On Mon, 12 Dec 2022 14:31:35 +1000 "Nicholas Piggin" <npiggin@xxxxxxxxx> wrote: > Have you run memtest on the system? Are the messages related to a > kernel upgrade? This and your KASAN bugs look possibly like random > corruption. About the time 5.4 was around I did not have such issues on the G4. Building stuff for hours natively or via distcc just went fine. So I suspect there is a flaw introduced in newer kernel versions. This memory corruption or part of it may be even here since kernel 5.10 if it's the same issue as https://bugzilla.kernel.org/show_bug.cgi?id=215389 But actually running memtest is a very good idea. Just to be sure I am not hunting ghosts. > Although with that KASAN one it's strange that kernfs_node_cache > was involved both times, it's strange that page tables are pointing > to that same slab memory. It could be a page table page use-after > -free maybe? Maybe with the page table fragment code. I'm sure other > people would have hit that before though, so I don't know what to > suggest. Does not necessarily need to be the case other people have experienced this issue. This G4 DP setup is a bit unusually as it runs 2 x G4 CPUs and I doubt there is much multicore testing done on ppc32. ;) A few issues I reported (and Christophe thankfully ironed out) originated from here. Next I'll check out a known good 5.4 LTS kernel and see what memtester (https://pyropus.ca./software/memtester/) tells me. Regards, Erhard
