[MagicAndre1981]

upload the new trace without the SPTD driver.

[rcollings]

Hi MagicAndre (and everybody else) - I wonder whether you could help with my problem

I am having a problem with high CPU usage by DPCs. The problem is intermittent. It doesn't happen when the machine is idle but kicks in after a period of use and as I start to use more programs. It will go on for 5-10 mins and will some time stop by itself. It usually stops if I close down a few programs. It is not tied to any one program as far as I can see.

I have followed your diagnostic procedure (I think - not being a developer, I am not an expert in these matters).

The first problem I had was that I got a message saying "Session NT Kernal logger lost 2267 event" and that I need to increase the buffer size and/or number of buffers. Is this a problem and if so, how do I do this)

Secondly, when I get the graph, there is only activity for the first 10s and only a few percent of the CPU. I ran the logging process for a lot longer than this and the high CPU usage only kicked in after a while

Anyway, I selected the area where there was some activity and ran the analysis. Here's what I got:
 26-04-2011 11-39-08.png (81.16K)
Number of downloads: 4

Is this any good? What does it mean? And what should I do next?

Many thanks for your help (in advance!)

[MagicAndre1981]

add this to the command line to increase the buffersize:

-BufferSize 1024

the USB port driver has some issue. Which USB devices (HDD, USB flash drives) to you use?

[rcollings]

MagicAndre1981, on 26 April 2011 - 05:18 AM, said:

add this to the command line to increase the buffersize:

-BufferSize 1024

the USB port driver has some issue. Which USB devices (HDD, USB flash drives) to you use?

Hi Andre - thanks for prompt rely.

Had rerun the trace for a shorter period when the DPC was already in progress and captured a trace without the error. Here is the resulting display:
 26-04-2011 13-39-50.png (183.71K)
Number of downloads: 7
which is not dramatically differnt to the first one

So its something to do with my usb devices/drivers?

In answer to your question about USB devices - I have lots: external drives, sound cards, graphics tablets, mice, printers, etc, etc.

Its a Dell Latitude D820 laptop. Early on the process, when it looked like it might be the USB devices, I pulled out the two USB cables that connect to two hubs that connect most of my devices. The high CPU usage did not go away immediately so I assumed that it wasn't anything to do with USB. However....

1) Does disconnecting the device remove the driver? Or to put it another way, if I disconnect a device and the problem doesn't go away immediately, can I assume that that that device and driver are not the cause of the problem?

2) I think the laptop has a number of devices that are connected internally via USB. I am not sure how I identify these.

Incidentally, as far as I am aware, this is a new problem on an existing machine which did not have the problem before. As far as I am aware, I have not installed any new devices recently

Can you suggest some next steps to try and track the problem down. I have uploaded the trace file here:
trace file

Many thanks for your help (and for your guidelines)

Update: problem resolved. Turns out it was an external Creative USB Sound card. I went back and pulled out all the USB cables and this time I was specifically monitoring the processor usage by the DPCs rather than just looking at the overall CPU usage. As before the overall CPU usage remained high but the DPC usage dropped to zero. And then it was just a question of trying each device in turn until I found the one that restarted the DPC usage. Kicking myself now for not being more thorough in my my original testing - could have saved myself several hours. But your diagnostic tools pointed me firmly back at it being a USB device somewhere so we got there in the end. I was close to going out an buying a new machine - which I do need - but I can do so in a more measured fashion.

Many thanks

Richard

Richard

This post has been edited by rcollings: 26 April 2011 - 08:57 AM

[MagicAndre1981]

remove each USB device 1 followed by the other (start with the WD USB HDD, please) to see which USB device causes the USBPort driver issue.

[hottroc]

Hi, I'm wondering if I spoke too soon in my last post. Maybe I hadn't rebooted or something (which I have now) but now, he says rather tentatively, it seems to be behaving itself. So I can't upload a new trace yet because it's not exhibiting the same behaviour. But I'm not totally sure its fixed yet so I'll keep my eye on it over the next few days and report back.

In the meantime can you tell me....1) Is this sptd driver problem also likely to have caused the other issues I mentioned briefly in post #390?...and 2) Are there likely to be any undesired effects from uninstalling the driver? I think Active ISO Burner may use it that I have installed but any others?

Many many thanks for your help on this. The problem has been bugging me for ages so if you have fixed it, which is looking promising at the mo, I will be really in your debt.
Regards,
hottroc.

[MagicAndre1981]

I have no idea what causes your other hard drive issues. The driver is used by ISO mounting tools, no burning tools.

[diddly]

I've started getting 40% cpu usage (windows cpu/mem gadget, although task manager reports 0%). I tried the steps at the start of this thread and the result is "Unknown" in the DPC CPU usage summary and "PCIIDEX.SYS" in the interrupt CPU usage summary. Any ideas on what to do next would be great.

[MagicAndre1981]

which PCI IDE devices (HDD, DVD ROM/Burner) do you still use?

[diddly]

I'm using SATA connectors for both DVD + HD. So hopefully I'm not using IDE at all. Does this mean I should disable any IDE controllers via device manager?

[MagicAndre1981]

diddly, on 29 April 2011 - 08:21 AM, said:

Does this mean I should disable any IDE controllers via device manager?

try it out. Create a system restore point first.

[diddly]

Tried it out and it seems to have worked, CPU now ~2%. Fingers crossed all is well. Cheers

[hottroc]

Me again, it seems yet again I spoke too soon. The issue has returned. I think this proves it is not the sptd driver which was completely uninstalled. I have attached a link to a new trace.
Please let me know if you have any other ideas what might cause the interrupt problem? Are there any other things that show as UNKNOWN in this situation?

Thanks.

Link:
http://cid-ac6270209...0209775C627!162

Regards,
hottroc.

PS I increased the buffersize in this trace as per one of your earlier posts.

EDIT - having looked at this new trace now myself I see that "UNKOWN" has disappeared, replaced by something else (USBPORT.SYS) so maybe the Unknown WAS to do with the sptd driver but it was somehow masking the real driver/problem. I wonder if this could point to the external USB-SATA drive caddy/cradle I have attached? That might make sense perhaps? Although I think I tried unplugging it in the past to no avail but I'll check.

This post has been edited by hottroc: 29 April 2011 - 09:35 AM

[MagicAndre1981]

diddly, on 29 April 2011 - 08:48 AM, said:

Tried it out and it seems to have worked, CPU now ~2%. Fingers crossed all is well. Cheers

ok, nice to hear this.

hottroc, on 29 April 2011 - 09:02 AM, said:

EDIT - having looked at this new trace now myself I see that "UNKOWN" has disappeared, replaced by something else (USBPORT.SYS) so maybe the Unknown WAS to do with the sptd driver but it was somehow masking the real driver/problem. I wonder if this could point to the external USB-SATA drive caddy/cradle I have attached? That might make sense perhaps? Although I think I tried unplugging it in the past to no avail but I'll check.

it maybe the USB HDD. Does it happen if you unplug the HDD?

[hottroc]

MagicAndre1981, on 29 April 2011 - 04:07 PM, said:

it maybe the USB HDD. Does it happen if you unplug the HDD?

Hi, well when the problem occurred I tried turning off and disconnecting the ext. HDD but it made no difference. So now I'm waiting to see if the problem occurs after having rebooted and leaving the ext HDD disconnected completeley. We should know soon enough.

[hottroc]

Hi, now we know, the problem re-occurs even with the external HDD caddy completely disconnected. So it's not that.

Tbh, I was hoping there was a way to find out the cause of the interrupt problem without going through a process of elimination of every hardware and software driver possible. I was hoping that was what this topic would give me.

[MagicAndre1981]

make a new trace and upload it.

[Illisvyel]

Hi guys, wanted to ask for some help.

I've been trying for ages to fix a very annoying problem with my computer which I've been experiencing since last year.

Whenever I use Skype and play a game or play music while playing, I can hear popping noises from my speakers, which lead me to believe that it was a problem with the drivers but updating did not fix it at all.
Yesterday I found this thread and I used Latencymon while the popping noises were occurring and it tells me that the responsible driver that is causing the popping noises is this one - "i8042prt.sys"

Isn't this a Windows 7 system file? If so, how can I fix my problem?

Thanks in advance.

This post has been edited by Illisvyel: 05 May 2011 - 02:09 PM

[MagicAndre1981]

i8042prt.sys = PS2 mouse/keyboard driver

which PS2 devices do you use?

[Illisvyel]

Only thing PS2 related I can think of is the PCSX2 emulator.
Could this be the cause?