On Thu, Aug 1, 2024 at 7:29 AM Hayato Kuroda (Fujitsu)
wrote:
>
> > Thanks for reviewing the patch, and your understanding is correct.
> >
> > Here is the updated patch 0001. I removed the comments as suggested by Amit.
> >
> > Since 0002 patch is only refactoring the code and I need some time to
Dear Hou,
> Thanks for reviewing the patch, and your understanding is correct.
>
> Here is the updated patch 0001. I removed the comments as suggested by Amit.
>
> Since 0002 patch is only refactoring the code and I need some time to review
> the comments for it, I will hold it until the 0001 is
On Wednesday, July 31, 2024 5:07 PM Kuroda, Hayato/黒田 隼人
wrote:
>
> Dear Hou,
>
> > When reviewing the code in logical/worker.c, I noticed that when
> > applying a cross-partition update action, it scans the old partition twice.
> > I am attaching the patch 0001 to remove this duplicate table s
Dear Hou,
Thanks for creating a patch!
> When reviewing the code in logical/worker.c, I noticed that when applying a
> cross-partition update action, it scans the old partition twice.
> I am attaching the patch 0001 to remove this duplicate table scan.
Just to clarify, you meant that FindReplTup
On Thu, Jul 25, 2024 at 5:56 PM Kirill Reshke wrote:
>
> > +/*
> > + * If the tuple to be modified could not be found, a log message is
> > emitted.
> > + */
> > +static void
> > +report_tuple_not_found(CmdType cmd, Relation targetrel, bool is_partition)
> > +{
> > + Assert(cmd == CMD_UPDATE || c
On Thu, Jul 25, 2024 at 4:00 PM Zhijie Hou (Fujitsu)
wrote:
>
> When reviewing the code in logical/worker.c, I noticed that when applying a
> cross-partition update action, it scans the old partition twice.
> I am attaching the patch 0001 to remove this duplicate table scan.
>
> The test shows tha
On Thu, Jul 25, 2024 at 5:56 PM Kirill Reshke wrote:
> > +/*
> > + * If the tuple to be modified could not be found, a log message is
> > emitted.
> > + */
> > +static void
> > +report_tuple_not_found(CmdType cmd, Relation targetrel, bool is_partition)
> > +{
> > + Assert(cmd == CMD_UPDATE || cm
Hi!
> When reviewing the code in logical/worker.c, I noticed that when applying a
> cross-partition update action, it scans the old partition twice.
Nice catch!
> -/*
> - * Workhorse for apply_handle_update()
> - * relinfo is for the relation we're actually updating in
> - * (could be a child p
Hi,
When reviewing the code in logical/worker.c, I noticed that when applying a
cross-partition update action, it scans the old partition twice.
I am attaching the patch 0001 to remove this duplicate table scan.
The test shows that it brings noticeable improvement:
Steps
-
Pub:
create table
