Skip to contentSkip to conversation

batman-adv: Fix netlink dumping of BLA claims

PATCH In review Sven Eckelmann 2 participants 3 messages Feb 24, 2018 – Feb 26, 2018 Batman advanced

lore
Share & tools
lore

Sven Eckelmann

author maintainer
#0 lore raw

PATCH 1/2 batman-adv: Fix netlink dumping of BLA claims

The function batadv_bla_claim_dump_bucket must be able to handle non-complete dumps of a single bucket. It tries to do that by saving the latest dumped index in *idx_skip to inform the caller about the current state.

But the caller only assumes that buckets were not completely dumped when the return code is non-zero. This function must therefore also return a non-zero index when the dumping of an entry failed. Otherwise the caller will just skip all remaining buckets.

And the function must also reset *idx_skip back to zero when it finished a bucket. Otherwise it will skip the same number of entries in the next bucket as the previous one had.

Fixes 3b7a63606020 ("batman-adv: add B.A.T.M.A.N. Dump BLA claims via netlink")
---
 net/batman-adv/bridge_loop_avoidance.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)
net/batman-adv/bridge_loop_avoidance.c +7 -4
index cdd8e8e4..60ce119e 100644
@@ -2161,22 +2161,25 @@ batadv_bla_claim_dump_bucket(struct sk_buff *msg, u32 portid, u32 seq,
 {
 	struct batadv_bla_claim *claim;
 	int idx = 0;
+	int ret = 0;
 
 	rcu_read_lock();
 	hlist_for_each_entry_rcu(claim, head, hash_entry) {
 		if (idx++ < *idx_skip)
 			continue;
-		if (batadv_bla_claim_dump_entry(msg, portid, seq,
-						primary_if, claim)) {
+
+		ret = batadv_bla_claim_dump_entry(msg, portid, seq,
+						  primary_if, claim);
+		if (ret) {
 			*idx_skip = idx - 1;
 			goto unlock;
 		}
 	}
 
-	*idx_skip = idx;
+	*idx_skip = 0;
 unlock:
 	rcu_read_unlock();
-	return 0;
+	return ret;
 }
 
 /**
Signature · 1 line
2.16.1
author maintainer
#1 lore raw

PATCH 2/2 batman-adv: Fix netlink dumping of BLA backbones

The function batadv_bla_backbone_dump_bucket must be able to handle non-complete dumps of a single bucket. It tries to do that by saving the latest dumped index in *idx_skip to inform the caller about the current state.

But the caller only assumes that buckets were not completely dumped when the return code is non-zero. This function must therefore also return a non-zero index when the dumping of an entry failed. Otherwise the caller will just skip all remaining buckets.

And the function must also reset *idx_skip back to zero when it finished a bucket. Otherwise it will skip the same number of entries in the next bucket as the previous one had.

Fixes 7f609cab5123 ("batman-adv: add backbone table netlink support")
---
 net/batman-adv/bridge_loop_avoidance.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)
net/batman-adv/bridge_loop_avoidance.c +7 -4
index 60ce119e..422ee16b 100644
@@ -2394,22 +2394,25 @@ batadv_bla_backbone_dump_bucket(struct sk_buff *msg, u32 portid, u32 seq,
 {
 	struct batadv_bla_backbone_gw *backbone_gw;
 	int idx = 0;
+	int ret = 0;
 
 	rcu_read_lock();
 	hlist_for_each_entry_rcu(backbone_gw, head, hash_entry) {
 		if (idx++ < *idx_skip)
 			continue;
-		if (batadv_bla_backbone_dump_entry(msg, portid, seq,
-						   primary_if, backbone_gw)) {
+
+		ret = batadv_bla_backbone_dump_entry(msg, portid, seq,
+						     primary_if, backbone_gw);
+		if (ret) {
 			*idx_skip = idx - 1;
 			goto unlock;
 		}
 	}
 
-	*idx_skip = idx;
+	*idx_skip = 0;
 unlock:
 	rcu_read_unlock();
-	return 0;
+	return ret;
 }
 
 /**
Signature · 1 line
2.16.1
maintainer
#2 lore raw
In reply to Sven Eckelmann
Quoted 5 lines · CET Sven Eckelmann

The function batadv_bla_claim_dump_bucket must be able to handle non-complete dumps of a single bucket. It tries to do that by saving the latest dumped index in *idx_skip to inform the caller about the current state.

Quoted 8 lines

But the caller only assumes that buckets were not completely dumped when the return code is non-zero. This function must therefore also return a non-zero index when the dumping of an entry failed. Otherwise the caller will just skip all remaining buckets.

And the function must also reset *idx_skip back to zero when it finished a bucket. Otherwise it will skip the same number of entries in the next bucket as the previous one had.

I've applied both patches in this series in 49197c00 and 29e4759e.

Thank you!
Simon