On Tue, Feb 25, 2014 at 3:05 PM, Alex Elder <elder@xxxxxxxx> wrote:
>>> The other thing is that the expected size is limited by
>>> rbd_image_header->obj_order, which is a single byte. I
>>> think you should encode this the same way. Even if the
>>> hint were for more than RBD, this level of granularity
>>> may be sufficient.
>>>
>>>> + u64 expected_write_size;
>>>
>>> Probably the same thing here, a byte might be enough
>>> to encode this by using log2(expected_write_size).
>>>
>>>> + __u8 expected_size_probability;
>>
>> I think in the interest of genericity expected_object_size should be an
>> arbitrary, not limited to a power-of-2 value. Now, given the current
>> 90M object size limit 64 bits might seem a lot, but extent offset and
>> length are 64 bit values and to be future-proof I followed that here.
>
> I have no objection to the 64-bit size but I still think
> a byte representing log2(size) is sufficient. Power-of-2
> granularity is most likely fine (and might even be what
> such a hint gets converted to anyway) for file systems
> or other backing store. But again, you can do that with
> a 64 bit value as well.
Filesystems of course round it, but probably not to a power-of-2.
I guess it's adjusted to a multiple of block size and then capped with
some value that the allocator can cope with. xfs for example would
happily take on say 5M. power-of-2 is sufficient for rbd, but
_probably_ not in general.
Thanks,
Ilya
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
