[v7,0/8,v7,0/8] fsdax: introduce fs query to support reflink

Message

Shiyang Ruan Sept. 24, 2021, 1:09 p.m. UTC

This patchset is aimed to support shared pages tracking for fsdax.

Changes from [V6 RESEND]:
  - Move ->memory_failure() into the patch who implements it
  - Change the parameter of ->memory_failure():
      unsigned long nr_pfns -> size_t size
  - Remove changes(2 patches) for mapped device
  - Add some necessary comments for functions or interfaces
  - P1: Make a pre-patch for changes for dax_{read,write}_lock()
  - P2: Change the parameter of ->notify_failure():
      void *data -> int flags
  - P3: keep the original dax_lock_page() logic
  - P5: Rewrite the lock function for file's mapping and index:
      dax_lock_mapping_entry()
  - P6: use the new dax_lock_mapping_entry() to lock dax entry, and add
      size parameter to handle a range of failure
  - P7: add a cross range calculation between memory_failure range and
      founded extent range
  - Rebased to v5.15-rc1

This patchset moves owner tracking from dax_assocaite_entry() to pmem
device driver, by introducing an interface ->memory_failure() for struct
pagemap.  This interface is called by memory_failure() in mm, and
implemented by pmem device.

Then call holder operations to find the filesystem which the corrupted
data located in, and call filesystem handler to track files or metadata
associated with this page.

Finally we are able to try to fix the corrupted data in filesystem and
do other necessary processing, such as killing processes who are using
the files affected.

The call trace is like this:
memory_failure()
|* fsdax case
|------------
|pgmap->ops->memory_failure()      => pmem_pgmap_memory_failure()
| dax_holder_notify_failure()      =>
|  dax_device->holder_ops->notify_failure() =>
|                                     - xfs_dax_notify_failure()
|  |* xfs_dax_notify_failure()
|  |--------------------------
|  |   xfs_rmap_query_range()
|  |    xfs_dax_notify_failure_fn()
|  |    * corrupted on metadata
|  |       try to recover data, call xfs_force_shutdown()
|  |    * corrupted on file data
|  |       try to recover data, call mf_dax_kill_procs()
|* normal case
|-------------
 mf_generic_kill_procs()

The fsdax & reflink support for XFS is not contained in this patchset.

(Rebased on v5.15-rc1)
==

Shiyang Ruan (8):
  dax: Use rwsem for dax_{read,write}_lock()
  dax: Introduce holder for dax_device
  mm: factor helpers for memory_failure_dev_pagemap
  pagemap,pmem: Introduce ->memory_failure()
  fsdax: Introduce dax_lock_mapping_entry()
  mm: Introduce mf_dax_kill_procs() for fsdax case
  xfs: Implement ->notify_failure() for XFS
  fsdax: add exception for reflinked files

 drivers/dax/device.c       |  11 +-
 drivers/dax/super.c        | 121 +++++++++++++++++---
 drivers/md/dm-writecache.c |   7 +-
 drivers/nvdimm/pmem.c      |  11 ++
 fs/dax.c                   | 115 ++++++++++++++-----
 fs/xfs/xfs_fsops.c         |   3 +
 fs/xfs/xfs_mount.h         |   1 +
 fs/xfs/xfs_super.c         | 188 +++++++++++++++++++++++++++++++
 include/linux/dax.h        |  80 ++++++++++++-
 include/linux/memremap.h   |   9 ++
 include/linux/mm.h         |   2 +
 mm/memory-failure.c        | 225 ++++++++++++++++++++++++++-----------
 12 files changed, 643 insertions(+), 130 deletions(-)