(off-list ancestor, not in this archive)
Possible BT deadlock due to recursive read locking (Was: Re: kernel BUG at kernel/locking/rtmutex.c:1059) · Sebastian Andrzej Siewior <bigeasy@linutronix.de> · 2017-08-31
Re: Possible BT deadlock due to recursive read locking (Was: Re: kernel BUG at kernel/locking/rtmutex.c:1059) · Thomas Gleixner <hidden> · 2017-08-31
Re: Possible BT deadlock due to recursive read locking (Was: Re: kernel BUG at kernel/locking/rtmutex.c:1059) · Mart van de Wege <hidden> · 2017-09-01
Re: Possible BT deadlock due to recursive read locking (Was: Re: kernel BUG at kernel/locking/rtmutex.c:1059) · Mart van de Wege <hidden> · 2017-09-02
[PATCH] Bluetooth: avoid recursive locking in hci_send_to_channel() · Sebastian Andrzej Siewior <bigeasy@linutronix.de> · 2017-09-21
Re: [PATCH] Bluetooth: avoid recursive locking in hci_send_to_channel() · Mart van de Wege <hidden> · 2017-09-24
Re: [PATCH] Bluetooth: avoid recursive locking in hci_send_to_channel() · Marcel Holtmann <marcel@holtmann.org> · 2017-10-30

Re: Possible BT deadlock due to recursive read locking (Was: Re: kernel BUG at kernel/locking/rtmutex.c:1059)

From: Mart van de Wege <hidden>
Date: 2017-09-02 13:58:14
Also in: linux-rt-users, lkml

On Fri, 1 Sep 2017 10:51:07 +0200
Mart van de Wege [off-list ref] wrote:

On Thu, 31 Aug 2017 16:52:12 +0200
Sebastian Andrzej Siewior [off-list ref] wrote:

quoted

On 2017-08-30 23:25:53 [+0200], Mart van de Wege wrote:

quoted

Hi,

Hi,

quoted

The issue persists up to v4.11.12-rt10

Does
	CONFIG_RWLOCK_RT_READER_BIASED=y
make it go away?

Yes, that does make it go away. -rt8 now boots cleanly, and works
fine. Building -rt10 now to be sure.

Confirm that as of 4.11.12-rt11 the deadlock still goes away when
enabling CONFIG_RWLOCK_RT_READER_BIASED

Mart