Re: [PATCH] proc: move the adding option Ngid to the end of proc/PID/status

[Alexey Dobriyan]

On Fri, Apr 17, 2015 at 5:26 PM, Tejun Heo <tj@xxxxxxxxxx> wrote:
> Hey,
>
> On Fri, Apr 17, 2015 at 04:23:48PM +0300, Alexey Dobriyan wrote:
>> It was demonstrated that Ngid addition as line 4 breaks apps,
>> but your "what if" remains "what if".
>>
>> I'd say Ngid should be moved to the end and every new field
>> must be added to the end from now on, people can't parse
>> simple file correctly, let's not create problems for them.
>
> If this were in -rc or we are only a couple releases out, sure, moving
> that to the end would be the right thing to do but that's not the case
> and it bothers me that the patch essentially trades in about the same
> magnitude of unknown risk.  No matter which way you spin it, unknown
> risk of similar magnitude is not better than known risk and it's
> pretty certain that we'll have all three variants out in the wild for
> the foreseeable future.
>
> If this has to happen, it should be moving Ngid right after TracerPid
> not at the end of file.

Moving Ngid to the end of file minimizes risk of breakage.
Correctly written code doesn't care.
Code which hardcodes layout won't notice.

It would be OK argument if gentlemen from Intel send "let's futureproof and
move Ngid because someone might depend on exact position" patch.

Primum non nocere.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/