Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Sender: Guenter Roeck <groeck7@gmail.com>
Date:   Wed, 15 Sep 2021 07:32:38 -0700
From:   Guenter Roeck <linux@roeck-us.net>
To:     Pavel Machek <pavel@denx.de>
Cc:     Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        linux-kernel@vger.kernel.org, stable@vger.kernel.org,
        Hsin-Yi Wang <hsinyi@chromium.org>,
        Kai-Heng Feng <kai.heng.feng@canonical.com>,
        Marcel Holtmann <marcel@holtmann.org>,
        Sasha Levin <sashal@kernel.org>,
        Mattijs Korpershoek <mkorpershoek@baylibre.com>
Subject: Re: [PATCH 5.10 157/236] Bluetooth: Move shutdown callback before
 flushing tx and rx queue
Message-ID: <20210915143238.GA2403125@roeck-us.net>
References: <20210913131100.316353015@linuxfoundation.org>
 <20210913131105.720088593@linuxfoundation.org>
 <20210915111843.GA16198@duo.ucw.cz>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20210915111843.GA16198@duo.ucw.cz>
Precedence: bulk

On Wed, Sep 15, 2021 at 01:18:43PM +0200, Pavel Machek wrote:
> Hi!
> 
> > [ Upstream commit 0ea53674d07fb6db2dd7a7ec2fdc85a12eb246c2 ]
> 
> Upstream commit is okay...
> 
> > So move the shutdown callback before flushing TX/RX queue to resolve the
> > issue.
> 
> ...but something went wrong in stable. This is not moving code, this
> is duplicating it:
> 
> > --- a/net/bluetooth/hci_core.c
> > +++ b/net/bluetooth/hci_core.c
> > @@ -1726,6 +1726,14 @@ int hci_dev_do_close(struct hci_dev *hdev)
> >  	hci_request_cancel_all(hdev);
> >  	hci_req_sync_lock(hdev);
> >  
> > +	if (!hci_dev_test_flag(hdev, HCI_UNREGISTER) &&
> > +	    !hci_dev_test_flag(hdev, HCI_USER_CHANNEL) &&
> > +	    test_bit(HCI_UP, &hdev->flags)) {
> > +		/* Execute vendor specific shutdown routine */
> > +		if (hdev->shutdown)
> > +			hdev->shutdown(hdev);
> > +	}
> > +
> >  	if (!test_and_clear_bit(HCI_UP, &hdev->flags)) {
> >  		cancel_delayed_work_sync(&hdev->cmd_timer);
> >  		hci_req_sync_unlock(hdev);
> 
> And yes, we end up with 2 copies in 5.10.
> 

Same problem in v5.4.y, unfortunately.

Guenter

> Best regards,
> 								Pavel
> --
> DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
> HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany