Maximilian Luz

Potentially related to #1271.

> The issue should be a bad order in which kernel modules are loaded. Yeah, it does look like that. > I should try to edit an initrd with mkinitcpio...

@Leo1998 Thanks!

This should be fixed. Please let me know if the issue still occurs.

I've received some reports via mail and here's the summary: Testing via [this module](https://github.com/linux-surface/surface-shutdown) (and variations) indicates that the issue is some problem in the firmware. In particular: When inserting...

Something more concrete to test: Since the module returns, can someone try replacing the [content of `surface_shutdown_init()`](https://github.com/linux-surface/surface-shutdown/blob/98dce8b674843c8fe1b4888847c806f06105b1bb/src/surface_shutdown.c#L21-L24) with ```c pr_warn("Calling efi.reset_system(EFI_RESET_SHUTDOWN, ...)"); efi.reset_system(EFI_RESET_SHUTDOWN, EFI_SUCCESS, 0, NULL); pr_warn("Call to efi.reset_system(EFI_RESET_SHUTDOWN, ...)...

@WolfWings This is very likely a different root problem. Can you open a new issue? I'll respond and answer to your questions there. On the SP9, we've now nailed this...

We managed to track down this issue to a fairly limited scope yesterday on our Matrix chat. It turns out that EFI's ResetSystem works fine before calling some shutdown related...

Limiting it further: The issue is likely caused by one (or multiple) of the following devices (SP9): ``` i915 0000:00:02.0 pcieport 0000:00:06.0 thunderbolt 0000:00:0d.3 xhci_hcd 0000:00:14.0 mei_me 0000:00:16.0 nvme 0000:02:00.0...

Another write-up from our findings in Matrix chat: First off, we have a workaround: https://github.com/linux-surface/kernel/commit/6f57d3f0780712e9b6cea1ee3b87605d2920e8ae. The problem seems to be a combination of the `pm_runtime_resume()` and the `drv->shutdown()` call, as...