Re: [RFC PATCH 1/3] btrfs-progs: introduce framework to check kernel supported features

[Eric Sandeen <sandeen@redhat.com>]

On 10/21/15 9:49 AM, Anand Jain wrote:
> 
> 
> On 10/21/2015 10:41 PM, Eric Sandeen wrote:
>> On 10/21/15 4:09 AM, Qu Wenruo wrote:
>>>> +static int get_kernel_code()
>>>> +{
>>>> +    int ret;
>>>> +    struct utsname utsbuf;
>>>> +    char *version;
>>>> +
>>>> +    ret = uname(&utsbuf);
>>>> +    if (ret)
>>>> +        return -ret;
>>>> +
>>>> +    version = strtok(utsbuf.release, "-");
>>>> +
>>>> +    return version_to_code(version);
>>>> +}
>>>
>>> The only problem is, kernel version is never reliable.
>>> If someone wants, uname output may even contain no numeric value.
>>
>> yep, I agree.  This will be misery for any custom kernel.
> 
>  How if we apply this only when kernel version is available ?

The problem is "kernel version" may not match btrfs version.
Distros backport and update subsystems without changing the
kernel version.

>  Otherwise progs will assume all features are supported as in
>  the current design.
> 
> Thanks, Anand

This is only a concern for kernels prior to 3.14, right?
v3.13 was released Sun Jan 19 18:40:23 2014, almost 2 years ago.

What has raised the current concern about these old kernels?
Why does this need fixing in upstream code?

-Eric