RPC: Fix two potential races in put_rpccred
It is possible for rpcauth_destroy_credcache() to cause the rpc credentials to be unhashed while put_rpccred is waiting for the rpc_credcache_lock on another cpu. Should this happen, then we can end up calling hlist_del_rcu(&cred->cr_hash) a second time in put_rpccred, thus causing list corruption. Should the credential actually be hashed, it is also possible for rpcauth_lookup_credcache to find and reference it before we get round to unhashing it. In this case, the call to rpcauth_unhash_cred will fail, and so we should just exit without destroying the cred. Reported-by: NNeil Brown <neilb@suse.de> Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>
Showing
想要评论请 注册 或 登录