Re: Remove an unused function GetWalRcvWriteRecPtr - Mailing list pgsql-hackers

From Bharath Rupireddy
Subject Re: Remove an unused function GetWalRcvWriteRecPtr
Date
Msg-id CALj2ACUvUZGQ=xDKy_9b9GjKVnBjyuj90m=BdGyNOhGb5h7yEA@mail.gmail.com
Whole thread Raw
In response to Re: Remove an unused function GetWalRcvWriteRecPtr  (Julien Rouhaud <rjuju123@gmail.com>)
List pgsql-hackers
On Sat, Mar 26, 2022 at 12:55 PM Julien Rouhaud <rjuju123@gmail.com> wrote:
>
> On Sat, Mar 26, 2022 at 02:52:29PM +0900, Michael Paquier wrote:
> > On Sat, Mar 26, 2022 at 10:51:15AM +0530, Bharath Rupireddy wrote:
> > > The function GetWalRcvWriteRecPtr isn't being used anywhere, however
> > > pg_atomic_read_u64(&walrcv->writtenUpto); (reading writtenUpto without
> > > spinlock) is being used directly in pg_stat_get_wal_receiver
> > > walreceiver.c. We either make use of the function instead of
> > > pg_atomic_read_u64(&walrcv->writtenUpto); or remove it. Since there's
> > > only one function using walrcv->writtenUpto right now, I prefer to
> > > remove the function to save some LOC (13).
> > >
> > > Attaching patch. Thoughts?
> >
> > This could be used by some external module, no?
>
> Maybe, but WalRcv is exposed so if an external module needs it it could still
> maintain its own version of GetWalRcvWriteRecPtr.

Yes. And the core extensions aren't using GetWalRcvWriteRecPtr. IMO,
let's not maintain that function.

Regards,
Bharath Rupireddy.



pgsql-hackers by date:

Previous
From: vignesh C
Date:
Subject: Re: Skipping schema changes in publication
Next
From: Bharath Rupireddy
Date:
Subject: Re: pg_relation_size on partitioned table