On Fri 2019-03-08 10:31:34, Sergey Senozhatsky wrote: > On (03/07/19 13:06), John Ogness wrote: > > On 2019-03-04, Sergey Senozhatsky <sergey.senozhatsky.work@xxxxxxxxx> wrote: > > > This, theoretically, creates a whole new world of possibilities for > > > console drivers. Now they can do GFP_KERNEL allocations and stall > > > printk_kthread during OOM; or they can explicitly reschedule from > > > ->write() callback (via console_conditional_schedule()) because > > > console_lock() sets console_may_schedule. > > > This can stall the entire printing pipeline > > OOM -> printk_kthread() -> console_lock() -> con_foo() -> kmalloc(GFP_KERNEL) -> OOM > > > Although, as I mentioned in a previous response[0], perhaps we should > > not loosen the requirements on write(). > > Right. Console drivers better stay restricted; very restricted. I agree here. > > It is exactly that disable_preempt() that is so harmful for realtime tasks. > I'll reply in another email (later today, or tomorrow). I see. But I have troubles to imagine a preemtible direct console output. The result would be mixed messages on a single line. Best Regards, Petr
