On Fri, Oct 2, 2020 at 8:14 AM Kyotaro Horiguchi <horikyota....@gmail.com> wrote: > > At Thu, 1 Oct 2020 12:55:34 +0000, "k.jami...@fujitsu.com" > <k.jami...@fujitsu.com> wrote in > > On Thursday, October 1, 2020 4:52 PM, Tsunakawa-san wrote: > > > > (I'm still mildly opposed to the function name, which seems exposing > detail too much.) >
Do you have any better proposal? BTW, I am still not sure whether it is a good idea to expose a new API for this especially because we do exactly the same thing in existing function smgrnblocks. Why not just add a new bool *cached parameter in smgrnblocks which will be set if we return cached value? I understand that we need to change the code wherever we call smgrnblocks or maybe even extensions if they call this function but it is not clear to me if that is a big deal. What do you think? I am not opposed to introducing the new API but I feel that adding a new parameter to the existing API to handle this case is a better option. -- With Regards, Amit Kapila.