Ron Mayer <rm_pg@cheapcomplexdevices.com> writes:
> Tom Lane wrote:
>> A backwards scan will get no such overlapping and thus be up to 2X
>> slower, unless the kernel is smart enough to do read-ahead for
>> descending-order read requests. Which seems not too probable.
> Linux's old adaptive readahead patches claimed to[1]:
I didn't say that there were *no* platforms that could do it.
regards, tom lane