Re: [RFC 0/3] block/file-posix: Work around XFS bug

[Max Reitz <mreitz@redhat.com>]

[-- Attachment #1.1: Type: text/plain, Size: 2200 bytes --]

On 04.11.19 16:12, Alberto Garcia wrote:
> On Mon 04 Nov 2019 03:25:12 PM CET, Max Reitz wrote:
>>>>>> So, it's obvious that c8bb23cbdbe32f5c326 is significant for 1M
>>>>>> cluster-size, even on rotational disk, which means that previous
>>>>>> assumption about calling handle_alloc_space() only for ssd is wrong,
>>>>>> we need smarter heuristics..
>>>>>>
>>>>>> So, I'd prefer (1) or (2).
>>>>
>>>> OK.  I wonder whether that problem would go away with Berto’s subcluster
>>>> series, though.
>>>
>>> Catching up with this now. I was told about this last week at the KVM
>>> Forum, but if the problems comes with the use of fallocate() and XFS,
>>> the I don't think subclusters will solve it.
>>>
>>> handle_alloc_space() is used to fill a cluster with zeroes when there's
>>> COW, and that happens the same with subclusters, just at the subcluster
>>> level instead of course.
>>>
>>> What can happen, if the subcluster size matches the filesystem block
>>> size, is that there's no need for any COW and therefore the bug is never
>>> triggered. But that's not quite the same as a fix :-)
>>
>> No, what I meant was that the original problem that led to c8bb23cbdbe
>> would go away.
> 
> Ah, right. Not quite, according to my numbers:
> 
> |--------------+----------------+-----------------+-------------|
> | Cluster size | subclusters=on | subclusters=off | fallocate() |
> |--------------+----------------+-----------------+-------------|
> |       256 KB |     10182 IOPS |        966 IOPS |  14007 IOPS |
> |       512 KB |      7919 IOPS |        563 IOPS |  13442 IOPS |
> |      1024 KB |      5050 IOPS |        465 IOPS |  13887 IOPS |
> |      2048 KB |      2465 IOPS |        271 IOPS |  13885 IOPS |
> |--------------+----------------+-----------------+-------------|
> 
> There's obviously no backing image, and only the last column uses
> handle_alloc_space() / fallocate().

Thanks for providing some numbers!

It was my impression, too, that subclusters wouldn’t solve it.  But it
didn’t seem like that big of a difference to me.  Did you run this with
aio=native?  (Because that’s where we have the XFS problem)

Max


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]