Maria Matejka
a861727d98
Split route data structure to storage (ro) / manipulation (rw) structures.
...
This should help a lot with keeping the route machinery and allocations
clean. This commit also changes behaviour of rte_update() with cached
rta. Newly, calling rte_update() keeps the number of rta references.
2020-04-30 17:38:30 +02:00
Maria Matejka
1967fe4fa6
Route: moved rte_src pointer from rta to rte
...
It is an auxiliary key in the routing table, not a route attribute.
2020-04-30 16:15:59 +02:00
Maria Matejka
52b6e35251
Route export: Common export function.
...
Also:
* Feeding request by net.
* Passing only net_addr in export structure.
* Cleanup of krt_replace_rte() call.
2020-04-30 16:15:59 +02:00
Maria Matejka
eb5298f3e1
Internal API change: passing all the rt_notify() data as a single structure.
2020-04-30 16:14:26 +02:00
Maria Matejka
1fd32f9562
Preference moved to RTA and set explicitly in protocols
2020-04-30 16:14:26 +02:00
Maria Matejka
86dbe0980c
Route import API redefinition.
...
Hidden rte_get_temp() into rte_update().
Split rte_update() / rte_withdraw().
2020-04-30 16:14:25 +02:00
Maria Matejka
dc042d87cb
Perf: changed route update pattern to be more like common protocols
2020-03-12 09:26:05 +01:00
Ondrej Zajicek (work)
759b204be3
Lib: Support for 64-bit numbers in bvsnprintf()
...
Use 'l' for s64/u64 instead of for long/ulong, as that is much more
useful. Also make number() correct with regard to signed/unsigned
typecasts.
2019-10-09 17:53:23 +02:00
Maria Matejka
6dda6931d1
Perf: allow testing with cached route attributes.
2019-10-09 17:47:14 +02:00
Maria Matejka
2f02c25e36
Perf: fixed stupid allocation bug
2019-03-15 16:34:26 +01:00
Maria Matejka
f9b97f1c62
Perf: Added forgotten all-protocol options
2019-02-01 14:09:01 +01:00
Maria Matejka
e85e37d91d
Perf: Prune the table after every loop to have clean state.
2019-01-31 15:05:40 +01:00
Maria Matejka
7411b694c3
Perf: Write also BIRD version to have all the needed data in the logfile
2019-01-31 15:03:43 +01:00
Maria Matějka
82b742533b
Perf: Protocol to measure BIRD performance internally
...
This protocol is highly experimental and nobody should use it in
production. Anyway it may help you getting some insight into what eats
so much time in filter processing.
2018-12-18 15:08:31 +01:00
