| From: | Simon Riggs <simon(at)2ndquadrant(dot)com> |
|---|---|
| To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
| Cc: | pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: VirtualXactLockTableInsert |
| Date: | 2008-06-30 06:46:36 |
| Message-ID: | 1214808396.3845.434.camel@ebony.site |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Fri, 2008-06-27 at 18:00 -0400, Tom Lane wrote:
> Simon Riggs <simon(at)2ndquadrant(dot)com> writes:
> > In cases where we know we will assign a real xid, can we just skip the
> > assignment of the virtual xid completely?
>
> Even if we could do this I doubt it would be a good idea. It'd destroy
> the invariant that all transactions have a vxid, which at the very least
> would create naming problems.
Ahh, no, I meant go straight to assigning a real xid, to avoid the
wasted effort in inserting a vxid *and* a real xid.
--
Simon Riggs www.2ndQuadrant.com
PostgreSQL Training, Services and Support
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Simon Riggs | 2008-06-30 06:51:08 | Re: VirtualXactLockTableInsert |
| Previous Message | Jamie Deppeler | 2008-06-30 05:55:10 | Postgresql 8.3 issue |