diff mbox series

[v3,18/30] null_blk: Introduce zone_append_max_sectors attribute

Message ID 20240328004409.594888-19-dlemoal@kernel.org (mailing list archive)
State New, archived
Headers show
Series Zone write plugging | expand

Commit Message

Damien Le Moal March 28, 2024, 12:43 a.m. UTC 
Add the zone_append_max_sectors configfs attribute and module parameter
to allow configuring the maximum number of 512B sectors of zone append
operations. This attribute is meaningful only for zoned null block
devices.

If not specified, the default is unchanged and the zoned device max
append sectors limit is set to the device max sectors limit.
If a non 0 value is used for this attribute, which is the default,
then native support for zone append operations is enabled.
Setting a 0 value disables native zone append operations support to
instead use the block layer emulation.

Signed-off-by: Damien Le Moal <dlemoal@kernel.org>
Reviewed-by: Hannes Reinecke <hare@suse.de>
---
 drivers/block/null_blk/main.c     | 12 ++++++++++--
 drivers/block/null_blk/null_blk.h |  1 +
 drivers/block/null_blk/zoned.c    | 22 +++++++++++++++++++---
 3 files changed, 30 insertions(+), 5 deletions(-)

Comments

Christoph Hellwig March 28, 2024, 4:51 a.m. UTC | #1 
Looks good:

Reviewed-by: Christoph Hellwig <hch@lst.de>
Bart Van Assche March 29, 2024, 9:35 p.m. UTC | #2 
On 3/27/24 5:43 PM, Damien Le Moal wrote:
> @@ -1953,7 +1961,7 @@ static int null_add_dev(struct nullb_device *dev)
>   
>   	rv = add_disk(nullb->disk);
>   	if (rv)
> -		goto out_ida_free;
> +		goto out_cleanup_zone;
>   

The above change is unrelated to the introduction of
zone_append_max_sectors and hence should not be in this patch.
Additionally, the order of cleanup actions in the error path seems
wrong to me. null_init_zoned_dev() is called before blk_mq_alloc_disk().
Hence, the put_disk() call in the error path should occur before the
null_free_zoned_dev() call.

Thanks,

Bart.
Damien Le Moal March 30, 2024, 12:33 a.m. UTC | #3 
On 3/30/24 06:35, Bart Van Assche wrote:
> On 3/27/24 5:43 PM, Damien Le Moal wrote:
>> @@ -1953,7 +1961,7 @@ static int null_add_dev(struct nullb_device *dev)
>>   
>>   	rv = add_disk(nullb->disk);
>>   	if (rv)
>> -		goto out_ida_free;
>> +		goto out_cleanup_zone;
>>   
> 
> The above change is unrelated to the introduction of
> zone_append_max_sectors and hence should not be in this patch.

Good catch. That is a bug of this patch. I removed this change as it is incorrect.

> Additionally, the order of cleanup actions in the error path seems
> wrong to me. null_init_zoned_dev() is called before blk_mq_alloc_disk().
> Hence, the put_disk() call in the error path should occur before the
> null_free_zoned_dev() call.

That is separate from this patch (and the series). But I will send a fix patch
for that.
Chaitanya Kulkarni April 2, 2024, 6:44 a.m. UTC | #4 
On 3/27/24 17:43, Damien Le Moal wrote:
> Add the zone_append_max_sectors configfs attribute and module parameter
> to allow configuring the maximum number of 512B sectors of zone append
> operations. This attribute is meaningful only for zoned null block
> devices.
> 
> If not specified, the default is unchanged and the zoned device max
> append sectors limit is set to the device max sectors limit.
> If a non 0 value is used for this attribute, which is the default,
> then native support for zone append operations is enabled.
> Setting a 0 value disables native zone append operations support to
> instead use the block layer emulation.
> 
> Signed-off-by: Damien Le Moal<dlemoal@kernel.org>
> Reviewed-by: Hannes Reinecke<hare@suse.de>


Looks good.

Reviewed-by: Chaitanya Kulkarni <kch@nvidia.com>

-ck
diff mbox series

Patch

diff --git a/drivers/block/null_blk/main.c b/drivers/block/null_blk/main.c
index 71c39bcd872c..cd6519360dbc 100644
--- a/drivers/block/null_blk/main.c
+++ b/drivers/block/null_blk/main.c
@@ -253,6 +253,11 @@  static unsigned int g_zone_max_active;
 module_param_named(zone_max_active, g_zone_max_active, uint, 0444);
 MODULE_PARM_DESC(zone_max_active, "Maximum number of active zones when block device is zoned. Default: 0 (no limit)");
 
+static int g_zone_append_max_sectors = INT_MAX;
+module_param_named(zone_append_max_sectors, g_zone_append_max_sectors, int, 0444);
+MODULE_PARM_DESC(zone_append_max_sectors,
+		 "Maximum size of a zone append command (in 512B sectors). Specify 0 for zone append emulation");
+
 static struct nullb_device *null_alloc_dev(void);
 static void null_free_dev(struct nullb_device *dev);
 static void null_del_dev(struct nullb *nullb);
@@ -436,6 +441,7 @@  NULLB_DEVICE_ATTR(zone_capacity, ulong, NULL);
 NULLB_DEVICE_ATTR(zone_nr_conv, uint, NULL);
 NULLB_DEVICE_ATTR(zone_max_open, uint, NULL);
 NULLB_DEVICE_ATTR(zone_max_active, uint, NULL);
+NULLB_DEVICE_ATTR(zone_append_max_sectors, uint, NULL);
 NULLB_DEVICE_ATTR(virt_boundary, bool, NULL);
 NULLB_DEVICE_ATTR(no_sched, bool, NULL);
 NULLB_DEVICE_ATTR(shared_tags, bool, NULL);
@@ -580,6 +586,7 @@  static struct configfs_attribute *nullb_device_attrs[] = {
 	&nullb_device_attr_zone_nr_conv,
 	&nullb_device_attr_zone_max_open,
 	&nullb_device_attr_zone_max_active,
+	&nullb_device_attr_zone_append_max_sectors,
 	&nullb_device_attr_zone_readonly,
 	&nullb_device_attr_zone_offline,
 	&nullb_device_attr_virt_boundary,
@@ -671,7 +678,7 @@  static ssize_t memb_group_features_show(struct config_item *item, char *page)
 			"shared_tags,size,submit_queues,use_per_node_hctx,"
 			"virt_boundary,zoned,zone_capacity,zone_max_active,"
 			"zone_max_open,zone_nr_conv,zone_offline,zone_readonly,"
-			"zone_size\n");
+			"zone_size,zone_append_max_sectors\n");
 }
 
 CONFIGFS_ATTR_RO(memb_group_, features);
@@ -751,6 +758,7 @@  static struct nullb_device *null_alloc_dev(void)
 	dev->zone_nr_conv = g_zone_nr_conv;
 	dev->zone_max_open = g_zone_max_open;
 	dev->zone_max_active = g_zone_max_active;
+	dev->zone_append_max_sectors = g_zone_append_max_sectors;
 	dev->virt_boundary = g_virt_boundary;
 	dev->no_sched = g_no_sched;
 	dev->shared_tags = g_shared_tags;
@@ -1953,7 +1961,7 @@  static int null_add_dev(struct nullb_device *dev)
 
 	rv = add_disk(nullb->disk);
 	if (rv)
-		goto out_ida_free;
+		goto out_cleanup_zone;
 
 	mutex_lock(&lock);
 	list_add_tail(&nullb->list, &nullb_list);
diff --git a/drivers/block/null_blk/null_blk.h b/drivers/block/null_blk/null_blk.h
index 477b97746823..a9c5df650ddb 100644
--- a/drivers/block/null_blk/null_blk.h
+++ b/drivers/block/null_blk/null_blk.h
@@ -82,6 +82,7 @@  struct nullb_device {
 	unsigned int zone_nr_conv; /* number of conventional zones */
 	unsigned int zone_max_open; /* max number of open zones */
 	unsigned int zone_max_active; /* max number of active zones */
+	unsigned int zone_append_max_sectors; /* Max sectors per zone append command */
 	unsigned int submit_queues; /* number of submission queues */
 	unsigned int prev_submit_queues; /* number of submission queues before change */
 	unsigned int poll_queues; /* number of IOPOLL submission queues */
diff --git a/drivers/block/null_blk/zoned.c b/drivers/block/null_blk/zoned.c
index 8e217f8fadcd..159746b0661c 100644
--- a/drivers/block/null_blk/zoned.c
+++ b/drivers/block/null_blk/zoned.c
@@ -62,6 +62,7 @@  int null_init_zoned_dev(struct nullb_device *dev,
 			struct queue_limits *lim)
 {
 	sector_t dev_capacity_sects, zone_capacity_sects;
+	sector_t zone_append_max_bytes;
 	struct nullb_zone *zone;
 	sector_t sector = 0;
 	unsigned int i;
@@ -103,6 +104,12 @@  int null_init_zoned_dev(struct nullb_device *dev,
 			dev->zone_nr_conv);
 	}
 
+	zone_append_max_bytes =
+		ALIGN_DOWN(dev->zone_append_max_sectors << SECTOR_SHIFT,
+			   dev->blocksize);
+	dev->zone_append_max_sectors =
+		min(zone_append_max_bytes >> SECTOR_SHIFT, zone_capacity_sects);
+
 	/* Max active zones has to be < nbr of seq zones in order to be enforceable */
 	if (dev->zone_max_active >= dev->nr_zones - dev->zone_nr_conv) {
 		dev->zone_max_active = 0;
@@ -154,7 +161,7 @@  int null_init_zoned_dev(struct nullb_device *dev,
 
 	lim->zoned = true;
 	lim->chunk_sectors = dev->zone_size_sects;
-	lim->max_zone_append_sectors = dev->zone_size_sects;
+	lim->max_zone_append_sectors = dev->zone_append_max_sectors;
 	lim->max_open_zones = dev->zone_max_open;
 	lim->max_active_zones = dev->zone_max_active;
 	return 0;
@@ -163,10 +170,16 @@  int null_init_zoned_dev(struct nullb_device *dev,
 int null_register_zoned_dev(struct nullb *nullb)
 {
 	struct request_queue *q = nullb->q;
+	struct gendisk *disk = nullb->disk;
 
 	blk_queue_flag_set(QUEUE_FLAG_ZONE_RESETALL, q);
-	nullb->disk->nr_zones = bdev_nr_zones(nullb->disk->part0);
-	return blk_revalidate_disk_zones(nullb->disk, NULL);
+	disk->nr_zones = bdev_nr_zones(disk->part0);
+
+	pr_info("%s: using %s zone append\n",
+		disk->disk_name,
+		queue_emulates_zone_append(q) ? "emulated" : "native");
+
+	return blk_revalidate_disk_zones(disk, NULL);
 }
 
 void null_free_zoned_dev(struct nullb_device *dev)
@@ -365,6 +378,9 @@  static blk_status_t null_zone_write(struct nullb_cmd *cmd, sector_t sector,
 
 	trace_nullb_zone_op(cmd, zno, zone->cond);
 
+	if (WARN_ON_ONCE(append && !dev->zone_append_max_sectors))
+		return BLK_STS_IOERR;
+
 	if (zone->type == BLK_ZONE_TYPE_CONVENTIONAL) {
 		if (append)
 			return BLK_STS_IOERR;