Re: [PATCHv6 0/7] tty: Hold write ldisc sem in tty_reopen()

[Pasi Kärkkäinen]

Hello Mikulas,

Any chance you could re-test these v6 patches? It seems you're able to relatively easily trigger the bug on your (pa-risc) environment.. I'd be good to have this series finally acked/merged!


Thanks a lot,

-- Pasi

On Thu, Nov 01, 2018 at 12:24:45AM +0000, Dmitry Safonov wrote:
> Hi all,
> 
> Here are fixes that worth to have in the @stable, as they were hit by
> different people, including Arista on v4.9 stable.
> 
> And for linux-next - adding lockdep asserts for line discipline changing
> code, verifying that write ldisc sem will be held forthwith.
> 
> Mikulas, can you add your tested-by on this patches set again, please?
> I tried to reproduce reboot issue on qemu-hppa and even built
> cross-compiler for pa-risc.. but was unlucky in reproducing.
> 
> Thanks,
> Dima
> 
> Changes since v5:
> - Better commit tags
> - Hopefully fixed issue with reboot on pa-risc with Debian 5
> 
> Changes since v4:
> - back to lock ldisc with (5*HZ) timeout in tty_reopen()
>   (LKP report link: lkml.kernel.org/r/<1536940609.3185.29.camel@xxxxxxxxxx>)
> - reordered 3/7 with 2/7 for LKP robot
> 
> Changes since v3:
> - Added tested-by Mark Rutland (thanks!)
> - Dropped patch with smp_wmb() - wrong idea
> - lockdep_assert_held() should be actually lockdep_assert_held_exclusive()
> - Described why tty_ldisc_open() can be called without ldisc_sem held
>   for pty slave end (o_tty).
> - Added Peter's patch for dropping self-made lockdep annotations
> - Fix for a reader(s) of ldisc semaphore waiting for an active reader(s)
> 
> Changes since v2:
> - Added reviewed-by tags
> - Hopefully, fixed reported by 0-day issue.
> - Added optional fix for wait_readers decrement
> 
> Changes since v1:
> - Added tested-by/reported-by tags
> - Dropped 3/4 (locking tty pair for lockdep sake),
>   Because of that - not adding lockdep_assert_held() in tty_ldisc_open()
> - Added 4/4 cleanup to inc tty->count only on success of
>   tty_ldisc_reinit()
> - lock ldisc without (5*HZ) timeout in tty_reopen()
> 
> v1 link: lkml.kernel.org/r/<20180829022353.23568-1-dima@xxxxxxxxxx>
> v2 link: lkml.kernel.org/r/<20180903165257.29227-1-dima@xxxxxxxxxx>
> v3 link: lkml.kernel.org/r/<20180911014821.26286-1-dima@xxxxxxxxxx>
> v4 link: lkml.kernel.org/r/<20180912001702.18522-1-dima@xxxxxxxxxx>
> 
> Cc: Daniel Axtens <dja@xxxxxxxxxx>
> Cc: Dmitry Vyukov <dvyukov@xxxxxxxxxx>
> Cc: Mark Rutland <mark.rutland@xxxxxxx>
> Cc: Michael Neuling <mikey@xxxxxxxxxxx>
> Cc: Mikulas Patocka <mpatocka@xxxxxxxxxx>
> Cc: Nathan March <nathan@xxxxxx>
> Cc: Pasi Kärkkäinen <pasik@xxxxxx>
> Cc: Peter Hurley <peter@xxxxxxxxxxxxxxxxxx>
> Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> Cc: "Rong, Chen" <rong.a.chen@xxxxxxxxx>
> Cc: Sergey Senozhatsky <sergey.senozhatsky.work@xxxxxxxxx>
> Cc: Tan Xiaojun <tanxiaojun@xxxxxxxxxx>
> Cc: Tetsuo Handa <penguin-kernel@xxxxxxxxxxxxxxxxxxx>
> (please, ignore if I Cc'ed you mistakenly)
> 
> Dmitry Safonov (6):
>   tty/ldsem: Wake up readers after timed out down_write()
>   tty: Hold tty_ldisc_lock() during tty_reopen()
>   tty: Don't block on IO when ldisc change is pending
>   tty: Simplify tty->count math in tty_reopen()
>   tty/ldsem: Add lockdep asserts for ldisc_sem
>   tty/ldsem: Decrement wait_readers on timeouted down_read()
> 
> Peter Zijlstra (1):
>   tty/ldsem: Convert to regular lockdep annotations
> 
>  drivers/tty/n_hdlc.c    |  4 +--
>  drivers/tty/n_r3964.c   |  2 +-
>  drivers/tty/n_tty.c     |  8 +++---
>  drivers/tty/tty_io.c    | 14 ++++++----
>  drivers/tty/tty_ldisc.c | 16 +++++++++++
>  drivers/tty/tty_ldsem.c | 62 +++++++++++++++++------------------------
>  include/linux/tty.h     |  7 +++++
>  7 files changed, 63 insertions(+), 50 deletions(-)
> 
> -- 
> 2.19.1
>