Asynchronous raid6 acceleration (part 3 of 3)
Get series with b4
b4 am [email protected]Versions newest first · linked by author and title
- v2 Latest posting 2/29 patches 4 repliesIn review
- v2 posting 1/24 patches 0 repliesSuperseded
- v18 patches 2 repliesSuperseded
Patches in v2 posting 2/2 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/9 | md/raid5: factor out mark_uptodate from ops_complete_compute5drivers/ | +16 | 0 replies | |
| 2/9 | md/raid6: asynchronous raid6 operationsdrivers/ drivers/ drivers/ | +299 | 1 reply | |
| 3/9 | md/raid5, 6: common schedule_reconstruction for raid5/6drivers/ | +18 | 0 replies | |
| 4/9 | md/raid6: asynchronous handle_stripe_fill6drivers/ | +93 | 0 replies | |
| 5/9 | md/raid6: asynchronous handle_stripe_dirtying6drivers/ | +37 | 0 replies | |
| 6/9 | md/raid6: asynchronous handle_parity_check6drivers/ | +139 | 0 replies | |
| 7/9 | md/raid6: asynchronous handle_stripe6drivers/ | +82 | 2 replies | |
| 8/9 | md/raid6: remove synchronous infrastructuredrivers/ | +0 | 0 replies | |
| 9/9 | md/raid456: distribute raid processing over multiple coresdrivers/ drivers/ | +45 | 0 replies |
Reviews trailers per patch
1 of 9 patches has a Reviewed-by or Acked-by.
| Patch | Andre Noll |
|---|---|
| 1/9 | |
| 2/9 | R-b |
| 3/9 | |
| 4/9 | |
| 5/9 | |
| 6/9 | |
| 7/9 | |
| 8/9 | |
| 9/9 |
Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.
Cover letter [PATCH v2 0/9] Asynchronous raid6 acceleration (part 3 of 3)
Modify the raid6 driver to offload stripe operations to the async_tx api.
git://git.kernel.org/pub/scm/linux/kernel/git/djbw/async_tx.git raid6
The changes since v1 are:
1/ Use a percpu scribble buffer for passing the address lists and a
raid6 spare page to the async_tx api.
2/ Fixed a potential assertion in handle_parity_checks6() 3/ An experimental multicore raid456 option. Given the new percpu
scheme it is a straightforward conversion to have raid5d issue work to the async thread pool.
As before let me know if it is suitable to take this into linux-next via async_tx.git/next, or if you would rather it go through md.git/for-next.
Thanks,
Dan
---
Dan Williams (4):
md/raid456: distribute raid processing over multiple cores md/raid6: asynchronous handle_parity_check6 md/raid6: asynchronous raid6 operations
md/raid5: factor out mark_uptodate from ops_complete_compute5
Yuri Tikhonov (5):
md/raid6: remove synchronous infrastructure md/raid6: asynchronous handle_stripe6 md/raid6: asynchronous handle_stripe_dirtying6 md/raid6: asynchronous handle_stripe_fill6 md/raid5,6: common schedule_reconstruction for raid5/6
drivers/md/Kconfig | 13 + drivers/md/raid5.c | 1246 ++++++++++++++++++++++++++++++---------------------- drivers/md/raid5.h | 8 3 files changed, 728 insertions(+), 539 deletions(-)