linux-kernel - Re: [PATCH] blutooth: try to improve CONFIG_SERIAL_DEV

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: <CAK8P3a2ZVZ_LjnNiVM-_ew-jR6=DyAjUhtU6do+sgAQv21kbJQ@mail.gmail.com>
Date:   Thu, 20 Apr 2017 09:02:19 +0200
From:   Arnd Bergmann <arnd@...db.de>
To:     Marcel Holtmann <marcel@...tmann.org>
Cc:     Sebastian Reichel <sre@...nel.org>,
        "Gustavo F. Padovan" <gustavo@...ovan.org>,
        Johan Hedberg <johan.hedberg@...il.com>,
        Rob Herring <robh@...nel.org>,
        Bjorn Andersson <bjorn.andersson@...aro.org>,
        "David S. Miller" <davem@...emloft.net>,
        Kalle Valo <kvalo@...eaurora.org>,
        linux-bluetooth@...r.kernel.org,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] blutooth: try to improve CONFIG_SERIAL_DEV_BUS dependency

On Thu, Apr 20, 2017 at 8:55 AM, Marcel Holtmann <marcel@...tmann.org> wrote:
> Hi Sebastian,
>
>>> With CONFIG_SERIAL_DEV_BUS=m, the hci_serdev.o file does not actually
>>> get built into hci_uart.o as the Makefile doesn't pick it up, leading
>>> to a link error with anything referring to it:
>>>
>>> ERROR: "hci_uart_register_device" [drivers/bluetooth/hci_nokia.ko] undefined!
>>> scripts/Makefile.modpost:91: recipe for target '__modpost' failed
>>>
>>> Changing this in the Makefile would cause another problem when
>>> hci_uart itself is built-in and cannot reference symbols from the
>>> serdev module.
>>>
>>> This tries to address both problems by introducing a new hidden
>>> Kconfig symbol that controls both the compilation of hci_serdev.o
>>> and whether the Nokia driver can be selected. This seems to address
>>> the problem for me, though there might be a better way to do it.
>>>
>>> Fixes: 7bb318680e86 ("Bluetooth: add nokia driver")
>>> Signed-off-by: Arnd Bergmann <arnd@...db.de>
>>> ---
>>> drivers/bluetooth/Kconfig  | 8 +++++++-
>>> drivers/bluetooth/Makefile | 2 +-
>>> 2 files changed, 8 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/drivers/bluetooth/Kconfig b/drivers/bluetooth/Kconfig
>>> index 8aafbed9e160..737d93ef27c5 100644
>>> --- a/drivers/bluetooth/Kconfig
>>> +++ b/drivers/bluetooth/Kconfig
>>> @@ -76,6 +76,12 @@ config BT_HCIUART
>>>        Say Y here to compile support for Bluetooth UART devices into the
>>>        kernel or say M to compile it as module (hci_uart).
>>>
>>> +config BT_HCIUART_SERDEV
>>> +    bool
>>> +    depends on SERIAL_DEV_BUS && BT_HCIUART
>>> +    depends on SERIAL_DEV_BUS=y || SERIAL_DEV_BUS=BT_HCIUART
>>> +    default y
>>> +
>>> config BT_HCIUART_H4
>>>      bool "UART (H4) protocol support"
>>>      depends on BT_HCIUART
>>> @@ -89,7 +95,7 @@ config BT_HCIUART_H4
>>> config BT_HCIUART_NOKIA
>>>      tristate "UART Nokia H4+ protocol support"
>>>      depends on BT_HCIUART
>>
>> I guess BT_HCIUART dependency is no longer needed
>> when we have BT_HCIUART_SERDEV dependency?
>
> if so then we need another patch since I already applied the one from Arnd.

We need both dependencies: BT_HCIUART_SERDEV is a 'bool'
symbol controlling whether subdrivers are able to use symbols
exported by the serdev code, while BT_HCIUART is a tristate
symbol and we need to depend on that to ensure that
BT_HCIUART_NOKIA cannot be set to '=y' when
BT_HCIUART=m.

No further patch should be needed.

    Arnd