Re: Relation extension scalability

On 22/03/16 10:15, Dilip Kumar wrote:
>
> On Mon, Mar 21, 2016 at 8:10 PM, Amit Kapila <amit(dot)kapila16(at)gmail(dot)com
> <mailto:amit(dot)kapila16(at)gmail(dot)com>> wrote:
> 11.
> +{
> +/*
> +* First try to get the lock in no-wait mode, if succeed extend one
> +
> * block, else get the lock in normal mode and after we get the lock
> -LockBuffer(buffer, BUFFER_LOCK_EXCLUSIVE);
> +buffer =
> RelationAddOneBlock(relation, otherBuffer, bistate);
>
> Won't this cause one extra block addition after backend extends the
> relation for multiple blocks, what is the need of same?
>
>
> This is the block for this backend, Extra extend for future request and
> already added to FSM. I could have added this count along with extra
> block in RelationAddExtraBlocks, But In that case I need to put some
> extra If for saving one buffer for this bakend and then returning that
> the specific buffer to caller, and In caller also need to distinguish
> between who wants to add one block or who have got one block added in
> along with extra block.
>
> I think this way code is simple.. That everybody comes down will add one
> block for self use. and all other functionality and logic is above, i.e.
> wether to take lock or not, whether to add extra blocks or not..
>
>

I also think the code simplicity makes this worth it.

--
Petr Jelinek http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services