Skip to contentSkip to conversation

rbd: miscellaneous cleanups

PATCH In review Alex Elder 3 participants 15 messages May 7, 2013 – May 8, 2013

Subsystems (2)
lore
Share & tools
lore

Alex Elder

author
#0 lore raw

cover rbd: miscellaneous cleanups

This is just some odds and ends, cleaning up code mostly. One might be considered a minor bug, but it's not serious.

-Alex

[PATCH 1/6] rbd: update capacity in rbd_dev_refresh() [PATCH 2/6] rbd: kill rbd_update_mapping_size()
[PATCH 3/6] rbd: don't print warning if not mapping a parent [PATCH 4/6] rbd: don't look up snapshot id in rbd_dev_mapping_set() [PATCH 5/6] rbd: kill rbd_dev_clear_mapping()
[PATCH 6/6] rbd: always set read-only flag in rbd_add()

author
#1 lore raw

PATCH 1/6 rbd: update capacity in rbd_dev_refresh()

When a mapped image changes size, we change the capacity recorded for the Linux disk associated with it, in rbd_update_mapping_size(). That function is called in two places--the format 1 and format 2 refresh routines.

There is no need to set the capacity while holding the header semaphore. Instead, do it in the common rbd_dev_refresh(), using the logic that's already there to initiate disk revalidation.

Add handling in the request function, just in case a request that exceeds the capacity of the device comes in (perhaps one that was started before a refresh shrunk the device).

---
 drivers/block/rbd.c |   23 +++++++++++++++--------
 1 file changed, 15 insertions(+), 8 deletions(-)
drivers/block/rbd.c +7 -0
index 5d5e3f0..a0f1fe5 100644
@@ -2874,6 +2874,13 @@ static void rbd_request_fn(struct request_queue *q)
 			goto end_request;	/* Shouldn't happen */
 		}
 
+		result = -ENOSPC;
+		if (offset + length > rbd_dev->mapping.size) {
+			rbd_warn(rbd_dev, "beyond EOD (%llu~%llu > %llu)\n",
+				offset, length, rbd_dev->mapping.size);
+			goto end_request;
+		}
+
 		result = -ENOMEM;
 		img_request = rbd_img_request_create(rbd_dev, offset, length,
 							write_request, false);
@@ -3116,14 +3123,8 @@ static void rbd_update_mapping_size(struct

rbd_device *rbd_dev)
if (rbd_dev->spec->snap_id != CEPH_NOSNAP)
return;

- if (rbd_dev->mapping.size != rbd_dev->header.image_size) {
- sector_t size;
-
+ if (rbd_dev->mapping.size != rbd_dev->header.image_size)
rbd_dev->mapping.size = rbd_dev->header.image_size;
- size = (sector_t)rbd_dev->mapping.size / SECTOR_SIZE;
- dout("setting size to %llu sectors", (unsigned long long)size);
- set_capacity(rbd_dev->disk, size);
- }
}

/*
@@ -3200,8 +3201,14 @@ static int rbd_dev_refresh(struct rbd_device *rbd_dev)

rbd_exists_validate(rbd_dev);
mutex_unlock(&ctl_mutex);
- if (mapping_size != rbd_dev->mapping.size)
+ if (mapping_size != rbd_dev->mapping.size) {
+ sector_t size;
+
+ size = (sector_t)rbd_dev->mapping.size / SECTOR_SIZE;
+ dout("setting size to %llu sectors", (unsigned long long)size);
+ set_capacity(rbd_dev->disk, size);
revalidate_disk(rbd_dev->disk);
+ }

return ret;
}

Signature · 1 line
1.7.9.5
author
#3 lore raw

PATCH 2/6 rbd: kill rbd_update_mapping_size()

In reply to Alex Elder

Since rbd_update_mapping_size() is now a trivial wrapper, just open code it in its two callers.

---
 drivers/block/rbd.c |   17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)
drivers/block/rbd.c +0 -0
index a0f1fe5..0a24cdf 100644
@@ -3118,15 +3118,6 @@ static int rbd_read_header(struct rbd_device

*rbd_dev,
return ret;
}

-static void rbd_update_mapping_size(struct rbd_device *rbd_dev) -{
- if (rbd_dev->spec->snap_id != CEPH_NOSNAP)
- return;
-
- if (rbd_dev->mapping.size != rbd_dev->header.image_size)
- rbd_dev->mapping.size = rbd_dev->header.image_size;
-}
-
/*
* only read the first part of the ondisk header, without the snaps info */
@@ -3143,7 +3134,9 @@ static int rbd_dev_v1_refresh(struct rbd_device *rbd_dev)

/* Update image size, and check for resize of mapped image */ rbd_dev->header.image_size = h.image_size;
- rbd_update_mapping_size(rbd_dev);
+ 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);
@@ -4074,7 +4067,9 @@ static int rbd_dev_v2_refresh(struct rbd_device *rbd_dev)
ret = rbd_dev_v2_image_size(rbd_dev);
if (ret)
goto out;
- rbd_update_mapping_size(rbd_dev);
+ 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);

Signature · 1 line
1.7.9.5
author
#6 lore raw

PATCH 3/6 rbd: don't print warning if not mapping a parent

In reply to Alex Elder

The presence of the LAYERING bit in an rbd image's feature mask does not guarantee the image actually has a parent image. Currently that bit is set only when a clone (i.e., image with a parent) is created, but it is (currently) not cleared if that clone gets flattened back into a "normal" image. A "parent_id" query will leave the parent_spec for the image being mapped a null pointer, but will not return an error.

Currently, whenever an image with the LAYERED feature gets mapped, a warning about the use of layered images gets printed. But we don't want to do this for a flattened image, so print the warning only if we find there is a parent spec after the probe.

---
 drivers/block/rbd.c |   11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)
drivers/block/rbd.c +0 -0
index 0a24cdf..7d93dbd 100644
@@ -4567,13 +4567,14 @@ static int rbd_dev_v2_probe(struct rbd_device

*rbd_dev)
ret = rbd_dev_v2_parent_info(rbd_dev);
if (ret)
goto out_err;
-
/*
- * Don't print a warning for parent images. We can
- * tell this point because we won't know its pool
- * name yet (just its pool id).
+ * 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->spec->pool_name)
+ if (rbd_dev->parent_spec && rbd_dev->spec->pool_name)
rbd_warn(rbd_dev, "WARNING: kernel layering "
"is EXPERIMENTAL!");
}

Signature · 1 line
1.7.9.5
author
#8 lore raw

PATCH 4/6 rbd: don't look up snapshot id in rbd_dev_mapping_set()

In reply to Alex Elder

Currently rbd_dev_mapping_set() looks up the snapshot id for the snapshot whose name is found in the rbd device's spec structure.

That function gets called by rbd_dev_device_setup(), which is called by rbd_add() *after* rbd_dev_image_probe(). If the image probe succeeds, the rbd device's spec will already have been updated to include names and ids for all fields.

Therefore there's no need to look up the snapshot id in rbd_dev_mapping_set().

---
 drivers/block/rbd.c |   11 +----------
 1 file changed, 1 insertion(+), 10 deletions(-)
drivers/block/rbd.c +0 -0
index 7d93dbd..5eebf6d 100644
@@ -936,20 +936,11 @@ static int rbd_snap_features(struct rbd_device

*rbd_dev, u64 snap_id,

static int rbd_dev_mapping_set(struct rbd_device *rbd_dev) {
- const char *snap_name = rbd_dev->spec->snap_name;
- u64 snap_id;
+ u64 snap_id = rbd_dev->spec->snap_id;
u64 size = 0;
u64 features = 0;
int ret;

- if (strcmp(snap_name, RBD_SNAP_HEAD_NAME)) {
- snap_id = rbd_snap_id_by_name(rbd_dev, snap_name);
- if (snap_id == CEPH_NOSNAP)
- return -ENOENT;
- } else {
- snap_id = CEPH_NOSNAP;
- }
-
ret = rbd_snap_size(rbd_dev, snap_id, &size); if (ret)
return ret;

Signature · 1 line
1.7.9.5
author
#10 lore raw

PATCH 5/6 rbd: kill rbd_dev_clear_mapping()

In reply to Alex Elder

This function is a duplicate of rbd_dev_mapping_clear(), and was added by mistake.

---
 drivers/block/rbd.c |    9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)
drivers/block/rbd.c +0 -0
index 5eebf6d..530793a 100644
@@ -966,13 +966,6 @@ static void rbd_dev_mapping_clear(struct rbd_device

*rbd_dev)
rbd_dev->mapping.read_only = true;
}

-static void rbd_dev_clear_mapping(struct rbd_device *rbd_dev) -{
- 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) {
char *name;
@@ -4910,7 +4903,7 @@ static void rbd_dev_device_release(struct device *dev)

rbd_free_disk(rbd_dev); clear_bit(RBD_DEV_FLAG_EXISTS, &rbd_dev->flags);
- rbd_dev_clear_mapping(rbd_dev);
+ rbd_dev_mapping_clear(rbd_dev);
unregister_blkdev(rbd_dev->major, rbd_dev->name); rbd_dev->major = 0; rbd_dev_id_put(rbd_dev);

Signature · 1 line
1.7.9.5
author
#12 lore raw

PATCH 6/6 rbd: always set read-only flag in rbd_add()

In reply to Alex Elder

Hold off setting the read-only flag in rbd_add() for an image being mapped until we have successfully probed the image. At that point we know whether it's a snapshot mapping or not, so we can set the read-only flag in that one place rather than doing so (for snapshots) in rbd_dev_mapping_set(). To do this, pass a flag to the image probe routine indicating whether we want a read-only mapping.

---
 drivers/block/rbd.c |   28 ++++++++++++++--------------
 1 file changed, 14 insertions(+), 14 deletions(-)
drivers/block/rbd.c +0 -0
index 530793a..0c72643 100644
@@ -359,7 +359,7 @@ static ssize_t rbd_add(struct bus_type *bus, const

char *buf,
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 read_only);

static struct bus_attribute rbd_bus_attrs[] = {
__ATTR(add, S_IWUSR, NULL, rbd_add),
@@ -951,11 +951,6 @@ static int rbd_dev_mapping_set(struct rbd_device *rbd_dev)
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;
}

@@ -963,7 +958,6 @@ static void rbd_dev_mapping_clear(struct rbd_device *rbd_dev)
{
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)
@@ -4620,7 +4614,7 @@ static int rbd_dev_probe_parent(struct rbd_device *rbd_dev)
if (!parent)
goto out_err;

- ret = rbd_dev_image_probe(parent);
+ ret = rbd_dev_image_probe(parent, true);
if (ret < 0)
goto out_err;
rbd_dev->parent = parent;
@@ -4743,7 +4737,7 @@ static void rbd_dev_image_release(struct rbd_device *rbd_dev)
* device. For format 2 images this includes determining the image
* id.
*/
-static int rbd_dev_image_probe(struct rbd_device *rbd_dev) +static int rbd_dev_image_probe(struct rbd_device *rbd_dev, bool read_only)
{
int ret;
int tmp;
@@ -4778,6 +4772,12 @@ static int rbd_dev_image_probe(struct rbd_device *rbd_dev)
if (ret)
goto err_out_probe;

+ /* 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;
+
ret = rbd_dev_probe_parent(rbd_dev);
if (!ret)
return 0;
@@ -4811,6 +4811,7 @@ static ssize_t rbd_add(struct bus_type *bus,
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))
@@ -4820,6 +4821,9 @@ static ssize_t rbd_add(struct bus_type *bus,
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)) {
@@ -4850,11 +4854,7 @@ static ssize_t rbd_add(struct bus_type *bus,
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, read_only);
if (rc < 0)
goto err_out_rbd_dev;

Signature · 1 line
1.7.9.5
Reviewed-by on v1 4/6
#9 lore raw

Re: [PATCH 4/6] rbd: don't look up snapshot id in rbd_dev_mapping_set()

In reply to Alex Elder
Quoted 42 lines · Alex Elder

Currently rbd_dev_mapping_set() looks up the snapshot id for the snapshot whose name is found in the rbd device's spec structure.

That function gets called by rbd_dev_device_setup(), which is called by rbd_add() *after* rbd_dev_image_probe(). If the image probe succeeds, the rbd device's spec will already have been updated to include names and ids for all fields.

Therefore there's no need to look up the snapshot id in rbd_dev_mapping_set().

---
drivers/block/rbd.c | 11 +----------

1 file changed, 1 insertion(+), 10 deletions(-)
drivers/block/rbd.c +1 -10
index 7d93dbd..5eebf6d 100644
@@ -936,20 +936,11 @@ static int rbd_snap_features(struct rbd_device
 *rbd_dev, u64 snap_id,
 
 static int rbd_dev_mapping_set(struct rbd_device *rbd_dev)
 {
-	const char *snap_name = rbd_dev->spec->snap_name;
-	u64 snap_id;
+	u64 snap_id = rbd_dev->spec->snap_id;
 	u64 size = 0;
 	u64 features = 0;
 	int ret;
 
-	if (strcmp(snap_name, RBD_SNAP_HEAD_NAME)) {
-		snap_id = rbd_snap_id_by_name(rbd_dev, snap_name);
-		if (snap_id == CEPH_NOSNAP)
-			return -ENOENT;
-	} else {
-		snap_id = CEPH_NOSNAP;
-	}
-
 	ret = rbd_snap_size(rbd_dev, snap_id, &size);
 	if (ret)
Quoted 1 line
		return ret;
Reviewed-by on v1 3/6
#7 lore raw

Re: [PATCH 3/6] rbd: don't print warning if not mapping a parent

In reply to Alex Elder
Quoted 42 lines · Alex Elder

The presence of the LAYERING bit in an rbd image's feature mask does not guarantee the image actually has a parent image. Currently that bit is set only when a clone (i.e., image with a parent) is created, but it is (currently) not cleared if that clone gets flattened back into a "normal" image. A "parent_id" query will leave the parent_spec for the image being mapped a null pointer, but will not return an error.

Currently, whenever an image with the LAYERED feature gets mapped, a warning about the use of layered images gets printed. But we don't want to do this for a flattened image, so print the warning only if we find there is a parent spec after the probe.

---
drivers/block/rbd.c | 11 ++++++-----

1 file changed, 6 insertions(+), 5 deletions(-)
drivers/block/rbd.c +6 -5
index 0a24cdf..7d93dbd 100644
@@ -4567,13 +4567,14 @@ static int rbd_dev_v2_probe(struct rbd_device
 *rbd_dev)
 		ret = rbd_dev_v2_parent_info(rbd_dev);
 		if (ret)
 			goto out_err;
-
 		/*
-		 * Don't print a warning for parent images.  We can
-		 * tell this point because we won't know its pool
-		 * name yet (just its pool id).
+		 * 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->spec->pool_name)
+		if (rbd_dev->parent_spec && rbd_dev->spec->pool_name)
 			rbd_warn(rbd_dev, "WARNING: kernel layering "
 					"is EXPERIMENTAL!");
Quoted 1 line
	}
Reviewed-by on v1 5/6
#11 lore raw

Re: [PATCH 5/6] rbd: kill rbd_dev_clear_mapping()

In reply to Alex Elder
Quoted 38 lines · Alex Elder

This function is a duplicate of rbd_dev_mapping_clear(), and was added by mistake.

---
drivers/block/rbd.c | 9 +--------

1 file changed, 1 insertion(+), 8 deletions(-)
drivers/block/rbd.c +0 -0
index 5eebf6d..530793a 100644
@@ -966,13 +966,6 @@ static void rbd_dev_mapping_clear(struct

rbd_device
*rbd_dev)

	rbd_dev->mapping.read_only = true;
}
drivers/block/rbd.c +0 -7
-static void rbd_dev_clear_mapping(struct rbd_device *rbd_dev)
-{
-	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)

{
	char *name;
drivers/block/rbd.c +1 -1
@@ -4910,7 +4903,7 @@ static void rbd_dev_device_release(struct device
 *dev)
 
 	rbd_free_disk(rbd_dev);
 	clear_bit(RBD_DEV_FLAG_EXISTS, &rbd_dev->flags);
-	rbd_dev_clear_mapping(rbd_dev);
+	rbd_dev_mapping_clear(rbd_dev);
 	unregister_blkdev(rbd_dev->major, rbd_dev->name);
 	rbd_dev->major = 0;
Quoted 1 line
	rbd_dev_id_put(rbd_dev);
Reviewed-by on v1 2/6
#4 lore raw

Re: [PATCH 2/6] rbd: kill rbd_update_mapping_size()

In reply to Alex Elder
Quoted 30 lines · Alex Elder

Since rbd_update_mapping_size() is now a trivial wrapper, just open code it in its two callers.

---
drivers/block/rbd.c | 17 ++++++-----------

1 file changed, 6 insertions(+), 11 deletions(-)
drivers/block/rbd.c +0 -9
index a0f1fe5..0a24cdf 100644
@@ -3118,15 +3118,6 @@ static int rbd_read_header(struct rbd_device
 *rbd_dev,
 	return ret;
 }
 
-static void rbd_update_mapping_size(struct rbd_device *rbd_dev)
-{
-	if (rbd_dev->spec->snap_id != CEPH_NOSNAP)
-		return;
-
-	if (rbd_dev->mapping.size != rbd_dev->header.image_size)
-		rbd_dev->mapping.size = rbd_dev->header.image_size;
-}
-
 /*
 * only read the first part of the ondisk header, without the snaps info

*/

drivers/block/rbd.c static int rbd_dev_v1_refresh(struct rbd_device · line 3140

Quoted 8 lines
drivers/block/rbd.c +3 -1
@@ -3143,7 +3134,9 @@ static int rbd_dev_v1_refresh(struct rbd_device
 *rbd_dev)
 
 	/* Update image size, and check for resize of mapped image */
 	rbd_dev->header.image_size = h.image_size;
-	rbd_update_mapping_size(rbd_dev);
+	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;

Using && instead of two conditions would be a bit cleaner. Maybe you have later patches that depend on this though. Open coding looks fine to me in any case.

Quoted 15 lines
drivers/block/rbd.c +3 -1
 	/* rbd_dev->header.object_prefix shouldn't change */
 	kfree(rbd_dev->header.snap_sizes);
@@ -4074,7 +4067,9 @@ static int rbd_dev_v2_refresh(struct rbd_device
 *rbd_dev)
 	ret = rbd_dev_v2_image_size(rbd_dev);
 	if (ret)
 		goto out;
-	rbd_update_mapping_size(rbd_dev);
+	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);
author
#5 lore raw

Re: [PATCH 2/6] rbd: kill rbd_update_mapping_size()

Quoted 34 lines · Josh Durgin
Quoted 32 lines · Alex Elder

Since rbd_update_mapping_size() is now a trivial wrapper, just open code it in its two callers.

---
drivers/block/rbd.c | 17 ++++++-----------

1 file changed, 6 insertions(+), 11 deletions(-)
drivers/block/rbd.c +0 -9
index a0f1fe5..0a24cdf 100644
@@ -3118,15 +3118,6 @@ static int rbd_read_header(struct rbd_device
 *rbd_dev,
    return ret;
 }
 
-static void rbd_update_mapping_size(struct rbd_device *rbd_dev)
-{
-    if (rbd_dev->spec->snap_id != CEPH_NOSNAP)
-        return;
-
-    if (rbd_dev->mapping.size != rbd_dev->header.image_size)
-        rbd_dev->mapping.size = rbd_dev->header.image_size;
-}
-
 /*
 * only read the first part of the ondisk header, without the snaps info

*/

drivers/block/rbd.c +0 -0
@@ -3143,7 +3134,9 @@ static int rbd_dev_v1_refresh(struct rbd_device
 *rbd_dev)
 
Quoted 10 lines

drivers/block/rbd.c static int rbd_dev_v1_refresh(struct rbd_device · line 3140

Quoted 6 lines
drivers/block/rbd.c +3 -1
    /* Update image size, and check for resize of mapped image */
    rbd_dev->header.image_size = h.image_size;
-    rbd_update_mapping_size(rbd_dev);
+    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;

Using && instead of two conditions would be a bit cleaner. Maybe you have later patches that depend on this though. Open coding looks fine to me in any case.

Honestly, that's the way I had it, but I looked at both and liked this better, I think because the lines ended up long or something.

I'll take another look before I commit it.

-Alex

Quoted 18 lines
Quoted 15 lines
drivers/block/rbd.c +3 -1
    /* rbd_dev->header.object_prefix shouldn't change */
    kfree(rbd_dev->header.snap_sizes);
@@ -4074,7 +4067,9 @@ static int rbd_dev_v2_refresh(struct rbd_device
 *rbd_dev)
    ret = rbd_dev_v2_image_size(rbd_dev);
    if (ret)
        goto out;
-    rbd_update_mapping_size(rbd_dev);
+    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);
Reviewed-by on v1 1/6
#2 lore raw

Re: [PATCH 1/6] rbd: update capacity in rbd_dev_refresh()

In reply to Alex Elder
Quoted 23 lines · Alex Elder

When a mapped image changes size, we change the capacity recorded for the Linux disk associated with it, in rbd_update_mapping_size(). That function is called in two places--the format 1 and format 2 refresh routines.

There is no need to set the capacity while holding the header semaphore. Instead, do it in the common rbd_dev_refresh(), using the logic that's already there to initiate disk revalidation.

Add handling in the request function, just in case a request that exceeds the capacity of the device comes in (perhaps one that was started before a refresh shrunk the device).

---
  drivers/block/rbd.c |   23 +++++++++++++++--------
  1 file changed, 15 insertions(+), 8 deletions(-)
drivers/block/rbd.c +0 -0
index 5d5e3f0..a0f1fe5 100644

drivers/block/rbd.c static void rbd_request_fn(struct request_queue *q) · line 2877

Quoted 4 lines
drivers/block/rbd.c +1 -0
@@ -2874,6 +2874,13 @@ static void rbd_request_fn(struct request_queue *q)
  			goto end_request;	/* Shouldn't happen */
  		}
 
+		result = -ENOSPC;

blk-core.c usually uses -EIO for this case. That's more expected from a read anyway, so I think -EIO would be better.

With that change:

Quoted 44 lines
drivers/block/rbd.c +7 -7
+		if (offset + length > rbd_dev->mapping.size) {
+			rbd_warn(rbd_dev, "beyond EOD (%llu~%llu > %llu)\n",
+				offset, length, rbd_dev->mapping.size);
+			goto end_request;
+		}
+
  		result = -ENOMEM;
  		img_request = rbd_img_request_create(rbd_dev, offset, length,
  							write_request, false);
@@ -3116,14 +3123,8 @@ static void rbd_update_mapping_size(struct
 rbd_device *rbd_dev)
  	if (rbd_dev->spec->snap_id != CEPH_NOSNAP)
  		return;
 
-	if (rbd_dev->mapping.size != rbd_dev->header.image_size) {
-		sector_t size;
-
+	if (rbd_dev->mapping.size != rbd_dev->header.image_size)
  		rbd_dev->mapping.size = rbd_dev->header.image_size;
-		size = (sector_t)rbd_dev->mapping.size / SECTOR_SIZE;
-		dout("setting size to %llu sectors", (unsigned long long)size);
-		set_capacity(rbd_dev->disk, size);
-	}
  }
 

/*

drivers/block/rbd.c +7 -1
@@ -3200,8 +3201,14 @@ static int rbd_dev_refresh(struct rbd_device
 *rbd_dev)
 
  	rbd_exists_validate(rbd_dev);
  	mutex_unlock(&ctl_mutex);
-	if (mapping_size != rbd_dev->mapping.size)
+	if (mapping_size != rbd_dev->mapping.size) {
+		sector_t size;
+
+		size = (sector_t)rbd_dev->mapping.size / SECTOR_SIZE;
+		dout("setting size to %llu sectors", (unsigned long long)size);
+		set_capacity(rbd_dev->disk, size);
  		revalidate_disk(rbd_dev->disk);
+	}
 
  	return ret;
  }
Reviewed-by on v1 6/6
#13 lore raw

Re: [PATCH 6/6] rbd: always set read-only flag in rbd_add()

In reply to Alex Elder
Quoted 101 lines · Alex Elder

Hold off setting the read-only flag in rbd_add() for an image being mapped until we have successfully probed the image. At that point we know whether it's a snapshot mapping or not, so we can set the read-only flag in that one place rather than doing so (for snapshots) in rbd_dev_mapping_set(). To do this, pass a flag to the image probe routine indicating whether we want a read-only mapping.

---
  drivers/block/rbd.c |   28 ++++++++++++++--------------
  1 file changed, 14 insertions(+), 14 deletions(-)
drivers/block/rbd.c +1 -1
index 530793a..0c72643 100644
@@ -359,7 +359,7 @@ static ssize_t rbd_add(struct bus_type *bus, const
 char *buf,
  		       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 read_only);
 
  static struct bus_attribute rbd_bus_attrs[] = {

__ATTR(add, S_IWUSR, NULL, rbd_add),

drivers/block/rbd.c +0 -6
@@ -951,11 +951,6 @@ static int rbd_dev_mapping_set(struct rbd_device
 *rbd_dev)
  	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;
  }
@@ -963,7 +958,6 @@ static void rbd_dev_mapping_clear(struct rbd_device
 *rbd_dev)
  {
  	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)

drivers/block/rbd.c +1 -1
@@ -4620,7 +4614,7 @@ static int rbd_dev_probe_parent(struct rbd_device
 *rbd_dev)
  	if (!parent)
  		goto out_err;
 
-	ret = rbd_dev_image_probe(parent);
+	ret = rbd_dev_image_probe(parent, true);
  	if (ret < 0)
  		goto out_err;
  	rbd_dev->parent = parent;
drivers/block/rbd.c +1 -1
@@ -4743,7 +4737,7 @@ static void rbd_dev_image_release(struct
 rbd_device *rbd_dev)
   * device.  For format 2 images this includes determining the image
   * id.
   */
-static int rbd_dev_image_probe(struct rbd_device *rbd_dev)
+static int rbd_dev_image_probe(struct rbd_device *rbd_dev, bool read_only)
  {
  	int ret;

int tmp;

drivers/block/rbd.c +6 -0
@@ -4778,6 +4772,12 @@ static int rbd_dev_image_probe(struct rbd_device
 *rbd_dev)
  	if (ret)
  		goto err_out_probe;
 
+	/* 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;
+
  	ret = rbd_dev_probe_parent(rbd_dev);
  	if (!ret)

return 0;

drivers/block/rbd.c +4 -0
@@ -4811,6 +4811,7 @@ static ssize_t rbd_add(struct bus_type *bus,
  	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))
@@ -4820,6 +4821,9 @@ static ssize_t rbd_add(struct bus_type *bus,
  	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)) {

drivers/block/rbd.c static ssize_t rbd_add(struct bus_type *bus, · line 4855

Quoted 6 lines
drivers/block/rbd.c +0 -3
@@ -4850,11 +4854,7 @@ static ssize_t rbd_add(struct bus_type *bus,
  	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 */

It looks like this was moved accidentally? Maybe needed by a later patch?

Quoted 6 lines
drivers/block/rbd.c +1 -2
-
-	rc = rbd_dev_image_probe(rbd_dev);
+	rc = rbd_dev_image_probe(rbd_dev, read_only);
  	if (rc < 0)
  		goto err_out_rbd_dev;
 
author
#14 lore raw

Re: [PATCH 6/6] rbd: always set read-only flag in rbd_add()

Quoted 13 lines · Josh Durgin
Quoted 12 lines · Alex Elder

Hold off setting the read-only flag in rbd_add() for an image being mapped until we have successfully probed the image. At that point we know whether it's a snapshot mapping or not, so we can set the read-only flag in that one place rather than doing so (for snapshots) in rbd_dev_mapping_set(). To do this, pass a flag to the image probe routine indicating whether we want a read-only mapping.

---
  drivers/block/rbd.c |   28 ++++++++++++++--------------
  1 file changed, 14 insertions(+), 14 deletions(-)
Quoted 4 lines

drivers/block/rbd.c

Quoted 4 lines
drivers/block/rbd.c +0 -0
index 530793a..0c72643 100644

. . .

Quoted 10 lines

drivers/block/rbd.c static ssize_t rbd_add(struct bus_type *bus, · line 4855

Quoted 7 lines
drivers/block/rbd.c +0 -3
@@ -4850,11 +4854,7 @@ static ssize_t rbd_add(struct bus_type *bus,
      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 */

It looks like this was moved accidentally? Maybe needed by a later patch?

No, I changed to grabbing the read-only flag in a local variable, and as long as I was doing that I moved it up right after argument parsing so I could free the options structure right away.

-Alex

Quoted 9 lines
Quoted 6 lines
drivers/block/rbd.c +1 -2
-
-    rc = rbd_dev_image_probe(rbd_dev);
+    rc = rbd_dev_image_probe(rbd_dev, read_only);
      if (rc < 0)
          goto err_out_rbd_dev;