From: Kaike Wan Date: Thu, 15 Aug 2019 19:20:39 +0000 (-0400) Subject: IB/hfi1: Unsafe PSN checking for TID RDMA READ Resp packet X-Git-Url: http://git.cdn.openwrt.org/?a=commitdiff_plain;h=35d5c8b82e2c32e8e29ca195bb4dac60ba7d97fc;p=openwrt%2Fstaging%2Fblogic.git IB/hfi1: Unsafe PSN checking for TID RDMA READ Resp packet When processing a TID RDMA READ RESP packet that causes KDETH EFLAGS errors, the packet's IB PSN is checked against qp->s_last_psn and qp->s_psn without the protection of qp->s_lock, which is not safe. This patch fixes the issue by acquiring qp->s_lock first. Fixes: 9905bf06e890 ("IB/hfi1: Add functions to receive TID RDMA READ response") Cc: Reviewed-by: Mike Marciniszyn Signed-off-by: Kaike Wan Signed-off-by: Dennis Dalessandro Link: https://lore.kernel.org/r/20190815192039.105923.7852.stgit@awfm-01.aw.intel.com Signed-off-by: Doug Ledford --- diff --git a/drivers/infiniband/hw/hfi1/tid_rdma.c b/drivers/infiniband/hw/hfi1/tid_rdma.c index 94070144fef5..01c8b0280700 100644 --- a/drivers/infiniband/hw/hfi1/tid_rdma.c +++ b/drivers/infiniband/hw/hfi1/tid_rdma.c @@ -2687,12 +2687,12 @@ static bool handle_read_kdeth_eflags(struct hfi1_ctxtdata *rcd, u32 fpsn; lockdep_assert_held(&qp->r_lock); + spin_lock(&qp->s_lock); /* If the psn is out of valid range, drop the packet */ if (cmp_psn(ibpsn, qp->s_last_psn) < 0 || cmp_psn(ibpsn, qp->s_psn) > 0) - return ret; + goto s_unlock; - spin_lock(&qp->s_lock); /* * Note that NAKs implicitly ACK outstanding SEND and RDMA write * requests and implicitly NAK RDMA read and atomic requests issued