| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Fixes: d284f8248c7 ("dm writecache: support optional offset for start of device")
Signed-off-by: Mikulas Patocka <[email protected]>
Signed-off-by: Mike Snitzer <[email protected]>
|
|
Add an optional parameter "start_sector" to allow the start of the
device to be offset by the specified number of 512-byte sectors. The
sectors below this offset are not used by the writecache device and are
left to be used for disk labels and/or userspace metadata (e.g. lvm).
Signed-off-by: Mikulas Patocka <[email protected]>
Signed-off-by: Mike Snitzer <[email protected]>
|
|
This adjusts the allocator calls to use the 2-factor argument style, as
already done treewide for better defense against allocator overflows.
Signed-off-by: Kees Cook <[email protected]>
[snitzer: tweaked code to leave assignment in a test alone]
Signed-off-by: Mike Snitzer <[email protected]>
|
|
The writecache target caches writes on persistent memory or SSD.
It is intended for databases or other programs that need extremely low
commit latency.
The writecache target doesn't cache reads because reads are supposed to
be cached in page cache in normal RAM.
If persistent memory isn't available this target can still be used in
SSD mode.
Signed-off-by: Mikulas Patocka <[email protected]>
Signed-off-by: Colin Ian King <[email protected]> # fix missing goto
Signed-off-by: Ross Zwisler <[email protected]> # fix compilation issue with !DAX
Signed-off-by: Dan Carpenter <[email protected]> # use msecs_to_jiffies
Acked-by: Dan Williams <[email protected]> # reworks to unify ARM and x86 flushing
Signed-off-by: Mike Snitzer <[email protected]>
|