diff options
Diffstat (limited to 'drivers/block')
| -rw-r--r-- | drivers/block/Kconfig | 9 | ||||
| -rw-r--r-- | drivers/block/Makefile | 2 | ||||
| -rw-r--r-- | drivers/block/aoe/aoecmd.c | 12 | ||||
| -rw-r--r-- | drivers/block/aoe/aoenet.c | 1 | ||||
| -rw-r--r-- | drivers/block/drbd/drbd_main.c | 4 | ||||
| -rw-r--r-- | drivers/block/drbd/drbd_nl.c | 6 | ||||
| -rw-r--r-- | drivers/block/drbd/drbd_receiver.c | 2 | ||||
| -rw-r--r-- | drivers/block/loop.c | 182 | ||||
| -rw-r--r-- | drivers/block/nbd.c | 19 | ||||
| -rw-r--r-- | drivers/block/null_blk_main.c | 14 | ||||
| -rw-r--r-- | drivers/block/rbd.c | 537 | ||||
| -rw-r--r-- | drivers/block/sunvdc.c | 2 | ||||
| -rw-r--r-- | drivers/block/sx8.c | 1586 | ||||
| -rw-r--r-- | drivers/block/virtio_blk.c | 7 | ||||
| -rw-r--r-- | drivers/block/xen-blkfront.c | 3 |
15 files changed, 450 insertions, 1936 deletions
diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig index 8a08cbb958d1..c867211f5ed7 100644 --- a/drivers/block/Kconfig +++ b/drivers/block/Kconfig @@ -299,15 +299,6 @@ config BLK_DEV_SKD Use device /dev/skd$N amd /dev/skd$Np$M. -config BLK_DEV_SX8 - tristate "Promise SATA SX8 support" - depends on PCI - ---help--- - Saying Y or M here will enable support for the - Promise SATA SX8 controllers. - - Use devices /dev/sx8/$N and /dev/sx8/$Np$M. - config BLK_DEV_RAM tristate "RAM block device support" ---help--- diff --git a/drivers/block/Makefile b/drivers/block/Makefile index a53cc1e3a2d3..d8f2f72c3b62 100644 --- a/drivers/block/Makefile +++ b/drivers/block/Makefile @@ -26,8 +26,6 @@ obj-$(CONFIG_BLK_DEV_NBD) += nbd.o obj-$(CONFIG_BLK_DEV_CRYPTOLOOP) += cryptoloop.o obj-$(CONFIG_VIRTIO_BLK) += virtio_blk.o -obj-$(CONFIG_BLK_DEV_SX8) += sx8.o - obj-$(CONFIG_XEN_BLKDEV_FRONTEND) += xen-blkfront.o obj-$(CONFIG_XEN_BLKDEV_BACKEND) += xen-blkback/ obj-$(CONFIG_BLK_DEV_DRBD) += drbd/ diff --git a/drivers/block/aoe/aoecmd.c b/drivers/block/aoe/aoecmd.c index 3cf9bc5d8d95..3d5117be57f9 100644 --- a/drivers/block/aoe/aoecmd.c +++ b/drivers/block/aoe/aoecmd.c @@ -420,13 +420,16 @@ aoecmd_cfg_pkts(ushort aoemajor, unsigned char aoeminor, struct sk_buff_head *qu rcu_read_lock(); for_each_netdev_rcu(&init_net, ifp) { dev_hold(ifp); - if (!is_aoe_netif(ifp)) - goto cont; + if (!is_aoe_netif(ifp)) { + dev_put(ifp); + continue; + } skb = new_skb(sizeof *h + sizeof *ch); if (skb == NULL) { printk(KERN_INFO "aoe: skb alloc failure\n"); - goto cont; + dev_put(ifp); + continue; } skb_put(skb, sizeof *h + sizeof *ch); skb->dev = ifp; @@ -441,9 +444,6 @@ aoecmd_cfg_pkts(ushort aoemajor, unsigned char aoeminor, struct sk_buff_head *qu h->major = cpu_to_be16(aoemajor); h->minor = aoeminor; h->cmd = AOECMD_CFG; - -cont: - dev_put(ifp); } rcu_read_unlock(); } diff --git a/drivers/block/aoe/aoenet.c b/drivers/block/aoe/aoenet.c index 63773a90581d..1e66c7a188a1 100644 --- a/drivers/block/aoe/aoenet.c +++ b/drivers/block/aoe/aoenet.c @@ -64,6 +64,7 @@ tx(int id) __must_hold(&txlock) pr_warn("aoe: packet could not be sent on %s. %s\n", ifp ? ifp->name : "netif", "consider increasing tx_queue_len"); + dev_put(ifp); spin_lock_irq(&txlock); } return 0; diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c index 5ece2fd70d9c..f3a96c76f5a4 100644 --- a/drivers/block/drbd/drbd_main.c +++ b/drivers/block/drbd/drbd_main.c @@ -2778,7 +2778,7 @@ static int init_submitter(struct drbd_device *device) enum drbd_ret_code drbd_create_device(struct drbd_config_context *adm_ctx, unsigned int minor) { struct drbd_resource *resource = adm_ctx->resource; - struct drbd_connection *connection; + struct drbd_connection *connection, *n; struct drbd_device *device; struct drbd_peer_device *peer_device, *tmp_peer_device; struct gendisk *disk; @@ -2906,7 +2906,7 @@ enum drbd_ret_code drbd_create_device(struct drbd_config_context *adm_ctx, unsig out_idr_remove_vol: idr_remove(&connection->peer_devices, vnr); out_idr_remove_from_resource: - for_each_connection(connection, resource) { + for_each_connection_safe(connection, n, resource) { peer_device = idr_remove(&connection->peer_devices, vnr); if (peer_device) kref_put(&connection->kref, drbd_destroy_connection); diff --git a/drivers/block/drbd/drbd_nl.c b/drivers/block/drbd/drbd_nl.c index 52dd517ff541..43c142ea364b 100644 --- a/drivers/block/drbd/drbd_nl.c +++ b/drivers/block/drbd/drbd_nl.c @@ -3426,7 +3426,7 @@ int drbd_adm_dump_devices(struct sk_buff *skb, struct netlink_callback *cb) { struct nlattr *resource_filter; struct drbd_resource *resource; - struct drbd_device *uninitialized_var(device); + struct drbd_device *device; int minor, err, retcode; struct drbd_genlmsghdr *dh; struct device_info device_info; @@ -3515,7 +3515,7 @@ int drbd_adm_dump_connections(struct sk_buff *skb, struct netlink_callback *cb) { struct nlattr *resource_filter; struct drbd_resource *resource = NULL, *next_resource; - struct drbd_connection *uninitialized_var(connection); + struct drbd_connection *connection; int err = 0, retcode; struct drbd_genlmsghdr *dh; struct connection_info connection_info; @@ -3677,7 +3677,7 @@ int drbd_adm_dump_peer_devices(struct sk_buff *skb, struct netlink_callback *cb) { struct nlattr *resource_filter; struct drbd_resource *resource; - struct drbd_device *uninitialized_var(device); + struct drbd_device *device; struct drbd_peer_device *peer_device = NULL; int minor, err, retcode; struct drbd_genlmsghdr *dh; diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c index 2b3103c30857..d94f41a0abbe 100644 --- a/drivers/block/drbd/drbd_receiver.c +++ b/drivers/block/drbd/drbd_receiver.c @@ -1298,7 +1298,7 @@ static void submit_one_flush(struct drbd_device *device, struct issue_flush_cont bio_set_dev(bio, device->ldev->backing_bdev); bio->bi_private = octx; bio->bi_end_io = one_flush_endio; - bio->bi_opf = REQ_OP_FLUSH | REQ_PREFLUSH; + bio->bi_opf = REQ_OP_WRITE | REQ_PREFLUSH; device->flush_jif = jiffies; set_bit(FLUSH_PENDING, &device->flags); diff --git a/drivers/block/loop.c b/drivers/block/loop.c index ad9b767ec40d..8396a4f99526 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -226,24 +226,30 @@ static void __loop_update_dio(struct loop_device *lo, bool dio) blk_mq_unfreeze_queue(lo->lo_queue); } -static int -figure_loop_size(struct loop_device *lo, loff_t offset, loff_t sizelimit) +/** + * loop_set_size() - sets device size and notifies userspace + * @lo: struct loop_device to set the size for + * @size: new size of the loop device + * + * Callers must validate that the size passed into this function fits into + * a sector_t, eg using loop_validate_size() + */ +static void loop_set_size(struct loop_device *lo, loff_t size) { - loff_t size = get_size(offset, sizelimit, lo->lo_backing_file); - sector_t x = (sector_t)size; struct block_device *bdev = lo->lo_device; - if (unlikely((loff_t)x != size)) - return -EFBIG; - if (lo->lo_offset != offset) - lo->lo_offset = offset; - if (lo->lo_sizelimit != sizelimit) - lo->lo_sizelimit = sizelimit; - set_capacity(lo->lo_disk, x); - bd_set_size(bdev, (loff_t)get_capacity(bdev->bd_disk) << 9); + set_capacity(lo->lo_disk, size); + bd_set_size(bdev, size << SECTOR_SHIFT); /* let user-space know about the new size */ kobject_uevent(&disk_to_dev(bdev->bd_disk)->kobj, KOBJ_CHANGE); - return 0; +} + +static void +figure_loop_size(struct loop_device *lo, loff_t offset, loff_t sizelimit) +{ + loff_t size = get_size(offset, sizelimit, lo->lo_backing_file); + + loop_set_size(lo, size); } static inline int @@ -1020,10 +1026,8 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode, !file->f_op->write_iter) lo_flags |= LO_FLAGS_READ_ONLY; - error = -EFBIG; size = get_loop_size(lo, file); - if ((loff_t)(sector_t)size != size) - goto out_unlock; + error = loop_prepare_queue(lo); if (error) goto out_unlock; @@ -1057,11 +1061,8 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode, loop_update_rotational(lo); loop_update_dio(lo); - set_capacity(lo->lo_disk, size); - bd_set_size(bdev, size << 9); loop_sysfs_init(lo); - /* let user-space know about the new size */ - kobject_uevent(&disk_to_dev(bdev->bd_disk)->kobj, KOBJ_CHANGE); + loop_set_size(lo, size); set_blocksize(bdev, S_ISBLK(inode->i_mode) ? block_size(inode->i_bdev) : PAGE_SIZE); @@ -1275,82 +1276,50 @@ static int loop_clr_fd(struct loop_device *lo) return __loop_clr_fd(lo, false); } +/** + * loop_set_status_from_info - configure device from loop_info + * @lo: struct loop_device to configure + * @info: struct loop_info64 to configure the device with + * + * Configures the loop device parameters according to the passed + * in loop_info64 configuration. + */ static int -loop_set_status(struct loop_device *lo, const struct loop_info64 *info) +loop_set_status_from_info(struct loop_device *lo, + const struct loop_info64 *info) { int err; struct loop_func_table *xfer; kuid_t uid = current_uid(); - struct block_device *bdev; - bool partscan = false; - - err = mutex_lock_killable(&loop_ctl_mutex); - if (err) - return err; - if (lo->lo_encrypt_key_size && - !uid_eq(lo->lo_key_owner, uid) && - !capable(CAP_SYS_ADMIN)) { - err = -EPERM; - goto out_unlock; - } - if (lo->lo_state != Lo_bound) { - err = -ENXIO; - goto out_unlock; - } - if ((unsigned int) info->lo_encrypt_key_size > LO_KEY_SIZE) { - err = -EINVAL; - goto out_unlock; - } - if (lo->lo_offset != info->lo_offset || - lo->lo_sizelimit != info->lo_sizelimit) { - sync_blockdev(lo->lo_device); - invalidate_bdev(lo->lo_device); - } - - /* I/O need to be drained during transfer transition */ - blk_mq_freeze_queue(lo->lo_queue); + if ((unsigned int) info->lo_encrypt_key_size > LO_KEY_SIZE) + return -EINVAL; err = loop_release_xfer(lo); if (err) - goto out_unfreeze; + return err; if (info->lo_encrypt_type) { unsigned int type = info->lo_encrypt_type; - if (type >= MAX_LO_CRYPT) { - err = -EINVAL; - goto out_unfreeze; - } + if (type >= MAX_LO_CRYPT) + return -EINVAL; xfer = xfer_funcs[type]; - if (xfer == NULL) { - err = -EINVAL; - goto out_unfreeze; - } + if (xfer == NULL) + return -EINVAL; } else xfer = NULL; err = loop_init_xfer(lo, xfer, info); if (err) - goto out_unfreeze; + return err; - if (lo->lo_offset != info->lo_offset || - lo->lo_sizelimit != info->lo_sizelimit) { - /* kill_bdev should have truncated all the pages */ - if (lo->lo_device->bd_inode->i_mapping->nrpages) { - err = -EAGAIN; - pr_warn("%s: loop%d (%s) has still dirty pages (nrpages=%lu)\n", - __func__, lo->lo_number, lo->lo_file_name, - lo->lo_device->bd_inode->i_mapping->nrpages); - goto out_unfreeze; - } - if (figure_loop_size(lo, info->lo_offset, info->lo_sizelimit)) { - err = -EFBIG; - goto out_unfreeze; - } - } + /* Avoid assigning overflow values */ + if (info->lo_offset > LLONG_MAX || info->lo_sizelimit > LLONG_MAX) + return -EOVERFLOW; - loop_config_discard(lo); + lo->lo_offset = info->lo_offset; + lo->lo_sizelimit = info->lo_sizelimit; memcpy(lo->lo_file_name, info->lo_file_name, LO_NAME_SIZE); memcpy(lo->lo_crypt_name, info->lo_crypt_name, LO_NAME_SIZE); @@ -1375,6 +1344,63 @@ loop_set_status(struct loop_device *lo, const struct loop_info64 *info) lo->lo_key_owner = uid; } + return 0; +} + +static int +loop_set_status(struct loop_device *lo, const struct loop_info64 *info) +{ + int err; + struct block_device *bdev; + kuid_t uid = current_uid(); + bool partscan = false; + bool size_changed = false; + + err = mutex_lock_killable(&loop_ctl_mutex); + if (err) + return err; + if (lo->lo_encrypt_key_size && + !uid_eq(lo->lo_key_owner, uid) && + !capable(CAP_SYS_ADMIN)) { + err = -EPERM; + goto out_unlock; + } + if (lo->lo_state != Lo_bound) { + err = -ENXIO; + goto out_unlock; + } + + if (lo->lo_offset != info->lo_offset || + lo->lo_sizelimit != info->lo_sizelimit) { + size_changed = true; + sync_blockdev(lo->lo_device); + invalidate_bdev(lo->lo_device); + } + + /* I/O need to be drained during transfer transition */ + blk_mq_freeze_queue(lo->lo_queue); + + if (size_changed && lo->lo_device->bd_inode->i_mapping->nrpages) { + /* If any pages were dirtied after invalidate_bdev(), try again */ + err = -EAGAIN; + pr_warn("%s: loop%d (%s) has still dirty pages (nrpages=%lu)\n", + __func__, lo->lo_number, lo->lo_file_name, + lo->lo_device->bd_inode->i_mapping->nrpages); + goto out_unfreeze; + } + + err = loop_set_status_from_info(lo, info); + if (err) + goto out_unfreeze; + + if (size_changed) { + loff_t new_size = get_size(lo->lo_offset, lo->lo_sizelimit, + lo->lo_backing_file); + loop_set_size(lo, new_size); + } + + loop_config_discard(lo); + /* update dio if lo_offset or transfer is changed */ __loop_update_dio(lo, lo->use_dio); @@ -1550,7 +1576,9 @@ static int loop_set_capacity(struct loop_device *lo) if (unlikely(lo->lo_state != Lo_bound)) return -ENXIO; - return figure_loop_size(lo, lo->lo_offset, lo->lo_sizelimit); + figure_loop_size(lo, lo->lo_offset, lo->lo_sizelimit); + + return 0; } static int loop_set_dio(struct loop_device *lo, unsigned long arg) @@ -2050,7 +2078,7 @@ static int loop_add(struct loop_device **l, int i) lo->tag_set.queue_depth = 128; lo->tag_set.numa_node = NUMA_NO_NODE; lo->tag_set.cmd_size = sizeof(struct loop_cmd); - lo->tag_set.flags = BLK_MQ_F_SHOULD_MERGE; + lo->tag_set.flags = BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_NO_SCHED; lo->tag_set.driver_data = lo; err = blk_mq_alloc_tag_set(&lo->tag_set); diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index 09323b0510f0..f2d847ffcbc7 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -1327,10 +1327,12 @@ static int nbd_start_device_ioctl(struct nbd_device *nbd, struct block_device *b mutex_unlock(&nbd->config_lock); ret = wait_event_interruptible(config->recv_wq, atomic_read(&config->recv_threads) == 0); - if (ret) + if (ret) { sock_shutdown(nbd); - flush_workqueue(nbd->recv_workq); + nbd_clear_que(nbd); + } + flush_workqueue(nbd->recv_workq); mutex_lock(&nbd->config_lock); nbd_bdev_reset(bdev); /* user requested, ignore socket errors */ @@ -1607,7 +1609,7 @@ static int nbd_dev_dbg_init(struct nbd_device *nbd) return -EIO; dir = debugfs_create_dir(nbd_name(nbd), nbd_dbg_dir); - if (!dir) { + if (IS_ERR(dir)) { dev_err(nbd_to_dev(nbd), "Failed to create debugfs dir for '%s'\n", nbd_name(nbd)); return -EIO; @@ -1633,7 +1635,7 @@ static int nbd_dbg_init(void) struct dentry *dbg_dir; dbg_dir = debugfs_create_dir("nbd", NULL); - if (!dbg_dir) + if (IS_ERR(dbg_dir)) return -EIO; nbd_dbg_dir = dbg_dir; @@ -1706,7 +1708,8 @@ static int nbd_dev_add(int index) if (err == -ENOSPC) err = -EEXIST; } else { - err = idr_alloc(&nbd_index_idr, nbd, 0, 0, GFP_KERNEL); + err = idr_alloc(&nbd_index_idr, nbd, 0, + (MINORMASK >> part_shift) + 1, GFP_KERNEL); if (err >= 0) index = err; } @@ -2310,6 +2313,12 @@ static int nbd_genl_status(struct sk_buff *skb, struct genl_info *info) } dev_list = nla_nest_start_noflag(reply, NBD_ATTR_DEVICE_LIST); + if (!dev_list) { + nlmsg_free(reply); + ret = -EMSGSIZE; + goto out; + } + if (index == -1) { ret = idr_for_each(&nbd_index_idr, &status_cb, reply); if (ret) { diff --git a/drivers/block/null_blk_main.c b/drivers/block/null_blk_main.c index 13eae973eaea..6cbdd8a691d2 100644 --- a/drivers/block/null_blk_main.c +++ b/drivers/block/null_blk_main.c @@ -1711,8 +1711,13 @@ static int null_add_dev(struct nullb_device *dev) blk_queue_flag_clear(QUEUE_FLAG_ADD_RANDOM, nullb->q); mutex_lock(&lock); - nullb->index = ida_simple_get(&nullb_indexes, 0, 0, GFP_KERNEL); - dev->index = nullb->index; + rv = ida_simple_get(&nullb_indexes, 0, 0, GFP_KERNEL); + if (rv < 0) { + mutex_unlock(&lock); + goto out_cleanup_zone; + } + nullb->index = rv; + dev->index = rv; mutex_unlock(&lock); blk_queue_logical_block_size(nullb->q, dev->blocksize); @@ -1724,13 +1729,16 @@ static int null_add_dev(struct nullb_device *dev) rv = null_gendisk_register(nullb); if (rv) - goto out_cleanup_zone; + goto out_ida_free; mutex_lock(&lock); list_add_tail(&nullb->list, &nullb_list); mutex_unlock(&lock); return 0; + +out_ida_free: + ida_free(&nullb_indexes, nullb->index); out_cleanup_zone: if (dev->zoned) null_zone_exit(dev); diff --git a/drivers/block/rbd.c b/drivers/block/rbd.c index 55745d6953f0..fd0cdced8024 100644 --- a/drivers/block/rbd.c +++ b/drivers/block/rbd.c @@ -626,9 +626,8 @@ void rbd_warn(struct rbd_device *rbd_dev, const char *fmt, ...) static void rbd_dev_remove_parent(struct rbd_device *rbd_dev); static int rbd_dev_refresh(struct rbd_device *rbd_dev); -static int rbd_dev_v2_header_onetime(struct rbd_device *rbd_dev); -static int rbd_dev_header_info(struct rbd_device *rbd_dev); -static int rbd_dev_v2_parent_info(struct rbd_device *rbd_dev); +static int rbd_dev_v2_header_onetime(struct rbd_device *rbd_dev, + struct rbd_image_header *header); static const char *rbd_dev_v2_snap_name(struct rbd_device *rbd_dev, u64 snap_id); static int _rbd_dev_v2_snap_size(struct rbd_device *rbd_dev, u64 snap_id, @@ -1098,15 +1097,24 @@ static void rbd_init_layout(struct rbd_device *rbd_dev) RCU_INIT_POINTER(rbd_dev->layout.pool_ns, NULL); } +static void rbd_image_header_cleanup(struct rbd_image_header *header) +{ + kfree(header->object_prefix); + ceph_put_snap_context(header->snapc); + kfree(header->snap_sizes); + kfree(header->snap_names); + + memset(header, 0, sizeof(*header)); +} + /* * Fill an rbd image header with information from the given format 1 * on-disk header. */ -static int rbd_header_from_disk(struct rbd_device *rbd_dev, - struct rbd_image_header_ondisk *ondisk) +static int rbd_header_from_disk(struct rbd_image_header *header, + struct rbd_image_header_ondisk *ondisk, + bool first_time) { - struct rbd_image_header *header = &rbd_dev->header; - bool first_time = header->object_prefix == NULL; struct ceph_snap_context *snapc; char *object_prefix = NULL; char *snap_names = NULL; @@ -1173,11 +1181,6 @@ static int rbd_header_from_disk(struct rbd_device *rbd_dev, if (first_time) { header->object_prefix = object_prefix; header->obj_order = ondisk->options.order; - rbd_init_layout(rbd_dev); - } else { - ceph_put_snap_context(header->snapc); - kfree(header->snap_names); - kfree(header->snap_sizes); } /* The remaining fields always get updated (when we refresh) */ @@ -1493,14 +1496,30 @@ static bool rbd_obj_is_tail(struct rbd_obj_request *obj_req) /* * Must be called after rbd_obj_calc_img_extents(). */ -static bool rbd_obj_copyup_enabled(struct rbd_obj_request *obj_req) +static void rbd_obj_set_copyup_enabled(struct rbd_obj_request *obj_req) { - if (!obj_req->num_img_extents || - (rbd_obj_is_entire(obj_req) && - !obj_req->img_request->snapc->num_snaps)) - return false; + rbd_assert(obj_req->img_request->snapc); - return true; + if (obj_req->img_request->op_type == OBJ_OP_DISCARD) { + dout("%s %p objno %llu discard\n", __func__, obj_req, + obj_req->ex.oe_objno); + return; + } + + if (!obj_req->num_img_extents) { + dout("%s %p objno %llu not overlapping\n", __func__, obj_req, + obj_req->ex.oe_objno); + return; + } + + if (rbd_obj_is_entire(obj_req) && + !obj_req->img_request->snapc->num_snaps) { + dout("%s %p objno %llu entire\n", __func__, obj_req, + obj_req->ex.oe_objno); + return; + } + + obj_req->flags |= RBD_OBJ_FLAG_COPYUP_ENABLED; } static u64 rbd_obj_img_extents_bytes(struct rbd_obj_request *obj_req) @@ -1599,6 +1618,7 @@ __rbd_obj_add_osd_request(struct rbd_obj_request *obj_req, static struct ceph_osd_request * rbd_obj_add_osd_request(struct rbd_obj_request *obj_req, int num_ops) { + rbd_assert(obj_req->img_request->snapc); return __rbd_obj_add_osd_request(obj_req, obj_req->img_request->snapc, num_ops); } @@ -1727,11 +1747,14 @@ static bool rbd_dev_parent_get(struct rbd_device *rbd_dev) * Caller is responsible for filling in the list of object requests * that comprises the image request, and the Linux request pointer * (if there is one). + * + * Only snap_id is captured here, for reads. For writes, snapshot + * context is captured in rbd_img_object_requests() after exclusive + * lock is ensured to be held. */ static struct rbd_img_request *rbd_img_request_create( struct rbd_device *rbd_dev, - enum obj_operation_type op_type, - struct ceph_snap_context *snapc) + enum obj_operation_type op_type) { struct rbd_img_request *img_request; @@ -1743,8 +1766,6 @@ static struct rbd_img_request *rbd_img_request_create( img_request->op_type = op_type; if (!rbd_img_is_write(img_request)) img_request->snap_id = rbd_dev->spec->snap_id; - else - img_request->snapc = snapc; if (rbd_dev_parent_get(rbd_dev)) img_request_layered_set(img_request); @@ -2087,7 +2108,7 @@ static int rbd_object_map_update_finish(struct rbd_obj_request *obj_req, struct rbd_device *rbd_dev = obj_req->img_request->rbd_dev; struct ceph_osd_data *osd_data; u64 objno; - u8 state, new_state, uninitialized_var(current_state); + u8 state, new_state, current_state; bool has_current_state; void *p; @@ -2389,9 +2410,6 @@ static int rbd_obj_init_write(struct rbd_obj_request *obj_req) if (ret) return ret; - if (rbd_obj_copyup_enabled(obj_req)) - obj_req->flags |= RBD_OBJ_FLAG_COPYUP_ENABLED; - obj_req->write_state = RBD_OBJ_WRITE_START; return 0; } @@ -2497,8 +2515,6 @@ static int rbd_obj_init_zeroout(struct rbd_obj_request *obj_req) if (ret) return ret; - if (rbd_obj_copyup_enabled(obj_req)) - obj_req->flags |= RBD_OBJ_FLAG_COPYUP_ENABLED; if (!obj_req->num_img_extents) { obj_req->flags |= RBD_OBJ_FLAG_NOOP_FOR_NONEXISTENT; if (rbd_obj_is_entire(obj_req)) @@ -2935,7 +2951,7 @@ static int rbd_obj_read_from_parent(struct rbd_obj_request *obj_req) int ret; child_img_req = rbd_img_request_create(img_req->rbd_dev->parent, - OBJ_OP_READ, NULL); + OBJ_OP_READ); if (!child_img_req) return -ENOMEM; @@ -3439,6 +3455,7 @@ again: case RBD_OBJ_WRITE_START: rbd_assert(!*result); + rbd_obj_set_copyup_enabled(obj_req); if (rbd_obj_write_is_noop(obj_req)) return true; @@ -3586,14 +3603,15 @@ static bool rbd_lock_add_request(struct rbd_img_request *img_req) static void rbd_lock_del_request(struct rbd_img_request *img_req) { struct rbd_device *rbd_dev = img_req->rbd_dev; - bool need_wakeup; + bool need_wakeup = false; lockdep_assert_held(&rbd_dev->lock_rwsem); spin_lock(&rbd_dev->lock_lists_lock); - rbd_assert(!list_empty(&img_req->lock_item)); - list_del_init(&img_req->lock_item); - need_wakeup = (rbd_dev->lock_state == RBD_LOCK_STATE_RELEASING && - list_empty(&rbd_dev->running_list)); + if (!list_empty(&img_req->lock_item)) { + list_del_init(&img_req->lock_item); + need_wakeup = (rbd_dev->lock_state == RBD_LOCK_STATE_RELEASING && + list_empty(&rbd_dev->running_list)); + } spin_unlock(&rbd_dev->lock_lists_lock); if (need_wakeup) complete(&rbd_dev->releasing_wait); @@ -3625,9 +3643,19 @@ static int rbd_img_exclusive_lock(struct rbd_img_request *img_req) static void rbd_img_object_requests(struct rbd_img_request *img_req) { + struct rbd_device *rbd_dev = img_req->rbd_dev; struct rbd_obj_request *obj_req; rbd_assert(!img_req->pending.result && !img_req->pending.num_pending); + rbd_assert(!need_exclusive_lock(img_req) || + __rbd_is_lock_owner(rbd_dev)); + + if (rbd_img_is_write(img_req)) { + rbd_assert(!img_req->snapc); + down_read(&rbd_dev->header_rwsem); + img_req->snapc = ceph_get_snap_context(rbd_dev->header.snapc); + up_read(&rbd_dev->header_rwsem); + } for_each_obj_request(img_req, obj_req) { int result = 0; @@ -3645,7 +3673,6 @@ static void rbd_img_object_requests(struct rbd_img_request *img_req) static bool rbd_img_advance(struct rbd_img_request *img_req, int *result) { - struct rbd_device *rbd_dev = img_req->rbd_dev; int ret; again: @@ -3666,9 +3693,6 @@ again: if (*result) return true; - rbd_assert(!need_exclusive_lock(img_req) || - __rbd_is_lock_owner(rbd_dev)); - rbd_img_object_requests(img_req); if (!img_req->pending.num_pending) { *result = img_req->pending.result; @@ -3974,14 +3998,19 @@ static void wake_lock_waiters(struct rbd_device *rbd_dev, int result) return; } - list_for_each_entry(img_req, &rbd_dev->acquiring_list, lock_item) { + while (!list_empty(&rbd_dev->acquiring_list)) { + img_req = list_first_entry(&rbd_dev->acquiring_list, + struct rbd_img_request, lock_item); mutex_lock(&img_req->state_mutex); rbd_assert(img_req->state == RBD_IMG_EXCLUSIVE_LOCK); + if (!result) + list_move_tail(&img_req->lock_item, + &rbd_dev->running_list); + else + list_del_init(&img_req->lock_item); rbd_img_schedule(img_req, result); mutex_unlock(&img_req->state_mutex); } - - list_splice_tail_init(&rbd_dev->acquiring_list, &rbd_dev->running_list); } static int get_lock_owner_info(struct rbd_device *rbd_dev, @@ -4130,6 +4159,10 @@ static int rbd_post_acquire_action(struct rbd_device *rbd_dev) { int ret; + ret = rbd_dev_refresh(rbd_dev); + if (ret) + return ret; + if (rbd_dev->header.features & RBD_FEATURE_OBJECT_MAP) { ret = rbd_object_map_open(rbd_dev); if (ret) @@ -4788,7 +4821,6 @@ static void rbd_queue_workfn(struct work_struct *work) struct request *rq = blk_mq_rq_from_pdu(work); struct rbd_device *rbd_dev = rq->q->queuedata; struct rbd_img_request *img_request; - struct ceph_snap_context *snapc = NULL; u64 offset = (u64)blk_rq_pos(rq) << SECTOR_SHIFT; u64 length = blk_rq_bytes(rq); enum obj_operation_type op_type; @@ -4853,10 +4885,6 @@ static void rbd_queue_workfn(struct work_struct *work) down_read(&rbd_dev->header_rwsem); mapping_size = rbd_dev->mapping.size; - if (op_type != OBJ_OP_READ) { - snapc = rbd_dev->header.snapc; - ceph_get_snap_context(snapc); - } up_read(&rbd_dev->header_rwsem); if (offset + length > mapping_size) { @@ -4866,13 +4894,12 @@ static void rbd_queue_workfn(struct work_struct *work) goto err_rq; } - img_request = rbd_img_request_create(rbd_dev, op_type, snapc); + img_request = rbd_img_request_create(rbd_dev, op_type); if (!img_request) { result = -ENOMEM; goto err_rq; } img_request->rq = rq; - snapc = NULL; /* img_request consumes a ref */ dout("%s rbd_dev %p img_req %p %s %llu~%llu\n", __func__, rbd_dev, img_request, obj_op_name(op_type), offset, length); @@ -4894,7 +4921,6 @@ err_rq: if (result) rbd_warn(rbd_dev, "%s %llx at %llx result %d", obj_op_name(op_type), length, offset, result); - ceph_put_snap_context(snapc); err: blk_mq_end_request(rq, errno_to_blk_status(result)); } @@ -4966,7 +4992,9 @@ out_req: * return, the rbd_dev->header field will contain up-to-date * information about the image. */ -static int rbd_dev_v1_header_info(struct rbd_device *rbd_dev) +static int rbd_dev_v1_header_info(struct rbd_device *rbd_dev, + struct rbd_image_header *header, + bool first_time) { struct rbd_image_header_ondisk *ondisk = NULL; u32 snap_count = 0; @@ -5014,7 +5042,7 @@ static int rbd_dev_v1_header_info(struct rbd_device *rbd_dev) snap_count = le32_to_cpu(ondisk->snap_count); } while (snap_count != want_count); - ret = rbd_header_from_disk(rbd_dev, ondisk); + ret = rbd_header_from_disk(header, ondisk, first_time); out: kfree(ondisk); @@ -5058,43 +5086,6 @@ static void rbd_dev_update_size(struct rbd_device *rbd_dev) } } -static int rbd_dev_refresh(struct rbd_device *rbd_dev) -{ - u64 mapping_size; - int ret; - - down_write(&rbd_dev->header_rwsem); - mapping_size = rbd_dev->mapping.size; - - ret = rbd_dev_header_info(rbd_dev); - if (ret) - goto out; - - /* - * If there is a parent, see if it has disappeared due to the - * mapped image getting flattened. - */ - if (rbd_dev->parent) { - ret = rbd_dev_v2_parent_info(rbd_dev); - if (ret) - goto out; - } - - if (rbd_dev->spec->snap_id == CEPH_NOSNAP) { - rbd_dev->mapping.size = rbd_dev->header.image_size; - } else { - /* validate mapped snapshot's EXISTS flag */ - rbd_exists_validate(rbd_dev); - } - -out: - up_write(&rbd_dev->header_rwsem); - if (!ret && mapping_size != rbd_dev->mapping.size) - rbd_dev_update_size(rbd_dev); - - return ret; -} - static int rbd_init_request(struct blk_mq_tag_set *set, struct request *rq, unsigned int hctx_idx, unsigned int numa_node) { @@ -5529,8 +5520,7 @@ static void rbd_dev_release(struct device *dev) module_put(THIS_MODULE); } -static struct rbd_device *__rbd_dev_create(struct rbd_client *rbdc, - struct rbd_spec *spec) +static struct rbd_device *__rbd_dev_create(struct rbd_spec *spec) { struct rbd_device *rbd_dev; @@ -5575,9 +5565,6 @@ static struct rbd_device *__rbd_dev_create(struct rbd_client *rbdc, rbd_dev->dev.parent = &rbd_root_dev; device_initialize(&rbd_dev->dev); - rbd_dev->rbd_client = rbdc; - rbd_dev->spec = spec; - return rbd_dev; } @@ -5590,12 +5577,10 @@ static struct rbd_device *rbd_dev_create(struct rbd_client *rbdc, { struct rbd_device *rbd_dev; - rbd_dev = __rbd_dev_create(rbdc, spec); + rbd_dev = __rbd_dev_create(spec); if (!rbd_dev) return NULL; - rbd_dev->opts = opts; - /* get an id and fill in device name */ rbd_dev->dev_id = ida_simple_get(&rbd_dev_id_ida, 0, minor_to_rbd_dev_id(1 << MINORBITS), @@ -5612,6 +5597,10 @@ static struct rbd_device *rbd_dev_create(struct rbd_client *rbdc, /* we have a ref from do_rbd_add() */ __module_get(THIS_MODULE); + rbd_dev->rbd_client = rbdc; + rbd_dev->spec = spec; + rbd_dev->opts = opts; + dout("%s rbd_dev %p dev_id %d\n", __func__, rbd_dev, rbd_dev->dev_id); return rbd_dev; @@ -5666,17 +5655,12 @@ static int _rbd_dev_v2_snap_size(struct rbd_device *rbd_dev, u64 snap_id, return 0; } -static int rbd_dev_v2_image_size(struct rbd_device *rbd_dev) -{ - return _rbd_dev_v2_snap_size(rbd_dev, CEPH_NOSNAP, - &rbd_dev->header.obj_order, - &rbd_dev->header.image_size); -} - -static int rbd_dev_v2_object_prefix(struct rbd_device *rbd_dev) +static int rbd_dev_v2_object_prefix(struct rbd_device *rbd_dev, + char **pobject_prefix) { size_t size; void *reply_buf; + char *object_prefix; int ret; void *p; @@ -5694,16 +5678,16 @@ static int rbd_dev_v2_object_prefix(struct rbd_device *rbd_dev) goto out; p = reply_buf; - rbd_dev->header.object_prefix = ceph_extract_encoded_string(&p, - p + ret, NULL, GFP_NOIO); + object_prefix = ceph_extract_encoded_string(&p, p + ret, NULL, + GFP_NOIO); + if (IS_ERR(object_prefix)) { + ret = PTR_ERR(object_prefix); + goto out; + } ret = 0; - if (IS_ERR(rbd_dev->header.object_prefix)) { - ret = PTR_ERR(rbd_dev->header.object_prefix); - rbd_dev->header.object_prefix = NULL; - } else { - dout(" object_prefix = %s\n", rbd_dev->header.object_prefix); - } + *pobject_prefix = object_prefix; + dout(" object_prefix = %s\n", object_prefix); out: kfree(reply_buf); @@ -5748,12 +5732,6 @@ static int _rbd_dev_v2_snap_features(struct rbd_device *rbd_dev, u64 snap_id, return 0; } -static int rbd_dev_v2_features(struct rbd_device *rbd_dev) -{ - return _rbd_dev_v2_snap_features(rbd_dev, CEPH_NOSNAP, - &rbd_dev->header.features); -} - /* * These are generic image flags, but since they are used only for * object map, store them in rbd_dev->object_map_flags. @@ -5790,6 +5768,14 @@ struct parent_image_info { u64 overlap; }; +static void rbd_parent_info_cleanup(struct parent_image_info *pii) +{ + kfree(pii->pool_ns); + kfree(pii->image_id); + + memset(pii, 0, sizeof(*pii)); +} + /* * The caller is responsible for @pii. */ @@ -5859,6 +5845,9 @@ static int __get_parent_info(struct rbd_device *rbd_dev, if (pii->has_overlap) ceph_decode_64_safe(&p, end, pii->overlap, e_inval); + dout("%s pool_id %llu pool_ns %s image_id %s snap_id %llu has_overlap %d overlap %llu\n", + __func__, pii->pool_id, pii->pool_ns, pii->image_id, pii->snap_id, + pii->has_overlap, pii->overlap); return 0; e_inval: @@ -5897,14 +5886,17 @@ static int __get_parent_info_legacy(struct rbd_device *rbd_dev, pii->has_overlap = true; ceph_decode_64_safe(&p, end, pii->overlap, e_inval); + dout("%s pool_id %llu pool_ns %s image_id %s snap_id %llu has_overlap %d overlap %llu\n", + __func__, pii->pool_id, pii->pool_ns, pii->image_id, pii->snap_id, + pii->has_overlap, pii->overlap); return 0; e_inval: return -EINVAL; } -static int get_parent_info(struct rbd_device *rbd_dev, - struct parent_image_info *pii) +static int rbd_dev_v2_parent_info(struct rbd_device *rbd_dev, + struct parent_image_info *pii) { struct page *req_page, *reply_page; void *p; @@ -5932,7 +5924,7 @@ static int get_parent_info(struct rbd_device *rbd_dev, return ret; } -static int rbd_dev_v2_parent_info(struct rbd_device *rbd_dev) +static int rbd_dev_setup_parent(struct rbd_device *rbd_dev) { struct rbd_spec *parent_spec; struct parent_image_info pii = { 0 }; @@ -5942,37 +5934,12 @@ static int rbd_dev_v2_parent_info(struct rbd_device *rbd_dev) if (!parent_spec) return -ENOMEM; - ret = get_parent_info(rbd_dev, &pii); + ret = rbd_dev_v2_parent_info(rbd_dev, &pii); if (ret) goto out_err; - dout("%s pool_id %llu pool_ns %s image_id %s snap_id %llu has_overlap %d overlap %llu\n", - __func__, pii.pool_id, pii.pool_ns, pii.image_id, pii.snap_id, - pii.has_overlap, pii.overlap); - - if (pii.pool_id == CEPH_NOPOOL || !pii.has_overlap) { - /* - * Either the parent never existed, or we have - * record of it but the image got flattened so it no - * longer has a parent. When the parent of a - * layered image disappears we immediately set the - * overlap to 0. The effect of this is that all new - * requests will be treated as if the image had no - * parent. - * - * If !pii.has_overlap, the parent image spec is not - * applicable. It's there to avoid duplication in each - * snapshot record. - */ - if (rbd_dev->parent_overlap) { - rbd_dev->parent_overlap = 0; - rbd_dev_parent_put(rbd_dev); - pr_info("%s: clone image has been flattened\n", - rbd_dev->disk->disk_name); - } - + if (pii.pool_id == CEPH_NOPOOL || !pii.has_overlap) goto out; /* No parent? No problem. */ - } /* The ceph file layout needs to fit pool id in 32 bits */ @@ -5984,58 +5951,46 @@ static int rbd_dev_v2_parent_info(struct rbd_device *rbd_dev) } /* - * The parent won't change (except when the clone is - * flattened, already handled that). So we only need to - * record the parent spec we have not already done so. + * The parent won't change except when the clone is flattened, + * so we only need to record the parent image spec once. */ - if (!rbd_dev->parent_spec) { - parent_spec->pool_id = pii.pool_id; - if (pii.pool_ns && *pii.pool_ns) { - parent_spec->pool_ns = pii.pool_ns; - pii.pool_ns = NULL; - } - parent_spec->image_id = pii.image_id; - pii.image_id = NULL; - parent_spec->snap_id = pii.snap_id; - - rbd_dev->parent_spec = parent_spec; - parent_spec = NULL; /* rbd_dev now owns this */ + parent_spec->pool_id = pii.pool_id; + if (pii.pool_ns && *pii.pool_ns) { + parent_spec->pool_ns = pii.pool_ns; + pii.pool_ns = NULL; } + parent_spec->image_id = pii.image_id; + pii.image_id = NULL; + parent_spec->snap_id = pii.snap_id; + + rbd_assert(!rbd_dev->parent_spec); + rbd_dev->parent_spec = parent_spec; + parent_spec = NULL; /* rbd_dev now owns this */ /* - * We always update the parent overlap. If it's zero we issue - * a warning, as we will proceed as if there was no parent. + * Record the parent overlap. If it's zero, issue a warning as + * we will proceed as if there is no parent. */ - if (!pii.overlap) { - if (parent_spec) { - /* refresh, careful to warn just once */ - if (rbd_dev->parent_overlap) - rbd_warn(rbd_dev, - "clone now standalone (overlap became 0)"); - } else { - /* initial probe */ - rbd_warn(rbd_dev, "clone is standalone (overlap 0)"); - } - } + if (!pii.overlap) + rbd_warn(rbd_dev, "clone is standalone (overlap 0)"); rbd_dev->parent_overlap = pii.overlap; out: ret = 0; out_err: - kfree(pii.pool_ns); - kfree(pii.image_id); + rbd_parent_info_cleanup(&pii); rbd_spec_put(parent_spec); return ret; } -static int rbd_dev_v2_striping_info(struct rbd_device *rbd_dev) +static int rbd_dev_v2_striping_info(struct rbd_device *rbd_dev, + u64 *stripe_unit, u64 *stripe_count) { struct { __le64 stripe_unit; __le64 stripe_count; } __attribute__ ((packed)) striping_info_buf = { 0 }; size_t size = sizeof (striping_info_buf); - void *p; int ret; ret = rbd_obj_method_sync(rbd_dev, &rbd_dev->header_oid, @@ -6047,27 +6002,33 @@ static int rbd_dev_v2_striping_info(struct rbd_device *rbd_dev) if (ret < size) return -ERANGE; - p = &striping_info_buf; - rbd_dev->header.stripe_unit = ceph_decode_64(&p); - rbd_dev->header.stripe_count = ceph_decode_64(&p); + *stripe_unit = le64_to_cpu(striping_info_buf.stripe_unit); + *stripe_count = le64_to_cpu(striping_info_buf.stripe_count); + dout(" stripe_unit = %llu stripe_count = %llu\n", *stripe_unit, + *stripe_count); + return 0; } -static int rbd_dev_v2_data_pool(struct rbd_device *rbd_dev) +static int rbd_dev_v2_data_pool(struct rbd_device *rbd_dev, s64 *data_pool_id) { - __le64 data_pool_id; + __le64 data_pool_buf; int ret; ret = rbd_obj_method_sync(rbd_dev, &rbd_dev->header_oid, &rbd_dev->header_oloc, "get_data_pool", - NULL, 0, &data_pool_id, sizeof(data_pool_id)); + NULL, 0, &data_pool_buf, + sizeof(data_pool_buf)); + dout("%s: rbd_obj_method_sync returned %d\n", __func__, ret); if (ret < 0) return ret; - if (ret < sizeof(data_pool_id)) + if (ret < sizeof(data_pool_buf)) return -EBADMSG; - rbd_dev->header.data_pool_id = le64_to_cpu(data_pool_id); - WARN_ON(rbd_dev->header.data_pool_id == CEPH_NOPOOL); + *data_pool_id = le64_to_cpu(data_pool_buf); + dout(" data_pool_id = %lld\n", *data_pool_id); + WARN_ON(*data_pool_id == CEPH_NOPOOL); + return 0; } @@ -6259,7 +6220,8 @@ out_err: return ret; } -static int rbd_dev_v2_snap_context(struct rbd_device *rbd_dev) +static int rbd_dev_v2_snap_context(struct rbd_device *rbd_dev, + struct ceph_snap_context **psnapc) { size_t size; int ret; @@ -6320,9 +6282,7 @@ static int rbd_dev_v2_snap_context(struct rbd_device *rbd_dev) for (i = 0; i < snap_count; i++) snapc->snaps[i] = ceph_decode_64(&p); - ceph_put_snap_context(rbd_dev->header.snapc); - rbd_dev->header.snapc = snapc; - + *psnapc = snapc; dout(" snap context seq = %llu, snap_count = %u\n", (unsigned long long)seq, (unsigned int)snap_count); out: @@ -6371,38 +6331,42 @@ out: return snap_name; } -static int rbd_dev_v2_header_info(struct rbd_device *rbd_dev) +static int rbd_dev_v2_header_info(struct rbd_device *rbd_dev, + struct rbd_image_header *header, + bool first_time) { - bool first_time = rbd_dev->header.object_prefix == NULL; int ret; - ret = rbd_dev_v2_image_size(rbd_dev); + ret = _rbd_dev_v2_snap_size(rbd_dev, CEPH_NOSNAP, + first_time ? &header->obj_order : NULL, + &header->image_size); if (ret) return ret; if (first_time) { - ret = rbd_dev_v2_header_onetime(rbd_dev); + ret = rbd_dev_v2_header_onetime(rbd_dev, header); if (ret) return ret; } - ret = rbd_dev_v2_snap_context(rbd_dev); - if (ret && first_time) { - kfree(rbd_dev->header.object_prefix); - rbd_dev->header.object_prefix = NULL; - } + ret = rbd_dev_v2_snap_context(rbd_dev, &header->snapc); + if (ret) + return ret; - return ret; + return 0; } -static int rbd_dev_header_info(struct rbd_device *rbd_dev) +static int rbd_dev_header_info(struct rbd_device *rbd_dev, + struct rbd_image_header *header, + bool first_time) { rbd_assert(rbd_image_format_valid(rbd_dev->image_format)); + rbd_assert(!header->object_prefix && !header->snapc); if (rbd_dev->image_format == 1) - return rbd_dev_v1_header_info(rbd_dev); + return rbd_dev_v1_header_info(rbd_dev, header, first_time); - return rbd_dev_v2_header_info(rbd_dev); + return rbd_dev_v2_header_info(rbd_dev, header, first_time); } /* @@ -6752,60 +6716,49 @@ out: */ static void rbd_dev_unprobe(struct rbd_device *rbd_dev) { - struct rbd_image_header *header; - rbd_dev_parent_put(rbd_dev); rbd_object_map_free(rbd_dev); rbd_dev_mapping_clear(rbd_dev); /* Free dynamic fields from the header, then zero it out */ - header = &rbd_dev->header; - ceph_put_snap_context(header->snapc); - kfree(header->snap_sizes); - kfree(header->snap_names); - kfree(header->object_prefix); - memset(header, 0, sizeof (*header)); + rbd_image_header_cleanup(&rbd_dev->header); } -static int rbd_dev_v2_header_onetime(struct rbd_device *rbd_dev) +static int rbd_dev_v2_header_onetime(struct rbd_device *rbd_dev, + struct rbd_image_header *header) { int ret; - ret = rbd_dev_v2_object_prefix(rbd_dev); + ret = rbd_dev_v2_object_prefix(rbd_dev, &header->object_prefix); if (ret) - goto out_err; + return ret; /* * Get the and check features for the image. Currently the * features are assumed to never change. */ - ret = rbd_dev_v2_features(rbd_dev); + ret = _rbd_dev_v2_snap_features(rbd_dev, CEPH_NOSNAP, + &header->features); if (ret) - goto out_err; + return ret; /* If the image supports fancy striping, get its parameters */ - if (rbd_dev->header.features & RBD_FEATURE_STRIPINGV2) { - ret = rbd_dev_v2_striping_info(rbd_dev); - if (ret < 0) - goto out_err; + if (header->features & RBD_FEATURE_STRIPINGV2) { + ret = rbd_dev_v2_striping_info(rbd_dev, &header->stripe_unit, + &header->stripe_count); + if (ret) + return ret; } - if (rbd_dev->header.features & RBD_FEATURE_DATA_POOL) { - ret = rbd_dev_v2_data_pool(rbd_dev); + if (header->features & RBD_FEATURE_DATA_POOL) { + ret = rbd_dev_v2_data_pool(rbd_dev, &header->data_pool_id); if (ret) - goto out_err; + return ret; } - rbd_init_layout(rbd_dev); return 0; - -out_err: - rbd_dev->header.features = 0; - kfree(rbd_dev->header.object_prefix); - rbd_dev->header.object_prefix = NULL; - return ret; } /* @@ -6827,7 +6780,7 @@ static int rbd_dev_probe_parent(struct rbd_device *rbd_dev, int depth) goto out_err; } - parent = __rbd_dev_create(rbd_dev->rbd_client, rbd_dev->parent_spec); + parent = __rbd_dev_create(rbd_dev->parent_spec); if (!parent) { ret = -ENOMEM; goto out_err; @@ -6837,8 +6790,8 @@ static int rbd_dev_probe_parent(struct rbd_device *rbd_dev, int depth) * Images related by parent/child relationships always share * rbd_client and spec/parent_spec, so bump their refcounts. */ - __rbd_get_client(rbd_dev->rbd_client); - rbd_spec_get(rbd_dev->parent_spec); + parent->rbd_client = __rbd_get_client(rbd_dev->rbd_client); + parent->spec = rbd_spec_get(rbd_dev->parent_spec); ret = rbd_dev_image_probe(parent, depth); if (ret < 0) @@ -6983,10 +6936,12 @@ static int rbd_dev_image_probe(struct rbd_device *rbd_dev, int depth) if (!depth) down_write(&rbd_dev->header_rwsem); - ret = rbd_dev_header_info(rbd_dev); + ret = rbd_dev_header_info(rbd_dev, &rbd_dev->header, true); if (ret) goto err_out_probe; + rbd_init_layout(rbd_dev); + /* * If this image is the one being mapped, we have pool name and * id, image name and id, and snap name - need to fill snap id. @@ -7020,7 +6975,7 @@ static int rbd_dev_image_probe(struct rbd_device *rbd_dev, int depth) } if (rbd_dev->header.features & RBD_FEATURE_LAYERING) { - ret = rbd_dev_v2_parent_info(rbd_dev); + ret = rbd_dev_setup_parent(rbd_dev); if (ret) goto err_out_probe; } @@ -7046,6 +7001,112 @@ err_out_format: return ret; } +static void rbd_dev_update_header(struct rbd_device *rbd_dev, + struct rbd_image_header *header) +{ + rbd_assert(rbd_image_format_valid(rbd_dev->image_format)); + rbd_assert(rbd_dev->header.object_prefix); /* !first_time */ + + if (rbd_dev->header.image_size != header->image_size) { + rbd_dev->header.image_size = header->image_size; + + if (rbd_dev->spec->snap_id == CEPH_NOSNAP) { + rbd_dev->mapping.size = header->image_size; + rbd_dev_update_size(rbd_dev); + } + } + + if (rbd_dev->spec->snap_id != CEPH_NOSNAP) { + /* validate mapped snapshot's EXISTS flag */ + rbd_exists_validate(rbd_dev); + } + + ceph_put_snap_context(rbd_dev->header.snapc); + rbd_dev->header.snapc = header->snapc; + header->snapc = NULL; + + if (rbd_dev->image_format == 1) { + kfree(rbd_dev->header.snap_names); + rbd_dev->header.snap_names = header->snap_names; + header->snap_names = NULL; + + kfree(rbd_dev->header.snap_sizes); + rbd_dev->header.snap_sizes = header->snap_sizes; + header->snap_sizes = NULL; + } +} + +static void rbd_dev_update_parent(struct rbd_device *rbd_dev, + struct parent_image_info *pii) +{ + if (pii->pool_id == CEPH_NOPOOL || !pii->has_overlap) { + /* + * Either the parent never existed, or we have + * record of it but the image got flattened so it no + * longer has a parent. When the parent of a + * layered image disappears we immediately set the + * overlap to 0. The effect of this is that all new + * requests will be treated as if the image had no + * parent. + * + * If !pii.has_overlap, the parent image spec is not + * applicable. It's there to avoid duplication in each + * snapshot record. + */ + if (rbd_dev->parent_overlap) { + rbd_dev->parent_overlap = 0; + rbd_dev_parent_put(rbd_dev); + pr_info("%s: clone has been flattened\n", + rbd_dev->disk->disk_name); + } + } else { + rbd_assert(rbd_dev->parent_spec); + + /* + * Update the parent overlap. If it became zero, issue + * a warning as we will proceed as if there is no parent. + */ + if (!pii->overlap && rbd_dev->parent_overlap) + rbd_warn(rbd_dev, + "clone has become standalone (overlap 0)"); + rbd_dev->parent_overlap = pii->overlap; + } +} + +static int rbd_dev_refresh(struct rbd_device *rbd_dev) +{ + struct rbd_image_header header = { 0 }; + struct parent_image_info pii = { 0 }; + int ret; + + dout("%s rbd_dev %p\n", __func__, rbd_dev); + + ret = rbd_dev_header_info(rbd_dev, &header, false); + if (ret) + goto out; + + /* + * If there is a parent, see if it has disappeared due to the + * mapped image getting flattened. + */ + if (rbd_dev->parent) { + ret = rbd_dev_v2_parent_info(rbd_dev, &pii); + if (ret) + goto out; + } + + down_write(&rbd_dev->header_rwsem); + rbd_dev_update_header(rbd_dev, &header); + if (rbd_dev->parent) + rbd_dev_update_parent(rbd_dev, &pii); + up_write(&rbd_dev->header_rwsem); + +out: + rbd_parent_info_cleanup(&pii); + rbd_image_header_cleanup(&header); + return ret; +} + static ssize_t do_rbd_add(struct bus_type *bus, const char *buf, size_t count) diff --git a/drivers/block/sunvdc.c b/drivers/block/sunvdc.c index 6b2fd630de85..6622dd1aa07b 100644 --- a/drivers/block/sunvdc.c +++ b/drivers/block/sunvdc.c @@ -983,6 +983,8 @@ static int vdc_port_probe(struct vio_dev *vdev, const struct vio_device_id *id) print_version(); hp = mdesc_grab(); + if (!hp) + return -ENODEV; err = -ENODEV; if ((vdev->dev_no << PARTITION_SHIFT) & ~(u64)MINORMASK) { diff --git a/drivers/block/sx8.c b/drivers/block/sx8.c deleted file mode 100644 index 4478eb7efee0..000000000000 --- a/drivers/block/sx8.c +++ /dev/null @@ -1,1586 +0,0 @@ -/* - * sx8.c: Driver for Promise SATA SX8 looks-like-I2O hardware - * - * Copyright 2004-2005 Red Hat, Inc. - * - * Author/maintainer: Jeff Garzik <jgarzik@pobox.com> - * - * This file is subject to the terms and conditions of the GNU General Public - * License. See the file "COPYING" in the main directory of this archive - * for more details. - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/init.h> -#include <linux/pci.h> -#include <linux/slab.h> -#include <linux/spinlock.h> -#include <linux/blk-mq.h> -#include <linux/sched.h> -#include <linux/interrupt.h> -#include <linux/compiler.h> -#include <linux/workqueue.h> -#include <linux/bitops.h> -#include <linux/delay.h> -#include <linux/ktime.h> -#include <linux/hdreg.h> -#include <linux/dma-mapping.h> -#include <linux/completion.h> -#include <linux/scatterlist.h> -#include <asm/io.h> -#include <linux/uaccess.h> - -#if 0 -#define CARM_DEBUG -#define CARM_VERBOSE_DEBUG -#else -#undef CARM_DEBUG -#undef CARM_VERBOSE_DEBUG -#endif -#undef CARM_NDEBUG - -#define DRV_NAME "sx8" -#define DRV_VERSION "1.0" -#define PFX DRV_NAME ": " - -MODULE_AUTHOR("Jeff Garzik"); -MODULE_LICENSE("GPL"); -MODULE_DESCRIPTION("Promise SATA SX8 block driver"); -MODULE_VERSION(DRV_VERSION); - -/* - * SX8 hardware has a single message queue for all ATA ports. - * When this driver was written, the hardware (firmware?) would - * corrupt data eventually, if more than one request was outstanding. - * As one can imagine, having 8 ports bottlenecking on a single - * command hurts performance. - * - * Based on user reports, later versions of the hardware (firmware?) - * seem to be able to survive with more than one command queued. - * - * Therefore, we default to the safe option -- 1 command -- but - * allow the user to increase this. - * - * SX8 should be able to support up to ~60 queued commands (CARM_MAX_REQ), - * but problems seem to occur when you exceed ~30, even on newer hardware. - */ -static int max_queue = 1; -module_param(max_queue, int, 0444); -MODULE_PARM_DESC(max_queue, "Maximum number of queued commands. (min==1, max==30, safe==1)"); - - -#define NEXT_RESP(idx) ((idx + 1) % RMSG_Q_LEN) - -/* 0xf is just arbitrary, non-zero noise; this is sorta like poisoning */ -#define TAG_ENCODE(tag) (((tag) << 16) | 0xf) -#define TAG_DECODE(tag) (((tag) >> 16) & 0x1f) -#define TAG_VALID(tag) ((((tag) & 0xf) == 0xf) && (TAG_DECODE(tag) < 32)) - -/* note: prints function name for you */ -#ifdef CARM_DEBUG -#define DPRINTK(fmt, args...) printk(KERN_ERR "%s: " fmt, __func__, ## args) -#ifdef CARM_VERBOSE_DEBUG -#define VPRINTK(fmt, args...) printk(KERN_ERR "%s: " fmt, __func__, ## args) -#else -#define VPRINTK(fmt, args...) -#endif /* CARM_VERBOSE_DEBUG */ -#else -#define DPRINTK(fmt, args...) -#define VPRINTK(fmt, args...) -#endif /* CARM_DEBUG */ - -#ifdef CARM_NDEBUG -#define assert(expr) -#else -#define assert(expr) \ - if(unlikely(!(expr))) { \ - printk(KERN_ERR "Assertion failed! %s,%s,%s,line=%d\n", \ - #expr, __FILE__, __func__, __LINE__); \ - } -#endif - -/* defines only for the constants which don't work well as enums */ -struct carm_host; - -enum { - /* adapter-wide limits */ - CARM_MAX_PORTS = 8, - CARM_SHM_SIZE = (4096 << 7), - CARM_MINORS_PER_MAJOR = 256 / CARM_MAX_PORTS, - CARM_MAX_WAIT_Q = CARM_MAX_PORTS + 1, - - /* command message queue limits */ - CARM_MAX_REQ = 64, /* max command msgs per host */ - CARM_MSG_LOW_WATER = (CARM_MAX_REQ / 4), /* refill mark */ - - /* S/G limits, host-wide and per-request */ - CARM_MAX_REQ_SG = 32, /* max s/g entries per request */ - CARM_MAX_HOST_SG = 600, /* max s/g entries per host */ - CARM_SG_LOW_WATER = (CARM_MAX_HOST_SG / 4), /* re-fill mark */ - - /* hardware registers */ - CARM_IHQP = 0x1c, - CARM_INT_STAT = 0x10, /* interrupt status */ - CARM_INT_MASK = 0x14, /* interrupt mask */ - CARM_HMUC = 0x18, /* host message unit control */ - RBUF_ADDR_LO = 0x20, /* response msg DMA buf low 32 bits */ - RBUF_ADDR_HI = 0x24, /* response msg DMA buf high 32 bits */ - RBUF_BYTE_SZ = 0x28, - CARM_RESP_IDX = 0x2c, - CARM_CMS0 = 0x30, /* command message size reg 0 */ - CARM_LMUC = 0x48, - CARM_HMPHA = 0x6c, - CARM_INITC = 0xb5, - - /* bits in CARM_INT_{STAT,MASK} */ - INT_RESERVED = 0xfffffff0, - INT_WATCHDOG = (1 << 3), /* watchdog timer */ - INT_Q_OVERFLOW = (1 << 2), /* cmd msg q overflow */ - INT_Q_AVAILABLE = (1 << 1), /* cmd msg q has free space */ - INT_RESPONSE = (1 << 0), /* response msg available */ - INT_ACK_MASK = INT_WATCHDOG | INT_Q_OVERFLOW, - INT_DEF_MASK = INT_RESERVED | INT_Q_OVERFLOW | - INT_RESPONSE, - - /* command messages, and related register bits */ - CARM_HAVE_RESP = 0x01, - CARM_MSG_READ = 1, - CARM_MSG_WRITE = 2, - CARM_MSG_VERIFY = 3, - CARM_MSG_GET_CAPACITY = 4, - CARM_MSG_FLUSH = 5, - CARM_MSG_IOCTL = 6, - CARM_MSG_ARRAY = 8, - CARM_MSG_MISC = 9, - CARM_CME = (1 << 2), - CARM_RME = (1 << 1), - CARM_WZBC = (1 << 0), - CARM_RMI = (1 << 0), - CARM_Q_FULL = (1 << 3), - CARM_MSG_SIZE = 288, - CARM_Q_LEN = 48, - - /* CARM_MSG_IOCTL messages */ - CARM_IOC_SCAN_CHAN = 5, /* scan channels for devices */ - CARM_IOC_GET_TCQ = 13, /* get tcq/ncq depth */ - CARM_IOC_SET_TCQ = 14, /* set tcq/ncq depth */ - - IOC_SCAN_CHAN_NODEV = 0x1f, - IOC_SCAN_CHAN_OFFSET = 0x40, - - /* CARM_MSG_ARRAY messages */ - CARM_ARRAY_INFO = 0, - - ARRAY_NO_EXIST = (1 << 31), - - /* response messages */ - RMSG_SZ = 8, /* sizeof(struct carm_response) */ - RMSG_Q_LEN = 48, /* resp. msg list length */ - RMSG_OK = 1, /* bit indicating msg was successful */ - /* length of entire resp. msg buffer */ - RBUF_LEN = RMSG_SZ * RMSG_Q_LEN, - - PDC_SHM_SIZE = (4096 << 7), /* length of entire h/w buffer */ - - /* CARM_MSG_MISC messages */ - MISC_GET_FW_VER = 2, - MISC_ALLOC_MEM = 3, - MISC_SET_TIME = 5, - - /* MISC_GET_FW_VER feature bits */ - FW_VER_4PORT = (1 << 2), /* 1=4 ports, 0=8 ports */ - FW_VER_NON_RAID = (1 << 1), /* 1=non-RAID firmware, 0=RAID */ - FW_VER_ZCR = (1 << 0), /* zero channel RAID (whatever that is) */ - - /* carm_host flags */ - FL_NON_RAID = FW_VER_NON_RAID, - FL_4PORT = FW_VER_4PORT, - FL_FW_VER_MASK = (FW_VER_NON_RAID | FW_VER_4PORT), - FL_DYN_MAJOR = (1 << 17), -}; - -enum { - CARM_SG_BOUNDARY = 0xffffUL, /* s/g segment boundary */ -}; - -enum scatter_gather_types { - SGT_32BIT = 0, - SGT_64BIT = 1, -}; - -enum host_states { - HST_INVALID, /* invalid state; never used */ - HST_ALLOC_BUF, /* setting up master SHM area */ - HST_ERROR, /* we never leave here */ - HST_PORT_SCAN, /* start dev scan */ - HST_DEV_SCAN_START, /* start per-device probe */ - HST_DEV_SCAN, /* continue per-device probe */ - HST_DEV_ACTIVATE, /* activate devices we found */ - HST_PROBE_FINISHED, /* probe is complete */ - HST_PROBE_START, /* initiate probe */ - HST_SYNC_TIME, /* tell firmware what time it is */ - HST_GET_FW_VER, /* get firmware version, adapter port cnt */ -}; - -#ifdef CARM_DEBUG -static const char *state_name[] = { - "HST_INVALID", - "HST_ALLOC_BUF", - "HST_ERROR", - "HST_PORT_SCAN", - "HST_DEV_SCAN_START", - "HST_DEV_SCAN", - "HST_DEV_ACTIVATE", - "HST_PROBE_FINISHED", - "HST_PROBE_START", - "HST_SYNC_TIME", - "HST_GET_FW_VER", -}; -#endif - -struct carm_port { - unsigned int port_no; - struct gendisk *disk; - struct carm_host *host; - - /* attached device characteristics */ - u64 capacity; - char name[41]; - u16 dev_geom_head; - u16 dev_geom_sect; - u16 dev_geom_cyl; -}; - -struct carm_request { - int n_elem; - unsigned int msg_type; - unsigned int msg_subtype; - unsigned int msg_bucket; - struct scatterlist sg[CARM_MAX_REQ_SG]; -}; - -struct carm_host { - unsigned long flags; - void __iomem *mmio; - void *shm; - dma_addr_t shm_dma; - - int major; - int id; - char name[32]; - - spinlock_t lock; - struct pci_dev *pdev; - unsigned int state; - u32 fw_ver; - - struct blk_mq_tag_set tag_set; - struct request_queue *oob_q; - unsigned int n_oob; - - unsigned int hw_sg_used; - - unsigned int resp_idx; - - unsigned int wait_q_prod; - unsigned int wait_q_cons; - struct request_queue *wait_q[CARM_MAX_WAIT_Q]; - - void *msg_base; - dma_addr_t msg_dma; - - int cur_scan_dev; - unsigned long dev_active; - unsigned long dev_present; - struct carm_port port[CARM_MAX_PORTS]; - - struct work_struct fsm_task; - - struct completion probe_comp; -}; - -struct carm_response { - __le32 ret_handle; - __le32 status; -} __attribute__((packed)); - -struct carm_msg_sg { - __le32 start; - __le32 len; -} __attribute__((packed)); - -struct carm_msg_rw { - u8 type; - u8 id; - u8 sg_count; - u8 sg_type; - __le32 handle; - __le32 lba; - __le16 lba_count; - __le16 lba_high; - struct carm_msg_sg sg[32]; -} __attribute__((packed)); - -struct carm_msg_allocbuf { - u8 type; - u8 subtype; - u8 n_sg; - u8 sg_type; - __le32 handle; - __le32 addr; - __le32 len; - __le32 evt_pool; - __le32 n_evt; - __le32 rbuf_pool; - __le32 n_rbuf; - __le32 msg_pool; - __le32 n_msg; - struct carm_msg_sg sg[8]; -} __attribute__((packed)); - -struct carm_msg_ioctl { - u8 type; - u8 subtype; - u8 array_id; - u8 reserved1; - __le32 handle; - __le32 data_addr; - u32 reserved2; -} __attribute__((packed)); - -struct carm_msg_sync_time { - u8 type; - u8 subtype; - u16 reserved1; - __le32 handle; - u32 reserved2; - __le32 timestamp; -} __attribute__((packed)); - -struct carm_msg_get_fw_ver { - u8 type; - u8 subtype; - u16 reserved1; - __le32 handle; - __le32 data_addr; - u32 reserved2; -} __attribute__((packed)); - -struct carm_fw_ver { - __le32 version; - u8 features; - u8 reserved1; - u16 reserved2; -} __attribute__((packed)); - -struct carm_array_info { - __le32 size; - - __le16 size_hi; - __le16 stripe_size; - - __le32 mode; - - __le16 stripe_blk_sz; - __le16 reserved1; - - __le16 cyl; - __le16 head; - - __le16 sect; - u8 array_id; - u8 reserved2; - - char name[40]; - - __le32 array_status; - - /* device list continues beyond this point? */ -} __attribute__((packed)); - -static int carm_init_one (struct pci_dev *pdev, const struct pci_device_id *ent); -static void carm_remove_one (struct pci_dev *pdev); -static int carm_bdev_getgeo(struct block_device *bdev, struct hd_geometry *geo); - -static const struct pci_device_id carm_pci_tbl[] = { - { PCI_VENDOR_ID_PROMISE, 0x8000, PCI_ANY_ID, PCI_ANY_ID, 0, 0, }, - { PCI_VENDOR_ID_PROMISE, 0x8002, PCI_ANY_ID, PCI_ANY_ID, 0, 0, }, - { } /* terminate list */ -}; -MODULE_DEVICE_TABLE(pci, carm_pci_tbl); - -static struct pci_driver carm_driver = { - .name = DRV_NAME, - .id_table = carm_pci_tbl, - .probe = carm_init_one, - .remove = carm_remove_one, -}; - -static const struct block_device_operations carm_bd_ops = { - .owner = THIS_MODULE, - .getgeo = carm_bdev_getgeo, -}; - -static unsigned int carm_host_id; -static unsigned long carm_major_alloc; - - - -static int carm_bdev_getgeo(struct block_device *bdev, struct hd_geometry *geo) -{ - struct carm_port *port = bdev->bd_disk->private_data; - - geo->heads = (u8) port->dev_geom_head; - geo->sectors = (u8) port->dev_geom_sect; - geo->cylinders = port->dev_geom_cyl; - return 0; -} - -static const u32 msg_sizes[] = { 32, 64, 128, CARM_MSG_SIZE }; - -static inline int carm_lookup_bucket(u32 msg_size) -{ - int i; - - for (i = 0; i < ARRAY_SIZE(msg_sizes); i++) - if (msg_size <= msg_sizes[i]) - return i; - - return -ENOENT; -} - -static void carm_init_buckets(void __iomem *mmio) -{ - unsigned int i; - - for (i = 0; i < ARRAY_SIZE(msg_sizes); i++) - writel(msg_sizes[i], mmio + CARM_CMS0 + (4 * i)); -} - -static inline void *carm_ref_msg(struct carm_host *host, - unsigned int msg_idx) -{ - return host->msg_base + (msg_idx * CARM_MSG_SIZE); -} - -static inline dma_addr_t carm_ref_msg_dma(struct carm_host *host, - unsigned int msg_idx) -{ - return host->msg_dma + (msg_idx * CARM_MSG_SIZE); -} - -static int carm_send_msg(struct carm_host *host, - struct carm_request *crq, unsigned tag) -{ - void __iomem *mmio = host->mmio; - u32 msg = (u32) carm_ref_msg_dma(host, tag); - u32 cm_bucket = crq->msg_bucket; - u32 tmp; - int rc = 0; - - VPRINTK("ENTER\n"); - - tmp = readl(mmio + CARM_HMUC); - if (tmp & CARM_Q_FULL) { -#if 0 - tmp = readl(mmio + CARM_INT_MASK); - tmp |= INT_Q_AVAILABLE; - writel(tmp, mmio + CARM_INT_MASK); - readl(mmio + CARM_INT_MASK); /* flush */ -#endif - DPRINTK("host msg queue full\n"); - rc = -EBUSY; - } else { - writel(msg | (cm_bucket << 1), mmio + CARM_IHQP); - readl(mmio + CARM_IHQP); /* flush */ - } - - return rc; -} - -static int carm_array_info (struct carm_host *host, unsigned int array_idx) -{ - struct carm_msg_ioctl *ioc; - u32 msg_data; - dma_addr_t msg_dma; - struct carm_request *crq; - struct request *rq; - int rc; - - rq = blk_mq_alloc_request(host->oob_q, REQ_OP_DRV_OUT, 0); - if (IS_ERR(rq)) { - rc = -ENOMEM; - goto err_out; - } - crq = blk_mq_rq_to_pdu(rq); - - ioc = carm_ref_msg(host, rq->tag); - msg_dma = carm_ref_msg_dma(host, rq->tag); - msg_data = (u32) (msg_dma + sizeof(struct carm_array_info)); - - crq->msg_type = CARM_MSG_ARRAY; - crq->msg_subtype = CARM_ARRAY_INFO; - rc = carm_lookup_bucket(sizeof(struct carm_msg_ioctl) + - sizeof(struct carm_array_info)); - BUG_ON(rc < 0); - crq->msg_bucket = (u32) rc; - - memset(ioc, 0, sizeof(*ioc)); - ioc->type = CARM_MSG_ARRAY; - ioc->subtype = CARM_ARRAY_INFO; - ioc->array_id = (u8) array_idx; - ioc->handle = cpu_to_le32(TAG_ENCODE(rq->tag)); - ioc->data_addr = cpu_to_le32(msg_data); - - spin_lock_irq(&host->lock); - assert(host->state == HST_DEV_SCAN_START || - host->state == HST_DEV_SCAN); - spin_unlock_irq(&host->lock); - - DPRINTK("blk_execute_rq_nowait, tag == %u\n", rq->tag); - blk_execute_rq_nowait(host->oob_q, NULL, rq, true, NULL); - - return 0; - -err_out: - spin_lock_irq(&host->lock); - host->state = HST_ERROR; - spin_unlock_irq(&host->lock); - return rc; -} - -typedef unsigned int (*carm_sspc_t)(struct carm_host *, unsigned int, void *); - -static int carm_send_special (struct carm_host *host, carm_sspc_t func) -{ - struct request *rq; - struct carm_request *crq; - struct carm_msg_ioctl *ioc; - void *mem; - unsigned int msg_size; - int rc; - - rq = blk_mq_alloc_request(host->oob_q, REQ_OP_DRV_OUT, 0); - if (IS_ERR(rq)) - return -ENOMEM; - crq = blk_mq_rq_to_pdu(rq); - - mem = carm_ref_msg(host, rq->tag); - - msg_size = func(host, rq->tag, mem); - - ioc = mem; - crq->msg_type = ioc->type; - crq->msg_subtype = ioc->subtype; - rc = carm_lookup_bucket(msg_size); - BUG_ON(rc < 0); - crq->msg_bucket = (u32) rc; - - DPRINTK("blk_execute_rq_nowait, tag == %u\n", rq->tag); - blk_execute_rq_nowait(host->oob_q, NULL, rq, true, NULL); - - return 0; -} - -static unsigned int carm_fill_sync_time(struct carm_host *host, - unsigned int idx, void *mem) -{ - struct carm_msg_sync_time *st = mem; - - time64_t tv = ktime_get_real_seconds(); - - memset(st, 0, sizeof(*st)); - st->type = CARM_MSG_MISC; - st->subtype = MISC_SET_TIME; - st->handle = cpu_to_le32(TAG_ENCODE(idx)); - st->timestamp = cpu_to_le32(tv); - - return sizeof(struct carm_msg_sync_time); -} - -static unsigned int carm_fill_alloc_buf(struct carm_host *host, - unsigned int idx, void *mem) -{ - struct carm_msg_allocbuf *ab = mem; - - memset(ab, 0, sizeof(*ab)); - ab->type = CARM_MSG_MISC; - ab->subtype = MISC_ALLOC_MEM; - ab->handle = cpu_to_le32(TAG_ENCODE(idx)); - ab->n_sg = 1; - ab->sg_type = SGT_32BIT; - ab->addr = cpu_to_le32(host->shm_dma + (PDC_SHM_SIZE >> 1)); - ab->len = cpu_to_le32(PDC_SHM_SIZE >> 1); - ab->evt_pool = cpu_to_le32(host->shm_dma + (16 * 1024)); - ab->n_evt = cpu_to_le32(1024); - ab->rbuf_pool = cpu_to_le32(host->shm_dma); - ab->n_rbuf = cpu_to_le32(RMSG_Q_LEN); - ab->msg_pool = cpu_to_le32(host->shm_dma + RBUF_LEN); - ab->n_msg = cpu_to_le32(CARM_Q_LEN); - ab->sg[0].start = cpu_to_le32(host->shm_dma + (PDC_SHM_SIZE >> 1)); - ab->sg[0].len = cpu_to_le32(65536); - - return sizeof(struct carm_msg_allocbuf); -} - -static unsigned int carm_fill_scan_channels(struct carm_host *host, - unsigned int idx, void *mem) -{ - struct carm_msg_ioctl *ioc = mem; - u32 msg_data = (u32) (carm_ref_msg_dma(host, idx) + - IOC_SCAN_CHAN_OFFSET); - - memset(ioc, 0, sizeof(*ioc)); - ioc->type = CARM_MSG_IOCTL; - ioc->subtype = CARM_IOC_SCAN_CHAN; - ioc->handle = cpu_to_le32(TAG_ENCODE(idx)); - ioc->data_addr = cpu_to_le32(msg_data); - - /* fill output data area with "no device" default values */ - mem += IOC_SCAN_CHAN_OFFSET; - memset(mem, IOC_SCAN_CHAN_NODEV, CARM_MAX_PORTS); - - return IOC_SCAN_CHAN_OFFSET + CARM_MAX_PORTS; -} - -static unsigned int carm_fill_get_fw_ver(struct carm_host *host, - unsigned int idx, void *mem) -{ - struct carm_msg_get_fw_ver *ioc = mem; - u32 msg_data = (u32) (carm_ref_msg_dma(host, idx) + sizeof(*ioc)); - - memset(ioc, 0, sizeof(*ioc)); - ioc->type = CARM_MSG_MISC; - ioc->subtype = MISC_GET_FW_VER; - ioc->handle = cpu_to_le32(TAG_ENCODE(idx)); - ioc->data_addr = cpu_to_le32(msg_data); - - return sizeof(struct carm_msg_get_fw_ver) + - sizeof(struct carm_fw_ver); -} - -static inline void carm_push_q (struct carm_host *host, struct request_queue *q) -{ - unsigned int idx = host->wait_q_prod % CARM_MAX_WAIT_Q; - - blk_mq_stop_hw_queues(q); - VPRINTK("STOPPED QUEUE %p\n", q); - - host->wait_q[idx] = q; - host->wait_q_prod++; - BUG_ON(host->wait_q_prod == host->wait_q_cons); /* overrun */ -} - -static inline struct request_queue *carm_pop_q(struct carm_host *host) -{ - unsigned int idx; - - if (host->wait_q_prod == host->wait_q_cons) - return NULL; - - idx = host->wait_q_cons % CARM_MAX_WAIT_Q; - host->wait_q_cons++; - - return host->wait_q[idx]; -} - -static inline void carm_round_robin(struct carm_host *host) -{ - struct request_queue *q = carm_pop_q(host); - if (q) { - blk_mq_start_hw_queues(q); - VPRINTK("STARTED QUEUE %p\n", q); - } -} - -static inline enum dma_data_direction carm_rq_dir(struct request *rq) -{ - return op_is_write(req_op(rq)) ? DMA_TO_DEVICE : DMA_FROM_DEVICE; -} - -static blk_status_t carm_queue_rq(struct blk_mq_hw_ctx *hctx, - const struct blk_mq_queue_data *bd) -{ - struct request_queue *q = hctx->queue; - struct request *rq = bd->rq; - struct carm_port *port = q->queuedata; - struct carm_host *host = port->host; - struct carm_request *crq = blk_mq_rq_to_pdu(rq); - struct carm_msg_rw *msg; - struct scatterlist *sg; - int i, n_elem = 0, rc; - unsigned int msg_size; - u32 tmp; - - crq->n_elem = 0; - sg_init_table(crq->sg, CARM_MAX_REQ_SG); - - blk_mq_start_request(rq); - - spin_lock_irq(&host->lock); - if (req_op(rq) == REQ_OP_DRV_OUT) - goto send_msg; - - /* get scatterlist from block layer */ - sg = &crq->sg[0]; - n_elem = blk_rq_map_sg(q, rq, sg); - if (n_elem <= 0) - goto out_ioerr; - - /* map scatterlist to PCI bus addresses */ - n_elem = dma_map_sg(&host->pdev->dev, sg, n_elem, carm_rq_dir(rq)); - if (n_elem <= 0) - goto out_ioerr; - - /* obey global hardware limit on S/G entries */ - if (host->hw_sg_used >= CARM_MAX_HOST_SG - n_elem) - goto out_resource; - - crq->n_elem = n_elem; - host->hw_sg_used += n_elem; - - /* - * build read/write message - */ - - VPRINTK("build msg\n"); - msg = (struct carm_msg_rw *) carm_ref_msg(host, rq->tag); - - if (rq_data_dir(rq) == WRITE) { - msg->type = CARM_MSG_WRITE; - crq->msg_type = CARM_MSG_WRITE; - } else { - msg->type = CARM_MSG_READ; - crq->msg_type = CARM_MSG_READ; - } - - msg->id = port->port_no; - msg->sg_count = n_elem; - msg->sg_type = SGT_32BIT; - msg->handle = cpu_to_le32(TAG_ENCODE(rq->tag)); - msg->lba = cpu_to_le32(blk_rq_pos(rq) & 0xffffffff); - tmp = (blk_rq_pos(rq) >> 16) >> 16; - msg->lba_high = cpu_to_le16( (u16) tmp ); - msg->lba_count = cpu_to_le16(blk_rq_sectors(rq)); - - msg_size = sizeof(struct carm_msg_rw) - sizeof(msg->sg); - for (i = 0; i < n_elem; i++) { - struct carm_msg_sg *carm_sg = &msg->sg[i]; - carm_sg->start = cpu_to_le32(sg_dma_address(&crq->sg[i])); - carm_sg->len = cpu_to_le32(sg_dma_len(&crq->sg[i])); - msg_size += sizeof(struct carm_msg_sg); - } - - rc = carm_lookup_bucket(msg_size); - BUG_ON(rc < 0); - crq->msg_bucket = (u32) rc; -send_msg: - /* - * queue read/write message to hardware - */ - VPRINTK("send msg, tag == %u\n", rq->tag); - rc = carm_send_msg(host, crq, rq->tag); - if (rc) { - host->hw_sg_used -= n_elem; - goto out_resource; - } - - spin_unlock_irq(&host->lock); - return BLK_STS_OK; -out_resource: - dma_unmap_sg(&host->pdev->dev, &crq->sg[0], n_elem, carm_rq_dir(rq)); - carm_push_q(host, q); - spin_unlock_irq(&host->lock); - return BLK_STS_DEV_RESOURCE; -out_ioerr: - carm_round_robin(host); - spin_unlock_irq(&host->lock); - return BLK_STS_IOERR; -} - -static void carm_handle_array_info(struct carm_host *host, - struct carm_request *crq, u8 *mem, - blk_status_t error) -{ - struct carm_port *port; - u8 *msg_data = mem + sizeof(struct carm_array_info); - struct carm_array_info *desc = (struct carm_array_info *) msg_data; - u64 lo, hi; - int cur_port; - size_t slen; - - DPRINTK("ENTER\n"); - - if (error) - goto out; - if (le32_to_cpu(desc->array_status) & ARRAY_NO_EXIST) - goto out; - - cur_port = host->cur_scan_dev; - - /* should never occur */ - if ((cur_port < 0) || (cur_port >= CARM_MAX_PORTS)) { - printk(KERN_ERR PFX "BUG: cur_scan_dev==%d, array_id==%d\n", - cur_port, (int) desc->array_id); - goto out; - } - - port = &host->port[cur_port]; - - lo = (u64) le32_to_cpu(desc->size); - hi = (u64) le16_to_cpu(desc->size_hi); - - port->capacity = lo | (hi << 32); - port->dev_geom_head = le16_to_cpu(desc->head); - port->dev_geom_sect = le16_to_cpu(desc->sect); - port->dev_geom_cyl = le16_to_cpu(desc->cyl); - - host->dev_active |= (1 << cur_port); - - strncpy(port->name, desc->name, sizeof(port->name)); - port->name[sizeof(port->name) - 1] = 0; - slen = strlen(port->name); - while (slen && (port->name[slen - 1] == ' ')) { - port->name[slen - 1] = 0; - slen--; - } - - printk(KERN_INFO DRV_NAME "(%s): port %u device %Lu sectors\n", - pci_name(host->pdev), port->port_no, - (unsigned long long) port->capacity); - printk(KERN_INFO DRV_NAME "(%s): port %u device \"%s\"\n", - pci_name(host->pdev), port->port_no, port->name); - -out: - assert(host->state == HST_DEV_SCAN); - schedule_work(&host->fsm_task); -} - -static void carm_handle_scan_chan(struct carm_host *host, - struct carm_request *crq, u8 *mem, - blk_status_t error) -{ - u8 *msg_data = mem + IOC_SCAN_CHAN_OFFSET; - unsigned int i, dev_count = 0; - int new_state = HST_DEV_SCAN_START; - - DPRINTK("ENTER\n"); - - if (error) { - new_state = HST_ERROR; - goto out; - } - - /* TODO: scan and support non-disk devices */ - for (i = 0; i < 8; i++) - if (msg_data[i] == 0) { /* direct-access device (disk) */ - host->dev_present |= (1 << i); - dev_count++; - } - - printk(KERN_INFO DRV_NAME "(%s): found %u interesting devices\n", - pci_name(host->pdev), dev_count); - -out: - assert(host->state == HST_PORT_SCAN); - host->state = new_state; - schedule_work(&host->fsm_task); -} - -static void carm_handle_generic(struct carm_host *host, - struct carm_request *crq, blk_status_t error, - int cur_state, int next_state) -{ - DPRINTK("ENTER\n"); - - assert(host->state == cur_state); - if (error) - host->state = HST_ERROR; - else - host->state = next_state; - schedule_work(&host->fsm_task); -} - -static inline void carm_handle_resp(struct carm_host *host, - __le32 ret_handle_le, u32 status) -{ - u32 handle = le32_to_cpu(ret_handle_le); - unsigned int msg_idx; - struct request *rq; - struct carm_request *crq; - blk_status_t error = (status == RMSG_OK) ? 0 : BLK_STS_IOERR; - u8 *mem; - - VPRINTK("ENTER, handle == 0x%x\n", handle); - - if (unlikely(!TAG_VALID(handle))) { - printk(KERN_ERR DRV_NAME "(%s): BUG: invalid tag 0x%x\n", - pci_name(host->pdev), handle); - return; - } - - msg_idx = TAG_DECODE(handle); - VPRINTK("tag == %u\n", msg_idx); - - rq = blk_mq_tag_to_rq(host->tag_set.tags[0], msg_idx); - crq = blk_mq_rq_to_pdu(rq); - - /* fast path */ - if (likely(crq->msg_type == CARM_MSG_READ || - crq->msg_type == CARM_MSG_WRITE)) { - dma_unmap_sg(&host->pdev->dev, &crq->sg[0], crq->n_elem, - carm_rq_dir(rq)); - goto done; - } - - mem = carm_ref_msg(host, msg_idx); - - switch (crq->msg_type) { - case CARM_MSG_IOCTL: { - switch (crq->msg_subtype) { - case CARM_IOC_SCAN_CHAN: - carm_handle_scan_chan(host, crq, mem, error); - goto done; - default: - /* unknown / invalid response */ - goto err_out; - } - break; - } - - case CARM_MSG_MISC: { - switch (crq->msg_subtype) { - case MISC_ALLOC_MEM: - carm_handle_generic(host, crq, error, - HST_ALLOC_BUF, HST_SYNC_TIME); - goto done; - case MISC_SET_TIME: - carm_handle_generic(host, crq, error, - HST_SYNC_TIME, HST_GET_FW_VER); - goto done; - case MISC_GET_FW_VER: { - struct carm_fw_ver *ver = (struct carm_fw_ver *) - (mem + sizeof(struct carm_msg_get_fw_ver)); - if (!error) { - host->fw_ver = le32_to_cpu(ver->version); - host->flags |= (ver->features & FL_FW_VER_MASK); - } - carm_handle_generic(host, crq, error, - HST_GET_FW_VER, HST_PORT_SCAN); - goto done; - } - default: - /* unknown / invalid response */ - goto err_out; - } - break; - } - - case CARM_MSG_ARRAY: { - switch (crq->msg_subtype) { - case CARM_ARRAY_INFO: - carm_handle_array_info(host, crq, mem, error); - break; - default: - /* unknown / invalid response */ - goto err_out; - } - break; - } - - default: - /* unknown / invalid response */ - goto err_out; - } - - return; - -err_out: - printk(KERN_WARNING DRV_NAME "(%s): BUG: unhandled message type %d/%d\n", - pci_name(host->pdev), crq->msg_type, crq->msg_subtype); - error = BLK_STS_IOERR; -done: - host->hw_sg_used -= crq->n_elem; - blk_mq_end_request(blk_mq_rq_from_pdu(crq), error); - - if (host->hw_sg_used <= CARM_SG_LOW_WATER) - carm_round_robin(host); -} - -static inline void carm_handle_responses(struct carm_host *host) -{ - void __iomem *mmio = host->mmio; - struct carm_response *resp = (struct carm_response *) host->shm; - unsigned int work = 0; - unsigned int idx = host->resp_idx % RMSG_Q_LEN; - - while (1) { - u32 status = le32_to_cpu(resp[idx].status); - - if (status == 0xffffffff) { - VPRINTK("ending response on index %u\n", idx); - writel(idx << 3, mmio + CARM_RESP_IDX); - break; - } - - /* response to a message we sent */ - else if ((status & (1 << 31)) == 0) { - VPRINTK("handling msg response on index %u\n", idx); - carm_handle_resp(host, resp[idx].ret_handle, status); - resp[idx].status = cpu_to_le32(0xffffffff); - } - - /* asynchronous events the hardware throws our way */ - else if ((status & 0xff000000) == (1 << 31)) { - u8 *evt_type_ptr = (u8 *) &resp[idx]; - u8 evt_type = *evt_type_ptr; - printk(KERN_WARNING DRV_NAME "(%s): unhandled event type %d\n", - pci_name(host->pdev), (int) evt_type); - resp[idx].status = cpu_to_le32(0xffffffff); - } - - idx = NEXT_RESP(idx); - work++; - } - - VPRINTK("EXIT, work==%u\n", work); - host->resp_idx += work; -} - -static irqreturn_t carm_interrupt(int irq, void *__host) -{ - struct carm_host *host = __host; - void __iomem *mmio; - u32 mask; - int handled = 0; - unsigned long flags; - - if (!host) { - VPRINTK("no host\n"); - return IRQ_NONE; - } - - spin_lock_irqsave(&host->lock, flags); - - mmio = host->mmio; - - /* reading should also clear interrupts */ - mask = readl(mmio + CARM_INT_STAT); - - if (mask == 0 || mask == 0xffffffff) { - VPRINTK("no work, mask == 0x%x\n", mask); - goto out; - } - - if (mask & INT_ACK_MASK) - writel(mask, mmio + CARM_INT_STAT); - - if (unlikely(host->state == HST_INVALID)) { - VPRINTK("not initialized yet, mask = 0x%x\n", mask); - goto out; - } - - if (mask & CARM_HAVE_RESP) { - handled = 1; - carm_handle_responses(host); - } - -out: - spin_unlock_irqrestore(&host->lock, flags); - VPRINTK("EXIT\n"); - return IRQ_RETVAL(handled); -} - -static void carm_fsm_task (struct work_struct *work) -{ - struct carm_host *host = - container_of(work, struct carm_host, fsm_task); - unsigned long flags; - unsigned int state; - int rc, i, next_dev; - int reschedule = 0; - int new_state = HST_INVALID; - - spin_lock_irqsave(&host->lock, flags); - state = host->state; - spin_unlock_irqrestore(&host->lock, flags); - - DPRINTK("ENTER, state == %s\n", state_name[state]); - - switch (state) { - case HST_PROBE_START: - new_state = HST_ALLOC_BUF; - reschedule = 1; - break; - - case HST_ALLOC_BUF: - rc = carm_send_special(host, carm_fill_alloc_buf); - if (rc) { - new_state = HST_ERROR; - reschedule = 1; - } - break; - - case HST_SYNC_TIME: - rc = carm_send_special(host, carm_fill_sync_time); - if (rc) { - new_state = HST_ERROR; - reschedule = 1; - } - break; - - case HST_GET_FW_VER: - rc = carm_send_special(host, carm_fill_get_fw_ver); - if (rc) { - new_state = HST_ERROR; - reschedule = 1; - } - break; - - case HST_PORT_SCAN: - rc = carm_send_special(host, carm_fill_scan_channels); - if (rc) { - new_state = HST_ERROR; - reschedule = 1; - } - break; - - case HST_DEV_SCAN_START: - host->cur_scan_dev = -1; - new_state = HST_DEV_SCAN; - reschedule = 1; - break; - - case HST_DEV_SCAN: - next_dev = -1; - for (i = host->cur_scan_dev + 1; i < CARM_MAX_PORTS; i++) - if (host->dev_present & (1 << i)) { - next_dev = i; - break; - } - - if (next_dev >= 0) { - host->cur_scan_dev = next_dev; - rc = carm_array_info(host, next_dev); - if (rc) { - new_state = HST_ERROR; - reschedule = 1; - } - } else { - new_state = HST_DEV_ACTIVATE; - reschedule = 1; - } - break; - - case HST_DEV_ACTIVATE: { - int activated = 0; - for (i = 0; i < CARM_MAX_PORTS; i++) - if (host->dev_active & (1 << i)) { - struct carm_port *port = &host->port[i]; - struct gendisk *disk = port->disk; - - set_capacity(disk, port->capacity); - add_disk(disk); - activated++; - } - - printk(KERN_INFO DRV_NAME "(%s): %d ports activated\n", - pci_name(host->pdev), activated); - - new_state = HST_PROBE_FINISHED; - reschedule = 1; - break; - } - - case HST_PROBE_FINISHED: - complete(&host->probe_comp); - break; - - case HST_ERROR: - /* FIXME: TODO */ - break; - - default: - /* should never occur */ - printk(KERN_ERR PFX "BUG: unknown state %d\n", state); - assert(0); - break; - } - - if (new_state != HST_INVALID) { - spin_lock_irqsave(&host->lock, flags); - host->state = new_state; - spin_unlock_irqrestore(&host->lock, flags); - } - if (reschedule) - schedule_work(&host->fsm_task); -} - -static int carm_init_wait(void __iomem *mmio, u32 bits, unsigned int test_bit) -{ - unsigned int i; - - for (i = 0; i < 50000; i++) { - u32 tmp = readl(mmio + CARM_LMUC); - udelay(100); - - if (test_bit) { - if ((tmp & bits) == bits) - return 0; - } else { - if ((tmp & bits) == 0) - return 0; - } - - cond_resched(); - } - - printk(KERN_ERR PFX "carm_init_wait timeout, bits == 0x%x, test_bit == %s\n", - bits, test_bit ? "yes" : "no"); - return -EBUSY; -} - -static void carm_init_responses(struct carm_host *host) -{ - void __iomem *mmio = host->mmio; - unsigned int i; - struct carm_response *resp = (struct carm_response *) host->shm; - - for (i = 0; i < RMSG_Q_LEN; i++) - resp[i].status = cpu_to_le32(0xffffffff); - - writel(0, mmio + CARM_RESP_IDX); -} - -static int carm_init_host(struct carm_host *host) -{ - void __iomem *mmio = host->mmio; - u32 tmp; - u8 tmp8; - int rc; - - DPRINTK("ENTER\n"); - - writel(0, mmio + CARM_INT_MASK); - - tmp8 = readb(mmio + CARM_INITC); - if (tmp8 & 0x01) { - tmp8 &= ~0x01; - writeb(tmp8, mmio + CARM_INITC); - readb(mmio + CARM_INITC); /* flush */ - - DPRINTK("snooze...\n"); - msleep(5000); - } - - tmp = readl(mmio + CARM_HMUC); - if (tmp & CARM_CME) { - DPRINTK("CME bit present, waiting\n"); - rc = carm_init_wait(mmio, CARM_CME, 1); - if (rc) { - DPRINTK("EXIT, carm_init_wait 1 failed\n"); - return rc; - } - } - if (tmp & CARM_RME) { - DPRINTK("RME bit present, waiting\n"); - rc = carm_init_wait(mmio, CARM_RME, 1); - if (rc) { - DPRINTK("EXIT, carm_init_wait 2 failed\n"); - return rc; - } - } - - tmp &= ~(CARM_RME | CARM_CME); - writel(tmp, mmio + CARM_HMUC); - readl(mmio + CARM_HMUC); /* flush */ - - rc = carm_init_wait(mmio, CARM_RME | CARM_CME, 0); - if (rc) { - DPRINTK("EXIT, carm_init_wait 3 failed\n"); - return rc; - } - - carm_init_buckets(mmio); - - writel(host->shm_dma & 0xffffffff, mmio + RBUF_ADDR_LO); - writel((host->shm_dma >> 16) >> 16, mmio + RBUF_ADDR_HI); - writel(RBUF_LEN, mmio + RBUF_BYTE_SZ); - - tmp = readl(mmio + CARM_HMUC); - tmp |= (CARM_RME | CARM_CME | CARM_WZBC); - writel(tmp, mmio + CARM_HMUC); - readl(mmio + CARM_HMUC); /* flush */ - - rc = carm_init_wait(mmio, CARM_RME | CARM_CME, 1); - if (rc) { - DPRINTK("EXIT, carm_init_wait 4 failed\n"); - return rc; - } - - writel(0, mmio + CARM_HMPHA); - writel(INT_DEF_MASK, mmio + CARM_INT_MASK); - - carm_init_responses(host); - - /* start initialization, probing state machine */ - spin_lock_irq(&host->lock); - assert(host->state == HST_INVALID); - host->state = HST_PROBE_START; - spin_unlock_irq(&host->lock); - schedule_work(&host->fsm_task); - - DPRINTK("EXIT\n"); - return 0; -} - -static const struct blk_mq_ops carm_mq_ops = { - .queue_rq = carm_queue_rq, -}; - -static int carm_init_disk(struct carm_host *host, unsigned int port_no) -{ - struct carm_port *port = &host->port[port_no]; - struct gendisk *disk; - struct request_queue *q; - - port->host = host; - port->port_no = port_no; - - disk = alloc_disk(CARM_MINORS_PER_MAJOR); - if (!disk) - return -ENOMEM; - - port->disk = disk; - sprintf(disk->disk_name, DRV_NAME "/%u", - (unsigned int)host->id * CARM_MAX_PORTS + port_no); - disk->major = host->major; - disk->first_minor = port_no * CARM_MINORS_PER_MAJOR; - disk->fops = &carm_bd_ops; - disk->private_data = port; - - q = blk_mq_init_queue(&host->tag_set); - if (IS_ERR(q)) - return PTR_ERR(q); - - blk_queue_max_segments(q, CARM_MAX_REQ_SG); - blk_queue_segment_boundary(q, CARM_SG_BOUNDARY); - - q->queuedata = port; - disk->queue = q; - return 0; -} - -static void carm_free_disk(struct carm_host *host, unsigned int port_no) -{ - struct carm_port *port = &host->port[port_no]; - struct gendisk *disk = port->disk; - - if (!disk) - return; - - if (disk->flags & GENHD_FL_UP) - del_gendisk(disk); - if (disk->queue) - blk_cleanup_queue(disk->queue); - put_disk(disk); -} - -static int carm_init_shm(struct carm_host *host) -{ - host->shm = dma_alloc_coherent(&host->pdev->dev, CARM_SHM_SIZE, - &host->shm_dma, GFP_KERNEL); - if (!host->shm) - return -ENOMEM; - - host->msg_base = host->shm + RBUF_LEN; - host->msg_dma = host->shm_dma + RBUF_LEN; - - memset(host->shm, 0xff, RBUF_LEN); - memset(host->msg_base, 0, PDC_SHM_SIZE - RBUF_LEN); - - return 0; -} - -static int carm_init_one (struct pci_dev *pdev, const struct pci_device_id *ent) -{ - struct carm_host *host; - int rc; - struct request_queue *q; - unsigned int i; - - printk_once(KERN_DEBUG DRV_NAME " version " DRV_VERSION "\n"); - - rc = pci_enable_device(pdev); - if (rc) - return rc; - - rc = pci_request_regions(pdev, DRV_NAME); - if (rc) - goto err_out; - - rc = dma_set_mask(&pdev->dev, DMA_BIT_MASK(32)); - if (rc) { - printk(KERN_ERR DRV_NAME "(%s): DMA mask failure\n", - pci_name(pdev)); - goto err_out_regions; - } - - host = kzalloc(sizeof(*host), GFP_KERNEL); - if (!host) { - printk(KERN_ERR DRV_NAME "(%s): memory alloc failure\n", - pci_name(pdev)); - rc = -ENOMEM; - goto err_out_regions; - } - - host->pdev = pdev; - spin_lock_init(&host->lock); - INIT_WORK(&host->fsm_task, carm_fsm_task); - init_completion(&host->probe_comp); - - host->mmio = ioremap(pci_resource_start(pdev, 0), - pci_resource_len(pdev, 0)); - if (!host->mmio) { - printk(KERN_ERR DRV_NAME "(%s): MMIO alloc failure\n", - pci_name(pdev)); - rc = -ENOMEM; - goto err_out_kfree; - } - - rc = carm_init_shm(host); - if (rc) { - printk(KERN_ERR DRV_NAME "(%s): DMA SHM alloc failure\n", - pci_name(pdev)); - goto err_out_iounmap; - } - - memset(&host->tag_set, 0, sizeof(host->tag_set)); - host->tag_set.ops = &carm_mq_ops; - host->tag_set.cmd_size = sizeof(struct carm_request); - host->tag_set.nr_hw_queues = 1; - host->tag_set.nr_maps = 1; - host->tag_set.queue_depth = max_queue; - host->tag_set.numa_node = NUMA_NO_NODE; - host->tag_set.flags = BLK_MQ_F_SHOULD_MERGE; - - rc = blk_mq_alloc_tag_set(&host->tag_set); - if (rc) - goto err_out_dma_free; - - q = blk_mq_init_queue(&host->tag_set); - if (IS_ERR(q)) { - rc = PTR_ERR(q); - blk_mq_free_tag_set(&host->tag_set); - goto err_out_dma_free; - } - - host->oob_q = q; - q->queuedata = host; - - /* - * Figure out which major to use: 160, 161, or dynamic - */ - if (!test_and_set_bit(0, &carm_major_alloc)) - host->major = 160; - else if (!test_and_set_bit(1, &carm_major_alloc)) - host->major = 161; - else - host->flags |= FL_DYN_MAJOR; - - host->id = carm_host_id; - sprintf(host->name, DRV_NAME "%d", carm_host_id); - - rc = register_blkdev(host->major, host->name); - if (rc < 0) - goto err_out_free_majors; - if (host->flags & FL_DYN_MAJOR) - host->major = rc; - - for (i = 0; i < CARM_MAX_PORTS; i++) { - rc = carm_init_disk(host, i); - if (rc) - goto err_out_blkdev_disks; - } - - pci_set_master(pdev); - - rc = request_irq(pdev->irq, carm_interrupt, IRQF_SHARED, DRV_NAME, host); - if (rc) { - printk(KERN_ERR DRV_NAME "(%s): irq alloc failure\n", - pci_name(pdev)); - goto err_out_blkdev_disks; - } - - rc = carm_init_host(host); - if (rc) - goto err_out_free_irq; - - DPRINTK("waiting for probe_comp\n"); - wait_for_completion(&host->probe_comp); - - printk(KERN_INFO "%s: pci %s, ports %d, io %llx, irq %u, major %d\n", - host->name, pci_name(pdev), (int) CARM_MAX_PORTS, - (unsigned long long)pci_resource_start(pdev, 0), - pdev->irq, host->major); - - carm_host_id++; - pci_set_drvdata(pdev, host); - return 0; - -err_out_free_irq: - free_irq(pdev->irq, host); -err_out_blkdev_disks: - for (i = 0; i < CARM_MAX_PORTS; i++) - carm_free_disk(host, i); - unregister_blkdev(host->major, host->name); -err_out_free_majors: - if (host->major == 160) - clear_bit(0, &carm_major_alloc); - else if (host->major == 161) - clear_bit(1, &carm_major_alloc); - blk_cleanup_queue(host->oob_q); - blk_mq_free_tag_set(&host->tag_set); -err_out_dma_free: - dma_free_coherent(&pdev->dev, CARM_SHM_SIZE, host->shm, host->shm_dma); -err_out_iounmap: - iounmap(host->mmio); -err_out_kfree: - kfree(host); -err_out_regions: - pci_release_regions(pdev); -err_out: - pci_disable_device(pdev); - return rc; -} - -static void carm_remove_one (struct pci_dev *pdev) -{ - struct carm_host *host = pci_get_drvdata(pdev); - unsigned int i; - - if (!host) { - printk(KERN_ERR PFX "BUG: no host data for PCI(%s)\n", - pci_name(pdev)); - return; - } - - free_irq(pdev->irq, host); - for (i = 0; i < CARM_MAX_PORTS; i++) - carm_free_disk(host, i); - unregister_blkdev(host->major, host->name); - if (host->major == 160) - clear_bit(0, &carm_major_alloc); - else if (host->major == 161) - clear_bit(1, &carm_major_alloc); - blk_cleanup_queue(host->oob_q); - blk_mq_free_tag_set(&host->tag_set); - dma_free_coherent(&pdev->dev, CARM_SHM_SIZE, host->shm, host->shm_dma); - iounmap(host->mmio); - kfree(host); - pci_release_regions(pdev); - pci_disable_device(pdev); -} - -module_pci_driver(carm_driver); diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c index 9b3ea86c20e5..3afc07b59477 100644 --- a/drivers/block/virtio_blk.c +++ b/drivers/block/virtio_blk.c @@ -1063,14 +1063,15 @@ static int virtblk_freeze(struct virtio_device *vdev) { struct virtio_blk *vblk = vdev->priv; + /* Ensure no requests in virtqueues before deleting vqs. */ + blk_mq_freeze_queue(vblk->disk->queue); + /* Ensure we don't receive any more interrupts */ vdev->config->reset(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); @@ -1088,7 +1089,7 @@ static int virtblk_restore(struct virtio_device *vdev) virtio_device_ready(vdev); - blk_mq_unquiesce_queue(vblk->disk->queue); + blk_mq_unfreeze_queue(vblk->disk->queue); return 0; } #endif diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index d0538c03f033..da67621ebc21 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -779,7 +779,8 @@ static int blkif_queue_rw_req(struct request *req, struct blkfront_ring_info *ri ring_req->u.rw.handle = info->handle; ring_req->operation = rq_data_dir(req) ? BLKIF_OP_WRITE : BLKIF_OP_READ; - if (req_op(req) == REQ_OP_FLUSH || req->cmd_flags & REQ_FUA) { + if (req_op(req) == REQ_OP_FLUSH || + (req_op(req) == REQ_OP_WRITE && (req->cmd_flags & REQ_FUA))) { /* * Ideally we can do an unordered flush-to-disk. * In case the backend onlysupports barriers, use that. |