Bruce Momjian <pgman@candle.pha.pa.us> writes:
> OK, clearly your looking at the bit is better than what we have now, so
> how about committing something that looks at the bit, but leave the
> default at zero. Then, let people test zero and non-zero delays and
> let's see what they find. That seems safe because we aren't enabling
> the problematic delay by default, at least until we find it is a help in
> most cases.
What I think I will do is write the code and try some pgbench tests with
scalefactor > 1. If that looks promising, I'll post or commit the code
and ask people to do more tests. We can hold off changing the default
delay back to nonzero until we have more data...
regards, tom lane