summaryrefslogtreecommitdiff
path: root/drivers/block/virtio_blk.c
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/block/virtio_blk.c')
-rw-r--r--drivers/block/virtio_blk.c466
1 files changed, 242 insertions, 224 deletions
diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index 5bf98fd6a651..30bca8cb7106 100644
--- a/drivers/block/virtio_blk.c
+++ b/drivers/block/virtio_blk.c
@@ -13,7 +13,6 @@
#include <linux/string_helpers.h>
#include <linux/idr.h>
#include <linux/blk-mq.h>
-#include <linux/blk-mq-virtio.h>
#include <linux/numa.h>
#include <linux/vmalloc.h>
#include <uapi/linux/virtio_ring.h>
@@ -227,7 +226,7 @@ static int virtblk_map_data(struct blk_mq_hw_ctx *hctx, struct request *req,
if (unlikely(err))
return -ENOMEM;
- return blk_rq_map_sg(hctx->queue, req, vbr->sg_table.sgl);
+ return blk_rq_map_sg(req, vbr->sg_table.sgl);
}
static void virtblk_cleanup_cmd(struct request *req)
@@ -471,18 +470,18 @@ static bool virtblk_prep_rq_batch(struct request *req)
return virtblk_prep_rq(req->mq_hctx, vblk, req, vbr) == BLK_STS_OK;
}
-static bool virtblk_add_req_batch(struct virtio_blk_vq *vq,
- struct request **rqlist)
+static void virtblk_add_req_batch(struct virtio_blk_vq *vq,
+ struct rq_list *rqlist)
{
+ struct request *req;
unsigned long flags;
- int err;
bool kick;
spin_lock_irqsave(&vq->lock, flags);
- while (!rq_list_empty(*rqlist)) {
- struct request *req = rq_list_pop(rqlist);
+ while ((req = rq_list_pop(rqlist))) {
struct virtblk_req *vbr = blk_mq_rq_to_pdu(req);
+ int err;
err = virtblk_add_req(vq->vq, vbr);
if (err) {
@@ -495,37 +494,32 @@ static bool virtblk_add_req_batch(struct virtio_blk_vq *vq,
kick = virtqueue_kick_prepare(vq->vq);
spin_unlock_irqrestore(&vq->lock, flags);
- return kick;
+ if (kick)
+ virtqueue_notify(vq->vq);
}
-static void virtio_queue_rqs(struct request **rqlist)
+static void virtio_queue_rqs(struct rq_list *rqlist)
{
- struct request *req, *next, *prev = NULL;
- struct request *requeue_list = NULL;
-
- rq_list_for_each_safe(rqlist, req, next) {
- struct virtio_blk_vq *vq = get_virtio_blk_vq(req->mq_hctx);
- bool kick;
-
- if (!virtblk_prep_rq_batch(req)) {
- rq_list_move(rqlist, &requeue_list, req, prev);
- req = prev;
- if (!req)
- continue;
- }
+ struct rq_list submit_list = { };
+ struct rq_list requeue_list = { };
+ struct virtio_blk_vq *vq = NULL;
+ struct request *req;
- if (!next || req->mq_hctx != next->mq_hctx) {
- req->rq_next = NULL;
- kick = virtblk_add_req_batch(vq, rqlist);
- if (kick)
- virtqueue_notify(vq->vq);
+ while ((req = rq_list_pop(rqlist))) {
+ struct virtio_blk_vq *this_vq = get_virtio_blk_vq(req->mq_hctx);
- *rqlist = next;
- prev = NULL;
- } else
- prev = req;
+ if (vq && vq != this_vq)
+ virtblk_add_req_batch(vq, &submit_list);
+ vq = this_vq;
+
+ if (virtblk_prep_rq_batch(req))
+ rq_list_add_tail(&submit_list, req);
+ else
+ rq_list_add_tail(&requeue_list, req);
}
+ if (vq)
+ virtblk_add_req_batch(vq, &submit_list);
*rqlist = requeue_list;
}
@@ -577,7 +571,7 @@ static int virtblk_submit_zone_report(struct virtio_blk *vblk,
vbr->out_hdr.type = cpu_to_virtio32(vblk->vdev, VIRTIO_BLK_T_ZONE_REPORT);
vbr->out_hdr.sector = cpu_to_virtio64(vblk->vdev, sector);
- err = blk_rq_map_kern(q, req, report_buf, report_len, GFP_KERNEL);
+ err = blk_rq_map_kern(req, report_buf, report_len, GFP_KERNEL);
if (err)
goto out;
@@ -720,25 +714,24 @@ fail_report:
return ret;
}
-static int virtblk_probe_zoned_device(struct virtio_device *vdev,
- struct virtio_blk *vblk,
- struct request_queue *q)
+static int virtblk_read_zoned_limits(struct virtio_blk *vblk,
+ struct queue_limits *lim)
{
+ struct virtio_device *vdev = vblk->vdev;
u32 v, wg;
dev_dbg(&vdev->dev, "probing host-managed zoned device\n");
- disk_set_zoned(vblk->disk);
- blk_queue_flag_set(QUEUE_FLAG_ZONE_RESETALL, q);
+ lim->features |= BLK_FEAT_ZONED;
virtio_cread(vdev, struct virtio_blk_config,
zoned.max_open_zones, &v);
- disk_set_max_open_zones(vblk->disk, v);
+ lim->max_open_zones = v;
dev_dbg(&vdev->dev, "max open zones = %u\n", v);
virtio_cread(vdev, struct virtio_blk_config,
zoned.max_active_zones, &v);
- disk_set_max_active_zones(vblk->disk, v);
+ lim->max_active_zones = v;
dev_dbg(&vdev->dev, "max active zones = %u\n", v);
virtio_cread(vdev, struct virtio_blk_config,
@@ -747,8 +740,8 @@ static int virtblk_probe_zoned_device(struct virtio_device *vdev,
dev_warn(&vdev->dev, "zero write granularity reported\n");
return -ENODEV;
}
- blk_queue_physical_block_size(q, wg);
- blk_queue_io_min(q, wg);
+ lim->physical_block_size = wg;
+ lim->io_min = wg;
dev_dbg(&vdev->dev, "write granularity = %u\n", wg);
@@ -764,13 +757,13 @@ static int virtblk_probe_zoned_device(struct virtio_device *vdev,
vblk->zone_sectors);
return -ENODEV;
}
- blk_queue_chunk_sectors(q, vblk->zone_sectors);
+ lim->chunk_sectors = vblk->zone_sectors;
dev_dbg(&vdev->dev, "zone sectors = %u\n", vblk->zone_sectors);
if (virtio_has_feature(vdev, VIRTIO_BLK_F_DISCARD)) {
dev_warn(&vblk->vdev->dev,
"ignoring negotiated F_DISCARD for zoned device\n");
- blk_queue_max_discard_sectors(q, 0);
+ lim->max_hw_discard_sectors = 0;
}
virtio_cread(vdev, struct virtio_blk_config,
@@ -785,25 +778,21 @@ static int virtblk_probe_zoned_device(struct virtio_device *vdev,
wg, v);
return -ENODEV;
}
- blk_queue_max_zone_append_sectors(q, v);
+ lim->max_hw_zone_append_sectors = v;
dev_dbg(&vdev->dev, "max append sectors = %u\n", v);
- return blk_revalidate_disk_zones(vblk->disk, NULL);
+ return 0;
}
-
#else
-
/*
- * Zoned block device support is not configured in this kernel.
- * Host-managed zoned devices can't be supported, but others are
- * good to go as regular block devices.
+ * Zoned block device support is not configured in this kernel, host-managed
+ * zoned devices can't be supported.
*/
#define virtblk_report_zones NULL
-
-static inline int virtblk_probe_zoned_device(struct virtio_device *vdev,
- struct virtio_blk *vblk, struct request_queue *q)
+static inline int virtblk_read_zoned_limits(struct virtio_blk *vblk,
+ struct queue_limits *lim)
{
- dev_err(&vdev->dev,
+ dev_err(&vblk->vdev->dev,
"virtio_blk: zoned devices are not supported");
return -EOPNOTSUPP;
}
@@ -828,7 +817,7 @@ static int virtblk_get_id(struct gendisk *disk, char *id_str)
vbr->out_hdr.type = cpu_to_virtio32(vblk->vdev, VIRTIO_BLK_T_GET_ID);
vbr->out_hdr.sector = 0;
- err = blk_rq_map_kern(q, req, id_str, VIRTIO_BLK_ID_BYTES, GFP_KERNEL);
+ err = blk_rq_map_kern(req, id_str, VIRTIO_BLK_ID_BYTES, GFP_KERNEL);
if (err)
goto out;
@@ -969,8 +958,7 @@ static int init_vq(struct virtio_blk *vblk)
{
int err;
unsigned short i;
- vq_callback_t **callbacks;
- const char **names;
+ struct virtqueue_info *vqs_info;
struct virtqueue **vqs;
unsigned short num_vqs;
unsigned short num_poll_vqs;
@@ -1007,28 +995,26 @@ static int init_vq(struct virtio_blk *vblk)
if (!vblk->vqs)
return -ENOMEM;
- names = kmalloc_array(num_vqs, sizeof(*names), GFP_KERNEL);
- callbacks = kmalloc_array(num_vqs, sizeof(*callbacks), GFP_KERNEL);
+ vqs_info = kcalloc(num_vqs, sizeof(*vqs_info), GFP_KERNEL);
vqs = kmalloc_array(num_vqs, sizeof(*vqs), GFP_KERNEL);
- if (!names || !callbacks || !vqs) {
+ if (!vqs_info || !vqs) {
err = -ENOMEM;
goto out;
}
for (i = 0; i < num_vqs - num_poll_vqs; i++) {
- callbacks[i] = virtblk_done;
+ vqs_info[i].callback = virtblk_done;
snprintf(vblk->vqs[i].name, VQ_NAME_LEN, "req.%u", i);
- names[i] = vblk->vqs[i].name;
+ vqs_info[i].name = vblk->vqs[i].name;
}
for (; i < num_vqs; i++) {
- callbacks[i] = NULL;
snprintf(vblk->vqs[i].name, VQ_NAME_LEN, "req_poll.%u", i);
- names[i] = vblk->vqs[i].name;
+ vqs_info[i].name = vblk->vqs[i].name;
}
/* Discover virtqueues and write information to configuration. */
- err = virtio_find_vqs(vdev, num_vqs, vqs, callbacks, names, &desc);
+ err = virtio_find_vqs(vdev, num_vqs, vqs, vqs_info, &desc);
if (err)
goto out;
@@ -1040,8 +1026,7 @@ static int init_vq(struct virtio_blk *vblk)
out:
kfree(vqs);
- kfree(callbacks);
- kfree(names);
+ kfree(vqs_info);
if (err)
kfree(vblk->vqs);
return err;
@@ -1094,14 +1079,6 @@ static int virtblk_get_cache_mode(struct virtio_device *vdev)
return writeback;
}
-static void virtblk_update_cache_mode(struct virtio_device *vdev)
-{
- u8 writeback = virtblk_get_cache_mode(vdev);
- struct virtio_blk *vblk = vdev->priv;
-
- blk_queue_write_cache(vblk->disk->queue, writeback, false);
-}
-
static const char *const virtblk_cache_types[] = {
"write through", "write back"
};
@@ -1113,6 +1090,7 @@ cache_type_store(struct device *dev, struct device_attribute *attr,
struct gendisk *disk = dev_to_disk(dev);
struct virtio_blk *vblk = disk->private_data;
struct virtio_device *vdev = vblk->vdev;
+ struct queue_limits lim;
int i;
BUG_ON(!virtio_has_feature(vblk->vdev, VIRTIO_BLK_F_CONFIG_WCE));
@@ -1121,7 +1099,15 @@ cache_type_store(struct device *dev, struct device_attribute *attr,
return i;
virtio_cwrite8(vdev, offsetof(struct virtio_blk_config, wce), i);
- virtblk_update_cache_mode(vdev);
+
+ lim = queue_limits_start_update(disk->queue);
+ if (virtblk_get_cache_mode(vdev))
+ lim.features |= BLK_FEAT_WRITE_CACHE;
+ else
+ lim.features &= ~BLK_FEAT_WRITE_CACHE;
+ i = queue_limits_commit_update_frozen(disk->queue, &lim);
+ if (i)
+ return i;
return count;
}
@@ -1192,7 +1178,8 @@ static void virtblk_map_queues(struct blk_mq_tag_set *set)
if (i == HCTX_TYPE_POLL)
blk_mq_map_queues(&set->map[i]);
else
- blk_mq_virtio_map_queues(&set->map[i], vblk->vdev, 0);
+ blk_mq_map_hw_queues(&set->map[i],
+ &vblk->vdev->dev, 0);
}
}
@@ -1220,11 +1207,12 @@ static int virtblk_poll(struct blk_mq_hw_ctx *hctx, struct io_comp_batch *iob)
while ((vbr = virtqueue_get_buf(vq->vq, &len)) != NULL) {
struct request *req = blk_mq_rq_from_pdu(vbr);
+ u8 status = virtblk_vbr_status(vbr);
found++;
if (!blk_mq_complete_request_remote(req) &&
- !blk_mq_add_to_batch(req, iob, virtblk_vbr_status(vbr),
- virtblk_complete_batch))
+ !blk_mq_add_to_batch(req, iob, status != VIRTIO_BLK_S_OK,
+ virtblk_complete_batch))
virtblk_request_done(req);
}
@@ -1248,31 +1236,17 @@ static const struct blk_mq_ops virtio_mq_ops = {
static unsigned int virtblk_queue_depth;
module_param_named(queue_depth, virtblk_queue_depth, uint, 0444);
-static int virtblk_probe(struct virtio_device *vdev)
+static int virtblk_read_limits(struct virtio_blk *vblk,
+ struct queue_limits *lim)
{
- struct virtio_blk *vblk;
- struct request_queue *q;
- int err, index;
-
- u32 v, blk_size, max_size, sg_elems, opt_io_size;
+ struct virtio_device *vdev = vblk->vdev;
+ u32 v, max_size, sg_elems, opt_io_size;
u32 max_discard_segs = 0;
u32 discard_granularity = 0;
u16 min_io_size;
u8 physical_block_exp, alignment_offset;
- unsigned int queue_depth;
size_t max_dma_size;
-
- if (!vdev->config->get) {
- dev_err(&vdev->dev, "%s failure: config access disabled\n",
- __func__);
- return -EINVAL;
- }
-
- err = ida_alloc_range(&vd_index_ida, 0,
- minor_to_index(1 << MINORBITS) - 1, GFP_KERNEL);
- if (err < 0)
- goto out;
- index = err;
+ int err;
/* We need to know how many segments before we allocate. */
err = virtio_cread_feature(vdev, VIRTIO_BLK_F_SEG_MAX,
@@ -1286,78 +1260,11 @@ static int virtblk_probe(struct virtio_device *vdev)
/* Prevent integer overflows and honor max vq size */
sg_elems = min_t(u32, sg_elems, VIRTIO_BLK_MAX_SG_ELEMS - 2);
- vdev->priv = vblk = kmalloc(sizeof(*vblk), GFP_KERNEL);
- if (!vblk) {
- err = -ENOMEM;
- goto out_free_index;
- }
-
- mutex_init(&vblk->vdev_mutex);
-
- vblk->vdev = vdev;
-
- INIT_WORK(&vblk->config_work, virtblk_config_changed_work);
-
- err = init_vq(vblk);
- if (err)
- goto out_free_vblk;
-
- /* Default queue sizing is to fill the ring. */
- if (!virtblk_queue_depth) {
- queue_depth = vblk->vqs[0].vq->num_free;
- /* ... but without indirect descs, we use 2 descs per req */
- if (!virtio_has_feature(vdev, VIRTIO_RING_F_INDIRECT_DESC))
- queue_depth /= 2;
- } else {
- queue_depth = virtblk_queue_depth;
- }
-
- memset(&vblk->tag_set, 0, sizeof(vblk->tag_set));
- vblk->tag_set.ops = &virtio_mq_ops;
- vblk->tag_set.queue_depth = queue_depth;
- vblk->tag_set.numa_node = NUMA_NO_NODE;
- vblk->tag_set.flags = BLK_MQ_F_SHOULD_MERGE;
- vblk->tag_set.cmd_size =
- sizeof(struct virtblk_req) +
- sizeof(struct scatterlist) * VIRTIO_BLK_INLINE_SG_CNT;
- vblk->tag_set.driver_data = vblk;
- vblk->tag_set.nr_hw_queues = vblk->num_vqs;
- vblk->tag_set.nr_maps = 1;
- if (vblk->io_queues[HCTX_TYPE_POLL])
- vblk->tag_set.nr_maps = 3;
-
- err = blk_mq_alloc_tag_set(&vblk->tag_set);
- if (err)
- goto out_free_vq;
-
- vblk->disk = blk_mq_alloc_disk(&vblk->tag_set, vblk);
- if (IS_ERR(vblk->disk)) {
- err = PTR_ERR(vblk->disk);
- goto out_free_tags;
- }
- q = vblk->disk->queue;
-
- virtblk_name_format("vd", index, vblk->disk->disk_name, DISK_NAME_LEN);
-
- vblk->disk->major = major;
- vblk->disk->first_minor = index_to_minor(index);
- vblk->disk->minors = 1 << PART_BITS;
- vblk->disk->private_data = vblk;
- vblk->disk->fops = &virtblk_fops;
- vblk->index = index;
-
- /* configure queue flush support */
- virtblk_update_cache_mode(vdev);
-
- /* If disk is read-only in the host, the guest should obey */
- if (virtio_has_feature(vdev, VIRTIO_BLK_F_RO))
- set_disk_ro(vblk->disk, 1);
-
/* We can handle whatever the host told us to handle. */
- blk_queue_max_segments(q, sg_elems);
+ lim->max_segments = sg_elems;
/* No real sector limit. */
- blk_queue_max_hw_sectors(q, UINT_MAX);
+ lim->max_hw_sectors = UINT_MAX;
max_dma_size = virtio_max_dma_size(vdev);
max_size = max_dma_size > U32_MAX ? U32_MAX : max_dma_size;
@@ -1369,50 +1276,39 @@ static int virtblk_probe(struct virtio_device *vdev)
if (!err)
max_size = min(max_size, v);
- blk_queue_max_segment_size(q, max_size);
+ lim->max_segment_size = max_size;
/* Host can optionally specify the block size of the device */
- err = virtio_cread_feature(vdev, VIRTIO_BLK_F_BLK_SIZE,
+ virtio_cread_feature(vdev, VIRTIO_BLK_F_BLK_SIZE,
struct virtio_blk_config, blk_size,
- &blk_size);
- if (!err) {
- err = blk_validate_block_size(blk_size);
- if (err) {
- dev_err(&vdev->dev,
- "virtio_blk: invalid block size: 0x%x\n",
- blk_size);
- goto out_cleanup_disk;
- }
-
- blk_queue_logical_block_size(q, blk_size);
- } else
- blk_size = queue_logical_block_size(q);
+ &lim->logical_block_size);
/* Use topology information if available */
err = virtio_cread_feature(vdev, VIRTIO_BLK_F_TOPOLOGY,
struct virtio_blk_config, physical_block_exp,
&physical_block_exp);
if (!err && physical_block_exp)
- blk_queue_physical_block_size(q,
- blk_size * (1 << physical_block_exp));
+ lim->physical_block_size =
+ lim->logical_block_size * (1 << physical_block_exp);
err = virtio_cread_feature(vdev, VIRTIO_BLK_F_TOPOLOGY,
struct virtio_blk_config, alignment_offset,
&alignment_offset);
if (!err && alignment_offset)
- blk_queue_alignment_offset(q, blk_size * alignment_offset);
+ lim->alignment_offset =
+ lim->logical_block_size * alignment_offset;
err = virtio_cread_feature(vdev, VIRTIO_BLK_F_TOPOLOGY,
struct virtio_blk_config, min_io_size,
&min_io_size);
if (!err && min_io_size)
- blk_queue_io_min(q, blk_size * min_io_size);
+ lim->io_min = lim->logical_block_size * min_io_size;
err = virtio_cread_feature(vdev, VIRTIO_BLK_F_TOPOLOGY,
struct virtio_blk_config, opt_io_size,
&opt_io_size);
if (!err && opt_io_size)
- blk_queue_io_opt(q, blk_size * opt_io_size);
+ lim->io_opt = lim->logical_block_size * opt_io_size;
if (virtio_has_feature(vdev, VIRTIO_BLK_F_DISCARD)) {
virtio_cread(vdev, struct virtio_blk_config,
@@ -1420,7 +1316,7 @@ static int virtblk_probe(struct virtio_device *vdev)
virtio_cread(vdev, struct virtio_blk_config,
max_discard_sectors, &v);
- blk_queue_max_discard_sectors(q, v ? v : UINT_MAX);
+ lim->max_hw_discard_sectors = v ? v : UINT_MAX;
virtio_cread(vdev, struct virtio_blk_config, max_discard_seg,
&max_discard_segs);
@@ -1429,7 +1325,7 @@ static int virtblk_probe(struct virtio_device *vdev)
if (virtio_has_feature(vdev, VIRTIO_BLK_F_WRITE_ZEROES)) {
virtio_cread(vdev, struct virtio_blk_config,
max_write_zeroes_sectors, &v);
- blk_queue_max_write_zeroes_sectors(q, v ? v : UINT_MAX);
+ lim->max_write_zeroes_sectors = v ? v : UINT_MAX;
}
/* The discard and secure erase limits are combined since the Linux
@@ -1455,8 +1351,7 @@ static int virtblk_probe(struct virtio_device *vdev)
if (!v) {
dev_err(&vdev->dev,
"virtio_blk: secure_erase_sector_alignment can't be 0\n");
- err = -EINVAL;
- goto out_cleanup_disk;
+ return -EINVAL;
}
discard_granularity = min_not_zero(discard_granularity, v);
@@ -1470,11 +1365,10 @@ static int virtblk_probe(struct virtio_device *vdev)
if (!v) {
dev_err(&vdev->dev,
"virtio_blk: max_secure_erase_sectors can't be 0\n");
- err = -EINVAL;
- goto out_cleanup_disk;
+ return -EINVAL;
}
- blk_queue_max_secure_erase_sectors(q, v);
+ lim->max_secure_erase_sectors = v;
virtio_cread(vdev, struct virtio_blk_config,
max_secure_erase_seg, &v);
@@ -1485,8 +1379,7 @@ static int virtblk_probe(struct virtio_device *vdev)
if (!v) {
dev_err(&vdev->dev,
"virtio_blk: max_secure_erase_seg can't be 0\n");
- err = -EINVAL;
- goto out_cleanup_disk;
+ return -EINVAL;
}
max_discard_segs = min_not_zero(max_discard_segs, v);
@@ -1502,45 +1395,144 @@ static int virtblk_probe(struct virtio_device *vdev)
if (!max_discard_segs)
max_discard_segs = sg_elems;
- blk_queue_max_discard_segments(q,
- min(max_discard_segs, MAX_DISCARD_SEGMENTS));
+ lim->max_discard_segments =
+ min(max_discard_segs, MAX_DISCARD_SEGMENTS);
if (discard_granularity)
- q->limits.discard_granularity = discard_granularity << SECTOR_SHIFT;
+ lim->discard_granularity =
+ discard_granularity << SECTOR_SHIFT;
else
- q->limits.discard_granularity = blk_size;
+ lim->discard_granularity = lim->logical_block_size;
}
- virtblk_update_capacity(vblk, false);
- virtio_device_ready(vdev);
-
- /*
- * All steps that follow use the VQs therefore they need to be
- * placed after the virtio_device_ready() call above.
- */
if (virtio_has_feature(vdev, VIRTIO_BLK_F_ZONED)) {
u8 model;
- virtio_cread(vdev, struct virtio_blk_config, zoned.model,
- &model);
+ virtio_cread(vdev, struct virtio_blk_config, zoned.model, &model);
switch (model) {
case VIRTIO_BLK_Z_NONE:
case VIRTIO_BLK_Z_HA:
- /* Present the host-aware device as non-zoned */
- break;
+ /* treat host-aware devices as non-zoned */
+ return 0;
case VIRTIO_BLK_Z_HM:
- err = virtblk_probe_zoned_device(vdev, vblk, q);
+ err = virtblk_read_zoned_limits(vblk, lim);
if (err)
- goto out_cleanup_disk;
+ return err;
break;
default:
- dev_err(&vdev->dev, "unsupported zone model %d\n",
- model);
- err = -EINVAL;
- goto out_cleanup_disk;
+ dev_err(&vdev->dev, "unsupported zone model %d\n", model);
+ return -EINVAL;
}
}
+ return 0;
+}
+
+static int virtblk_probe(struct virtio_device *vdev)
+{
+ struct virtio_blk *vblk;
+ struct queue_limits lim = {
+ .features = BLK_FEAT_ROTATIONAL,
+ .logical_block_size = SECTOR_SIZE,
+ };
+ int err, index;
+ unsigned int queue_depth;
+
+ if (!vdev->config->get) {
+ dev_err(&vdev->dev, "%s failure: config access disabled\n",
+ __func__);
+ return -EINVAL;
+ }
+
+ err = ida_alloc_range(&vd_index_ida, 0,
+ minor_to_index(1 << MINORBITS) - 1, GFP_KERNEL);
+ if (err < 0)
+ goto out;
+ index = err;
+
+ vdev->priv = vblk = kmalloc(sizeof(*vblk), GFP_KERNEL);
+ if (!vblk) {
+ err = -ENOMEM;
+ goto out_free_index;
+ }
+
+ mutex_init(&vblk->vdev_mutex);
+
+ vblk->vdev = vdev;
+
+ INIT_WORK(&vblk->config_work, virtblk_config_changed_work);
+
+ err = init_vq(vblk);
+ if (err)
+ goto out_free_vblk;
+
+ /* Default queue sizing is to fill the ring. */
+ if (!virtblk_queue_depth) {
+ queue_depth = vblk->vqs[0].vq->num_free;
+ /* ... but without indirect descs, we use 2 descs per req */
+ if (!virtio_has_feature(vdev, VIRTIO_RING_F_INDIRECT_DESC))
+ queue_depth /= 2;
+ } else {
+ queue_depth = virtblk_queue_depth;
+ }
+
+ memset(&vblk->tag_set, 0, sizeof(vblk->tag_set));
+ vblk->tag_set.ops = &virtio_mq_ops;
+ vblk->tag_set.queue_depth = queue_depth;
+ vblk->tag_set.numa_node = NUMA_NO_NODE;
+ vblk->tag_set.cmd_size =
+ sizeof(struct virtblk_req) +
+ sizeof(struct scatterlist) * VIRTIO_BLK_INLINE_SG_CNT;
+ vblk->tag_set.driver_data = vblk;
+ vblk->tag_set.nr_hw_queues = vblk->num_vqs;
+ vblk->tag_set.nr_maps = 1;
+ if (vblk->io_queues[HCTX_TYPE_POLL])
+ vblk->tag_set.nr_maps = 3;
+
+ err = blk_mq_alloc_tag_set(&vblk->tag_set);
+ if (err)
+ goto out_free_vq;
+
+ err = virtblk_read_limits(vblk, &lim);
+ if (err)
+ goto out_free_tags;
+
+ if (virtblk_get_cache_mode(vdev))
+ lim.features |= BLK_FEAT_WRITE_CACHE;
+
+ vblk->disk = blk_mq_alloc_disk(&vblk->tag_set, &lim, vblk);
+ if (IS_ERR(vblk->disk)) {
+ err = PTR_ERR(vblk->disk);
+ goto out_free_tags;
+ }
+
+ virtblk_name_format("vd", index, vblk->disk->disk_name, DISK_NAME_LEN);
+
+ vblk->disk->major = major;
+ vblk->disk->first_minor = index_to_minor(index);
+ vblk->disk->minors = 1 << PART_BITS;
+ vblk->disk->private_data = vblk;
+ vblk->disk->fops = &virtblk_fops;
+ vblk->index = index;
+
+ /* If disk is read-only in the host, the guest should obey */
+ if (virtio_has_feature(vdev, VIRTIO_BLK_F_RO))
+ set_disk_ro(vblk->disk, 1);
+
+ virtblk_update_capacity(vblk, false);
+ virtio_device_ready(vdev);
+
+ /*
+ * All steps that follow use the VQs therefore they need to be
+ * placed after the virtio_device_ready() call above.
+ */
+ if (IS_ENABLED(CONFIG_BLK_DEV_ZONED) &&
+ (lim.features & BLK_FEAT_ZONED)) {
+ err = blk_revalidate_disk_zones(vblk->disk);
+ if (err)
+ goto out_cleanup_disk;
+ }
+
err = device_add_disk(&vdev->dev, vblk->disk, virtblk_attr_groups);
if (err)
goto out_cleanup_disk;
@@ -1588,10 +1580,16 @@ static void virtblk_remove(struct virtio_device *vdev)
put_disk(vblk->disk);
}
-#ifdef CONFIG_PM_SLEEP
-static int virtblk_freeze(struct virtio_device *vdev)
+static int virtblk_freeze_priv(struct virtio_device *vdev)
{
struct virtio_blk *vblk = vdev->priv;
+ struct request_queue *q = vblk->disk->queue;
+ unsigned int memflags;
+
+ /* Ensure no requests in virtqueues before deleting vqs. */
+ memflags = blk_mq_freeze_queue(q);
+ blk_mq_quiesce_queue_nowait(q);
+ blk_mq_unfreeze_queue(q, memflags);
/* Ensure we don't receive any more interrupts */
virtio_reset_device(vdev);
@@ -1599,15 +1597,13 @@ static int virtblk_freeze(struct virtio_device *vdev)
/* Make sure no work handler is accessing the device. */
flush_work(&vblk->config_work);
- blk_mq_quiesce_queue(vblk->disk->queue);
-
vdev->config->del_vqs(vdev);
kfree(vblk->vqs);
return 0;
}
-static int virtblk_restore(struct virtio_device *vdev)
+static int virtblk_restore_priv(struct virtio_device *vdev)
{
struct virtio_blk *vblk = vdev->priv;
int ret;
@@ -1617,12 +1613,33 @@ static int virtblk_restore(struct virtio_device *vdev)
return ret;
virtio_device_ready(vdev);
-
blk_mq_unquiesce_queue(vblk->disk->queue);
+
return 0;
}
+
+#ifdef CONFIG_PM_SLEEP
+static int virtblk_freeze(struct virtio_device *vdev)
+{
+ return virtblk_freeze_priv(vdev);
+}
+
+static int virtblk_restore(struct virtio_device *vdev)
+{
+ return virtblk_restore_priv(vdev);
+}
#endif
+static int virtblk_reset_prepare(struct virtio_device *vdev)
+{
+ return virtblk_freeze_priv(vdev);
+}
+
+static int virtblk_reset_done(struct virtio_device *vdev)
+{
+ return virtblk_restore_priv(vdev);
+}
+
static const struct virtio_device_id id_table[] = {
{ VIRTIO_ID_BLOCK, VIRTIO_DEV_ANY_ID },
{ 0 },
@@ -1650,7 +1667,6 @@ static struct virtio_driver virtio_blk = {
.feature_table_legacy = features_legacy,
.feature_table_size_legacy = ARRAY_SIZE(features_legacy),
.driver.name = KBUILD_MODNAME,
- .driver.owner = THIS_MODULE,
.id_table = id_table,
.probe = virtblk_probe,
.remove = virtblk_remove,
@@ -1659,6 +1675,8 @@ static struct virtio_driver virtio_blk = {
.freeze = virtblk_freeze,
.restore = virtblk_restore,
#endif
+ .reset_prepare = virtblk_reset_prepare,
+ .reset_done = virtblk_reset_done,
};
static int __init virtio_blk_init(void)