diff options
| author | Donald Sharp <sharpd@cumulusnetworks.com> | 2017-09-25 08:10:24 -0400 |
|---|---|---|
| committer | Donald Sharp <sharpd@cumulusnetworks.com> | 2017-09-25 08:10:24 -0400 |
| commit | 1ca60f2c3ebe0a33c508de1ab81edb2bde5d9c4c (patch) | |
| tree | 21aaae4de4406c6da1bfe91dd6bcbeabb6da6a43 /zebra/zebra_rib.c | |
| parent | c84ed8700efc07d66d2ea7373dcc2d9e116f1003 (diff) | |
zebra: Do not allow same rib_dest_t be queued multiple times to meta queue
If we have already scheduled a node to be on the meta_queue, there is no
need to schedule it up again.
On startup we are calling rib_update() multiple times per connected route.
Due to the multiple ways we can get callbacks for adding a connected route
I decided it was best to just improve meta_queue performance as opposed
to trying to figure out all the different ways across all the platforms
that we can decide that a connected route has changed. This appears
to solve the issue with a very large # of interfaces coming up
at the same time on startup.
Signed-off-by: Donald Sharp <sharpd@cumulusnetworks.com>
Diffstat (limited to 'zebra/zebra_rib.c')
| -rw-r--r-- | zebra/zebra_rib.c | 9 |
1 files changed, 9 insertions, 0 deletions
diff --git a/zebra/zebra_rib.c b/zebra/zebra_rib.c index bb49663319..00550fa07d 100644 --- a/zebra/zebra_rib.c +++ b/zebra/zebra_rib.c @@ -2552,6 +2552,15 @@ static void rib_update_table(struct route_table *table, * the trigger event. */ for (rn = route_top(table); rn; rn = srcdest_route_next(rn)) { + /* + * If we are looking at a route node and the node + * has already been queued we don't + * need to queue it up again + */ + if (rn->info + && CHECK_FLAG(rib_dest_from_rnode(rn)->flags, + RIB_ROUTE_ANY_QUEUED)) + continue; switch (event) { case RIB_UPDATE_IF_CHANGE: /* Examine all routes that won't get processed by the |