Hi Finn, Geert,
Am 14.03.2024 um 13:59 schrieb Michael Schmitz:
Hi Finn,
On 13/03/24 17:40, Finn Thain wrote:
On Wed, 13 Mar 2024, Michael Schmitz wrote:
Running a stack-ng stack-fill stressor triggered the panic, too. I'll
see how reliable that is.
I see an immediate failure with full preemption in QEMU when I run
stress-ng -t 60 --zombie -1
Seems to work quite well.
Though what I got on the first attempt was a deadlock with full
preemption in ARAnyM. No panic.
Time to look for a solution now we've got a nice test case.
No luck with whatever I tried around signals, cache maintenance and mm.
The 'BUG: Bad rss-counter state' message suggests we're freeing the same
page ranges twice, sometimes in many cases. I cannot quite see how
preempting the kernel on interupt return would cause this. Signal
forcing process exit but process exiting before signal is received due
to preemption? But skipping preemption when a signal is pending did not
change anything in my tests...
Running out of ideas here, sorry.
Cheers,
Michael
Cheers,
Michael
