Re: reducing the footprint of ScanKeyword (was Re: Large writable variables)

On 2019-01-04 12:26:18 -0500, Tom Lane wrote:
> I wrote:
> > Andres Freund <andres(at)anarazel(dot)de> writes:
> >> On 2018-12-29 16:59:52 -0500, John Naylor wrote:
> >>> I think 0001 with complete keyword lookup replacement is in decent
> >>> enough shape to post. Make check-world passes. A few notes and
> >>> caveats:
>
> >> I tried to take this for a spin, an for me the build fails because various
> >> frontend programs don't have KeywordOffsets/Strings defined, but reference it
> >> through various functions exposed to the frontend (like fmtId()). That I see
> >> that error but you don't is probably related to me using -fuse-ld=gold in
> >> CFLAGS.
>
> > I was just about to point out that the cfbot is seeing that too ...
>
> Aside from the possible linkage problem, this will need a minor rebase
> over 4879a5172, which rearranged some of plpgsql's calls of
> ScanKeywordLookup.
>
> While I don't think it's going to be hard to resolve these issues,
> I'm wondering where we want to go with this. Is anyone excited
> about pursuing the perfect-hash-function idea? (Joerg's example
> function looked pretty neat to me.) If we are going to do that,
> does it make sense to push this version beforehand?

I think it does make sense to push this version beforehand. Most of
the code would be needed anyway, so it's not like this is going to
cause a lot of churn.

Greetings,

Andres Freund