FAILED: patch "[PATCH] IB/hfi1: Fix deadlock with txreq allocation slow path" failed to apply to 4.6-stable tree

The patch below does not apply to the 4.6-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From: Mike Marciniszyn <[email protected]>
Date: Fri, 17 Jun 2016 19:17:49 -0700
Subject: [PATCH] IB/hfi1: Fix deadlock with txreq allocation slow path

A failure in the get_txreq() inline will result in a slow path retry using __get_txreq().

__get_txreq() attempts to procure the qp s_lock, which is already held in all callers.

Fix by deleting the s_lock maintenance in __get_txreq() and add sparse syntax hooks to future proof the code.

index bc95c4112c61..d8fb056526f8 100644
@@ -92,11 +92,10 @@ void hfi1_put_txreq(struct verbs_txreq *tx)
 
 struct verbs_txreq *__get_txreq(struct hfi1_ibdev *dev,
 				struct rvt_qp *qp)
+	__must_hold(&qp->s_lock)
 {
 	struct verbs_txreq *tx = ERR_PTR(-EBUSY);
-	unsigned long flags;
 
-	spin_lock_irqsave(&qp->s_lock, flags);
 	write_seqlock(&dev->iowait_lock);
 	if (ib_rvt_state_ops[qp->state] & RVT_PROCESS_RECV_OK) {
 		struct hfi1_qp_priv *priv;

@@ -116,7 +115,6 @@ struct verbs_txreq *__get_txreq(struct hfi1_ibdev *dev,
 	}
 out:
 	write_sequnlock(&dev->iowait_lock);
-	spin_unlock_irqrestore(&qp->s_lock, flags);
 	return tx;
 }
 

index 1cf69b2fe4a5..a1d6e0807f97 100644
@@ -73,6 +73,7 @@ struct verbs_txreq *__get_txreq(struct hfi1_ibdev *dev,
 
 static inline struct verbs_txreq *get_txreq(struct hfi1_ibdev *dev,
 					    struct rvt_qp *qp)
+	__must_hold(&qp->slock)
 {
 	struct verbs_txreq *tx;
 	struct hfi1_qp_priv *priv = qp->priv;

A 4.6 version is in http://marc.info/?l=linux-rdma&m=146904918032092&w=2.

Oddly it cherry-picked for me ok. Is this due to a flaw in the stable triggering method in the upstream patch?

Mike

I can't do much with a web-link, can you send me the patch?

Hm, if cherry-pick worked, that's a huge hint, can you send me the cherry-picked patch output? Odds are git can handle a merge issue better than quilt can.

thanks,

greg k-h

The web link was intended as an FYI.

I submitting the actual patch to [email protected].

Mike

Were you able to see this patch on the list?

Mike