Compiler Hosts and Targets

Compiler Hosts and Targets

[Tucker Taft]

As part of building up a set of materials for companies evaluating
Ada for their programming needs, here is an initial list of
hosts and targets supported by various Ada compiler vendors,
as well as a list of companies selling Ada-supportive tools.
We would appreciate any comments, additions or deletions,
etc.

Thanks in advance.
-Tucker Taft   stt@averstar.com   http://www.averstar.com/~stt/
Technical Director, Intermetrics, Inc.  Burlington, MA  USA
An AverStar Company (www.averstar.com)

and also, President, Ada Resource Association (www.adaresource.org).
-------------------
            A d a    C o m p i l e r s   a n d   T o o l s

The following is a partial list of Hosts, Targets, and operating systems 
supported by various Ada compiler vendors:

    Hosts:  
       SPARC/Solaris, HP9000/HP-UX, RS6000/AIX, IBM390/MVS,
       Intelx86/Win-NT/9X, SGI/IRIX, Concurrent/PowerMAX, Intelx86/Linux,
       Intelx86/OS/2, PowerMAC/Tenon, VAX/VMS, Alpha/Open-VMS,
       Intelx86/DOS, Alpha/DEC-Unix, Siemens-Nixdorf RM200/SINIX

    Targets: All the above hosts plus:
       PowerPC/VxWorks, PowerPC/Raven, Intelx86/PharLapETS,
       PowerPC/Integrity, HP7xx/HP-RT, MIPS/VxWorks, RAD6000/VxWorks,
       68K/VxWorks, Pentium/VxWorks, ADI-21020/Bare, i960/HAOS,
       PowerPC/LynxOS, ADI-SHARC/Virtuoso, IBM390/CICS, Nighthawk 6800

The Hosts and Targets list is not complete, as some compilers
are used in-house without being officially validated, while others
are in the process of development.  In particular, there are a 
number of GNAT-derivatives being used for production work for targets
or hosts not listed above.

There are also Ada 95 compilers that generate Java byte codes, as well
as those that generate optimized ISO/ANSI C as their intermediate code,
allowing integration with JVM-based environments, as well as with any existing
ISO/ANSI-C-supportive environment.  In addition, there is the GNAT
compiler supported by Ada Core Technologies, which targets the
GNU "GCC" optimizer and backend.  As with GCC, the GNAT compiler operates 
under the "open source" principle (the GNAT front end is written
in Ada 95 itself).  Given these three integration approaches 
(JVM, ISO/ANSI C, and GCC), plus the existing retargetable
technologies of the other compiler vendors, an Ada 95 capability
is readily available on essentially any platform of interest now
or in the future.


The Ada compiler and tool market is served by a number of software vendors.  

The following vendors provide multiple Ada compilers (those marked with a
"*" are members of the Ada Resource Association -- www.adaresource.org):

    *Ada Core Technologies  (www.gnat.com)
    *Aonix (www.aonix.com)
    *AverStar (formerly Intermetrics)  (www.averstar.com)
    *DDC-I  (www.ddci.com)
    *Green Hills Software  (www.ghs.com)
     Irvine Compiler Corp (www.irvine.com)
    *O. C. Systems (www.ocsystems.com)
    *Rational Software (www.rational.com)
     RR Software (www.rrsoftware.com)
     TLD Systems

There are also companies who just offer compilers for their own
hardware or chips; for example:

    Analog Devices (for the SHARC) (www.analog.com)
    Concurrent (www.ccur.com)
    SGI (www.sgi.com)

There are also a number of tool companies which provide Ada-supportive tools;
for example:

    Ainslie Software (AdaJNI -- www.ainslie-software.com)
    GrammaTech (Ada-Assured -- www.grammatech.com)
    Integrated Computer Solutions (ICS Ada Xcessories -- www.ics.com)
    IPL (AdaTest -- www.iplbath.com)
    McCabe and Associates (McCabe QA/Test/Reengineer/2000 -- www.mccabe.com)
    Objective Interface Systems (ORBExpress/Ada -- www.ois.com)
    Praxis Critical Systems (SPARK -- www.praxis-cs.co.uk)
    Vector Software (AdaCAST -- www.vectors.com)

A number of the compiler vendors also sell Ada-supportive tools
The Ada compiler and tool market is approximately $100 Million.

Ada training and mentoring services are also available from many
of the above product companies, plus from various other companies 
and independent consultants.

For a sampling of projects that have used Ada successfully, see
http://www.adaic.org/projects/successes.html.

Re: Compiler Hosts and Targets

[Robert I. Eachus]

In article <F3z902.DyI.0.-s@inmet.camb.inmet.com> stt@houdini.camb.inmet.com (Tucker Taft) writes:

  > As part of building up a set of materials for companies evaluating
  > Ada for their programming needs, here is an initial list of
  > hosts and targets supported by various Ada compiler vendors,
  > as well as a list of companies selling Ada-supportive tools.
  > We would appreciate any comments, additions or deletions,
  > etc...

  > The Hosts and Targets list is not complete, as some compilers
  > are used in-house without being officially validated, while others
  > are in the process of development.  In particular, there are a 
  > number of GNAT-derivatives being used for production work for targets
  > or hosts not listed above.

    Good start, but it would be nice to include some of those mot
officially validated compilers and where to find them.  I use Gnat on
a 68K-based Amiga running AmigaDOS at home (see www.ninemoons.com) and
still on SunOS as well as Solaris at work.
--

					Robert I. Eachus

with Standard_Disclaimer;
use  Standard_Disclaimer;
function Message (Text: in Clever_Ideas) return Better_Ideas is...

Re: Compiler Hosts and Targets

[dennison]

In article <F3z902.DyI.0.-s@inmet.camb.inmet.com>,
  stt@houdini.camb.inmet.com (Tucker Taft) wrote:

>     Targets: All the above hosts plus:
>        68K/VxWorks, Pentium/VxWorks, ADI-21020/Bare, i960/HAOS,

Non-Ada issue: There is no such thing as Pentium/VxWorks, unfortunately.
There is an Intel3(or4)86/VxWorks target, which may be used on Pentiums
(although it will run *slower* unless you change the caching behavior!). But
if you use that on a Pentium it will not generated Pentium optimized code.

> There are also a number of tool companies which provide Ada-supportive tools;
> for example:
>
>     Ainslie Software (AdaJNI -- www.ainslie-software.com)
>     GrammaTech (Ada-Assured -- www.grammatech.com)
>     Integrated Computer Solutions (ICS Ada Xcessories -- www.ics.com)
>     IPL (AdaTest -- www.iplbath.com)
>     McCabe and Associates (McCabe QA/Test/Reengineer/2000 -- www.mccabe.com)
>     Objective Interface Systems (ORBExpress/Ada -- www.ois.com)
>     Praxis Critical Systems (SPARK -- www.praxis-cs.co.uk)
>     Vector Software (AdaCAST -- www.vectors.com)

"Black and White software" ( http://www.blackwhite.com ) makes UIM/Ada, a
Motif GUI builder that generates Ada code (like Aonix's TeleUSE). I'd
definitely throw in anyone who makes a GUI builder.

--
T.E.D.

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

Re: Compiler Hosts and Targets

[dewarr]

In article <755s4v$7vh$1@nnrp1.dejanews.com>,
  dennison@telepath.com wrote:
> Non-Ada issue: There is no such thing as Pentium/VxWorks,
> unfortunately.
> There is an Intel3(or4)86/VxWorks target, which may be
> used on Pentiums
> (although it will run *slower* unless you change the
> caching behavior!). But
> if you use that on a Pentium it will not generated
> Pentium optimized code.

Not quite clear what "it" refers to here. VxWorks is not
in the business of "generating" code at all! It is of
course the case that compilers typically used with VxWorks,
including gcc, and GNAT, can of course generate Pentium
optimized code.


-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

Re: Compiler Hosts and Targets

[dennison]

In article <75c6ir$l61$1@nnrp1.dejanews.com>,
  dewarr@my-dejanews.com wrote:
> In article <755s4v$7vh$1@nnrp1.dejanews.com>,
>   dennison@telepath.com wrote:
> > Non-Ada issue: There is no such thing as Pentium/VxWorks,
> > unfortunately.
> > There is an Intel3(or4)86/VxWorks target, which may be
> > used on Pentiums
> > (although it will run *slower* unless you change the
> > caching behavior!). But
> > if you use that on a Pentium it will not generated
> > Pentium optimized code.
>
> Not quite clear what "it" refers to here. VxWorks is not
> in the business of "generating" code at all! It is of
> course the case that compilers typically used with VxWorks,
> including gcc, and GNAT, can of course generate Pentium
> optimized code.

Of course "it" was a simplification. I didn't think anyone here was
interested in the gory details. But since you seem to insist upon them I'll go
ahead. The rest of you may want to go get a sandwich or something. :-)

Tornado, the system that came w/ our VxWorks (targetted to a PC clone)
contained the source code for the OS, a BSP for the PC, gcc, and a GUI
environment for building the OS and user programs. The BSP's that are
available are for PC386 and PC486 (no "PCPentium", no "PCPentiumII"). The gcc
we get is a prebuilt version for the 386 (not "Pentium", 386). With this
system as shipped, you cannot build Pentium targetted code. (I suppose its
possible they have a bare Pentium BSP, but that wouldn't be much fun to use
on a PC unless you enjoy writing device drivers).

The compiler we use is the GreenHills Ada compiler, which to my knowledge is
the *only* validated Ada compiler on this platform. (Rational does have some
VxWorks compilers for other CPU's). The GreenHills compiler cannot build
PentiumII or Pentuim optimized code; only 368 and 486.

The bug I was referrring to is a known VxWorks bug in this configuration. The
OS performs a cache invalidate whenever a task switch occurs. Apparently this
instruction takes roughly 5 cycles on the Pentium, but 2000 on the Pentium
II. If you run your system with relatively high rate tasks (2KHz in our
case), this causes a Pentium II system to run *slower* than Pentium I. We
have measured this in house (and we are not the only ones). The folks at
WindRiver actually sent someone out here to track it down, and ended up
changing our OS source files to get around the problem (*much* better now).

Everyone back from the fridge yet? Good. The point is that while you can run a
VxWorks system on a Pentium PC, neither VxWorks, not Tornado, nor Green Hills
are really designed for it.

--
T.E.D.

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

Re: Compiler Hosts and Targets

[dewarr]

In article <75dr0b$h$1@nnrp1.dejanews.com>,
  dennison@telepath.com wrote:
> The gcc
> we get is a prebuilt version for the 386 (not "Pentium",
> 386). With this
> system as shipped, you cannot build Pentium targetted
> code.

That is surely wrong, there is no such thing as gcc
specifically targetted to the 386, you build a version of
gcc for the x86, and then use the switches in the normal
manner to control which target (386/486, 586=pentium,
686=pentium-pro) you want.

Are you really saying that Wind River cripples this
standard gcc feature? I doubt it!

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

Re: Compiler Hosts and Targets

[dennison]

In article <75ffh3$cgk$1@nnrp1.dejanews.com>,
  dewarr@my-dejanews.com wrote:
> In article <75dr0b$h$1@nnrp1.dejanews.com>,
>   dennison@telepath.com wrote:
> > The gcc
> > we get is a prebuilt version for the 386 (not "Pentium",
> > 386). With this
> > system as shipped, you cannot build Pentium targetted
> > code.
>
> That is surely wrong, there is no such thing as gcc
> specifically targetted to the 386, you build a version of
> gcc for the x86, and then use the switches in the normal
> manner to control which target (386/486, 586=pentium,
> 686=pentium-pro) you want.
>
> Are you really saying that Wind River cripples this
> standard gcc feature? I doubt it!

No. But the system gets built from the Tornado GUI. There's no Pentium option
in Tornado. There's no mention in any of the VxWorks manuals about a "586"
option. When WindRiver came here to sell VxWorks to us, we asked them about
Pentium code generation, and they said they couldn't do it yet, and didn't
have a BSP for it (but of course the 486 one would work). There may a way to
turn on any Pentium II optimizations their version of gcc may have by hand
modifying their makefiles. But that implies spending the engieering effort
required to reverse engineer their whole build process, and debug the
results. Frankly just figuring out how to get our hardware working has taken
too much of our engineering effort. I have already mentioned in another post
one of the problems we encountered due to the BSP being 486 targetted.

So some parts of the system (like gcc) may have explicit support for the
Pentium (as opposed to the 486). But VxWorks/Tornado itself does not. A 486
targetted system may work on a Pentium II, but that's not the same thing as
having a Pentuim II targetted system.

--
T.E.D.

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

Re: Compiler Hosts and Targets

[dewarr]

In article <75lqn8$8p9$1@nnrp1.dejanews.com>,
  dennison@telepath.com wrote:
> Pentium (as opposed to the 486). But VxWorks/Tornado
> itself does not. A 486
> targetted system may work on a Pentium II, but that's not
> the same thing as
> having a Pentuim II targetted system.


This is far easier to deal with than you imply. Your Ada
vendor should certainly be able to tell you how to do what
you want!

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

Re: Compiler Hosts and Targets

[dennison]

In article <75nbge$ier$1@nnrp1.dejanews.com>,
  dewarr@my-dejanews.com wrote:
> In article <75lqn8$8p9$1@nnrp1.dejanews.com>,
>   dennison@telepath.com wrote:
> > Pentium (as opposed to the 486). But VxWorks/Tornado
> > itself does not. A 486
> > targetted system may work on a Pentium II, but that's not
> > the same thing as
> > having a Pentuim II targetted system.
>
> This is far easier to deal with than you imply. Your Ada
> vendor should certainly be able to tell you how to do what
> you want!

Well, yes. Since Pentiums run 486 code you can still *use* the system on a
Pentium II. It just runs a lot slower than a Pentuim targetted system would
have (which is more or less what they did tell us). But there is a least one
arcane performance problem we have discovered because this is not an
officially supported target. Our Ada vendor was no help here since it was a
VxWorks problem. It was only fixed when a VxWorks representative flew out
here and hacked our system in person. There may be more.

--
T.E.D.

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

Re: Compiler Hosts and Targets

[dewarr]

In article <75ob4g$b6j$1@nnrp1.dejanews.com>,
  dennison@telepath.com wrote:
>> Well, yes. Since Pentiums run 486 code you can still
>*use* the system on a
> Pentium II. It just runs a lot slower than a Pentuim
> targetted system would
> have (which is more or less what they did tell us). But
> there is a least one
> arcane performance problem we have discovered because
> this is not an
> officially supported target. Our Ada vendor was no help
> here since it was a
> VxWorks problem. It was only fixed when a VxWorks
> representative flew out
> here and hacked our system in person. There may be more.

No, I am saying something different, it is perfectly
possible to run Pentium optimized code under VxWorks,
and any decent Ada compiler for the x86 should be able
to generate such code, and your Ada vendor should be able
to tell you how! For gnu-c code, gcc has the capability
of generating Pentium-optimized code that can be used
under gcc, and again your Ada vendor should be able to
tell you how to do this.

If they can't then this really is not a VxWorks limitation!

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

Re: Compiler Hosts and Targets

[Steven Hovater]

Rational does indeed have a validated Ada (83 and 95) compiler (ApexCross)  for
the Tornado/Intel.

Cheers,
Steve

dennison@telepath.com wrote:

> In article <75c6ir$l61$1@nnrp1.dejanews.com>,
>   dewarr@my-dejanews.com wrote:
> > In article <755s4v$7vh$1@nnrp1.dejanews.com>,
> >   dennison@telepath.com wrote:
> > > Non-Ada issue: There is no such thing as Pentium/VxWorks,
> > > unfortunately.
> > > There is an Intel3(or4)86/VxWorks target, which may be
> > > used on Pentiums
> > > (although it will run *slower* unless you change the
> > > caching behavior!). But
> > > if you use that on a Pentium it will not generated
> > > Pentium optimized code.
> >
> > Not quite clear what "it" refers to here. VxWorks is not
> > in the business of "generating" code at all! It is of
> > course the case that compilers typically used with VxWorks,
> > including gcc, and GNAT, can of course generate Pentium
> > optimized code.
>
> Of course "it" was a simplification. I didn't think anyone here was
> interested in the gory details. But since you seem to insist upon them I'll go
> ahead. The rest of you may want to go get a sandwich or something. :-)
>
> Tornado, the system that came w/ our VxWorks (targetted to a PC clone)
> contained the source code for the OS, a BSP for the PC, gcc, and a GUI
> environment for building the OS and user programs. The BSP's that are
> available are for PC386 and PC486 (no "PCPentium", no "PCPentiumII"). The gcc
> we get is a prebuilt version for the 386 (not "Pentium", 386). With this
> system as shipped, you cannot build Pentium targetted code. (I suppose its
> possible they have a bare Pentium BSP, but that wouldn't be much fun to use
> on a PC unless you enjoy writing device drivers).
>
> The compiler we use is the GreenHills Ada compiler, which to my knowledge is
> the *only* validated Ada compiler on this platform. (Rational does have some
> VxWorks compilers for other CPU's). The GreenHills compiler cannot build
> PentiumII or Pentuim optimized code; only 368 and 486.
>
> The bug I was referrring to is a known VxWorks bug in this configuration. The
> OS performs a cache invalidate whenever a task switch occurs. Apparently this
> instruction takes roughly 5 cycles on the Pentium, but 2000 on the Pentium
> II. If you run your system with relatively high rate tasks (2KHz in our
> case), this causes a Pentium II system to run *slower* than Pentium I. We
> have measured this in house (and we are not the only ones). The folks at
> WindRiver actually sent someone out here to track it down, and ended up
> changing our OS source files to get around the problem (*much* better now).
>
> Everyone back from the fridge yet? Good. The point is that while you can run a
> VxWorks system on a Pentium PC, neither VxWorks, not Tornado, nor Green Hills
> are really designed for it.
>
> --
> T.E.D.
>
> -----------== Posted via Deja News, The Discussion Network ==----------
> http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own

--
Steven
Hovater
svh@rational.com
Software Engineering Consultant
Phone/fax:781-676-2565/2500
Rational
Software                                                                   Pager:
888-906-2209
83 Hartwell Ave, Lexington, MA                                             Amateur
radio: AA1YH

Re: Compiler Hosts and Targets

[dennison]

In article <3685A09B.78400E9B@mediaone.net>,
  Steven Hovater <nh-ho@mediaone.net> wrote:
> Rational does indeed have a validated Ada (83 and 95) compiler (ApexCross)
for
> the Tornado/Intel.
>

Hmm. They do have a certificate registered with EDS. It appears to be quite
new (3 months old). I didn't see it in a search last week because the OS is
erroniously listed as "Tornado 1.1" (Tornado is the development suite which
includes vxWorks.) If you want to use it, you have to host from a solaris
box, though.

--
T.E.D.

-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own