bgpd: Do not reset peers on suppress-fib toggling

If the desired state is the same - do nothing instead of resetting once again.

Fixes: bdb5ae8bce94432eb5e581f04f48dc4aa5db7ca4 ("bgpd: Make suppress-fib-pending clear peering")
Signed-off-by: Donatas Abraitis <donatas@opensourcerouting.org>
(cherry picked from commit 7fb4c03f5b1c9cd4e8acb85b4cf953ac85aa0bbf)

diff --git a/bgpd/bgpd.c b/bgpd/bgpd.c
index 4dfc967a7e28ed35219c622abd5ff7f41d6a0334..c53f4965cc931174e54ccbae9c60e696221badce 100644 (file)
--- a/bgpd/bgpd.c
+++ b/bgpd/bgpd.c
@@ -461,6 +461,10 @@ void bgp_suppress_fib_pending_set(struct bgp *bgp, bool set)
        if (bgp->inst_type == BGP_INSTANCE_TYPE_VIEW)
                return;
 
+       /* Do nothing if already in a desired state */
+       if (set == !!CHECK_FLAG(bgp->flags, BGP_FLAG_SUPPRESS_FIB_PENDING))
+               return;
+
        if (set) {
                SET_FLAG(bgp->flags, BGP_FLAG_SUPPRESS_FIB_PENDING);
                /* Send msg to zebra for the first instance of bgp enabled