* [PATCH] rbd: Use RBD fast-diff for querying actual allocation
@ 2020-06-09 7:30 Yi Li
2020-06-09 7:54 ` 李义
2020-06-09 12:20 ` Jason Dillaman
0 siblings, 2 replies; 7+ messages in thread
From: Yi Li @ 2020-06-09 7:30 UTC (permalink / raw)
To: qemu-devel; +Cc: yilikernel, yili
Since Ceph version Infernalis (9.2.0) the new fast-diff mechanism
of RBD allows for querying actual rbd image usage.
Prior to this version there was no easy and fast way to query how
much allocation a RBD image had inside a Ceph cluster.
To use the fast-diff feature it needs to be enabled per RBD image
and is only supported by Ceph cluster running version Infernalis
(9.2.0) or newer.
Without the fast-diff feature enabled qemu-img will report an allocation
identical to the image capacity.
'qemu-img info rbd:cepharm/liyi-rbd' might output for example:
image: json:{"driver": "raw", "file": {"pool": "cepharm",
"image": "liyi-rbd", "driver": "rbd"}}
file format: raw
virtual size: 20 GiB (21474836480 bytes)
disk size: 0 B
cluster_size: 4194304
Newly created rbds will have the fast-diff feature enabled.
Signed-off-by: Yi Li <yili@winhong.com>
---
block/rbd.c | 60 +++++++++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 60 insertions(+)
diff --git a/block/rbd.c b/block/rbd.c
index 617553b022..f231653f7b 100644
--- a/block/rbd.c
+++ b/block/rbd.c
@@ -1107,6 +1107,65 @@ static int64_t qemu_rbd_getlength(BlockDriverState *bs)
return info.size;
}
+#if LIBRBD_VERSION_CODE > 265
+static int disk_usage_callback(uint64_t offset, size_t len, int exists,
+ void *arg)
+{
+ uint64_t *used_size = (uint64_t *)(arg);
+ if (exists) {
+ (*used_size) += len;
+ }
+ return 0;
+}
+#endif
+
+static int64_t qemu_rbd_allocated_file_size(BlockDriverState *bs)
+{
+ BDRVRBDState *s = bs->opaque;
+ rbd_image_info_t info;
+ int r;
+ uint64_t used_size = 0;
+ uint64_t features = 0;
+
+ r = rbd_stat(s->image, &info, sizeof(info));
+ if (r < 0) {
+ return r;
+ }
+
+ r = rbd_get_features(s->image, &features);
+ if (r < 0) {
+ return r;
+ }
+
+ /*
+ * rbd_diff_iterate2() is available in versions above Ceph 0.94 (Hammer)
+ * It uses a object map inside Ceph which is faster than rbd_diff_iterate()
+ * which iterates all objects.
+ * LIBRBD_VERSION_CODE for Ceph 0.94 is 265. In 266 and upwards diff_iterate2
+ * is available
+ */
+#if LIBRBD_VERSION_CODE > 265
+ if (features & RBD_FEATURE_FAST_DIFF) {
+
+ /*
+ * RBD image fast-diff feature enabled
+ * Querying for actual allocation.
+ */
+ r = rbd_diff_iterate2(s->image, NULL, 0, info.size, 0, 1,
+ &disk_usage_callback,
+ &used_size);
+ if (r < 0) {
+ return r;
+ }
+ } else {
+ used_size = info.size;
+ }
+#else
+ used_size = info.size;
+#endif
+ return used_size;
+}
+
static int coroutine_fn qemu_rbd_co_truncate(BlockDriverState *bs,
int64_t offset,
bool exact,
@@ -1316,6 +1375,7 @@ static BlockDriver bdrv_rbd = {
.bdrv_get_info = qemu_rbd_getinfo,
.create_opts = &qemu_rbd_create_opts,
.bdrv_getlength = qemu_rbd_getlength,
+ .bdrv_get_allocated_file_size = qemu_rbd_allocated_file_size,
.bdrv_co_truncate = qemu_rbd_co_truncate,
.protocol_name = "rbd",
--
2.25.3
^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [PATCH] rbd: Use RBD fast-diff for querying actual allocation
2020-06-09 7:30 [PATCH] rbd: Use RBD fast-diff for querying actual allocation Yi Li
@ 2020-06-09 7:54 ` 李义
2020-06-09 12:20 ` Jason Dillaman
1 sibling, 0 replies; 7+ messages in thread
From: 李义 @ 2020-06-09 7:54 UTC (permalink / raw)
To: Yi Li, qemu-devel; +Cc: kwolf, dillaman, qemu-block, mreitz
CC qemu-block and rbd maintainers
On 6/9/20, Yi Li <yili@winhong.com> wrote:
> Since Ceph version Infernalis (9.2.0) the new fast-diff mechanism
> of RBD allows for querying actual rbd image usage.
>
> Prior to this version there was no easy and fast way to query how
> much allocation a RBD image had inside a Ceph cluster.
>
> To use the fast-diff feature it needs to be enabled per RBD image
> and is only supported by Ceph cluster running version Infernalis
> (9.2.0) or newer.
>
> Without the fast-diff feature enabled qemu-img will report an allocation
> identical to the image capacity.
>
> 'qemu-img info rbd:cepharm/liyi-rbd' might output for example:
>
> image: json:{"driver": "raw", "file": {"pool": "cepharm",
> "image": "liyi-rbd", "driver": "rbd"}}
> file format: raw
> virtual size: 20 GiB (21474836480 bytes)
> disk size: 0 B
> cluster_size: 4194304
>
> Newly created rbds will have the fast-diff feature enabled.
>
> Signed-off-by: Yi Li <yili@winhong.com>
> ---
> block/rbd.c | 60 +++++++++++++++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 60 insertions(+)
>
> diff --git a/block/rbd.c b/block/rbd.c
> index 617553b022..f231653f7b 100644
> --- a/block/rbd.c
> +++ b/block/rbd.c
> @@ -1107,6 +1107,65 @@ static int64_t qemu_rbd_getlength(BlockDriverState
> *bs)
> return info.size;
> }
>
> +#if LIBRBD_VERSION_CODE > 265
> +static int disk_usage_callback(uint64_t offset, size_t len, int exists,
> + void *arg)
> +{
> + uint64_t *used_size = (uint64_t *)(arg);
> + if (exists) {
> + (*used_size) += len;
> + }
> + return 0;
> +}
> +#endif
> +
> +static int64_t qemu_rbd_allocated_file_size(BlockDriverState *bs)
> +{
> + BDRVRBDState *s = bs->opaque;
> + rbd_image_info_t info;
> + int r;
> + uint64_t used_size = 0;
> + uint64_t features = 0;
> +
> + r = rbd_stat(s->image, &info, sizeof(info));
> + if (r < 0) {
> + return r;
> + }
> +
> + r = rbd_get_features(s->image, &features);
> + if (r < 0) {
> + return r;
> + }
> +
> + /*
> + * rbd_diff_iterate2() is available in versions above Ceph 0.94
> (Hammer)
> + * It uses a object map inside Ceph which is faster than
> rbd_diff_iterate()
> + * which iterates all objects.
> + * LIBRBD_VERSION_CODE for Ceph 0.94 is 265. In 266 and upwards
> diff_iterate2
> + * is available
> + */
> +#if LIBRBD_VERSION_CODE > 265
> + if (features & RBD_FEATURE_FAST_DIFF) {
> +
> + /*
> + * RBD image fast-diff feature enabled
> + * Querying for actual allocation.
> + */
> + r = rbd_diff_iterate2(s->image, NULL, 0, info.size, 0, 1,
> + &disk_usage_callback,
> + &used_size);
> + if (r < 0) {
> + return r;
> + }
> + } else {
> + used_size = info.size;
> + }
> +#else
> + used_size = info.size;
> +#endif
> + return used_size;
> +}
> +
> static int coroutine_fn qemu_rbd_co_truncate(BlockDriverState *bs,
> int64_t offset,
> bool exact,
> @@ -1316,6 +1375,7 @@ static BlockDriver bdrv_rbd = {
> .bdrv_get_info = qemu_rbd_getinfo,
> .create_opts = &qemu_rbd_create_opts,
> .bdrv_getlength = qemu_rbd_getlength,
> + .bdrv_get_allocated_file_size = qemu_rbd_allocated_file_size,
> .bdrv_co_truncate = qemu_rbd_co_truncate,
> .protocol_name = "rbd",
>
> --
> 2.25.3
>
>
>
>
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] rbd: Use RBD fast-diff for querying actual allocation
2020-06-09 7:30 [PATCH] rbd: Use RBD fast-diff for querying actual allocation Yi Li
2020-06-09 7:54 ` 李义
@ 2020-06-09 12:20 ` Jason Dillaman
2020-06-09 13:56 ` 李义
1 sibling, 1 reply; 7+ messages in thread
From: Jason Dillaman @ 2020-06-09 12:20 UTC (permalink / raw)
To: Yi Li; +Cc: yilikernel, qemu-devel
On Tue, Jun 9, 2020 at 3:31 AM Yi Li <yili@winhong.com> wrote:
>
> Since Ceph version Infernalis (9.2.0) the new fast-diff mechanism
> of RBD allows for querying actual rbd image usage.
>
> Prior to this version there was no easy and fast way to query how
> much allocation a RBD image had inside a Ceph cluster.
>
> To use the fast-diff feature it needs to be enabled per RBD image
> and is only supported by Ceph cluster running version Infernalis
> (9.2.0) or newer.
>
> Without the fast-diff feature enabled qemu-img will report an allocation
> identical to the image capacity.
>
> 'qemu-img info rbd:cepharm/liyi-rbd' might output for example:
>
> image: json:{"driver": "raw", "file": {"pool": "cepharm",
> "image": "liyi-rbd", "driver": "rbd"}}
> file format: raw
> virtual size: 20 GiB (21474836480 bytes)
> disk size: 0 B
> cluster_size: 4194304
>
> Newly created rbds will have the fast-diff feature enabled.
>
> Signed-off-by: Yi Li <yili@winhong.com>
> ---
> block/rbd.c | 60 +++++++++++++++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 60 insertions(+)
>
> diff --git a/block/rbd.c b/block/rbd.c
> index 617553b022..f231653f7b 100644
> --- a/block/rbd.c
> +++ b/block/rbd.c
> @@ -1107,6 +1107,65 @@ static int64_t qemu_rbd_getlength(BlockDriverState *bs)
> return info.size;
> }
>
> +#if LIBRBD_VERSION_CODE > 265
> +static int disk_usage_callback(uint64_t offset, size_t len, int exists,
> + void *arg)
> +{
> + uint64_t *used_size = (uint64_t *)(arg);
> + if (exists) {
> + (*used_size) += len;
> + }
> + return 0;
> +}
> +#endif
> +
> +static int64_t qemu_rbd_allocated_file_size(BlockDriverState *bs)
> +{
> + BDRVRBDState *s = bs->opaque;
> + rbd_image_info_t info;
> + int r;
> + uint64_t used_size = 0;
> + uint64_t features = 0;
> +
> + r = rbd_stat(s->image, &info, sizeof(info));
> + if (r < 0) {
> + return r;
> + }
> +
> + r = rbd_get_features(s->image, &features);
> + if (r < 0) {
> + return r;
> + }
You should probably test the flags to ensure that the
RBD_FLAG_FAST_DIFF_INVALID flag is not set [1]. It's potentially very
slow and expensive to calculate the disk usage w/o a fast-diff (on
large images) since it requires iterating over every possible 4MiB
backing data object (by default) to query its actual usage.
> + /*
> + * rbd_diff_iterate2() is available in versions above Ceph 0.94 (Hammer)
> + * It uses a object map inside Ceph which is faster than rbd_diff_iterate()
> + * which iterates all objects.
> + * LIBRBD_VERSION_CODE for Ceph 0.94 is 265. In 266 and upwards diff_iterate2
> + * is available
> + */
> +#if LIBRBD_VERSION_CODE > 265
> + if (features & RBD_FEATURE_FAST_DIFF) {
> +
> + /*
> + * RBD image fast-diff feature enabled
> + * Querying for actual allocation.
> + */
> + r = rbd_diff_iterate2(s->image, NULL, 0, info.size, 0, 1,
> + &disk_usage_callback,
> + &used_size);
> + if (r < 0) {
> + return r;
> + }
> + } else {
> + used_size = info.size;
> + }
> +#else
> + used_size = info.size;
> +#endif
> + return used_size;
> +}
> +
> static int coroutine_fn qemu_rbd_co_truncate(BlockDriverState *bs,
> int64_t offset,
> bool exact,
> @@ -1316,6 +1375,7 @@ static BlockDriver bdrv_rbd = {
> .bdrv_get_info = qemu_rbd_getinfo,
> .create_opts = &qemu_rbd_create_opts,
> .bdrv_getlength = qemu_rbd_getlength,
> + .bdrv_get_allocated_file_size = qemu_rbd_allocated_file_size,
> .bdrv_co_truncate = qemu_rbd_co_truncate,
> .protocol_name = "rbd",
>
> --
> 2.25.3
>
>
>
>
[1] https://github.com/libvirt/libvirt/commit/21deeaf02fdf216b08210fc899579736973ca81d#diff-107c5451015e5980c90048ff615becb8
--
Jason
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] rbd: Use RBD fast-diff for querying actual allocation
2020-06-09 12:20 ` Jason Dillaman
@ 2020-06-09 13:56 ` 李义
2020-06-09 15:13 ` [PATCH v2] " Yi Li
2020-06-11 15:36 ` Yi Li
0 siblings, 2 replies; 7+ messages in thread
From: 李义 @ 2020-06-09 13:56 UTC (permalink / raw)
To: dillaman; +Cc: kwolf, 李义, qemu-block, qemu-devel, mreitz, Yi Li
On 6/9/20, Jason Dillaman <jdillama@redhat.com> wrote:
> On Tue, Jun 9, 2020 at 3:31 AM Yi Li <yili@winhong.com> wrote:
>>
>> Since Ceph version Infernalis (9.2.0) the new fast-diff mechanism
>> of RBD allows for querying actual rbd image usage.
>>
>> Prior to this version there was no easy and fast way to query how
>> much allocation a RBD image had inside a Ceph cluster.
>>
>> To use the fast-diff feature it needs to be enabled per RBD image
>> and is only supported by Ceph cluster running version Infernalis
>> (9.2.0) or newer.
>>
>> Without the fast-diff feature enabled qemu-img will report an allocation
>> identical to the image capacity.
>>
>> 'qemu-img info rbd:cepharm/liyi-rbd' might output for example:
>>
>> image: json:{"driver": "raw", "file": {"pool": "cepharm",
>> "image": "liyi-rbd", "driver": "rbd"}}
>> file format: raw
>> virtual size: 20 GiB (21474836480 bytes)
>> disk size: 0 B
>> cluster_size: 4194304
>>
>> Newly created rbds will have the fast-diff feature enabled.
>>
>> Signed-off-by: Yi Li <yili@winhong.com>
>> ---
>> block/rbd.c | 60 +++++++++++++++++++++++++++++++++++++++++++++++++++++
>> 1 file changed, 60 insertions(+)
>>
>> diff --git a/block/rbd.c b/block/rbd.c
>> index 617553b022..f231653f7b 100644
>> --- a/block/rbd.c
>> +++ b/block/rbd.c
>> @@ -1107,6 +1107,65 @@ static int64_t qemu_rbd_getlength(BlockDriverState
>> *bs)
>> return info.size;
>> }
>>
>> +#if LIBRBD_VERSION_CODE > 265
>> +static int disk_usage_callback(uint64_t offset, size_t len, int exists,
>> + void *arg)
>> +{
>> + uint64_t *used_size = (uint64_t *)(arg);
>> + if (exists) {
>> + (*used_size) += len;
>> + }
>> + return 0;
>> +}
>> +#endif
>> +
>> +static int64_t qemu_rbd_allocated_file_size(BlockDriverState *bs)
>> +{
>> + BDRVRBDState *s = bs->opaque;
>> + rbd_image_info_t info;
>> + int r;
>> + uint64_t used_size = 0;
>> + uint64_t features = 0;
>> +
>> + r = rbd_stat(s->image, &info, sizeof(info));
>> + if (r < 0) {
>> + return r;
>> + }
>> +
>> + r = rbd_get_features(s->image, &features);
>> + if (r < 0) {
>> + return r;
>> + }
>
> You should probably test the flags to ensure that the
> RBD_FLAG_FAST_DIFF_INVALID flag is not set [1]. It's potentially very
> slow and expensive to calculate the disk usage w/o a fast-diff (on
> large images) since it requires iterating over every possible 4MiB
> backing data object (by default) to query its actual usage.
>
Thanks for your review and remind me of this issue.
>> + /*
>> + * rbd_diff_iterate2() is available in versions above Ceph 0.94
>> (Hammer)
>> + * It uses a object map inside Ceph which is faster than
>> rbd_diff_iterate()
>> + * which iterates all objects.
>> + * LIBRBD_VERSION_CODE for Ceph 0.94 is 265. In 266 and upwards
>> diff_iterate2
>> + * is available
>> + */
>> +#if LIBRBD_VERSION_CODE > 265
>> + if (features & RBD_FEATURE_FAST_DIFF) {
>> +
>> + /*
>> + * RBD image fast-diff feature enabled
>> + * Querying for actual allocation.
>> + */
>> + r = rbd_diff_iterate2(s->image, NULL, 0, info.size, 0, 1,
>> + &disk_usage_callback,
>> + &used_size);
>> + if (r < 0) {
>> + return r;
>> + }
>> + } else {
>> + used_size = info.size;
>> + }
>> +#else
>> + used_size = info.size;
>> +#endif
>> + return used_size;
>> +}
>> +
>> static int coroutine_fn qemu_rbd_co_truncate(BlockDriverState *bs,
>> int64_t offset,
>> bool exact,
>> @@ -1316,6 +1375,7 @@ static BlockDriver bdrv_rbd = {
>> .bdrv_get_info = qemu_rbd_getinfo,
>> .create_opts = &qemu_rbd_create_opts,
>> .bdrv_getlength = qemu_rbd_getlength,
>> + .bdrv_get_allocated_file_size = qemu_rbd_allocated_file_size,
>> .bdrv_co_truncate = qemu_rbd_co_truncate,
>> .protocol_name = "rbd",
>>
>> --
>> 2.25.3
>>
>>
>>
>>
>
> [1]
> https://github.com/libvirt/libvirt/commit/21deeaf02fdf216b08210fc899579736973ca81d#diff-107c5451015e5980c90048ff615becb8
>
> --
> Jason
>
>
^ permalink raw reply [flat|nested] 7+ messages in thread
* [PATCH v2] rbd: Use RBD fast-diff for querying actual allocation
2020-06-09 13:56 ` 李义
@ 2020-06-09 15:13 ` Yi Li
2020-06-11 15:36 ` Yi Li
1 sibling, 0 replies; 7+ messages in thread
From: Yi Li @ 2020-06-09 15:13 UTC (permalink / raw)
To: dillaman; +Cc: kwolf, yilikernel, qemu-block, qemu-devel, mreitz, yili
Since Ceph version Infernalis (9.2.0) the new fast-diff mechanism
of RBD allows for querying actual rbd image usage.
Prior to this version there was no easy and fast way to query how
much allocation a RBD image had inside a Ceph cluster.
To use the fast-diff feature it needs to be enabled per RBD image
and is only supported by Ceph cluster running version Infernalis
(9.2.0) or newer.
The fast-diff feature disabled or fast-diff map is marked as invalid,
qemu-img will report an allocation identical to the image capacity.
'qemu-img info rbd:cepharm/liyi-rbd' might output for example:
image: json:{"driver": "raw", "file": {"pool": "cepharm",
"image": "liyi-rbd", "driver": "rbd"}}
file format: raw
virtual size: 20 GiB (21474836480 bytes)
disk size: 0 B
cluster_size: 4194304
Newly created rbds will have the fast-diff feature enabled.
Signed-off-by: Yi Li <yili@winhong.com>
---
block/rbd.c | 103 ++++++++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 103 insertions(+)
diff --git a/block/rbd.c b/block/rbd.c
index 617553b022..c1e68ff7e9 100644
--- a/block/rbd.c
+++ b/block/rbd.c
@@ -1107,6 +1107,108 @@ static int64_t qemu_rbd_getlength(BlockDriverState *bs)
return info.size;
}
+#if LIBRBD_VERSION_CODE > 265
+static int disk_usage_callback(uint64_t offset, size_t len, int exists,
+ void *arg)
+{
+ uint64_t *used_size = (uint64_t *)(arg);
+ if (exists) {
+ (*used_size) += len;
+ }
+ return 0;
+}
+
+static int qemu_rbd_getflags(rbd_image_t image, uint64_t *flags)
+{
+ int r;
+
+ r = rbd_get_flags(image, flags);
+ if (r < 0) {
+ return r;
+ }
+ return 0;
+}
+
+static bool qemu_rbd_use_fastdiff(uint64_t features, uint64_t flags)
+{
+ return (((features & RBD_FEATURE_FAST_DIFF) != 0ULL) &&
+ ((flags & RBD_FLAG_FAST_DIFF_INVALID) == 0ULL));
+}
+
+static int qemu_rbd_set_allocation(rbd_image_t image,
+ rbd_image_info_t *info,
+ uint64_t *used_size)
+{
+ int r;
+ /*
+ * RBD image fast-diff feature enabled
+ * Querying for actual allocation.
+ */
+ r = rbd_diff_iterate2(image, NULL, 0, info->size, 0, 1,
+ &disk_usage_callback,
+ used_size);
+ if (r < 0) {
+ return r;
+ }
+ return 0;
+}
+
+#else
+static int qemu_rbd_getflags(rbd_image_t image G_GNUC_UNUSED, uint64_t *flags)
+{
+ *flags = 0;
+ return 0;
+}
+
+static bool qemu_rbd_use_fastdiff(uint64_t features G_GNUC_UNUSED,
+ uint64_t feature_flags G_GNUC_UNUSED)
+{
+ return false;
+}
+
+static int qemu_rbd_set_allocation(rbd_image_t image G_GNUC_UNUSED,
+ rbd_image_info_t *info _GNUC_UNUSED,
+ uint64_t *used_size _GNUC_UNUSED)
+{
+ return 0;
+}
+#endif
+
+static int64_t qemu_rbd_allocated_file_size(BlockDriverState *bs)
+{
+ BDRVRBDState *s = bs->opaque;
+ rbd_image_info_t info;
+ int r;
+ uint64_t used_size = 0;
+ uint64_t features = 0;
+ uint64_t flags = 0;
+
+ r = rbd_stat(s->image, &info, sizeof(info));
+ if (r < 0) {
+ return r;
+ }
+
+ r = rbd_get_features(s->image, &features);
+ if (r < 0) {
+ return r;
+ }
+
+ r = qemu_rbd_getflags(s->image, &flags);
+ if (r < 0) {
+ return r;
+ }
+
+ if (qemu_rbd_use_fastdiff(features, flags)) {
+ r = qemu_rbd_set_allocation(s->image, &info, &used_size);
+ if (r < 0) {
+ return r;
+ }
+ } else {
+ used_size = info.size;
+ }
+ return used_size;
+}
+
static int coroutine_fn qemu_rbd_co_truncate(BlockDriverState *bs,
int64_t offset,
bool exact,
@@ -1316,6 +1418,7 @@ static BlockDriver bdrv_rbd = {
.bdrv_get_info = qemu_rbd_getinfo,
.create_opts = &qemu_rbd_create_opts,
.bdrv_getlength = qemu_rbd_getlength,
+ .bdrv_get_allocated_file_size = qemu_rbd_allocated_file_size,
.bdrv_co_truncate = qemu_rbd_co_truncate,
.protocol_name = "rbd",
--
2.25.3
^ permalink raw reply related [flat|nested] 7+ messages in thread
* [PATCH v2] rbd: Use RBD fast-diff for querying actual allocation
2020-06-09 13:56 ` 李义
2020-06-09 15:13 ` [PATCH v2] " Yi Li
@ 2020-06-11 15:36 ` Yi Li
2020-06-17 14:13 ` Yi Li
1 sibling, 1 reply; 7+ messages in thread
From: Yi Li @ 2020-06-11 15:36 UTC (permalink / raw)
To: jdillama; +Cc: yilikernel, qemu-devel, yili
Since Ceph version Infernalis (9.2.0) the new fast-diff mechanism
of RBD allows for querying actual rbd image usage.
Prior to this version there was no easy and fast way to query how
much allocation a RBD image had inside a Ceph cluster.
To use the fast-diff feature it needs to be enabled per RBD image
and is only supported by Ceph cluster running version Infernalis
(9.2.0) or newer.
The fast-diff feature disabled or fast-diff map is marked as invalid,
qemu-img will report an allocation identical to the image capacity.
'qemu-img info rbd:cepharm/liyi-rbd' might output for example:
image: json:{"driver": "raw", "file": {"pool": "cepharm",
"image": "liyi-rbd", "driver": "rbd"}}
file format: raw
virtual size: 20 GiB (21474836480 bytes)
disk size: 0 B
cluster_size: 4194304
Newly created rbds will have the fast-diff feature enabled.
Signed-off-by: Yi Li <yili@winhong.com>
---
block/rbd.c | 103 ++++++++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 103 insertions(+)
diff --git a/block/rbd.c b/block/rbd.c
index 617553b022..c1e68ff7e9 100644
--- a/block/rbd.c
+++ b/block/rbd.c
@@ -1107,6 +1107,108 @@ static int64_t qemu_rbd_getlength(BlockDriverState *bs)
return info.size;
}
+#if LIBRBD_VERSION_CODE > 265
+static int disk_usage_callback(uint64_t offset, size_t len, int exists,
+ void *arg)
+{
+ uint64_t *used_size = (uint64_t *)(arg);
+ if (exists) {
+ (*used_size) += len;
+ }
+ return 0;
+}
+
+static int qemu_rbd_getflags(rbd_image_t image, uint64_t *flags)
+{
+ int r;
+
+ r = rbd_get_flags(image, flags);
+ if (r < 0) {
+ return r;
+ }
+ return 0;
+}
+
+static bool qemu_rbd_use_fastdiff(uint64_t features, uint64_t flags)
+{
+ return (((features & RBD_FEATURE_FAST_DIFF) != 0ULL) &&
+ ((flags & RBD_FLAG_FAST_DIFF_INVALID) == 0ULL));
+}
+
+static int qemu_rbd_set_allocation(rbd_image_t image,
+ rbd_image_info_t *info,
+ uint64_t *used_size)
+{
+ int r;
+ /*
+ * RBD image fast-diff feature enabled
+ * Querying for actual allocation.
+ */
+ r = rbd_diff_iterate2(image, NULL, 0, info->size, 0, 1,
+ &disk_usage_callback,
+ used_size);
+ if (r < 0) {
+ return r;
+ }
+ return 0;
+}
+
+#else
+static int qemu_rbd_getflags(rbd_image_t image G_GNUC_UNUSED, uint64_t *flags)
+{
+ *flags = 0;
+ return 0;
+}
+
+static bool qemu_rbd_use_fastdiff(uint64_t features G_GNUC_UNUSED,
+ uint64_t feature_flags G_GNUC_UNUSED)
+{
+ return false;
+}
+
+static int qemu_rbd_set_allocation(rbd_image_t image G_GNUC_UNUSED,
+ rbd_image_info_t *info _GNUC_UNUSED,
+ uint64_t *used_size _GNUC_UNUSED)
+{
+ return 0;
+}
+#endif
+
+static int64_t qemu_rbd_allocated_file_size(BlockDriverState *bs)
+{
+ BDRVRBDState *s = bs->opaque;
+ rbd_image_info_t info;
+ int r;
+ uint64_t used_size = 0;
+ uint64_t features = 0;
+ uint64_t flags = 0;
+
+ r = rbd_stat(s->image, &info, sizeof(info));
+ if (r < 0) {
+ return r;
+ }
+
+ r = rbd_get_features(s->image, &features);
+ if (r < 0) {
+ return r;
+ }
+
+ r = qemu_rbd_getflags(s->image, &flags);
+ if (r < 0) {
+ return r;
+ }
+
+ if (qemu_rbd_use_fastdiff(features, flags)) {
+ r = qemu_rbd_set_allocation(s->image, &info, &used_size);
+ if (r < 0) {
+ return r;
+ }
+ } else {
+ used_size = info.size;
+ }
+ return used_size;
+}
+
static int coroutine_fn qemu_rbd_co_truncate(BlockDriverState *bs,
int64_t offset,
bool exact,
@@ -1316,6 +1418,7 @@ static BlockDriver bdrv_rbd = {
.bdrv_get_info = qemu_rbd_getinfo,
.create_opts = &qemu_rbd_create_opts,
.bdrv_getlength = qemu_rbd_getlength,
+ .bdrv_get_allocated_file_size = qemu_rbd_allocated_file_size,
.bdrv_co_truncate = qemu_rbd_co_truncate,
.protocol_name = "rbd",
--
2.25.3
^ permalink raw reply related [flat|nested] 7+ messages in thread
* Re: [PATCH v2] rbd: Use RBD fast-diff for querying actual allocation
2020-06-11 15:36 ` Yi Li
@ 2020-06-17 14:13 ` Yi Li
0 siblings, 0 replies; 7+ messages in thread
From: Yi Li @ 2020-06-17 14:13 UTC (permalink / raw)
To: jdillama, qemu-devel; +Cc: kwolf, qemu-block, mreitz
ping ?
On 6/11/20, Yi Li <yili@winhong.com> wrote:
> Since Ceph version Infernalis (9.2.0) the new fast-diff mechanism
> of RBD allows for querying actual rbd image usage.
>
> Prior to this version there was no easy and fast way to query how
> much allocation a RBD image had inside a Ceph cluster.
>
> To use the fast-diff feature it needs to be enabled per RBD image
> and is only supported by Ceph cluster running version Infernalis
> (9.2.0) or newer.
>
> The fast-diff feature disabled or fast-diff map is marked as invalid,
> qemu-img will report an allocation identical to the image capacity.
>
> 'qemu-img info rbd:cepharm/liyi-rbd' might output for example:
>
> image: json:{"driver": "raw", "file": {"pool": "cepharm",
> "image": "liyi-rbd", "driver": "rbd"}}
> file format: raw
> virtual size: 20 GiB (21474836480 bytes)
> disk size: 0 B
> cluster_size: 4194304
>
> Newly created rbds will have the fast-diff feature enabled.
>
> Signed-off-by: Yi Li <yili@winhong.com>
> ---
> block/rbd.c | 103 ++++++++++++++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 103 insertions(+)
>
> diff --git a/block/rbd.c b/block/rbd.c
> index 617553b022..c1e68ff7e9 100644
> --- a/block/rbd.c
> +++ b/block/rbd.c
> @@ -1107,6 +1107,108 @@ static int64_t qemu_rbd_getlength(BlockDriverState
> *bs)
> return info.size;
> }
>
> +#if LIBRBD_VERSION_CODE > 265
> +static int disk_usage_callback(uint64_t offset, size_t len, int exists,
> + void *arg)
> +{
> + uint64_t *used_size = (uint64_t *)(arg);
> + if (exists) {
> + (*used_size) += len;
> + }
> + return 0;
> +}
> +
> +static int qemu_rbd_getflags(rbd_image_t image, uint64_t *flags)
> +{
> + int r;
> +
> + r = rbd_get_flags(image, flags);
> + if (r < 0) {
> + return r;
> + }
> + return 0;
> +}
> +
> +static bool qemu_rbd_use_fastdiff(uint64_t features, uint64_t flags)
> +{
> + return (((features & RBD_FEATURE_FAST_DIFF) != 0ULL) &&
> + ((flags & RBD_FLAG_FAST_DIFF_INVALID) == 0ULL));
> +}
> +
> +static int qemu_rbd_set_allocation(rbd_image_t image,
> + rbd_image_info_t *info,
> + uint64_t *used_size)
> +{
> + int r;
> + /*
> + * RBD image fast-diff feature enabled
> + * Querying for actual allocation.
> + */
> + r = rbd_diff_iterate2(image, NULL, 0, info->size, 0, 1,
> + &disk_usage_callback,
> + used_size);
> + if (r < 0) {
> + return r;
> + }
> + return 0;
> +}
> +
> +#else
> +static int qemu_rbd_getflags(rbd_image_t image G_GNUC_UNUSED, uint64_t
> *flags)
> +{
> + *flags = 0;
> + return 0;
> +}
> +
> +static bool qemu_rbd_use_fastdiff(uint64_t features G_GNUC_UNUSED,
> + uint64_t feature_flags G_GNUC_UNUSED)
> +{
> + return false;
> +}
> +
> +static int qemu_rbd_set_allocation(rbd_image_t image G_GNUC_UNUSED,
> + rbd_image_info_t *info _GNUC_UNUSED,
> + uint64_t *used_size _GNUC_UNUSED)
> +{
> + return 0;
> +}
> +#endif
> +
> +static int64_t qemu_rbd_allocated_file_size(BlockDriverState *bs)
> +{
> + BDRVRBDState *s = bs->opaque;
> + rbd_image_info_t info;
> + int r;
> + uint64_t used_size = 0;
> + uint64_t features = 0;
> + uint64_t flags = 0;
> +
> + r = rbd_stat(s->image, &info, sizeof(info));
> + if (r < 0) {
> + return r;
> + }
> +
> + r = rbd_get_features(s->image, &features);
> + if (r < 0) {
> + return r;
> + }
> +
> + r = qemu_rbd_getflags(s->image, &flags);
> + if (r < 0) {
> + return r;
> + }
> +
> + if (qemu_rbd_use_fastdiff(features, flags)) {
> + r = qemu_rbd_set_allocation(s->image, &info, &used_size);
> + if (r < 0) {
> + return r;
> + }
> + } else {
> + used_size = info.size;
> + }
> + return used_size;
> +}
> +
> static int coroutine_fn qemu_rbd_co_truncate(BlockDriverState *bs,
> int64_t offset,
> bool exact,
> @@ -1316,6 +1418,7 @@ static BlockDriver bdrv_rbd = {
> .bdrv_get_info = qemu_rbd_getinfo,
> .create_opts = &qemu_rbd_create_opts,
> .bdrv_getlength = qemu_rbd_getlength,
> + .bdrv_get_allocated_file_size = qemu_rbd_allocated_file_size,
> .bdrv_co_truncate = qemu_rbd_co_truncate,
> .protocol_name = "rbd",
>
> --
> 2.25.3
>
>
>
>
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2020-06-17 14:14 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-06-09 7:30 [PATCH] rbd: Use RBD fast-diff for querying actual allocation Yi Li
2020-06-09 7:54 ` 李义
2020-06-09 12:20 ` Jason Dillaman
2020-06-09 13:56 ` 李义
2020-06-09 15:13 ` [PATCH v2] " Yi Li
2020-06-11 15:36 ` Yi Li
2020-06-17 14:13 ` Yi Li
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).