Re: [REGRESSION][BISECTED] "xHCI host controller not responding, assume dead" on stable kernel > 6.8.7

[Mario Limonciello]

On 5/20/2024 09:39, Christian Heusel wrote:
> On 24/05/06 02:53PM, Linux regression tracking (Thorsten Leemhuis) wrote:
> > [CCing Mario, who asked for the two suspected commits to be backported]
> >
> > On 06.05.24 14:24, Gia wrote:
> > > Hello, from 6.8.7=>6.8.8 I run into a similar problem with my Caldigit
> > > TS3 Plus Thunderbolt 3 dock.
> > >
> > > After the update I see this message on boot "xHCI host controller not
> > > responding, assume dead" and the dock is not working anymore. Kernel
> > > 6.8.7 works great.
>
> We now have some further information on the matter as somebody was kind
> enough to bisect the issue in the [Arch Linux Forums][0]:
>
>      cc4c94a5f6c4 ("thunderbolt: Reset topology created by the boot firmware")
>
> This is a stable commit id, the relevant mainline commit is:
>
>      59a54c5f3dbd ("thunderbolt: Reset topology created by the boot firmware")
>
> The other reporter created [a issue][1] in our bugtracker, which I'll
> leave here just for completeness sake.
>
> Reported-by: Benjamin Böhmke <benjamin@xxxxxxxxxxx>
> Reported-by: Gia <giacomo.gio@xxxxxxxxx>
> Bisected-by: Benjamin Böhmke <benjamin@xxxxxxxxxxx>
>
> The person doing the bisection also offered to chime in here if further
> debugging is needed!
>
> Also CC'ing the Commitauthors & Subsystem Maintainers for this report.
>
> Cheers,
> Christian
>
> [0]: https://bbs.archlinux.org/viewtopic.php?pid=2172526
> [1]: https://gitlab.archlinux.org/archlinux/packaging/packages/linux/-/issues/48
>
> #regzbot introduced: 59a54c5f3dbd
> #regzbot link: https://gitlab.archlinux.org/archlinux/packaging/packages/linux/-/issues/48

As I mentioned in my other email I would like to collate logs onto a 
kernel Bugzilla.  With these two cases:

thunderbolt.dyndbg=+p
thunderbolt.dyndbg=+p thunderbolt.host_reset=false

Also what is the value for:

$ cat /sys/bus/thunderbolt/devices/domain0/iommu_dma_protection