Re: xact.c state machine redesign - Mailing list pgsql-hackers

From Alvaro Herrera
Subject Re: xact.c state machine redesign
Date
Msg-id 20040915223326.GA7813@dcc.uchile.cl
Whole thread Raw
In response to xact.c state machine redesign  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
On Wed, Sep 15, 2004 at 01:34:01PM -0400, Tom Lane wrote:
> After looking over the state machine in xact.c, I'm thinking of removing
> the TBLOCK_SUBENDABORT_ALL and TBLOCK_SUBENDABORT_RELEASE states in
> favor of having the ROLLBACK command mark the whole transaction state
> stack similarly to what is now done for COMMIT.

FYI, this idea works for me.

> This isn't any fewer states than we have now, but the states seem much
> more clearly organized to me --- in particular, other than the RESTART
> states there's full symmetry between outer-level and subtransaction
> states.

Excellent.  This was my initial idea, but I messed up in the transition
to savepoints.

-- 
Alvaro Herrera (<alvherre[a]dcc.uchile.cl>)
"Postgres is bloatware by design: it was built to house
PhD theses." (Joey Hellerstein, SIGMOD annual conference 2002)



pgsql-hackers by date:

Previous
From: Mark Wong
Date:
Subject: Re: some PITR performance data with DBT-2
Next
From: "Marc G. Fournier"
Date:
Subject: PostgreSQL Core Committee Welcomes New Member