Error message in centos 7 AMD-Vi: Completion-Wait loop timed out

Issues related to hardware problems
Post Reply
Terminator7000
Posts: 1
Joined: 2018/08/25 22:31:39

Error message in centos 7 AMD-Vi: Completion-Wait loop timed out

Post by Terminator7000 » 2018/08/25 23:00:32

I used Rufus to compile a bootable usb drive to install Centos 7 when I try to install I keep getting this error message "AMD-Vi: Completion-Wait loop timed out" After researching in detail I found out that the problem is possibly due to the ryzen 7 1700 I also get error messages after that "Kernel Offset: 0x3e00000 from 0xfffffff81000000 (relocationrange: 0xffffffff800000000-0xfffffffffbfffffff". Current Kernel running is 3.10.0-862.e17.x86_64 and all of the solves that I found on the net told me to install a 3.xx kernel which I have. Does any one know how solve the issue with the AMD-VI... error because it keeps repeating and the installation crashes.

The Current build is
CPU: Ryzen 7 1700
Motherboard: Mohawk B350
Ram: G.SKILL Aegis 16GB 288-Pin DDR4 SDRAM DDR4 2400
Video Cards: RX 480

User avatar
TrevorH
Site Admin
Posts: 33202
Joined: 2009/09/24 10:40:56
Location: Brighton, UK

Re: Error message in centos 7 AMD-Vi: Completion-Wait loop timed out

Post by TrevorH » 2018/08/26 10:38:40

I used Rufus to compile a bootable usb drive to install Centos 7
Did it install or are the problems you have during the install? If they're during the install then I would recommend recreating the USB stick from the iso image, only this time use plain old 'dd' to do it. Rufus in some configurations can alter the iso image as it copies it and will break it.

I did a quick search via google and everything I saw says this problem can be bypassed by passing iommu=soft via the kernel command line. Depending on whether you are using UEFI or legacy BIOS, during the install, when you see the first CentOS menu asking you if you would like to Install, Verify Media and Install, or do Troubleshooting, you should press either the tab key (Legacy BIOS mode) or hit 'e' to edit the menu (UEFI mode) then scroll to the end of the line and remove the words 'rhgb' and 'quiet' (if present) and append a space followed by iommu=soft then hit either Ctrl-X (UEFI mode) or press Enter (Legacy BIOS mode) to boot up.

You might also want to check for more up to date firmware/BIOS for your motherboard.
The future appears to be RHEL or Debian. I think I'm going Debian.
Info for USB installs on http://wiki.centos.org/HowTos/InstallFromUSBkey
CentOS 5 and 6 are deadest, do not use them.
Use the FAQ Luke

calcatinge
Posts: 8
Joined: 2018/04/16 20:43:38
Location: Bucharest, Romania

Re: Error message in centos 7 AMD-Vi: Completion-Wait loop timed out

Post by calcatinge » 2019/04/07 11:15:13

Hey Friends,

I have the same issue with the following setups:

Setup 1:
CPU: AMD Ryzen 7 1700
Motherboard: Gigabyte B450 Aorus M (Latest BIOS installed)
RAM: 32 GB DDR4 Crucial 3000 Mhz
GPU1: Sapphire Radeon RX 580 Nitro+ Special Edition 8GB GDDR5
GPU2: Asus Mining Radeon RX 470 4G GDDR5
SSD: 240 GB WD Green

Setup 2:
CPU: AMD Ryzen 7 1700
Motherboard: ASRock AB350M Pro4 (Latest BIOS installed)
RAM: 32 GB DDR4 Crucial 3000 Mhz
GPU1: Sapphire Radeon RX 580 Nitro+ Special Edition 8GB GDDR5
GPU2: Sapphire Radeon RX 550 2G GDDR5
SSD: 240 GB WD Green

Both systems run CentOS 7.

The issue first started after installing the second GPU on each of the systems (which was last week). I can't even boot the systems, as the errors appear right after BIOS initialization. I can't even boot from a live usb, as I have tried Fedora 29 and it gives me the sames errors.

The errors are like this:

[exerpt]
AMD-Vi: Completion-Wait loop timed out
AMD-Vi: Completion-Wait loop timed out
AMD-Vi: Completion-Wait loop timed out
...
AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=22:00.0 address=0x0000000174bb7560]

Entering emergecy mode. Exit the shell to continue.
Type "journalctl" to view system logs.
....
AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=22:00.0 address=0x0000000174bb75d0]
AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=22:00.0 address=0x0000000174bb7700]
AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=22:00.0 address=0x0000000174bb7630]
AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=22:00.0 address=0x0000000174bb7660]
AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=22:00.0 address=0x0000000174bb7690]
AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=22:00.0 address=0x0000000174bb76c0]
AMD-Vi: Event logged [IOTLB_INV_TIMEOUT device=22:00.0 address=0x0000000174bb76f0]
...
AMD-Vi: Completion-Wait loop timed out
AMD-Vi: Event logged [IO_PAGE_FAULT device=23:00.3 domain=0x000 address=0x00000000fffd3990 flags=0x0070]
...
Buffer I/O error on dev dm-0, logical block 104857472, async page read
Buffer I/O error on dev dm-0, logical block 104857473, async page read
Buffer I/O error on dev dm-0, logical block 104857474, async page read
Buffer I/O error on dev dm-0, logical block 104857475, async page read
Buffer I/O error on dev dm-0, logical block 104857476, async page read
Buffer I/O error on dev dm-0, logical block 104857477, async page read
.....


After some research I discovered that it is a IOMMU issue, but I hoped that it should have been fixed by now...
Thus, I turned IOMMU off on both motherboards, and I managed to boot the system.


I am an architect and I use blender for GPU rendering (this is the idea of having two cards), but the AMDGPU-Pro driver from the AMD's website (the only one that Blender uses for GPU rendering, as it can't use the open one) affects the way Gnome works on Xorg. It affects it in the way that it doesn't start at all, it keeps restarting to the GDM screen...

Now, I don't really know what to do, as a second card would dramatically influence my rendering times so I would like to keep those and not return them to the shop. For example, on the Classroom benchmark in Blender, the rendering time get better by almost 50% while both cards for GPU rendering. This is a great thing, as all clients prefer their projects to be delivered as soon as possible...

Any ideas?

Thanks.

User avatar
TrevorH
Site Admin
Posts: 33202
Joined: 2009/09/24 10:40:56
Location: Brighton, UK

Re: Error message in centos 7 AMD-Vi: Completion-Wait loop timed out

Post by TrevorH » 2019/04/07 13:00:31

Does it not work with iommu=soft then?
The future appears to be RHEL or Debian. I think I'm going Debian.
Info for USB installs on http://wiki.centos.org/HowTos/InstallFromUSBkey
CentOS 5 and 6 are deadest, do not use them.
Use the FAQ Luke

calcatinge
Posts: 8
Joined: 2018/04/16 20:43:38
Location: Bucharest, Romania

Re: Error message in centos 7 AMD-Vi: Completion-Wait loop timed out

Post by calcatinge » 2019/04/08 07:22:21

Well, partially. With iommu=soft it boots, but I can't see anaconda installer, I only have a blank screen and a working mouse pointer...

What I also did was to install using the troubleshoot menu. Anaconda started, I managed to install CentOS, but then, after reboot, I ended up with the following errors (see the atachments).

I managed to make it work using Ubuntu, but then, after a reboot, the system hangs waiting for snapd to initialize...
Attachments
Errors after boot.
Errors after boot.
IMG_20190407_160114x.jpg (68.98 KiB) Viewed 2822 times

Post Reply