On Thu, Nov 18, 2021 at 2:17 PM Ian Kumlien <ian.kumlien@xxxxxxxxx> wrote: > > On Thu, Nov 18, 2021 at 2:09 PM Rafael J. Wysocki <rafael@xxxxxxxxxx> wrote: > > On Thu, Nov 18, 2021 at 1:43 PM Ian Kumlien <ian.kumlien@xxxxxxxxx> wrote: > > > > > > Hi, > > > > > > I started updating kernels to 5.14 a while ago and the laptop kept > > > deadlocking, it's my work laptop so it's quite locked down wrt bios > > > and bios updates.... > > > > > > But, i have since tried with 5.15.2 and it still crashes... > > > > > > So the little testing i have is: > > > acpi=off -> works > > > acpi=ht -> *boom* > > > > This need not mean that ACPI is the source of the problem, though. > > > > > 5.13.x did work however and i know there was updates wrt to acpi in 5.14 > > > > > > Older kernel oops can be found here: > > > https://photos.app.goo.gl/ZygV5Ja39C7cQCR39 > > > > But the Oops is indeed related to ACPI. It looks like the AML > > interpreter is crashing when trying to execute the _OFF method of a > > power resource. > > > > > I'd basically just want to know what kind of information can I provide > > > to help you to find and fix this issue? > > > > Would you be able to collect a trace with dynamic debug in > > drivers/acpi/power.c enabled, that is, add dyndbg="file power.c p+" to > > the kernel command line before booting and capture an oops then? > > Done, it seems like it's crashing mid screen refresh or so though, > hope it's useful > > https://photos.app.goo.gl/kzgDv8EV75sq6FYf7 The string passed in the kernel command line is not exactly the above, so it didn't turn on the dynamic debug (one of the " characters is misplaced). > I could also dump all the acpi data, if that helps It may help later.
