comp.lang.ada
 help / color / mirror / Atom feed
From: Robert Dewar <dewar@gnat.com>
Subject: Re: Optimization Performance of GNAT Ada
Date: 2000/04/20
Date: 2000-04-20T00:00:00+00:00	[thread overview]
Message-ID: <8dn2a7$f07$1@nnrp1.deja.com> (raw)
In-Reply-To: 1e9dq0j.1f1x7cv1pnzoaoN%herwin@gmu.edu

In article <1e9dq0j.1f1x7cv1pnzoaoN%herwin@gmu.edu>,
  herwin@gmu.edu (Harry Erwin) wrote:
> How good is it?

Obviously this is not a question with a simple answer. GNAT
uses the gcc code generator, which has a very effective
optimizer with appropriate optimization levels turned on.
Consult the section in the GNAT manual on performance issues
for further details.

After all the question above is misleading, it's not really
what Harry is interested in, they are interested in "How will
GNAT do on my code?" and that's a harder question to answer!
(since it is very dependent on the code that is being run).

Most of our customers have found the performance meets their
requirements, and compares favorably to other compilers. There
are some cases where performance is a problem (e.g. the use of
packed arrays in GNORT mode) but these are problems that are
in the process of being addressed (we just fixed this GNORT
problem for example).

If you have specific performance requirements or problems,
by all means let us know at report@gnat.com (always remembering
to put your customer number in the subject line!)

Robert Dewar
Ada Core Technologies


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




      reply	other threads:[~2000-04-20  0:00 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2000-04-20  0:00 Optimization Performance of GNAT Ada Harry Erwin
2000-04-20  0:00 ` Robert Dewar [this message]
replies disabled

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox