On 2022-03-09 23:23, Maximilian Engelhardt wrote: > I have no idea what's going on here. Is this a bug in the Linux kernel? The > kernel I did my tests with is 5.16.0-2-armmp-lpae. > I could imagine this is a race somewhere, which might also explain > Christian's > observations with arm64. > > I still have an old autopkgtest armhf image from Jun 6 2021 that has linux > 5.10.0-7-armmp-lpae and doesn't show that behaviour. So some change after > that > must have broken the console numbering.
Another data point in support of this possibly being a kernel issue: I created buster and bullseye images for armhf (kernel armmp-lpae), both worked fine. In the bullseye image, I then replaced the default 5.10 kernel with 5.14 (the earliest released version after 5.10) from snapshots.d.o, and then it hanged again. Manual boot continued to work fine, so it couldn't have been that. I'm also leaning towards a race, but the question is whether this is really a potential bug, or whether determinism is assumed where none is guaranteed (similar to the issue of storage devices like sda and sdb being non-deterministic). Anybody got an idea whom we could ask? Seems to me that the QEMU devs would be those with the largest interest...
