Subject: Re: [PATCH v2 1/1] mux: consumer: Add dummy functions for
 !CONFIG_MULTIPLEXER case
To: Andy Shevchenko <andy.shevchenko@gmail.com>,
        Kuppuswamy Sathyanarayanan 
        <sathyanarayanan.kuppuswamy@linux.intel.com>
Cc: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        Sathyanarayanan Kuppuswamy Natarajan <sathyaosid@gmail.com>
References: <83ff1554464c781df142117ed30430476ec73634.1499537263.git.sathyanarayanan.kuppuswamy@linux.intel.com>
 <CAHp75VdbVx1YrO_8USJmRVjni2SW9g=3-fTWJQZaEJWOQ-RHCw@mail.gmail.com>
From: Peter Rosin <peda@axentia.se>
Organization: Axentia Technologies AB
Message-ID: <d8673232-5bb2-1679-71a8-2e5af944dd99@axentia.se>
Date: Sun, 9 Jul 2017 08:59:57 +0200
User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:52.0) Gecko/20100101
 Thunderbird/52.2.1
MIME-Version: 1.0
In-Reply-To: <CAHp75VdbVx1YrO_8USJmRVjni2SW9g=3-fTWJQZaEJWOQ-RHCw@mail.gmail.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 7bit
SpamDiagnosticOutput: 1:99
SpamDiagnosticMetadata: NSPM
X-MS-Exchange-CrossTenant-OriginalArrivalTime: 09 Jul 2017 07:00:01.4995 (UTC)
X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted
X-MS-Exchange-Transport-CrossTenantHeadersStamped: AM5PR0202MB2548
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org
Content-Length: 3290
Lines: 117

On 2017-07-08 23:22, Andy Shevchenko wrote:
> On Sat, Jul 8, 2017 at 9:12 PM,
> <sathyanarayanan.kuppuswamy@linux.intel.com> wrote:
>> From: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com>
>>
>> Add dummy functions to avoid compile time issues when CONFIG_MULTIPLEXER
>> is not enabled.
>>
> 
> I don't think the error return code is okay to all of them. The return
> value should be choosen carefully (for some functions it's okay IMO to
> return 0).

BTW, is ENODEV correct for this situation? I have this nagging feeling
that ENODEV is over-used?

And again, all these stubs should all be inlines, or things will break it
this file is included more than once.

>> Signed-off-by: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com>
>> ---
>>  include/linux/mux/consumer.h | 38 ++++++++++++++++++++++++++++++++++++++
>>  1 file changed, 38 insertions(+)
>>
>> Changes since v1:
>>  * Changed #ifdef to #if IS_ENABLED.
>>
>> diff --git a/include/linux/mux/consumer.h b/include/linux/mux/consumer.h
>> index 5577e1b..df78988 100644
>> --- a/include/linux/mux/consumer.h
>> +++ b/include/linux/mux/consumer.h
>> @@ -16,6 +16,7 @@
>>  struct device;
>>  struct mux_control;
>>
>> +#if IS_ENABLED(CONFIG_MULTIPLEXER)
>>  unsigned int mux_control_states(struct mux_control *mux);
>>  int __must_check mux_control_select(struct mux_control *mux,
>>                                     unsigned int state);
>> @@ -29,4 +30,41 @@ void mux_control_put(struct mux_control *mux);
>>  struct mux_control *devm_mux_control_get(struct device *dev,
>>                                          const char *mux_name);
>>
>> +#else
>> +unsigned int mux_control_states(struct mux_control *mux)
>> +{
>> +       return -ENODEV;
> 
> Peter, is here we are obliged to return error code in such case?

Since it will presumably be difficult to obtain a mux_control
w/o the mux-core being present, it doesn't matter much what
most of these stubs return.

For this stub, 0 is perhaps best, since the kernel-doc for
mux_control_states mentions nothing about any error possibility.

>> +}
>> +
>> +int __must_check mux_control_select(struct mux_control *mux,
>> +                                   unsigned int state)
>> +{
>> +       return -ENODEV;
> 
> return 0; ?

Maybe. But it doesn't matter much, but in this case the consumer must
handle errors. See above.

>> +}
>> +
>> +int __must_check mux_control_try_select(struct mux_control *mux,
>> +                                       unsigned int state)
>> +{
>> +       return -ENODEV;
>> +}
> 
> return 0; ?

Maybe. But it doesn't matter much, but in this case the consumer must
handle errors. See above.

>> +
>> +int mux_control_deselect(struct mux_control *mux)
>> +{
>> +       return -ENODEV;
>> +}
> 
> return 0; ?

Probably. See above.

Cheers,
peda

>> +
>> +struct mux_control *mux_control_get(struct device *dev, const char *mux_name)
>> +{
>> +       return ERR_PTR(-ENODEV);
>> +}
>> +
>> +void mux_control_put(struct mux_control *mux) {}
>> +
>> +struct mux_control *devm_mux_control_get(struct device *dev,
>> +                                        const char *mux_name)
>> +{
>> +       return ERR_PTR(-ENODEV);
>> +}
>> +#endif
>> +
>>  #endif /* _LINUX_MUX_CONSUMER_H */
>> --
>> 2.7.4
>>
> 
> 
>