Bug 451845
| Summary: | Unable to throttle a Pentium 4 CPU. | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Joonas Sarajärvi <muep> | ||||||||||||
| Component: | kernel | Assignee: | John Feeney <jfeeney> | ||||||||||||
| Status: | CLOSED NOTABUG | QA Contact: | Fedora Extras Quality Assurance <extras-qa> | ||||||||||||
| Severity: | medium | Docs Contact: | |||||||||||||
| Priority: | low | ||||||||||||||
| Version: | 9 | CC: | boggiano, jfrieben, kernel-maint | ||||||||||||
| Target Milestone: | --- | ||||||||||||||
| Target Release: | --- | ||||||||||||||
| Hardware: | i386 | ||||||||||||||
| OS: | Linux | ||||||||||||||
| Whiteboard: | |||||||||||||||
| Fixed In Version: | Doc Type: | Bug Fix | |||||||||||||
| Doc Text: | Story Points: | --- | |||||||||||||
| Clone Of: | Environment: | ||||||||||||||
| Last Closed: | 2008-11-26 20:18:19 UTC | Type: | --- | ||||||||||||
| Regression: | --- | Mount Type: | --- | ||||||||||||
| Documentation: | --- | CRM: | |||||||||||||
| Verified Versions: | Category: | --- | |||||||||||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||||||||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||||||||||
| Embargoed: | |||||||||||||||
| Attachments: |
|
||||||||||||||
|
Description
Joonas Sarajärvi
2008-06-17 19:07:34 UTC
Created attachment 309658 [details]
Output of /sbin/lspci
Created attachment 309659 [details]
Output of /sbin/lsmod
Created attachment 309660 [details]
Output of cat /proc/cpuinfo
Created attachment 309661 [details]
Output of cat /proc/acpi/processor/CPU0/info
Created attachment 309663 [details]
Output of modprobe acpi_cpufreq
I'm assuming by 'throttling' you mean 'run at different speeds'. Your CPU doesn't actually support this feature. In earlier releases, you were using 'p4-clockmod', which faked the appearance of running at different speeds, by inserting 'pauses' between running instructions. This doesn't actually save any power (which is why it got removed). Matthew has been working on making ACPI 'do the right thing' in the absence of working throttling states. Faking P-states is completely the wrong thing to do however, so kpowersave/cpuspeed/cpufreq-applet will never work on your hardware. Looking at your acpi/processor/info output though, it looks like your cpu/bios doesn't expose T-states either. Thank you for the fast response. It seems I was then just misinformed about the actual functionality. I have noticed that after loading p4-clockmod.ko on an Intel Celeron 2 GHz desktop system running kernel 2.6.18-92.1.10.el5, minimum and maximum frequencies are both set to 20000000. Thus, there is no scaling at all whereas the clock is adjusted dynamically for kernel 2.6.22.14-72.fc6 which has been added for testing purposes. Is this change an intended one follwing the reasoning of comment #6? (In reply to comment #8) more /proc/acpi/processor/CPU1/info processor id: 0 acpi id: 1 bus mastering control: no power management: no throttling control: yes limit interface: yes more /proc/acpi/processor/CPU1/throttling state count: 8 active state: T0 states: *T0: 00% T1: 12% T2: 25% T3: 37% T4: 50% T5: 62% T6: 75% T7: 87% (In reply to comment #9) After installing F10 plus updates including kernel-2.6.27.5-123.fc10.i686, the p4-clockmod module is present and can be loaded. However, the cpuspeed daemon cannot be started, and /sys/devices/system/cpu/cpu0/cpufreq/ is empty. However, dmesg actually reports: "p4-clockmod: P4/Xeon(TM) CPU On-Demand Clock Modulation available" upon loading of module p4-clockmod. this is expected behaviour. The user-interface is meant to be disabled. The ACPI code will call into the p4-clockmod code directly when throttling is necessary. |