netdev - Re: [PATCH v3] bluetooth: Fix Advertisement Monitor Suspend/Resume

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-Id: <9D93B868-B8DE-4C0E-B055-A2BB3B93528F@holtmann.org>
Date:   Fri, 24 Sep 2021 12:14:22 +0200
From:   Marcel Holtmann <marcel@...tmann.org>
To:     Manish Mandlik <mmandlik@...gle.com>
Cc:     Luiz Augusto von Dentz <luiz.dentz@...il.com>,
        CrosBT Upstreaming <chromeos-bluetooth-upstreaming@...omium.org>,
        linux-bluetooth <linux-bluetooth@...r.kernel.org>,
        Archie Pusaka <apusaka@...gle.com>,
        Miao-chen Chou <mcchou@...gle.com>,
        "David S. Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>,
        Johan Hedberg <johan.hedberg@...il.com>,
        linux-kernel@...r.kernel.org, netdev@...r.kernel.org
Subject: Re: [PATCH v3] bluetooth: Fix Advertisement Monitor Suspend/Resume

Hi Manish,

> During system suspend, advertisement monitoring is disabled by setting
> the HCI_VS_MSFT_LE_Set_Advertisement_Filter_Enable to False. This
> disables the monitoring during suspend, however, if the controller is
> monitoring a device, it sends HCI_VS_MSFT_LE_Monitor_Device_Event to
> indicate that the monitoring has been stopped for that particular
> device. This event may occur after suspend depending on the
> low_threshold_timeout and peer device advertisement frequency, which
> causes early wake up.
> 
> Right way to disable the monitoring for suspend is by removing all the
> monitors before suspend and re-monitor after resume to ensure no events
> are received during suspend. This patch fixes this suspend/resume issue.
> 
> Following tests are performed:
> - Add monitors before suspend and make sure DeviceFound gets triggered
> - Suspend the system and verify that all monitors are removed by kernel
>  but not Released by bluetoothd
> - Wake up and verify that all monitors are added again and DeviceFound
>  gets triggered
> 
> Signed-off-by: Manish Mandlik <mmandlik@...gle.com>
> Reviewed-by: Archie Pusaka <apusaka@...gle.com>
> Reviewed-by: Miao-chen Chou <mcchou@...gle.com>
> ---
> 
> Changes in v3:
> - Updated the msft_{suspend/resume} function names
> 
> Changes in v2:
> - Updated the Reviewd-by names
> 
> net/bluetooth/hci_request.c |  15 +++--
> net/bluetooth/msft.c        | 117 +++++++++++++++++++++++++++++++-----
> net/bluetooth/msft.h        |   5 ++
> 3 files changed, 116 insertions(+), 21 deletions(-)

patch has been applied to bluetooth-next tree.

Regards

Marcel