[PATCH 01/24] bpfilter: reject kernel addresses
From: Christoph Hellwig <hch@lst.de>
Date: 2020-07-20 12:53:04
Also in:
bpf, bridge, linux-can, linux-crypto, linux-hams, linux-s390, linux-sctp, lkml, lvs-devel, mptcp, netdev, netfilter-devel
Subsystem:
networking [general], the rest · Maintainers:
"David S. Miller", Eric Dumazet, Jakub Kicinski, Paolo Abeni, Linus Torvalds
When feeding addresses to userspace we can't support kernel addresses that were fed under set_fs(KERNEL_DS) from bpf-cgroup. Signed-off-by: Christoph Hellwig <hch@lst.de> --- net/bpfilter/bpfilter_kern.c | 5 +++++ 1 file changed, 5 insertions(+)
diff --git a/net/bpfilter/bpfilter_kern.c b/net/bpfilter/bpfilter_kern.c
index 2c31e82cb953af..977e9dad72ca4f 100644
--- a/net/bpfilter/bpfilter_kern.c
+++ b/net/bpfilter/bpfilter_kern.c@@ -41,6 +41,11 @@ static int __bpfilter_process_sockopt(struct sock *sk, int optname, ssize_t n; int ret = -EFAULT; + if (uaccess_kernel()) { + pr_err("kernel access not supported\n"); + return -EFAULT; + } + req.is_set = is_set; req.pid = current->pid; req.cmd = optname;
--
2.27.0