xfrm: policy: increment xfrm_hash_generation on hash rebuild

Hash rebuild will re-set all the inexact entries, then re-insert them.
Lookups that can occur in parallel will therefore not find any policies.

This was safe when lookups were still guarded by rwlock.
After rcu-ification, lookups check the hash_generation seqcount to detect
when a hash resize takes place.  Hash rebuild missed the needed increment.

Hash resizes and hash rebuilds cannot occur in parallel (both acquire
hash_resize_mutex), so just increment xfrm_hash_generation, like resize.

Fixes: a7c44247f7 ("xfrm: policy: make xfrm_policy_lookup_bytype lockless")
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
This commit is contained in:
Florian Westphal 2019-01-04 14:17:01 +01:00 committed by Steffen Klassert
parent 355b00d1e1
commit 7a474c3658
1 changed files with 2 additions and 0 deletions

View File

@ -1235,6 +1235,7 @@ static void xfrm_hash_rebuild(struct work_struct *work)
} while (read_seqretry(&net->xfrm.policy_hthresh.lock, seq));
spin_lock_bh(&net->xfrm.xfrm_policy_lock);
write_seqcount_begin(&xfrm_policy_hash_generation);
/* make sure that we can insert the indirect policies again before
* we start with destructive action.
@ -1334,6 +1335,7 @@ static void xfrm_hash_rebuild(struct work_struct *work)
out_unlock:
__xfrm_policy_inexact_flush(net);
write_seqcount_end(&xfrm_policy_hash_generation);
spin_unlock_bh(&net->xfrm.xfrm_policy_lock);
mutex_unlock(&hash_resize_mutex);