Index: share/man/man9/vnode_pager_setsize.9
===================================================================
--- share/man/man9/vnode_pager_setsize.9
+++ share/man/man9/vnode_pager_setsize.9
@@ -27,7 +27,7 @@
 .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 .\" SUCH DAMAGE.
 .\"
-.Dd April 6, 2021
+.Dd April 8, 2021
 .Dt VNODE_PAGER_SETSIZE 9
 .Os
 .Sh NAME
@@ -41,26 +41,31 @@
 .Fn vnode_pager_setsize "struct vnode *vp" "vm_ooffset_t nsize"
 .Sh DESCRIPTION
 .Nm
-lets the VM system know about a change in size for a file.
-Content beyond the new EOF specified by the
+lets the VM system know about a change in size for a file,
+and updates the object size and vnode pager size of the vm object in
+.Fa vp
+with
+.Fa nsize .
+Subsequent page faults happened beyond the new object size will result in
+.Va SIGBUS .
+.Pp
+Pages between the old EOF and the new EOF are removed from the object queue
+if the file size shrinks.
+In case the new EOF specified by the
 .Fa nsize
-argument will be purged from the cache.
+argument is not aligned to page boundary,
+partial-page area starting beyond the EOF will be zeroed
+if the page exists resident.
+.Pp
+In case the vnode
+.Fa vp
+does not have a VM object allocated, the effect of calling this function is no-op.
+.Pp
 This function is useful for use within file system code to implement
 truncation in
 .Xr VOP_SETATTR 9 .
-.Sh IMPLEMENTATION NOTES
-In case the new EOF specified by the
-.Fa nsize
-argument is not aligned to page boundaries,
-partial-page area starting beyond the EOF will be zeroed.
-In partial-page area,
-for content occupying whole blocks within block
-boundaries,
-the dirty bits for the corresponding blocks will be cleared.
-.Sh LOCKING
-Writer lock of the VM object of
-.Fa vp
-will be held within the function.
+.Sh LOCKS
+The vnode should be exclusively locked on entry and will still be locked on exit.
 .Sh SEE ALSO
 .Xr vnode 9
 .Sh HISTORY