Extend the ionice man page to explain the "none" class and how the cpu-nice => io-priority inheritance works. Signed-off-by: Jakob Unterwurzacher <jakobunt@xxxxxxxxx> --- Reading the cfq source [1] cleared this up for me. The formula in the patch is from Documentation/block/ioprio.txt . Preview of the patched man page: http://pastebin.com/f5caa733c [1] http://tomoyo.sourceforge.jp/cgi-bin/lxr/source/block/cfq-iosched.c#L1366 diff -up a/ionice.1 b/ionice.1 --- a/ionice.1 2008-12-27 21:34:49.000000000 +0100 +++ b/ionice.1 2008-12-27 21:08:11.000000000 +0100 @@ -11,7 +11,7 @@ This program sets or gets the io schedul If no arguments or just \fI-p\fR is given, \fIionice\fR will query the current io scheduling class and priority for that process. -As of this writing, Linux supports 3 scheduling classes: +As of this writing, a process can be in one of four scheduling classes: .IP "\fBIdle\fP" A program running with idle io priority will only get disk time when no other @@ -21,9 +21,9 @@ class does not take a priority argument. is permitted for an ordinary user (since kernel 2.6.25). .IP "\fBBest effort\fP" -This is the default scheduling class for any process that hasn't asked for -a specific io priority. Programs inherit the CPU nice setting for io -priorities. This class takes a priority argument from \fI0-7\fR, with lower +This is the effective scheduling class for any process that hasn't asked for +a specific io priority. +This class takes a priority argument from \fI0-7\fR, with lower number being higher priority. Programs running at the same best effort priority are served in a round-robin fashion. @@ -35,6 +35,14 @@ some care, as it can starve other proces will receive on each scheduling window. This scheduling class is not permitted for an ordinary (i.e., non-root) user. +.IP "\fBNone\fP" +A process that that hasn't asked for a specific io priority will formally be +in this class. The io scheduler will treat a process that is in this class as +if it were in the best effort class. The priority within the best effort +class will be dynamically derived from the cpu nice level of the process: +io_priority = (cpu_nice + 20) / 5 + + .SH OPTIONS .IP "\fB-c \fIclass\fP" The scheduling class. \fI1\fR for real time, \fI2\fR for best-effort, \fI3\fR for idle. -- To unsubscribe from this list: send the line "unsubscribe util-linux-ng" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
