search for: governor

This fixes an actual bug (failure to exit from a function after an allocation failure), an inconsistency (not removing the cpufreq_dom list member upon failure), and a latent bug (not clearing the current governor upon governor initialization failure when there was no old one; latent because the only current code path leading to this situation frees the policy upon failure and hence the governor not getting cleared is benign). Signed-off-by: Jan Beulich <jbeulich@suse.com> --- a/xen/drivers/cpufreq/c...

Hi, The default "stop" action for the cpupower service seems to be to set the ondemand governor, but this fails. I can reproduce the issue directly by running cpupower: $ sudo cpupower frequency-set -g ondemand Setting cpu: 0 Error setting new values. Common errors: - Do you have proper administration rights? (super-user?) - Is the governor you requested available and modprobed? - Trying to...

Add cpufreq governors: performance, powersave, userspace This patch add 3 more governors beside original running ondemand cpufreq governor. performance governor is with best performance, keeping cpu always running at highest freq; powersave governor is with best power save effect, keeping cpu always running at lowest...

Back in c/s 18950 the default cpufreq governor was set to userspace (it had previously been performance). However, since there is no supplied userspace program or script that will change the frequency this is at best a no-op. Worse, on some hardware with some BIOS revisions, this actually sets the CPUs running at their lowest frequency ra...

I have xeons 5520 how do I configure xen 4.3 to lower cpu clocks ? cpufreq=dom0-kernel cpuidle I read power management under xen wiki but command xenpm set-scaling-governor ondeman Gives [CPU0] failed to set governor name (19 - No such device) [CPU1] failed to set governor name (19 - No such device) [CPU2] failed to set governor name (19 - No such device) [CPU3] failed to set governor name (19 - No such device) [CPU4] failed to set governor name (19 - No such device...

...vg freq -1749837312 KHz CPU13: Residency(ms) Avg Res(ms) Avg freq -1749837312 KHz CPU14: Residency(ms) Avg Res(ms) Avg freq -1749837312 KHz CPU15: Residency(ms) Avg Res(ms) Avg freq -1749837312 KHz xenpm set-scaling-governor userspace [CPU0] failed to set governor name [CPU1] failed to set governor name [CPU2] failed to set governor name [CPU3] failed to set governor name [CPU4] failed to set governor name [CPU5] failed to set governor name [CPU6] failed to set governor name [CPU7] failed to set governor name [CPU8] fa...

Modify the cpufreq ondemand governor so that it can get idle and total ticks from the Xen hypervisor. Linux and Xen have different ideas of what an idle tick is, so the Xen values for both have to be returned in the same platform hypercall. Signed-off-by: Mark Langsdorf <mark.langsdorf@amd.com> diff -r b4278beaf354 xen/arch/x...

Set userspace governor as default, which stays same effect as when cpufreq in xen is not enabled. As a result, enable cpufreq in xen by default to avoid reboot to activate cpufreq. Now it''s always on but w/o performance impact if user doesn''t tempt to change governor. We also add some warning info for d...

...s. In term of the performance benefit of this patch, I ran the high_systime workload (which does a lot of fork() and exit()) at various load levels (500, 1000, 1500 and 2000 users) on a 4-socket IvyBridge-EX bare-metal system (60 cores, 120 threads) with intel_pstate driver and performance scaling governor. The JPM (jobs/minutes) and execution time results were as follows: Kernel JPM Execution Time ------ --- -------------- At 500 users: 3.19 118857.14 26.25s 3.19-qspinlock 134889.75 23.13s % change +13.5%...

...s. In term of the performance benefit of this patch, I ran the high_systime workload (which does a lot of fork() and exit()) at various load levels (500, 1000, 1500 and 2000 users) on a 4-socket IvyBridge-EX bare-metal system (60 cores, 120 threads) with intel_pstate driver and performance scaling governor. The JPM (jobs/minutes) and execution time results were as follows: Kernel JPM Execution Time ------ --- -------------- At 500 users: 3.19 118857.14 26.25s 3.19-qspinlock 134889.75 23.13s % change +13.5%...

Hi list, I'm using C7 and noticed that cpu freq is enabled with governor conservative. I've used cpupower utils for governor change, but at the next boot it restart with conservative governor. I've tried also to start cpupower service and setting OPT in /etc/sysconfig/cpupower but it does not works. Can someone point me in the right direction? Thanks in adv...

Modify the cpufreq ondemand governor so that it can get idle and total ticks from the Xen hypervisor. Linux and Xen have different ideas of what an idle tick is, so the Xen values for both have to be returned in the same platform hypercall. Otherwise, use basically the same scheme as native Linux. Signed-off-by: Mark Langsdorf &lt...

Modify the cpufreq ondemand governor so that it can get idle and total nsecs from the Xen hypervisor.  Xen uses nsecs to measure idle time, while Linux uses ticks. Other than accounting for that difference, use the same algorithm to calculate idle time as Linux does. Signed-off-by: Mark Langsdorf <mark.langsdorf@amd.com> dif...

...cy period may be longer than accouting period (30ms), csched_tick should be aware of resume event to adjust elapsed credits. ---- c) when cpu''s freq is scaled dynamically When cpufreq/Px is enabled, cpu''s frequency is adjusted to different operation points driven by a on-demand governor. So csched_acct may need take frequency difference among cpus into consideration and total available credits won''t be a simple 300 * online cpu_number. ---- Of course there''re bunch of research areas to add more power factor into scheduler policy. But above is fundamental stuff...

...ning is disabled? What is the point of enabling dynamic tuning, especially when using the supplied profiles like throughput-performance? I haven't customized those by specifying any thresholds, and I don't see any point in doing things like dynamically changing disk schedulers or CPU governors depending on something recognized by tuned. Why is there no measurable difference in power consumption between using the powersave, throughput-performance and virtual-host profiles but a very noticeable difference in performance? The difference in performance appears to depend mainly on which...

After upgrade my laptop, I discovered a low performance on it. The laptop is a 2,73GHz Intel centrino and from some kernels versions to now, it had a good performance, the cpu scaling worked fine. Now, I discover that forcing in /etc/sysconfig/cpuspeed the governors and/or max/min speeds, it don't want to run more quickly. The laptop boots fine with the correct speed (usual speed), but when the init scripts run the service cpuspeed, the performance goes low: Only wants 800MHz!! Yes, I played with the documentation and the values on it, but the problem b...

Hello I'm building a router/proxy/firewall, the default governor is ondemand, but for this application, what governor is best suited? (for example conservative?) Thanks at all! Best regards

On 09/23/2017 03:52 AM, hw wrote: > Thank you very much for the notice!? Looking at a couple machines, I found > that the automatic choice of profile isn?t what I would want. > > Now I wonder how everyone deals with this, i. e. do you set a profile once > and never change it, or do you keep changing the profile according to > circumstances?? Is changing it even advisable, i. e.

..._cpufreq_para(xc_handle, cpuid, UP_THRESHOLD, threshold) ) - fprintf(stderr, "failed to set up scaling threshold\n"); + fprintf(stderr, "failed to set up scaling threshold (%d - %s)\n", + errno, strerror(errno)); } } void scaling_governor_func(int argc, char *argv[]) { int cpuid = -1; - char *name = NULL; + char *name; if ( argc >= 2 ) { - name = strdup(argv[1]); - if ( name == NULL ) - goto out; - if ( sscanf(argv[0], "%d", &cpuid) != 1 ) - { -...

Hi! Attached patch makes xen-tools and stubdom-dm going independent from hardcoded pathes. It is no possible to install into /usr/local or any other non-default directory and use it out-of-the box. This allows us to have different Xen versions in different directories, simplifies packaging for distributions. It also finds ''hvmloader'' and