public inbox for linux-nvme@lists.infradead.org
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@lst.de>
To: Jens Axboe <axboe@kernel.dk>, Keith Busch <kbusch@kernel.org>,
	Sagi Grimberg <sagi@grimberg.me>, Chao Leng <lengchao@huawei.com>
Cc: Ming Lei <ming.lei@redhat.com>,
	linux-nvme@lists.infradead.org, linux-block@vger.kernel.org
Subject: [PATCH 8/8] nvme: use blk_mq_[un]quiesce_tagset
Date: Thu, 20 Oct 2022 12:56:08 +0200	[thread overview]
Message-ID: <20221020105608.1581940-9-hch@lst.de> (raw)
In-Reply-To: <20221020105608.1581940-1-hch@lst.de>

From: Chao Leng <lengchao@huawei.com>

All controller namespaces share the same tagset, so we can use this
interface which does the optimal operation for parallel quiesce based on
the tagset type(e.g. blocking tagsets and non-blocking tagsets).

nvme connect_q should not be quiesced when quiesce tagset, so set the
QUEUE_FLAG_SKIP_TAGSET_QUIESCE to skip it when init connect_q.

Currently we use NVME_NS_STOPPED to ensure pairing quiescing and
unquiescing. If use blk_mq_[un]quiesce_tagset, NVME_NS_STOPPED will be
invalided, so introduce NVME_CTRL_STOPPED to replace NVME_NS_STOPPED.
In addition, we never really quiesce a single namespace. It is a better
choice to move the flag from ns to ctrl.

Signed-off-by: Chao Leng <lengchao@huawei.com>
[hch: rebased on top of prep patches]
Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 drivers/nvme/host/core.c | 38 +++++++++-----------------------------
 drivers/nvme/host/nvme.h |  2 +-
 2 files changed, 10 insertions(+), 30 deletions(-)

diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
index 0ab3a18fd9f85..cc71f1001144f 100644
--- a/drivers/nvme/host/core.c
+++ b/drivers/nvme/host/core.c
@@ -4891,6 +4891,7 @@ int nvme_alloc_io_tag_set(struct nvme_ctrl *ctrl, struct blk_mq_tag_set *set,
 			ret = PTR_ERR(ctrl->connect_q);
 			goto out_free_tag_set;
 		}
+		blk_queue_flag_set(QUEUE_FLAG_SKIP_TAGSET_QUIESCE, ctrl->connect_q);
 	}
 
 	ctrl->tagset = set;
@@ -5090,20 +5091,6 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev,
 }
 EXPORT_SYMBOL_GPL(nvme_init_ctrl);
 
-static void nvme_start_ns_queue(struct nvme_ns *ns)
-{
-	if (test_and_clear_bit(NVME_NS_STOPPED, &ns->flags))
-		blk_mq_unquiesce_queue(ns->queue);
-}
-
-static void nvme_stop_ns_queue(struct nvme_ns *ns)
-{
-	if (!test_and_set_bit(NVME_NS_STOPPED, &ns->flags))
-		blk_mq_quiesce_queue(ns->queue);
-	else
-		blk_mq_wait_quiesce_done(ns->queue->tag_set);
-}
-
 /**
  * nvme_kill_queues(): Ends all namespace queues
  * @ctrl: the dead controller that needs to end
@@ -5120,10 +5107,9 @@ void nvme_kill_queues(struct nvme_ctrl *ctrl)
 	if (ctrl->admin_q && !blk_queue_dying(ctrl->admin_q))
 		nvme_start_admin_queue(ctrl);
 	if (!test_and_set_bit(NVME_CTRL_NS_DEAD, &ctrl->flags)) {
-		list_for_each_entry(ns, &ctrl->namespaces, list) {
+		list_for_each_entry(ns, &ctrl->namespaces, list)
 			blk_mark_disk_dead(ns->disk);
-			nvme_start_ns_queue(ns);
-		}
+		nvme_start_queues(ctrl);
 	}
 	up_read(&ctrl->namespaces_rwsem);
 }
@@ -5179,23 +5165,17 @@ EXPORT_SYMBOL_GPL(nvme_start_freeze);
 
 void nvme_stop_queues(struct nvme_ctrl *ctrl)
 {
-	struct nvme_ns *ns;
-
-	down_read(&ctrl->namespaces_rwsem);
-	list_for_each_entry(ns, &ctrl->namespaces, list)
-		nvme_stop_ns_queue(ns);
-	up_read(&ctrl->namespaces_rwsem);
+	if (!test_and_set_bit(NVME_CTRL_STOPPED, &ctrl->flags))
+		blk_mq_quiesce_tagset(ctrl->tagset);
+	else
+		blk_mq_wait_quiesce_done(ctrl->tagset);
 }
 EXPORT_SYMBOL_GPL(nvme_stop_queues);
 
 void nvme_start_queues(struct nvme_ctrl *ctrl)
 {
-	struct nvme_ns *ns;
-
-	down_read(&ctrl->namespaces_rwsem);
-	list_for_each_entry(ns, &ctrl->namespaces, list)
-		nvme_start_ns_queue(ns);
-	up_read(&ctrl->namespaces_rwsem);
+	if (test_and_clear_bit(NVME_CTRL_STOPPED, &ctrl->flags))
+		blk_mq_unquiesce_tagset(ctrl->tagset);
 }
 EXPORT_SYMBOL_GPL(nvme_start_queues);
 
diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h
index 82989a3322130..54d8127fb6dc5 100644
--- a/drivers/nvme/host/nvme.h
+++ b/drivers/nvme/host/nvme.h
@@ -238,6 +238,7 @@ enum nvme_ctrl_flags {
 	NVME_CTRL_ADMIN_Q_STOPPED	= 1,
 	NVME_CTRL_STARTED_ONCE		= 2,
 	NVME_CTRL_NS_DEAD     		= 3,
+	NVME_CTRL_STOPPED		= 4,
 };
 
 struct nvme_ctrl {
@@ -487,7 +488,6 @@ struct nvme_ns {
 #define NVME_NS_ANA_PENDING	2
 #define NVME_NS_FORCE_RO	3
 #define NVME_NS_READY		4
-#define NVME_NS_STOPPED		5
 
 	struct cdev		cdev;
 	struct device		cdev_device;
-- 
2.30.2



  parent reply	other threads:[~2022-10-20 12:02 UTC|newest]

Thread overview: 61+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-10-20 10:56 per-tagset SRCU struct and quiesce Christoph Hellwig
2022-10-20 10:56 ` [PATCH 1/8] block: set the disk capacity to 0 in blk_mark_disk_dead Christoph Hellwig
2022-10-20 13:16   ` Sagi Grimberg
2022-10-21  1:09   ` Ming Lei
2022-10-21 13:11     ` Christoph Hellwig
2022-10-21  1:53   ` Chao Leng
2022-10-21  6:49   ` Hannes Reinecke
2022-10-21 13:13     ` Christoph Hellwig
2022-10-21 21:12   ` Bart Van Assche
2022-10-20 10:56 ` [PATCH 2/8] blk-mq: skip non-mq queues in blk_mq_quiesce_queue Christoph Hellwig
2022-10-20 13:16   ` Sagi Grimberg
2022-10-21  1:13   ` Ming Lei
2022-10-21 13:19     ` Christoph Hellwig
2022-10-21 15:08       ` Ming Lei
2022-10-21  2:47   ` Chao Leng
2022-10-21  3:16     ` Chao Leng
2022-10-21  6:49   ` Hannes Reinecke
2022-10-20 10:56 ` [PATCH 3/8] blk-mq: move the srcu_struct used for quiescing to the tagset Christoph Hellwig
2022-10-20 13:23   ` Sagi Grimberg
2022-10-20 17:26   ` Keith Busch
2022-10-21 13:20     ` Christoph Hellwig
2022-10-21  1:41   ` Ming Lei
2022-10-21  2:49   ` Chao Leng
2022-10-21  6:50   ` Hannes Reinecke
2022-10-21  7:16   ` Chao Leng
2022-10-21 13:22     ` Christoph Hellwig
2022-10-20 10:56 ` [PATCH 4/8] blk-mq: pass a tagset to blk_mq_wait_quiesce_done Christoph Hellwig
2022-10-20 13:23   ` Sagi Grimberg
2022-10-21  1:46   ` Ming Lei
2022-10-21 13:23     ` Christoph Hellwig
2022-10-21  2:49   ` Chao Leng
2022-10-21  6:50   ` Hannes Reinecke
2022-10-21 21:18   ` Bart Van Assche
2022-10-20 10:56 ` [PATCH 5/8] blk-mq: add tagset quiesce interface Christoph Hellwig
2022-10-20 13:24   ` Sagi Grimberg
2022-10-21  1:53   ` Ming Lei
2022-10-21  2:49   ` Chao Leng
2022-10-21  6:51   ` Hannes Reinecke
2022-10-21 21:22   ` Bart Van Assche
2022-10-24  1:57     ` Chao Leng
2022-10-24 13:35       ` Bart Van Assche
2022-10-25  1:38         ` Chao Leng
2022-10-20 10:56 ` [PATCH 6/8] nvme: move the NS_DEAD flag to the controller Christoph Hellwig
2022-10-20 13:30   ` Sagi Grimberg
2022-10-21 13:28     ` Christoph Hellwig
2022-10-24  8:43       ` Sagi Grimberg
2022-10-24  8:50         ` Sagi Grimberg
2022-10-21  2:49   ` Chao Leng
2022-10-21  6:51   ` Hannes Reinecke
2022-10-20 10:56 ` [PATCH 7/8] nvme: remove nvme_set_queue_dying Christoph Hellwig
2022-10-20 13:10   ` Sagi Grimberg
2022-10-21 13:29     ` Christoph Hellwig
2022-10-24  8:48       ` Sagi Grimberg
2022-10-21  2:50   ` Chao Leng
2022-10-21  6:52   ` Hannes Reinecke
2022-10-20 10:56 ` Christoph Hellwig [this message]
2022-10-20 13:35   ` [PATCH 8/8] nvme: use blk_mq_[un]quiesce_tagset Sagi Grimberg
2022-10-21  2:50   ` Chao Leng
2022-10-21  6:52   ` Hannes Reinecke
2022-10-20 13:16 ` per-tagset SRCU struct and quiesce Sagi Grimberg
2022-10-21 18:06 ` Keith Busch

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20221020105608.1581940-9-hch@lst.de \
    --to=hch@lst.de \
    --cc=axboe@kernel.dk \
    --cc=kbusch@kernel.org \
    --cc=lengchao@huawei.com \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-nvme@lists.infradead.org \
    --cc=ming.lei@redhat.com \
    --cc=sagi@grimberg.me \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox