Thread

1. VAX portt was Re: [HACKERS] configure on linux

   Andrew Martin <martin@biochemistry.ucl.ac.uk> — 1998-02-06T11:22:57Z

   > You misunderstand me.  I didn't suggest removing the S_LOCK() et al
   > macros.  What I meant was that the actual assembly implementation of
   > tas() itself might be better off in a separate source file.  As an
   > example, here is my current version of the locking code for the VAX,
   > in s_lock.h (bbssi is "branch on bit set and set, interlocked"):
   > 
   > static int tas(slock_t *lock) {
   
   With the difficult bit done, maybe I should take a look at getting it
   running on the old VMS uVAX-II I have sitting at home :-)
   
   Andrew
   
   ----------------------------------------------------------------------------
   Dr. Andrew C.R. Martin                             University College London
   EMAIL: (Work) martin@biochem.ucl.ac.uk    (Home) andrew@stagleys.demon.co.uk
   URL:   http://www.biochem.ucl.ac.uk/~martin
   Tel:   (Work) +44(0)171 419 3890                    (Home) +44(0)1372 275775
   
   

2. Re: VAX portt was Re: [HACKERS] configure on linux

   Tom Ivar Helbekkmo <tih@hamartun.priv.no> — 1998-02-06T19:58:27Z

   On Fri, 6 Feb 1998, Andrew Martin wrote:
   
   > With the difficult bit done, maybe I should take a look at getting
   > it running on the old VMS uVAX-II I have sitting at home :-)
   
   That would be cool -- but no, the difficult bit is not done.  If you
   want to get PostgreSQL to work well under VMS, especially on a II,
   where you can't run anything newer than 5.5, and thus can't use the
   POSIX interface stuff from VMS 6, you'll need to rewrite the access
   methods at the lowest level to utilize RMS properly, or performance
   will absolutely suck.  As an example, I got well over an order of
   magnitude of speed improvement for the IOZONE benchmark under VMS by
   writing a couple of tens of lines of RMS-savvy interface code...
   
   Oh.  You weren't serious.  OK.
   
   However, the VAX port is running nicely under NetBSD.  It seems to
   be sane -- reading through 100 Kbytes of regression.diffs for the
   Sparc port and 60 Kbytes for the VAX port is hard work, especially
   since the diffs are all either slightly differently worded error
   messages, tiny precision differences in floating point math hardware,
   zero instead of minus zero in math results or different limits on
   max/min representable numbers.  In addition, there are a number of
   date and time differences that I don't understand -- like why a minute
   is sometimes "1 min" and sometimes "1 min 60.00 sec"...
   
   Anyway, the patches for NetBSD/vax are sent separately, with comments.
   
   -tih
   -- 
   Popularity is the hallmark of mediocrity.  --Niles Crane, "Frasier"