X61s Hardware Interrupt up to 40% of CPU resource

Hello, i recently updated my harddrive of my x61s to a seagate Momentus ST9320421ASG 7200.3 GForce - 320GB 7200rpm 16MB. After that i have done fresh Installation with Windows XP Professional on the new drive twice!! Each time when i finish installation i begin to get a wired Hardware Interrupt with constantly eats up to 40% of my cpu resource and cuts my battery life to half. i have tried to deactived the component one after another in order to locate the problem(which componet is sending the interrupt). But i found nothing. And the interrupt does not start right away when i boot into XP. it starts sometimes later after login or start some programms like opera or Firefox browser. Does anyone has an idea about this issue? BTW, i already use the system update to get all my drivers and bios to the newst version. Thanks!
Solved!
Go to Solution.

T his issue annoyed me for a long time , but final!! eventually !!! i fixed it!
its because of the latest intel display dirver.
so everyone just need to download an older one.

Similar Messages

X200 Hardware Interrupts constantly at 40% of CPU resources

As many other thinkpad owners I started to experience excessive "Hardware Interrupts" using between 40-45% of my CPU resources (on both cores) on my x200. Note that Task Manager will _not_ show this, you will need something like Microsoft's "Process Explorer" to show this. This started about a week ago, following a slew of updates I had been rolling in from "System Update" on my Vista x64 Business.
I searched the Internet high and low and found many similar experiences (not only with Lenovo models, not only with Vista). Obviously, it is nearly impossible to find out what's causing this. There is a useful tool called "KrView" by Microsoft that will display more info about which driver is causing the interrupts, but unfortunately it does not run under Vista x64.There is also an on-board Vista tool that allows you to check for unsigned drivers but that didn't seem to be the problem (all drivers were certified ok).
Other have reported problems with their network drivers where such interrupts would occur as soon as they turn on WiFi. Some were successful by removing their installation of Deamon Tools (which I have not installed, so that didn't help).Others suggested hard disk drivers or CD-ROM driver issues.
I thus tried to home in on the probably culprit, and found the following puzzling effect: This only occurs when my x200 is in the docking station! After a clean startup, it will be nice and idle for some time, but after some 5 minutes it will start to constantly show the 40-45% of "Hardware Interrupts." Once this happens, it almost never goes back down, even if you undock it. However, if you _start_ the system in an undocked state, the "Hardware Interrupts" do not appear!! I even tried to manually connect all the different pieces that I have attached to my docking station one by one to my undocked laptop (e.g., network cable, external keyboard and mouse, USB harddisk) - none did trigger this. But as soon as you redock your laptop, it will spike up again to its 40-50% "Hardware Interrupts".
I _assume_ that it might be the display driver that causes this, as this is the only piece of hardware that I couldn't "test" on my standalone laptop: I use a displayport cable to connect by 24" monitor, and the x200 does not have a connector for this... I tried installing the latest Intel graphics drivers but the won't install and tell me that I must use the OEM lenovo versions. I also could not "rollback" the driver (option greyed out) and the old driver version doesn't seem to be available on the lenovo post...
Alternatively, it might be some sort of "docking station driver" that's causing this. Although I'm not sure what this driver would be and how to update it :-(
Any hints? (sorry, ultra-long post)

And this is the kernrate log after I uninstalled the sata controller from control panel. No CPU load.
I restarted the machine after removing the sata controller and it installed the controller again, it has prompted me to reboot again to finish the installation but I haven't done that yet. Until now, no high load has happened.
C:\Documents and Settings\ctrlER>"C:\Program Files\KrView\Kernrates\Kernrate_i38
6_XP.exe"
/==============================\
< KERNRATE LOG >
\==============================/
Date: 2009/02/19 Time: 13:35:13
Machine Name: TUTTLE
Number of Processors: 2
PROCESSOR_ARCHITECTURE: x86
PROCESSOR_LEVEL: 6
PROCESSOR_REVISION: 0f0a
Physical Memory: 2535 MB
Pagefile Total: 4422 MB
Virtual Total: 2047 MB
PageFile1: \??\C:\pagefile.sys, 2046MB
OS Version: 5.1 Build 2600 Service-Pack: 3.0
WinDir: C:\WINDOWS
Kernrate User-Specified Command Line:
C:\Program Files\KrView\Kernrates\Kernrate_i386_XP.exe
Kernel Profile (PID = 0): Source= Time,
Using Kernrate Default Rate of 25000 events/hit
Starting to collect profile data
***> Press ctrl-c to finish collecting profile data
===> Finished Collecting Data, Starting to Process Results
------------Overall Summary:--------------
P0 K 0:00:00.125 ( 1.2%) U 0:00:00.031 ( 0.3%) I 0:00:10.312 (98.5%) DPC
0:00:00.031 ( 0.3%) Interrupt 0:00:00.000 ( 0.0%)
Interrupts= 1673, Interrupt Rate= 160/sec.
P1 K 0:00:00.250 ( 2.4%) U 0:00:00.000 ( 0.0%) I 0:00:10.218 (97.6%) DPC
0:00:00.015 ( 0.1%) Interrupt 0:00:00.031 ( 0.3%)
Interrupts= 1673, Interrupt Rate= 160/sec.
TOTAL K 0:00:00.375 ( 1.8%) U 0:00:00.031 ( 0.1%) I 0:00:20.531 (98.1%) DPC
0:00:00.046 ( 0.2%) Interrupt 0:00:00.031 ( 0.1%)
Total Interrupts= 3346, Total Interrupt Rate= 320/sec.
Total Profile Time = 10468 msec
BytesStart BytesStop Byt
esDiff.
Available Physical Memory , 2158850048, 2162970624, 4120
576
Available Pagefile(s) , 4226367488, 4225921024, -446
464
Available Virtual , 2132660224, 2131611648, -1048
576
Available Extended Virtual , 0, 0,
0
Total Avg. Rate
Context Switches , 9745, 931/sec.
System Calls , 29229, 2792/sec.
Page Faults , 1707, 163/sec.
I/O Read Operations , 72, 7/sec.
I/O Write Operations , 37, 4/sec.
I/O Other Operations , 1000, 96/sec.
I/O Read Bytes , 24076, 334/ I/O
I/O Write Bytes , 1380, 37/ I/O
I/O Other Bytes , 1382780, 1383/ I/O
Results for Kernel Mode:
OutputResults: KernelModuleCount = 129
Percentage in the following table is based on the Total Hits for the Kernel
Time 493 hits, 25000 events per hit --------
Module Hits msec %Total Events/Sec
ntkrnlpa 395 10468 80 % 943351
hal 37 10468 7 % 88364
KSecDD 33 10468 6 % 78811
NETw5x32 12 10468 2 % 28658
win32k 7 10468 1 % 16717
Tppwrif 2 10468 0 % 4776
spiv 2 10468 0 % 4776
avipbb 1 10468 0 % 2388
TPHKDRV 1 10468 0 % 2388
igxpmp32 1 10468 0 % 2388
NDIS 1 10468 0 % 2388
Ntfs 1 10468 0 % 2388
================================= END OF RUN ==================================
============================== NORMAL END OF RUN ==============================
C:\Documents and Settings\ctrlER>
C:\Program Files\KrView\Kernrates\Kernrate_i386_XP.exe -z ntkrnlpa
Kernel Profile (PID = 0): Source= Time,
Using Kernrate Default Rate of 25000 events/hit
CallBack: Finished Attempt to Load symbols for 804d7000 \WINDOWS\system32\ntkrnl
pa.exe
Starting to collect profile data
***> Press ctrl-c to finish collecting profile data
===> Finished Collecting Data, Starting to Process Results
------------Overall Summary:--------------
P0 K 0:00:00.156 ( 1.2%) U 0:00:00.078 ( 0.6%) I 0:00:12.296 (98.1%) DPC
0:00:00.031 ( 0.2%) Interrupt 0:00:00.015 ( 0.1%)
Interrupts= 1736, Interrupt Rate= 139/sec.
P1 K 0:00:00.171 ( 1.4%) U 0:00:00.031 ( 0.2%) I 0:00:12.328 (98.4%) DPC
0:00:00.015 ( 0.1%) Interrupt 0:00:00.015 ( 0.1%)
Interrupts= 1735, Interrupt Rate= 138/sec.
TOTAL K 0:00:00.328 ( 1.3%) U 0:00:00.109 ( 0.4%) I 0:00:24.625 (98.3%) DPC
0:00:00.046 ( 0.2%) Interrupt 0:00:00.031 ( 0.1%)
Total Interrupts= 3471, Total Interrupt Rate= 277/sec.
Total Profile Time = 12531 msec
BytesStart BytesStop Byt
esDiff.
Available Physical Memory , 2166996992, 2164838400, -2158
592
Available Pagefile(s) , 4227276800, 4225929216, -1347
584
Available Virtual , 2132660224, 2130460672, -2199
552
Available Extended Virtual , 0, 0,
0
Total Avg. Rate
Context Switches , 10583, 845/sec.
System Calls , 30623, 2444/sec.
Page Faults , 1704, 136/sec.
I/O Read Operations , 45, 4/sec.
I/O Write Operations , 26, 2/sec.
I/O Other Operations , 725, 58/sec.
I/O Read Bytes , 17252, 383/ I/O
I/O Write Bytes , 280, 11/ I/O
I/O Other Bytes , 1377388, 1900/ I/O
Results for Kernel Mode:
OutputResults: KernelModuleCount = 129
Percentage in the following table is based on the Total Hits for the Kernel
Time 502 hits, 25000 events per hit --------
Module Hits msec %Total Events/Sec
ntkrnlpa 397 12531 79 % 792035
KSecDD 39 12531 7 % 77807
hal 39 12531 7 % 77807
NETw5x32 17 12531 3 % 33915
win32k 4 12531 0 % 7980
igxpmp32 2 12531 0 % 3990
avgntflt 1 12531 0 % 1995
mrxsmb 1 12531 0 % 1995
NDIS 1 12531 0 % 1995
ACPI 1 12531 0 % 1995
===> Processing Zoomed Module ntkrnlpa.exe...
----- Zoomed module ntkrnlpa.exe (Bucket size = 16 bytes, Rounding Down) -------
Percentage in the following table is based on the Total Hits for this Zoom Modul
e
Time 397 hits, 25000 events per hit --------
Module Hits msec %Total Events/Sec
NtBuildNumber 322 12531 81 % 642406
PoShutdownBugCheck 20 12531 5 % 39901
ZwYieldExecution 11 12531 2 % 21945
Kei386EoiHelper 5 12531 1 % 9975
RtlIpv6StringToAddressA 5 12531 1 % 9975
ProbeForRead 4 12531 1 % 7980
KeTickCount 4 12531 1 % 7980
KeSynchronizeExecution 4 12531 1 % 7980
wctomb 4 12531 1 % 7980
LsaDeregisterLogonProcess 3 12531 0 % 5985
IoWMISetNotificationCallback 1 12531 0 % 1995
SeTokenIsWriteRestricted 1 12531 0 % 1995
PsSetProcessPriorityByClass 1 12531 0 % 1995
PsEstablishWin32Callouts 1 12531 0 % 1995
PoQueueShutdownWorkItem 1 12531 0 % 1995
ObQueryNameString 1 12531 0 % 1995
ExRaiseStatus 1 12531 0 % 1995
mbstowcs 1 12531 0 % 1995
PoSetPowerState 1 12531 0 % 1995
PoStartNextPowerIrp 1 12531 0 % 1995
MmTrimAllSystemPagableMemory 1 12531 0 % 1995
MmIsThisAnNtAsSystem 1 12531 0 % 1995
MmProtectMdlSystemAddress 1 12531 0 % 1995
MmIsDriverVerifying 1 12531 0 % 1995
KeRundownQueue 1 12531 0 % 1995
================================= END OF RUN ==================================
============================== NORMAL END OF RUN ==============================
C:\Documents and Settings\ctrlER>
Message Edited by ctrler on 02-19-2009 05:43 AM

MSI 945GCM5-F V2 FSB 1333 / hardware interrupts

I struggle bigtime with hardware interrupts on this MB.
On the MB i got a Samsung spinpoint 320GB sata2 disk, a standard dualcore intel and 2GB corsair RAM
Latest BIOS.
Tried to update drivers for everything, the disk is running with DMA (UDMA5)
I ran out of ideas, so I'm hoping someone have a few?
When it happens always core#0 uses 90-100% cpu and it lags/freezes the computer in a different way than if I use almost 100% load on a task.
I'm suspecting the MB, since it don't have much connected to it.....

Quote
On the PCI-slot I think it says 200, but it's hard to see.
I asked for the ERP number, because there are quite a few boards with the model number MS-7267. The ERP number is one possible way to keep them appart. If the ERP number of you board is in fact 200 you have the right BIOS (If it was 170, BIOS v17.7 [=vH.7] would be the wrong one):
Quote
601-7267-010 BIOS version 1.x = 945GZM3-L
601-7267-020 BIOS version 3.x = 945GZM3-L
601-7267-030 BIOS version 5.x = 945GM3-F
601-7267-070 BIOS version 7.x = 945GM4-FI
601-7267-080 BIOS version 9.x = 945GCM5-F
601-7267-100 BIOS version 9.x = 945GM5-F
601-7267-110 BIOS version 9.x = 945GZM5-F
601-7267-130 BIOS version B.x = 945GZM6-LV
601-7267-140 BIOS version D.x = 945GZM6-L
601-7267-150 BIOS version C.x = 945GZM6-F
601-7267-160 BIOS version E.x = 945GCM5-L V2
601-7267-170 BIOS version F.x = 945GCM5-F V2
601-7267-190 BIOS version G.x = 945GCM5-L V2
601-7267-200 BIOS version H.x = 945GCM5-F V2
Quote
PSU: NorthQ 4001EXT, Silent ATX 400W, 140mm Fan, 12-17dBA, 2xSATA, PCI-E, 20/24pin
Quote from: AaronYuri on 14-July-08, 21:27:44
Just 15A on the +12v rail? Try with a better PSU.
Considering the fact, that he uses the onboard graphics chip with just an E2200, one HDD and one ODD connected to the board, 15A @12V may be just enough to power this configuration. Of course, the question is always if a given PSU really holds what it promises on the PSU sticker. Even for a 400W device, 15A is not a whole lot (a 450W Corsair PSU for example has a 33A-Limit). The PSU is not my first suspicion in your case, but I agree to AaronJuri, that NorthQ Device looks weak enough to consider testing a better one just to rule out the possibility that your problem is in any way PSU related.
asdfsdf, a couple of more question and suggestions:
Questions:
-What OS are you using?
-Is that fresh installation of the OS or is it taken from another system?
-Did you update the BIOS yourself at any point, and if you did, how did you do it and did you clear CMOS with main A/C power cable removed from PSU after the flash?
-Are (IO)APIC and ACPI enabled in BIOS?
Suggestions:
1. Clear CMOS with main A/C power cable removed from PSU and load "Optimized Defaults" in BIOS after restarting your system.
--> Test if the problem still occurs. If it does, do this:
2. Unplug your SATA Optical Drive from both PSU & board, remove all USB Devices and Disable the following peripherals in BIOS:
- USB Controller & USB Legacy Support (Use PS2 Keyboard & Mouse for testing
- OnBoard LAN Controller
- HD Audio Controller
- LAN Option ROM
- OnBoard IDE Controller
- OnBoard Floppy Controller
- COM Port & Parallel Port
- Intel EIST
- Spread Spectrum
- Auto Disable DIMM/PCI Frequency
Retest and see if there are any changes. If the lags disappear, re-enable each Device/function in BIOS (one by one) to find out, which one causes the problem.

T61: Problems with Hardware Interrupts when on battery

Not entirely sure, but was wondering if anyone else had experienced this.
I recently updated my T61 laptop with the included System Update tool. I'm fairly regular in my checks (every other week or so)... and noticed some disturbing behavior after updating the Power Management Software and driver off this utility.
Specifically, this behavior manifests itself as 30-40% of my CPU being taken up as "Hardware Interrupts" according to Process Explorer. Task Manager shows the CPU used, but reports 99% as "Idle Process". This wouldn't be a big problem except that it only occurs when the laptop is running on battery (generally after about a minute on battery). This saps my battery time, and it also lags -everything- down (even the mouse cursor).
A temporary solution is to sleep the machine and then power it back on, and then I get another minute before the Hardware Interrupts kick on.
Any suggestions?
maderator note: type added.
Message Edited by Agotthelf on 02-04-2009 10:32 AM
Solved!
Go to Solution.

Hello,
did the interrupt start after quiet the same time, when on battery?
Then please go to start, control panel, task planer and delete the PM task.
Follow @LenovoForums on Twitter! Try the forum search, before first posting: Forum Search Option
Please insert your type, model (not S/N) number and used OS in your posts.
I´m a volunteer here using New X1 Carbon, ThinkPad Yoga, Yoga 11s, Yoga 13, T430s,T510, X220t, IdeaCentre B540.
TIP: If your computer runs satisfactorily now, it may not be necessary to update the system.
English Community Deutsche Community Comunidad en Español

Hardware Interrupts HELP please

Hello,
My problem is this: when i open my PC everything works perfect, but when i execute a "heavy" programme(like games or a full system scan etc) after a while the CPU reaches 100% and the system is responding really slow(30-45 sec after the mouse click).I have downloaded the processexplorer and clearly the problem is the Hardware Interrupts.
I had the same problem 6 months ago, i formated the HD and everything was normal again.Yesterday the Hardware Interrupts started again.I couldn`t find anything to do(i`ve searched a couple of forums) so i decided to format the disk again hopping that everything was going to fix again(at least for some months) but unfortunately today and after trying to play Command and Conquer Tiberium Wars the system crashed and (based to processexplorer) the hardware interrupts was the problem again.
Somebody help please and thanks for your time

First of all thank you very much for your very quick response.
I also thought that maybe the Video Card or the CPU temperature is increasing and causes the hardware interrupts but the CPU temperature doesn`t go higher than 37-38 degrees Celsius and the SYS temp is 1 or 2 degrees lower. For example right now i am copying some files from a DVD and the Hardware interrupts are using avg. 30% of the CPU and the CPU temp is avg. 32 degrees Celsius and the SYS temp is avg 29 degrees Celsius. The temperatures are taken from the Core Center programme from MSI.If there is any other way to see my CPU`s and system`s temperature please tell me to check it.
As for the Hard Disks I am checking now one of them and there may be a problem.I will do some tests and i will inform you as soon as i can.
Thank you very much

LX195 WHS High hardware interrupts after wake up from sleep

I have a HP-LX195 windows home server which is set to sleep every day between 2AM and 7 AM. On wake up from sleep the CPU usage is triggered to 35-40%. No tasks are executing during this period and if any they add up the original usage. Using process explorer notices these CPU utilization is due to hardware interrupts. Using kernrate noticed the high usage is on intelppm.sys.
I have determined this is due to sleep. If the server is kept running without sleep, the hardware interrupt is minimal on normal operations. But if i manually set the server to sleep and wake up, the usage triggers back to 35 or 40. The high cpu utilization doesn't go away until restart.
I have tested this with other LX195 and it seems consistent and repeatable. I have posted this on microsoft WHS forum as well. This doesn't happen with EX490.
Is it possible to get a fix from HP regarding the same. Thanks Raja

does hp support monitor these forums?

Using ioctl to wait for hardware interrupt in user space in Petalinux

I have a custom Network device driver for a device that generates two hardware interrupts, one every 20 ms and another every second. I have a separate interrupt handler registered for each interrupt. Each interrupt handler does nothing more than call "up" on a semaphore. There is a different semaphore associated with each interrupt.
I want a user-space program to be notified when the interrupt occurs, so I have an ioctl defined with two commands. One is a command to wait for the 20 ms interrupt and the other is a command to wait for the 1 second interrupt. In each case, the ioctl routine simply calls down_interruptible on the corresonding semaphore (the same semaphore modified by the interrupt handlers).
In my user-space program, I create two pthreads. Each has an infinite loop to call each of the two ioctls described above.
I am seeing that the user-space thread that waits for the 1 second interrupt is going to sleep and seems to be blocking the overall process. When the 1 second interrupt comes in, the thread wakes up and, in the process of handling the interrupt, allows the other thread to be tasked. The result is that exactly one 20 ms interrupt is handled each second, whenever a 1 second interrupt fires.
I have tried enabling the O_NONBLOCK mode on the socket, but this did not make any difference. I'm not sure how to allow the second thread to continue to run, even while the first thread is asleep. Seems like this is the whole point of using pthreads.
Any help would be appreciated.

So, the issue was that I was trying to implement the ioctl in a network device. When you call ioctl on a socket, it apparently blocks the process until the ioctl completes.
I implemented a character device inside my network driver and used this to handle the ioctls instead. Once I did this, I found that only the calling thread blocked on ioctl, not the entire process.

6602: Rising edge capturing hardware interrupt

I have this signal in input:
I have to do something on rising edge as soon as possible.
My idea is to receive an hw interrupt from the board, and to write an isr soubroutine.
I read about callback function so i write some code.
But when i call:
    DAQmxErrChk ( DAQmxCfgChangeDetectionTiming ( udpTaskHandle,
                                                "Dev1/Ctr0",
                                                "Dev1/Ctr0",
                                                DAQmx_Val_ContSamps,
                                                1));
I have an error for the ctr0, it seems i can use onli the port0.
Now my question.... is there on port0 a software thread that read continuosly the value of the pins?
Or is there an hardware interrupt anyway ?
I think that a software thread has a big latency for me, at this point don't need the 6602 board,
i could read the signal on LPT standard port.
Thank you for the answer.
There are 10 kinds of people. Those who understand binary notation, and those who do not.
Solved!
Go to Solution.

Sorry if it's passed a long time, i tryed to do as you suggested, but if i set DAQmx_Val_HWTimedSinglePoint how can i have a continuos acquisition?
that's my code, it seems that thecallback function is called only once.
#include <stdio.h>
#include <NIDAQmx.h>
#define DAQmxErrChk(functionCall) if( DAQmxFailed(error=(functionCall)) ) goto Error; else
int32 CVICALLBACK TriggerDetectionCallback(TaskHandle taskHandle, int32 signalID, void *callbackData);
void Cleanup (void);
static TaskHandle taskHandle=0;
int main(void)
    int         error=0;
    int32       read;
    uInt32      data[1000];
    char        errBuff[2048]={'\0'};
    // DAQmx Configure Code
    DAQmxErrChk (DAQmxCreateTask("",&taskHandle));
    DAQmxErrChk (DAQmxCreateCICountEdgesChan(taskHandle,
                                            "Dev1/ctr1",
                                            DAQmx_Val_Rising,
                                            0,
                                            DAQmx_Val_CountUp));
    DAQmxErrChk (DAQmxCfgSampClkTiming(taskHandle,
                                       "/Dev1/PFI35",
                                       1000.0,
                                       DAQmx_Val_Rising,
                                       DAQmx_Val_HWTimedSinglePoint ,
                                       1000));
    DAQmxErrChk (DAQmxRegisterSignalEvent(taskHandle,
                                        DAQmx_Val_SampleClock ,
                                        0,
                                        TriggerDetectionCallback,
                                        NULL)
    // DAQmx Start Code
    DAQmxErrChk (DAQmxStartTask(taskHandle));
    printf("Continuously reading. Press Ctrl+C to interrupt\n");
    while( 1 ) {
Error:
    puts("");
    if( DAQmxFailed(error) )
        DAQmxGetExtendedErrorInfo(errBuff,2048);
    if( taskHandle!=0 ) {
        // DAQmx Stop Code
        DAQmxStopTask(taskHandle);
        DAQmxClearTask(taskHandle);
    if( DAQmxFailed(error) )
        printf("DAQmx Error: %s\n",errBuff);
    printf("End of program, press Enter key to quit\n");
    getchar();
    return 0;
int32 CVICALLBACK TriggerDetectionCallback(TaskHandle taskHandle, int32 signalID, void *callbackData)
    int32   error=0;
    uInt32 data[200]={0};
    int32   numRead;
    uInt32 i=0;
    char    errBuff[2048]={'\0'};
    static int count =0;
    count++;
    printf("\n\n%d", count);
    if( taskHandle ) {
        // DAQmx Read Code
        DAQmxErrChk (DAQmxReadCounterU32(taskHandle,1000,10.0,data,1000,&numRead,NULL));
        printf("\n%d",data);
    return 0;
Error:
    if( DAQmxFailed(error) )
        DAQmxGetExtendedErrorInfo(errBuff,2048);
        Cleanup();
        printf("DAQmx Error: %s\n",errBuff);
    return 0;
void Cleanup (void)
    if( taskHandle!=0 )
        // DAQmx Stop Code
        DAQmxStopTask(taskHandle);
        DAQmxClearTask(taskHandle);
        taskHandle = 0;
Message Edited by blacksocket on 30-03-2010 03:56 PM
There are 10 kinds of people. Those who understand binary notation, and those who do not.

While being idle online for a few minutes, why does Firefox start cyclically using 50% to 5% CPU resources, with a 10 to 1, hi to lo ratio

This behavior became apparent after the last couple of Firefox version updates. Current Firefox version is 21.0. My PC is an IBM ThinkCentre running @ 3.0 Ghz Intel CPU with 1.5 Gigs PC3200 DDR ram. With or without any other Windows apps open, whether I'm on line just idling on my home page (Google) or any other web page or entering / receiving data, several minutes into a session, the MS XP pro SP3 task manager shows Firefox cyclically using 50% (and more) CPU resources for 8-10 seconds then 1-5% CPU resources for 1-2 seconds. Closing & restarting Firefox, allows me a few minutes of speedy surfing. My ISP is AT&T 2nd tier DSL with a download thruput of 300 (+) MB (not Mb) / second (altho, most of the time I'm just viewing a web page or idling, without downloading or uploading any data).
What has changed? My PC & all of its software have remained the same with the exception of MS security updates, Adobe Reader updates, & Firefox updates (without adding any additional plug ins, add ons, etc., with each update).
What the...! For the past several minutes that I've been inputting data here, the task manager showed a jump of only 13% cpu usage (ever so briefly) by Firefox & only then when I switched between the task manager & Firefox. Intermittent anomalies are gonna be the death of me...
Well, this is an annoyance, albeit an intermittent one. Ah, I pine for the days when I heard about & believed in "seamless integration"...any thots that might help resolve this issue sure would be appreciated.

Hello,
The Reset Firefox feature can fix many issues by restoring Firefox to its factory default state while saving your essential information.
Note: ''This will cause you to lose any Extensions, Open websites, and some Preferences.''
To Reset Firefox do the following:
#Go to Firefox > Help > Troubleshooting Information.
#Click the "Reset Firefox" button.
#Firefox will close and reset. After Firefox is done, it will show a window with the information that is imported. Click Finish.
#Firefox will open with all factory defaults applied.
Further information can be found in the [[Reset Firefox – easily fix most problems]] article.
Did this fix your problems? Please report back to us!
Thank you.

Firefox using massive CPU resources, even with no addons?

I've been trying to figure out why Firefox is using ridiculously large amounts of CPU resources, but can't seem to find anything. At the time of writing this, I have 5 tabs open, none of which are highly graphic- or sound-intensive (facebook, 3 articles and this) and no add-on toolbars running. I updated to FF7 the other day, primarily because FF wouldn't open without doing so, and it's currently using a little over 625K in CPU.
I've tried restarting, but it starts at around 105K and skyrockets to around 300K within 5 minutes and I can't think what to do to sort it out. I've tried running it in safe mode, but there's no difference to the CPU usage. Add to that the fact that I can't even run another program at the same time (making me choose between music or internet), and frankly, it's driving me up the wall.
I don't want to change browsers, for the simple reason that I prefer FF in terms of usability and appearance, so I'd prefer a fix if possible. The troubleshooting information included is from my current session, which does have some addons re-enabled, since disabling them did no good whatsoever.

Post to the developer forums under OS X Technologies.

SQL Developer occasionally hogs all CPU resources

Client OS: Windows XP Professional
Connecting to : Oracle 9.2.0.3 on Sun Solaris server
I encountered a few instances where SQL Developer hogs all CPU resources on my Windows XP notebook.
I usually have to kill the SQL Developer process.
I have noticed this happening occasionally when I :
1. hit Enter at the SQL Worksheet window
2. hit F9 at the SQL Worksheet window - the query does not start
How do I collect a dump or some kind of debug information when the above happens ?
Who can I send the information to ?

How do I collect a dump or some kind of debug
information when the above happens ?Start SQLDeveloper using sqldeveloper.exe in <sqldeveloper>\jdev\bin
When sqldeveloper starts hogging cpu, go to the console window and type ctrl-break. You should get a thread dump.
Who can I send the information to ?If you have a support contract, you can raise a TAR, otherwise post the thread dump here.

Cpu resource is available but many run-queue from vmstat

Hi, all
I coundn't find appropriate category for this system related questions.
I am sorry for this.
The system has more than 100 CPUs on a linux machine, but there are more than 10 run queues.
As you see, "r" is around 10, but "id" of cpu is about 90%.
=> I cound not understand why processes do not use OTHER available CPU resource in stead of waiting in run-queue.
uname -a
Linux aaa 2.6.18-194.el5 #1 SMP Mon Mar 29 22:10:29 EDT 2010 x86_64 x86_64 x86_64 GNU/Linux
vmstat 1 10
procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu------
r b   swpd   free   buff cache   si   so    bi    bo   in   cs us sy id wa st
6 13      0 117346968 3604540 105912264    0    0 1013    60    0    0 6 2 90 3 0
9 14      0 117347072 3604540 105912240    0    0 27752 11759 40266 79569 4 2 89 5 0
11 7      0 117348512 3604540 105912272    0    0 36680 3085 37745 74248 5 2 89 4 0
7 16      0 117344720 3604540 105912192    0    0 27720 15913 36344 74105 4 1 89 6 0
7 13      0 117346896 3604540 105912288    0    0 9944 15803 41294 80441 6 2 88 5 0
5 9      0 117348656 3604540 105912288    0    0 11368 2051 39088 77034 5 1 89 4 0Thanks in advance.
Best Regards.

Thanks for your reply.
"r" is the number of processes ON CPU plus the number of processes WAITING FOR CPU.
The following is from "man vmstat".
Procs
       r: The number of processes waiting for run time.
       b: The number of processes in uninterruptible sleep.I am confused whether or not "r" contains processes which are corrently on cpu, not in run queue.
Could you give me more detailed explaination or some article?
Thanks.
Best Regards.
Edited by: 869578 on 2013. 2. 15 오전 12:48

Firefox sometimes freezes up for a few minutes and begins using 47% of cpu resources and sometimes also extra 100-200kb of memory

since updating from 3.6.xx to firefox 5, 6 and now 7, firefox sometimes will freeze while loading a site and use 42-50% of cpu resource. it also sometimes doubles or trebles memory it uses at same time as freezing, and freezes for 2 or more minutes. when it comes back sometimes it will have failed to load a page and will display the url but a white screen. this can happen any site, but seems to occur more when on image sites. usually i have more than one ff window open, one of which is an iboard, and the freezing seems to occur more often when either posting, or when opening a subboard or more often a thread in tabs or new windows. problem seemed to be helped when i disabled a recently installed java plugin but didnt disappear, and it can still happen anytime with only one window/tab open. the only common thing seems to be the site is either refreshing, or loading.

Try:
*http://blog.bonardo.net/2011/09/30/is-your-firefor-freezing-at-regular-intervals
*Places Maintenance: https://addons.mozilla.org/firefox/addon/places-maintenance/

Moble device support uses 50 percent of cpu resource

I tried uninstalling Itunes and all of it's components. I have a pentium 4 with 3 gigs of ram. Apple Mobile Device Support uses 50% of CPU resources and slows my computer to a craw when Itunes is not running. I don't mind if it uses resources when using the product, but it uses the resources even when your not running any Apple software... Windows XP 32 bit pro all service packs and upgrades. Itunes 10.5 or latest. Any help please..

I am having same problem. Many say reload iTunes in a particular way. Some say that works, others say it doesn't. You can run "netsh winsock reset" and it will temporarily free up the memory once you reboot, but it impacted my IE from finding the server. Still looking. Apple must know this is a problem. As one use said, have they heard of regression testing?

How to limiting CPU resource per process

Solaris 8 OS
Sun 4800 Server
I have 4 CPU modules. There is a runaway process that consumes 25% of the CPU resources. Which is one whole CPU in my situation.
Is there a way for me to limit the CPU consumption from user processes? IE: process can't use more than 5-10% of total CPU at a given time.
There is a command 'limit'. Limit can only control the amount of CPU TIME. This is not applicable because sometimes user need to run processes which last for days.
There is a way to do this . . By buying Solaris Resource Manager. This software cost over $10,000. It will come with a lot of things that I do not need.
There's got to be a way to limit this.
I have done some extensive researching and NO lead.
Any help would be greatly appreciated.

I was thinking about this a bit more. I remember a while back Sun has a BATCH scheduling class available for download. Unfortunately, I can't find it anymore. Maybe Sun turned it into a for purchase product. Anyway, the comments about priority are relevant here.
Just because a process is chewing up your CPU doesn't necessarily mean its a bad thing. If the process is doing work, that shouldn't be considered a problem. If the job is impacting other jobs in the systems that are considering more important, then its a problem. Moving this application into a batch scheduling class would definately take care of this problem OR using NICE to lower the process nice value to say 19 will also work. This would make it so that other processes with higher nice values would preempt this application off the CPU. A BATCH scheduling class would do the same thing.
OR MAYBE and I don't now if this is possible and its way out in left field, maybe investigate changing the default scheduling class to IA so that all normal processes startup in the IA scheduling class and put this abusive process in the TS scheduling class. This way IA scheduling class apps will always preempt anything in the TS scheduling class. Sorta like turning your TS scheduling class into a BATCH scheduling class. Of course, using NICE is probably much more easy.

X61s Hardware Interrupt up to 40% of CPU resource

Similar Messages

Maybe you are looking for