On Wed, Oct 19, 2022 at 5:30 PM Alvaro Herrera <alvherre@xxxxxxxxxxxxxx> wrote: > That's exactly what I was trying to say. > If there's no compression, we don't read prior chunks. Great to read that. I'll probably try to benchmark w/ and w/o compression eventually. Need to deal with other issues first, will take a while to report back on this. > This can be seen in detoast_attr_slice() Thank you for the pointer. I'll be sure to have a look at that code. And thanks again for chiming in, with very useful info Alvaro.
