Re: [PATCH] xfs: remove experimental tag for reflinks

[Amir Goldstein <amir73il@gmail.com>]

On Wed, Nov 15, 2017 at 3:10 AM, Darrick J. Wong
<darrick.wong@oracle.com> wrote:
> On Wed, Aug 30, 2017 at 04:54:00PM +0200, Christoph Hellwig wrote:
>> But reject reflink + DAX file systems for now until the code to
>> support reflinks on DAX is actually implemented.
>>
>> Signed-off-by: Christoph Hellwig <hch@lst.de>
>> ---
>>  fs/xfs/xfs_super.c | 8 +++-----
>>  1 file changed, 3 insertions(+), 5 deletions(-)
>>
>> diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c
>> index 38aaacdbb8b3..92521032468e 100644
>> --- a/fs/xfs/xfs_super.c
>> +++ b/fs/xfs/xfs_super.c
>> @@ -1634,7 +1634,9 @@ xfs_fs_fill_super(
>>               }
>>               if (xfs_sb_version_hasreflink(&mp->m_sb))
>>                       xfs_alert(mp,
>> -             "DAX and reflink have not been tested together!");
>> +             "DAX and reflink can not be used together!");
>> +                     error = -EINVAL;
>> +                     goto out_filestream_unmount;
>>       }
>>
>>       if (xfs_sb_version_hasrmapbt(&mp->m_sb)) {
>> @@ -1648,10 +1650,6 @@ xfs_fs_fill_super(
>>       "EXPERIMENTAL reverse mapping btree feature enabled. Use at your own risk!");
>>       }
>>
>> -     if (xfs_sb_version_hasreflink(&mp->m_sb))
>> -             xfs_alert(mp,
>> -     "EXPERIMENTAL reflink feature enabled. Use at your own risk!");
>
> Ok, now that the incore extent map rework has landed upstream, can we
> please get everyone's QA to stress the reflink/cow code vigorously
> during this release cycle so that we can re-consider this patch for
> 4.16?
>
> (And if you're particularly mean, force cowextsize = 1fsb to age the
> filesystem prematurely?)
>
> Right now I think the only problem I know about is that log recovery bug
> where the defer ops of replayed defer ops get logged in the wrong order,
> but that's it.
>

Last time you wrote about this bug you had a "hard question" about transaction
reservation for the solution and said your're going to go have a think about it:
https://marc.info/?l=linux-xfs&m=150766311924170&w=2
Did you come to any conclusions?
That sounds like one of those nasty CoW corner cases, so I'd be happy to know
there is at least a well thought design for a solution - if not a fix.

Practically, I would love if that bug could be solved soon so that we can all
start running generic/503 for more than a few iterations to stress
test reflink/cow
with power failure. Success on this front could be a big upside before
turning off
EXPERIMENTAL.

Thanks,
Amir.