Re: [RFC PATCH 5/7] sound: core: Avoid using timespec for struct snd_timer_status

[Baolin Wang]

On 21 September 2017 at 21:14, Arnd Bergmann <arnd@xxxxxxxx> wrote:
> On Thu, Sep 21, 2017 at 8:18 AM, Baolin Wang <baolin.wang@xxxxxxxxxx> wrote:
>
>>  static long snd_timer_user_ioctl_compat(struct file *file, unsigned int cmd, unsigned long arg)
>> @@ -158,12 +151,10 @@ static long snd_timer_user_ioctl_compat(struct file *file, unsigned int cmd, uns
>>                 return snd_timer_user_gparams_compat(file, argp);
>>         case SNDRV_TIMER_IOCTL_INFO32:
>>                 return snd_timer_user_info_compat(file, argp);
>> -       case SNDRV_TIMER_IOCTL_STATUS32:
>> +       case SNDRV_TIMER_IOCTL_STATUS_COMPAT32:
>>                 return snd_timer_user_status_compat(file, argp);
>> -#ifdef CONFIG_X86_X32
>> -       case SNDRV_TIMER_IOCTL_STATUS_X32:
>> -               return snd_timer_user_status_x32(file, argp);
>> -#endif /* CONFIG_X86_X32 */
>> +       case SNDRV_TIMER_IOCTL_STATUS_COMPAT64:
>> +               return snd_timer_user_status64(file, argp);
>>         }
>
> I think the last line would fail to build since snd_timer_user_status64()
> is defined 'static' in a different file.

I saw the timer_compat.c file will be included into timer.c file, so I
think it will not. (My arm32 platform can not build compat mode, but I
will try again to make sure it can build successfully.)

>
> Also, snd_timer_user_status_compat() seems to be the same as
> snd_timer_user_status32(), so I think you can redirect it the same
> way as snd_timer_user_status64 after making both functions globally
> visible.

OK. Let me check again.

-- 
Baolin.wang
Best Regards