Horton

Re: pl/perl extension fails on Windows

Robert Haas <robertmhaas@gmail.com>

View thread View on postgr.es
From: Robert Haas <robertmhaas@gmail.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Ashutosh Sharma <ashu.coek88@gmail.com>, Andrew Dunstan <andrew.dunstan@2ndquadrant.com>, Dave Page <dpage@pgadmin.org>, Sandeep Thakkar <sandeep.thakkar@enterprisedb.com>, "pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>
Date: 2017-07-25T15:39:33Z
Lists: pgsql-hackers 
On Tue, Jul 25, 2017 at 11:32 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Robert Haas <robertmhaas@gmail.com> writes:
>> On Tue, Jul 25, 2017 at 11:00 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>>> Hm, I had the idea that we were already asking ExtUtils::Embed for that,
>>> but now I see we only inquire about LDFLAGS not CCFLAGS.  Yes, this sounds
>>> like a promising avenue to pursue.
>>>
>>> It would be useful to see the results of
>>> perl -MExtUtils::Embed -e ccopts
>>> on one of the affected installations, and compare that to the problematic
>>> field(s).
>
>> Why ccopts rather than ccflags?
>
> I was looking at the current code which fetches ldopts, and analogizing.
> Don't know the difference between ccflags and ccopts.

Oh, here I was thinking you were 3 steps ahead of me.  :-)

Per "perldoc ExtUtils::Embed", ccopts() = perl_inc() plus ccflags()
plus ccdlflags().

On my system:

[rhaas pgsql]$ perl -MExtUtils::Embed -e 'for (qw(ccopts ccflags
ccdlflags perl_inc)) { print "==$_==\n"; eval "$_()"; print "\n\n";
};'
==ccopts==
 -fno-common -DPERL_DARWIN -mmacosx-version-min=10.12 -pipe -Os
-fno-strict-aliasing -fstack-protector-strong -I/opt/local/include
-DPERL_USE_SAFE_PUTENV
-I/opt/local/lib/perl5/5.24/darwin-thread-multi-2level/CORE

==ccflags==
 -fno-common -DPERL_DARWIN -mmacosx-version-min=10.12 -pipe -Os
-fno-strict-aliasing -fstack-protector-strong -I/opt/local/include
-DPERL_USE_SAFE_PUTENV

==ccdlflags==


==perl_inc==
 -I/opt/local/lib/perl5/5.24/darwin-thread-multi-2level/CORE

I don't have a clear sense of whether ccopts() or ccflags() is what we
want here, but FWIW ccopts() is more inclusive.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

Commits

  1. MSVC: Test whether 32-bit Perl needs -D_USE_32BIT_TIME_T.

  2. Further tweaks to compiler flags for PL/Perl on Windows.

  3. Absorb -D_USE_32BIT_TIME_T switch from Perl, if relevant.

  4. PL/Perl portability fix: absorb relevant -D switches from Perl.

  5. PL/Perl portability fix: avoid including XSUB.h in plperl.c.