| Age | Commit message (Collapse) | Author |
|---|
|
fuse_ext_map_iomap_begin() runs on every mapping fault of a DAX
extent-mapped file, meaning the TLB-miss-plus-page-table-miss case,
and on every chunk of every read/write. These are not page faults in
the usual sense: nothing is read in from storage. The fault only
installs page table entries for memory that is already there, so there
is no I/O to amortize the resolver's cost against; whatever it costs
is added to the cost of a memory access.
The rbtree it walks is built from individually allocated nodes. Each
level is a dependent, un-prefetchable dereference and a potential LLC
miss (~100ns), so a striped file with S strips pays log2(S) of them
per resolution, and that is a large fraction of a mapping fault that
otherwise costs little more than the page table update itself.
Replace the rbtree in struct fuse_backing with a packed, sorted array
of fuse_iext extents, allocated once per map rather than once per
extent. The array packs the extents into the fewest cache lines, and
fault handling is O(1) when the map allows it:
- A cyclic (striped) map is required to have equal-length extents
contiguous from zero, so a file offset converts arithmetically to a
strip index and cycle number. This patch validates that and does
it.
- A simple map whose extents happen to be equal-length and contiguous
from zero (the common famfs allocation) resolves as offset / ext_len.
This patch detects that at populate time and does it.
- An irregular map falls back to a binary search of the packed array.
famfs currently always produces maps that resolve in O(1).
Because the extents are sorted at populate time, a cyclic map's extents
no longer have to arrive in offset order, and the overlap check is an
adjacent-pair compare over the sorted array rather than a check at each
rbtree insertion. Error returns are unchanged: -EIO with a message for
malformed extents and overlaps, -EINVAL for a non-uniform cyclic map.
Signed-off-by: John Groves <john@groves.net>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Add the FUSE_BACKING_MAP_CYCLIC flag for FUSE_NOTIFY_BACKING_MAP
notifications to enable creating striped maps.
This is just a set of regular extents that repeats after the end of the
last extent using a striping pattern (i.e. on the target backings the
stripes have a contiguous footprint).
Two additional limits apply compared to non-cyclic extent map population:
- the size of the extents (chunk size) must be equal
- the extents must start from zero offset and follow each other without
gaps.
Originally-by: John Groves <john@groves.net>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Wire up read, write, splice and mmap operations for extent-mapped
files through the iomap/DAX infrastructure.
When a passthrough file is opened with an EXTMAP backing, I/O is
dispatched to the dax devices referenced by the extent map.
If an address is not mapped, EIO is returned on the I/O operation.
Originally-by: John Groves <john@groves.net>
Reviewed-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Add support for creating and managing extent maps that map file regions to
dax device regions.
Introduce FUSE_NOTIFY_BACKING_MAP for populating extent maps from
userspace.
Extent maps are handled as a new type of backing and are assigned a 64 bit
backing ID by the server.
One extent consists of
- offset within the containing backing
- length of extent
- target backing ID
- offset into target backing
Extents must be non-overlapping, and can only refer to dax device backings
for now.
At this point only support creating and populating the extent map in one
operation, though the interface is not limited by this and later may be
changed, so that partial population of an extent map would be possible.
Originally-by: John Groves <john@groves.net>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
