From: "Yannick Duchêne (Hibou57)" <yannick_duchene@yahoo.fr>
Subject: Re: bit numbers in packed arrays of Boolean
Date: Tue, 31 Aug 2010 13:34:00 +0200
Date: 2010-08-31T13:34:00+02:00 [thread overview]
Message-ID: <op.viavqyjmule2fv@garhos> (raw)
In-Reply-To: 82r5hfghjr.fsf@stephe-leake.org
Le Tue, 31 Aug 2010 13:14:48 +0200, Stephen Leake
<stephen_leake@stephe-leake.org> a écrit:
> LRM 13.9(5) says Bit_Array_16_Type and Unsigned_16 have the same
> representation, but it does not specifically address the index order.
At least ARM 3.6 Array Types does not states anything. But as you used
Unchecked_Conversion, it is unlikely to be portable or this result could
be enforced. Let say this is the result of a compiler with descent default
behaviors.
ARM 13.9 Unchecked Type Conversions says:
11.a/2 Implementation defined: The effect of unchecked conversion for
instances with nonscalar result types whose effect is not defined by
the language.
And do not believe the reference defines anything for an array of 16
bit-layout booleans.
Possibly depends on how the index is interpreted: as an index in the
polynomial representation of the word value or as a bit index ? Both would
be legitimate for a compiler.
That was my humble opinion on the topic.
--
“Dual licensing is the Perl's way to disinfect the GNU General Public
Virus!” (anonymous)
next prev parent reply other threads:[~2010-08-31 11:34 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-08-31 11:14 bit numbers in packed arrays of Boolean Stephen Leake
2010-08-31 11:34 ` Yannick Duchêne (Hibou57) [this message]
2010-08-31 12:34 ` Niklas Holsti
2010-08-31 12:41 ` Yannick Duchêne (Hibou57)
2010-08-31 13:08 ` Dmitry A. Kazakov
2010-08-31 13:40 ` Yannick Duchêne (Hibou57)
2010-08-31 13:57 ` sjw
2010-08-31 14:07 ` Dmitry A. Kazakov
2010-08-31 14:30 ` Niklas Holsti
2010-09-02 20:09 ` Randy Brukardt
2010-08-31 18:13 ` Jeffrey Carter
replies disabled
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox