Re: Geomview & Cygwin setup
(adding in the geomview and cygwin lists.)
Chris,
Looking at the changes in:
http://cvs.sourceforge.net/viewcvs.py/geomview/geomview/src/bin/animate/glob.c
there are specific fixes for Cygwin portability that have been
included in Geomview 1.8.2 alpha that affect d_ino use; try changing
the sense of !defined(__CYGWIN__) to just defined(__CYGWIN__) and
see if that works with your compiler. I imagine it will.
I see the output of gcc -v has changed in current cygwin:
Thread model: posix
gcc version 3.4.4 (cygming special) (gdc 0.12, using dmd 0.125)
It used to be just e.g.:
gcc version 3.3.3 (cygwin special)
cygming, not cygwin? ('ming' is a strong insult in the UK. I get the
impression the writer doesn't like cygwin.) No idea what pre-release
gdc or dmd are, or where they've come from, or why I should care about them.
It's awfully tempting to conclude that the compiler is screwing you
around (because the compiler's authors have been screwing it around,
because they hate Cygwin?), and you might want to to try an earlier
version of that. I also wonder if __CYGWIN__ is being defined and
picked up in the compiler environment correctly.
I'm actually running Geomview 1.8.1 and Geomview 1.8.2-alpha on top
of cygwin base 1.5.19-4... but I compiled them on a much earlier
cygwin with an earlier version of gcc, and on an entirely different
machine that I recently migrated the files from. (I haven't tried
recompiling, and your note suggests I would be unwise to.)
But I have compiled SaVi and Geomview on an earlier Cygwin;
http://cvs.sourceforge.net/viewcvs.py/savi/savi-dev/README?rev=1.36&view=markup
tells me the first base Cygwin I used successfully was 1.5.9-1, and
also lists the successful versions of gcc.
cheers,
L.
> Hi Lloyd,
>
> I'm a masters student at UBC in Vancouver and I'm trying to get Geomview
> installed on my Windows machine. I'm following the instructions that
> you've posted via geomview.org but my
compilation is running into an
> error. It seems in glob.c (in directory
> geomview-1.8.2-alpha/src/bin/animate) there is a reference to d_ino of
> the dirent structure. Now from my research into fixing this it seems
> that d_ino is not supported in version 1.5.19-4 but will be in 1.5.20, I
> also tried with version 1.5.8-1 with the same result. Do you know of any
> way I can get around this problem? I hope this is description is
> satisfactory, if you need any clarification please let me know.
Cheers,
Chris Elliott
<[EMAIL PROTECTED]>
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup
On 28 April 2006 12:30, Lloyd Wood wrote:
> cygming, not cygwin? ('ming' is a strong insult in the UK.
Yes, and we also do tongue-in-cheek humour quite well. Well, at least some
of us do.
> I get the impression the writer doesn't like cygwin.)
Who precisely do you suppose "the writer" would have been? Think hard
before answering: who could possibly have been responsible for writing the
cygwin port of gcc, used for compiling programs for cygwin? Give up? The
answer is "someone who would have to be a significant contributor to the
cygwin project". It would be strange to think that someone would spend all
that time and effort doing unpaid voluntary work if they didn't like cygwin
quite a lot. As a general guideline for logical reasoning, you should reject
any assumption you make that leads you to the conclusion that everyone else is
insane / stupid / irrational / entirely oblivious to their own best interests.
Occam's razor and all that.
> It's awfully tempting to conclude that the compiler is screwing you
> around (because the compiler's authors have been screwing it around,
> because they hate Cygwin?),
Well, then you are saying that ridiculous, paranoid, ludicrous conspiracy
theorising, based on zero information and even less research, is quite
tempting. Now go away and find out what MinGW is and why cygwin's compiler
might want to refer to it. Once you stop placing a superfluous emotional
colouring on the word you will be less distracted by irrelevancies and more
able to solve your bug rationally. Stop spreading misinformed FUD.
cheers,
DaveK
--
Can't think of a witty .sigline today
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
Re: Geomview & Cygwin setup
__CYGWIN__ is still valid under gcc version 3.4.4 (cygming special),
as this test shows:
#include
int main(void) {
printf("hello. Testing defines.\n");
#if defined(__CYGWIN__)
printf("__CYGWIN__");
#endif
#if defined(__CYGMING__)
printf("__CYGMING__");
#endif
}
[EMAIL PROTECTED] /home/lloyd
$ gcc -o cygtest cygwin.c
[EMAIL PROTECTED] /home/lloyd
$ ./cygtest
hello. Testing defines.
__CYGWIN__
(Why #define says __CYGWIN__ and gcc -v says something else is
silly.) I suspect something has changed affecting the CYGWIN test in glob.c
L.
At Friday 2006-04-28 12:29 +0100, Lloyd Wood wrote:
(adding in the geomview and cygwin lists.)
Chris,
Looking at the changes in:
http://cvs.sourceforge.net/viewcvs.py/geomview/geomview/src/bin/animate/glob.c
there are specific fixes for Cygwin portability that have been
included in Geomview 1.8.2 alpha that affect d_ino use; try changing
the sense of !defined(__CYGWIN__) to just defined(__CYGWIN__) and
see if that works with your compiler. I imagine it will.
I see the output of gcc -v has changed in current cygwin:
Thread model: posix
gcc version 3.4.4 (cygming special) (gdc 0.12, using dmd 0.125)
It used to be just e.g.:
gcc version 3.3.3 (cygwin special)
cygming, not cygwin? ('ming' is a strong insult in the UK. I get the
impression the writer doesn't like cygwin.) No idea what pre-release
gdc or dmd are, or where they've come from, or why I should care about them.
It's awfully tempting to conclude that the compiler is screwing you
around (because the compiler's authors have been screwing it around,
because they hate Cygwin?), and you might want to to try an earlier
version of that. I also wonder if __CYGWIN__ is being defined and
picked up in the compiler environment correctly.
I'm actually running Geomview 1.8.1 and Geomview 1.8.2-alpha on top
of cygwin base 1.5.19-4... but I compiled them on a much earlier
cygwin with an earlier version of gcc, and on an entirely different
machine that I recently migrated the files from. (I haven't tried
recompiling, and your note suggests I would be unwise to.)
But I have compiled SaVi and Geomview on an earlier Cygwin;
http://cvs.sourceforge.net/viewcvs.py/savi/savi-dev/README?rev=1.36&view=markup
tells me the first base Cygwin I used successfully was 1.5.9-1, and
also lists the successful versions of gcc.
cheers,
L.
> Hi Lloyd,
>
> I'm a masters student at UBC in Vancouver and I'm trying to get Geomview
> installed on my Windows machine. I'm following the instructions that
> you've posted via geomview.org but my
compilation is running into an
> error. It seems in glob.c (in directory
> geomview-1.8.2-alpha/src/bin/animate) there is a reference to d_ino of
> the dirent structure. Now from my research into fixing this it seems
> that d_ino is not supported in version 1.5.19-4 but will be in 1.5.20, I
> also tried with version 1.5.8-1 with the same result. Do you know of any
> way I can get around this problem? I hope this is description is
> satisfactory, if you need any clarification please let me know.
Cheers,
Chris Elliott
<[EMAIL PROTECTED]>
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup
At Friday 2006-04-28 12:46 +0100, Dave Korn wrote:
> > cygming, not cygwin? ('ming' is a strong insult in the UK.
>
> Yes, and we also do tongue-in-cheek humour quite well. Well, at least some
> of us do.
humour is intended to generate an emotional response - laughter.
>> I get the impression the writer doesn't like cygwin.)
>
> Who precisely do you suppose "the writer" would have been?
More accurately, someone who doesn't like Windows.
Now go away and find out what MinGW is and why cygwin's compiler
might want to refer to it.
A referral to it and bad pun would be CygMinGW, which would at least
include the full MinGW 'clue'. Before mailing I attempted some
searches on 'ming' with 'gcc' and, unsurprisingly, got nowhere.
Clearer, and backwardly compatible, would be a gcc version report of
something like:
gcc version 3.4.4 (cygwin special) (MinGW, gdc 0.12, using dmd 0.125)
which has the advantages of being unambiguous and providing
searchable clues for context.
Once you stop placing a superfluous emotional
colouring on the word you will be less distracted by irrelevancies and more
able to solve your bug rationally. Stop spreading misinformed FUD.
The attempted pun to generate humour is itself (surprise) 'a
superfluous emotional colouring on the word', and is the distraction
leading me to irrelevancies. That superfluous attempt at humour has
certainly generated an emotional response.
Saying 'ming' instead of 'MinGW' is a deliberately misleading
distraction. I was deliberately misinformed by gcc -v.
__CYGWIN__ is defined for gcc; gcc -v should report 'cygwin special'.
I do not recommend changing the environment to __CYGMING__ to match
in a further attempt at humour.
If you want people to use the information you provide rationally, do
not colour that information with irrational attempts at humour to
provide misinformation that causes doubt.
We need to trust the output of gcc -v.
L.
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup
On 28 April 2006 14:35, Lloyd Wood wrote: > __CYGWIN__ is still valid under gcc version 3.4.4 (cygming special), > as this test shows: You really need to read some manuals, as you wouldn't have to do things like this if you had read about the -dM option. (You also wouldn't have led yourself up the garden path about mingw if you had read about -mno-cygwin). Hint: gcc -dM -E - < /dev/null > (Why #define says __CYGWIN__ and gcc -v says something else is > silly.) No it isn't. Your assumption that the two have to be in some way related is silly. As is attempting to parse the output from "gcc -v" to detect a given target rather than using one of the predefined macros, which is what they are for and how it is supposed to be done. The output from "gcc -v" is for *humans* to read, and version strings are allowed to have free-format text and no guarantees are provided regarding the content or formatting of that text or how it may or may not change in the future. >I suspect something has changed affecting the CYGWIN test in glob.c First you said that __CYGWIN__ is still valid. Then you think that something has changed affecting the '#ifdef __CYGWIN__' test in glob.c. Since you acknowledge that __CYGWIN__ is still valid, presumably you believe that #ifdef has been changed to only succeed if a symbol is not defined? Now, if you stop posting random guesses and uninformed speculation and try and tell us what the actual *problem* is, perhaps we can answer some questions for you about the recent changes to d_ino under cygwin? cheers, DaveK -- Can't think of a witty .sigline today -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup
On 28 April 2006 15:04, Lloyd Wood wrote: [most of what I want to say is in the response I just sent before I just received this reply, so I'm only going to reiterate one single *very* important point]. > Clearer, and backwardly compatible, would be a gcc version report of > something like: > gcc version 3.4.4 (cygwin special) (MinGW, gdc 0.12, using dmd 0.125) > Saying 'ming' instead of 'MinGW' is a deliberately misleading > distraction. No, it's a linguistic constuct called a "contraction". You saw "cyg" without the "win" and didn't complain about that. > I was deliberately misinformed by gcc -v. No, you misinformed yourself by your false assumptions about the nature of gcc -v. > __CYGWIN__ is defined for gcc; gcc -v should report 'cygwin special'. > I do not recommend changing the environment to __CYGMING__ to match > in a further attempt at humour. > > If you want people to use the information you provide rationally, do > not colour that information with irrational attempts at humour to > provide misinformation that causes doubt. > > We need to trust the output of gcc -v. No, this is completely wrong and absolutely not what you want to do. The output of "gcc -v" is a human-readable summary of various internal information which can be obtained by more reliable means for programmatic use. Your attempt to indirectly deduce the same info from a freeform text string presupposes many assumptions that simply do not hold, and therefore is doomed to incorrectness. cheers, DaveK -- Can't think of a witty .sigline today -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup
At Friday 2006-04-28 15:17 +0100, Dave Korn wrote: > If you want people to use the information you provide rationally, do > not colour that information with irrational attempts at humour to > provide misinformation that causes doubt. > > We need to trust the output of gcc -v. No, this is completely wrong and absolutely not what you want to do. The output of "gcc -v" is a human-readable summary of various internal information which can be obtained by more reliable means for programmatic use. So you admit that gcc -v's output is deliberately unreliable and will lie to the human reader, then. L. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup
At Friday 2006-04-28 15:11 +0100, Dave Korn wrote: perhaps we can answer some questions for you about the recent changes to d_ino under cygwin? So, something has changed after all? Perhaps you could just point me at a document detailing those changes to d_ino that you mention? L. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup
On 28 April 2006 15:42, Lloyd Wood wrote: > At Friday 2006-04-28 15:17 +0100, Dave Korn wrote: >>> If you want people to use the information you provide rationally, do >>> not colour that information with irrational attempts at humour to >>> provide misinformation that causes doubt. >>> >>> We need to trust the output of gcc -v. >> >> No, this is completely wrong and absolutely not what you want to do. The >> output of "gcc -v" is a human-readable summary of various internal >> information which can be obtained by more reliable means for programmatic >> use. > > So you admit that gcc -v's output is deliberately unreliable and will > lie to the human reader, then. What are you gibbering about? "Unreliable"? It's the exact verbatim same every single time you run it. That seems 100% reliable to me. Of course, it's pretty unreliable if you try to use it for purposes it is not intended for, like matching different versions of gcc that target the same arch, because there can be more than one version of gcc for any given target and they will have different version strings. It's also unreliable if you try and use it to make the tea or predict the weather. All three of these uses are not what it was intended for and hence your fault, not the compiler's or anyone else's. And "lie"? Now you're attributing intentionality to an inanimate object. Give it up. It's only *you* who believes that the output of gcc -v is meant to be some kind of identification mechanism. AND YOU ARE WRONG. Sure, there's enough information there for an intelligent human being to extract the regularities and notice the differences, but it's a false inference that those regularities are systematic, or that a simple pattern-matching test can suffice to replace the kind of intellectual skills that a human being brings to bear on the problem. cheers, DaveK -- Can't think of a witty .sigline today -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup
At Friday 2006-04-28 15:17 +0100, Dave Korn wrote: > __CYGWIN__ is defined for gcc; gcc -v should report 'cygwin special'. > I do not recommend changing the environment to __CYGMING__ to match > in a further attempt at humour. > > If you want people to use the information you provide rationally, do > not colour that information with irrational attempts at humour to > provide misinformation that causes doubt. > > We need to trust the output of gcc -v. No, this is completely wrong and absolutely not what you want to do. The output of "gcc -v" is a human-readable summary of various internal information which can be obtained by more reliable means for programmatic use. Your attempt to indirectly deduce the same info from a freeform text string presupposes many assumptions that simply do not hold, and therefore is doomed to incorrectness. In that case, gcc -v may as well return an empty string, since, as you say, it's always going to be wrong, and because obviously even humans have trouble parsing its output. L. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup
On 28 April 2006 15:45, Lloyd Wood wrote: > At Friday 2006-04-28 15:11 +0100, Dave Korn wrote: >> perhaps we can answer some questions >> for you about the recent changes to d_ino under cygwin? > > So, something has changed after all? If you don't read *carefully* and with attention to detail, you're not going to get anywhere. Your question is a straw man: nobody has claimed that nothing has changed anywhere in the universe. Please try and remember YOUR OWN words: >I suspect something has changed affecting the CYGWIN test in glob.c I pointed out that nothing has changed affecting the cygwin test in glob.c; it is a total invention on your part to suppose that is the same as claiming nothing else has changed elsewhere. Don't go attempting to put words in my mouth because it won't wash. > Perhaps you could just point me at a document detailing those changes > to d_ino that you mention? No, because no such document exists. It has been discussed at great length over the past couple of months on the cygwin mailing list. You are at a university and presumably are familiar with the concept of "research", and "googling for information", aren't you? cheers, DaveK -- Can't think of a witty .sigline today -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup
At Friday 2006-04-28 16:01 +0100, Dave Korn wrote: On 28 April 2006 15:45, Lloyd Wood wrote: > At Friday 2006-04-28 15:11 +0100, Dave Korn wrote: >> perhaps we can answer some questions >> for you about the recent changes to d_ino under cygwin? > > So, something has changed after all? If you don't read *carefully* and with attention to detail, you're not going to get anywhere. Your question is a straw man: nobody has claimed that nothing has changed anywhere in the universe. [..] > Perhaps you could just point me at a document detailing those changes > to d_ino that you mention? No, because no such document exists. It has been discussed at great length over the past couple of months on the cygwin mailing list. Thankyou for that hint. From that we have established: yes, something has changed in d_ino; yes, documents exist. (Get over yourself.) L. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Geomview & Cygwin setup
On Fri, Apr 28, 2006 at 12:29:52PM +0100, Lloyd Wood wrote: >It's awfully tempting to conclude that the compiler is screwing you >around (because the compiler's authors have been screwing it around, >because they hate Cygwin?), and you might want to to try an earlier >version of that. I also wonder if __CYGWIN__ is being defined and >picked up in the compiler environment correctly. As the person who started using the term "cygming" in the gcc context, I want to assure you that the software was not intended to maliciously cause people problems. I understand that this happens quite often in the free software world -- people spend hours cackling with glee as they produce a binary package whose sole intent is to provide the illusion of usefulness, when it will really just format a hard drive or something. However, although I did not actually produce the package in question, I can vouch for the person who did. So please set aside your fears and concerns about gcc. The word "ming" in this context merely reflects an unforgivable lack of knowledge of British slang. I should point out that it is unlikely that anyone is going to be changing gcc to accommodate the needs of Geomview (whatever that is) so perhaps you should mark Geomview as "unsupported on cygwin" since this version number difference is apparently close to an insurmountable problem. cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Geomview & Cygwin setup
On Fri, Apr 28, 2006 at 04:17:14PM +0100, Lloyd Wood wrote: >At Friday 2006-04-28 16:01 +0100, Dave Korn wrote: >>On 28 April 2006 15:45, Lloyd Wood wrote: >> >>> At Friday 2006-04-28 15:11 +0100, Dave Korn wrote: perhaps we can answer some questions for you about the recent changes to d_ino under cygwin? >>> >>> So, something has changed after all? >> >> If you don't read *carefully* and with attention to detail, >>you're not going >>to get anywhere. Your question is a straw man: nobody has claimed that >>nothing has changed anywhere in the universe. >[..] >>> Perhaps you could just point me at a document detailing those changes >>> to d_ino that you mention? >> >> No, because no such document exists. It has been discussed at great >> length >>over the past couple of months on the cygwin mailing list. > >Thankyou for that hint. From that we have established: yes, something >has changed in d_ino; yes, documents exist. > >(Get over yourself.) Oh Chinese Dynasty you. cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup
On 28 April 2006 16:17, Lloyd Wood wrote: > (Get over yourself.) You are the one who believes that reality should be changed to match your false assumptions about the intent of "gcc -v". You are also the one who has publicly accused Gerrit et al of committing an act of massive unprofessionalism and abusing their responsibilities as maintainers. Hence I suggest you take a big spoonful of your own medicine. [ This space intentionally left blank for L.Wood to explain once more why he is right and everyone else in the world is wrong. ] cheers, DaveK -- Can't think of a witty .sigline today -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup
If you need to find out what gcc is targeting, perhaps you should
use "-dumpmachine" instead.
$ gcc -dumpmachine
i686-pc-cygwin
$ gcc -dumpmachine -mno-cygwin
i686-pc-mingw32
Lloyd Wood wrote:
> cygming, not cygwin? ('ming' is a strong insult in the UK. I get
> the impression the writer doesn't like cygwin.)
You think that's bad? When the company I was working for spun off
from AT&T, they decided to name themselves "loo scent". :-) I must
admit I'd never heard of the UK "ming" until you mentioned it. I
suspect we at least pronounce it differently, though.
gsw
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup
At Friday 2006-04-28 11:45 -0400, Williams, Gerald S \(Jerry\) wrote:
If you need to find out what gcc is targeting, perhaps you should
use "-dumpmachine" instead.
$ gcc -dumpmachine
i686-pc-cygwin
$ gcc -dumpmachine -mno-cygwin
i686-pc-mingw32
Having identification behaviour dependent on a cygwin-specific flag
like this is.. insane. Doing the former (which is what you'd expect)
makes you none the wiser about mingw. The latter is a special case
where you already have some idea what the target may be, so why would
you issue the command? Begs the question.
Lloyd Wood wrote:
> cygming, not cygwin? ('ming' is a strong insult in the UK. I get
> the impression the writer doesn't like cygwin.)
You think that's bad? When the company I was working for spun off
from AT&T, they decided to name themselves "loo scent". :-)
Ah, the famous brown ring of quality, which, alas, won't exist for much longer.
cheers,
L.
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup
On 28 April 2006 17:07, Lloyd Wood wrote: > At Friday 2006-04-28 11:45 -0400, Williams, Gerald S \(Jerry\) wrote: >> If you need to find out what gcc is targeting, perhaps you should use >> "-dumpmachine" instead. >> >> $ gcc -dumpmachine >> i686-pc-cygwin >> $ gcc -dumpmachine -mno-cygwin >> i686-pc-mingw32 > > Having identification behaviour dependent on a cygwin-specific flag > like this is.. insane. When you use -mno-cygwin, you are invoking A DIFFERENT compiler. Having the *same* identification for two different compilers that target different targets would be insane. cheers, DaveK -- Can't think of a witty .sigline today -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup
At Friday 2006-04-28 17:44 +0100, Dave Korn wrote: On 28 April 2006 17:07, Lloyd Wood wrote: > At Friday 2006-04-28 11:45 -0400, Williams, Gerald S \(Jerry\) wrote: >> If you need to find out what gcc is targeting, perhaps you should use >> "-dumpmachine" instead. >> >> $ gcc -dumpmachine >> i686-pc-cygwin >> $ gcc -dumpmachine -mno-cygwin >> i686-pc-mingw32 > > Having identification behaviour dependent on a cygwin-specific flag > like this is.. insane. When you use -mno-cygwin, you are invoking A DIFFERENT compiler. Having the *same* identification for two different compilers that target different targets would be insane. But gee, that's exactly what gcc -v provides. A single identification for both compilers. Why isn't there a gcc -v -mno-cygwin, then? It's all positively minging. (Chris: note the usage tip. First g pronounced as j, so it's unlike any Chinese dynasties or Flash Gordon's nemesis.) L. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup
On 28 April 2006 18:04, Lloyd Wood wrote: > At Friday 2006-04-28 17:44 +0100, Dave Korn wrote: >> On 28 April 2006 17:07, Lloyd Wood wrote: >> >>> At Friday 2006-04-28 11:45 -0400, Williams, Gerald S \(Jerry\) wrote: If you need to find out what gcc is targeting, perhaps you should use "-dumpmachine" instead. $ gcc -dumpmachine i686-pc-cygwin $ gcc -dumpmachine -mno-cygwin i686-pc-mingw32 >>> >>> Having identification behaviour dependent on a cygwin-specific flag >>> like this is.. insane. >> >> When you use -mno-cygwin, you are invoking A DIFFERENT >> compiler. Having the >> *same* identification for two different compilers that target different >> targets would be insane. > > But gee, that's exactly what gcc -v provides. A single identification > for both compilers. How many times, for crying out loud? The output of "gcc -v" IS NOT A FORMAL IDENTIFIER OF ANY SORT WHATSOEVER. > Why isn't there a gcc -v -mno-cygwin, then? Because it's still the same compiler package? Because nobody has ever cared about it because nobody has ever been daft enough to attempt to misuse the "gcc -v" output in this way before? [ This space intentionally left blank for L.Wood to explain why "gcc -v" ought to be a formal identifier and why everyone else in the world should change things round so we can do it all his way. ] > It's all positively minging. (Chris: note the usage tip. First g > pronounced as j, so it's unlike any Chinese dynasties or Flash > Gordon's nemesis.) Actually it's minging to rhyme with singing, as anyone who has watched Little Britain or Catherine Tate's show should know. [ This space intentionally left blank for L.Wood to insist that everyone should pronounce words the way that /he/ does. ] cheers, DaveK -- Can't think of a witty .sigline today -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup
At Friday 2006-04-28 18:16 +0100, Dave Korn wrote: On 28 April 2006 18:04, Lloyd Wood wrote: > At Friday 2006-04-28 17:44 +0100, Dave Korn wrote: >> On 28 April 2006 17:07, Lloyd Wood wrote: >> >>> At Friday 2006-04-28 11:45 -0400, Williams, Gerald S \(Jerry\) wrote: If you need to find out what gcc is targeting, perhaps you should use "-dumpmachine" instead. $ gcc -dumpmachine i686-pc-cygwin $ gcc -dumpmachine -mno-cygwin i686-pc-mingw32 >>> >>> Having identification behaviour dependent on a cygwin-specific flag >>> like this is.. insane. >> >> When you use -mno-cygwin, you are invoking A DIFFERENT >> compiler. Having the >> *same* identification for two different compilers that target different >> targets would be insane. > > But gee, that's exactly what gcc -v provides. A single identification > for both compilers. How many times, for crying out loud? The output of "gcc -v" IS NOT A FORMAL IDENTIFIER OF ANY SORT WHATSOEVER. > Why isn't there a gcc -v -mno-cygwin, then? Because it's still the same compiler package? Because nobody has ever cared about it because nobody has ever been daft enough to attempt to misuse the "gcc -v" output in this way before? What, reading gcc -v's output using eyes and a screen is misuse? > It's all positively minging. (Chris: note the usage tip. First g > pronounced as j, so it's unlike any Chinese dynasties or Flash > Gordon's nemesis.) Actually it's minging to rhyme with singing, as anyone who has watched Little Britain or Catherine Tate's show should know. Ah, that would be the BBC's "received pronunciation" take on the term. L. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Geomview & Cygwin setup
Lloyd Wood wrote: At Friday 2006-04-28 18:16 +0100, Dave Korn wrote: On 28 April 2006 18:04, Lloyd Wood wrote: > At Friday 2006-04-28 17:44 +0100, Dave Korn wrote: >> On 28 April 2006 17:07, Lloyd Wood wrote: >> >>> At Friday 2006-04-28 11:45 -0400, Williams, Gerald S \(Jerry\) wrote: If you need to find out what gcc is targeting, perhaps you should use "-dumpmachine" instead. $ gcc -dumpmachine i686-pc-cygwin $ gcc -dumpmachine -mno-cygwin i686-pc-mingw32 >>> >>> Having identification behaviour dependent on a cygwin-specific flag >>> like this is.. insane. >> >> When you use -mno-cygwin, you are invoking A DIFFERENT >> compiler. Having the >> *same* identification for two different compilers that target different >> targets would be insane. > > But gee, that's exactly what gcc -v provides. A single identification > for both compilers. How many times, for crying out loud? The output of "gcc -v" IS NOT A FORMAL IDENTIFIER OF ANY SORT WHATSOEVER. > Why isn't there a gcc -v -mno-cygwin, then? Because it's still the same compiler package? Because nobody has ever cared about it because nobody has ever been daft enough to attempt to misuse the "gcc -v" output in this way before? What, reading gcc -v's output using eyes and a screen is misuse? > It's all positively minging. (Chris: note the usage tip. First g > pronounced as j, so it's unlike any Chinese dynasties or Flash > Gordon's nemesis.) Actually it's minging to rhyme with singing, as anyone who has watched Little Britain or Catherine Tate's show should know. Ah, that would be the BBC's "received pronunciation" take on the term. Isn't this thread getting a bit off-topic? Llyod, if you feel there is a problem here that needs a resolution, may I suggest that you submit a patch? At least then the discussion could turn from visual and linguistic aspects of gcc back to technical. -- Larry Hall http://www.rfk.com RFK Partners, Inc. (508) 893-9779 - RFK Office 838 Washington Street (508) 893-9889 - FAX Holliston, MA 01746 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup
At Friday 2006-04-28 17:44 +0100, you wrote: On 28 April 2006 17:07, Lloyd Wood wrote: > At Friday 2006-04-28 11:45 -0400, Williams, Gerald S \(Jerry\) wrote: >> If you need to find out what gcc is targeting, perhaps you should use >> "-dumpmachine" instead. >> >> $ gcc -dumpmachine >> i686-pc-cygwin >> $ gcc -dumpmachine -mno-cygwin >> i686-pc-mingw32 > > Having identification behaviour dependent on a cygwin-specific flag > like this is.. insane. When you use -mno-cygwin, you are invoking A DIFFERENT compiler. If you don't install any of the gcc-mingw- upgrade stuff (devel tab in Cygwin), you only want ONE compiler. L. -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Geomview & Cygwin setup
Lloyd Wood wrote: > If you don't install any of the gcc-mingw- upgrade stuff (devel tab > in Cygwin), you only want ONE compiler. It's not possible to install one without the other, at least not without overriding setup.exe's wishes, because gcc-core includes gcc-mingw-core on the "requries" line, and similarly for all the other modules. This means if you select one the other will be added automatically. Brian -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Geomview & Cygwin setup
Brian Dessent wrote: Lloyd Wood wrote: If you don't install any of the gcc-mingw- upgrade stuff (devel tab in Cygwin), you only want ONE compiler. It's not possible to install one without the other, at least not without overriding setup.exe's wishes, because gcc-core includes gcc-mingw-core on the "requries" line, and similarly for all the other modules. This means if you select one the other will be added automatically. And, implied by Brian's response but is worthwhile to state explicitly, if you choose to override setup.exe's defaults here, it's assumed that you know what you're doing (i.e. what the consequences are, etc.) But then you're creating a custom installation of Cygwin and such installs are not supported by this list. That means you're on your own if you run into problems. I don't know if this situation applies to Lloyd's case or not. But I thought it prudent to tie up any perceived loophole here. :-) -- Larry Hall http://www.rfk.com RFK Partners, Inc. (508) 893-9779 - RFK Office 838 Washington Street (508) 893-9889 - FAX Holliston, MA 01746 -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Geomview & Cygwin setup
On Fri, Apr 28, 2006 at 03:39:14PM -0400, Larry Hall (Cygwin) wrote: >Brian Dessent wrote: >>Lloyd Wood wrote: >> >>>If you don't install any of the gcc-mingw- upgrade stuff (devel tab >>>in Cygwin), you only want ONE compiler. >> >>It's not possible to install one without the other, at least not without >>overriding setup.exe's wishes, because gcc-core includes gcc-mingw-core >>on the "requries" line, and similarly for all the other modules. This >>means if you select one the other will be added automatically. > >And, implied by Brian's response but is worthwhile to state explicitly, >if you choose to override setup.exe's defaults here, it's assumed that >you know what you're doing (i.e. what the consequences are, etc.) But >then you're creating a custom installation of Cygwin and such installs >are not supported by this list. That means you're on your own if you >run into problems. I don't know if this situation applies to Lloyd's >case or not. But I thought it prudent to tie up any perceived loophole >here. :-) Let me state something else, too. We're not going to be changing Cygwin's gcc. I haven't seen anything in this thread to support the notion that gcc is doing anything wrong. I will admit, however, to being biased by the fact that the initial message in this discussion was rather insulting even if it was arguably intended to be "humorous". So, that means that "Geomview" will either have to use what we provide or decide not to support Cygwin. Given this statement of fact, further complaints about how things work or offering of opinions about how things should work would be fruitless. So, I would appreciate it if this thread would now die or move to cygwin-talk. Gerald Williams has already valiantly tried to do this once. Let's follow his example. cygwin-talk AT cygwin PERIOD com cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Geomview & Cygwin setup
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Lloyd Wood wrote: >> Actually it's minging to rhyme with singing, as anyone who has watched >> Little Britain or Catherine Tate's show should know. > > Ah, that would be the BBC's "received pronunciation" take on the term. Well, its the same pronunciation in my part of the country. Never heard it used any other way... Anyway, give it up. - -- Al Slater -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.0 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFEVN1iz4fTOFL/EDYRAlZiAJ9ryWIuG7mfhQTShLuzJXT9br4c3QCfY2zo 7vwupz7H7vtZEA/5zzPrPbo= =th6b -END PGP SIGNATURE- -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
RE: Geomview & Cygwin setup [Attn gcc maintainer]
Dave Korn wrote: > > (Why #define says __CYGWIN__ and gcc -v says something else is > > silly.) > > No it isn't. Your assumption that the two have to be in some way related is > silly. As is attempting to parse the output from "gcc -v" to detect a given > target rather than using one of the predefined macros, which is what they are > for and how it is supposed to be done. The output from "gcc -v" is for > *humans* to read, and version strings are allowed to have free-format text and > no guarantees are provided regarding the content or formatting of that text or > how it may or may not change in the future. First, I agree with most of the points Dave has made - cygwin gcc is a single package (hence -v output that uses a phrase intended to show that both cygwin native and mingw cross-compilation is supported, and in no way was intended to slight cygwin). I agree that Lloyd is reading too much into the current output of -v. I also agree that -v is not the appropriate way to determine what the compiler supports - on cygwin, gcc without options will compile a cygwin native executable, and #define __CYGWIN__ to a nonzero value, while gcc with the -mno-cygwin option will cross-compile mingw native executables, and leave __CYGWIN__ undefined. Using the output of -dumpmachine, or better yet, using the autoconf tradition of compiling a test program to see what the actual compiler's behavior is instead of relying on version strings, is more appropriate. However, I feel that the output of gcc -v is not entirely freeform, because part of it is also used in gcc --version. In this recent autoconf thread, http://lists.gnu.org/archive/html/autoconf/2006-04/msg00116.html, it was pointed out that Debian has a similar bug where their gcc's version output violates GNU coding standards by not ending in a version number. Gerritt, I would suggest that the next time you package gcc for cygwin, that you alter the version string such that it ends in a version number, perhaps something like: gcc (GCC) 3.4.4 (cygwin/mingw special, gdc 0.12, dmd 0.125) 3.4.4-2 Now, as to the real question about d_ino semantics. Cygwin versions up to 1.5.18 declared d_ino, but were buggy in their implementation such that d_ino sometimes disagreed with st_ino (for example, on NTFS drives). This bug forced me to compile cygwin coreutils as though d_ino did not exist, since there are applications that do care about the correct value of d_ino. Cygwin 1.5.19 removed the declaration of d_ino altogether, as permitted by POSIX, because of this bug. Therefore, if your app fails to compile when d_ino is not present, it is a bug in your app for disobeying POSIX/SUSv3, which explicitly state that d_ino is optional. Cygwin 1.5.20 will introduce a patch that ensures d_ino is always correct, at the expense of a slowdown in cases where Windows is not helpful (mainly when dealing with a drive on NT machines), so it will reintroduce a d_ino declaration in the headers. You can test this out now by using a cygwin snapshot. -- Eric Blake -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
Re: Geomview & Cygwin setup [Attn gcc maintainer]
On Sat, Apr 29, 2006 at 04:46:31PM +, Eric Blake wrote: >However, I feel that the output of gcc -v is not entirely freeform, because >part of it is also used in gcc --version. In this recent autoconf thread, >http://lists.gnu.org/archive/html/autoconf/2006-04/msg00116.html, >it was pointed out that Debian has a similar bug where their gcc's >version output violates GNU coding standards by not ending in a >version number. Gerritt, I would suggest that the next time you >package gcc for cygwin, that you alter the version string such that >it ends in a version number, perhaps something like: >gcc (GCC) 3.4.4 (cygwin/mingw special, gdc 0.12, dmd 0.125) 3.4.4-2 When I first introduced the parentheses in the version, I tried to follow advice of one of the gcc developers and I patterned the output vaguely on some version of Red Hat's gcc. I've just run 'gcc --version' on Fedora, Gentoo, Cygwin, and Debian,. All of their version strings end with a parenthesized expression. The only thing that they seem to do that Cygwin doesn't is include some variation of the gcc version within the parentheses. So, if you know you're running on "Gentoo Linux" you'll be able to find out what release of, e.g., gcc 3.3.4 you're using). This particular bit of information would not solve the intractable problem which started this discussion. I don't see any reason for Gerrit to change anything except possibly to put everything within one parenthesized expression. "GNU standards" aside, we seem to be in good company. Now, let's move on from this topic? Please? I really don't think it deserves this much attention. cgf -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
