| From: | Nathan Bossart <nathandbossart(at)gmail(dot)com> |
|---|---|
| To: | "Devanga(dot)Susmitha(at)fujitsu(dot)com" <Devanga(dot)Susmitha(at)fujitsu(dot)com> |
| Cc: | "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org>, "Chiranmoy(dot)Bhattacharya(at)fujitsu(dot)com" <Chiranmoy(dot)Bhattacharya(at)fujitsu(dot)com>, "Ragesh(dot)Hajela(at)fujitsu(dot)com" <Ragesh(dot)Hajela(at)fujitsu(dot)com> |
| Subject: | Re: [PATCH] Hex-coding optimizations using SVE on ARM. |
| Date: | 2025-01-09 18:25:51 |
| Message-ID: | Z4AUr-Y-TPaqtg0y@nathan |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Thu, Jan 09, 2025 at 11:22:05AM +0000, Devanga(dot)Susmitha(at)fujitsu(dot)com wrote:
> This email aims to discuss the contribution of optimized hex_encode and
> hex_decode functions for ARM (aarch64) machines. These functions are
> widely used for encoding and decoding binary data in the bytea data type.
Thank you for sharing this work! I'm not able to review this in depth at
the moment, but I am curious if you considered trying to enable
auto-vectorization on the code or using the higher-level SIMD support in
src/include/port/simd.h. Those may not show as impressive of gains as your
patch, but they would likely require much less code and apply to a wider
set of architectures.
--
nathan
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Nathan Bossart | 2025-01-09 18:27:20 | Re: Fix a wrong errmsg in AlterRole() |
| Previous Message | Andres Freund | 2025-01-09 18:24:40 | Re: Eagerly scan all-visible pages to amortize aggressive vacuum |