On (11/22/18 11:16), Peter Zijlstra wrote: > > So maybe we need to switch debug objects print-outs to _always_ > > printk_deferred(). Debug objects can be used in code which cannot > > do direct printk() - timekeeping is just one example. > > No, printk_deferred() is a disease, it needs to be eradicated, not > spread around. deadlock-free printk() is deferred, but OK. Another idea then: --- diff --git a/lib/debugobjects.c b/lib/debugobjects.c index 70935ed91125..3928c2b2f77c 100644 --- a/lib/debugobjects.c +++ b/lib/debugobjects.c @@ -323,10 +323,13 @@ static void debug_print_object(struct debug_obj *obj, char *msg) void *hint = descr->debug_hint ? descr->debug_hint(obj->object) : NULL; limit++; + + bust_spinlocks(1); WARN(1, KERN_ERR "ODEBUG: %s %s (active state %u) " "object type: %s hint: %pS\n", msg, obj_states[obj->state], obj->astate, descr->name, hint); + bust_spinlocks(0); } debug_objects_warnings++; } --- This should make serial consoles re-entrant. So printk->console_driver_write() hopefully will not deadlock. IOW, this turns serial consoles into early_consoles, for a while ;) -ss
