RDS: Let rds_message_alloc_sgs() return NULL
Even with the previous fix, we still are reading the iovecs once
to determine SGs needed, and then again later on. Preallocating
space for sg lists as part of rds_message seemed like a good idea
but it might be better to not do this. While working to redo that
code, this patch attempts to protect against userspace rewriting
the rds_iovec array between the first and second accesses.
The consequences of this would be either a too-small or too-large
sg list array. Too large is not an issue. This patch changes all
callers of message_alloc_sgs to handle running out of preallocated
sgs, and fail gracefully.
Signed-off-by: Andy Grover <andy.grover@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/net/rds/rdma.c b/net/rds/rdma.c
index caa4d98..8920f2a 100644
--- a/net/rds/rdma.c
+++ b/net/rds/rdma.c
@@ -607,6 +607,10 @@
op->op_recverr = rs->rs_recverr;
WARN_ON(!nr_pages);
op->op_sg = rds_message_alloc_sgs(rm, nr_pages);
+ if (!op->op_sg) {
+ ret = -ENOMEM;
+ goto out;
+ }
if (op->op_notify || op->op_recverr) {
/* We allocate an uninitialized notifier here, because
@@ -807,6 +811,10 @@
rm->atomic.op_active = 1;
rm->atomic.op_recverr = rs->rs_recverr;
rm->atomic.op_sg = rds_message_alloc_sgs(rm, 1);
+ if (!rm->atomic.op_sg) {
+ ret = -ENOMEM;
+ goto err;
+ }
/* verify 8 byte-aligned */
if (args->local_addr & 0x7) {