Jump to content

How to get the cause of high CPU usage by DPC / Interrupt


Recommended Posts

Hi there,I followed your guide and got a rar file for you. I know who are the culprits but I am wondering how to handle this and why are there so many with the value uSec over 250. I try to keep my hardwares up to date as much as possible. Only my BIOS has not been updated yet but there is a side story to this.

Rar: http://www.megaupload.com/?d=7O7QN817

Edited by iRic
Link to comment
Share on other sites


Hi Andre,

This is my first post on the forum. I have a very similar issue as many people on the forum, but I can not solve it, by ordinary methods.

The DPC spikes issue has been firstly noticed after I have installed an high end RME Babyface USB audiocard last friday. The card itself is not the cause, I have latest drivers, latest mixing console and the card was running perfectly. Everything went fine until I used well known MPC-HC player to play some 20GB+ m2ts files (blu-ray) with HD audio streams and set RME for audio playback. Before the purchase, I havent noticed the crackles or audiodropouts, because somehow the integrated Realtek Audio chip on the Gigabyte board has hidden the DPC audio issues, which went on with the new RME Card ( the card has selectable buffer sample size ).

Please look at these images when the RME card has been used as playback:

MPC-HC RME Babyface, LatencyMon

MPC-HC, RME Babyface, LatencyMon2

The RME card started to display USB transmission errors - rising with playback:

RME Babyface USB transmission errors

I tried to attach the RME card USB port to another port or better said, controller ( before it was Southbridge USB 2.0 ), namely NEC USB3.0, but the result was the same, DPC issues. Only the name of the usbport.sys was named NEC or so.

When I uninstalled the RME card to check, if its the card which causes the spikes, Realtek on the board has been set as default device, but the issues were still present. I tried to disable CPU powersafe features, Turbo Boost, used GPU clock tool and MSI Afterburner ( to set GPU to fixed frequencies ), but nothing helped:

MPC-HC, Realtek audio chip,LatencyMon+DPC Latency Checker

To sum it up, huge DPC spikes has been notified by the LatencyMon at following drivers/sys files:

usbport.sys ( or any name assigned by the usb controller )

portcls.sys ( occasionally )

scsiport.sys ( everytime there )

ataport.sys

hdaudiobus.sys ( occasionally using realtek )

Now my setup:

i7 980x

Win7 64bit

Gigabyte GA-X58a UD5 Rev. 2, BIOS FD ( released a week ago ), the chipset driver is actual 9.1.2.1007 ( from here )

OCZ Revodrive (not x2), 50GB with the latest driver (for OS )

DELL Perc H700 SAS/SATA Raid Controller with latest driver (Raid 10), using 4x WD RE4-GP

3x Samsung HD204UI

Pioneer BDR-205

Sapphire ATI Radeon HD5670 Ultimate - tested Catalyst 10.10e, 11.1, currently 11.1a

Please take a notice, that Im using MPC-HC with madvr video renderer ( curently best of ). Using other renderers like EVR based, I couldnt reproduce these huge scsiport.sys spikes ( 2000-3999 ).

During playback with EVR rendrer, the scsiport.sys events were displayed as the highest in the Latencymon, but the DPC Count was under 500 and ISR under 250 ( execution time ).

What do you think?

Thanks

Pluto

Edited by Plutotype
Link to comment
Share on other sites

Power Savings Options: http://www.megaupload.com/?d=9NOOMHAX ***Please Note that I deactivated Sleep function after 15 minutes WHILE taking those screnshots.

I am currently disscussing with gateway and Acer customer service about my BIOS.

To be honest, I have no idea if I should flash my BIOS. I don't want to screw up my system because this BIOS update is not for the BIOS I use or something like that.

My question to acer:

I currently have a Gateway Fx6802-07c bought from Futurshop in january 2010. I would like to know if the current BIOS update available for the 6802 on the Gateway site (R01.B1, 06/29/2010) is valid for my system. Motherboard: TBGM01 / BIOS: P01-A0. I ask the question here since the Gateway support told me: Flashing the wrong BIOS can cause harm to the system. Acer recommends that you should only upgrade your firmware/drivers if you have been instructed to do so by an Acer Customer Care representative. By using these firmware/drivers you agree to accept the possibility of product failure.

Acer's response:

Response 02/14/2011 11:52 AM Dear Ric

Ric, we does not recommend the BIOS update. If the system is working fine then there is no need for a BIOS update. If the system have any known issue and if the BIOS update

is available to resolve that issue then you may do so.

I sent a reply that says: Well that's the problem, ACPI.SYS hogs 2 of my cores under NT & Kernel by using a steady 15-30% of my CPU. I want to know If the BIOS Update is valid for my BIOS so I can update it just like all of my other drivers. I am currently trying to resolve this problem by elemination process simply by updating my drivers one by one. Reinstalling Windows 7 did not fix it.

Acer answer: As per your mail we understand that you are experiencing an issue with CPU usage and you need to know about BIOS upgrade. I apologize for the inconvenience that you have experienced.

You may please go through the following link to find the BIOS update available for your system.

http://global-download.gateway.com/GDFiles/BIOS/BIOS/BIOS_Gateway_R01.B1_A_A.zip?acerid=634133888894332578&Step1=DESKTOP&Step2=FXSERIES&Step3=FX6802&OS=ALL&LC=en&BC=GATEWAY&SC=PA_6G

Since it is a BIOS update for this model, it won't cause any harm to your system. We don't recommend flashing the wrong BIOS, since it cause harm to the system.

By using these firmware/drivers you agree to accept the possibility of product failure.

I stripped their posts of the un-usefull info for you and me lol.

My question to people who read this post, would it be safe (please investigate) to update my BIOS and would it resolve the ACPI.SYS (DPC_INTERRUPT) issue which causes an average 15-30% CPU usage on my i7 920 ?

Edited by iRic
Link to comment
Share on other sites

@iRic

I never had issues while updating the BIOS. try it out. Also load the default BIOS settings.

but will it SURELY resolve the ACPI.SYS problem ?

Edited by iRic
Link to comment
Share on other sites

My BIOS is ¨P01-A0 and the update is named R01-B1, does it matter....? I mean, my pc is 2 years old,I bought in 2010. They might have changed the BIOs since then...Can you please tell me the signification of these P01-A0 and etc ?

EDIT: I believe my BIOS is not the same since their "updates" are older (they were last modified in 23/5/2009) than my BIOS's (dates from 08/08/2009).

I also tried disabling my devices in device manager but with no success. By devices I mean; Realtek on-board soundcard, USB Razer DeadthAdder Mouse, USB HP Multimedia Keyboard, Nvidia GeForceGTX 260 (GPU), DVD read-writting hardware.

I tried re-updating to the lastest drivers but changed nothing.

Edited by iRic
Link to comment
Share on other sites

What do you think?

contact the support of the RME Babyface USB audiocard. Their driver cause the issue.

Hi Andre,

Thanks for the check of my post. I have already posted the issue on the RME forum. I was asked to check my drivers and install the RME soundcard correctly. Both actions were done earlier. The install of the card is very simply. You take out the newest drivers, extract them somewhere in your computer and then you connect the card via USB to some stable powered USB slot. In device Manager you click babyface on "update driver" and locate the extracted driver folder. Click ok, done. My card is installed corectly, but behaves bad. In my setup.

The worst thing on the card is a fact, that I have problems with directsound wav,mp3 playback (no video) in general. Tried 128/512/1024/2048 buffer size at 44.1/48kHz, but the audio playback is not smooth - skips and crackles occasionally, gets played faster. The card has definately issues with any playback in directsound mode. Again - only in my setup.

Today, I did some tests where I wanted to check how does the DPC spikes turn on when there is lower GPU load than at the mentioned madVR . So I have tried another renderer EVR, then DXVA based EVR-CP and the shader based madVR. Here are the results:

LatencyMon results, same video file (14GB):

EVR:

Your system appears to be suitable for playing real time audio without dropouts.

Highest DPC routine execution time (µs): 108

Responsible driver: dxgkrnl.sys (DirectX Graphics Kernel, Microsoft Corporation)

DPC count (execution time <500 µs): 2835330

DPC count (execution time 500-999 µs): 0

DPC count (execution time 1000-1999 µs): 0

DPC count (execution time 2000-3999 µs): 0

DPC count (execution time >=4000 µs): 0

Highest ISR routine execution time (µs): 29

Responsible driver: ataport.SYS (ATAPI Driver Extension, Microsoft Corporation)

ISR count (execution time <250 µs): 3438554

ISR count (execution time 250-499 µs): 0

ISR count (execution time 500-999 µs): 0

ISR count (execution time 1000-1999 µs): 0

ISR count (execution time >=2000 µs): 0

Highest pagefault resolution time (µs): 0

Hard pagefault count (total): 0

Number of processes hit: 0

Pagefault count of hardest hit process: 0

Process with highest pagefault count: none

EVR-CP:

Your system appears to be suitable for playing real time audio without dropouts.

Highest DPC routine execution time (µs): 101

Responsible driver: dxgkrnl.sys (DirectX Graphics Kernel, Microsoft Corporation)

DPC count (execution time <500 µs): 5631255

DPC count (execution time 500-999 µs): 0

DPC count (execution time 1000-1999 µs): 0

DPC count (execution time 2000-3999 µs): 0

DPC count (execution time >=4000 µs): 0

Highest ISR routine execution time (µs): 19

Responsible driver: SCSIPORT.SYS (SCSI Port Driver, Microsoft Corporation)

ISR count (execution time <250 µs): 6182521

ISR count (execution time 250-499 µs): 0

ISR count (execution time 500-999 µs): 0

ISR count (execution time 1000-1999 µs): 0

ISR count (execution time >=2000 µs): 0

Highest pagefault resolution time (µs): 0

Hard pagefault count (total): 0

Number of processes hit: 0

Pagefault count of hardest hit process: 0

Process with highest pagefault count: none

madVR:

One or more DPC and ISR routines that execute on behalf of drivers running in your system may be causing problems when processing real time audio. You may be experiencing audio dropouts, clicks and pops.

Highest DPC routine execution time (µs): 3666

Responsible driver: SCSIPORT.SYS (SCSI Port Driver, Microsoft Corporation)

DPC count (execution time <500 µs): 2168838

DPC count (execution time 500-999 µs): 0

DPC count (execution time 1000-1999 µs): 0

DPC count (execution time 2000-3999 µs): 378

DPC count (execution time >=4000 µs): 0

Highest ISR routine execution time (µs): 3614

Responsible driver: SCSIPORT.SYS (SCSI Port Driver, Microsoft Corporation)

ISR count (execution time <250 µs): 1692748

ISR count (execution time 250-499 µs): 0

ISR count (execution time 500-999 µs): 0

ISR count (execution time 1000-1999 µs): 0

ISR count (execution time >=2000 µs): 232

Highest pagefault resolution time (µs): 0

Hard pagefault count (total): 0

Number of processes hit: 0

Pagefault count of hardest hit process: 0

Process with highest pagefault count: none

I turned GPU-Z on and there definately a relation between GPU load and DPC spikes. EVR and EVR-CP generate approx. 0-14% GPU load. MadVR´s GPU load was between 35-43%.DPC spikes and GPU-Z looked like this:

EVR

EVR CP dxva

madVR

Tests has been done with the RME Babyface on the audio output.

To understand. EVR does not use DXVA, its passing everything to CPU. EVR-CP with DXVA based decoders offloads this work to the GPU processors. MadVR is different, its basically programmed like a D3D video game renderer - all work is done via GPU shaders, it does not support DXVA at all. MadVR outputs in YV12 color format only and its programmed to achieve the best results - with the costs of GPU load.

What I try to say, is that the DPC latency issues rise with the GPU load. Im not skilled in this, but few things come off due to this:

- IRQ issue

- PCIe issue ( check my PCIe configuration - have 3 cards there )

- motherboard issue, chipset issue, bus communication

- GPU issue

-...

I dont believe RME has produced a faulty card with faulty driver. I disabled both audio audiocards ( Internal Realtek and Babyface ) and also ATI HDMI output card on my computer and then run the LatencyMon with madVR in MPC-HC:

MadVR - NO AUDIO CARD

again same bad results on SCSIport.sys and ataport.sys

What I believe is that my system, better said, the way how my drives and PCIe cards are configured can behave badly from the beginning. What should I do?-What kind of test do I need to do, to target the issue/cause? It is the motherboard, it is the PCIe OCZ Revodrive, it is the Perc H700 Raid Controller ( disabled - same results - tested ) or maybe the ATI 5670 card? Why does the DPC spikes show their face when RME USB card is turned on on its USB port - usbport.sys/portcls.sys ? Why does realtek getting same issues when this audio chip is on - hdaudiobus.sys?

Thanks for your opinion and please correct me if my questions are not in the right direction.

Pluto

Edited by Plutotype
Link to comment
Share on other sites

My BIOS is ¨P01-A0 and the update is named R01-B1, does it matter....? I mean, my pc is 2 years old,I bought in 2010. They might have changed the BIOs since then...Can you please tell me the signification of these P01-A0 and etc ?

Gateway R01.B1 1.4 MB 06/29/2010

This is the latest BIOS from 2010:

http://global-download.gateway.com/GDFiles/BIOS/BIOS/BIOS_Gateway_R01.B1_A_A.zip?acerid=634133888894332578&Step1=DESKTOP&Step2=FX%20SERIES&Step3=FX6802&OS=ALL&LC=en&BC=GATEWAY&SC=PA_6G

try it out.

Link to comment
Share on other sites

madVR:

One or more DPC and ISR routines that execute on behalf of drivers running in your system may be causing problems when processing real time audio. You may be experiencing audio dropouts, clicks and pops.

Highest DPC routine execution time (µs): 3666

Responsible driver: SCSIPORT.SYS (SCSI Port Driver, Microsoft Corporation)

DPC count (execution time <500 µs): 2168838

DPC count (execution time 500-999 µs): 0

DPC count (execution time 1000-1999 µs): 0

DPC count (execution time 2000-3999 µs): 378

DPC count (execution time >=4000 µs): 0

Highest ISR routine execution time (µs): 3614

Responsible driver: SCSIPORT.SYS (SCSI Port Driver, Microsoft Corporation)

ISR count (execution time <250 µs): 1692748

ISR count (execution time 250-499 µs): 0

ISR count (execution time 500-999 µs): 0

ISR count (execution time 1000-1999 µs): 0

ISR count (execution time >=2000 µs): 232

Highest pagefault resolution time (µs): 0

Hard pagefault count (total): 0

Number of processes hit: 0

Pagefault count of hardest hit process: 0

Process with highest pagefault count: none

Tests has been done with the RME Babyface on the audio output.

To understand. EVR does not use DXVA, its passing everything to CPU. EVR-CP with DXVA based decoders offloads this work to the GPU processors. MadVR is different, its basically programmed like a D3D video game renderer - all work is done via GPU shaders, it does not support DXVA at all. MadVR outputs in YV12 color format only and its programmed to achieve the best results - with the costs of GPU load.

What I try to say, is that the DPC latency issues rise with the GPU load. Im not skilled in this, but few things come off due to this:

- IRQ issue

- PCIe issue ( check my PCIe configuration - have 3 cards there )

- motherboard issue, chipset issue, bus communication

- GPU issue

-...

I disabled both audio audiocards ( Internal Realtek and Babyface ) and also ATI HDMI output card on my computer and then run the LatencyMon with madVR in MPC-HC:

MadVR - NO AUDIO CARD

again same bad results on SCSIport.sys and ataport.sys

this is confusing. using the GPU decoding instead of CPU causes HDD issues. Which ATI Catalyst drivers do you use? Do you use the 11.2?

What I believe is that my system, better said, the way how my drives and PCIe cards are configured can behave badly from the beginning. What should I do?-What kind of test do I need to do, to target the issue/cause? It is the motherboard, it is the PCIe OCZ Revodrive, it is the Perc H700 Raid Controller ( disabled - same results - tested ) or maybe the ATI 5670 card?

Which motherboard do you use? Do you use the latest BIOS and drivers? Do yo use the latest OCZ Revodrive drivers? Is the Revodrive detected as SCSI device in the device manager?

http://www.ocztechnology.com/drivers/RevoDrive_and_RevoDrive_X2/

Can you give me a "bad" xperf trace?

Link to comment
Share on other sites

this is confusing. using the GPU decoding instead of CPU causes HDD issues. Which ATI Catalyst drivers do you use? Do you use the 11.2?

Which motherboard do you use? Do you use the latest BIOS and drivers? Do yo use the latest OCZ Revodrive drivers? Is the Revodrive detected as SCSI device in the device manager?

http://www.ocztechnology.com/drivers/RevoDrive_and_RevoDrive_X2/

Can you give me a "bad" xperf trace?

- ATI Catalyst - currently 11.1a, will try to update during weekend to 11.2

- device manager

- I have Gigabyte GA-X58a-UD5 Rev 2.0, BIOS FD - latest, chipset driver is actual 9.1.2.1007

- will try to prepare the x perf file

thanks

Edited by Plutotype
Link to comment
Share on other sites

I resolved the issue by disabling core parking in reg edit and also by installing some intel softwares such as; Intel Turbo Boost 2.0, Intel Processor ID utility, Intel Rapid Storage Technology. As for the BIOS, I'm skeptic over it. If you look carefully at my last post you'll know why. Thx a lot for your time and help.

Edited by iRic
Link to comment
Share on other sites

How to get the cause of high CPU usage by DPC (Deferred Procedure Call) and interrupts?

To get started you need the Windows Performance Tools Kit. Read here how to install it:

http://www.msfn.org/board/index.php?showtopic=146919

Now open a command prompt with admin rights (http://windows.microsoft.com/en-US/windows7/How-do-I-run-an-application-once-with-a-full-administrator-access-token) and run the following commands:

xperf -on latency -stackwalk profile

now wait a time while the high DPC and Interrupt usage occurs.

to stop the trace run the following command:

xperf -d DPC_Interrupt.etl

This closes the trace and writes the result to the file DPC_Interrupt.etl.

In the next step, make a double click on the etl file to run the Viewer.

Where I can find the etl file please? There is no such etl file in my computer after the trace is closed. When I start the trace, nothing happens. When I close it, black command prompt appears for a little second, but thats all. Is this correct?

Thanks

Plutotype

Edited by Plutotype
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...