]>
git.ipfire.org Git - thirdparty/bird.git/log
Igor Putovny [Thu, 23 Jan 2025 16:12:01 +0000 (17:12 +0100)]
Move code around
Igor Putovny [Fri, 17 Jan 2025 14:10:24 +0000 (15:10 +0100)]
Implement removing prefix from the trie
Igor Putovny [Thu, 16 Jan 2025 13:04:59 +0000 (14:04 +0100)]
Extend logging functions
Igor Putovny [Wed, 15 Jan 2025 16:08:53 +0000 (17:08 +0100)]
Increase readability of old code
Igor Putovny [Mon, 13 Jan 2025 15:41:14 +0000 (16:41 +0100)]
Add comments
Igor Putovny [Mon, 16 Dec 2024 12:46:11 +0000 (13:46 +0100)]
Bugfix
Igor Putovny [Thu, 12 Dec 2024 13:51:43 +0000 (14:51 +0100)]
Recycle bucket ID when bucket is empty
Igor Putovny [Wed, 11 Dec 2024 15:42:28 +0000 (16:42 +0100)]
Set px_origin where it makes sense
Igor Putovny [Wed, 11 Dec 2024 15:32:10 +0000 (16:32 +0100)]
Delete parent pointer when removing node from trie
Igor Putovny [Wed, 11 Dec 2024 14:14:42 +0000 (15:14 +0100)]
Leaves don't have to be IN_FIB since we are not removing NON_FIB leaves anymore
Igor Putovny [Mon, 9 Dec 2024 17:28:28 +0000 (18:28 +0100)]
Small changes
Igor Putovny [Mon, 9 Dec 2024 17:06:15 +0000 (18:06 +0100)]
Incorporate updates into trie
Igor Putovny [Fri, 6 Dec 2024 17:31:23 +0000 (18:31 +0100)]
Implement processing of incremental updates
Igor Putovny [Fri, 6 Dec 2024 15:01:55 +0000 (16:01 +0100)]
Add prefix origin
Igor Putovny [Thu, 5 Dec 2024 15:03:48 +0000 (16:03 +0100)]
Trie dump now contains all node information
Igor Putovny [Mon, 2 Dec 2024 14:15:19 +0000 (15:15 +0100)]
Small additions
Igor Putovny [Mon, 2 Dec 2024 14:04:30 +0000 (15:04 +0100)]
Add dumping trie
Igor Putovny [Mon, 2 Dec 2024 12:56:09 +0000 (13:56 +0100)]
Process incremental updates in rt_notify()
Igor Putovny [Fri, 29 Nov 2024 16:44:49 +0000 (17:44 +0100)]
Implement function to process incremental update
Igor Putovny [Fri, 29 Nov 2024 14:00:35 +0000 (15:00 +0100)]
Rewrite function for allocating bucket IDs
Igor Putovny [Mon, 25 Nov 2024 15:51:15 +0000 (16:51 +0100)]
Root has a depth 0 instead of 1
Igor Putovny [Wed, 20 Nov 2024 12:42:11 +0000 (13:42 +0100)]
Do not delete potential buckets in the third pass
Igor Putovny [Mon, 18 Nov 2024 14:41:05 +0000 (15:41 +0100)]
Rename bucket to original_bucket
Igor Putovny [Mon, 18 Nov 2024 14:04:50 +0000 (15:04 +0100)]
choose_lowest_id_bucket() now returns the bucket, it does not assign it
Igor Putovny [Mon, 18 Nov 2024 13:42:22 +0000 (14:42 +0100)]
Check ancestors after aggregation
Igor Putovny [Fri, 15 Nov 2024 16:35:51 +0000 (17:35 +0100)]
Small changes
Igor Putovny [Fri, 11 Oct 2024 14:12:18 +0000 (16:12 +0200)]
Cleanup
Igor Putovny [Fri, 11 Oct 2024 12:41:49 +0000 (14:41 +0200)]
Don't assign selected bucket to imaginary node in the third pass
Igor Putovny [Fri, 11 Oct 2024 12:40:35 +0000 (14:40 +0200)]
Silence unused variable warning
Igor Putovny [Fri, 11 Oct 2024 12:36:52 +0000 (14:36 +0200)]
Add more assertions
Igor Putovny [Wed, 9 Oct 2024 15:46:28 +0000 (17:46 +0200)]
Create function to choose bucket with the lowest ID from the set of potential buckets
Igor Putovny [Mon, 7 Oct 2024 15:02:01 +0000 (17:02 +0200)]
trie_insert_prefix() now allows updating a bucket of existing prefix
Igor Putovny [Mon, 7 Oct 2024 10:34:17 +0000 (12:34 +0200)]
Fix wording in comments
Igor Putovny [Fri, 4 Oct 2024 14:29:08 +0000 (16:29 +0200)]
Do not delete trie after finishing aggregation
Igor Putovny [Wed, 2 Oct 2024 15:00:21 +0000 (17:00 +0200)]
Modify merge_potential_buckets() to return whether the target set has changed
Igor Putovny [Thu, 26 Sep 2024 17:29:02 +0000 (19:29 +0200)]
Add selected bucket and FIB status to node
Igor Putovny [Wed, 25 Sep 2024 11:00:57 +0000 (13:00 +0200)]
Second pass doesn't need protocol
Igor Putovny [Tue, 24 Sep 2024 17:00:00 +0000 (19:00 +0200)]
node_insert_potential_bucket() accepts bucket pointer, not an ID
Igor Putovny [Thu, 19 Sep 2024 15:12:15 +0000 (17:12 +0200)]
Remove settle timer
Igor Putovny [Wed, 18 Sep 2024 14:00:40 +0000 (16:00 +0200)]
Refactor
Igor Putovny [Tue, 17 Sep 2024 12:21:13 +0000 (14:21 +0200)]
Rename
Igor Putovny [Tue, 17 Sep 2024 10:36:38 +0000 (12:36 +0200)]
Use hmap for assigning bucket IDs
Igor Putovny [Mon, 16 Sep 2024 14:52:49 +0000 (16:52 +0200)]
Log time to measure duration of individual aggregation phases
Igor Putovny [Tue, 27 Aug 2024 12:01:31 +0000 (14:01 +0200)]
Implement bitmaps for storing potential buckets
Igor Putovny [Fri, 23 Aug 2024 15:36:43 +0000 (17:36 +0200)]
Delete unnecessary code
Igor Putovny [Mon, 19 Aug 2024 13:24:52 +0000 (15:24 +0200)]
Bugfix
During implementation of the new mechanism of storing potential buckets it was
found that aggregator allocates more than 4000 buckets despite routes in the
configuration file being grouped to approximately 200 buckets.
Upon closer inspection it became clear that tmp_bucket in aggregator_rt_notify()
was allocated in the linpool when it should have been allocated on stack.
Moreover, tmp_bucket was allocated without additional space for aggr_data.
These data were never copied from tmp_bucket to new_bucket.
This was the source of false negative results of HASH_FIND. Buckets are compared
not only by their hashes, but also lists of aggr_data. Because same_val_list()
was reading beyond allocated memory, buckets were never equal. HASH_FIND
therefore returned NULL, which prompted aggregator to create many more buckets
despite already existing buckets with the same hash.
Igor Putovny [Thu, 15 Aug 2024 09:31:09 +0000 (11:31 +0200)]
Add logs about number of nodes
Igor Putovny [Wed, 14 Aug 2024 14:47:22 +0000 (16:47 +0200)]
Avoid unnecessary allocation of nodes in the first pass
Igor Putovny [Wed, 31 Jul 2024 16:12:50 +0000 (18:12 +0200)]
Split third pass function into two functions
(in order to get rid of redundant condition testing)
Igor Putovny [Wed, 17 Jul 2024 12:04:19 +0000 (14:04 +0200)]
Add ancestor pointer to enable faster lookup of ancestor with non-null bucket
Igor Putovny [Wed, 10 Jul 2024 13:12:49 +0000 (15:12 +0200)]
Write documentation
Igor Putovny [Wed, 10 Jul 2024 13:10:57 +0000 (15:10 +0200)]
Update time before and after aggregation
Igor Putovny [Wed, 10 Jul 2024 13:07:54 +0000 (15:07 +0200)]
Add logging option to log prefix aggregation progress
Igor Putovny [Wed, 10 Jul 2024 12:06:29 +0000 (14:06 +0200)]
Change order of parameters, make aesthetic changes
Igor Putovny [Wed, 10 Jul 2024 11:09:48 +0000 (13:09 +0200)]
Add unique ID to buckets and compare them by this ID
Igor Putovny [Wed, 10 Jul 2024 10:41:18 +0000 (12:41 +0200)]
Add counters of nodes with assigned bucket
Igor Putovny [Thu, 4 Jul 2024 13:34:29 +0000 (15:34 +0200)]
Add protocol cleanup
Igor Putovny [Thu, 4 Jul 2024 13:12:52 +0000 (15:12 +0200)]
Small changes
Igor Putovny [Thu, 4 Jul 2024 12:41:47 +0000 (14:41 +0200)]
Remove unused includes
Igor Putovny [Thu, 4 Jul 2024 12:36:55 +0000 (14:36 +0200)]
Add comments
Igor Putovny [Wed, 3 Jul 2024 14:04:44 +0000 (16:04 +0200)]
Change order of parameters in create_route()
Igor Putovny [Wed, 3 Jul 2024 13:45:21 +0000 (15:45 +0200)]
Rename
Igor Putovny [Wed, 3 Jul 2024 13:44:36 +0000 (15:44 +0200)]
Small changes, add assert
Igor Putovny [Thu, 27 Jun 2024 15:22:17 +0000 (17:22 +0200)]
Remove old first_pass()
Igor Putovny [Thu, 27 Jun 2024 15:14:01 +0000 (17:14 +0200)]
Small changes
Igor Putovny [Wed, 19 Jun 2024 13:33:05 +0000 (15:33 +0200)]
Bugfix
Igor Putovny [Wed, 19 Jun 2024 10:30:34 +0000 (12:30 +0200)]
Add get_status() function
Igor Putovny [Thu, 13 Jun 2024 14:37:10 +0000 (16:37 +0200)]
Modify aggregator_start() and shutdown()
Igor Putovny [Thu, 13 Jun 2024 14:36:12 +0000 (16:36 +0200)]
Kick settle timer when receiving updates in rt_notify()
Igor Putovny [Thu, 13 Jun 2024 11:08:42 +0000 (13:08 +0200)]
Request feed when settle timer triggers and initialize trie just before aggregation
Igor Putovny [Wed, 12 Jun 2024 16:27:36 +0000 (18:27 +0200)]
Set aggregation mode in the configuration file
Igor Putovny [Wed, 12 Jun 2024 15:27:05 +0000 (17:27 +0200)]
Rename
Igor Putovny [Tue, 11 Jun 2024 11:47:07 +0000 (13:47 +0200)]
Run aggregation on feed end from src channel and request feeding after receiving update
Igor Putovny [Tue, 11 Jun 2024 11:27:15 +0000 (13:27 +0200)]
Move trie initialization to separate function
Igor Putovny [Tue, 4 Jun 2024 15:56:26 +0000 (17:56 +0200)]
Remove delete_trie()
Igor Putovny [Tue, 4 Jun 2024 15:44:28 +0000 (17:44 +0200)]
Change slabs for linpools
Igor Putovny [Tue, 4 Jun 2024 12:15:32 +0000 (14:15 +0200)]
Add more assertions
Igor Putovny [Tue, 4 Jun 2024 10:25:54 +0000 (12:25 +0200)]
Run aggregation only once (temporary solution)
Igor Putovny [Tue, 4 Jun 2024 10:17:39 +0000 (12:17 +0200)]
Change logs
Igor Putovny [Fri, 31 May 2024 14:20:52 +0000 (16:20 +0200)]
Ignore updates if protocol is not UP
Igor Putovny [Fri, 31 May 2024 14:19:56 +0000 (16:19 +0200)]
Use sl_allocz() for node allocation
Igor Putovny [Thu, 30 May 2024 14:44:44 +0000 (16:44 +0200)]
Compare pointers to NULL using implicit conversion to bool
Igor Putovny [Thu, 30 May 2024 12:36:54 +0000 (14:36 +0200)]
Don't run aggregation on feed end
Igor Putovny [Thu, 30 May 2024 11:07:48 +0000 (13:07 +0200)]
Fix incorrect use of SKIP_BACK
Igor Putovny [Fri, 17 May 2024 13:28:08 +0000 (15:28 +0200)]
Add settle timer
Igor Putovny [Fri, 17 May 2024 10:48:03 +0000 (12:48 +0200)]
Replace goto with else
Igor Putovny [Thu, 16 May 2024 12:48:14 +0000 (14:48 +0200)]
Move aggregation algorithm to separate function
Igor Putovny [Tue, 14 May 2024 12:57:47 +0000 (14:57 +0200)]
Add assertions
Igor Putovny [Tue, 14 May 2024 12:27:45 +0000 (14:27 +0200)]
Use memcpy with size argument based on destination size
Igor Putovny [Fri, 10 May 2024 13:58:02 +0000 (15:58 +0200)]
Remove unused code
Igor Putovny [Fri, 10 May 2024 10:33:56 +0000 (12:33 +0200)]
Replace goto with else
Igor Putovny [Thu, 9 May 2024 16:00:44 +0000 (18:00 +0200)]
Replace for loops with memcpy
Igor Putovny [Thu, 9 May 2024 15:57:28 +0000 (17:57 +0200)]
Replace for loops with memcpy
Igor Putovny [Thu, 9 May 2024 15:35:20 +0000 (17:35 +0200)]
Reorder parameters to stay consistent with the codebase
Igor Putovny [Thu, 9 May 2024 15:29:03 +0000 (17:29 +0200)]
Add logs and declarations, remove unused code
Igor Putovny [Tue, 7 May 2024 16:09:29 +0000 (18:09 +0200)]
Make small changes to increase readability
Igor Putovny [Tue, 7 May 2024 14:36:27 +0000 (16:36 +0200)]
Rename
Igor Putovny [Fri, 12 Apr 2024 10:59:37 +0000 (12:59 +0200)]
Rename
Igor Putovny [Thu, 11 Apr 2024 15:34:42 +0000 (17:34 +0200)]
Construct trie only after protocol feed ends
Igor Putovny [Thu, 11 Apr 2024 14:53:40 +0000 (16:53 +0200)]
Don't schedule aggregation as event but run it at the protocol feed end