ydb/library/cpp/threading/bounded_queue, branch main Mirror of YDB github repos https://code.mastervirt.ru/ydb/atom?h=main 2023-07-28T03:50:19Z Intermediate changes 2023-07-28T03:50:19Z robot-piglet robot-piglet@yandex-team.com 2023-07-28T03:50:19Z urn:sha1:85ef4ee49c3edbb700d0ef903d01177bf9984018 revert rXXXXXX (see discusstion in pr), will commit again more pci-dss friendly way 2023-07-27T09:28:50Z kulikov kulikov@yandex-team.com 2023-07-27T09:28:50Z urn:sha1:4a7691c519e6114e013dc1dd0c3b2528154507f9 try to get rid of locks and allocations for elastic queue thread pool 2023-07-21T10:59:33Z kulikov kulikov@yandex-team.com 2023-07-21T10:59:33Z urn:sha1:5706cb392271ea40eab053314e7c0f4d9d4547ba In case of heavy load and high rps current thread pool implementation seems to have problems at least with contention on lock inside condvar (long futex wait calls from http server listener thread), so try to implement something more efficient: - replace condvar with TEventCounter implementation without internal lock (pthread condvar maintains waiters wakeup order, thread pool doesn't need it); - introduce well-known bounded mpmc queue over ring buffer; - get rid of TDecrementingWrapper; - add options to turn on new pool in library/cpp/http/server and search/daemons (will remove after adoption); - make elastic queue ut check both versions; - workaround problems with android/arm build targets.