Re: [RFC][PATCH v3 0/4] let printk()/console_trylock() callers to cond_resched()

From: Sergey Senozhatsky
Date: Tue Feb 02 2016 - 22:48:25 EST


Hello,

On (01/23/16 17:15), Sergey Senozhatsky wrote:
> [was: cond_resched() some of console_trylock callers]
>
> Hello,
>
> console_unlock() allows to cond_resched() if its caller has
> set `console_may_schedule' to 1 (this functionality present
> since commit 'printk: do cond_resched() between lines while
> outputting to consoles').

gentlemen, did you get a chance to look into this?

-ss

> The rules are:
> -- console_lock() always sets `console_may_schedule' to 1
> -- console_trylock() always sets `console_may_schedule' to 0
>
> printk() calls console_unlock() with preemption desabled, which
> basically can lead to RCU stalls, watchdog soft lockups, etc. if
> something is simultaneously calling printk() frequent enough (IOW,
> console_sem owner always has new data to send to console divers
> and can't leave console_unlock() for a long time).
>
> printk()->console_trylock() callers do not necessarily execute in
> atomic contexts, and some of them can cond_resched() in
> console_unlock(). console_trylock() can set `console_may_schedule'
> to 1 (allow cond_resched() later in consoe_unlock()) when it's safe.
>
> 0001 addresses a theoretical loss of printk messages.
> 0002 is an optional optimization patch
> 0003 drops console_trylock_for_printk
> 0004 lets some of printk()/console_trylock() callers to cond_resched()
> in console_unlock()
>
> v2-v3 (thanks to Petr Mladek for reviews):
> -- do not call can_use_console() on every iteration in console_unlock() (Petr Mladek)
> -- move "This stops the holder of console_sem.." comment (noted by Petr Mladek)
> -- take extra care of !PREEMPT_COUNT kernels (Petr Mladek)
> -- call_console_drivers() still must check cpu_online && CON_ANYTIME (Petr Mladek)
> -- removed console_trylock_for_printk() (noted by Petr Mladek)
>
> v1-v2:
> -- make have_callable_console() available for !PRINTK configs (lkp@xxxxxxxxx)
> -- take care of RCU preempt kernels in console_trylock()
>
>
> Sergey Senozhatsky (4):
> printk: move can_use_console out of console_trylock_for_printk
> printk: do not console_cont_flush() on every jump to again
> printk: remove console_trylock_for_printk
> printk: set may_schedule for some of console_trylock callers
>
> kernel/printk/printk.c | 121 ++++++++++++++++++++++---------------------------
> 1 file changed, 55 insertions(+), 66 deletions(-)
>
> --
> 2.7.0
>