| From: | Peter Eisentraut <e99re41(at)DoCS(dot)UU(dot)SE> |
|---|---|
| To: | Jan Wieck <wieck(at)debis(dot)com> |
| Cc: | PostgreSQL HACKERS <pgsql-hackers(at)postgreSQL(dot)org> |
| Subject: | Re: [HACKERS] TODO item |
| Date: | 2000-02-08 12:57:15 |
| Message-ID: | Pine.GSO.4.02A.10002081348140.12742-100000@Krokodil.DoCS.UU.SE |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Tue, 8 Feb 2000, Jan Wieck wrote:
> And if this storage manager would work with some amount of
> preallocated blocks, it would be totally happy with a
> fdatasync() instead of a fsync(). Some per tablespace
> configurable options like initial number of blocks, next
> extent size and percentage increase would be fine.
On Linux, fdatasync() does exactly the same as fsync(). On FreeBSD (3.4),
fdatasync() isn't even documented and I can't find it in any of the
include files either. What I'm saying is that for the vast majority of our
users this would most likely buy exactly nothing. I just wanted to point
that out, not dismiss the idea.
--
Peter Eisentraut Sernanders vaeg 10:115
peter_e(at)gmx(dot)net 75262 Uppsala
http://yi.org/peter-e/ Sweden
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Lamar Owen | 2000-02-08 13:02:36 | Re: [HACKERS] Questions on 7.0 for RPM building |
| Previous Message | Lamar Owen | 2000-02-08 12:53:23 | Re: [HACKERS] Questions on 7.0 for RPM building |