* | Fix compile error | Craig Tiller | 2016-07-15 |
| | |||
* | Merge branch 'histo' of github.com:vjpai/grpc into delayed-write | Craig Tiller | 2016-07-15 |
|\ | |||
* | | Cleanup from code review | Craig Tiller | 2016-07-14 |
| | | |||
| * | clang-format | Vijay Pai | 2016-07-13 |
| | | |||
| * | Resolve pernicious race between destructor and thread functions by insisting ↵ | Vijay Pai | 2016-07-13 |
| | | | | | | | | that destructor is invoked after the class has gone back to being a harmless single-threaded thing. | ||
* | | Fix some failures | Craig Tiller | 2016-07-13 |
| | | |||
| * | Temporary workaround by setting up a deadline on server shutdown. | Vijay Pai | 2016-07-13 |
| | | | | | | | | | | This reveals an issue that needs to be solved: D0713 14:48:28.049861213 14503 server.c:704] Waiting for 44 channels and 0/1 listeners to be destroyed before shutting down server | ||
| * | Unify and make consistent the per-thread shutdown process | Vijay Pai | 2016-07-13 |
| | | |||
* | | Dont stall shutting down server | Craig Tiller | 2016-07-13 |
| | | |||
* | | get timeouts working on run_tests.py again | Craig Tiller | 2016-07-13 |
| | | |||
* | | Better fine-tuning | Craig Tiller | 2016-07-12 |
| | | |||
* | | Get json_run_localhost working again | Craig Tiller | 2016-07-12 |
| | | |||
* | | Get json_run_localhost working again | Craig Tiller | 2016-07-12 |
| | | |||
* | | Actually run json_run_localhost tests | Craig Tiller | 2016-07-12 |
| | | |||
* | | Make request freelists per cq, to allow drastically higher requested call counts | Craig Tiller | 2016-07-12 |
| | | |||
* | | Ensure timely response to marks | Craig Tiller | 2016-07-12 |
| | | |||
| * | Fix some shutdown errors related to CQ/join ordering | Vijay Pai | 2016-07-08 |
| | | |||
| * | Merge branch 'verbose_errors' into histo | Vijay Pai | 2016-07-08 |
| |\ | |||
| * | | Stop holding histogram for a long time | Vijay Pai | 2016-07-08 |
| | | | |||
* | | | Fix high seqno tests, up runtimes for benchmark tests to flush more bugs | Craig Tiller | 2016-07-07 |
|/ / | |||
| * | Add more information so that we can have a meaningful exit code | Vijay Pai | 2016-07-07 |
| | | |||
| * | Reduce assertions, use status codes, increase verbosity on errors | vjpai | 2016-07-06 |
|/ | |||
* | Colocate calls for the same channel on the same CQ | Craig Tiller | 2016-06-27 |
| | |||
* | Merge branch 'master' into wheezy | Vijay Pai | 2016-06-17 |
|\ | |||
| * | Merge pull request #6838 from vjpai/trigger_happy | Jan Tattermusch | 2016-06-16 |
| |\ | | | | | | | Revert unneeded change in qps worker | ||
* | | | Include <cinttypes> for PRIuPTR. Not clear why this is needed. | Vijay Pai | 2016-06-16 |
| | | | |||
* | | | Merge branch 'master' into wheezy | Vijay Pai | 2016-06-15 |
|\| | | |||
* | | | Specify template parent explicitly | Vijay Pai | 2016-06-15 |
| | | | |||
| * | | Fix gpr_log format mismatches in c++ tests | Yuchen Zeng | 2016-06-10 |
| | | | |||
| * | | Factor out json seriailzation code and move it into parse_json. | David Klempner | 2016-06-08 |
|/ / | | | | | | | | | | | This is for the same reasons as ParseJson, that is so that we can limit the scope of the proto namespace differences between internal and external. | ||
| * | I was trigger-happy with changes to qps_worker in the hope of improving | vjpai | 2016-06-08 |
|/ | | | | | | stability. I believe that this change was unneeded and actually may hurt matters by holding a lock for too long. Partially undoes #5444 , particular commit e7042b5 | ||
* | merge and resolve conflicts | yang-g | 2016-06-03 |
|\ | |||
* | | Make grpc++ independent of protobuf | yang-g | 2016-06-03 |
| | | |||
| * | Remove unused gflags and gtest includes | vjpai | 2016-05-31 |
| | | |||
| * | Deprecate perf_db use | vjpai | 2016-05-31 |
|/ | |||
* | Speed up tests | Craig Tiller | 2016-05-23 |
| | |||
* | Cleanup redundant tests | Craig Tiller | 2016-05-23 |
| | |||
* | Fix timeout on async server shutdown | Craig Tiller | 2016-05-23 |
| | | | | | Server was continuing to make requests for new calls forever, which were starving out the shutdown sequence. Change order and win. | ||
* | Merge branch 'guess_cpu' into test_affine | Craig Tiller | 2016-05-20 |
|\ | |||
| * | Better guesses at benchmarks turned unit tests cpu cost | Craig Tiller | 2016-05-20 |
| | | |||
* | | Better testing | Craig Tiller | 2016-05-16 |
|/ | |||
* | clang-format | Craig Tiller | 2016-05-13 |
| | |||
* | Make qps workers connect before even warmup stage begins | Craig Tiller | 2016-05-13 |
| | |||
* | fix formatting | Jan Tattermusch | 2016-05-11 |
| | |||
* | add support for scenario categories | Jan Tattermusch | 2016-05-05 |
| | |||
* | Merge pull request #6340 from ctiller/fail_fast_breaks_qps | Jan Tattermusch | 2016-05-02 |
|\ | | | | | Disable fail fast for qps driver | ||
* | | clang-format, make code more self-documenting | Craig Tiller | 2016-04-29 |
| | | |||
* | | Add a shard channel arg to client channel construction | Craig Tiller | 2016-04-28 |
| | | | | | | | | | | | | C core automatically shares subchannels between channels. Therefore our multiple channel performance tests were really testing single channel performance. | ||
| * | Disable fail fast for qps driver | Craig Tiller | 2016-04-28 |
|/ |