RE: SIMD optimization for list_sort - Mailing list pgsql-hackers

From Shankaran, Akash
Subject RE: SIMD optimization for list_sort
Date
Msg-id PH0PR11MB5000875C4A45C9B653F28200F22DA@PH0PR11MB5000.namprd11.prod.outlook.com
Whole thread Raw
In response to Re: SIMD optimization for list_sort  (Álvaro Herrera <alvherre@kurilemu.de>)
List pgsql-hackers
Hi Alvaro and John,

> I think John's reservations on whether to absorb another requisite shared library is something to strongly consider
--I notice that x86-simd-sort is not yet packaged by Debian, for instance.
 
> Maybe your needs *are* served better by introducing simd-sort in pgvector after all?  Alternatively: would it be too
badto reimplement simdsort in C within Postgres rather than using a third party library?
 

Thanks for your feedback and discussions here. 

> I've marked this commitfest item[1] as Returned with Feedback.  Please feel free to reopen it once you have an
updatedpatch to share.
 
At this time, I'd like to withdraw the commitfest item [1]. The original developer who proposed the patch is no longer
atIntel, and the x86-simd-sort library's future is also uncertain. Due to ongoing changes on our end at this time, we
won'tbe able to support this work going forward.  :- (
 

[1] https://commitfest.postgresql.org/patch/5607/

Akash Shankaran

pgsql-hackers by date:

Previous
From: Pavel Stehule
Date:
Subject: Re: event trigger support for PL/Python
Next
From: Peter Geoghegan
Date:
Subject: Re: index prefetching