On Thu, Jun 23, 2022 at 2:09 AM Robert Haas <robertmh...@gmail.com> wrote: > On Wed, Jun 22, 2022 at 12:34 AM Thomas Munro <thomas.mu...@gmail.com> wrote: > > > For the record, the third idea proposed was to use 1 for the first > > > byte, so that 0 is reserved for NULL and works with memset(0). Here's > > > an attempt at that. > > > > ... erm, though, duh, I forgot to adjust Assert(val > base). One more time. > > I like this idea and think this might have the side benefit of making > it harder to get away with accessing relptr_off directly.
Thanks. Pushed, and back-patched to 14, where min_dynamic_shared_memory arrived. I wondered in passing if the stuff about relptr_declare() was still needed to avoid confusing pgindent, since we tweaked the indent code a bit for macros that take a typename, but it seems that it still mangles "relptr(FooBar) some_struct_member;", putting extra whitespace in front of it. Hmmph.
