Home > mailing lists

Confusing comment in tidbitmap.c - Mailing list pgsql-hackers

From	David Rowley
Subject	Confusing comment in tidbitmap.c
Date	December 14, 2014 11:03:37
Msg-id	CAApHDvrYG5ooAMkxbiEs3fs_wkqk0ibfii=ghmtNJ9rTb_nVKg@mail.gmail.com Whole thread Raw
Responses	Re: Confusing comment in tidbitmap.c (Tom Lane <tgl@sss.pgh.pa.us>)
List	pgsql-hackers

The following comment above #define PAGES_PER_CHUNK in tibbitmap.c appears to be incorrect:

"But we

* also want PAGES_PER_CHUNK to be a power of 2 to avoid expensive integer

* remainder operations. So, define it like this:"

I don't quite follow this as it does nothing of the kind. Check tbm_page_is_lossy() where we do: bitno = pageno % PAGES_PER_CHUNK;

Or am I missing something about the compiler optimizing that to: bitno = pageno & 255; ?

Regards

David Rowley

From: Amit Kapila
Date: 14 December 2014, 09:24:23
Subject: Re: pg_basebackup vs. Windows and tablespaces

From: David Rowley
Date: 14 December 2014, 14:58:16
Subject: Re: speedup tidbitmap patch: cache page