[PATCH v5 0/9] fsdax: introduce fs query to support reflink

From: Shiyang Ruan
Date: Sun Jun 27 2021 - 20:03:04 EST


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

Change from V4:
- Add nr_pfns for ->memory_failure() to support range based notification
- Remove struct bdev in dax_holder_notify_failure()
- Add rwsem for dax_holder
- Rename functions to (*_)notify_failure()
- Remove sb->corrupted_range(), implement holder_ops in filesystem and
MD driver instead
- Reorganize the patchset, make it easy to review
- Rebased to v5.13-rc7

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()
| - md_dax_notify_failure()
| |* xfs_dax_notify_failure()
| |--------------------------
| | xfs_rmap_query_range()
| | xfs_currupt_helper()
| | * corrupted on metadata
| | try to recover data, call xfs_force_shutdown()
| | * corrupted on file data
| | try to recover data, call mf_dax_kill_procs()
| |* md_dax_notify_failure()
| |-------------------------
| md_targets->iterate_devices()
| md_targets->rmap() => linear_rmap()
| dax_holder_notify_failure()
|* normal case
|-------------
mf_generic_kill_procs()

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

(Rebased on v5.13-rc7)
==

Shiyang Ruan (9):
pagemap: Introduce ->memory_failure()
dax: Introduce holder for dax_device
mm: factor helpers for memory_failure_dev_pagemap
pmem,mm: Implement ->memory_failure in pmem driver
mm: Introduce mf_dax_kill_procs() for fsdax case
xfs: Implement ->corrupted_range() for XFS
dm: Introduce ->rmap() to find bdev offset
md: Implement dax_holder_operations
fs/dax: Remove useless functions

block/genhd.c | 30 +++++++
drivers/dax/super.c | 49 ++++++++++
drivers/md/dm-linear.c | 20 +++++
drivers/md/dm.c | 126 +++++++++++++++++++++++++-
drivers/nvdimm/pmem.c | 13 +++
fs/dax.c | 73 ++++-----------
fs/xfs/xfs_fsops.c | 5 ++
fs/xfs/xfs_mount.h | 1 +
fs/xfs/xfs_super.c | 140 +++++++++++++++++++++++++++++
include/linux/dax.h | 27 ++++++
include/linux/device-mapper.h | 5 ++
include/linux/genhd.h | 1 +
include/linux/memremap.h | 9 ++
include/linux/mm.h | 10 +++
mm/memory-failure.c | 165 ++++++++++++++++++++++------------
15 files changed, 564 insertions(+), 110 deletions(-)

--
2.32.0