[v4,00/14] btrfs: more RST delete fixes

Message

Johannes Thumshirn Jan. 13, 2025, 7:31 p.m. UTC

Here's another set of fixes for the delete path on RAID stripe-tree backed
filesystems.

Josef's CI system started tripping over a bad key order due to the usage
of btrfs_set_item_key_safe() in btrfs_partially_delete_raid_extent() and
while investigating what is happening there I found more bugs and not
handled corner cases, which resulted in more fixes and test-cases.

Unfortunately I couldn't fix the bad key order problem and had to resort
to re-creating the item in btrfs_partially_delete_raid_extent() and insert
the new one after deleting the old.

Fstests btrfs/06* are extremely good in exhibiting these failures and
btrfs/060 has been extensively run while developing this series.

A full CI run of v1 can be found here:
https://github.com/btrfs/linux/actions/runs/12291668397

Changes to v1:
- Handle extent_map lookup failure in 1/14
- Don't use key.offset = -1 for initial search in 3/14
- Don't break before calling btrfs_previous_item if we're on slot 0 in
  6/14
- Remove btrfs_mark_buffer_dirty calls
- Remove line breaks at 80 chars if we're just a bit over
- Fix multiple issues on comment styling

Link to v1:
https://lore.kernel.org/linux-btrfs/[e-mail]

Note:
I did not copy the implementation of btrfs_drop_extents() as I'd like to
have feedback on this variant first, before putting the time and energy in
a "completely new" implementation.

---
Changes in v4:
- Handle case when btrfs_previous_item returns 1
- Explicitly document that the root cause of the tree corruption is
  unclear
- Link to v3: https://lore.kernel.org/r/[e-mail]

Changes in v3:
- Drop patch "btrfs: fix search when deleting a RAID stripe-extent"
- Split out setting the incompat flag in the seftests code from patch 1
- Rework changelog of 2/13
- Rename the new stripe_extent item to newitem instead of just 'new' in
  8/13
- Link to v2: https://lore.kernel.org/r/[e-mail]

---
Johannes Thumshirn (14):
      btrfs: selftests: correct RAID stripe-tree feature flag setting
      btrfs: don't try to delete RAID stripe-extents if we don't need to
      btrfs: assert RAID stripe-extent length is always greater than 0
      btrfs: fix front delete range calculation for RAID stripe extents
      btrfs: fix tail delete of RAID stripe-extents
      btrfs: fix deletion of a range spanning parts two RAID stripe extents
      btrfs: implement hole punching for RAID stripe extents
      btrfs: don't use btrfs_set_item_key_safe on RAID stripe-extents
      btrfs: selftests: check for correct return value of failed lookup
      btrfs: selftests: don't split RAID extents in half
      btrfs: selftests: test RAID stripe-tree deletion spanning two items
      btrfs: selftests: add selftest for punching holes into the RAID stripe extents
      btrfs: selftests: add test for punching a hole into 3 RAID stripe-extents
      btrfs: selftests: add a selftest for deleting two out of three extents

 fs/btrfs/ctree.c                        |   1 +
 fs/btrfs/raid-stripe-tree.c             | 144 ++++++-
 fs/btrfs/tests/raid-stripe-tree-tests.c | 661 +++++++++++++++++++++++++++++++-
 3 files changed, 776 insertions(+), 30 deletions(-)
---
base-commit: e6a77e22aa3760314b67b96c8f3d79658a5746e3
change-id: 20241218-rst-delete-fixes-f2659047f627

Best regards,

Comments

David Sterba Jan. 14, 2025, 3:01 p.m. UTC | #1

On Mon, Jan 13, 2025 at 08:31:41PM +0100, Johannes Thumshirn wrote:
> Here's another set of fixes for the delete path on RAID stripe-tree backed
> filesystems.
> 
> Josef's CI system started tripping over a bad key order due to the usage
> of btrfs_set_item_key_safe() in btrfs_partially_delete_raid_extent() and
> while investigating what is happening there I found more bugs and not
> handled corner cases, which resulted in more fixes and test-cases.
> 
> Unfortunately I couldn't fix the bad key order problem and had to resort
> to re-creating the item in btrfs_partially_delete_raid_extent() and insert
> the new one after deleting the old.
> 
> Fstests btrfs/06* are extremely good in exhibiting these failures and
> btrfs/060 has been extensively run while developing this series.
> 
> A full CI run of v1 can be found here:
> https://github.com/btrfs/linux/actions/runs/12291668397
> 
> Changes to v1:
> - Handle extent_map lookup failure in 1/14
> - Don't use key.offset = -1 for initial search in 3/14
> - Don't break before calling btrfs_previous_item if we're on slot 0 in
>   6/14
> - Remove btrfs_mark_buffer_dirty calls
> - Remove line breaks at 80 chars if we're just a bit over
> - Fix multiple issues on comment styling
> 
> Link to v1:
> https://lore.kernel.org/linux-btrfs/[e-mail]
> 
> Note:
> I did not copy the implementation of btrfs_drop_extents() as I'd like to
> have feedback on this variant first, before putting the time and energy in
> a "completely new" implementation.
> 
> ---
> Changes in v4:
> - Handle case when btrfs_previous_item returns 1
> - Explicitly document that the root cause of the tree corruption is
>   unclear
> - Link to v3: https://lore.kernel.org/r/[e-mail]

I've added the series to for-next, with some minor fixups.