Re: SunOS patch for memcmp()

> > > > > OK, what do people want with the memcmp() fix?  Tatsuo and I say apply,
> > > > > Tom is yes, or was,
> > > > Still is.  I don't want to gin up a strtol fix from scratch at this
> > > > late date in our cycle, but I think that the memcmp fix is safe.
> > > OK, good.  I will put the strtol on my list for 7.3.  The memcmp is much
> > > more significant.  Overflow is minor for most uses.
> >
> > Right. I'll plop SunOS back into the list of supported platforms for
> > this release. Thanks for the work Tatsuo and Bruce!
>
> OK, I have four votes for the patch, and one against.  I will apply it
> now.  We can consider SunOS supported.  There is the the problem that
> overflow is not detected by strtol but that is not a critical feature:
>
>   INSERT INTO INT4_TBL(f1) VALUES ('1000000000000');
> - ERROR:  pg_atoi: error reading "1000000000000": Numerical result out of range
>
> I will try to get that fixed for 7.3.

Actually, here is a fix for that too.  Patch attached but not applied.
I merely enabled our existing strtol.c for SunOS.  Peter, if you would
prefer memcmp.c to be handled the same way, I can do that too, with no
configure test at all, merely forcing use on SunOS.

--
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026
Index: src/backend/port/Makefile.in
===================================================================
RCS file: /cvsroot/pgsql/src/backend/port/Makefile.in,v
retrieving revision 1.30
diff -c -r1.30 Makefile.in
*** src/backend/port/Makefile.in    2001/12/20 21:23:05    1.30
--- src/backend/port/Makefile.in    2001/12/20 21:25:04
***************
*** 36,41 ****
--- 36,44 ----
  ifeq ($(PORTNAME), darwin)
  OBJS += darwin/SUBSYS.o
  endif
+ ifeq ($(PORTNAME), sunos)
+ OBJS += strtol.o
+ endif

  all: SUBSYS.o