| From: | Greg Stark <stark(at)mit(dot)edu> |
|---|---|
| To: | Heikki Linnakangas <heikki(dot)linnakangas(at)enterprisedb(dot)com> |
| Cc: | Peter Geoghegan <peter(at)2ndquadrant(dot)com>, Robert Haas <robertmhaas(at)gmail(dot)com>, Bruce Momjian <bruce(at)momjian(dot)us>, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>, Simon Riggs <simon(at)2ndquadrant(dot)com>, Peter Eisentraut <peter_e(at)gmx(dot)net>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org>, Jan Urbański <wulczer(at)wulczer(dot)org> |
| Subject: | Re: Large C files |
| Date: | 2011-09-09 22:54:18 |
| Message-ID: | CAM-w4HOjJQu9J8skM3CODRCoyJo2pwukGisxsjRTvSfAF4bPMA@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Fri, Sep 9, 2011 at 6:57 AM, Heikki Linnakangas
<heikki(dot)linnakangas(at)enterprisedb(dot)com> wrote:
>> In particular, I'd like to know what
>> boundaries it is envisaged that the code should be refactored along to
>> increase its conceptual integrity, or to better separate concerns. I
>> assume that that's the idea, since each new .c file would have to have
>> a discrete purpose.
>
> I'd like to see it split into routines involved in writing WAL, and those
> involved in recovery. And maybe a third file for archiving-related stuff.
Having a clean API for working with WAL independently of recovery
would let us have a maintainable xlogdump tool that doesn't constantly
get out of sync with the wal archive format. It would also make it
easier to write things like prefretching logic that requires reading
the upcoming xlog before its time to actually replay it.
--
greg
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Bruce Momjian | 2011-09-09 23:06:49 | Re: Patch to improve reliability of postgresql on linux nfs |
| Previous Message | Greg Stark | 2011-09-09 22:48:17 | Re: Protecting against multiple instances per cluster |