linux-kernel - Re: [PATCH] staging: wfx: fix an undefined reference error when CONFIG

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: <5DA0ACD4.4060101@huawei.com>
Date:   Sat, 12 Oct 2019 00:24:52 +0800
From:   zhong jiang <zhongjiang@...wei.com>
To:     Greg KH <gregkh@...uxfoundation.org>
CC:     Jerome Pouiller <Jerome.Pouiller@...abs.com>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] staging: wfx: fix an undefined reference error when CONFIG_MMC=m

On 2019/10/12 0:16, Greg KH wrote:
> On Fri, Oct 11, 2019 at 11:51:16PM +0800, zhong jiang wrote:
>> On 2019/10/11 17:02, Greg KH wrote:
>>> On Fri, Oct 11, 2019 at 08:40:08AM +0000, Jerome Pouiller wrote:
>>>> On Friday 11 October 2019 06:26:16 CEST Greg KH wrote:
>>>>> CAUTION: This email originated from outside of the organization. Do not 
>>>> click links or open attachments unless you recognize the sender and know the 
>>>> content is safe.
>>>>> On Fri, Oct 11, 2019 at 11:02:19AM +0800, zhong jiang wrote:
>>>>>> I hit the following error when compile the kernel.
>>>>>>
>>>>>> drivers/staging/wfx/main.o: In function `wfx_core_init':
>>>>>> /home/z00352263/linux-next/linux-next/drivers/staging/wfx/main.c:488: 
>>>> undefined reference to `sdio_register_driver'
>>>>>> drivers/staging/wfx/main.o: In function `wfx_core_exit':
>>>>>> /home/z00352263/linux-next/linux-next/drivers/staging/wfx/main.c:496: 
>>>> undefined reference to `sdio_unregister_driver'
>>>>>> drivers/staging/wfx/main.o:(.debug_addr+0x1a8): undefined reference to 
>>>> `sdio_register_driver'
>>>>>> drivers/staging/wfx/main.o:(.debug_addr+0x6f0): undefined reference to 
>>>> `sdio_unregister_driver'
>>>>>> Signed-off-by: zhong jiang <zhongjiang@...wei.com>
>>>>>> ---
>>>>>>  drivers/staging/wfx/Kconfig  | 3 ++-
>>>>>>  drivers/staging/wfx/Makefile | 5 +++--
>>>>>>  2 files changed, 5 insertions(+), 3 deletions(-)
>>>>>>
>>>>>> diff --git a/drivers/staging/wfx/Kconfig b/drivers/staging/wfx/Kconfig
>>>>>> index 9b8a1c7..4d045513 100644
>>>>>> --- a/drivers/staging/wfx/Kconfig
>>>>>> +++ b/drivers/staging/wfx/Kconfig
>>>>>> @@ -1,7 +1,8 @@
>>>>>>  config WFX
>>>>>>       tristate "Silicon Labs wireless chips WF200 and further"
>>>>>>       depends on MAC80211
>>>>>> -     depends on (SPI || MMC)
>>>>>> +     depends on SPI
>>>>>> +     select MMC
>>>>> How about:
>>>>>         depends on (SPI && MMC)
>>>> I dislike to force user to enable both buses while only one of them is 
>>>> sufficient. I would prefer to keep current dependencies and to add
>>>> #ifdef around problematic functions.
>>> Yes, that's the better thing to do here overall.
>>>
>>> zhong, can you work on that?
>> How about the following patch ?
>>
>> diff --git a/drivers/staging/wfx/Makefile b/drivers/staging/wfx/Makefile
>> index 0d9c1ed..77d68b7 100644
>> --- a/drivers/staging/wfx/Makefile
>> +++ b/drivers/staging/wfx/Makefile
>> @@ -19,6 +19,6 @@ wfx-y := \
>>         sta.o \
>>         debug.o
>>  wfx-$(CONFIG_SPI) += bus_spi.o
>> -wfx-$(subst m,y,$(CONFIG_MMC)) += bus_sdio.o
>> +wfx-$(CONFIG_MMC) += bus_sdio.o
>>
>>  obj-$(CONFIG_WFX) += wfx.o
>> diff --git a/drivers/staging/wfx/main.c b/drivers/staging/wfx/main.c
>> index d2508bc..26720a4 100644
>> --- a/drivers/staging/wfx/main.c
>> +++ b/drivers/staging/wfx/main.c
>> @@ -484,16 +484,19 @@ static int __init wfx_core_init(void)
>>
>>         if (IS_ENABLED(CONFIG_SPI))
>>                 ret = spi_register_driver(&wfx_spi_driver);
>> -       if (IS_ENABLED(CONFIG_MMC) && !ret)
>> +#ifdef CONFIG_MMC
>> +       if (!ret)
>>                 ret = sdio_register_driver(&wfx_sdio_driver);
> Put this in an inline function in the .h file so that you just call:
> 	wfx_register_sdio_driver()
> and it does what is needed to be done depending on if CONFIG_MMC is
> enabled or not (note, your check here isn't quite correct, I think you
> need to do IS_ENABLED())
>
> Same for spi.
>
> We really do not like #ifdefs in .c files.
You're right.  I should not use #ifdef here.

Thanks,
zhong jiang
> thanks,
>
> greg k-h
>
> .
>