This option disables the TSC clocksource watchdog
for isolated CPUs (which avoids interruptions
by the per-CPU clocksource watchdog timer).
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Variables in force_latency, min_perf_pct, max_perf_pct and no_turbo
options were not being expanded. Fix it.
Fixes#203
Signed-off-by: Ondřej Lysoněk <olysonek@redhat.com>
In the kernel table of exit latencies of various CPUs the worst
exit latency for C1 seems to be 3 us, so increase the backup
value latency requirement to 3 us to match the worst C1 exit
latency.
Related: rhbz#1737628
Signed-off-by: Jaroslav Škarvada <jskarvad@redhat.com>
Now multiple profiles can be included by:
[main]
include=profile1;profile2
Also comma "," can be used instead of the semicolon ";".
No escaping is done, so it's assumed there will be no comma or semicolon
in the profile name.
Signed-off-by: Jaroslav Škarvada <jskarvad@redhat.com>
For sap-hana and virtual-host profiles used state3 (i.e. usually C3)
C-state in latency specification instead of the hardcoded value 70 us,
i.e.:
force_latency=cstate.id:3|70
Also dropped obsoleted TODO note from the plugin_cpu.
Related: rhbz#1737628
Signed-off-by: Jaroslav Škarvada <jskarvad@redhat.com>
Maximal latency can be now specified multiple ways:
- directly in usec (this is the same as before), e.g. for 10 us:
force_latency = 10
- as an ID of maximal cstate allowed, e.g. for the kernel state1:
force_latency = cstate.id:1
- as a name (case sensitive) of maximal cstate allowed, e.g. for the state named C1:
force_latency = cstate.name:C1
It is also possible to specify multiple fallback values separated by '|', e.g.:
force_latency = cstate.name:C6|cstate.id:4|10
This will try to obtain latency of cstate named C6, if it fails (e.g.
there is no such cstate), it will try kernel state4 and if it also fails
it finally fallbacks to 10 us.
This commit also changes force_latency settings of latency-performance
profile to:
force_latency=cstate.id:1|1
I.e. it tries kernel state1 and fallbacks to 1 us.
Resolves: rhbz#1737628
Signed-off-by: Jaroslav Škarvada <jskarvad@redhat.com>
It fixed the following problem e.g. the profile:
[sysctl]
kernel.pid_max=>131072
and if kernel.pid_max is already 131072 Tuned shows traceback.
Resolves: rhbz#1739418
Signed-off-by: Jaroslav Škarvada <jskarvad@redhat.com>
RHEL will include a backport of the kernel patch
"do not raise timer softirq unconditionally (spinlockless version)"
(https://lkml.org/lkml/2019/4/15/1215), but disabled by default.
Enable it in realtime-virtual-{host/guest} profiles as it decreases
maximum cyclictest latency.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Realtime profiles require to restrict irqbalance too. Apply them to
the realtime profile so that realtime-virtual-{host|guest} can benefit
from that too.
Signed-off-by: Peter Xu <peterx@redhat.com>
Due to commit d4038a7e64, if a script fails to execute or its exit
code is non-zero, no subsequent scripts are executed. This seems
logically wrong and it causes problems especially during rollback as some
tunings may not be reverted due to this behaviour. Also, it appears it
was not the intention of that commit to change this behaviour - the
intention rather seems to have been to make _call_scripts return
success/error information for use by the verification mechanism.
So let's aggregate the success/error information instead.
Related: https://github.com/redhat-performance/tuned/pull/194
Signed-off-by: Ondřej Lysoněk <olysonek@redhat.com>
So that QEMU uses the in kernel irqchip.
Original patch provided by Luiz Capitulino <lcapitulino@redhat.com>.
Resolves: rhbz#1554458
Signed-off-by: Jaroslav Škarvada <jskarvad@redhat.com>
The kernel.shmall and kernel.shmmax parameters were previously set using
a script that calculated their values based on the system's memory size.
However the script was broken for high values of memory size.
The default value of kernel.shmall and kernel.shmmax in RHEL-8 is
18446744073692774399, which is sufficiently high and it's recommended to
use it. So we could simply drop the tuning from the profile, however it
seems better to set the value explicitly so that the profile overrides
any changes that may have been applied to the parameters due to
misconfiguration.
Resolves: rhbz#1708418
Signed-off-by: Ondřej Lysoněk <olysonek@redhat.com>
The default value of kernel.sem in RHEL-8 is '32000 1024000000 500
32000', which is much higher than what we currently have in the profile.
It is unnecessary to lower the settings, the RHEL-8 defaults are safe to
use. So we could simply drop the tuning from the profile, however it
seems better to set the value explicitly so that the profile overrides
any changes that may have been applied to the parameter due to
misconfiguration.
Resolves: rhbz#1701394
Signed-off-by: Ondřej Lysoněk <olysonek@redhat.com>
The default value of kernel.sem in RHEL-8 is '32000 1024000000 500
32000', which is much higher than what we currently have in the profile.
It is unnecessary to lower the settings, the RHEL-8 defaults are safe to
use. So we could simply drop the tuning from the profile, however it
seems better to set the value explicitly so that the profile overrides
any changes that may have been applied to the parameter due to
misconfiguration.
Resolves: rhbz#1701394
Signed-off-by: Ondřej Lysoněk <olysonek@redhat.com>
As discussed in rhbz#1672213, the sap-hana-vmware profile is not needed
anymore and it only confuses users. Let's drop it.
Resolves: rhbz#1715541
Signed-off-by: Ondřej Lysoněk <olysonek@redhat.com>
Ignore non-existent sysctl settings from the system configuration files
(/etc/sysctl.conf, etc.). Logging errors about these settings hurts user
experience, if the non-existent settings are in fact real settings that
are just temporarily unavailable. For example, the following settings
(from /usr/lib/sysctl.d/00-system.conf on RHEL-7) are not available until
the br_netfilter module is loaded. However once that module is loaded,
it is often desirable to set these, so having them in a RHEL-provided
configuration file makes sense.
net.bridge.bridge-nf-call-ip6tables = 0
net.bridge.bridge-nf-call-iptables = 0
net.bridge.bridge-nf-call-arptables = 0
This change restores the old behaviour before the recent rewrite of the
sysctl plugin away from using the 'sysctl' program (running
'sysctl --system' ignores missing settings).
Resolves: rhbz#1714595
Signed-off-by: Ondřej Lysoněk <olysonek@redhat.com>
A bare 'except' statement can catch (and hide) all sorts of errors,
including NameError. Fix it to catch only the errors we really expect.
Signed-off-by: Ondřej Lysoněk <olysonek@redhat.com>
sysctl options such as net.ipv4.tcp_wmem and net.ipv4.tcp_rmem include tabs. When you use tuned-adm verify, the changes report back as broken as the whitespaces are different because one side has been sanitized but the other side has not. This pull request will fix that.
cpulist_unpack, which is used in the function, returns an ordered cpu
list, however by converting it to a set, the order is lost. Rewrite the
operation in a way that preserves the order.
Resolves: rhbz#1706171
Signed-off-by: Ondřej Lysoněk <olysonek@redhat.com>
As these interfaces allow loading and thus executing code from plugins
the default permission policy is lowered from:
yes:yes:yes
to
auth_admin:auth_admin:yes
So only users logged into the X server are allowed to use this interface
without extra root/admin authentication.
This was found by security review by Matthias Gerstner
<matthias.gerstner@suse.com>
Also see:
https://github.com/redhat-performance/tuned/issues/180
The force_latency parameter belongs to the cpu plugin, so that's the
section it needs to be in.
Fixes#132
Resolves: rhbz#1569375
Signed-off-by: Ondřej Lysoněk <olysonek@redhat.com>
