Re: asm in ada

Re: asm in ada

[William J. Thomas]

If its the intel version, look for Machine_Operations_386. Its a real piece
of work, a real gem.

"Etienne" <etienne.baudin@wanadoo.fr> wrote in message
news:Ihps5.149$xr6.264251@nnrp6.proxad.net...
> Hello,
> I'd like to know if inserting asm code is possible in objectAda (how?)
> Thanks
> Etienne
>
>

Re: asm in ada

[Robert Dewar]

In article <fTit5.192$N13.24942@eagle.america.net>,
  "William J. Thomas" <wjthomas@wcvt.com> wrote:
> If its the intel version, look for Machine_Operations_386. Its
a real piece
> of work, a real gem.
>
> "Etienne" <etienne.baudin@wanadoo.fr> wrote in message
> news:Ihps5.149$xr6.264251@nnrp6.proxad.net...
> > Hello,
> > I'd like to know if inserting asm code is possible in
objectAda (how?)
> > Thanks
> > Etienne


Well that's a nice note (the design of the Object Ada facility
comes from my original design for Alsys Ada, back in the old
days when I worked for Alsys). I agree it's a nice design :-)

Note that nearly all compilers will provide a facility for
either machine code insertions, or package machine code, since
this is required by the systems programming annex which most
Ada 95 compilers implement. However, the facility is in general
VERY target dependent (except on the Motorola 88000 where
interestingly the official system ABI dictated what package
Machined_Code must look like for any Ada compiler for that
system, at least so I am told :-) That means you have to look
into the specific compiler documentation from the vendor.

In the case of GNAT, for example, both package machine code
and intrinsics are available, and they are based on interfacing
directly with the internal register transfer language used
by the compiler. This is a little complex, but has the advantage
that the optimizer can fully deal with insertions, they can
be inlined etc. Anyway, you have to go to the particular
compiler in question for this information, and the resulting
program is likely to be highly non-portable.

It is often better to either try to avoid the use of ASM
completely (it would be interesting to know why you are using
it, so often there are ways to do what you want in Ada after
all), or to write a small assembler routine (which surprisingly
is likely to be more portable than writing asm inserts).

Robert Dewar


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: asm in ada

[Gautier de Montmollin]

Robert Dewar:

> It is often better to either try to avoid the use of ASM
> completely (it would be interesting to know why you are using
> it, so often there are ways to do what you want in Ada after
> all), or to write a small assembler routine (which surprisingly
> is likely to be more portable than writing asm inserts).

Apart of hardware-oriented instructions, asm code seems to me
useless if the compiler provides advanced optimizations, like GNAT.
It is in general useless to try to outperform it. It's better
to see the feedback given by produced asm ("gcc -S") and
execution speed, for choosing the most convenient data structures,
"to inline or not to inline", etc. . A good surprise among others
is that passing and using an elegant unconstrained vector seems
faster than passing N numbers as parameters - surely it is more
"speaking" to the optimizer. With 3.13 it is surely even nicer...

As for ObjectAda, despite its wonderful programming environment,
I fear that it doesn't optimize at all (slow code and no option
beyond suppress_all). Unless I'm wrong ?...

In the Borland Pascal days, "to optimize" meant "to write
inline assembler code for that"...

______________________________________________________
Gautier  --  http://members.xoom.com/gdemont/gsoft.htm

Re: asm in ada

[Samuel T. Harris]

Gautier de Montmollin wrote:
> 
> Robert Dewar:
> 
> > It is often better to either try to avoid the use of ASM
> > completely (it would be interesting to know why you are using
> > it, so often there are ways to do what you want in Ada after
> > all), or to write a small assembler routine (which surprisingly
> > is likely to be more portable than writing asm inserts).
> 
> Apart of hardware-oriented instructions, asm code seems to me
> useless if the compiler provides advanced optimizations, like GNAT.
> It is in general useless to try to outperform it.

When our hardware vendor gave us some machine code to do
semaphores which outperforms an Ada task for the same purpose,
and that semaphore is used hundreds of times a second, then
we use a machine code subprogram.

I would say it is usually useless to try to outperform the
compiler. But the thoughful developer will recognize the
appropriate opportunities and exploit them.

-- 
Samuel T. Harris, Principal Engineer
Raytheon, Aerospace Engineering Services
"If you can make it, We can fake it!"