Re: [PATCH 0/2 v2] btrfs: reject device with CHANGING_FSID_V2 flag

[Anand Jain <anand.jain@oracle.com>]

On 28/09/2023 01:43, David Sterba wrote:
> On Tue, Sep 26, 2023 at 07:56:57AM +0800, Anand Jain wrote:
>> On 26/09/2023 00:58, David Sterba wrote:
>>> On Thu, Sep 21, 2023 at 05:51:12AM +0800, Anand Jain wrote:
>>>> v2: Splits the patch into two parts: one for the new code to reject
>>>> devices with CHANGING_FSID_V2 and the second to remove the unused code.
>>>>
>>>> The btrfs-progs code to reunite devices with the CHANGING_FSID_V2 flag,
>>>> which is ported from the kernel, can be found in the following btrfs-progs
>>>> patchset:
>>>>
>>>>    [PATCH 0/4 v4] btrfs-progs: recover from failed metadata_uuid port kernel
>>>>       btrfs-progs: tune use the latest bdev in fs_devices for super_copy
>>>>       btrfs-progs: add support to fix superblock with CHANGING_FSID_V2 flag
>>>>       btrfs-progs: recover from the failed btrfstune -m|M
>>>>       btrfs-progs: test btrfstune -m|M ability to fix previous failures
>>>>
>>>> Furthermore, when the kernel stops supporting the CHANGING_FSID_V2 flag,
>>>> we will need to update the btrfs-progs test case accordingly:
>>>>
>>>>       btrfs-progs: misc-tests/034-metadata-uuid remove kernel support
>>>>
>>>> v1:
>>>> https://lore.kernel.org/linux-btrfs/02d59bdd7a8b778deb17e300354558498db59376.1692178060.git.anand.jain@oracle.com
>>>>
>>>>
>>>> Anand Jain (2):
>>>>     btrfs: reject device with CHANGING_FSID_V2
>>>>     btrfs: remove unused code related to the CHANGING_FSID_V2 flag
>>>
>>
>>
>>> Added to misc-next, thanks.
>>
>>> I've updated the 2nd patch with some
>>> historical background.
>>
>> Now much more complete! Thanks.
>>
>>> The temp-fsid patch now does not apply cleanly,
>>> I'll do a refresh on top of this series. Once it's in for-next, please
>>> have a look. Thanks.
>>
>> Sure. But, temp-fsid v4 still has a subvol mount bug, as reported
>> before. I have a fix that is in-memory only. I am trying to get a
>> reasonable list of fstests passed before sending it out. Guilherme
>> can add the super-block flag on top of this, although it is not
>> mandatory from the btrfs internals pov. I will send out the
>> in-memory based temp-fsid soon I get the fstests (with some fix)
>> working today.
> 
> Feel free the send the patches even if there are still failing tests,
> I'd like to get an idea of how are you fixing the remaining problem(s).
> Thanks.

Yep, it is almost ready. A few minor cleanups and code optimizations
are still due. However, I have sent them for the review comments.
Below are the patches in the mailing list now. Also, after sending
them, I realized it would have been easier for the searches if the
subject included 'temp-fsid' instead of 'cloned-device'

  [PATCH 0/2] btrfs: support cloned-device mount capability

  [PATCH 0/2] btrfs-progs: mkfs: testing cloned-device


Thanks, Anand