[3/4,RFC] btrfs: export global block reserve size as space_info
diff mbox

Message ID 21dfbfe3b1361092ad1dc286f636c1fff4a2a4f6.1391779611.git.dsterba@suse.cz
State Accepted
Headers show

Commit Message

David Sterba Feb. 7, 2014, 1:34 p.m. UTC
Introduce a block group type bit for a global reserve and fill the space
info for SPACE_INFO ioctl. This should replace the newly added ioctl
(01e219e8069516cdb98594d417b8bb8d906ed30d) to get just the 'size' part
of the global reserve, while the actual usage can be now visible in the
'btrfs fi df' output during ENOSPC stress.

The unpatched userspace tools will show the blockgroup as 'unknown'.

CC: Jeff Mahoney <jeffm@suse.com>
CC: Josef Bacik <jbacik@fb.com>
Signed-off-by: David Sterba <dsterba@suse.cz>
---
 fs/btrfs/ctree.h |  9 ++++++++-
 fs/btrfs/ioctl.c | 20 ++++++++++++++++++++
 2 files changed, 28 insertions(+), 1 deletion(-)

Comments

Chris Mason Feb. 10, 2014, 5:03 p.m. UTC | #1
On 02/07/2014 08:34 AM, David Sterba wrote:
> Introduce a block group type bit for a global reserve and fill the space
> info for SPACE_INFO ioctl. This should replace the newly added ioctl
> (01e219e8069516cdb98594d417b8bb8d906ed30d) to get just the 'size' part
> of the global reserve, while the actual usage can be now visible in the
> 'btrfs fi df' output during ENOSPC stress.
>
> The unpatched userspace tools will show the blockgroup as 'unknown'.
>

This wasn't in my rc2 pull because I wanted to sync up with Jeff on it. 
  I like the idea of combining this into SPACE_INFO, any objections?

-chris
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Jeff Mahoney April 4, 2014, 2:56 p.m. UTC | #2
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 2/10/14, 12:03 PM, Chris Mason wrote:
> On 02/07/2014 08:34 AM, David Sterba wrote:
>> Introduce a block group type bit for a global reserve and fill
>> the space info for SPACE_INFO ioctl. This should replace the
>> newly added ioctl (01e219e8069516cdb98594d417b8bb8d906ed30d) to
>> get just the 'size' part of the global reserve, while the actual
>> usage can be now visible in the 'btrfs fi df' output during
>> ENOSPC stress.
>> 
>> The unpatched userspace tools will show the blockgroup as
>> 'unknown'.
>> 
> 
> This wasn't in my rc2 pull because I wanted to sync up with Jeff on
> it. I like the idea of combining this into SPACE_INFO, any
> objections?

Sorry, was on vacation when this went by and just got the ping. Yeah,
I have no objections here.

- -Jeff

- -- 
Jeff Mahoney
SUSE Labs
-----BEGIN PGP SIGNATURE-----
Version: GnuPG/MacGPG2 v2.0.22 (Darwin)
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=fiJ0
-----END PGP SIGNATURE-----
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Patch
diff mbox

diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 2c1a42ca519f..8bf1890ea21f 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -985,7 +985,8 @@  struct btrfs_dev_replace_item {
 #define BTRFS_BLOCK_GROUP_RAID10	(1ULL << 6)
 #define BTRFS_BLOCK_GROUP_RAID5         (1ULL << 7)
 #define BTRFS_BLOCK_GROUP_RAID6         (1ULL << 8)
-#define BTRFS_BLOCK_GROUP_RESERVED	BTRFS_AVAIL_ALLOC_BIT_SINGLE
+#define BTRFS_BLOCK_GROUP_RESERVED	(BTRFS_AVAIL_ALLOC_BIT_SINGLE | \
+					 BTRFS_SPACE_INFO_GLOBAL_RSV)
 
 enum btrfs_raid_types {
 	BTRFS_RAID_RAID10,
@@ -1017,6 +1018,12 @@  enum btrfs_raid_types {
  */
 #define BTRFS_AVAIL_ALLOC_BIT_SINGLE	(1ULL << 48)
 
+/*
+ * A fake block group type that is used to communicate global block reserve
+ * size to userspace via the SPACE_INFO ioctl.
+ */
+#define BTRFS_SPACE_INFO_GLOBAL_RSV	(1ULL << 49)
+
 #define BTRFS_EXTENDED_PROFILE_MASK	(BTRFS_BLOCK_GROUP_PROFILE_MASK | \
 					 BTRFS_AVAIL_ALLOC_BIT_SINGLE)
 
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 383ab455bfa7..0d938f8f173e 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -3465,6 +3465,11 @@  static long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg)
 		up_read(&info->groups_sem);
 	}
 
+	/*
+	 * Global block reserve, exported as a space_info
+	 */
+	slot_count++;
+
 	/* space_slots == 0 means they are asking for a count */
 	if (space_args.space_slots == 0) {
 		space_args.total_spaces = slot_count;
@@ -3523,6 +3528,21 @@  static long btrfs_ioctl_space_info(struct btrfs_root *root, void __user *arg)
 		up_read(&info->groups_sem);
 	}
 
+	/*
+	 * Add global block reserve
+	 */
+	if (slot_count) {
+		struct btrfs_block_rsv *block_rsv = &root->fs_info->global_block_rsv;
+
+		spin_lock(&block_rsv->lock);
+		space.total_bytes = block_rsv->size;
+		space.used_bytes = block_rsv->size - block_rsv->reserved;
+		spin_unlock(&block_rsv->lock);
+		space.flags = BTRFS_SPACE_INFO_GLOBAL_RSV;
+		memcpy(dest, &space, sizeof(space));
+		space_args.total_spaces++;
+	}
+
 	user_dest = (struct btrfs_ioctl_space_info __user *)
 		(arg + sizeof(struct btrfs_ioctl_space_args));