summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2022-02-03BGP: Static global linpools replaced by private linpoolsMaria Matejka
2022-02-03Socket cork fixesMaria Matejka
2022-02-03Asynchronous neighbor notifications fixesMaria Matejka
2022-02-03Explicitly storing and checking loop information in socketsMaria Matejka
2022-02-03Neighbors: Locking and asynchronous notificationsMaria Matejka
2022-02-03Global interface list renamed to not clash with local listsMaria Matejka
2022-02-03Interfaces and neighbor notifications do properly enter protocol loopsMaria Matejka
2022-02-03Page cleanup routines updated to allow nested birdloopsMaria Matejka
2022-02-03Route table fast prune fixupMaria Matejka
2022-02-03Babel: fixed interface pool removalMaria Matejka
2022-02-03Dropping a nonsensical assert which broke BabelMaria Matejka
2022-02-03Removing periodic wakeup of mainloop in debug version to search for ↵Maria Matejka
notification issues
2022-02-03Performance data of sark vs. v2.0.8 with a chart generatorMaria Matejka
2021-12-22Thread documentation: Chapter 4 on memory managementMaria Matejka
2021-12-08Thread documentation: Completely rewritten chapter 3 on loops and locksMaria Matejka
2021-12-08Final version of asynchronous export documentationMaria Matejka
2021-12-08Thread documentation: chapter 3, coroutines and lockingMaria Matejka
2021-12-08Thread documentation: chapters 0, 1 and 2Maria Matejka
2021-12-07Fixed build errors for OpenBSDMaria Jan Matejka
2021-12-07Fixed standby memory page counters on shutdownMaria Matejka
Bug introduced by commit 38278d94ba0a179d5eeb061a59850a4e1c150e5b.
2021-12-07GDB: io loop printing support in resource dumps, better iterating over ↵Maria Matejka
linpool chunks
2021-12-07Fixed a race condition in channel aux table cleanupMaria Matejka
2021-12-07Event list asserts to prevent running into rotten dataMaria Matejka
2021-12-07Using more Python-ish constructions in BIRD linked-list accessorsMaria Matejka
2021-12-07Main IO loop shouldn't skip sockets when ping is receivedMaria Matejka
2021-12-02Standby memory pages are accounted for as overhead in show memory commandMaria Matejka
2021-12-01No memory unmapping when shutting downMaria Matejka
All the memory is just freed implicitly on exit, no need for page-by-page unmapping.
2021-12-01Faster shutdown and cleanups by freeing route attributes strictly from main loopMaria Matejka
2021-12-01Faster prune on table deletionMaria Matejka
2021-12-01Table prune routines request export announcements directlyMaria Matejka
2021-12-01fixed duplicate routes propagationMaria Matejka
2021-12-01Merge branch 'master' into HEADMaria Matejka
2021-12-01Fix of shutdown: premature log cleanup led to use-after-freeMaria Matejka
2021-12-01Page allocator moved from pools to IO loops.Maria Matejka
The resource pool system is highly hierarchical and keeping spare pages in pools leads to unnecessarily complex memory management. Loops have a flat hiearchy, at least for now, and it is therefore much easier to keep care of pages, especially in cases of excessive virtual memory fragmentation.
2021-11-30For safer memory allocations, resources are bound to loops.Maria Matejka
Also all loops have their basic resource pool for allocations which are auto-freed when the loop is stopping.
2021-11-27Memory statistics split into Effective and OverheadMaria Matejka
This feature is intended mostly for checking that BIRD's allocation strategies don't consume much memory space. There are some cases where withdrawing routes in a specific order lead to memory fragmentation and this output should give the user at least a notion of how much memory is actually used for data storage and how much memory is "just allocated" or used for overhead. Also raising the "system allocator overhead estimation" from 8 to 16 bytes; it is probably even more. I've found 16 as a local minimum in best scenarios among reachable machines. I couldn't find any reasonable method to estimate this value when BIRD starts up. This commit also fixes the inaccurate computation of memory overhead for slabs where the "system allocater overhead estimation" was improperly added to the size of mmap-ed memory.
2021-11-24fixup of table reload bugMaria Matejka
2021-11-24Directly mapped pages are kept for future use if temporarily not neededMaria Matejka
2021-11-24GDB pretty printer for resource poolsMaria Matejka
2021-11-23Fixed pipe reload/refeedMaria Matejka
2021-11-23Stored pages release routineMaria Matejka
2021-11-23Fixed channel export map confusionMaria Matejka
2021-11-22BGP: End route refresh before another startsMaria Matejka
2021-11-22RPKI has its own loopMaria Matejka
2021-11-22RPKI: Use the route refresh mechanism also for the first loadMaria Matejka
2021-11-22More information on RPKI errorsMaria Matejka
2021-11-22Higher settle times when route refresh in the source table is runningMaria Matejka
2021-11-22Corking also feed start to keep BIRD running when refeeds would easily cause ↵Maria Matejka
congestion
2021-11-22Route attribute cache is now lockless on read / clone.Maria Matejka
Lots of time was spent locking when accessing route attribute cache. This overhead should be now reduced to a minimum.
2021-11-22Route sources have their separate global lockMaria Matejka
generated by cgit v1.2.3 (git 2.39.1) at 2025-01-04 07:52:28 +0000