Merge branch 'koct9i-flush'

Signed-off-by: Sebastien GODARD <sysstat@users.noreply.github.com>

diff --cc man/iostat.in
index 8e991a624c192a1be8d55075bc9346ddf6e3d0a1,437161a495c01343c00d08e126fe2696e6da2b60..d83e8ce934b972239fcc2b2a600bb767c3d3e3e1
--- 1/man/iostat.in
--- 2/man/iostat.in
+++ b/man/iostat.in
@@@ -1,4 -1,4 +1,4 @@@
--.TH IOSTAT 1 "NOVEMBER 2018" Linux "Linux User's Manual" -*- nroff -*-
++.TH IOSTAT 1 "DECEMBER 2019" Linux "Linux User's Manual" -*- nroff -*-
  .SH NAME
  iostat \- Report Central Processing Unit (CPU) statistics and input/output
  statistics for devices and partitions.
@@@ -216,6 -216,12 +216,12 @@@ The number (after merges) of write requ
  .RS
  The number (after merges) of discard requests completed per second for the device.
  
 -Origin (unmerged) flush operations are counted as writes (likely with zero size).
+ .RE
+ .B f/s
+ .RS
+ The number (after merges) of flush requests completed per second for the device.
++This counts flush requests executed by disks. Flush requests are not tracked for partitions.
+ 
  .RE
  .B sec/s (kB/s, MB/s)
  .RS
@@@ -326,6 -332,13 +332,15 @@@ The average time (in milliseconds) for 
  to be served. This includes the time spent by the requests in queue and
  the time spent servicing them.
  
 -to be served. Flush requests are issued one at a time, thus flush operation
 -could be twice longer: wait current flush request, then issue and wait next.
+ .RE
+ .B f_await
+ .RS
+ The average time (in milliseconds) for flush requests issued to the device
++to be served.
++The block layer combines flush requests and executes at most one at a time.
++Thus flush operations could be twice as long: Wait for current flush request,
++then execute it, then wait for the next one.
+ 
  .RE
  .B aqu-sz
  .RS