Registered Member
|
When I try using ctrl+alt+F1 the screen fills with "powernow_k8: failing targ, change pending bit set" and "cpufreq: --target --index failed to change cpu frequency: -5". Please help newbie.
Update: The terminal screen quit displaying the errors after a few hours, but the KSystemLog is still printing the errors every second. After rebooting, terminal screen started displaying errors again continuously. Additional Info: - computer Acer T135 - AMD Sempron Processor - added Radeon HD 3450 PCI video board System Log indicates that these errors are being provided by kernel Process.
Last edited by dougc on Sat Apr 11, 2015 1:06 am, edited 1 time in total.
|
Registered Member
|
Update: I suspect that there is a problem with frequency control of the processor. I went into Bios and changed freq control from "Default" to "X8 1600 Mhz". My options were 1. Default, 2. X5 1000 Mhz, 3. X6 1200 Mhz, 4. X7 1400 Mhz, 5. X8 1600Mhz and 6. X9 1800 Mhz. After reboot THIS STOPPED THE ERRORS.
With "Default" option SystemLog showed: 10/04/15 01:39:02 AM M110 kernel [ 95.464109] powernow_k8: Hardware error - pending bit very stuck - no further pstate changes possible 10/04/15 01:39:02 AM M110 kernel [ 95.464120] tsc: Marking TSC unstable due to cpufreq changes 10/04/15 01:39:02 AM M110 kernel [ 95.464125] powernow_k8: transition frequency failed 10/04/15 01:39:02 AM M110 kernel [ 95.464337] cpufreq: __target_index: Failed to change cpu frequency: 1 10/04/15 01:39:02 AM M110 kernel [ 95.464417] Switched to clocksource acpi_pm 10/04/15 01:39:03 AM M110 kernel [ 95.632056] powernow_k8: failing targ, change pending bit set 10/04/15 01:39:03 AM M110 kernel [ 95.632071] cpufreq: __target_index: Failed to change cpu frequency: -5 The last two lines would continuously repeat. With "X8 1600 Mhz" option the SystemLog showed: 10/04/15 01:51:34 AM M110 kernel [ 0.052442] ..TIMER: vector=0x30 apic1=0 pin1=2 apic2=0 pin2=0 10/04/15 01:51:34 AM M110 kernel [ 0.092132] smpboot: CPU0: AMD Sempron(tm) Processor 3000+ (fam: 0f, model: 2c, stepping: 02) 10/04/15 01:51:34 AM M110 kernel [ 0.096000] Performance Events: AMD PMU driver. 10/04/15 01:51:34 AM M110 kernel [ 0.096000] ... version: 0 10/04/15 01:51:34 AM M110 kernel [ 0.096000] ... bit width: 48 10/04/15 01:51:34 AM M110 kernel [ 0.096000] ... generic registers: 4 10/04/15 01:51:34 AM M110 kernel [ 0.096000] ... value mask: 0000ffffffffffff 10/04/15 01:51:34 AM M110 kernel [ 0.096000] ... max period: 00007fffffffffff 10/04/15 01:51:34 AM M110 kernel [ 0.096000] ... fixed-purpose events: 0 10/04/15 01:51:34 AM M110 kernel [ 0.096000] ... event mask: 000000000000000f 10/04/15 01:51:34 AM M110 kernel [ 0.096000] x86: Booted up 1 node, 1 CPUs 10/04/15 01:51:34 AM M110 kernel [ 0.096000] smpboot: Total of 1 processors activated (3616.59 BogoMIPS) How or why the "Default" option would cause a problem I don't know. Perhaps something needs to be changed in script for kernel. This work around may help others. If anyone has any further insight, please let me know. Additional Note: The problem is repeatable. If I select "Default" in Bios for CPU frequency, the error messages appear continuously. Thanks for looking. |
Registered users: bancha, Bing [Bot], Evergrowing, Google [Bot], lockheed, mesutakcan, mickae, Sogou [Bot]