Re: Large writable variables - Mailing list pgsql-hackers

From Andres Freund
Subject Re: Large writable variables
Date
Msg-id 20181016063317.xjfd2dnlhy6tg5gx@alap3.anarazel.de
Whole thread Raw
In response to Re: Large writable variables  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: Large writable variables  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
Hi,

On 2018-10-16 01:59:00 -0400, Tom Lane wrote:
> I wrote:
> > Andres Freund <andres@anarazel.de> writes:
> >> top unitialized allocations:
> >> 0000000008435040 0000000000085280 b DCHCache
> >> 0000000008391168 0000000000043840 b NUMCache
> 
> > Here's a patch to improve that situation.
> 
> Hm, looking at that more closely, there's a problem with the idea of
> allocating the cache slots one-at-a-time.  Currently,
> sizeof(DCHCacheEntry) and sizeof(NUMCacheEntry) are each just a bit more
> than a power of 2, which would cause palloc to waste nearly 50% of the
> allocation it makes for them.

Hm, that's a bit annoying...


> We could forget the one-at-a-time idea and just allocate the whole
> array on first use, but I feel like that's probably not a good answer.

I suspect it'd be fine, but obviously we can do better.


> Also, I noticed that the biggest part of those structs are arrays of
> FormatNode, which has been designed with complete lack of thought about
> size or padding issues.  We can very easily cut it in half on 64-bit
> machines.

Heh, neat. I feel like we've paid very little attention to that in a
myriad of places :(

Greetings,

Andres Freund


pgsql-hackers by date:

Previous
From: Andres Freund
Date:
Subject: Re: Large writable variables
Next
From: Amit Kapila
Date:
Subject: Re: Pluggable Storage - Andres's take