![]() |
Hardware interrupts spiking |
Post Reply
|
Page 12> |
| Author | ||
Buggy Buddha
Senior Member
Joined: 03 March 2006 Status: Offline Points: 116 |
Post Options
Thanks(0)
Quote Reply
Topic: Hardware interrupts spikingPosted: 05 September 2006 at 4:50am |
|
|
I must confess I don't understand what the hardware interrupts displayed in Process Explorer are and how they work. I'd like to know if it's normal for hardware interrupts to ever spike to 50% levels (this is on a dual core system, so it's probably taking one of the cores to 100% CPU usage)? I recently noticed that this would happen when I was using a program to compare a large amount of files on DVD to the same files on HDD. The program, I think, counts hashes of files both on HDD and DVD and compares the hashes. Should something like this cause hardware interrupts to spike like that, or is it abnormal?
In general, I'd like to know what would cause hardware interrupts and how much CPU they'd normally take. TIA. :) |
||
![]() |
||
namrehto
Senior Member
Joined: 23 June 2005 Location: Scotland Status: Offline Points: 3866 |
Post Options
Thanks(0)
Quote Reply
Posted: 05 September 2006 at 5:02am |
|
|
Hardware interrupts occur when a peripheral wants attention from the CPU. It sounds like your DVD drive might have got into PIO transfer mode. Have a look at Interrupts spikes to 80% of CPU and see if it helps.
|
||
|
Gil
|
||
![]() |
||
Buggy Buddha
Senior Member
Joined: 03 March 2006 Status: Offline Points: 116 |
Post Options
Thanks(0)
Quote Reply
Posted: 05 September 2006 at 6:24am |
|
|
Thanks, I'll check if it's in PIO mode and try that solution if it is. It's a new system, though, and I don't really like the idea it would just revert to PIO mode on its own. Broken hardware? I kind of hope not, seeing how new the system is.
Is it ever normal to see hardware interrupts take a notable amount of CPU time? Or does it always imply some kind of a software/hardware issue? Edit: one more thing, this is an XP SP 2 system. Edited by Buggy Buddha - 05 September 2006 at 6:33am |
||
![]() |
||
namrehto
Senior Member
Joined: 23 June 2005 Location: Scotland Status: Offline Points: 3866 |
Post Options
Thanks(0)
Quote Reply
Posted: 05 September 2006 at 6:34am |
|
|
In general interrupts shouldn't be excessive. If either they or DPCs are high it's commonly a driver issue, sometimes hardware.
From the PE help file: "Interrupts and DPCs On Windows NT-based systems Process Explorer shows
two artificial processes: Interrupts and DPCs. These processes reflect the
amount of time the system spends servicing hardware interrupts and Deferred
Procedure Calls (DPCs), respectively. High CPU consumption by these activities
can indicate a hardware problem or device driver bug. To see the total number of
interrupts and DPCs executed since the system booted add the Context Switch
column. Another sometimes useful metric is the number of interrupts and DPCs
generated per refresh interval, which you see when you add the CSwitches Delta
column." |
||
|
Gil
|
||
![]() |
||
Buggy Buddha
Senior Member
Joined: 03 March 2006 Status: Offline Points: 116 |
Post Options
Thanks(0)
Quote Reply
Posted: 05 September 2006 at 6:44am |
|
|
I see. I read that from the manual, but didn't quite understand whether it's always abnormal to see high CPU consumption by hardware interrupts or if it can even be expected in certain normal situations. Comparing the hashes of files has to stress the CPU, memory, and the hard drive and DVD drive alike, right? The program that did the hashing showed practically no CPU usage during the operation, but hardware interrupts stayed high. Originally, I thought it made sense, since the system was reading a lot of files from the HDD and DVD at the same time. The operation completed, too, and I didn't notice any system slowdown, though that may be just because it's a dual core system. But it'll be interesting to see if one of the devices has gone to PIO mode. If it hasn't, then it's indicative of an actual hardware failure, then?
Edited by Buggy Buddha - 05 September 2006 at 6:45am |
||
![]() |
||
namrehto
Senior Member
Joined: 23 June 2005 Location: Scotland Status: Offline Points: 3866 |
Post Options
Thanks(0)
Quote Reply
Posted: 05 September 2006 at 7:26am |
|
|
If an IDE controller kept reverting to PIO mode (as a result of cumulative errors) there's a possibility of a hardware issue. As a first instance checking all the cables/connectors would be a good idea, and making sure there's nothing else on the same IDE cable that could be interfering. Also confirm that the correct driver's installed (usually the standard Windows one). Manufacturer's diagnostics might help.
One wouldn't have thought that hash checking of files should be any more stressful on the interface to the DVD drive than simply copying files, and that ought to impose a pretty low interrupt load. |
||
|
Gil
|
||
![]() |
||
Buggy Buddha
Senior Member
Joined: 03 March 2006 Status: Offline Points: 116 |
Post Options
Thanks(0)
Quote Reply
Posted: 05 September 2006 at 8:31am |
|
|
Looks like you were absolutely right. I checked the controller properties, and according to that, the DVD drive is in fact running in PIO mode. Windows even gave a nice warning in that menu, explaining that: "The transfer mode for this device was set to a mode lower than what it is capable of because of excessive transfer errors to that device. This will cause a loss of system performance. Please check the cabling to the device..." Nice. Going to set it back to Ultra DMA 2 mode and reboot. That does leave me to wonder what caused it to revert to PIO mode in the first place, though.
|
||
![]() |
||
Icon
Newbie
Joined: 05 September 2006 Status: Offline Points: 3 |
Post Options
Thanks(0)
Quote Reply
Posted: 05 September 2006 at 12:37pm |
|
|
My Hardware Interrupts are also showing at 50% but I do not think it's the DVD drive.I only have a Samsung CD writer/DVD reader and the IDE controllers are showing 'DMA if available'
Go easy on me as I hope I have understood the above but if I haven't please let me know. |
||
![]() |
||
namrehto
Senior Member
Joined: 23 June 2005 Location: Scotland Status: Offline Points: 3866 |
Post Options
Thanks(0)
Quote Reply
Posted: 05 September 2006 at 1:52pm |
|
|
@Icon: Open a new thread in the Troubleshooting forum for your problem. If the interrupts are pretty much constant - i.e. don't only occur during I/O - then one would suspect an issue with sound/video, possibly a driver problem.
|
||
|
Gil
|
||
![]() |
||
Buggy Buddha
Senior Member
Joined: 03 March 2006 Status: Offline Points: 116 |
Post Options
Thanks(0)
Quote Reply
Posted: 05 September 2006 at 3:23pm |
|
Quoting myself, but I've been trying to find the reason the DVD-drive went to PIO. I found this MS article that may be useful to others, as well: http://www.microsoft.com/whdc/device/storage/IDE-DMA.mspx The most interesting thing I read there is this:
So, any errors that cause Windows to revert to PIO mode should be logged in the system event log. My event log absolutely does not have any such log entries. So are we to assume that Windows did not switch PIO mode on because of errors, but something else happened instead? |
||
![]() |
||
Post Reply
|
Page 12> |
|
Tweet
|
| Forum Jump | Forum Permissions ![]() You cannot post new topics in this forum You cannot reply to topics in this forum You cannot delete your posts in this forum You cannot edit your posts in this forum You cannot create polls in this forum You cannot vote in polls in this forum |