Skip to contentSkip to conversation

btrfs: add owner and fs_info for btrfs_device::alloc_state

PATCH In review Qu Wenruo 3 participants 3 messages Aug 20, 2020 Btrfs file system

lore
Share & tools
lore

Qu Wenruo

author
#0 lore raw

PATCH 1/1 btrfs: add owner and fs_info for btrfs_device::alloc_state

Commit 1c11b63eff2a ("btrfs: replace pending/pinned chunks lists with io tree") introduced btrfs_device::alloc_state extent io tree, but it doesn't initialize the fs_info and owner member.

This means the following features are not properly supported:
- Fs owner report for insert_state() error
Without fs_info initialized, although btrfs_err() won't panic, it won't output which fs is causing the error.

- Wrong owner for trace events
alloc_state will get the owner as pinned extents.

Fix this by assiging proper fs_info and owner for
btrfs_device::alloc_state.

Fixes 1c11b63eff2a ("btrfs: replace pending/pinned chunks lists with io tree")
---
 fs/btrfs/extent-io-tree.h | 1 +
 fs/btrfs/volumes.c        | 7 ++++---
 2 files changed, 5 insertions(+), 3 deletions(-)
fs/btrfs/extent-io-tree.h +1 -0
index 219a09a2b734..250b8cbaaf97 100644
@@ -48,6 +48,7 @@ enum {
 	IO_TREE_INODE_FILE_EXTENT,
 	IO_TREE_LOG_CSUM_RANGE,
 	IO_TREE_SELFTEST,
+	IO_TREE_DEVICE_ALLOC_STATE,
 };
 
 struct extent_io_tree {
fs/btrfs/volumes.c +4 -3
index 1d5026fdfc75..4683533f3de3 100644
@@ -405,7 +405,7 @@ void __exit btrfs_cleanup_fs_uuids(void)
  * Returned struct is not linked onto any lists and must be destroyed using
  * btrfs_free_device.
  */
-static struct btrfs_device *__alloc_device(void)
+static struct btrfs_device *__alloc_device(struct btrfs_fs_info *fs_info)
 {
 	struct btrfs_device *dev;
 
@@ -432,7 +432,8 @@ static struct btrfs_device *__alloc_device(void)
 	btrfs_device_data_ordered_init(dev);
 	INIT_RADIX_TREE(&dev->reada_zones, GFP_NOFS & ~__GFP_DIRECT_RECLAIM);
 	INIT_RADIX_TREE(&dev->reada_extents, GFP_NOFS & ~__GFP_DIRECT_RECLAIM);
-	extent_io_tree_init(NULL, &dev->alloc_state, 0, NULL);
+	extent_io_tree_init(fs_info, &dev->alloc_state,
+			    IO_TREE_DEVICE_ALLOC_STATE, NULL);
 
 	return dev;
 }
@@ -6524,7 +6525,7 @@ struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info,
 	if (WARN_ON(!devid && !fs_info))
 		return ERR_PTR(-EINVAL);
 
-	dev = __alloc_device();
+	dev = __alloc_device(fs_info);
 	if (IS_ERR(dev))
 		return dev;
 
Signature · 1 line
2.28.0
Reviewed-by on v1 1/1
#1 lore raw
Quoted 9 lines · Qu Wenruo

Commit 1c11b63eff2a ("btrfs: replace pending/pinned chunks lists with io tree") introduced btrfs_device::alloc_state extent io tree, but it doesn't initialize the fs_info and owner member.

This means the following features are not properly supported:
- Fs owner report for insert_state() error
Without fs_info initialized, although btrfs_err() won't panic, it won't output which fs is causing the error.

Quoted 8 lines

- Wrong owner for trace events
alloc_state will get the owner as pinned extents.

Fix this by assiging proper fs_info and owner for
btrfs_device::alloc_state.

Fixes 1c11b63eff2a ("btrfs: replace pending/pinned chunks lists with io tree")
#2 lore raw
In reply to Qu Wenruo
Quoted 9 lines · Qu Wenruo

Commit 1c11b63eff2a ("btrfs: replace pending/pinned chunks lists with io tree") introduced btrfs_device::alloc_state extent io tree, but it doesn't initialize the fs_info and owner member.

This means the following features are not properly supported:
- Fs owner report for insert_state() error
Without fs_info initialized, although btrfs_err() won't panic, it won't output which fs is causing the error.

Quoted 8 lines

- Wrong owner for trace events
alloc_state will get the owner as pinned extents.

Fix this by assiging proper fs_info and owner for
btrfs_device::alloc_state.

Fixes 1c11b63eff2a ("btrfs: replace pending/pinned chunks lists with io tree")

Added to misc-next, thanks.