Re: Speed up COPY TO text/CSV parsing using SIMD

From: sahil verma <sahilvermatokyo1(at)gmail(dot)com>
To: pgsql-hackers(at)lists(dot)postgresql(dot)org
Cc: ayoub(dot)kazar(at)data-bene(dot)io
Subject: Re: Speed up COPY TO text/CSV parsing using SIMD
Date: 2026-09-12 03:12:12
Message-ID: CAJ6aHM_DoDHOayJrQnJNJcHSX+AE1cK7mEe6iJZFcbRLEbVrhQ@mail.gmail.com
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi Ayoub,

I applied the patch to the latest master branch. It applied cleanly,
compiled without warnings, and all regression tests passed
successfully. I also manually verified the CSV output using a test
table.

The code changes are clear. The optimization to avoid the redundant
strlen() pass by reusing the tptr offset, and bulk-sending the clean
prefix when escapec == quotec, makes perfect sense and introduces no
logical regressions.

Regards,
Sahil

Browse pgsql-hackers by date

  From Date Subject
Next Message Chao Li 2026-09-12 03:14:08 Re: fixes for a few GUC descriptions
Previous Message Chao Li 2026-09-12 02:57:01 Re: REPACK (CONCURRENTLY) decoding worker is canceled by lock_timeout