linux-stable.git/drivers/md/md-bitmap.c, branch v6.6.2

md/md-bitmap: hold 'reconfig_mutex' in backlog_store()

2023-07-27T07:13:30+00:00

Several reasons why 'reconfig_mutex' should be held:

1) rdev_for_each() is not safe to be called without the lock, because
   rdev can be removed concurrently.
2) mddev_destroy_serial_pool() and mddev_create_serial_pool() should not
   be called concurrently.
3) mddev_suspend() from mddev_destroy/create_serial_pool() should be
   protected by the lock.

Fixes: 10c92fca636e ("md-bitmap: create and destroy wb_info_pool with the change of backlog")
Signed-off-by: Yu Kuai 
Link: https://lore.kernel.org/r/20230706083727.608914-3-yukuai1@huaweicloud.com
Signed-off-by: Song Liu

md/md-bitmap: remove unnecessary local variable in backlog_store()

2023-07-27T07:13:30+00:00

Local variable is definied first in the beginning of backlog_store(),
there is no need to define it again.

Fixes: 8c13ab115b57 ("md/bitmap: don't set max_write_behind if there is no write mostly device")
Signed-off-by: Yu Kuai 
Link: https://lore.kernel.org/r/20230706083727.608914-2-yukuai1@huaweicloud.com
Signed-off-by: Song Liu

md: make bitmap file support optional

2023-07-27T07:13:29+00:00

The support for write intent bitmaps in files on an external files in md
is a hot mess that abuses ->bmap to map file offsets into physical device
objects, and also abuses buffer_heads in a creative way.

Make this code optional so that MD can be built into future kernels
without buffer_head support, and so that we can eventually deprecate it.

Note this does not affect the internal bitmap support, which has none of
the problems.

Signed-off-by: Christoph Hellwig 
Reviewed-by: Hannes Reinecke 
Reviewed-by: Himanshu Madhani 
Signed-off-by: Song Liu 
Link: https://lore.kernel.org/r/20230615064840.629492-11-hch@lst.de

md-bitmap: don't use ->index for pages backing the bitmap file

2023-07-27T07:13:29+00:00

The md driver allocates pages for storing the bitmap file data, which
are not page cache pages, and then stores the page granularity file
offset in page->index, which is a field that isn't really valid except
for page cache pages.

Use a separate index for the superblock, and use the scheme used at
read size to recalculate the index for the bitmap pages instead.

Signed-off-by: Christoph Hellwig 
Signed-off-by: Song Liu 
Link: https://lore.kernel.org/r/20230615064840.629492-10-hch@lst.de

md-bitmap: account for mddev->bitmap_info.offset in read_sb_page

2023-07-27T07:13:29+00:00

Diretly apply mddev->bitmap_info.offset to the sector number to read
instead of doing that in both callers.

Signed-off-by: Christoph Hellwig 
Reviewed-by: Hannes Reinecke 
Reviewed-by: Himanshu Madhani 
Signed-off-by: Song Liu 
Link: https://lore.kernel.org/r/20230615064840.629492-9-hch@lst.de

md-bitmap: cleanup read_sb_page

2023-07-27T07:13:29+00:00

Convert read_sb_page to the normal kernel coding style, calculate the
target sector only once, and add a local iosize variable to make the call
to sync_page_io more readable.

Signed-off-by: Christoph Hellwig 
Reviewed-by: Hannes Reinecke 
Reviewed-by: Johannes Thumshirn 
Reviewed-by: Himanshu Madhani 
Signed-off-by: Song Liu 
Link: https://lore.kernel.org/r/20230615064840.629492-8-hch@lst.de

md-bitmap: refactor md_bitmap_init_from_disk

2023-07-27T07:13:29+00:00

Split the confusing loop in md_bitmap_init_from_disk that iterates over
all chunks but also needs to read and map the pages into three separate
loops: one that iterates over the pages to read them, a second optional
one to iterate over the pages to mark them invalid if the bitmaps are
out of date, and a final one that actually iterates over the chunks.

Reported-by: kernel test robot 
Closes: https://lore.kernel.org/oe-kbuild-all/202306160552.smw0qbmb-lkp@intel.com/
Signed-off-by: Christoph Hellwig 
Signed-off-by: Song Liu 
Link: https://lore.kernel.org/r/20230615064840.629492-7-hch@lst.de

md-bitmap: rename read_page to read_file_page

2023-07-27T07:13:29+00:00

Make the difference to read_sb_page clear.

Signed-off-by: Christoph Hellwig 
Reviewed-by: Hannes Reinecke 
Reviewed-by: Johannes Thumshirn 
Reviewed-by: Himanshu Madhani 
Signed-off-by: Song Liu 
Link: https://lore.kernel.org/r/20230615064840.629492-6-hch@lst.de

md-bitmap: split file writes into a separate helper

2023-07-27T07:13:29+00:00

Split the file write code out of write_page into a separate helper.

Signed-off-by: Christoph Hellwig 
Reviewed-by: Hannes Reinecke 
Reviewed-by: Johannes Thumshirn 
Reviewed-by: Himanshu Madhani 
Signed-off-by: Song Liu 
Link: https://lore.kernel.org/r/20230615064840.629492-5-hch@lst.de

md-bitmap: use %pD to print the file name in md_bitmap_file_kick

2023-07-27T07:13:29+00:00

Don't bother allocating an extra buffer in the I/O failure handler and
instead use the printk built-in format to print the last 4 path name
components.

Signed-off-by: Christoph Hellwig 
Reviewed-by: Hannes Reinecke 
Reviewed-by: Johannes Thumshirn 
Reviewed-by: Himanshu Madhani 
Signed-off-by: Song Liu 
Link: https://lore.kernel.org/r/20230615064840.629492-4-hch@lst.de