Re: index prefetching

Hi,

On 2025-08-14 15:30:16 -0400, Peter Geoghegan wrote:
> On Thu Aug 14, 2025 at 3:15 PM EDT, Peter Geoghegan wrote:
> > On Thu, Aug 14, 2025 at 2:53 PM Andres Freund <andres(at)anarazel(dot)de> wrote:
> >> I think this is just an indicator of being IO bound.
> >
> > Then why does the exact same pair of runs show "I/O Timings: shared
> > read=194.629" for the sequential table backwards scan (with total
> > execution time 1132.360 ms), versus "I/O Timings: shared read=352.88"
> > (with total execution time 697.681 ms) for the random table backwards
> > scan?
>
> Is there any particular significance to the invalid op reports I also see in
> the same log files?

> $ cat sequential.txt | grep invalid | head
> 2025-08-14 14:35:03.278 EDT [2516983][client backend] [[unknown]][0/1:0] DEBUG: 00000: io 0 |op invalid|target invalid|state IDLE : wait_one io_gen: 2, ref_gen: 1, cycle 1
> 2025-08-14 14:35:03.278 EDT [2516983][client backend] [[unknown]][0/1:0] DEBUG: 00000: io 0 |op invalid|target invalid|state IDLE : wait_one io_gen: 3, ref_gen: 2, cycle 1

No - that's likely just that the IO completed and thus the handle was made
reusable (i.e. state IDLE). Note that the generation of IO we're waiting for
(ref_gen) is lower than the IO handle's (io_gen).

Greetings,

Andres Freund