On Tue, Apr 9, 2024 at 3:33 PM Jeff Davis wrote:
> Should I go ahead and commit something like that now, or hold it until
> the other thread concludes, or hold it until the July CF?
I think it's fine to commit it now if it makes it usefully easier to
fix an open item, and otherwise it should wait
On Tue, Apr 9, 2024 at 3:30 PM Jeff Davis wrote:
> Pages of warnings is not ideal, though. We should either support
> "SH_SCOPE static", or have some kind of useful #error that makes it
> clear that we don't support it (and/or don't think it's a good idea).
Fair.
> > I'm not sure that I like th
Hi,
On Tue, 2024-04-09 at 11:56 -0700, Andres Freund wrote:
> FWIW, with just about any modern-ish compiler just using "inline"
> doesn't
> actually force inlining, it just changes the cost model to make it
> more
> likely.
OK.
In the linked thread, I didn't see a good reason to encourage the
co
On Tue, 2024-04-09 at 14:49 -0400, Robert Haas wrote:
> Hmm. I'm pretty sure that I've run into this problem, but I concluded
> that I should use either "static inline" or "extern" and didn't think
> any more of it.
Pages of warnings is not ideal, though. We should either support
"SH_SCOPE static"
Hi,
On 2024-04-09 11:10:15 -0700, Jeff Davis wrote:
> The reason I'm suggesting it there is because the hash table is used
> only for the indexed binary heap, not an ordinary binary heap, so I'd
> like to leave it up to the compiler whether to do any inlining or not.
FWIW, with just about any mod
On Tue, Apr 9, 2024 at 2:10 PM Jeff Davis wrote:
> If using "SH_SCOPE static" with simplehash.h, it causes a bunch of
> warnings about functions that are defined but not used. It's simple
> enough to fix by appending pg_attribute_unused() to the declarations
> (attached).
Hmm. I'm pretty sure tha
If using "SH_SCOPE static" with simplehash.h, it causes a bunch of
warnings about functions that are defined but not used. It's simple
enough to fix by appending pg_attribute_unused() to the declarations
(attached).
There are currently no callers that use "SH_SCOPE static", but I'm
suggesting its
