md/raid10: fix missing discard IO accounting

md_account_bio() is not called from raid10_handle_discard(), now that we
handle bitmap inside md_account_bio(), also fix missing
bitmap_startwrite for di

md/raid10: fix missing discard IO accounting

md_account_bio() is not called from raid10_handle_discard(), now that we
handle bitmap inside md_account_bio(), also fix missing
bitmap_startwrite for discard.

Test whole disk discard for 20G raid10:

Before:
Device d/s dMB/s drqm/s %drqm d_await dareq-sz
md0 48.00 16.00 0.00 0.00 5.42 341.33

After:
Device d/s dMB/s drqm/s %drqm d_await dareq-sz
md0 68.00 20462.00 0.00 0.00 2.65 308133.65

Link: https://lore.kernel.org/linux-raid/[email protected]
Fixes: 528bc2cf2fcc ("md/raid10: enable io accounting")
Signed-off-by: Yu Kuai <[email protected]>
Acked-by: Coly Li <[email protected]>

show more ...

badblocks: use sector_t instead of int to avoid truncation of badblocks length

There is a truncation of badblocks length issue when set badblocks as
follow:

echo "2055 4294967299" > bad_blocks
cat

badblocks: use sector_t instead of int to avoid truncation of badblocks length

There is a truncation of badblocks length issue when set badblocks as
follow:

echo "2055 4294967299" > bad_blocks
cat bad_blocks
2055 3

Change 'sectors' argument type from 'int' to 'sector_t'.

This change avoids truncation of badblocks length for large sectors by
replacing 'int' with 'sector_t' (u64), enabling proper handling of larger
disk sizes and ensuring compatibility with 64-bit sector addressing.

Fixes: 9e0e252a048b ("badblocks: Add core badblock management code")
Signed-off-by: Zheng Qixing <[email protected]>
Reviewed-by: Yu Kuai <[email protected]>
Acked-by: Coly Li <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>

show more ...

md: improve return types of badblocks handling functions

rdev_set_badblocks() only indicates success/failure, so convert its return
type from int to boolean for better semantic clarity.

rdev_clear_

md: improve return types of badblocks handling functions

rdev_set_badblocks() only indicates success/failure, so convert its return
type from int to boolean for better semantic clarity.

rdev_clear_badblocks() return value is never used by any caller, convert it
to void. This removes unnecessary value returns.

Also update narrow_write_error() in both raid1 and raid10 to use boolean
return type to match rdev_set_badblocks().

Signed-off-by: Zheng Qixing <[email protected]>
Reviewed-by: Yu Kuai <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>

show more ...

md/raid10: wait barrier before returning discard request with REQ_NOWAIT

raid10_handle_discard should wait barrier before returning a discard bio
which has REQ_NOWAIT. And there is no need to print

md/raid10: wait barrier before returning discard request with REQ_NOWAIT

raid10_handle_discard should wait barrier before returning a discard bio
which has REQ_NOWAIT. And there is no need to print warning calltrace
if a discard bio has REQ_NOWAIT flag. Quality engineer usually checks
dmesg and reports error if dmesg has warning/error calltrace.

Fixes: c9aa889b035f ("md: raid10 add nowait support")
Signed-off-by: Xiao Ni <[email protected]>
Acked-by: Coly Li <[email protected]>
Link: https://lore.kernel.org/linux-raid/[email protected]
Signed-off-by: Yu Kuai <[email protected]>

show more ...

md/raid1,raid10: don't ignore IO flags

If blk-wbt is enabled by default, it's found that raid write performance
is quite bad because all IO are throttled by wbt of underlying disks,
due to flag REQ_

md/raid1,raid10: don't ignore IO flags

If blk-wbt is enabled by default, it's found that raid write performance
is quite bad because all IO are throttled by wbt of underlying disks,
due to flag REQ_IDLE is ignored. And turns out this behaviour exist since
blk-wbt is introduced.

Other than REQ_IDLE, other flags should not be ignored as well, for
example REQ_META can be set for filesystems, clearing it can cause priority
reverse problems; And REQ_NOWAIT should not be cleared as well, because
io will wait instead of failing directly in underlying disks.

Fix those problems by keep IO flags from master bio.

Fises: f51d46d0e7cb ("md: add support for REQ_NOWAIT")
Fixes: e34cbd307477 ("blk-wbt: add general throttling mechanism")
Fixes: 5404bc7a87b9 ("[PATCH] Allow file systems to differentiate between data and meta reads")
Link: https://lore.kernel.org/linux-raid/[email protected]
Signed-off-by: Yu Kuai <[email protected]>

show more ...

md: don't export md_cluster_ops

Add a new field 'cluster_ops' and initialize it md_setup_cluster(), so
that the gloable variable 'md_cluter_ops' doesn't need to be exported.
Also prepare to switch m

md: don't export md_cluster_ops

Add a new field 'cluster_ops' and initialize it md_setup_cluster(), so
that the gloable variable 'md_cluter_ops' doesn't need to be exported.
Also prepare to switch md-cluster to use md_submod_head.

Link: https://lore.kernel.org/linux-raid/[email protected]
Signed-off-by: Yu Kuai <[email protected]>
Reviewed-by: Su Yue <[email protected]>

show more ...

md: switch personalities to use md_submodule_head

Remove the global list 'pers_list', and switch to use md_submodule_head,
which is managed by xarry. Prepare to unify registration and unregistration

md: switch personalities to use md_submodule_head

Remove the global list 'pers_list', and switch to use md_submodule_head,
which is managed by xarry. Prepare to unify registration and unregistration
for all sub modules.

Link: https://lore.kernel.org/linux-raid/[email protected]
Signed-off-by: Yu Kuai <[email protected]>

show more ...

md: only include md-cluster.h if necessary

md-cluster is only supportted by raid1 and raid10, there is no need to
include md-cluster.h for other personalities.

Also move APIs that is only used in m

md: only include md-cluster.h if necessary

md-cluster is only supportted by raid1 and raid10, there is no need to
include md-cluster.h for other personalities.

Also move APIs that is only used in md-cluster.c from md.h to
md-cluster.h.

Link: https://lore.kernel.org/linux-raid/[email protected]
Signed-off-by: Yu Kuai <[email protected]>
Reviewed-by: Su Yue <[email protected]>

show more ...

md/raid*: Fix the set_queue_limits implementations

queue_limits_cancel_update() must only be called if
queue_limits_start_update() is called first. Remove the
queue_limits_cancel_update() calls from

md/raid*: Fix the set_queue_limits implementations

queue_limits_cancel_update() must only be called if
queue_limits_start_update() is called first. Remove the
queue_limits_cancel_update() calls from the raid*_set_limits() functions
because there is no corresponding queue_limits_start_update() call.

Cc: Christoph Hellwig <[email protected]>
Fixes: c6e56cf6b2e7 ("block: move integrity information into queue_limits")
Signed-off-by: Bart Van Assche <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Link: https://lore.kernel.org/linux-raid/[email protected]/
Signed-off-by: Yu Kuai <[email protected]>

show more ...

block: Add common atomic writes enable flag

Currently only stacked devices need to explicitly enable atomic writes by
setting BLK_FEAT_ATOMIC_WRITES_STACKED flag.

This does not work well for device

block: Add common atomic writes enable flag

Currently only stacked devices need to explicitly enable atomic writes by
setting BLK_FEAT_ATOMIC_WRITES_STACKED flag.

This does not work well for device mapper stacking devices, as there many
sets of limits are stacked and what is the 'bottom' and 'top' device can
swapped. This means that BLK_FEAT_ATOMIC_WRITES_STACKED needs to be set
for many queue limits, which is messy.

Generalize enabling atomic writes enabling by ensuring that all devices
must explicitly set a flag - that includes NVMe, SCSI sd, and md raid.

Signed-off-by: John Garry <[email protected]>
Reviewed-by: Mike Snitzer <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>

show more ...

md/md-bitmap: move bitmap_{start, end}write to md upper layer

There are two BUG reports that raid5 will hang at
bitmap_startwrite([1],[2]), root cause is that bitmap start write and end
write is unb

md/md-bitmap: move bitmap_{start, end}write to md upper layer

There are two BUG reports that raid5 will hang at
bitmap_startwrite([1],[2]), root cause is that bitmap start write and end
write is unbalanced, it's not quite clear where, and while reviewing raid5
code, it's found that bitmap operations can be optimized. For example,
for a 4 disks raid5, with chunksize=8k, if user issue a IO (0 + 48k) to
the array:

┌────────────────────────────────────────────────────────────┐
│chunk 0 │
│ ┌────────────┬─────────────┬─────────────┬────────────┼
│ sh0 │A0: 0 + 4k │A1: 8k + 4k │A2: 16k + 4k │A3: P │
│ ┼────────────┼─────────────┼─────────────┼────────────┼
│ sh1 │B0: 4k + 4k │B1: 12k + 4k │B2: 20k + 4k │B3: P │
┼──────┴────────────┴─────────────┴─────────────┴────────────┼
│chunk 1 │
│ ┌────────────┬─────────────┬─────────────┬────────────┤
│ sh2 │C0: 24k + 4k│C1: 32k + 4k │C2: P │C3: 40k + 4k│
│ ┼────────────┼─────────────┼─────────────┼────────────┼
│ sh3 │D0: 28k + 4k│D1: 36k + 4k │D2: P │D3: 44k + 4k│
└──────┴────────────┴─────────────┴─────────────┴────────────┘

Before this patch, 4 stripe head will be used, and each sh will attach
bio for 3 disks, and each attached bio will trigger
bitmap_startwrite() once, which means total 12 times.
- 3 times (0 + 4k), for (A0, A1 and A2)
- 3 times (4 + 4k), for (B0, B1 and B2)
- 3 times (8 + 4k), for (C0, C1 and C3)
- 3 times (12 + 4k), for (D0, D1 and D3)

After this patch, md upper layer will calculate that IO range (0 + 48k)
is corresponding to the bitmap (0 + 16k), and call bitmap_startwrite()
just once.

Noted that this patch will align bitmap ranges to the chunks, for example,
if user issue a IO (0 + 4k) to array:

- Before this patch, 1 time (0 + 4k), for A0;
- After this patch, 1 time (0 + 8k) for chunk 0;

Usually, one bitmap bit will represent more than one disk chunk, and this
doesn't have any difference. And even if user really created a array
that one chunk contain multiple bits, the overhead is that more data
will be recovered after power failure.

Also remove STRIPE_BITMAP_PENDING since it's not used anymore.

[1] https://lore.kernel.org/all/CAJpMwyjmHQLvm6zg1cmQErttNNQPDAAXPKM3xgTjMhbfts986Q@mail.gmail.com/
[2] https://lore.kernel.org/all/[email protected]/

Signed-off-by: Yu Kuai <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Song Liu <[email protected]>

show more ...

md/md-bitmap: remove the last parameter for bimtap_ops->endwrite()

For the case that IO failed for one rdev, the bit will be mark as NEEDED
in following cases:

1) If badblocks is set and rdev is no

md/md-bitmap: remove the last parameter for bimtap_ops->endwrite()

For the case that IO failed for one rdev, the bit will be mark as NEEDED
in following cases:

1) If badblocks is set and rdev is not faulty;
2) If rdev is faulty;

Case 1) is useless because synchronize data to badblocks make no sense.
Case 2) can be replaced with mddev->degraded.

Also remove R1BIO_Degraded, R10BIO_Degraded and STRIPE_DEGRADED since
case 2) no longer use them.

Signed-off-by: Yu Kuai <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Song Liu <[email protected]>

show more ...

md/md-bitmap: factor behind write counters out from bitmap_{start/end}write()

behind_write is only used in raid1, prepare to refactor
bitmap_{start/end}write(), there are no functional changes.

Sig

md/md-bitmap: factor behind write counters out from bitmap_{start/end}write()

behind_write is only used in raid1, prepare to refactor
bitmap_{start/end}write(), there are no functional changes.

Signed-off-by: Yu Kuai <[email protected]>
Reviewed-by: Xiao Ni <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Song Liu <[email protected]>

show more ...

md/raid10: Atomic write support

Set BLK_FEAT_ATOMIC_WRITES_STACKED to enable atomic writes.

For an attempt to atomic write to a region which has bad blocks, error
the write as we just cannot do thi

md/raid10: Atomic write support

Set BLK_FEAT_ATOMIC_WRITES_STACKED to enable atomic writes.

For an attempt to atomic write to a region which has bad blocks, error
the write as we just cannot do this. It is unlikely to find devices which
support atomic writes and bad blocks.

Reviewed-by: Yu Kuai <[email protected]>
Signed-off-by: John Garry <[email protected]>
Reviewed-by: Martin K. Petersen <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>

show more ...

md/raid10: Handle bio_split() errors

Add proper bio_split() error handling. For any error, call
raid_end_bio_io() and return. Except for discard, where we end the bio
directly.

Reviewed-by: Yu Kuai

md/raid10: Handle bio_split() errors

Add proper bio_split() error handling. For any error, call
raid_end_bio_io() and return. Except for discard, where we end the bio
directly.

Reviewed-by: Yu Kuai <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Signed-off-by: John Garry <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Jens Axboe <[email protected]>

show more ...

md/raid10: don't wait for Faulty rdev in wait_blocked_rdev()

Faulty rdev should never be accessed anymore, hence there is no point to
wait for bad block to be acknowledged in this case while handlin

md/raid10: don't wait for Faulty rdev in wait_blocked_rdev()

Faulty rdev should never be accessed anymore, hence there is no point to
wait for bad block to be acknowledged in this case while handling write
request.

Signed-off-by: Yu Kuai <[email protected]>
Tested-by: Mariusz Tkaczyk <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Song Liu <[email protected]>

show more ...

md/raid10: fix null ptr dereference in raid10_size()

In raid10_run() if raid10_set_queue_limits() succeed, the return value
is set to zero, and if following procedures failed raid10_run() will
retur

md/raid10: fix null ptr dereference in raid10_size()

In raid10_run() if raid10_set_queue_limits() succeed, the return value
is set to zero, and if following procedures failed raid10_run() will
return zero while mddev->private is still NULL, causing null ptr
dereference in raid10_size().

Fix the problem by only overwrite the return value if
raid10_set_queue_limits() failed.

Fixes: 3d8466ba68d4 ("md/raid10: use the atomic queue limit update APIs")
Cc: [email protected]
Reported-and-tested-by: ValdikSS <[email protected]>
Closes: https://lore.kernel.org/all/[email protected]/
Signed-off-by: Yu Kuai <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Song Liu <[email protected]>

show more ...

md/md-bitmap: merge md_bitmap_resize() into bitmap_operations

So that the implementation won't be exposed, and it'll be possible
to invent a new bitmap by replacing bitmap_operations.

Signed-off-by

md/md-bitmap: merge md_bitmap_resize() into bitmap_operations

So that the implementation won't be exposed, and it'll be possible
to invent a new bitmap by replacing bitmap_operations.

Signed-off-by: Yu Kuai <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Song Liu <[email protected]>

show more ...

md/md-bitmap: pass in mddev directly for md_bitmap_resize()

And move the condition "if (mddev->bitmap)" into md_bitmap_resize() as
well, on the one hand make code cleaner, on the other hand try not

md/md-bitmap: pass in mddev directly for md_bitmap_resize()

And move the condition "if (mddev->bitmap)" into md_bitmap_resize() as
well, on the one hand make code cleaner, on the other hand try not to
access bitmap directly.

Since we are here, also change the parameter 'init' from int to bool.

Signed-off-by: Yu Kuai <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Song Liu <[email protected]>

show more ...

md/md-bitmap: merge md_bitmap_unplug_async() into md_bitmap_unplug()

Add a parameter 'bool sync' to distinguish them, and
md_bitmap_unplug_async() won't be exported anymore, hence
bitmap_operations

md/md-bitmap: merge md_bitmap_unplug_async() into md_bitmap_unplug()

Add a parameter 'bool sync' to distinguish them, and
md_bitmap_unplug_async() won't be exported anymore, hence
bitmap_operations only need one op to cover them.

Signed-off-by: Yu Kuai <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Song Liu <[email protected]>

show more ...

md/md-bitmap: merge md_bitmap_cond_end_sync() into bitmap_operations

So that the implementation won't be exposed, and it'll be possible
to invent a new bitmap by replacing bitmap_operations.

Also c

md/md-bitmap: merge md_bitmap_cond_end_sync() into bitmap_operations

So that the implementation won't be exposed, and it'll be possible
to invent a new bitmap by replacing bitmap_operations.

Also change the parameter from bitmap to mddev, to avoid access
bitmap outside md-bitmap.c as much as possible.

Signed-off-by: Yu Kuai <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Song Liu <[email protected]>

show more ...

md/md-bitmap: merge md_bitmap_close_sync() into bitmap_operations

So that the implementation won't be exposed, and it'll be possible
to invent a new bitmap by replacing bitmap_operations.

Also chan

md/md-bitmap: merge md_bitmap_close_sync() into bitmap_operations

So that the implementation won't be exposed, and it'll be possible
to invent a new bitmap by replacing bitmap_operations.

Also change the parameter from bitmap to mddev, to avoid access
bitmap outside md-bitmap.c as much as possible.

Signed-off-by: Yu Kuai <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Song Liu <[email protected]>

show more ...

md/md-bitmap: merge md_bitmap_end_sync() into bitmap_operations

So that the implementation won't be exposed, and it'll be possible
to invent a new bitmap by replacing bitmap_operations.

Also change

md/md-bitmap: merge md_bitmap_end_sync() into bitmap_operations

So that the implementation won't be exposed, and it'll be possible
to invent a new bitmap by replacing bitmap_operations.

Also change the parameter from bitmap to mddev, to avoid access
bitmap outside md-bitmap.c as much as possible.

Signed-off-by: Yu Kuai <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Song Liu <[email protected]>

show more ...

md/md-bitmap: remove the parameter 'aborted' for md_bitmap_end_sync()

For internal callers, aborted are always set to false, while for
external callers, aborted are always set to true.

Hence there

md/md-bitmap: remove the parameter 'aborted' for md_bitmap_end_sync()

For internal callers, aborted are always set to false, while for
external callers, aborted are always set to true.

Hence there is no need to always pass in true for exported api.

Signed-off-by: Yu Kuai <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Song Liu <[email protected]>

show more ...

md/md-bitmap: merge md_bitmap_start_sync() into bitmap_operations

So that the implementation won't be exposed, and it'll be possible
to invent a new bitmap by replacing bitmap_operations.

Also chan

md/md-bitmap: merge md_bitmap_start_sync() into bitmap_operations

So that the implementation won't be exposed, and it'll be possible
to invent a new bitmap by replacing bitmap_operations.

Also change the parameter from bitmap to mddev, to avoid access
bitmap outside md-bitmap.c as much as possible.

Also fix lots of code style.

Signed-off-by: Yu Kuai <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: Song Liu <[email protected]>

show more ...