Re: CustomScan support on readfuncs.c

> -----Original Message-----
> From: Robert Haas [mailto:robertmhaas(at)gmail(dot)com]
> Sent: Saturday, October 03, 2015 5:44 AM
> To: Kaigai Kouhei(海外 浩平)
> Cc: Amit Kapila; Andres Freund; pgsql-hackers
> Subject: ##freemail## Re: [HACKERS] CustomScan support on readfuncs.c
>
> On Tue, Sep 29, 2015 at 6:19 PM, Kouhei Kaigai <kaigai(at)ak(dot)jp(dot)nec(dot)com> wrote:
> >> On Mon, Sep 28, 2015 at 8:31 PM, Kouhei Kaigai <kaigai(at)ak(dot)jp(dot)nec(dot)com> wrote:
> >> >> Instead of doing this:
> >> >>
> >> >> + /* Dump library and symbol name instead of raw pointer */
> >> >> appendStringInfoString(str, " :methods ");
> >> >> - _outToken(str, node->methods->CustomName);
> >> >> + _outToken(str, node->methods->methods_library_name);
> >> >> + appendStringInfoChar(str, ' ');
> >> >> + _outToken(str, node->methods->methods_symbol_name);
> >> >>
> >> >> Suppose we just make library_name and symbol_name fields in the node
> >> >> itself, that are dumped and loaded like any others.
> >> >>
> >> >> Would that be better?
> >> >>
> >> > I have no preference here.
> >> >
> >> > Even if we dump library_name/symbol_name as if field in CustomScan,
> >> > not CustomScanMethods, in a similar way, we cannot use WRITE_STRING_FIELD
> >> > here, because its 'fldname' assumes these members are direct field of
> >> > CustomScan.
> >> >
> >> > /* Write a character-string (possibly NULL) field */
> >> > #define WRITE_STRING_FIELD(fldname) \
> >> > (appendStringInfo(str, " :" CppAsString(fldname) " "), \
> >> > _outToken(str, node->fldname))
> >>
> >> Well that's exactly what I was suggesting: making them a direct field
> >> of CustomScan.
> >>
> > Let me confirm. Are you suggesting to have library_name/symbol_name
> > in CustomScan, not CustomScanMethods?
> > I prefer these fields are in CustomScanMethods because we don't need
> > to setup them again once PG_init set up these symbols. CustomScan has
> > to be created every time when it is chosen by planner.
>
> True. But that doesn't cost much. I'm not sure it's better, so if
> you don't like it, don't worry about it.
>
Yep, I like library_name and symbol_name are in CustomScanMethods
because the table of callbacks and its identifiers are not separable

> >> > One other question I have. Do we have a portable way to lookup
> >> > a pair of library and symbol by address?
> >> > Glibc has dladdr() functions that returns these information,
> >> > however, manpage warned it is not defined by POSIX.
> >> > If we would be able to have any portable way, it may make the
> >> > interface simpler.
> >>
> >> Yes: load_external_function.
> >>
> > It looks up an address by a pair of library and symbol name....
> > What I'm looking for is a portable function that looks up a pair
> > of library and symbol name by an address, like dladdr() of glibc.
> > I don't know whether other *nix or windows have these infrastructure.
>
> No, that doesn't exist, and the chances of us trying add that
> infrastructure for this feature are nil.
>
OK, probably, it is the only way to expect extension put correct
values on the pair of library and symbol names.

I try to check whether the current patch workable with this direction
using my extension. Please wait for a few days.

Thanks,
--
NEC Business Creation Division / PG-Strom Project
KaiGai Kohei <kaigai(at)ak(dot)jp(dot)nec(dot)com>