Also added the -i and --ignore-missing options to the tuned-adm manual page.
Related: rhbz#2075827
Signed-off-by: Jaroslav Škarvada <jskarvad@redhat.com>
When the reapply_sysctl option in tuned-main.conf is set to 1, "system"
sysctls from /run/sysctl.d/*.conf, /etc/sysctl.d/*.conf and
/etc/sysctl.conf will be set *after* TuneD sysctl settings are applied.
As reapply_sysctl=1 is TuneD default, this can be confusing to TuneD
users who only rely on TuneD profile settings and do not know about this
functionality. We also need a way to inform other components using
TuneD about this.
This PR adds information about reapplied sysctls in TuneD logs.
Signed-off-by: Jiri Mencak <jmencak@users.noreply.github.com>
Extends the runtime API by the method:
instance_acquire_devices(devices, instance_name)
E.g. consider the following TuneD profile:
[cpus_perf]
type=cpu
devices=cpu0, cpu1
governor=performance
[cpus_idle]
type=cpu
devices=${f:cpulist2devs:2-3}
governor=ondemand
[cpus_idle2]
type=cpu
devices=${f:cpulist2devs:4-5,7}
governor=ondemand
Notice that it's possible to use internal function 'cpulist2devs' to easily
specify cpulists.
After the following API call:
instance_acquire_devices("cpulist:2-5,7", "cpus_perf")
It will result in cpu0 - cpu5 and cpu7 having the "performance" governor.
It's also possible to specify full device names in the API instead of the
cpulist, e.g.:
instance_acquire_devices("cpu2,cpu3,cpu4,cpu5,cpu7", "cpus_perf")
In case the comma is part of the device name, it can be escaped by "\,".
In case the device name starts with the string "cpulist:", e.g. there is
by accident device named "cpulist:abcd", it's possible to use
"cpulist:cpulist:abcd" to escape it and don't recognize it as the cpulist.
The cpulist expand feature also supports bitmasks variant prefixed by the
0x and written as a string, e.g. "0xffffffff", i.e. all the standard
TuneD cpulist syntaxes are supported.
Resolves: rhbz#2113925
Signed-off-by: Jaroslav Škarvada <jskarvad@redhat.com>
Now it should allow duplicate options and sections in the INI files, the
latest value is taken, the previous values are ignored, e.g.:
[selinux]
avc_cache_threshold=4096
avc_cache_threshold=8192
The avc_cache_threshold will be set to the 8192.
Related: rhbz#2071418
Signed-off-by: Jaroslav Škarvada <jskarvad@redhat.com>
The new profile sets the intel_acpi driver to passive mode to
be able to set the standard acpi governors (ondemand/userspace)
and provides more flexibility on the C-states, as it is exposed
as a new variable.
Signed-off-by: Christophe Fontaine <cfontain@redhat.com>
Even when we instruct TuneD to not apply changes to the bootloader config
(skip_grub_config=true) it's useful to expose the generated cmdline.
External tools could then read TuneD's desired cmdline, and adapt configurations
of (so far) unsupported bootloaders like systemd-boot.
Signed-off-by: Adriaan Schmidt <adriaan.schmidt@siemens.com>
The polkit has D-bus activation thus it will be activated on demand and
we don't need to explicitly activate it.
Resolves: rhbz#2065591
Signed-off-by: Jaroslav Škarvada <jskarvad@redhat.com>
The PREEMPT_RT upstream tree [1] is reintroducing a per-CPU kthread for
handling timer expiry named ktimers/${cpu}, and RHEL-RT is following this
approach [2].
The recommended priority tuning for that kthread is to simply match that of
ksoftirqd, which is what is implemented here.
Note that ktimersoftd is the predecessor of ktimers, and isn't present on
RT kernels newer than v5.0.19-rt11 / RHEL-RT 7.
[1]: https://git.kernel.org/pub/scm/linux/kernel/git/rt/linux-rt-devel.git
[2]: https://bugzilla.redhat.com/show_bug.cgi?id=2122220
Resolves: rhbz#2122220
Signed-off-by: Valentin Schneider <vschneid@redhat.com>
In the old kernels, vm.hugepages_treat_as_movable tunable was aimed at
reducing memory fragmentation. Things have changed since then and this
tunable is no longer present in recent kernels:
https://lore.kernel.org/lkml/20171003072619.8654-1-mhocko@kernel.org/t/
Its presence in the hpc-compute profile causes TuneD to log errors.
Remove it.
Signed-off-by: Jiri Mencak <jmencak@users.noreply.github.com>
