[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference bulova::decw_jan-89_to_nov-90

Title:DECWINDOWS 26-JAN-89 to 29-NOV-90
Notice:See 1639.0 for VMS V5.3 kit; 2043.0 for 5.4 IFT kit
Moderator:STAR::VATNE
Created:Mon Oct 30 1989
Last Modified:Mon Dec 31 1990
Last Successful Update:Fri Jun 06 1997
Number of topics:3726
Total number of notes:19516

2873.0. "Compilation flags used on DECwindows libaries" by RANDIE::ANDY (Make me rich, and you'll be happy) Tue Jun 05 1990 09:10

          <<< VOGON::DISK$JB:[NOTES$LIBRARY]COUNTRY_PORTING.NOTE;1 >>>
                -< European Country Porting centre conference >-
================================================================================
Note 61.0               DECwindows libs compilation flags             No replies
SAC::REILLY_G "Gerry"                                 7 lines  31-MAY-1990 15:34
--------------------------------------------------------------------------------
    Does anyone out there know what the compilation flags used for the
    DECwindows libraries was? 
    
    My reason - a customer wishes to dynamically load the libaries and
    therefore he can only handle libraries compiled with -G0.
    
    Cheers, Gerry

T.RTitleUserPersonal
Name
DateLines
2873.1DECWIN::FISHERPrune Juice: A Warrior&#039;s Drink!Wed Jun 06 1990 13:286
I presume this is an Ultrix question by the form of the option you mentioned,
correct?  But how do you dynamically load a library in Ultrix?

If this is VMS, I don't think I understand "-GO".

Burns
2873.2f & g floating?DECWIN::FISHERPrune Juice: A Warrior&#039;s Drink!Wed Jun 06 1990 13:319
Are you talking about, perhaps, the g-floating vs f-floating bit?  On VMS,
they are compiled and linked for f-floating.  On Ultrix, they are compiled with
pcc, not vcc.  I would venture a guess (but no certain knowledge) that pcc
just uses vanilla f-float as well.

Sometime in the future, the C folks have promised us a way to have mixed g/f
floating modules, but so far as I know, that time has not come yet.

Burns
2873.3QUARK::LIONELFree advice is worth every centWed Jun 06 1990 16:373
You do mean D_floating vs. G_floating, don't you Burns?

		Steve
2873.4DECWIN::FISHERPrune Juice: A Warrior&#039;s Drink!Wed Jun 06 1990 18:043
Whoops.  Yes, that is what I meant.  Thanks, Steve.

Burns
2873.5C has it nowTOOLEY::B_WACKERThu Jun 07 1990 13:507
>Sometime in the future, the C folks have promised us a way to have
>mixed g/f floating modules, but so far as I know, that time has not
>come yet. 

Mixed d and g float is possible now.  See the C 3.1 release notes.

Bruce
2873.6DECWIN::FISHERPrune Juice: A Warrior&#039;s Drink!Fri Jun 08 1990 13:344
Ok, thanks.  Apparently the relnotes are not on CD, so I don't
have them.

Burns
2873.7DECWIN::FISHERPrune Juice: A Warrior&#039;s Drink!Fri Jun 08 1990 13:398
However, the latest guide to VAX C says that you must not link together modules
complied with /G_floating and those without /G_floating (section 8.4).

In any case, we compile without /G_floating, and link with the non-G RTL, and
as of VMS V5.4, we use C V3.0.  I assume that means that you can't link with
G stuff.

Burnsf
2873.8Should be able to mix in /g using /defineTOOLEY::B_WACKERFri Jun 08 1990 14:3133
RE .6:
Release notes should be on your system for layered products.  On mine
$help cc release gives:

CC

  Release_Notes

     The release notes for VAX C are contained in the following file:

     SYS$HELP:VAXC031.RELEASE_NOTES

     You can  type  or  print  this  file  to  read  the  release  notes
     information.

RE .7:

The user guide is older than the (3.1) release notes.  If you use the
define mentioned in the release notes for "new" modules then you 
should be able to mix new g_float modules with DW because the /define 
will cause references to c$$g_double_mumble instead of just double in 
the new modules.

Before when there was only one name for "double" you got the
appropriate routine by only letting the linker see the right one,
which produced the limitation.  Now that there is alternate entry
points we can distinguish between double (g) and double (d) and make
it work. 

BTW I haven't actually done this, but looking at the include files the 
implementation is pretty pretty simple so it should work ;-).

Bruce
2873.9DECWIN::FISHERPrune Juice: A Warrior&#039;s Drink!Mon Jun 11 1990 13:435
Found it.  So this implies that even though we in DW land have not done anything
specific to support mixed-float, that customers should be able to use g-float
with DECwindows by recompiling with DEFINE=CC$mixedfloat or some such, right?

Burns
2873.10well kept secret??TOOLEY::B_WACKERMon Jun 11 1990 14:069
>Found it.  So this implies that even though we in DW land have not
>done anything specific to support mixed-float, that customers should
>be able to use g-float with DECwindows by recompiling with
>DEFINE=CC$mixedfloat or some such, right?

Right.  If the new code uses mixedfloat it should work with the the DW 
non-mixed stuff.

Bruce