C Language, unnamed UNIONS

unnamed UNIONS

Author	Message
Lee Westla #1 / 8	unnamed UNIONS I am currently developing a tool for use on Linux and Solaris, using GCC to recompile my source between platforms. However, GCC on our sparc station (running solaris), refuses to compile code containing unnamed union/structure members. Does anyone know why this could be? Also, does anyone know a way to get VC++ to behave like a true C compiler, and allow the incorporation of nested functions? Any help would be greatly appreciated. Cheers Lee.
Sun, 10 Apr 2005 05:25:08 GMT

Mike Wahle #2 / 8	unnamed UNIONS Quote: > I am currently developing a tool for use on Linux and Solaris, using > GCC to recompile my source between platforms. However, GCC on our > sparc station (running solaris), refuses to compile code containing > unnamed union/structure members. Does anyone know why this could be? > Also, does anyone know a way to get VC++ to behave like a true C > compiler, and allow the incorporation of nested functions? What do you call 'true C'? Obviously not ISO standard C (nor any of its atecedents I know of), since they do not allow nested functions. Quote: > Any help would be greatly appreciated. Get help here: http://www.accu.org/bookreviews/public/reviews/0sb/beginner_s_c.htm and here: http://www.eskimo.com/~scs/C-faq/top.html -Mike
Sun, 10 Apr 2005 05:47:55 GMT

Richard Heathfiel #3 / 8	unnamed UNIONS Quote: > I am currently developing a tool for use on Linux and Solaris, using > GCC to recompile my source between platforms. However, GCC on our > sparc station (running solaris), refuses to compile code containing > unnamed union/structure members. Does anyone know why this could be? Possibly because unnamed union/structure members are not part of the C language. Quote: > Also, does anyone know a way to get VC++ to behave like a true C > compiler, and allow the incorporation of nested functions? It behaves like a true C compiler by not allowing the incorporation of nested functions. If you want Pascal, you know where to find it. <snip> -- "Usenet is a strange place." - Dennis M Ritchie, 29 July 1999. C FAQ: http://www.eskimo.com/~scs/C-faq/top.html K&R answers, C books, etc: http://users.powernet.co.uk/eton
Sun, 10 Apr 2005 06:01:50 GMT

Mike Wahle #4 / 8	unnamed UNIONS Quote: > > I am currently developing a tool for use on Linux and Solaris, using > > GCC to recompile my source between platforms. However, GCC on our > > sparc station (running solaris), refuses to compile code containing > > unnamed union/structure members. Does anyone know why this could be? > Possibly because unnamed union/structure members are not part of the C > language. Except bit fields. :-) I was poking around the standard and noted this exception, but didn't see an immediately evident statement about this for other types, so I left this question alone. -Mike
Sun, 10 Apr 2005 07:22:30 GMT

Ben Pfaf #5 / 8	unnamed UNIONS Quote: > I am currently developing a tool for use on Linux and Solaris, using > GCC to recompile my source between platforms. However, GCC on our > sparc station (running solaris), refuses to compile code containing > unnamed union/structure members. Does anyone know why this could be? Unnamed structure and union members, except for bit-fields, aren't in the C standard. But if you upgrade to a recent version of GCC, they are supported for compatibility with other compilers.
Sun, 10 Apr 2005 07:32:17 GMT

Lee Westla #6 / 8	unnamed UNIONS Originating from a low-level background, I've only ever used GCC for compilation of C/C++ source code, yet have experienced no problems when generating code containing nested functions. As many of my projects involve a combination of assembly + high-level source, and consist of sub-functions only relevant within the context of another, it is often more efficient to eliminate parameter passing and maintain cache coherence by availing an existing stack frame during the execution of a nested sub-function. In order to facilitate the polymorphic nature of C++, many compilation systems employ stack convolution and thus assume an indeterminate finite stack frame. As a "TRUE" C platform, by definition but possibly not implementation, should always resolve a determinable finite stack frame, nested function generation is both an attractive and optimising feature. With this in mind, maybe I should rephrase the question as follows: "Does anyone know if VC++ has a compilation switch, which when activated, enables it to return a determinable finite stack frame?" That said, if this is an "undocumented" language feature infrequently deemed appropriate by a majority high-level user base, I guess it is specific to the GCC implementation. Cheers Lee
Mon, 11 Apr 2005 02:55:42 GMT

CBFalcone #7 / 8	unnamed UNIONS Quote: > Originating from a low-level background, I've only ever used GCC for > compilation of C/C++ source code, yet have experienced no problems > when generating code containing nested functions. > As many of my projects involve a combination of assembly + high-level > source, and consist of sub-functions only relevant within the context > of another, it is often more efficient to eliminate parameter passing > and maintain cache coherence by availing an existing stack frame > during the execution of a nested sub-function. > In order to facilitate the polymorphic nature of C++, many compilation > systems employ stack convolution and thus assume an indeterminate > finite stack frame. As a "TRUE" C platform, by definition but > possibly not implementation, should always resolve a determinable > finite stack frame, nested function generation is both an attractive > and optimising feature. > With this in mind, maybe I should rephrase the question as follows: > "Does anyone know if VC++ has a compilation switch, which when > activated, enables it to return a determinable finite stack frame?" > That said, if this is an "undocumented" language feature infrequently > deemed appropriate by a majority high-level user base, I guess it is > specific to the GCC implementation. While your motives may be admirable, the subject of specific implementation features remains OT on c.l.c. It cannot be answered by reference to the ISO standard. I suspect the only reason gcc has that extension is to cater to languages with the feature, such as Pascal and Ada. Microsoft caters to nobody other than Gammon, let alone industry standards. It might be possible with some form of non-standard attribute, but nobody here is likely to have any such knowledge. -- Available for consulting/temporary embedded and systems. <http://cbfalconer.home.att.net> USE worldnet address!
Mon, 11 Apr 2005 06:33:03 GMT

those who know me have no need of my nam #8 / 8	unnamed UNIONS in comp.lang.c i read: Quote: >Originating from a low-level background, I've only ever used GCC for >compilation of C/C++ source code, yet have experienced no problems >when generating code containing nested functions. gcc is very much like other compilers, it provides extensions to the language, which this must be since c contains no such provisions. unfortunately what defaults to enabled in one version might be disabled, missing or operate differently in another. but in this group we discuss the standardized language. you want the gcc-help mailing list -- see <http://www.gnu.org/ml/> for more information. Quote: >"Does anyone know if VC++ has a and for this sort of thing you want microsoft.public.vc.ide_general, i think. -- bringing you boring signatures for 17 years
Wed, 13 Apr 2005 16:10:29 GMT

Page 1 of 1

[ 8 post ]