On Sat, 28 Mar 2026 21:20:24 +0100 Salvatore Bonaccorso
<[email protected]> wrote:
> Hi Ariel,
>
> On Sat, Mar 28, 2026 at 09:07:21PM +0100, Ariel wrote:
> > The problem is that the patches mentioned in message #69 aren't
available
> > (404), so I looked for patches with similar names to see if they were
> > available elsewhere (those are the ones I linked to in my previous
message).
> > I tried the Debian test version to see if they were included and
would fix
> > the problem. But it doesn't seem so.
>
> The patch to test from message #69 is really just the quirk one, it is
> not commited anywhere. What I would be interested to hear if the quirk
> fixes your problem, independly of the Debian specific patches we have.
>
>
> --- a/drivers/iommu/intel/iommu.c 2022-06-03 14:50:52.248268257 +0200
> +++ b/drivers/iommu/intel/iommu.c 2022-06-03 14:48:12.695769217 +0200
> @@ -6186,6 +6186,9 @@
> dmar_map_gfx = 0;
> }
>
> +/* Q35 integrated gfx dmar support is totally busted. */
> +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x29b2, quirk_iommu_igfx);
> +
> /* G4x/GM45 integrated gfx dmar support is totally busted. */
> DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2a40, quirk_iommu_igfx);
> DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2e00, quirk_iommu_igfx);
>
> Regards,
> Salvatore
>
>
I can confirm that after applying this kernel patch, the system no
longer hangs and boots correctly even with VT for Direct I/O enabled.
The source I used for the test is linux-source 6.19.8 in forky/sid.
Kind regards,
Ariel.
