Commit 0c696bb3 authored by Jens Axboe's avatar Jens Axboe
Browse files

Merge tag 'md-next-20231020' of... 

Merge tag 'md-next-20231020' of https://git.kernel.org/pub/scm/linux/kernel/git/song/md into for-6.7/block

Pull MD changes from Song:

"1. Handle timeout in md-cluster, by Denis Plotnikov;
 2. Cleanup pers->prepare_suspend, by Yu Kuai."

* tag 'md-next-20231020' of https://git.kernel.org/pub/scm/linux/kernel/git/song/md:
  md: cleanup pers->prepare_suspend()
  md-cluster: check for timeout while a new disk adding
parents 41195755 78b7b13f

drivers/md/md-cluster.c

+11 −4
Original line number Diff line number Diff line
@@ -501,7 +501,7 @@ static void process_suspend_info(struct mddev *mddev, 
	mddev->pers->quiesce(mddev, 0);

}



static void process_add_new_disk(struct mddev *mddev, struct cluster_msg *cmsg)

static int process_add_new_disk(struct mddev *mddev, struct cluster_msg *cmsg)

{

	char disk_uuid[64];

	struct md_cluster_info *cinfo = mddev->cluster_info;
@@ -509,6 +509,7 @@ static void process_add_new_disk(struct mddev *mddev, struct cluster_msg *cmsg) 
	char raid_slot[16];

	char *envp[] = {event_name, disk_uuid, raid_slot, NULL};

	int len;

	int res = 0;



	len = snprintf(disk_uuid, 64, "DEVICE_UUID=");

	sprintf(disk_uuid + len, "%pU", cmsg->uuid);
@@ -517,9 +518,14 @@ static void process_add_new_disk(struct mddev *mddev, struct cluster_msg *cmsg) 
	init_completion(&cinfo->newdisk_completion);

	set_bit(MD_CLUSTER_WAITING_FOR_NEWDISK, &cinfo->state);

	kobject_uevent_env(&disk_to_dev(mddev->gendisk)->kobj, KOBJ_CHANGE, envp);

	wait_for_completion_timeout(&cinfo->newdisk_completion,

			NEW_DEV_TIMEOUT);

	if (!wait_for_completion_timeout(&cinfo->newdisk_completion,

					NEW_DEV_TIMEOUT)) {

		pr_err("md-cluster(%s:%d): timeout on a new disk adding\n",

			__func__, __LINE__);

		res = -1;

	}

	clear_bit(MD_CLUSTER_WAITING_FOR_NEWDISK, &cinfo->state);

	return res;

}
@@ -594,7 +600,8 @@ static int process_recvd_msg(struct mddev *mddev, struct cluster_msg *msg) 
				     le64_to_cpu(msg->high));

		break;

	case NEWDISK:

		process_add_new_disk(mddev, msg);

		if (process_add_new_disk(mddev, msg))

			ret = -1;

		break;

	case REMOVE:

		process_remove_disk(mddev, msg);

drivers/md/md.c

+16 −1
Original line number Diff line number Diff line
@@ -91,6 +91,18 @@ static void mddev_detach(struct mddev *mddev); 
static void export_rdev(struct md_rdev *rdev, struct mddev *mddev);

static void md_wakeup_thread_directly(struct md_thread __rcu *thread);



enum md_ro_state {

	MD_RDWR,

	MD_RDONLY,

	MD_AUTO_READ,

	MD_MAX_STATE

};



static bool md_is_rdwr(struct mddev *mddev)

{

	return (mddev->ro == MD_RDWR);

}



/*

 * Default number of read corrections we'll attempt on an rdev

 * before ejecting it from the array. We divide the read error
@@ -333,6 +345,10 @@ EXPORT_SYMBOL_GPL(md_new_event); 
static LIST_HEAD(all_mddevs);

static DEFINE_SPINLOCK(all_mddevs_lock);



static bool is_md_suspended(struct mddev *mddev)

{

	return percpu_ref_is_dying(&mddev->active_io);

}

/* Rather than calling directly into the personality make_request function,

 * IO requests come here first so that we can check if the device is

 * being suspended pending a reconfiguration.
@@ -9144,7 +9160,6 @@ void md_do_sync(struct md_thread *thread) 
	spin_unlock(&mddev->lock);



	wake_up(&resync_wait);

	wake_up(&mddev->sb_wait);

	md_wakeup_thread(mddev->thread);

	return;

}

drivers/md/md.h

+0 −18
Original line number Diff line number Diff line
@@ -565,23 +565,6 @@ enum recovery_flags { 
	MD_RESYNCING_REMOTE,	/* remote node is running resync thread */

};



enum md_ro_state {

	MD_RDWR,

	MD_RDONLY,

	MD_AUTO_READ,

	MD_MAX_STATE

};



static inline bool md_is_rdwr(struct mddev *mddev)

{

	return (mddev->ro == MD_RDWR);

}



static inline bool is_md_suspended(struct mddev *mddev)

{

	return percpu_ref_is_dying(&mddev->active_io);

}



static inline int __must_check mddev_lock(struct mddev *mddev)

{

	return mutex_lock_interruptible(&mddev->reconfig_mutex);
@@ -641,7 +624,6 @@ struct md_personality 
	int (*start_reshape) (struct mddev *mddev);

	void (*finish_reshape) (struct mddev *mddev);

	void (*update_reshape_pos) (struct mddev *mddev);

	void (*prepare_suspend) (struct mddev *mddev);

	/* quiesce suspends or resumes internal processing.

	 * 1 - stop new actions and wait for action io to complete

	 * 0 - return to normal behaviour

drivers/md/raid5.c

+1 −43
Original line number Diff line number Diff line
@@ -5953,19 +5953,6 @@ static int add_all_stripe_bios(struct r5conf *conf, 
	return ret;

}



static bool reshape_inprogress(struct mddev *mddev)

{

	return test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery) &&

	       test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) &&

	       !test_bit(MD_RECOVERY_DONE, &mddev->recovery) &&

	       !test_bit(MD_RECOVERY_INTR, &mddev->recovery);

}



static bool reshape_disabled(struct mddev *mddev)

{

	return is_md_suspended(mddev) || !md_is_rdwr(mddev);

}



static enum stripe_result make_stripe_request(struct mddev *mddev,

		struct r5conf *conf, struct stripe_request_ctx *ctx,

		sector_t logical_sector, struct bio *bi)
@@ -5997,8 +5984,7 @@ static enum stripe_result make_stripe_request(struct mddev *mddev, 
			if (ahead_of_reshape(mddev, logical_sector,

					     conf->reshape_safe)) {

				spin_unlock_irq(&conf->device_lock);

				ret = STRIPE_SCHEDULE_AND_RETRY;

				goto out;

				return STRIPE_SCHEDULE_AND_RETRY;

			}

		}

		spin_unlock_irq(&conf->device_lock);
@@ -6077,15 +6063,6 @@ static enum stripe_result make_stripe_request(struct mddev *mddev, 


out_release:

	raid5_release_stripe(sh);

out:

	if (ret == STRIPE_SCHEDULE_AND_RETRY && !reshape_inprogress(mddev) &&

	    reshape_disabled(mddev)) {

		bi->bi_status = BLK_STS_IOERR;

		ret = STRIPE_FAIL;

		pr_err("md/raid456:%s: io failed across reshape position while reshape can't make progress.\n",

		       mdname(mddev));

	}



	return ret;

}
@@ -9027,22 +9004,6 @@ static int raid5_start(struct mddev *mddev) 
	return r5l_start(conf->log);

}



static void raid5_prepare_suspend(struct mddev *mddev)

{

	struct r5conf *conf = mddev->private;



	wait_event(mddev->sb_wait, !reshape_inprogress(mddev) ||

				    percpu_ref_is_zero(&mddev->active_io));

	if (percpu_ref_is_zero(&mddev->active_io))

		return;



	/*

	 * Reshape is not in progress, and array is suspended, io that is

	 * waiting for reshpape can never be done.

	 */

	wake_up(&conf->wait_for_overlap);

}



static struct md_personality raid6_personality =

{

	.name		= "raid6",
@@ -9063,7 +9024,6 @@ static struct md_personality raid6_personality = 
	.check_reshape	= raid6_check_reshape,

	.start_reshape  = raid5_start_reshape,

	.finish_reshape = raid5_finish_reshape,

	.prepare_suspend = raid5_prepare_suspend,

	.quiesce	= raid5_quiesce,

	.takeover	= raid6_takeover,

	.change_consistency_policy = raid5_change_consistency_policy,
@@ -9088,7 +9048,6 @@ static struct md_personality raid5_personality = 
	.check_reshape	= raid5_check_reshape,

	.start_reshape  = raid5_start_reshape,

	.finish_reshape = raid5_finish_reshape,

	.prepare_suspend = raid5_prepare_suspend,

	.quiesce	= raid5_quiesce,

	.takeover	= raid5_takeover,

	.change_consistency_policy = raid5_change_consistency_policy,
@@ -9114,7 +9073,6 @@ static struct md_personality raid4_personality = 
	.check_reshape	= raid5_check_reshape,

	.start_reshape  = raid5_start_reshape,

	.finish_reshape = raid5_finish_reshape,

	.prepare_suspend = raid5_prepare_suspend,

	.quiesce	= raid5_quiesce,

	.takeover	= raid4_takeover,

	.change_consistency_policy = raid5_change_consistency_policy,