On Thu, May 28, 2020 at 02:20:45PM +0000, Luis Chamberlain wrote: > On Thu, May 28, 2020 at 09:04:02AM -0500, Eric W. Biederman wrote: > > I see some recent (within the last year) fixes to proc_sysctl.c in this > > area. Do you have those? It looks like bridge up and down is stressing > > this code. Either those most recent fixes are wrong, your kernel is > > missing them or this needs some more investigation. > > Thanks for the review Eric. > Seconded. My first try at fixing it was incorrect, hopefully the new attempt ([PATCH] get_subdir: do not drop new subdir if returning it) I've just sent will fare better. > Boris, the elaborate deatils you provided would also be what would be > needed for a commit log, specially since this is fixing a crash. If > you confirm this is still present upstream by reproducing with a test > case it would be wonderful. Yes. It is present in the upstream kernel. In my patch I've added warning to catch this bad condition. The warning fires easily during boot. Thanks, Boris.
