Hi Salvatore, On Sat, Feb 12, 2022 at 07:11:40PM +0100, Salvatore Bonaccorso wrote: > > […] > > after upgrading a A20-OLinuXino-LIME2 board to bullseye, I had problems > > with the > > network interface which could be solved by preparing a tweaked u-boot (c.f. > > [1]). > > However, a few hours later, a kernel oops happended and the board could not > > be > > reached from the LAN again: > > > > […] > > > > I downgraded to the buster kernel for now which was working reliably for > > months > > without issues on the same board. > > > > Many thanks and best regards, > > > > Andi > > > > > > [1] https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=927397#62 > > Just stumbled over this bugreport from last year, is this still a > problem with a recent kernel? >
Hard to say: I replaced the board by an "identical" board but with a newer revision. Another board (same revision as the failing one) has a different hardware configuration: It has no eMMC flash. Both boards work fine since months without any issues. I kept the board that shows the issue here on my desk to experiment, where the lockup (after ~days) is no problem. Just now I powered the board again and upgraded to the latest kernel. Let's see what happens. The last times (some time ago already) it still failed after some time. That said, I can imagine that this is a hardware issue, which has not been triggered by the buster kernel (as I wrote in the initial report, it worked without any issues there) but now, related to kernel changes/optimizations or the like, it is triggered (on that specific, individual board). Note that unfortunately the board is not perfectly 'stable' with regard to the hardware (#927397 #911560 #911560). Best regards, Andi
