The man page suggest that the cpu_shares parameter of schedinfo
allows values 0-262144, but the kernel remaps values 0 and 1 to
the minimum 2, just document that behaviour:
[root@test ~]# cat /cgroup/cpu/libvirt/qemu/cpu.shares
1024
[root@test ~]# echo 0 > /cgroup/cpu/libvirt/qemu/cpu.shares
[root@test ~]# cat /cgroup/cpu/libvirt/qemu/cpu.shares
2
[root@test ~]# echo 1 > /cgroup/cpu/libvirt/qemu/cpu.shares
[root@test ~]# cat /cgroup/cpu/libvirt/qemu/cpu.shares
2
[root@test ~]#
* tools/virsh.pod: update description of the cpu_shares parameter
to indicate the values 0 and 1 are automatically changed by the
kernel to minimal value 2
diff --git a/tools/virsh.pod b/tools/virsh.pod
index a01d723..6e3febb 100644
--- a/tools/virsh.pod
+++ b/tools/virsh.pod
@@ -914,7 +914,9 @@ If I<--current> is specified, affect the current guest state.
B<Note>: The cpu_shares parameter has a valid value range of 0-262144; Negative
values are wrapped to positive, and larger values are capped at the maximum.
-Therefore, -1 is a useful shorthand for 262144.
+Therefore, -1 is a useful shorthand for 262144. The values 0 and 1 seems to
+also be automatically changed by the Linux kernel to 2 which consider this
+the minimal value.
B<Note>: The weight and cap parameters are defined only for the
XEN_CREDIT scheduler and are now I<DEPRECATED>.
--
Daniel Veillard | libxml Gnome XML XSLT toolkit
http://xmlsoft.org/
daniel(a)veillard.com | Rpmfind RPM search engine
http://rpmfind.net/
http://veillard.com/ | virtualization library
http://libvirt.org/