Patch "nvmet: confirm sq percpu has scheduled and switched to atomic" has been added to the 4.9-stable tree

Patch "nvmet: confirm sq percpu has scheduled and switched to atomic" has been added to the 4.9-stable tree

[gregkh]

This is a note to let you know that I've just added the patch titled

    nvmet: confirm sq percpu has scheduled and switched to atomic

to the 4.9-stable tree which can be found at:
    http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary

The filename of the patch is:
     nvmet-confirm-sq-percpu-has-scheduled-and-switched-to-atomic.patch
and it can be found in the queue-4.9 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@vger.kernel.org> know about it.


>From foo@baz Mon Dec 18 14:12:34 CET 2017
From: Sagi Grimberg <sagi@grimberg.me>
Date: Mon, 6 Mar 2017 18:46:20 +0200
Subject: nvmet: confirm sq percpu has scheduled and switched to atomic

From: Sagi Grimberg <sagi@grimberg.me>


[ Upstream commit d11ea004a458b982e19b188c386e25a9b66ec446 ]

percpu_ref_kill is not enough to prevent subsequent
percpu_ref_tryget_live from failing. Hence call
perfcpu_ref_kill_confirm to make it safe.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Sasha Levin <alexander.levin@verizon.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 drivers/nvme/target/core.c  |   11 ++++++++++-
 drivers/nvme/target/nvmet.h |    1 +
 2 files changed, 11 insertions(+), 1 deletion(-)

--- a/drivers/nvme/target/core.c
+++ b/drivers/nvme/target/core.c
@@ -422,6 +422,13 @@ void nvmet_sq_setup(struct nvmet_ctrl *c
 	ctrl->sqs[qid] = sq;
 }
 
+static void nvmet_confirm_sq(struct percpu_ref *ref)
+{
+	struct nvmet_sq *sq = container_of(ref, struct nvmet_sq, ref);
+
+	complete(&sq->confirm_done);
+}
+
 void nvmet_sq_destroy(struct nvmet_sq *sq)
 {
 	/*
@@ -430,7 +437,8 @@ void nvmet_sq_destroy(struct nvmet_sq *s
 	 */
 	if (sq->ctrl && sq->ctrl->sqs && sq->ctrl->sqs[0] == sq)
 		nvmet_async_events_free(sq->ctrl);
-	percpu_ref_kill(&sq->ref);
+	percpu_ref_kill_and_confirm(&sq->ref, nvmet_confirm_sq);
+	wait_for_completion(&sq->confirm_done);
 	wait_for_completion(&sq->free_done);
 	percpu_ref_exit(&sq->ref);
 
@@ -458,6 +466,7 @@ int nvmet_sq_init(struct nvmet_sq *sq)
 		return ret;
 	}
 	init_completion(&sq->free_done);
+	init_completion(&sq->confirm_done);
 
 	return 0;
 }
--- a/drivers/nvme/target/nvmet.h
+++ b/drivers/nvme/target/nvmet.h
@@ -73,6 +73,7 @@ struct nvmet_sq {
 	u16			qid;
 	u16			size;
 	struct completion	free_done;
+	struct completion	confirm_done;
 };
 
 /**


Patches currently in stable-queue which might be from sagi@grimberg.me are

queue-4.9/blk-mq-fix-tagset-reinit-in-the-presence-of-cpu-hot-unplug.patch
queue-4.9/nvmet-confirm-sq-percpu-has-scheduled-and-switched-to-atomic.patch
queue-4.9/nvme-use-kref_get_unless_zero-in-nvme_find_get_ns.patch
queue-4.9/nvme-loop-fix-a-possible-use-after-free-when-destroying-the-admin-queue.patch
queue-4.9/nvmet-rdma-fix-a-possible-uninitialized-variable-dereference.patch