IO Problems (USB) seem to hang on Boot (AMD64)

[Micheal Smith]

To all Glorious Saviors:

Installation goes smooth. When asked, during installation, I install
device-specific drivers. For the record, this is the AMD64 distribution.

So, whenever my IOMMU Contoller is enabled on the UEFI of the motherboard,
I get this error: [ 1.669503] AMD-Vi: Event Logged [IO_PAGE_FAULT
device=2:00.0 domain=0x0017 address=0x00000000be9f9880 flags=0x0010]
filling my screen.

When IOMMU is disabled, I get along the lines of [ 2 OR 3 OR 4 OR 5 .
(randomnumber)]usb 3-5 device descriptor read/64, error -32.

Somehow, I've also generated an error along the lines of: [ 20.979743]
nouveau 0000:01:00.0: PMC unhandled INTR 0x44000000

The only success I've had in dealing with this is when I tried to use
clonezilla, it was having serious problems. I was able to get *it* working
(the Ubuntu-based version) with the following command in the boot
parameters: "quiet splash usbcore.old_scheme_first=1". However, no such
luck with Debian....

The install process is fine. I just *can't* get it to boot.

Sincerely,
A User

[Darac Marjal]

Installation goes smooth.  When asked, during installation, I install
device-specific drivers.  For the record, this is the AMD64 distribution. 
So, whenever my IOMMU Contoller is enabled on the UEFI of the motherboard,
I get this error: [  1.669503] AMD-Vi: Event Logged [IO_PAGE_FAULT
device=2:00.0 domain=0x0017 address=0x00000000be9f9880 flags=0x0010]
filling my screen.

Apparently, this is a known issue with the AMD-VI IOMMU. The solution
(according to the interwebs) is to add "iommu=soft" to the kernel
command line (and disable the hardware IOMMU). Hopefully, the errors
below are due to the lack of proper IOMMU and using the software IOMMU
will help.

When IOMMU is disabled, I get along the lines of [    2 OR 3 OR 4 OR 5 .
(randomnumber)]usb 3-5 device descriptor read/64, error -32.  
Somehow, I've also generated an error along the lines of: [   20.979743]
 nouveau 0000:01:00.0: PMC unhandled INTR 0x44000000
The only success I've had in dealing with this is when I tried to use
clonezilla, it was having serious problems.  I was able to get *it*
working (the Ubuntu-based version) with the following command in the boot
parameters: "quiet splash usbcore.old_scheme_first=1".  However, no such
luck with Debian....
The install process is fine.  I just *can't* get it to boot.
Sincerely, 
A User

[Micheal Smith]

So, I'm still having serious issues with this thing: I used the following
procedure to update the grub-bootloader, from Linux Mint:

mount /dev/sda6 /mnt

sudo mount --bind /dev/ /mnt/dev

sudo mount --bind /proc/ /mnt/proc

sudo mount --bind /sys /mnt/sys

cp /etc/resolv.conf /mnt/etc/resolav.conf

sudo chroot /mnt

I then changed the line:
GRUB_CMDLINE_LINUX="" to GRUB_CMDLINE_LINUX="iommu=soft" and ran
update-grub. I do have IOMMU disabled, and I did the same thing to my
Linux Mint Intallation, and now I can run *that* without IOMMU support
enabled. I reloaded optimized defaults on the UEFI, and I saw my change in
GRUB2.

Now, I consistently get another error message when trying to boot.
[ 3.(something)] [drm] 0000:01:00.0: PMC unhandled INTR 0x44000000

The error messages from recovery mode may be slightly more insightful:


[ 3.(something)] [drm] nouveau 0000:01:00.0: Register 0x00680520 not
found in PLL limits table
[ 3.(something)] [drm] Supports Vblank Timestamp Caching Rev 1
(10.10.2010)
[ 3.(something)] [drm] No driver support for vblank timestamp query.
[ 3.(something)] [drm] nouveau 0000:01:00.0 voltage table 0x50 Unknown
[ 3.(something)] [drm] nouveau 0000:01:00.0: Register 0x.... not found
in PLL limits table
[ 3.(something)] [drm] nouveau 0000:01:00.0: Failed to determine boot
perflvl
[ 3.(something)] [drm] nouveau 0000:01:00.0: Pixel Clock Comparison
table not found

Otherwise this is a 100% squeeky clean fresh install of Debian. A great
column of the item listed above this list precedes it.

Now I have no idea what is wrong....

Thanks Again,
A user