linux-toradex.git/drivers/md/persistent-data, branch v4.9-rc3

dm array: introduce cursor api

2016-09-22T15:15:04+00:00

More efficient way to iterate an array due to prefetching (makes use of
the new dm_btree_cursor_* api).

Signed-off-by: Joe Thornber 
Signed-off-by: Mike Snitzer

dm btree: introduce cursor api

2016-09-22T15:15:04+00:00

This uses prefetching to speed up iteration through a btree.

Signed-off-by: Joe Thornber 
Signed-off-by: Mike Snitzer

dm array: add dm_array_new()

2016-09-22T15:12:23+00:00

dm_array_new() creates a new, populated array more efficiently than
starting with an empty one and resizing.

Signed-off-by: Joe Thornber 
Signed-off-by: Mike Snitzer

dm btree: fix a bug in dm_btree_find_next_single()

2016-07-20T16:43:34+00:00

dm_btree_find_next_single() can short-circuit the search for a block
with a return of -ENODATA if all entries are higher than the search key
passed to lower_bound().

This hasn't been a problem because of the way the btree has been used by
DM thinp.  But it must be fixed now in preparation for fixing the race
in DM thinp's handling of simultaneous block discard vs allocation.
Otherwise, once that fix is in place, some of the blocks in a discard
would not be unmapped as expected.

Signed-off-by: Joe Thornber 
Signed-off-by: Mike Snitzer

dm space map metadata: remove unused variable in brb_pop()

2015-12-14T14:26:01+00:00

Remove the unused struct block_op pointer that was inadvertantly
introduced, via cut-and-paste of previous brb_op() code, as part of
commit 50dd842ad.

(Cc'ing stable@ because commit 50dd842ad did)

Fixes: 50dd842ad ("dm space map metadata: fix ref counting bug when bootstrapping a new space map")
Reported-by: David Binderman 
Signed-off-by: Mike Snitzer 
Cc: stable@vger.kernel.org

dm btree: factor out need_insert() helper

2015-12-10T15:38:59+00:00

Eliminates code duplication within insert().

Signed-off-by: Mike Snitzer

dm bufio: store stacktrace in buffers to help find buffer leaks

2015-12-10T15:38:58+00:00

The option DM_DEBUG_BLOCK_STACK_TRACING is moved from persistent-data
directory to device mapper directory because it will now be used by
persistent-data and bufio.  When the option is enabled, each bufio buffer
stores the stacktrace of the last dm_bufio_get(), dm_bufio_read() or
dm_bufio_new() call that increased the hold count to 1.  The buffer's
stacktrace is printed if the buffer was not released before the bufio
client is destroyed.

When DM_DEBUG_BLOCK_STACK_TRACING is enabled, any bufio buffer leaks are
considered warnings - i.e. the kernel continues afterwards.  If not
enabled, buffer leaks are considered BUGs and the kernel with crash.
Reasoning on this disposition is: if we only ever warned on buffer leaks
users would generally ignore them and the problematic code would never
get fixed.

Successfully used to find source of bufio leaks fixed with commit
fce079f63c3 ("dm btree: fix bufio buffer leaks in dm_btree_del() error
path").

Signed-off-by: Mikulas Patocka 
Signed-off-by: Mike Snitzer

dm block manager: cleanup code that prints stacktrace

2015-12-10T15:38:56+00:00

There is no need to record stack trace and immediately print it.  Just
use dump_stack() to print the current stack.

Signed-off-by: Mikulas Patocka 
Signed-off-by: Mike Snitzer

dm btree: fix bufio buffer leaks in dm_btree_del() error path

2015-12-10T15:30:18+00:00

If dm_btree_del()'s call to push_frame() fails, e.g. due to
btree_node_validator finding invalid metadata, the dm_btree_del() error
path must unlock all frames (which have active dm-bufio buffers) that
were pushed onto the del_stack.

Otherwise, dm_bufio_client_destroy() will BUG_ON() because dm-bufio
buffers have leaked, e.g.:
  device-mapper: bufio: leaked buffer 3, hold count 1, list 0

Signed-off-by: Joe Thornber 
Signed-off-by: Mike Snitzer 
Cc: stable@vger.kernel.org

dm space map metadata: fix ref counting bug when bootstrapping a new space map

2015-12-09T18:27:25+00:00

When applying block operations (BOPs) do not remove them from the
uncommitted BOP ring-buffer until after they've been applied -- in case
we recurse.

Also, perform BOP_INC operation, in dm_sm_metadata_create() and
sm_metadata_extend(), in terms of the uncommitted BOP ring-buffer rather
than using direct calls to sm_ll_inc().

Signed-off-by: Joe Thornber 
Signed-off-by: Mike Snitzer 
Cc: stable@vger.kernel.org