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

[Alberto Garcia]

On Mon 04 Nov 2019 04:14:56 PM CET, Max Reitz wrote:

>>> 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)

Here's with aio=native

|--------------+----------------+-----------------+-------------|
| Cluster size | subclusters=on | subclusters=off | fallocate() |
|--------------+----------------+-----------------+-------------|
|       256 KB |     19897 IOPS |        891 IOPS |  32194 IOPS |
|       512 KB |     17881 IOPS |        436 IOPS |  33092 IOPS |
|      1024 KB |     17050 IOPS |        341 IOPS |  32768 IOPS |
|      2048 KB |      7854 IOPS |        207 IOPS |  30944 IOPS |
|--------------+----------------+-----------------+-------------|

Berto