btrfs: scrub: use a more reader friendly code to implement scrub_simple_mirror()
Get series with b4
b4 am [email protected]Versions newest first · linked by author and title
- v8 Latest12 patches 7 repliesIn review
- v713 patches 18 repliesSuperseded
- v613 patches 0 repliesSuperseded
Earlier versions · 5
| Version | Posted | Patches | Replies | Status |
|---|---|---|---|---|
| v5 | 13 patches | 8 replies | Superseded | |
| v4 | 13 patches | 7 replies | Superseded | |
| v3 | 12 patches | 23 replies | Superseded | |
| v2 | 12 patches | 19 replies | Superseded | |
| v1 | 11 patches | 5 replies | Superseded |
Changes in v8 from the cover letter
- Better bio layer interface
* No special scrub read helper
Just go btrfs_submit_bio() for read bios
* Shared code base for repair writes
A common helper, btrfs_map_repair_block(), is shared between read-repair and scrub.
* Some code style enhancement related to bio layer
- Full RAID56 code cleanup
A huge -2500 lines cleanup, although not done in this series.
Patches in v8 No applies recorded yet
Reviews trailers per patch
2 of 12 patches have a Reviewed-by or Acked-by.
| Patch | Christoph Hellwig | Anand Jain | Johannes Thumshirn |
|---|---|---|---|
| 1/12 | R-b | R-b | R-b |
| 2/12 | R-b | ||
| 3/12 | |||
| 4/12 | |||
| 5/12 | |||
| 6/12 | |||
| 7/12 | |||
| 8/12 | |||
| 9/12 | |||
| 10/12 | |||
| 11/12 | |||
| 12/12 |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes. Scroll the table sideways for all 3 reviewers.
Cover letter [PATCH v8 00/12] btrfs: scrub: use a more reader friendly code to implement scrub_simple_mirror()
This series can be found in my github repo:
https://github.com/adam900710/linux/tree/scrub_stripe
It's recommended to fetch from the repo, as our misc-next seems to change pretty rapidly.
[Changelog]
v8:
- Better bio layer interface
* No special scrub read helper
Just go btrfs_submit_bio() for read bios
* Shared code base for repair writes
A common helper, btrfs_map_repair_block(), is shared between read-repair and scrub.
* Some code style enhancement related to bio layer
- Full RAID56 code cleanup
A huge -2500 lines cleanup, although not done in this series.
The cleanup is done in an dedicated series: "btrfs: scrub: finish the switch to scrub_stripe and cleanup the old code".
Can be fetched from github repo.
v7:
- Fix a bug that scrub_stripe::extent_sector_bitmap is not cleared Exposed during my development for RAID56 new scrub code.
Extent_sector_bitmap indicates whether the sector is utilized by any extent.
If that bitmap is not cleared before running the next stripe, we can treat unused sectors as NODATASUM data.
This is not a big deal for non-RAID56 profiles, as they skip empty stripe through other methods.
But can be very problematic for RAID56, as they need to scrub data stripes then P/Q stripes.
Such inherited bitmap makes RAID56 to scrub all those unused full stripes and greatly slow down the scrub.
v6:
- Fix a bug in zoned block group repair
Exposed during my development for RAID56 new scrub code.
There is a bug that we may use @stripe to determine if we need to queue a block group for repair.
But @stripe is the last stripe we checked, it may not have any error.
The correct way is to go through all the stripes and queue the repair if we found any error.
v5:
- Fix a bug that unconditionally repairs a zoned block group Only trigger the repair if we had any initial read failure
Huge thanks to Johannes for the initial ZNS tests.
v4:
- Add a dedicated patch to add btrfs_bio::fs_info
Along with dedicated allocator for scrub btrfs bios.
The dedicated allocator is due to the fact that scrub and regular btrfs bios have very different mandatory members (fs_info vs inode + file_offset).
For now I believe a different allocator would be better.
- Some code style change
* No more single letter temporray structure copied from old scrub code
* Use "for (int i = 0; ...)" when possible
* Some new lines fixes
* Extra brackets for tenrary operators
* A new macro for (BTRFS_STRIPE_LEN >> PAGE_SHIFT)
* Use enum for scrub_stripe::state bit flags
* Use extra brackets for double shifting
* Use explicit != 0 or == 0 comparing memcmp() results
* Remove unnecessary ASSERT()s after btrfs_bio allocation
v3:
- Add a dedicated @fs_info member for btrfs_bio
Unfortunately although we have a 32 bytes hole between @end_io_work and @bio, compiler still choose not to use that hole for whatever reasons. Thus this would increase the size of btrfs_bio by 4 bytes.
- Rebased to lastest misc-next
- Fix various while space error not caught by btrfs-workflow
v2:
- Use batched scrub_stripe submission
This allows much better performance compared to the old scrub code
- Add scrub specific bio layer helpers
This makes the scrub code to be completely rely on logical bytenr + mirror_num.
[PROBLEMS OF OLD SCRUB]
- Too many delayed jumps, making it hard to read
Even starting from scrub_simple_mirror(), we have the following functions:
scrub_extent()
|
v
scrub_sectors()
|
v
scrub_add_sector_to_rd_bio()
| endio function
v
scrub_bio_end_io()
| delayed work
v
scrub_bio_end_io_worker()
|
v
scrub_block_complete()
|
v
scrub_handle_errored_blocks()
|
v
scrub_recheck_block()
|
v scrub_repair_sector_from_good_copy()
Not to mention the hidden jumps in certain branches.
- IOPS inefficient for fragmented extents
The real block size of scrub read is between 4K and 128K. If the extents are not adjacent, the blocksize drops to 4K and would be an IOPS disaster.
- All hardcoded to do the logical -> physical mapping by scrub itself No usage of any existing bio facilities.
And even implemented a RAID56 recovery wrapper.
[NEW SCRUB_STRIPE BASED SOLUTION]
- Overall streamlined code base
queue_scrub_stripe()
|
v
scrub_find_fill_first_stripe()
|
vdone
Or
queue_scrub_stripe()
|
v
flush_scrub_stripes()
|
v
scrub_submit_initial_read()
| endio function
v
scrub_read_endio()
| delayed work
v
scrub_stripe_read_repair_worker()
|
v
scrub_verify_one_stripe()
|
v
scrub_stripe_submit_repair_read()
|
v
scrub_write_sectors()
|
v scrub_stripe_report_errors()
Only one endio and delayed work, all other work are properly done in a sequential workflow.
- Always read in 64KiB block size
The real blocksize of read starts at 64KiB, and ends at 512K. This already results a better performance even for the worst case:
With patchset: 404.81MiB/s
Without patchset: 369.30MiB/s
Around 10% performance improvement on an SATA SSD.
- All logical bytenr/mirror_num based read and write
With the new single stripe fast path in btrfs_submit_bio(), scrub can reuse most of the bio layer code, result much simpler scrub code.
[TODO]
- More testing on zoned devices
Now the patchset can already pass all scrub/replace/raid/repair groups with regular devices.
Johannes helped some zoned tests, but more zoned tests would still be very appreciated.
Qu Wenruo (12):
btrfs: scrub: use dedicated super block verification function to scrub
one super block
btrfs: introduce btrfs_bio::fs_info member
btrfs: introduce a new helper to submit write bio for repair
btrfs: scrub: introduce the structure for new BTRFS_STRIPE_LEN based
interface
btrfs: scrub: introduce a helper to find and fill the sector info for
a scrub_stripe
btrfs: scrub: introduce a helper to verify one metadata
btrfs: scrub: introduce a helper to verify one scrub_stripe
btrfs: scrub: introduce the main read repair worker for scrub_stripe
btrfs: scrub: introduce a writeback helper for scrub_stripe
btrfs: scrub: introduce error reporting functionality for scrub_stripe
btrfs: scrub: introduce the helper to queue a stripe for scrub
btrfs: scrub: switch scrub_simple_mirror() to scrub_stripe
infrastructure
fs/btrfs/bio.c | 138 ++-- fs/btrfs/bio.h | 14 +- fs/btrfs/compression.c | 3 +- fs/btrfs/extent_io.c | 3 +- fs/btrfs/file-item.c | 9 +- fs/btrfs/file-item.h | 3 +- fs/btrfs/inode.c | 13 +- fs/btrfs/raid56.c | 2 +- fs/btrfs/raid56.h | 5 + fs/btrfs/scrub.c | 1668 ++++++++++++++++++++++++++++------------ fs/btrfs/volumes.c | 73 ++ fs/btrfs/volumes.h | 3 + fs/btrfs/zoned.c | 4 +- 13 files changed, 1386 insertions(+), 552 deletions(-)
Signature · 1 line
2.39.2