Re: [PATCH 0/3] md: raid helper cleanups

在 2026/8/27 15:56, Li Youhong 写道:

From: Li Youhong <redacted>

This series factors a few repeated md/raid patterns into small helpers.
There is no intentional behaviour change; each patch is a mechanical
extract of existing logic.

Patch 1 adds rdev_record_write_error() in md.h for the
WriteErrorSeen + WantReplacement + MD_RECOVERY_NEEDED sequence shared by
raid1, raid10 and raid5.

Patches 2 and 3 move raid1/raid10-only helpers into raid1-10.c, which is
#included by both drivers:

   - rdev_bb_block_sectors() for the badblocks-aligned block size used in
     narrow_write_error()
   - flush_bio_list() for the bitmap plug flush path previously duplicated
     in raid10's flush_pending_writes() and raid10_unplug()

The helpers in raid1-10.c follow the existing static inline convention
used by raid1_submit_write(), check_decay_read_errors(), and others in
that file.

Li Youhong (3):
       md: add rdev_record_write_error() helper
       md/raid1-10: add rdev_bb_block_sectors() helper
       md/raid1-10: share flush_bio_list between raid1 and raid10
---
  drivers/md/md.h       |  7 +++++++
  drivers/md/raid1-10.c | 27 +++++++++++++++++++++++++++
  drivers/md/raid1.c    | 42 +++++++-----------------------------------
  drivers/md/raid10.c   | 50 +++++++-------------------------------------------
  drivers/md/raid5.c    |  5 +----
  5 files changed, 48 insertions(+), 83 deletions(-)

Reviewed-by: Yu Kuai <yukuai@fygo.io>

Noted only bug fix can be accepted for current merge window, cleanup patches
will be applied later for 6.4.

-- 
Thanks,
Kuai