freebsd-dev/cddl
Alexander Motin d3c6ba3214 MFOpenZFS: make zil max block size tunable
We've observed that on some highly fragmented pools, most metaslab
allocations are small (~2-8KB), but there are some large, 128K
allocations.  The large allocations are for ZIL blocks.  If there is a
lot of fragmentation, the large allocations can be hard to satisfy.

The most common impact of this is that we need to check (and thus load)
lots of metaslabs from the ZIL allocation code path, causing sync writes
to wait for metaslabs to load, which can take a second or more.  In the
worst case, we may not be able to satisfy the allocation, in which case
the ZIL will resort to txg_wait_synced() to ensure the change is on
disk.

To provide a workaround for this, this change adds a tunable that can
reduce the size of ZIL blocks.

External-issue: DLPX-61719
Reviewed-by: George Wilson <george.wilson@delphix.com>
Reviewed-by: Paul Dagnelie <pcd@delphix.com>
Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Matthew Ahrens <mahrens@delphix.com>
Closes #8865
openzfs/zfs@b8738257c2

MFC after:	2 weeks
2020-03-19 01:05:54 +00:00
..
compat/opensolaris
contrib/opensolaris MFOpenZFS: make zil max block size tunable 2020-03-19 01:05:54 +00:00
lib sack_newdata and snd_recover hold the same value. Therefore, use only 2020-02-13 15:14:46 +00:00
sbin pkgbase: Force zfs(8) and zpool(8) to be in the runtime package 2019-09-05 14:07:49 +00:00
tests
usr.bin fix up r354804, link zstreamdump with libzfs 2019-11-18 10:34:27 +00:00
usr.sbin DTrace: add a top level makefile to the new test suit 2019-06-09 22:45:07 +00:00
Makefile
Makefile.inc