[PATCHv6 1/7] tty/ldsem: Wake up readers after timed out down_write()

From: Dmitry Safonov
Date: Wed Oct 31 2018 - 20:25:02 EST


ldsem_down_read() will sleep if there is pending writer in the queue.
If the writer times out, readers in the queue should be woken up,
otherwise they may miss a chance to acquire the semaphore until the last
active reader will do ldsem_up_read().

There was a couple of reports where there was one active reader and
other readers soft locked up:
Showing all locks held in the system:
2 locks held by khungtaskd/17:
#0: (rcu_read_lock){......}, at: watchdog+0x124/0x6d1
#1: (tasklist_lock){.+.+..}, at: debug_show_all_locks+0x72/0x2d3
2 locks held by askfirst/123:
#0: (&tty->ldisc_sem){.+.+.+}, at: ldsem_down_read+0x46/0x58
#1: (&ldata->atomic_read_lock){+.+...}, at: n_tty_read+0x115/0xbe4

Prevent readers wait for active readers to release ldisc semaphore.

Link: lkml.kernel.org/r/20171121132855.ajdv4k6swzhvktl6@xxxxxxxxxxxxxxxxxxxxxx
Link: lkml.kernel.org/r/20180907045041.GF1110@shao2-debian
Cc: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
Cc: Jiri Slaby <jslaby@xxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: stable@xxxxxxxxxxxxxxx
Reported-by: kernel test robot <rong.a.chen@xxxxxxxxx>
Signed-off-by: Dmitry Safonov <dima@xxxxxxxxxx>
---
drivers/tty/tty_ldsem.c | 10 ++++++++++
1 file changed, 10 insertions(+)

diff --git a/drivers/tty/tty_ldsem.c b/drivers/tty/tty_ldsem.c
index 0c98d88f795a..b989ca26fc78 100644
--- a/drivers/tty/tty_ldsem.c
+++ b/drivers/tty/tty_ldsem.c
@@ -293,6 +293,16 @@ down_write_failed(struct ld_semaphore *sem, long count, long timeout)
if (!locked)
atomic_long_add_return(-LDSEM_WAIT_BIAS, &sem->count);
list_del(&waiter.list);
+
+ /*
+ * In case of timeout, wake up every reader who gave the right of way
+ * to writer. Prevent separation readers into two groups:
+ * one that helds semaphore and another that sleeps.
+ * (in case of no contention with a writer)
+ */
+ if (!locked && list_empty(&sem->write_wait))
+ __ldsem_wake_readers(sem);
+
raw_spin_unlock_irq(&sem->wait_lock);

__set_current_state(TASK_RUNNING);
--
2.19.1