| Message ID | 20241011011630.2002803-1-yukuai1@huaweicloud.com (mailing list archive) |
|---|---|
| Headers | show |
| Series | md: enhance faulty checking for blocked handling | expand |
Mariusz, you have run some tests on v1, but didn't give your Tested-by tag. Would you mind rerun the test and reply with the tag? Thanks, Song On Thu, Oct 10, 2024 at 6:18 PM Yu Kuai <yukuai1@huaweicloud.com> wrote: > > From: Yu Kuai <yukuai3@huawei.com> > > Changes in v2: > - add more comments and commit message in patch 3; > - fix some typo; > > The lifetime of badblocks: > > 1) IO error, and decide to record badblocks, and record sb_flags; > 2) write IO found rdev has badblocks and not yet acknowledged, then this > IO is blocked; > 3) daemon found sb_flags is set, update superblock and flush badblocks; > 4) write IO continue; > > Main idea is that badblocks will be set in memory fist, before badblocks > are acknowledged, new write request must be blocked to prevent reading > old data after power failure, and this behaviour is not necessary if rdev > is faulty in the first place. > > Yu Kuai (7): > md: add a new helper rdev_blocked() > md: don't wait faulty rdev in md_wait_for_blocked_rdev() > md: don't record new badblocks for faulty rdev > md/raid1: factor out helper to handle blocked rdev from > raid1_write_request() > md/raid1: don't wait for Faulty rdev in wait_blocked_rdev() > md/raid10: don't wait for Faulty rdev in wait_blocked_rdev() > md/raid5: don't set Faulty rdev for blocked_rdev > > drivers/md/md.c | 15 +++++++-- > drivers/md/md.h | 24 +++++++++++++++ > drivers/md/raid1.c | 75 +++++++++++++++++++++++---------------------- > drivers/md/raid10.c | 40 +++++++++++------------- > drivers/md/raid5.c | 13 ++++---- > 5 files changed, 99 insertions(+), 68 deletions(-) > > -- > 2.39.2 >
On Thu, 17 Oct 2024 23:46:58 -0700 Song Liu <song@kernel.org> wrote: > Mariusz, you have run some tests on v1, but didn't give your > Tested-by tag. Would you mind rerun the test and reply with > the tag? > > Thanks, > Song Hi Song, I see no functional difference between v1 and v2 feel free to add it. I will be hard to rerun these tests right now. Tested-by: Mariusz Tkaczyk <mariusz.tkaczyk@linux.intel.com> Thanks, Mariusz
On Fri, Oct 18, 2024 at 5:56 AM Mariusz Tkaczyk <mariusz.tkaczyk@linux.intel.com> wrote: > > On Thu, 17 Oct 2024 23:46:58 -0700 > Song Liu <song@kernel.org> wrote: > > > Mariusz, you have run some tests on v1, but didn't give your > > Tested-by tag. Would you mind rerun the test and reply with > > the tag? > > > > Thanks, > > Song > > Hi Song, > I see no functional difference between v1 and v2 feel free to add it. > I will be hard to rerun these tests right now. > > Tested-by: Mariusz Tkaczyk <mariusz.tkaczyk@linux.intel.com> Applied to md-6.13 branch. Thanks, Song
From: Yu Kuai <yukuai3@huawei.com> Changes in v2: - add more comments and commit message in patch 3; - fix some typo; The lifetime of badblocks: 1) IO error, and decide to record badblocks, and record sb_flags; 2) write IO found rdev has badblocks and not yet acknowledged, then this IO is blocked; 3) daemon found sb_flags is set, update superblock and flush badblocks; 4) write IO continue; Main idea is that badblocks will be set in memory fist, before badblocks are acknowledged, new write request must be blocked to prevent reading old data after power failure, and this behaviour is not necessary if rdev is faulty in the first place. Yu Kuai (7): md: add a new helper rdev_blocked() md: don't wait faulty rdev in md_wait_for_blocked_rdev() md: don't record new badblocks for faulty rdev md/raid1: factor out helper to handle blocked rdev from raid1_write_request() md/raid1: don't wait for Faulty rdev in wait_blocked_rdev() md/raid10: don't wait for Faulty rdev in wait_blocked_rdev() md/raid5: don't set Faulty rdev for blocked_rdev drivers/md/md.c | 15 +++++++-- drivers/md/md.h | 24 +++++++++++++++ drivers/md/raid1.c | 75 +++++++++++++++++++++++---------------------- drivers/md/raid10.c | 40 +++++++++++------------- drivers/md/raid5.c | 13 ++++---- 5 files changed, 99 insertions(+), 68 deletions(-)