Re: Vacuum statistics - Mailing list pgsql-hackers

Hi, all!

A few other changes in this version, beyond the five points above:

• I split the patch set so that each commit introduces one category of statistics. 
  In particular, recently_dead_tuples now lives in the same commit as missed_dead_tuples (patch 0003), where 
  it logically belongs - both are dead-tuple counters derived the same way, so keeping them together makes 
  the series easier to review.
• I also changed the order in which the counters are added across the commits. I'm still working out the right way 
  to verify the buffer statistics, timings and he frozen / visibility-map counters.

Thanks again for the review.

I have noticed that patches need to be rebased, so I have rebased them. 

I have noticed that I messed up with version indexes before and fixed it now - added 40th versions.

-- 
-----------
Best regards,
Alena Rybakina
Yandex Cloud