Horton

Re: Scaling XLog insertion (was Re: Moving more work outside WALInsertLock)

Fujii Masao <masao.fujii@gmail.com>

View thread
From: Fujii Masao <masao.fujii@gmail.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Heikki Linnakangas <heikki.linnakangas@enterprisedb.com>, Pg Hackers <pgsql-hackers@postgresql.org>, Simon Riggs <simon@2ndquadrant.com>, Alvaro Herrera <alvherre@commandprompt.com>, Jeff Janes <jeff.janes@gmail.com>, Robert Haas <robertmhaas@gmail.com>
Date: 2012-03-22T01:32:35Z
Lists: pgsql-hackers 
On Wed, Mar 21, 2012 at 10:59 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Heikki Linnakangas <heikki.linnakangas@enterprisedb.com> writes:
>> ... although none of the issues alone is a show-stopper, but considering
>> all these things together, I'm starting to feel that this needs to be
>> pushed to 9.3. Thoughts?
>
> Agreed.  In particular, I think you are right that it'd be prudent to
> simplify the WAL-location arithmetic and then rebase this code onto
> that.  And since no code at all has been written for the arithmetic
> change, I think we have to consider that it's not 9.2 material.

Agreed.

BTW, the patch changes some functions so that they use volatile pointer,
as follows:

@@ -8448,7 +9232,7 @@ XLogReportParameters(void)
 void
 UpdateFullPageWrites(void)
 {
-	XLogCtlInsert *Insert = &XLogCtl->Insert;
+	volatile XLogCtlInsert *Insert = &XLogCtl->Insert;

These changes should be applied?

Regards,

-- 
Fujii Masao
NIPPON TELEGRAPH AND TELEPHONE CORPORATION
NTT Open Source Software Center