From: Bart Van Assche <bart.vanassche@wdc.com>
Date: Wed, 11 Oct 2017 10:48:50 -0700
Subject: RDMA/cxgb3: Annotate locking assumptions
Patch-mainline: v4.15-rc1
Git-commit: 1f5461b2021ae100d0171e6e0d99c009f84f3ac6
References: bsc#1103992 FATE#326009

Tell sparse what the locking assumptions are for __flush_qp() such
that it does not complain about the locking operations inside that
function.

Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Reviewed-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Acked-by: Thomas Bogendoerfer <tbogendoerfer@suse.de>
---
 drivers/infiniband/hw/cxgb3/iwch_qp.c |    3 +++
 1 file changed, 3 insertions(+)

--- a/drivers/infiniband/hw/cxgb3/iwch_qp.c
+++ b/drivers/infiniband/hw/cxgb3/iwch_qp.c
@@ -722,10 +722,13 @@ int iwch_post_terminate(struct iwch_qp *
  */
 static void __flush_qp(struct iwch_qp *qhp, struct iwch_cq *rchp,
 				struct iwch_cq *schp)
+	__releases(&qhp->lock)
+	__acquires(&qhp->lock)
 {
 	int count;
 	int flushed;
 
+	lockdep_assert_held(&qhp->lock);
 
 	pr_debug("%s qhp %p rchp %p schp %p\n", __func__, qhp, rchp, schp);
 	/* take a ref on the qhp since we must release the lock */