bug kernel "AMD-Vi : Event logged [IO_PAGE_FAULT device=01:00.0 domain=0x0013 address=0x0000000000001000 flags=0x0000]"
How to disable (fixing ?) this permanently bug (only for ubuntu) from any kernel >=3.5.0 ?
Can use dmesg only this messages - /var/log/syslog* and /var/log/kern.log* = size this file is many, many GB (hehe).
Question information
- Language:
- English Edit question
- Status:
- Expired
- For:
- Ubuntu Edit question
- Assignee:
- No assignee Edit question
- Last query:
- Last reply:
Revision history for this message
|
#1 |
My komputer:
iAMD Bulldozer 8x (first model); graphics card Nvidia GTX260; 8 GB RAM DDR3; 2 x OCZ-VERTEX3 120 GB (Raid-0); 2x WD1002FAEX-00Z3A0 1 TB (Raid-0); motherboard MSI 990FXA-GD80
Revision history for this message
|
#2 |
Actually use ubuntu-13.04-x86_64 - last stable kernel 3.8.0-29-generic.
Revision history for this message
|
#3 |
Have you tested your RAM using Memtest86+ from Grub?
Revision history for this message
|
#4 |
Yes - its all ok - no anny errors - test x 5 - many, many hours...
Revision history for this message
|
#5 |
RAM config to 1600 Hz - native frequency
Revision history for this message
|
#7 |
What is the output of :
lspci
Thanks
Revision history for this message
|
#8 |
My lspci output:
http://
Revision history for this message
|
#9 |
Windows 8 not any bad report and bugs from my computer - normaly and stable working. This situation only for ubuntu and meayby debian... (any other distro based to *.deb or not ? - like linux-kernel >= 3.5.0* )
Revision history for this message
|
#10 |
01:00.0 VGA compatible controller: NVIDIA Corporation GT200 [GeForce GTX 260] (rev a1)
Corelates to your error nicely. What is the output of:
sudo lshw -C display; lsb_release -a;uname -a;dpkg -l | grep nvidia
Thanks
Revision history for this message
|
#12 |
can use native drive (opensource) its the same error
Revision history for this message
|
#13 |
can use any other nonfree drive - its the same bug - only for kernel .>= 3.5.0
Revision history for this message
|
#14 |
ok try:
sudo apt-get --purge remove `dpkg -l | grep nvidia | grep -v nvidia-common | awk {'print $2'}`
sudo apt-get --purge autoremove
sudo apt-get clean
sudo apt-get update
sudo apt-get install nvidia-current
reboot to test
Revision history for this message
|
#15 |
*drive=drivers=(?) => sorry my bad english
can use nvidia-current its the same...testing before...ok. use this tips again and print result and paste link
Revision history for this message
|
#16 |
http://
can i'am and you discussion for this session - /var/log/kern.log = 58, 7 MB !!! - minutes...
Anny change...its the same bug:
AMD-Vi: Event logged [IO_PAGE_FAULT device=01:00.0 domain=0x0013 address=
Revision history for this message
|
#17 |
Xorg.0.log
http://
Revision history for this message
|
#19 |
This question was expired because it remained in the 'Open' state without activity for the last 15 days.
Revision history for this message
|
#20 |
Actually using this idioten patented, usig this command for /etc/rc.local:
rm -f /var/log/kern*
rm -f /var/log/dmesg*
rm -f /var/log/syslog*
Revision history for this message
|
#21 |
This question was expired because it remained in the 'Open' state without activity for the last 15 days.
Revision history for this message
|
#22 |
HI,
I have same problem.
Is there any solution ?
Revision history for this message
|
#24 |
latest kernel official or not officjal (3.12.6-
Revision history for this message
|
#25 |
Solved with the last available nVidia driver nvidia-319-updates
(319.60-0ubuntu1).
Linux 3.11.0-15-generic
Le 23/01/2014 23:16, lulo a écrit :
> Question #234598 on Ubuntu changed:
> https:/
>
> lulo posted a new comment:
> latest kernel official or not officjal (3.12.6-
> #201312201218 SMP Fri Dec 20 17:20:06 UTC 2013 x86_64 x86_64 x86_64
> GNU/Linux) its the same problem...
>
Revision history for this message
|
#26 |
Oooops, error, yesterday the bug seemed lost, but now he's back.
Arghhhh !!!
What can we do, please help, hundreds of messages per second in kernel.log
Revision history for this message
|
#27 |
I've hit this bug too with a GeForce GTX 260, my entire disk was filled with syslog/kern messages like:
Feb 13 10:58:11 chie kernel: [ 119.126741] AMD-Vi: Event logged [IO_PAGE_FAULT device=01:00.0 domain=0x0014 address=
lspci shows device 01:00.0:
01:00.0 VGA compatible controller: NVIDIA Corporation GT200 [GeForce GTX 260] (rev a1)
Downgraded my kernel from 3.5.0-45 (recent system update) to 3.5.0-28-generic (version before system update) and issue was resolved.
Revision history for this message
|
#28 |
Anything new here? I have the same problem with a "NVIDIA Corporation G92 [GeForce 8800 GT] (rev a2)" on a Gigabyte 990XA-UD3 with AMD FX-6300 processor.
I tried all availlable drivers and all show this error save the NOUVEAU which shows another error in the same verbosity ;)
Revision history for this message
|
#29 |
Solved using this kernel commandline option : iommu=pt
This problem appears when IOMMU is enabled in the BIOS, otherwhise USB2 and network are broken.
We can work around this by disabling the hw iommu. To do so, add iommu=pt to the kernel commandline and make sure you have no other iommu options set.
In file /etc/default/grub, change
GRUB_CMDLINE_
to
GRUB_CMDLINE_
Revision history for this message
|
#30 |
oboumati: CONFIRMED
1. Disabling IOMMU to my UEFI (disabling iommu hw)
2. Add "iommu=pt" to my /etc/default/grub
3. sudo update-grub2
4. reboot
5. dmesg not any line type:
"AMD-Vi: Event logged [IO_PAGE_FAULT device=01:00.0 domain=0x0014 address=
6. SOLVED
PS.:
Question:
How to enabling HW IOMMU to my UEFI and not this bug ?
Revision history for this message
|
#31 |
In my case, hardware IOMMU MUST be ENABLED in bios configuration, otherwhise USB2 and network are broken !!
The "iommu=pt" command line option is required in this case only.
If you disable hardware IOMMU in your uefi bios, the command line option is unnecessary
Revision history for this message
|
#32 |
oboumati:
UPS...CONFIRMED:
1. Enabling this futers to my UEFI again [IOMMU=64M]
2. Add "iommu=pt" to my /etc/default/grub
3. sudo update-grub2
4. reboot
5. dmesg not any line type:
"AMD-Vi: Event logged [IO_PAGE_FAULT device=01:00.0 domain=0x0014 address=
6. SOLVED
Hmmm - no any problems to USB2 and network - meaby only to other version (old - confirmed) firmware and any old kernel versions to this motherboard (my motherboard: MSI 990FXA-GD80; firmware v13.5), and other motherboards...
Revision history for this message
|
#33 |
actualy using this tips to ubuntu-14.04 x86_64 (packages state: 29-03-2014 r. - 16:03)
no any problems to IOMMU (this bug)
SOLVED
Revision history for this message
|
#34 |
working on ubuntu 14.04 final, bug still exists.
Adding the opion "iommu=pt" to the grub-commandline and update-grub2 solves the problem.
Revision history for this message
|
#36 |
Effectively helps to use the same option in the boot loader "iommu=pt"