Re: SC_IS_PRERELEASE and TLM_IS_PRERELEASE

Works fine for me.

Thanks,
Philipp

On 18/01/11 13:59, john.aynsley@doulos.com wrote:
> Philipp,
>
> I would go with David and vote for a simple 0 or 1.
>
> John A
>
>
>
> From:
> David C Black <dcblack@xtreme-eda.com>
> To:
> Philipp A. Hartmann <philipp.hartmann@offis.de>
> Cc:
> john.aynsley@doulos.com, systemc-p1666-technical@eda.org
> Date:
> 18/01/2011 03:16
> Subject:
> Re: SC_IS_PRERELEASE and TLM_IS_PRERELEASE
>
>
>
> The standard definitely says lowercase true/false are the correct values.
> However,
> there is another consideration, CPP, the C preprocessor. The GNU 4.0
> manual states:
>
>
> The preprocessor does not know anything about types in the language.
> Therefore, sizeof operators are not recognized in ?#if?, and neither are
> enum constants. They will be taken as identifiers which are not macros,
> and replaced by zero. In the case of sizeof, this is likely to cause the
> expression to be invalid.
>
> The preprocessor calculates the value of expression. It carries out all
> calculations in the widest integer type known to the compiler; on most
> machines supported by GCC this is 64 bits. This is not the same rule as
> the compiler uses to calculate the value of a constant expression, and may
> give different results in some cases. If the value comes out to be nonzero
> , the ?#if? succeeds and the controlled text is included; otherwise it is
> skipped.
>
> So I think we should use 1 and 0.
>
> On Jan 17, 2011, at 5:31 PM, Philipp A. Hartmann wrote:
>
> John,
>
> I've checked this issue on MS Visual C++ as well. There, TRUE/FALSE are
> defined (due to windows.h) and unfortunaly true/false are not supported
> as constant expression (although they should, according to my
> understanding of the C++ standard).
>
> A platform independent definition should then use 1/0, respectively.
> Thus, the extended wording from my last mail seems to be the best way to
> go. The implementer can then provide a suitable definition for all the
> supported platforms.
>
> Opinions?
>
> Greetings from Oldenburg,
> Philipp
>
> On 12/01/11 15:42, Philipp A. Hartmann wrote:
> John,
>
> On 12/01/11 15:13, john.aynsley@doulos.com wrote:
>
> What precisely are you proposing we do instead?
>
> For simplicity, I would propose to just spell 'false' and 'true' in
> lower-case. These are valid constant expressions for the C(++)
> preprocessor, evaluating to 0 and 1.
>
> A more elaborate definition could use the following:
>
> The TLM_IS_PRERELEASE flag shall be a controlling constant expression
> (see ISO/IEC14882:2003, 16.1(4)), evaluating to 0 or 1, respectively.
>
> Greetings from Oldenburg,
> Philipp
>
>
>
> From:
> "Philipp A. Hartmann" <philipp.hartmann@offis.de>
> To:
> john.aynsley@doulos.com
> Cc:
> systemc-p1666-technical@eda.org
> Date:
> 12/01/2011 01:45
> Subject:
> Re: SC_IS_PRERELEASE and TLM_IS_PRERELEASE
>
>
>
> John, all,
>
> I've checked this again. TRUE and FALSE are neither part of the C nor
> the C++ standard.
>
> They are defined by some platforms (e.g. on Windows), and also
> commonly defined in C applications, but the latter mostly due to the
> lack of proper bool support.
>
> Wrt. to backwards compatibility, I highly doubt that there are
> frequent uses of this, since the current TLM-2 kit is just broken in
> that regard. Try the following on a non-Windows platform:
>
> // ---8<---
> #include <systemc>
> #include <tlm.h>
>
> int sc_main( int, char*[] )
> {
> # if TLM_IS_PRERELEASE == TRUE
> std::cout << "TLM pre-release" << std::endl;
> # endif
>
> # if TLM_IS_PRERELEASE == FALSE
> std::cout << "TLM not a pre-release" << std::endl;
> # endif
> return 0;
> }
> // --->8---
>
> This prints, _both_ lines on my default Linux/GCC 4.4 platform, since
> TRUE and FALSE are undefined. And comparing two undefined symbols is
> quite ? undefined. ;-)
>
> Greetings from Oldenburg,
> Philipp
>
> On 11/01/11 13:37, john.aynsley@doulos.com wrote:
> All,
>
> Philipp writes:
>
>
> 9.6.5 SystemC Version
> 11.8.1 TLM version information
>
> SC_IS_PRERELEASE
> TLM_IS_PRERELEASE
>
> "... flag shall be either FALSE or TRUE, not enclosed in quotation
> marks."
>
> Hmmm, maybe 'true' or 'false' (lower-case) would be more portable?
> Or even an application-defined integral constant expression evaluating
>
> to 0/1 (or false/true).
>
>
> [JA] I guess FALSE and TRUE are still required for backward
> compatibility
> in the TLM-2.0 world?
>
>

-- 
Philipp A. Hartmann
Hardware/Software Design Methodology Group
OFFIS Institute for Information Technology
R&D Division Transportation · FuE-Bereich Verkehr
Escherweg 2 · 26121 Oldenburg · Germany · http://offis.de/en/
Phone/Fax: +49-441-9722-420/282 · PGP: 0x9161A5C0 · Skype: phi.har
-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.