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] [thread-next>] [day] [month] [year] [list]
Message-Id: <5742801D-881A-45BC-A8A7-28D694179D8E@holtmann.org>
Date:   Mon, 19 Dec 2022 12:42:03 +0100
From:   Marcel Holtmann <marcel@...tmann.org>
To:     Luiz Augusto von Dentz <luiz.dentz@...il.com>
Cc:     Neeraj Sanjay Kale <neeraj.sanjaykale@....com>,
        Johan Hedberg <johan.hedberg@...il.com>,
        Paul Menzel <pmenzel@...gen.mpg.de>, amitkumar.karwar@....com,
        rohit.fule@....com, sherry.sun@....com,
        LKML <linux-kernel@...r.kernel.org>,
        linux-bluetooth@...r.kernel.org
Subject: Re: [PATCH v5] Bluetooth: Add hci_nxp to hci_uart module to support
 NXP BT chipsets

Hi Luiz,

>>> Add hci_nxp to the hci_uart module which adds support for the NXP BT
>>> chips. This driver has Power Save feature that will put the NXP bluetooth
>>> chip into sleep state, whenever there is no activity for certain duration
>>> of time (2000ms), and will be woken up when any activity is to be
>>> initiated.
>>> 
>>> The Power Save feature can be configured with the following set
>>> of commands (optional):
>>> hcitool -i hci0 cmd 3F 23 02 00 00    (enable Power Save)
>>> hcitool -i hci0 cmd 3F 23 03 00 00    (disable Power Save)
>>> where,
>>> OGF = 0x3F (vendor specific command)
>>> OCF = 0x23 (command to set Power Save state)
>>> arg[0] = 0x02 (disable Power Save)
>>> arg[0] = 0x03 (enable Power Save)
>>> arg[1,2,...] = XX (don't care)
>>> 
>>> The sleep/wake-up source can be configured with the following set
>>> of commands (optional):
>>> hcitool -i hci0 cmd 3F 53 03 14 01 FF    (set UART break method)
>>> hcitool -i hci0 cmd 3F 53 03 14 00 FF    (set UART DSR method)
>>> where,
>>> OGF = 0x3F (vendor specific command)
>>> OCF = 0x53 (command to set sleep and wake-up source)
>>> arg[0] = 0x00 (Chip to host method NONE)
>>> arg[0] = 0x01 (Chip to host method UART DTR)
>>> arg[0] = 0x02 (Chip to host method UART BREAK)
>>> arg[0] = 0x03 (Chip to host method GPIO)
>>> arg[1] = 0x14 (Chip to host GPIO[20] if arg[0] is 0x03, else 0xFF)
>>> arg[2] = 0x00 (Host to chip method UART DSR)
>>> arg[2] = 0x01 (Host to chip method UART BREAK)
>>> arg[3] = 0xXX (Reserved for future use)
>>> 
>>> By default, the hci_nxp sets power save enable, chip to host wake-up
>>> source as GPIO and host to chip sleep and wake-up source as UART
>>> break during driver initialization, by sending the respective
>>> commands to the chip.
>>> 
>>> Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@....com>
>>> ---
>>> v2: Changed the subject/summary lines and added more details in the
>>> description. (Paul Menzel)
>>> v3: Made internal functions static, optimized the code, added few
>>> comments. (Sherry Sun)
>>> v4: Reworked entire code to send vendor commands cmd23 and cmd53 by
>>> using __hci_cmd_sync. (Luiz Augusto von Dentz)
>>> v5: Used hci_command_hdr and combined OGF+OCF into a single opcode.
>>> (Luiz Augusto von Dentz)
>>> ---
>>> MAINTAINERS                   |   6 +
>>> drivers/bluetooth/Kconfig     |  10 +
>>> drivers/bluetooth/Makefile    |   1 +
>>> drivers/bluetooth/hci_ldisc.c |   6 +
>>> drivers/bluetooth/hci_nxp.c   | 592 ++++++++++++++++++++++++++++++++++
>>> drivers/bluetooth/hci_nxp.h   |  94 ++++++
>>> drivers/bluetooth/hci_uart.h  |   8 +-
>>> 7 files changed, 716 insertions(+), 1 deletion(-)
>>> create mode 100644 drivers/bluetooth/hci_nxp.c
>>> create mode 100644 drivers/bluetooth/hci_nxp.h
>> 
>> so this is a clear NAK. Add this as serdev driver and not hook further into the
>> mess that is the HCI line discipline.
> 
> I wonder if we should make it more clear somehow, perhaps include a
> text on the likes of BT_HCIUART that is deprecated and new drivers
> shall use BT_HCIUART_SERDEV instead.

not even that. They need to be separate drivers. A long time ago I posted
the skeleton for btuart.ko and bt3wire.ko and that is where this has to go.

Regards

Marcel

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ