Bug#1102914:

[Salvatore Bonaccorso <carnil@debian.org>]

Hi,

On Mon, Apr 14, 2025 at 02:01:07PM +0200, Milan Broz wrote:
> Hi,
> 
> we have the same issue with 6.1.133-1 (previous 6.1.129-1 works fine).
> 
> We use passthrough of NVMe device, VM start now fails with
> 
> qemu-system-x86_64: -device {"driver":"vfio-pci","host":"0000:03:00.0","id":"hostdev0","bus":"pci.3","addr":"0x0"}: VFIO_MAP_DMA failed: Cannot allocate memory
> qemu-system-x86_64: -device {"driver":"vfio-pci","host":"0000:03:00.0","id":"hostdev0","bus":"pci.3","addr":"0x0"}: vfio 0000:03:00.0: failed to setup container
> 
> I cannot easily do bisect on this machine, but if you have problem to reproduce it, I can probably try to setup other machine and run upstream bisect there.
> 
> Anyway, I think this is a major regression.

As asked in #1102889 as well, would it be possible to get a full
kernel log as well please?

I have a suspect that it might be related to
https://lore.kernel.org/stable/20250407181218.GA737271@ax162/
so far, but need to verify it.

Regards,
Salvatore