On 03.09.2013 16:22, Merlin Moncure wrote:
On Mon, Sep 2, 2013 at 10:32 PM, Bruce Momjian wrote:
On Mon, Sep 2, 2013 at 10:14:03AM +0300, Heikki Linnakangas wrote:
diff --git a/src/backend/access/transam/xlog.c
b/src/backend/access/transam/xlog.c
index 39c58d0..28e62ea 100644
--- a/src/backe
Bruce Momjian wrote:
> On Mon, Sep 2, 2013 at 10:14:03AM +0300, Heikki Linnakangas wrote:
> > diff --git a/src/backend/access/transam/xlog.c
> > b/src/backend/access/transam/xlog.c
> > index 39c58d0..28e62ea 100644
> > - XLogInsertSlotPadded *insertSlots;
> > + /*
> > +* Make sure the ab
On Mon, Sep 2, 2013 at 10:32 PM, Bruce Momjian wrote:
> On Mon, Sep 2, 2013 at 10:14:03AM +0300, Heikki Linnakangas wrote:
>> diff --git a/src/backend/access/transam/xlog.c
>> b/src/backend/access/transam/xlog.c
>> index 39c58d0..28e62ea 100644
>> --- a/src/backend/access/transam/xlog.c
>> +++ b
On Mon, Sep 2, 2013 at 10:14:03AM +0300, Heikki Linnakangas wrote:
> diff --git a/src/backend/access/transam/xlog.c
> b/src/backend/access/transam/xlog.c
> index 39c58d0..28e62ea 100644
> --- a/src/backend/access/transam/xlog.c
> +++ b/src/backend/access/transam/xlog.c
> @@ -428,8 +428,14 @@ type
Now that I've had a little break from the big XLogInsert scaling patch,
I went back to do some testing and profiling of it. I saw a lot of
contention on the first access of RedoRecPtr and force/fullPageWrites,
which made me realize that I put those variables right next to the
heavily-contended