Fix manual description of zfs_arc_dnode_limit

In arc_evict_state() we start pruning when arc_dnode_size >
arc_dnode_limit, i.e. arc_dnode_limit is a ceiling rather than a
floor.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Signed-off-by: Chris Dunlop <chris@onthe.net.au>
Closes #6228
This commit is contained in:
chrisrd 2017-06-15 06:23:02 +10:00 committed by Brian Behlendorf
parent 8f7933fec9
commit 627791f3c0

View File

@ -412,7 +412,7 @@ Default value: \fB2\fR.
.RS 12n .RS 12n
When the number of bytes consumed by dnodes in the ARC exceeds this number of When the number of bytes consumed by dnodes in the ARC exceeds this number of
bytes, try to unpin some of it in response to demand for non-metadata. This bytes, try to unpin some of it in response to demand for non-metadata. This
value acts as a floor to the amount of dnode metadata, and defaults to 0 which value acts as a ceiling to the amount of dnode metadata, and defaults to 0 which
indicates that a percent which is based on \fBzfs_arc_dnode_limit_percent\fR of indicates that a percent which is based on \fBzfs_arc_dnode_limit_percent\fR of
the ARC meta buffers that may be used for dnodes. the ARC meta buffers that may be used for dnodes.