On 07/15/2016 04:09 AM, Peter Lieven wrote:
> Am 05.07.2016 um 17:09 schrieb Paolo Bonzini:
>>
>> On 05/07/2016 16:59, Eric Blake wrote:
>>> And yes, we could probably switch to (potentially slower) / % * instead
>>> of bit operations in block/io.c to accommodate a non-power-of-2 optimal
>>> size, but it would require a careful audit to make sure we don't have
>>> even more bit-wise operations lurking that were assuming a power of 2.
>> Yes, it would require some auditing but it has always worked.  Regarding
>> speed, I'm sure that / % * are not slower than a system call! :)
> 
> Eric, whats the status of this? Will you send a follow-up to your series
> restoring the old behaviour?

Still on my plate.  As it is a bug-fix, the patch can go in even after
hard freeze, so at the moment I've unfortunately been more focused on
other patches (such as NBD_CMD_WRITE_ZEROES support) that were posted
before soft freeze, but which must be committed prior to hard freeze or
else they miss 2.7.  But rest assured that I'll make sure 2.7 doesn't
regress on your setup.


-- 
Eric Blake   eblake redhat com    +1-919-301-3266
Libvirt virtualization library http://libvirt.org