On 3/29/16 7:16 PM, Dave Chinner wrote:
> On Tue, Mar 29, 2016 at 02:28:28PM -0500, Eric Sandeen wrote:
>> Today, a kernel which refuses to mount a filesystem read-write
>> due to unknown ro-compat features can still transition to read-write
>> via the remount path. The old kernel is most likely none the wiser,
>> because it's unaware of the new feature, and isn't using it. However,
>> writing to the filesystem may well corrupt metadata related to that
>> new feature, and moving to a newer kernel which understand the feature
>> will have problems.
>>
>> Right now the only ro-compat feature we have is the free inode btree,
>> which showed up in v3.16. It would be good to push this back to
>> all the active stable kernels, I think, so that if anyone is using
>> newer mkfs (which enables the finobt feature) with older kernel
>> releases, they'll be protected.
>
> Ok, so the bug was introduced with the original extended feature
> masks in commit e721f50 ("xfs: implement extended feature masks"),
> which was introduced in 3.10. So it will need to go back to stable
> kernels all the way back to 3.10, right?
>
> Cheers,
>
> Dave.
>
Yeah, that seems right; it's not when the first ro-compat feature
was introduced, it was when the handling was introduced, right?
If you need to massage the cc:stable line feel free, of course.
ie:
Cc: <stable@xxxxxxxxxxxxxxx> # 3.10.x-
I think.
-Eric
|