On Tue, Sep 16, 2014 at 6:49 AM, Heikki Linnakangas
<hlinnakangas@vmware.com> wrote:
>>>> As it happens, I also wrote an implementation of Slice-by-4 the other
>>>> day
>>>
>> If Heikki's version works I see little need to use my/Abhijit's
>> patch. That version has part of it under the zlib license. If Heikki's
>> version is a 'clean room', then I'd say we go with it. It looks really
>> quite similar though... We can make minor changes like additional
>> unrolling without problems lateron.
>
>
> I used http://create.stephan-brumme.com/crc32/#slicing-by-8-overview as
> reference - you can probably see the similarity. Any implementation is going
> to look more or less the same, though; there aren't that many ways to write
> the implementation.
So, it seems like the status of this patch is:
1. It probably has a bug, since Amit's testing seemed to show that it
wasn't returning the same results as unpatched master.
2. The performance tests showed a significant win on an important workload.
3. It's not in any CommitFest anywhere.
Given point #2, it's seems like we ought to find a way to keep this
from sliding into oblivion.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
