- mixing compilers (g++ but Intel others)? - 25 Updates
| Jerry Stuckle <jstucklex@attglobal.net>: Apr 29 08:36PM -0400 On 4/29/2016 3:50 PM, Christian Gollwitzer wrote: > undefined behaviour. (Though, practically it's doing what it is intened > to do) > Christian Not at all. The determination is made at compile time, as the requirements specified. They didn't say HOW it was to be determined. And you obviously don't understand how unions work. Show me where it is "undefined behavior". -- ================== Remove the "x" from my email address Jerry Stuckle jstucklex@attglobal.net ================== |
| Jerry Stuckle <jstucklex@attglobal.net>: Apr 29 08:38PM -0400 On 4/29/2016 5:02 PM, Ben Bacarisse wrote: > the endianess of the machine (i.e. how consecutively addressed bytes > correspond to value bytes). > <snip> Not at all. You just don't understand how unions work. They are NOT the same as structures. And if sizeof(short) == 1, then endian is immaterial. But I dare you to show me a single implementation where sizeof(short) == 1. But if you continue to be pedantic about it, then use int instead of short. -- ================== Remove the "x" from my email address Jerry Stuckle jstucklex@attglobal.net ================== |
| "Öö Tiib" <ootiib@hot.ee>: Apr 29 07:54PM -0700 On Friday, 29 April 2016 19:23:50 UTC+3, Jerry Stuckle wrote: > No, I have moved no goalposts. My statements have been consistent since > the start. Just because your limited experience hasn't exposed you to > other options does not mean they don't exist. You started with patronizing "very experienced" claim that some compilers use different .o and .a formats. What concretely? That is "demand" and "not important". Instead you move goalposts to nonsense claim that no one "is required" to use same .o and .a formats. Who has claimed opposite? > > format specified by ISO/IEC 21320-1:2015. Everybody just wants to. > Now who's moving the goalposts - or at least bringing up another red > herring? It was just an example how that "are not required to" is nonsense that no one has argued with. |
| Jerry Stuckle <jstucklex@attglobal.net>: Apr 29 10:59PM -0400 On 4/29/2016 10:54 PM, Öö Tiib wrote: > use different .o and .a formats. What concretely? That is "demand" and > "not important". Instead you move goalposts to nonsense claim that no > one "is required" to use same .o and .a formats. Who has claimed opposite? Patronizing - IN YOUR OPINION. But you have multiple times proven you are just a pig someone (not only me) tries to teach to sing. I have moved NOT goalposts - the only "movement" is in your mind. And multiple people, including you, have claimed all .o files are the same in Linux. >> herring? > It was just an example how that "are not required to" is nonsense that > no one has argued with. No, it's just another example of trying to teach a pig to sing. Every time you are proven wrong, YOU try to move the goalposts. It doesn't work. -- ================== Remove the "x" from my email address Jerry Stuckle jstucklex@attglobal.net ================== |
| "Öö Tiib" <ootiib@hot.ee>: Apr 29 08:39PM -0700 On Saturday, 30 April 2016 05:59:25 UTC+3, Jerry Stuckle wrote: > moved NOT goalposts - the only "movement" is in your mind. > And multiple people, including you, have claimed all .o files are the > same in Linux. I have written nothing about Linux or about files being same. It was you who claimed that 'ar' is gcc tool and I wrote that it has been Unix utility for ages. > > no one has argued with. > No, it's just another example of trying to teach a pig to sing. Every > time you are proven wrong, YOU try to move the goalposts. It doesn't work. Sorry, if I tried to teach you to sing. I know you can only oink back. |
| Jerry Stuckle <jstucklex@attglobal.net>: Apr 29 11:42PM -0400 On 4/29/2016 11:39 PM, Öö Tiib wrote: > I have written nothing about Linux or about files being same. It was you > who claimed that 'ar' is gcc tool and I wrote that it has been Unix > utility for ages. So? That still doesn't mean another toolkit can't have its own version. >> No, it's just another example of trying to teach a pig to sing. Every >> time you are proven wrong, YOU try to move the goalposts. It doesn't work. > Sorry, if I tried to teach you to sing. I know you can only oink back. ROFLMAO! Is that all the better you can do, pig? -- ================== Remove the "x" from my email address Jerry Stuckle jstucklex@attglobal.net ================== |
| "Öö Tiib" <ootiib@hot.ee>: Apr 29 08:52PM -0700 On Saturday, 30 April 2016 06:43:02 UTC+3, Jerry Stuckle wrote: > > who claimed that 'ar' is gcc tool and I wrote that it has been Unix > > utility for ages. > So? That still doesn't mean another toolkit can't have its own version. And that I also already answered that no one argues that they can't. They can. Also it is impossible to prove negative that no one does use that liberty. However you argue that they do. So it is up to you to prove that. > >> time you are proven wrong, YOU try to move the goalposts. It doesn't work. > > Sorry, if I tried to teach you to sing. I know you can only oink back. > ROFLMAO! Is that all the better you can do, pig? Oh yes I forgot that you can also roll on the floor oinking your ass off. :D |
| Jerry Stuckle <jstucklex@attglobal.net>: Apr 29 11:58PM -0400 On 4/29/2016 11:52 PM, Öö Tiib wrote: > can. Also it is impossible to prove negative that no one does use that > liberty. However you argue that they do. So it is up to you to prove > that. You don't read very well either, do you? >>> Sorry, if I tried to teach you to sing. I know you can only oink back. >> ROFLMAO! Is that all the better you can do, pig? > Oh yes I forgot that you can also roll on the floor oinking your ass off. :D Ah, yes. Sorry about that. I shouldn't insult pigs like I did. They are much better than you - as you have so astutely pointed out. -- ================== Remove the "x" from my email address Jerry Stuckle jstucklex@attglobal.net ================== |
| Paavo Helde <myfirstname@osa.pri.ee>: Apr 30 09:46AM +0300 On 30.04.2016 3:38, Jerry Stuckle wrote: > And if sizeof(short) == 1, then endian is immaterial. But I dare you to > show me a single implementation where sizeof(short) == 1. Most DSP-s, e.g.: Analog Devices 32-bit SHARC DSP (all integer types are 32 bits) http://www.verycomputer.com/45_793d80b65c0ca44f_1.htm Texas Instruments TMS32F28xx DSP (all integer types are 16 bits) "http://www.thecodingforums.com/threads/implementations-with-char_bit-32.440062/#post-2435265" |
| Jerry Stuckle <jstucklex@attglobal.net>: Apr 30 08:40AM -0400 On 4/30/2016 2:46 AM, Paavo Helde wrote: > Texas Instruments TMS32F28xx DSP (all integer types are 16 bits) > "http://www.thecodingforums.com/threads/implementations-with-char_bit-32.440062/#post-2435265" How does a 16 bit or 32 bit integer line up with a short size of 1 byte (8 bits)? -- ================== Remove the "x" from my email address Jerry Stuckle jstucklex@attglobal.net ================== |
| "Öö Tiib" <ootiib@hot.ee>: Apr 30 06:57AM -0700 On Saturday, 30 April 2016 15:40:35 UTC+3, Jerry Stuckle wrote: > > "http://www.thecodingforums.com/threads/implementations-with-char_bit-32.440062/#post-2435265" > How does a 16 bit or 32 bit integer line up with a short size of 1 byte > (8 bits)? Read the links? |
| "Öö Tiib" <ootiib@hot.ee>: Apr 30 07:01AM -0700 On Saturday, 30 April 2016 06:59:03 UTC+3, Jerry Stuckle wrote: > > liberty. However you argue that they do. So it is up to you to prove > > that. > You don't read very well either, do you? Posting only cheap insults? That indicates you can have nothing else left. Sorry. |
| Jerry Stuckle <jstucklex@attglobal.net>: Apr 30 10:49AM -0400 On 4/30/2016 10:01 AM, Öö Tiib wrote: >> You don't read very well either, do you? > Posting only cheap insults? That indicates you can have nothing else left. > Sorry. The truth hurts, doesn't it? But I'm through here - you can have the last word. Trolls like you need it. -- ================== Remove the "x" from my email address Jerry Stuckle jstucklex@attglobal.net ================== |
| Jerry Stuckle <jstucklex@attglobal.net>: Apr 30 10:50AM -0400 On 4/30/2016 9:57 AM, Öö Tiib wrote: >> How does a 16 bit or 32 bit integer line up with a short size of 1 byte >> (8 bits)? > Read the links? Once again you don't read very well, do you? And you obviously don't understand even the simplest code. But then trolls are like that. -- ================== Remove the "x" from my email address Jerry Stuckle jstucklex@attglobal.net ================== |
| Chris Vine <chris@cvine--nospam--.freeserve.co.uk>: Apr 30 05:05PM +0100 On Sat, 30 Apr 2016 08:40:24 -0400 > > "http://www.thecodingforums.com/threads/implementations-with-char_bit-32.440062/#post-2435265" > How does a 16 bit or 32 bit integer line up with a short size of 1 > byte (8 bits)? The sizeof operator references the size of char, not the size of an octet (8 bits). sizeof(char) is required to evaluate to 1 in both the C and C++ standards. In the 32 bit case cited, all integer types (char, short, int and long) appear to be comprised of 32 bits. As your test requires sizeof(char) to be less than sizeof(short), it does not work for that case. But it is still quite a neat way of doing it and will work with the vast majority of systems. You ought to put a -std=c99 or -std=c11 compiler flag in though, as reading a different union member than the one last stored gives undefined behaviour with C89 (but not C99/C11). |
| "Öö Tiib" <ootiib@hot.ee>: Apr 30 09:50AM -0700 On Saturday, 30 April 2016 17:50:57 UTC+3, Jerry Stuckle wrote: > Once again you don't read very well, do you? And you obviously don't > understand even the simplest code. > But then trolls are like that. Obviously you lie that you can't read at all. You *did* read the links and *saw* that you are *wrong* again. Therefore you can again only write pathetic cheap insults. It is becoming uninteresting to bust your smart aleck bubbles in a row. |
| Ben Bacarisse <ben.usenet@bsb.me.uk>: Apr 30 08:12PM +0100 >> <snip> > Not at all. You just don't understand how unions work. They are NOT > the same as structures. Unless you address the issue of padding bits in integer types I'll have to assume you don't know what the problem is. It is only by understanding both padding bits *and* unions that you will be able to see how the code can go wrong. > And if sizeof(short) == 1, then endian is immaterial. But I dare you to > show me a single implementation where sizeof(short) == 1. You've had some links already but here is another: http://focus.ti.com/lit/ug/spru281f/spru281f.pdf The C compiler for the TMS320C55x has sizeof(short) == 1, but it also has an integer type with a size > 1. (It also has integer types with padding bits, as it happens.) > But if you continue to be pedantic about it, then use int instead of > short. sizeof(int) == 1 for the compiler I have just cited. -- Ben. |
| David Brown <david.brown@hesbynett.no>: Apr 30 09:21PM +0200 On 29/04/16 20:28, Jerry Stuckle wrote: > printf("BIGENDIAN"); > return 0; > } That is runtime only - it does not generate a constant expression. And it is not portable to systems where sizeof(short) == 1. > 
Subscribe to:
Post Comments (Atom)
|
No comments:
Post a Comment