* block device image metadata (in-memory version)
*/
struct rbd_image_header {
- /* These four fields never change for a given rbd image */
+ /* These six fields never change for a given rbd image */
char *object_prefix;
- u64 features;
__u8 obj_order;
__u8 crypt_type;
__u8 comp_type;
+ u64 stripe_unit;
+ u64 stripe_count;
+ u64 features; /* Might be changeable someday? */
/* The remaining fields need to be updated occasionally */
u64 image_size;
struct ceph_snap_context *snapc;
- char *snap_names;
- u64 *snap_sizes;
-
- u64 stripe_unit;
- u64 stripe_count;
+ char *snap_names; /* format 1 only */
+ u64 *snap_sizes; /* format 1 only */
};
/*
};
};
struct page **copyup_pages;
+ u32 copyup_page_count;
struct ceph_osd_request *osd_req;
struct rbd_obj_request *obj_request; /* obj req initiator */
};
struct page **copyup_pages;
+ u32 copyup_page_count;
spinlock_t completion_lock;/* protects next_completion */
u32 next_completion;
rbd_img_callback_t callback;
size_t count);
static ssize_t rbd_remove(struct bus_type *bus, const char *buf,
size_t count);
-static int rbd_dev_image_probe(struct rbd_device *rbd_dev);
+static int rbd_dev_image_probe(struct rbd_device *rbd_dev, bool mapping);
static struct bus_attribute rbd_bus_attrs[] = {
__ATTR(add, S_IWUSR, NULL, rbd_add),
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_refresh(struct rbd_device *rbd_dev);
+static int rbd_dev_v2_header_onetime(struct rbd_device *rbd_dev);
+static int rbd_dev_v2_header_info(struct rbd_device *rbd_dev);
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,
}
/*
- * Create a new header structure, translate header format from the on-disk
- * header.
+ * Fill an rbd image header with information from the given format 1
+ * on-disk header.
*/
-static int rbd_header_from_disk(struct rbd_image_header *header,
+static int rbd_header_from_disk(struct rbd_device *rbd_dev,
struct rbd_image_header_ondisk *ondisk)
{
+ 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;
+ u64 *snap_sizes = NULL;
u32 snap_count;
- size_t len;
size_t size;
+ int ret = -ENOMEM;
u32 i;
- memset(header, 0, sizeof (*header));
+ /* Allocate this now to avoid having to handle failure below */
- snap_count = le32_to_cpu(ondisk->snap_count);
+ if (first_time) {
+ size_t len;
- len = strnlen(ondisk->object_prefix, sizeof (ondisk->object_prefix));
- header->object_prefix = kmalloc(len + 1, GFP_KERNEL);
- if (!header->object_prefix)
- return -ENOMEM;
- memcpy(header->object_prefix, ondisk->object_prefix, len);
- header->object_prefix[len] = '\0';
+ len = strnlen(ondisk->object_prefix,
+ sizeof (ondisk->object_prefix));
+ object_prefix = kmalloc(len + 1, GFP_KERNEL);
+ if (!object_prefix)
+ return -ENOMEM;
+ memcpy(object_prefix, ondisk->object_prefix, len);
+ object_prefix[len] = '\0';
+ }
+ /* Allocate the snapshot context and fill it in */
+
+ snap_count = le32_to_cpu(ondisk->snap_count);
+ snapc = ceph_create_snap_context(snap_count, GFP_KERNEL);
+ if (!snapc)
+ goto out_err;
+ snapc->seq = le64_to_cpu(ondisk->snap_seq);
if (snap_count) {
+ struct rbd_image_snap_ondisk *snaps;
u64 snap_names_len = le64_to_cpu(ondisk->snap_names_len);
- /* Save a copy of the snapshot names */
+ /* We'll keep a copy of the snapshot names... */
+
+ if (snap_names_len > (u64)SIZE_MAX)
+ goto out_2big;
+ snap_names = kmalloc(snap_names_len, GFP_KERNEL);
+ if (!snap_names)
+ goto out_err;
+
+ /* ...as well as the array of their sizes. */
- if (snap_names_len > (u64) SIZE_MAX)
- return -EIO;
- header->snap_names = kmalloc(snap_names_len, GFP_KERNEL);
- if (!header->snap_names)
+ size = snap_count * sizeof (*header->snap_sizes);
+ snap_sizes = kmalloc(size, GFP_KERNEL);
+ if (!snap_sizes)
goto out_err;
+
/*
- * Note that rbd_dev_v1_header_read() guarantees
- * the ondisk buffer we're working with has
+ * Copy the names, and fill in each snapshot's id
+ * and size.
+ *
+ * Note that rbd_dev_v1_header_info() guarantees the
+ * ondisk buffer we're working with has
* snap_names_len bytes beyond the end of the
* snapshot id array, this memcpy() is safe.
*/
- memcpy(header->snap_names, &ondisk->snaps[snap_count],
- snap_names_len);
+ memcpy(snap_names, &ondisk->snaps[snap_count], snap_names_len);
+ snaps = ondisk->snaps;
+ for (i = 0; i < snap_count; i++) {
+ snapc->snaps[i] = le64_to_cpu(snaps[i].id);
+ snap_sizes[i] = le64_to_cpu(snaps[i].image_size);
+ }
+ }
- /* Record each snapshot's size */
+ /* We won't fail any more, fill in the header */
- size = snap_count * sizeof (*header->snap_sizes);
- header->snap_sizes = kmalloc(size, GFP_KERNEL);
- if (!header->snap_sizes)
- goto out_err;
- for (i = 0; i < snap_count; i++)
- header->snap_sizes[i] =
- le64_to_cpu(ondisk->snaps[i].image_size);
+ down_write(&rbd_dev->header_rwsem);
+ if (first_time) {
+ header->object_prefix = object_prefix;
+ header->obj_order = ondisk->options.order;
+ header->crypt_type = ondisk->options.crypt_type;
+ header->comp_type = ondisk->options.comp_type;
+ /* The rest aren't used for format 1 images */
+ header->stripe_unit = 0;
+ header->stripe_count = 0;
+ header->features = 0;
} else {
- header->snap_names = NULL;
- header->snap_sizes = NULL;
+ ceph_put_snap_context(header->snapc);
+ kfree(header->snap_names);
+ kfree(header->snap_sizes);
}
- header->features = 0; /* No features support in v1 images */
- header->obj_order = ondisk->options.order;
- header->crypt_type = ondisk->options.crypt_type;
- header->comp_type = ondisk->options.comp_type;
-
- /* Allocate and fill in the snapshot context */
+ /* The remaining fields always get updated (when we refresh) */
header->image_size = le64_to_cpu(ondisk->image_size);
+ header->snapc = snapc;
+ header->snap_names = snap_names;
+ header->snap_sizes = snap_sizes;
- header->snapc = ceph_create_snap_context(snap_count, GFP_KERNEL);
- if (!header->snapc)
- goto out_err;
- header->snapc->seq = le64_to_cpu(ondisk->snap_seq);
- for (i = 0; i < snap_count; i++)
- header->snapc->snaps[i] = le64_to_cpu(ondisk->snaps[i].id);
+ /* Make sure mapping size is consistent with header info */
- return 0;
+ if (rbd_dev->spec->snap_id == CEPH_NOSNAP || first_time)
+ if (rbd_dev->mapping.size != header->image_size)
+ rbd_dev->mapping.size = header->image_size;
+
+ up_write(&rbd_dev->header_rwsem);
+ return 0;
+out_2big:
+ ret = -EIO;
out_err:
- kfree(header->snap_sizes);
- header->snap_sizes = NULL;
- kfree(header->snap_names);
- header->snap_names = NULL;
- kfree(header->object_prefix);
- header->object_prefix = NULL;
+ kfree(snap_sizes);
+ kfree(snap_names);
+ ceph_put_snap_context(snapc);
+ kfree(object_prefix);
- return -ENOMEM;
+ return ret;
}
static const char *_rbd_dev_v1_snap_name(struct rbd_device *rbd_dev, u32 which)
rbd_dev->mapping.size = size;
rbd_dev->mapping.features = features;
- /* If we are mapping a snapshot it must be marked read-only */
-
- if (snap_id != CEPH_NOSNAP)
- rbd_dev->mapping.read_only = true;
-
return 0;
}
{
rbd_dev->mapping.size = 0;
rbd_dev->mapping.features = 0;
- rbd_dev->mapping.read_only = true;
}
static const char *rbd_segment_name(struct rbd_device *rbd_dev, u64 offset)
kref_put(&obj_request->kref, rbd_obj_request_destroy);
}
-static void rbd_img_request_get(struct rbd_img_request *img_request)
-{
- dout("%s: img %p (was %d)\n", __func__, img_request,
- atomic_read(&img_request->kref.refcount));
- kref_get(&img_request->kref);
-}
-
+static bool img_request_child_test(struct rbd_img_request *img_request);
+static void rbd_parent_request_destroy(struct kref *kref);
static void rbd_img_request_destroy(struct kref *kref);
static void rbd_img_request_put(struct rbd_img_request *img_request)
{
rbd_assert(img_request != NULL);
dout("%s: img %p (was %d)\n", __func__, img_request,
atomic_read(&img_request->kref.refcount));
- kref_put(&img_request->kref, rbd_img_request_destroy);
+ if (img_request_child_test(img_request))
+ kref_put(&img_request->kref, rbd_parent_request_destroy);
+ else
+ kref_put(&img_request->kref, rbd_img_request_destroy);
}
static inline void rbd_img_obj_request_add(struct rbd_img_request *img_request,
smp_mb();
}
+static void img_request_child_clear(struct rbd_img_request *img_request)
+{
+ clear_bit(IMG_REQ_CHILD, &img_request->flags);
+ smp_mb();
+}
+
static bool img_request_child_test(struct rbd_img_request *img_request)
{
smp_mb();
kmem_cache_free(rbd_obj_request_cache, obj_request);
}
+/* It's OK to call this for a device with no parent */
+
+static void rbd_spec_put(struct rbd_spec *spec);
+static void rbd_dev_unparent(struct rbd_device *rbd_dev)
+{
+ rbd_dev_remove_parent(rbd_dev);
+ rbd_spec_put(rbd_dev->parent_spec);
+ rbd_dev->parent_spec = NULL;
+ rbd_dev->parent_overlap = 0;
+}
+
/*
* Caller is responsible for filling in the list of object requests
* that comprises the image request, and the Linux request pointer
static struct rbd_img_request *rbd_img_request_create(
struct rbd_device *rbd_dev,
u64 offset, u64 length,
- bool write_request,
- bool child_request)
+ bool write_request)
{
struct rbd_img_request *img_request;
} else {
img_request->snap_id = rbd_dev->spec->snap_id;
}
- if (child_request)
- img_request_child_set(img_request);
- if (rbd_dev->parent_spec)
+ if (rbd_dev->parent_overlap)
img_request_layered_set(img_request);
spin_lock_init(&img_request->completion_lock);
img_request->next_completion = 0;
INIT_LIST_HEAD(&img_request->obj_requests);
kref_init(&img_request->kref);
- rbd_img_request_get(img_request); /* Avoid a warning */
- rbd_img_request_put(img_request); /* TEMPORARY */
-
dout("%s: rbd_dev %p %s %llu/%llu -> img %p\n", __func__, rbd_dev,
write_request ? "write" : "read", offset, length,
img_request);
if (img_request_write_test(img_request))
ceph_put_snap_context(img_request->snapc);
- if (img_request_child_test(img_request))
- rbd_obj_request_put(img_request->obj_request);
-
kmem_cache_free(rbd_img_request_cache, img_request);
}
+static struct rbd_img_request *rbd_parent_request_create(
+ struct rbd_obj_request *obj_request,
+ u64 img_offset, u64 length)
+{
+ struct rbd_img_request *parent_request;
+ struct rbd_device *rbd_dev;
+
+ rbd_assert(obj_request->img_request);
+ rbd_dev = obj_request->img_request->rbd_dev;
+
+ parent_request = rbd_img_request_create(rbd_dev->parent,
+ img_offset, length, false);
+ if (!parent_request)
+ return NULL;
+
+ img_request_child_set(parent_request);
+ rbd_obj_request_get(obj_request);
+ parent_request->obj_request = obj_request;
+
+ return parent_request;
+}
+
+static void rbd_parent_request_destroy(struct kref *kref)
+{
+ struct rbd_img_request *parent_request;
+ struct rbd_obj_request *orig_request;
+
+ parent_request = container_of(kref, struct rbd_img_request, kref);
+ orig_request = parent_request->obj_request;
+
+ parent_request->obj_request = NULL;
+ rbd_obj_request_put(orig_request);
+ img_request_child_clear(parent_request);
+
+ rbd_img_request_destroy(kref);
+}
+
static bool rbd_img_obj_end_request(struct rbd_obj_request *obj_request)
{
struct rbd_img_request *img_request;
{
struct rbd_img_request *img_request;
struct rbd_device *rbd_dev;
- u64 length;
+ struct page **pages;
u32 page_count;
rbd_assert(obj_request->type == OBJ_REQUEST_BIO);
rbd_dev = img_request->rbd_dev;
rbd_assert(rbd_dev);
- length = (u64)1 << rbd_dev->header.obj_order;
- page_count = (u32)calc_pages_for(0, length);
- rbd_assert(obj_request->copyup_pages);
- ceph_release_page_vector(obj_request->copyup_pages, page_count);
+ pages = obj_request->copyup_pages;
+ rbd_assert(pages != NULL);
obj_request->copyup_pages = NULL;
+ page_count = obj_request->copyup_page_count;
+ rbd_assert(page_count);
+ obj_request->copyup_page_count = 0;
+ ceph_release_page_vector(pages, page_count);
/*
* We want the transfer count to reflect the size of the
struct ceph_osd_client *osdc;
struct rbd_device *rbd_dev;
struct page **pages;
+ u32 page_count;
int result;
- u64 obj_size;
- u64 xferred;
+ u64 parent_length;
+ u64 offset;
+ u64 length;
rbd_assert(img_request_child_test(img_request));
pages = img_request->copyup_pages;
rbd_assert(pages != NULL);
img_request->copyup_pages = NULL;
+ page_count = img_request->copyup_page_count;
+ rbd_assert(page_count);
+ img_request->copyup_page_count = 0;
orig_request = img_request->obj_request;
rbd_assert(orig_request != NULL);
- rbd_assert(orig_request->type == OBJ_REQUEST_BIO);
+ rbd_assert(obj_request_type_valid(orig_request->type));
result = img_request->result;
- obj_size = img_request->length;
- xferred = img_request->xferred;
+ parent_length = img_request->length;
+ rbd_assert(parent_length == img_request->xferred);
+ rbd_img_request_put(img_request);
- rbd_dev = img_request->rbd_dev;
+ rbd_assert(orig_request->img_request);
+ rbd_dev = orig_request->img_request->rbd_dev;
rbd_assert(rbd_dev);
- rbd_assert(obj_size == (u64)1 << rbd_dev->header.obj_order);
-
- rbd_img_request_put(img_request);
if (result)
goto out_err;
- /* Allocate the new copyup osd request for the original request */
-
+ /*
+ * The original osd request is of no use to use any more.
+ * We need a new one that can hold the two ops in a copyup
+ * request. Allocate the new copyup osd request for the
+ * original request, and release the old one.
+ */
result = -ENOMEM;
- rbd_assert(!orig_request->osd_req);
osd_req = rbd_osd_req_create_copyup(orig_request);
if (!osd_req)
goto out_err;
+ rbd_osd_req_destroy(orig_request->osd_req);
orig_request->osd_req = osd_req;
orig_request->copyup_pages = pages;
+ orig_request->copyup_page_count = page_count;
/* Initialize the copyup op */
osd_req_op_cls_init(osd_req, 0, CEPH_OSD_OP_CALL, "rbd", "copyup");
- osd_req_op_cls_request_data_pages(osd_req, 0, pages, obj_size, 0,
+ osd_req_op_cls_request_data_pages(osd_req, 0, pages, parent_length, 0,
false, false);
/* Then the original write request op */
+ offset = orig_request->offset;
+ length = orig_request->length;
osd_req_op_extent_init(osd_req, 1, CEPH_OSD_OP_WRITE,
- orig_request->offset,
- orig_request->length, 0, 0);
- osd_req_op_extent_osd_data_bio(osd_req, 1, orig_request->bio_list,
- orig_request->length);
+ offset, length, 0, 0);
+ if (orig_request->type == OBJ_REQUEST_BIO)
+ osd_req_op_extent_osd_data_bio(osd_req, 1,
+ orig_request->bio_list, length);
+ else
+ osd_req_op_extent_osd_data_pages(osd_req, 1,
+ orig_request->pages, length,
+ offset & ~PAGE_MASK, false, false);
rbd_osd_req_format_write(orig_request);
int result;
rbd_assert(obj_request_img_data_test(obj_request));
- rbd_assert(obj_request->type == OBJ_REQUEST_BIO);
+ rbd_assert(obj_request_type_valid(obj_request->type));
img_request = obj_request->img_request;
rbd_assert(img_request != NULL);
rbd_dev = img_request->rbd_dev;
rbd_assert(rbd_dev->parent != NULL);
- /*
- * First things first. The original osd request is of no
- * use to use any more, we'll need a new one that can hold
- * the two ops in a copyup request. We'll get that later,
- * but for now we can release the old one.
- */
- rbd_osd_req_destroy(obj_request->osd_req);
- obj_request->osd_req = NULL;
-
/*
* Determine the byte range covered by the object in the
* child image to which the original request was to be sent.
}
result = -ENOMEM;
- parent_request = rbd_img_request_create(rbd_dev->parent,
- img_offset, length,
- false, true);
+ parent_request = rbd_parent_request_create(obj_request,
+ img_offset, length);
if (!parent_request)
goto out_err;
- rbd_obj_request_get(obj_request);
- parent_request->obj_request = obj_request;
result = rbd_img_request_fill(parent_request, OBJ_REQUEST_PAGES, pages);
if (result)
goto out_err;
parent_request->copyup_pages = pages;
+ parent_request->copyup_page_count = page_count;
parent_request->callback = rbd_img_obj_parent_read_full_callback;
result = rbd_img_request_submit(parent_request);
return 0;
parent_request->copyup_pages = NULL;
+ parent_request->copyup_page_count = 0;
parent_request->obj_request = NULL;
rbd_obj_request_put(obj_request);
out_err:
static void rbd_img_parent_read(struct rbd_obj_request *obj_request)
{
- struct rbd_device *rbd_dev;
struct rbd_img_request *img_request;
int result;
rbd_assert(obj_request_img_data_test(obj_request));
rbd_assert(obj_request->img_request != NULL);
rbd_assert(obj_request->result == (s32) -ENOENT);
- rbd_assert(obj_request->type == OBJ_REQUEST_BIO);
+ rbd_assert(obj_request_type_valid(obj_request->type));
- rbd_dev = obj_request->img_request->rbd_dev;
- rbd_assert(rbd_dev->parent != NULL);
/* rbd_read_finish(obj_request, obj_request->length); */
- img_request = rbd_img_request_create(rbd_dev->parent,
+ img_request = rbd_parent_request_create(obj_request,
obj_request->img_offset,
- obj_request->length,
- false, true);
+ obj_request->length);
result = -ENOMEM;
if (!img_request)
goto out_err;
- rbd_obj_request_get(obj_request);
- img_request->obj_request = obj_request;
-
- result = rbd_img_request_fill(img_request, OBJ_REQUEST_BIO,
- obj_request->bio_list);
+ if (obj_request->type == OBJ_REQUEST_BIO)
+ result = rbd_img_request_fill(img_request, OBJ_REQUEST_BIO,
+ obj_request->bio_list);
+ else
+ result = rbd_img_request_fill(img_request, OBJ_REQUEST_PAGES,
+ obj_request->pages);
if (result)
goto out_err;
* Request sync osd watch/unwatch. The value of "start" determines
* whether a watch request is being initiated or torn down.
*/
-static int rbd_dev_header_watch_sync(struct rbd_device *rbd_dev, int start)
+static int rbd_dev_header_watch_sync(struct rbd_device *rbd_dev, bool start)
{
struct ceph_osd_client *osdc = &rbd_dev->rbd_client->client->osdc;
struct rbd_obj_request *obj_request;
rbd_dev->watch_request->osd_req);
osd_req_op_watch_init(obj_request->osd_req, 0, CEPH_OSD_OP_WATCH,
- rbd_dev->watch_event->cookie, 0, start);
+ rbd_dev->watch_event->cookie, 0, start ? 1 : 0);
rbd_osd_req_format_write(obj_request);
ret = rbd_obj_request_submit(osdc, obj_request);
result = -ENOMEM;
img_request = rbd_img_request_create(rbd_dev, offset, length,
- write_request, false);
+ write_request);
if (!img_request)
goto end_request;
}
/*
- * Read the complete header for the given rbd device.
- *
- * Returns a pointer to a dynamically-allocated buffer containing
- * the complete and validated header. Caller can pass the address
- * of a variable that will be filled in with the version of the
- * header object at the time it was read.
- *
- * Returns a pointer-coded errno if a failure occurs.
+ * Read the complete header for the given rbd device. On successful
+ * return, the rbd_dev->header field will contain up-to-date
+ * information about the image.
*/
-static struct rbd_image_header_ondisk *
-rbd_dev_v1_header_read(struct rbd_device *rbd_dev)
+static int rbd_dev_v1_header_info(struct rbd_device *rbd_dev)
{
struct rbd_image_header_ondisk *ondisk = NULL;
u32 snap_count = 0;
size += names_size;
ondisk = kmalloc(size, GFP_KERNEL);
if (!ondisk)
- return ERR_PTR(-ENOMEM);
+ return -ENOMEM;
ret = rbd_obj_read_sync(rbd_dev, rbd_dev->header_name,
0, size, ondisk);
if (ret < 0)
- goto out_err;
+ goto out;
if ((size_t)ret < size) {
ret = -ENXIO;
rbd_warn(rbd_dev, "short header read (want %zd got %d)",
size, ret);
- goto out_err;
+ goto out;
}
if (!rbd_dev_ondisk_valid(ondisk)) {
ret = -ENXIO;
rbd_warn(rbd_dev, "invalid header");
- goto out_err;
+ goto out;
}
names_size = le64_to_cpu(ondisk->snap_names_len);
snap_count = le32_to_cpu(ondisk->snap_count);
} while (snap_count != want_count);
- return ondisk;
-
-out_err:
- kfree(ondisk);
-
- return ERR_PTR(ret);
-}
-
-/*
- * reload the ondisk the header
- */
-static int rbd_read_header(struct rbd_device *rbd_dev,
- struct rbd_image_header *header)
-{
- struct rbd_image_header_ondisk *ondisk;
- int ret;
-
- ondisk = rbd_dev_v1_header_read(rbd_dev);
- if (IS_ERR(ondisk))
- return PTR_ERR(ondisk);
- ret = rbd_header_from_disk(header, ondisk);
+ ret = rbd_header_from_disk(rbd_dev, ondisk);
+out:
kfree(ondisk);
return ret;
}
-/*
- * only read the first part of the ondisk header, without the snaps info
- */
-static int rbd_dev_v1_refresh(struct rbd_device *rbd_dev)
-{
- int ret;
- struct rbd_image_header h;
-
- ret = rbd_read_header(rbd_dev, &h);
- if (ret < 0)
- return ret;
-
- down_write(&rbd_dev->header_rwsem);
-
- /* Update image size, and check for resize of mapped image */
- rbd_dev->header.image_size = h.image_size;
- if (rbd_dev->spec->snap_id == CEPH_NOSNAP)
- if (rbd_dev->mapping.size != rbd_dev->header.image_size)
- rbd_dev->mapping.size = rbd_dev->header.image_size;
-
- /* rbd_dev->header.object_prefix shouldn't change */
- kfree(rbd_dev->header.snap_sizes);
- kfree(rbd_dev->header.snap_names);
- /* osd requests may still refer to snapc */
- ceph_put_snap_context(rbd_dev->header.snapc);
-
- rbd_dev->header.image_size = h.image_size;
- rbd_dev->header.snapc = h.snapc;
- rbd_dev->header.snap_names = h.snap_names;
- rbd_dev->header.snap_sizes = h.snap_sizes;
- /* Free the extra copy of the object prefix */
- if (strcmp(rbd_dev->header.object_prefix, h.object_prefix))
- rbd_warn(rbd_dev, "object prefix changed (ignoring)");
- kfree(h.object_prefix);
-
- up_write(&rbd_dev->header_rwsem);
-
- return ret;
-}
-
/*
* Clear the rbd device's EXISTS flag if the snapshot it's mapped to
* has disappeared from the (just updated) snapshot context.
mapping_size = rbd_dev->mapping.size;
mutex_lock_nested(&ctl_mutex, SINGLE_DEPTH_NESTING);
if (rbd_dev->image_format == 1)
- ret = rbd_dev_v1_refresh(rbd_dev);
+ ret = rbd_dev_v1_header_info(rbd_dev);
else
- ret = rbd_dev_v2_refresh(rbd_dev);
+ ret = rbd_dev_v2_header_info(rbd_dev);
/* If it's a mapped snapshot, validate its EXISTS flag */
__le64 snapid;
void *p;
void *end;
+ u64 pool_id;
char *image_id;
u64 overlap;
int ret;
p = reply_buf;
end = reply_buf + ret;
ret = -ERANGE;
- ceph_decode_64_safe(&p, end, parent_spec->pool_id, out_err);
- if (parent_spec->pool_id == CEPH_NOPOOL)
+ ceph_decode_64_safe(&p, end, pool_id, out_err);
+ if (pool_id == CEPH_NOPOOL)
goto out; /* No parent? No problem. */
/* The ceph file layout needs to fit pool id in 32 bits */
ret = -EIO;
- if (parent_spec->pool_id > (u64)U32_MAX) {
+ if (pool_id > (u64)U32_MAX) {
rbd_warn(NULL, "parent pool id too large (%llu > %u)\n",
- (unsigned long long)parent_spec->pool_id, U32_MAX);
+ (unsigned long long)pool_id, U32_MAX);
goto out_err;
}
+ parent_spec->pool_id = pool_id;
image_id = ceph_extract_encoded_string(&p, end, NULL, GFP_KERNEL);
if (IS_ERR(image_id)) {
ceph_decode_64_safe(&p, end, parent_spec->snap_id, out_err);
ceph_decode_64_safe(&p, end, overlap, out_err);
- rbd_dev->parent_overlap = overlap;
- rbd_dev->parent_spec = parent_spec;
- parent_spec = NULL; /* rbd_dev now owns this */
+ if (overlap) {
+ rbd_spec_put(rbd_dev->parent_spec);
+ rbd_dev->parent_spec = parent_spec;
+ parent_spec = NULL; /* rbd_dev now owns this */
+ rbd_dev->parent_overlap = overlap;
+ } else {
+ rbd_warn(rbd_dev, "ignoring parent of clone with overlap 0\n");
+ }
out:
ret = 0;
out_err:
return snap_name;
}
-static int rbd_dev_v2_refresh(struct rbd_device *rbd_dev)
+static int rbd_dev_v2_header_info(struct rbd_device *rbd_dev)
{
+ bool first_time = rbd_dev->header.object_prefix == NULL;
int ret;
down_write(&rbd_dev->header_rwsem);
+ if (first_time) {
+ ret = rbd_dev_v2_header_onetime(rbd_dev);
+ if (ret)
+ goto out;
+ }
+
+ /*
+ * If the image supports layering, get the parent info. We
+ * need to probe the first time regardless. Thereafter we
+ * only need to if there's a parent, to see if it has
+ * disappeared due to the mapped image getting flattened.
+ */
+ if (rbd_dev->header.features & RBD_FEATURE_LAYERING &&
+ (first_time || rbd_dev->parent_spec)) {
+ bool warn;
+
+ ret = rbd_dev_v2_parent_info(rbd_dev);
+ if (ret)
+ goto out;
+
+ /*
+ * Print a warning if this is the initial probe and
+ * the image has a parent. Don't print it if the
+ * image now being probed is itself a parent. We
+ * can tell at this point because we won't know its
+ * pool name yet (just its pool id).
+ */
+ warn = rbd_dev->parent_spec && rbd_dev->spec->pool_name;
+ if (first_time && warn)
+ rbd_warn(rbd_dev, "WARNING: kernel layering "
+ "is EXPERIMENTAL!");
+ }
+
ret = rbd_dev_v2_image_size(rbd_dev);
if (ret)
goto out;
+
if (rbd_dev->spec->snap_id == CEPH_NOSNAP)
if (rbd_dev->mapping.size != rbd_dev->header.image_size)
rbd_dev->mapping.size = rbd_dev->header.image_size;
ret = rbd_dev_v2_snap_context(rbd_dev);
dout("rbd_dev_v2_snap_context returned %d\n", ret);
- if (ret)
- goto out;
out:
up_write(&rbd_dev->header_rwsem);
{
struct rbd_image_header *header;
- rbd_dev_remove_parent(rbd_dev);
- rbd_spec_put(rbd_dev->parent_spec);
- rbd_dev->parent_spec = NULL;
- rbd_dev->parent_overlap = 0;
+ rbd_dev_unparent(rbd_dev);
/* Free dynamic fields from the header, then zero it out */
memset(header, 0, sizeof (*header));
}
-static int rbd_dev_v1_probe(struct rbd_device *rbd_dev)
+static int rbd_dev_v2_header_onetime(struct rbd_device *rbd_dev)
{
int ret;
- /* Populate rbd image metadata */
-
- ret = rbd_read_header(rbd_dev, &rbd_dev->header);
- if (ret < 0)
- goto out_err;
-
- /* Version 1 images have no parent (no layering) */
-
- rbd_dev->parent_spec = NULL;
- rbd_dev->parent_overlap = 0;
-
- dout("discovered version 1 image, header name is %s\n",
- rbd_dev->header_name);
-
- return 0;
-
-out_err:
- kfree(rbd_dev->header_name);
- rbd_dev->header_name = NULL;
- kfree(rbd_dev->spec->image_id);
- rbd_dev->spec->image_id = NULL;
-
- return ret;
-}
-
-static int rbd_dev_v2_probe(struct rbd_device *rbd_dev)
-{
- int ret;
-
- ret = rbd_dev_v2_image_size(rbd_dev);
- if (ret)
- goto out_err;
-
- /* Get the object prefix (a.k.a. block_name) for the image */
-
ret = rbd_dev_v2_object_prefix(rbd_dev);
if (ret)
goto out_err;
- /* Get the and check features for the image */
-
+ /*
+ * Get the and check features for the image. Currently the
+ * features are assumed to never change.
+ */
ret = rbd_dev_v2_features(rbd_dev);
if (ret)
goto out_err;
- /* If the image supports layering, get the parent info */
-
- if (rbd_dev->header.features & RBD_FEATURE_LAYERING) {
- ret = rbd_dev_v2_parent_info(rbd_dev);
- if (ret)
- goto out_err;
- /*
- * Print a warning if this image has a parent.
- * Don't print it if the image now being probed
- * is itself a parent. We can tell at this point
- * because we won't know its pool name yet (just its
- * pool id).
- */
- if (rbd_dev->parent_spec && rbd_dev->spec->pool_name)
- rbd_warn(rbd_dev, "WARNING: kernel layering "
- "is EXPERIMENTAL!");
- }
-
/* If the image supports fancy striping, get its parameters */
if (rbd_dev->header.features & RBD_FEATURE_STRIPINGV2) {
if (ret < 0)
goto out_err;
}
-
- /* crypto and compression type aren't (yet) supported for v2 images */
-
- rbd_dev->header.crypt_type = 0;
- rbd_dev->header.comp_type = 0;
-
- /* Get the snapshot context, plus the header version */
-
- ret = rbd_dev_v2_snap_context(rbd_dev);
- if (ret)
- goto out_err;
-
- dout("discovered version 2 image, header name is %s\n",
- rbd_dev->header_name);
+ /* No support for crypto and compression type format 2 images */
return 0;
out_err:
- rbd_dev->parent_overlap = 0;
- rbd_spec_put(rbd_dev->parent_spec);
- rbd_dev->parent_spec = NULL;
- kfree(rbd_dev->header_name);
- rbd_dev->header_name = NULL;
+ rbd_dev->header.features = 0;
kfree(rbd_dev->header.object_prefix);
rbd_dev->header.object_prefix = NULL;
if (!parent)
goto out_err;
- ret = rbd_dev_image_probe(parent);
+ ret = rbd_dev_image_probe(parent, false);
if (ret < 0)
goto out_err;
rbd_dev->parent = parent;
return 0;
out_err:
if (parent) {
- rbd_spec_put(rbd_dev->parent_spec);
+ rbd_dev_unparent(rbd_dev);
kfree(rbd_dev->header_name);
rbd_dev_destroy(parent);
} else {
{
int ret;
- ret = rbd_dev_mapping_set(rbd_dev);
- if (ret)
- return ret;
-
/* generate unique id: find highest unique id, add one */
rbd_dev_id_get(rbd_dev);
if (ret)
goto err_out_blkdev;
- ret = rbd_bus_add_dev(rbd_dev);
+ ret = rbd_dev_mapping_set(rbd_dev);
if (ret)
goto err_out_disk;
+ set_capacity(rbd_dev->disk, rbd_dev->mapping.size / SECTOR_SIZE);
+
+ ret = rbd_bus_add_dev(rbd_dev);
+ if (ret)
+ goto err_out_mapping;
/* Everything's ready. Announce the disk to the world. */
- set_capacity(rbd_dev->disk, rbd_dev->mapping.size / SECTOR_SIZE);
set_bit(RBD_DEV_FLAG_EXISTS, &rbd_dev->flags);
add_disk(rbd_dev->disk);
return ret;
+err_out_mapping:
+ rbd_dev_mapping_clear(rbd_dev);
err_out_disk:
rbd_free_disk(rbd_dev);
err_out_blkdev:
static void rbd_dev_image_release(struct rbd_device *rbd_dev)
{
- int ret;
-
rbd_dev_unprobe(rbd_dev);
- ret = rbd_dev_header_watch_sync(rbd_dev, 0);
- if (ret)
- rbd_warn(rbd_dev, "failed to cancel watch event (%d)\n", ret);
kfree(rbd_dev->header_name);
rbd_dev->header_name = NULL;
rbd_dev->image_format = 0;
/*
* Probe for the existence of the header object for the given rbd
- * device. For format 2 images this includes determining the image
- * id.
+ * device. If this image is the one being mapped (i.e., not a
+ * parent), initiate a watch on its header object before using that
+ * object to get detailed information about the rbd image.
*/
-static int rbd_dev_image_probe(struct rbd_device *rbd_dev)
+static int rbd_dev_image_probe(struct rbd_device *rbd_dev, bool mapping)
{
int ret;
int tmp;
if (ret)
goto err_out_format;
- ret = rbd_dev_header_watch_sync(rbd_dev, 1);
- if (ret)
- goto out_header_name;
+ if (mapping) {
+ ret = rbd_dev_header_watch_sync(rbd_dev, true);
+ if (ret)
+ goto out_header_name;
+ }
if (rbd_dev->image_format == 1)
- ret = rbd_dev_v1_probe(rbd_dev);
+ ret = rbd_dev_v1_header_info(rbd_dev);
else
- ret = rbd_dev_v2_probe(rbd_dev);
+ ret = rbd_dev_v2_header_info(rbd_dev);
if (ret)
goto err_out_watch;
goto err_out_probe;
ret = rbd_dev_probe_parent(rbd_dev);
- if (!ret)
- return 0;
+ if (ret)
+ goto err_out_probe;
+
+ dout("discovered format %u image, header name is %s\n",
+ rbd_dev->image_format, rbd_dev->header_name);
+ return 0;
err_out_probe:
rbd_dev_unprobe(rbd_dev);
err_out_watch:
- tmp = rbd_dev_header_watch_sync(rbd_dev, 0);
- if (tmp)
- rbd_warn(rbd_dev, "unable to tear down watch request\n");
+ if (mapping) {
+ tmp = rbd_dev_header_watch_sync(rbd_dev, false);
+ if (tmp)
+ rbd_warn(rbd_dev, "unable to tear down "
+ "watch request (%d)\n", tmp);
+ }
out_header_name:
kfree(rbd_dev->header_name);
rbd_dev->header_name = NULL;
struct rbd_spec *spec = NULL;
struct rbd_client *rbdc;
struct ceph_osd_client *osdc;
+ bool read_only;
int rc = -ENOMEM;
if (!try_module_get(THIS_MODULE))
rc = rbd_add_parse_args(buf, &ceph_opts, &rbd_opts, &spec);
if (rc < 0)
goto err_out_module;
+ read_only = rbd_opts->read_only;
+ kfree(rbd_opts);
+ rbd_opts = NULL; /* done with this */
rbdc = rbd_get_client(ceph_opts);
if (IS_ERR(rbdc)) {
rbdc = NULL; /* rbd_dev now owns this */
spec = NULL; /* rbd_dev now owns this */
- rbd_dev->mapping.read_only = rbd_opts->read_only;
- kfree(rbd_opts);
- rbd_opts = NULL; /* done with this */
-
- rc = rbd_dev_image_probe(rbd_dev);
+ rc = rbd_dev_image_probe(rbd_dev, true);
if (rc < 0)
goto err_out_rbd_dev;
+ /* If we are mapping a snapshot it must be marked read-only */
+
+ if (rbd_dev->spec->snap_id != CEPH_NOSNAP)
+ read_only = true;
+ rbd_dev->mapping.read_only = read_only;
+
rc = rbd_dev_device_setup(rbd_dev);
if (!rc)
return count;
spin_unlock_irq(&rbd_dev->lock);
if (ret < 0)
goto done;
- ret = count;
rbd_bus_del_dev(rbd_dev);
+ ret = rbd_dev_header_watch_sync(rbd_dev, false);
+ if (ret)
+ rbd_warn(rbd_dev, "failed to cancel watch event (%d)\n", ret);
rbd_dev_image_release(rbd_dev);
module_put(THIS_MODULE);
+ ret = count;
done:
mutex_unlock(&ctl_mutex);
projects / ~andy / linux / blobdiff