Bug 572550

Summary:	Input devices stutter and system freezes occur
Product:	[Fedora] Fedora	Reporter:	Kamil Páral <kparal>
Component:	kernel	Assignee:	Kernel Maintainer List <kernel-maint>
Status:	CLOSED RAWHIDE	QA Contact:	Fedora Extras Quality Assurance <extras-qa>
Severity:	medium	Docs Contact:
Priority:	low
Version:	13	CC:	anton, dougsland, dwmw2, gansalmon, itamar, jonathan, kernel-maint, peter, virt-maint
Target Milestone:	---
Target Release:	---
Hardware:	All
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2010-03-22 11:56:20 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:
Bug Blocks:	507681

Description Kamil Páral 2010-03-11 14:20:00 UTC

Description of problem:
I just installed Fedora 13 Alpha. My mouse works, but it stutters heavily. Every few seconds it lags and the movement is not fluent. My keyboards also shows a little stuttering every few seconds when I hold some key.

I think it could be related to these messages in dmesg output, there's a lot of them:

Mar 11 15:14:51 localhost kernel: DMAR:[DMA Write] Request device [00:02.0] fault addr 121151000 
Mar 11 15:14:51 localhost kernel: DMAR:[fault reason 05] PTE Write access is not set
Mar 11 15:14:51 localhost kernel: DRHD: handling fault status reg 3

Logs are attached.

Version-Release number of selected component (if applicable):
Fedora 13 Alpha
kernel-2.6.33-0.52.rc8.git6.fc13.x86_64
ThinkPad T500

How reproducible:
all the time

Steps to Reproduce:
1.
2.
3.
  
Actual results:


Expected results:


Additional info:

Comment 1 Kamil Páral 2010-03-11 14:42:46 UTC

Also Rawhide freezes for me after circa 10 minutes of usage. I think it could be related.

Comment 2 Kamil Páral 2010-03-11 14:43:10 UTC

s/Rawhide/F13 Alpha

Comment 3 Kamil Páral 2010-03-18 14:06:47 UTC

I have had another system freeze, this time after around 30 minutes. Every time I see those DMA errors in dmesg long before that.

I have also observed the same system freeze on another Thinkpad T500 notebook.

Comment 4 Kamil Páral 2010-03-19 09:05:39 UTC

jskladan, you have the same hardware, could you please confirm this issue too (whether you see those DMA errors and whether it freezes for you)? Thanks.

Those DMA errors seems not to be triggered right after boot, but only after some time. When they happen and flood the /var/log/messages, then the system usually freezes in a few minutes. Maybe it is related to disk activity.

Also I have a suspicion that I see exactly those three error lines right after booting kernel in grub, in the system console (it lasts only a second and then changes to graphical/text boot screen).

Comment 5 Kamil Páral 2010-03-19 11:56:40 UTC

I have found bug #479996, that concerns probably the same issue.

Following information from that bug:
1. disabling Intel VT technology resolves the problem (no DMA errors in dmesg)
2. keeping Intel VT technology enabled but appending intel_iommu=off kernel option resolves the problem too

Comment 6 David Woodhouse 2010-03-19 12:40:02 UTC

Does it go away if you boot with 'intel_iommu=igfx_off'?

If so, it should be fixed in F-13 and rawhide now that I've ported over the fix for bug #538163. As soon as those kernels finish building.

Comment 7 Kamil Páral 2010-03-19 13:03:19 UTC

David, it looks good, I think that kernel option helps too. I will re-test the new kernels and report. Can you tell me in which version it should be fixed so I can verify it?

Comment 8 David Woodhouse 2010-03-19 13:06:30 UTC

For rawhide, kernel-2.6.34-0.13.rc1.git1.fc14 at http://koji.fedoraproject.org/koji/taskinfo?taskID=2062865

For F-13, kernel-2.6.33.1-17.fc13 at http://koji.fedoraproject.org/koji/taskinfo?taskID=2062867

Comment 9 Kamil Páral 2010-03-22 09:08:37 UTC

David, I have installed kernel-2.6.33.1-17.fc13 and I don't see any problems now. I believe it's fixed. Well done.

Comment 10 David Woodhouse 2010-03-22 11:56:20 UTC

Just to be clear, this workaround applies _only_ to spurious faults on the built-in graphics device, seen on this _specific_ chipset (rev. 07):

00:00.0 Host bridge [0600]: Intel Corporation Mobile 4 Series Chipset Memory Controller Hub [8086:2a40] (rev 07)