On 04/24, Roman Gushchin wrote: > > On Wed, Apr 24, 2019 at 05:46:19PM +0200, Oleg Nesterov wrote: > > > > OK, how about the ABSOLUTELY UNTESTED patch below? For the start. > > It looks good to me (and all freezer selftests pass). > > Just to be sure, is it a solution to avoid the busy loop in the signal handling > loop, right? Yes, > Because it doesn't allow to drop the ->frozen check from recalc(). Yes. Because we can race with unfreeze after leave_frozen(). > The JOBCTL_TRAP_FREEZE check without siglock initially looked dangerous to me, > but after some thoughts I didn't find any case when it's wrong. I think this is fine... Yes, JOBCTL_TRAP_FREEZE can be already set when we take siglock, but I don't think we need to recheck this flag. The only important thing (afaics) is that CGRP_FREEZE is stable under css_set_lock, so we can't wrongly set TRAP_FREEZE. > Do you prefer me to master a patch Yes please ;) Oleg.
