Re: Coding in WalSndWaitForWal - Mailing list pgsql-hackers

From Kyotaro Horiguchi
Subject Re: Coding in WalSndWaitForWal
Date
Msg-id 20191113.171810.985192835121707471.horikyota.ntt@gmail.com
Whole thread Raw
In response to Re: Coding in WalSndWaitForWal  (Michael Paquier <michael@paquier.xyz>)
List pgsql-hackers
Ah, my stupid.

At Wed, 13 Nov 2019 16:34:49 +0900, Michael Paquier <michael@paquier.xyz> wrote in 
> On Tue, Nov 12, 2019 at 11:27:16AM -0800, Andres Freund wrote:
> > It seems to me it'd be better to just remove the "get a more recent
> > flush pointer" block - it doesn't seem to currently surve a meaningful
> > purpose.
> 
> +1.  That was actually my suggestion upthread :)

Actually it is useless as it is. But the code still seems to me an
incomplete fast path (that lacks immediate return after it) for the
case where just one call to GetFlushRecPtr advances RecentFlushPtr is
enough.

However, I'm not confident taht removing the (intended) fast path
impacts perforance significantly. So I don't object to remove it.

regards.

-- 
Kyotaro Horiguchi
NTT Open Source Software Center



pgsql-hackers by date:

Previous
From: Dilip Kumar
Date:
Subject: Re: [HACKERS] Block level parallel vacuum
Next
From: Fujii Masao
Date:
Subject: Re: Recovery performance of DROP DATABASE with many tablespaces